3 EPA
United States       Office of Air Quality
Environmental Protection  Planning and Standards
Agency          Research Triangle Park NC 27711

Air
Gasoline Distribution
Industry (Stage I) -
Background Information
for Proposed Standards
                                            EPA-453/R-94-002a
                                            January 1994

-------
                                        EPA-453/R-94-002a
Gasoline  Distribution Industry (Stage I) -
        Background  Information for
             Proposed  Standards
                 Emission Standards Division
             U.S. ENVIRONMENTAL PROTECTION AGENCY
                   Office of Air and Radiation
              Office of Air Quality Planning and Standards
              Research Triangle Park, North Carolina 27711

                     January 1994

-------
This report has been reviewed by the Emission Standards Division
of the Office of Air Quality Planning and Standards, EPA, and
approved for publication.  Mention of trade names or commercial
products is not intended 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 NC 27711, (919) 541-
2777,  or from National Technical Information Services, 5285 Port
Royal Road, Springfield VA 22161, (703)  487-4650.

-------
                        ENVIRONMENTAL PROTECTION AGENCY

                            Background Information
                                  and Draft
                        Environmental Impact Statement
                     for Gasoline Distribution Facilities

                                 Prepared by:
      C. Jopdan                                         (Date)''
Director, /Emission Standards Division
U.S. Environmental Protection Agency
Research Triangle Park, N.C.  27711

1.    The proposed standards of performance would limit hazardous air
      pollutant (HAP) emissions from existing and new major source bulk
      gasoline terminals and pipeline breakout stations.  Under section 112(d)
      of the 1990 Clean Air Act, EPA is required to regulate sources of HAPs
      listed pursuant to section 112(c).

2.    Copies of this document have been sent to the following Federal
      Departments:  Labor, Health and Human Services, Defense, Transportation,
      Agriculture, Commerce, Interior, and Energy; the National Science
      Foundation; the Council on Environmental Quality; members of the State
      and Territorial Air Pollution Program Administrators; the Association of
      Local Air Pollution Control Officials; EPA Regional  Administrators; and
      other interested parties.

3.    The comment period for review of this document is 60 days from the date
      of publication of the proposed standards in the Federal Register.  Mr.
      Steve Shedd may be contacted at (919) 541-5397 regarding the date of the
      comment period.

4.    For additional information contact:

      Mr. Steve Shedd
      Chemicals and Petroleum Branch (MD-13)
      U.S.  Environmental Protection Agency
      Research Triangle Park, N.C.  27711
      Telephone:   (919) 541-5397

5.    Copies of this document may be obtained from:

      U.S.  EPA Library (MD-35)
      Research Triangle Park, N.C.  27711
      Telephone:   (919) 541-2777

      National  Technical Information Service
      5285  Port Royal Road
      Springfield, VA  22161
      Telephone:   (703) 487-4650

-------
                        TABLE OF CONTENTS


                                                            Page
                                            i
1.0  SUMMARY	1-1

     1.1  STATUTORY AUTHORITY 	 1-1
     1.2  REGULATORY ALTERNATIVES 	 1-1
     1.3  ENVIRONMENTAL IMPACT	1-3
     1.4  ECONOMIC IMPACT	1-5

2.0  INTRODUCTION	2-1

     2.1  BACKGROUND AND AUTHORITY FOR STANDARDS	2-1
     2.2  SELECTION OF POLLUTANTS AND SOURCE CATEGORIES  .  . 2-6
     2.3  PROCEDURE FOR DEVELOPMENT OF NESHAP  	 2-7
     2.4  CONSIDERATION OF COSTS    	2-9
     2.5  CONSIDERATION OF ENVIRONMENTAL IMPACTS   	 2-10
     2.6  RESIDUAL RISK STANDARDS	2-11

3.0  PROCESSES AND POLLUTANT EMISSIONS  	 3-1

     3.1  GENERAL	3-1
     3.2  FACILITIES AND THEIR EMISSIONS  	 3-3
     3.3  BASELINE EMISSIONS  	 3-40
     3.4  REFERENCES	3-46

4.0  EMISSION CONTROL TECHNIQUES  	 4-1

     4.1  CONTROL TECHNIQUES  	 4-1
     4.2  REFERENCES	4-41

5.0  MODEL PLANTS AND REGULATORY ALTERNATIVES  	 5-1

     5.1  MODEL PLANTS	5-1
     5.2  REGULATORY ALTERNATIVES 	 5-19
     5.3  REFERENCES	5-31

6.0  ENVIRONMENTAL AND ENERGY IMPACTS	6-1

     6.1  AIR POLLUTION EMISSION IMPACTS  	 6-1
     6.2  WATER POLLUTION IMPACTS .... 	 6-19
     6.3  SOLID WASTE IMPACTS 	 6-19
     6.4  ENERGY IMPACTS	6-20
     6.5  OTHER ENVIRONMENTAL IMPACTS 	 6-24
     6.6  REFERENCES	6-26

-------
                  TABLE OF CONTENTS (Concluded)
                                                            Paqe
7.0  CONTROL COSTS	7-1

     7.1  MODEL PLANT COSTS 	 7-1
     7.2  COST ANALYSIS OF REGULATORY ALTERNATIVES   .... 7-30
     7.3  REFERENCES	7-51


8.0  ECONOMIC IMPACT ANALYSIS 	 8-1

     8.1  PROFILE OF THE U.S. GASOLINE
          DISTRIBUTION INDUSTRY 	 8-1
     8.2  ESTIMATES OF BASELINE YEAR CONDITIONS  	 8-48
     8.3  ESTIMATION OF ECONOMIC AND FINANCIAL
          IMPACTS	8-64
     8.4  REFERENCES	8-88


APPENDIX A   EVOLUTION OF THE BACKGROUND INFORMATION
             DOCUMENT	A-l
APPENDIX B   INDEX TO ENVIRONMENTAL IMPACT CONSIDERATIONS  . B-l
APPENDIX C   CALCULATION OF HAP VAPOR PROFILES
             FOR GASOLINE	C-l
APPENDIX D   BASELINE EMISSIONS ANALYSIS 	  D-l
                                11

-------
                         LIST OF FIGURES
                                                             Page
Figure 3-1     Gasoline Distribution Facilities -
               United States  	  3-2

Figure 4-1     Gasoline Tank Truck Loading Methods  	  4-3
Figure 4-2     A Simplified Example of Controls at
               Bulk Gasoline Terminals  	  4-5
Figure 4-3     A Simplified Schematic of a Typical
               Carbon Adsorption System 	  4-7
Figure 4-4     A Simplified Flow Diagram for a
               Typical Elevated Flare System  	  4-9
Figure 4-5     A Simplified Flow Diagram for a
               Typical Enclosed Flare System  	   4-11
Figure 4-6     A Typical Temperature Controlled Flare and
               Simplified Flow Diagram  	   4-13
Figure 4-7     A Simplified Diagram of a Refrigeration
               System	4-15
Figure 4-8     Tank Truck Vapor Collection Equipment
               For Bottom Loading Operations  	   4-29
Figure 4-9     Vapor Balance System at a Bulk Gasoline
               Plant	4-33
Figure 4-10    Vapor Balance System at a Service
               Station	4-36

Figure 8-1     The U.S. Gasoline System	8-3
Figure 8-2     SIC 5171 and 5172 Characteristics	8-18
Figure 8-3     Wholesale Gasoline Establishment
               Ownership and Sales Trends:  SIC 5171  .  . .   8-21
Figure 8-4     1987 Sales Per Establishment for SICs
               5171 and 5172	8-22
Figure 8-5     Refiner vs. Non-refiner Firm Ownership
               of Wholesaling Establishments  	   8-23
Figure 8-6     Trend in Bulk Station Throughput 	   8-32
Figure 8-7     Transportation of Petroleum Products
               1974-1989 (Relative proportion of total
               ton-miles shipped by various modes)  ....   8-33
Figure 8-8     Estimated Number of Retail Gasoline
               Outlets: 1982, 1985,  1990  	   8-43
Figure 8-9     Gasoline Distribution:   Factor and
               Product Flows  	   8-65
Figure 8-10    Hypothetical Bulk Terminal Services
               Other Inputs Market	8-70
                               iii

-------
                          LIST OF TABLES
Table 1-1


Table 3-1
Table 3-2
Table 3-3
Table 3-4

Table 3-5

Table 3-6

Table 3-7

Table 3-8
Table 3-9

Table 3-10

Table 3-11

Table 3-12


Table 4-1

Table 4-2

Table 4-3

Table 4-4



Table 5-1

Table 5-2

Table 5-3

Table 5-4

Table 5-5
                                               Page

ENVIRONMENTAL AND ECONOMIC IMPACTS
OF REGULATORY ALTERNATIVES 	   1-4

VAPOR PROFILE OF NORMAL GASOLINE  	   3-7
VAPOR PROFILES USED IN ANALYSIS	3-11
RVP BY STATE BY MONTH	3-14
SATURATION  (S) FACTORS FOR CALCULATING
GASOLINE LOADING LOSSES   	   3-18
UNCONTROLLED EMISSIONS FROM EXAMPLE PIPELINE
FACILITIES	3-22
STORAGE TANK EMISSION FACTORS FOR BULK
TERMINAL STORAGE TANKS 	   3-26
STORAGE TANK EMISSION FACTORS FOR
PIPELINE BREAKOUT STATION STORAGE TANKS   .  .   3-27
UNCONTROLLED EMISSIONS FROM BULK  TERMINALS  .   3-30
UNCONTROLLED EMISSIONS FROM AN EXAMPLE BULK
PLANT	3-33
UNCONTROLLED EMISSIONS FROM AN EXAMPLE
SERVICE STATION  	   3-35
1998 BASELINE PARAMETERS USED IN  EMISSIONS
ANALYSIS	3-42
1998 BASELINE EMISSIONS FROM GASOLINE
DISTRIBUTION SOURCES 	   3-45

SUMMARY OF  EMISSION TEST DATA FOR BULK
GASOLINE TERMINAL VAPOR PROCESSORS  	   4-16
TANK SEAL CONTROL EFFICIENCIES -  INTERNAL
FLOATING ROOF TANKS	4-26
TANK SEAL CONTROL EFFICIENCIES -  EXTERNAL
FLOATING ROOF TANKS	4-27
ESTIMATED CONTROL EFFECTIVENESS FOR
LEAK DETECTION AND REPAIR PROGRAMS FOR
VALVES AND  PUMPS	4-40

PIPELINE PUMPING STATION MODEL PLANT
PARAMETERS	5-3
PIPELINE BREAKOUT STATION MODEL PLANT
PARAMETERS	5-6
BULK GASOLINE TERMINAL MODEL PLANT
PARAMETERS	5-9
RAILCAR LOADING BULK GASOLINE TERMINAL
MODEL PLANT PARAMETERS  	   5-10
BULK GASOLINE PLANT MODEL PLANT PARAMETERS  .   5-12
                                iv

-------
                    LIST OF TABLES (Continued)
Table 5-6

Table 5-7
Table 5-8
Table 5-9
Table 5-10

Table 5-11

Table 5-12
Table 5-13

Table 5-14
Table 5-15
Table 5-16
Table 6-1

Table 6-2


Table 6-3


Table 6-4


Table 6-5

Table 6-6


Table 6-7



Table 7-1


Table 7-2


Table 7-3

Table 7-4

Table 7-5
                                               Page

CHARACTERIZATION OF NATIONWIDE TANK TRUCK
POPULATION	5-15
SERVICE STATION MODEL PLANT PARAMETERS  . .  .   5-18
MODEL PLANT POTENTIAL TOTAL HAP EMISSIONS   .   5-20
MODEL PLANT MAXIMUM INDIVIDUAL HAP EMISSIONS   5-22
NEW PIPELINE FACILITIES REGULATORY
ALTERNATIVES 	   5-24
EXISTING PIPELINE FACILITIES REGULATORY
ALTERNATIVES 	   5-25
NEW BULK TERMINAL REGULATORY ALTERNATIVES   .   5-26
EXISTING BULK TERMINAL REGULATORY
ALTERNATIVES 	   5-27
NEW BULK PLANT REGULATORY ALTERNATIVES  . .  .   5-28
EXISTING BULK PLANT REGULATORY ALTERNATIVES   5-29
NEW AND EXISTING SERVICE STATION
REGULATORY ALTERNATIVES  	   5-30

SUMMARY OF HAP VAPOR PROFILES USED IN
ANALYSIS	6-4
SUMMARY OF REGULATORY ALTERNATIVE EMISSION
REDUCTIONS FOR EXISTING FACILITIES IN THE
BASE YEAR (1998)	6-5
SUMMARY OF REGULATORY ALTERNATIVE EMISSION
REDUCTIONS FOR NEW FACILITIES IN THE
BASE YEAR (1998)	6-7
SUMMARY OF REGULATORY ALTERNATIVE EMISSION
REDUCTIONS FOR ALL FACILITIES IN THE BASE
YEAR (1998)  	6-9
ESTIMATED SOLID WASTE IMPACTS FROM CARBON
DISPOSAL AT BULK GASOLINE TERMINALS  ....   6-21
ESTIMATED NET ENERGY SAVINGS  (GASOLINE  SAVED)
FROM GASOLINE DISTRIBUTION CONTROL
ALTERNATIVES 	   6-22
ESTIMATED PARTICULATE, NOX/  AND SOX
EMISSIONS FROM INCINERATION AT BULK GASOLINE
TERMINALS	6-25

COSTS OF INSTALLING A BOLTED INTERNAL FLOATING
ROOF ON AN EXISTING FIXED-ROOF TANK
(THIRD QUARTER 1990 DOLLARS)  	  7-3
COSTS OF INSTALLING A SECONDARY SEAL ON AN
EXISTING EXTERNAL FLOATING ROOF TANK
(THIRD QUARTER 1990 DOLLARS)  	  7-4
ANNUAL COST FOR QUARTERLY LEAK DETECTION
AND REPAIR	7-6
ANNUAL COST FOR MONTHLY LEAK DETECTION
AND REPAIR	7-7
SUMMARY OF LEAK DETECTION AND REPAIR NET
COSTS PER MONITORING EVENT	7-8

-------
                    LIST OF TABLES (Continued)
Table 7-6


Table 7-7


Table 7-8



Table 7-9


Table 7-10



Table 7-11


Table 7-12


Table 7-13



Table 7-14


Table 7-15


Table 7-16

Table 7-17

Table 7-18


Table 7-19


Table 7-20

Table 7-21

Table 7-22

Table 7-23
BULK TERMINAL LOADING RACK COSTS - NEW 35
mg/1 UNIT (THOUSANDS OF THIRD QUARTER 1990
DOLLARS)	7-11
BULK TERMINAL LOADING RACK COSTS - NEW 10
mg/1 UNIT (THOUSANDS OF THIRD QUARTER 1990
DOLLARS)	7-12
BULK TERMINAL LOADING RACK COSTS -
UNCONTROLLED TO 35 mg/1 UNIT WITH LOADING
RACK CONVERSION (THOUSANDS OF THIRD QUARTER
1990 DOLLARS)  	7-13
BULK TERMINAL LOADING RACK COSTS -
UPGRADE OF 80 mg/1 TO 35 mg/1 UNIT
(THOUSANDS OF THIRD QUAixTEx 1990 DOLLARS)   .   7-14
BULK TERMINAL LOADING RACK COSTS -
UNCONTROLLED TO 10 mg/1 UNIT WITH RACK
CONVERSION (THOUSANDS OF THIRD QUARTER 1990
DOLLARS)	7-15
BULK TERMINAL LOADING RACK COSTS -
UPGRADE OF 80 mg/1 TO 10 mg/1 UNIT
(THOUSANDS OF THIRD QUARTER 1990 DOLLARS)   .   7-16
BULK TERMINAL LOADING RACK COSTS -
UPGRADE OF 35 mg/1 TO 10 mg/1 UNIT
(THOUSAND? OF THIRD QUARTER 1990 DOLLARS)   .   7-17
BULK TERMINAL LOADING RACK COSTS -
UPGRADE OF 35 rag/1 TO 5 mg/1 UNIT OR NEW
5 mg/1 UNIT  (THOUSANDS OF THIRD QUARTER
1990 DOLLARS)  	7-18
BULK TERMINAL LOADING RACK COSTS -
UPGRADE OF 10 mg/1 TO 5 mg/1 UNIT
(THOUSANDS OF THIRD QUARTER 1990 DOLLARS)   .   7-19
BULK TERMINAL LOADING RACK COSTS - THERMAL
OXIDIZER ADD-ON (THOUSANDS OF THIRD QUARTER
1990 DOLLARS)  	7-20
RAILCAR VAPOR CONTROL COSTS FOR 10 mg/1
(THIRD QUARTER 1990 DOLLARS)  	   7-25
AVERAGE CONTROL COSTS FOR BULK PLANTS
(NO EXEMPTIONS) (THIRD QUARTER 1990 DOLLARS)   7-28
ESTIMATED CONTROL COSTS FOR BULK PLANTS
(EXEMPT <4,000 GAL/DAY)  (THIRD QUARTER 1990
DOLLARS)	7-29
SERVICE STATION STAGE I CAPITAL AND
ANNUALIZED COST ESTIMATES  (THIRD QUARTER
1990 DOLLARS)  	7-31
EXISTING PIPELINE FACILITIES NATIONWIDE
CONTROL LEVEL COSTS  	   7-32
NEW PIPELINE FACILITIES NATIONWIDE CONTROL
LEVEL COSTS	7-33
EXISTING BULK TERMINALS NATIONWIDE CONTROL
LEVEL COSTS	7-34
NEW BULK TERMINALS NATIONWIDE CONTROL
LEVEL COSTS	7-36
                                VI

-------
                    LIST OF TABLES (Continued)
Table 7-24

Table 7-25

Table 7-26

Table 7-27

Table 7-28



Table 8-1


Table 8-2

Table 8-3

Table 8-4

Table 8-5



Table 8-6

Table 8-7


Table 8-8

Table 8-9

Table 8-10

Table 8-11

Table 8-12

Table 8-13

Table 8-14

Table 8-15

Table 8-16

Table 8-17
                                               Page

EXISTING BULK PLANTS NATIONWIDE CONTROL
LEVEL COSTS	7-38
NEW BULK PLANTS NATIONWIDE CONTROL
LEVEL COSTS	7-39
EXISTING AND NEW SERVICE STATIONS NATIONWIDE
CONTROL LEVEL COSTS  	   7-40
SUMMARY OF ALTERNATIVE IMPACTS (5 mg/1 FOR
NEW TERMINALS)	7-49
SUMMARY OF ALTERNATIVE IMPACTS (10 mg/1  FOR
NEW TERMINALS IN ALTERNATIVES IV, IV-Q,
AND IV-M	7-50

TRENDS IN GASOLINE MARKETING:  U.S. GASOLINE
PRODUCTION AND DISPOSITION (IN MILLIONS  OF
LITERS)  	8-2
CONSUMPTION OF GASOLINE:  1982, 1987, 1989
(IN THOUSANDS OF LITERS)	8-7
TRENDS IN RETAIL MOTOR GASOLINE PRICES
(IN CENTS PER GALLON, INCLUDING TAXES)  ...   8-8
ESTIMATES OF MARGINS AT VARIOUS POINTS IN
THE GASOLINE DISTRIBUTION CHAIN   	   8-10
ESTIMATED NUMBER OF PEOPLE EMPLOYED BY
PMAA-MEMBER INDEPENDENT PETROLEUM MARKETERS,
BY EMPLOYMENT TYPE AND JOB CATEGORY:
1985, 1987, AND 1989	8-14
CONCENTRATION RATIOS FOR GASOLINE SALES
(PERCENTAGE OF U.S. TOTAL)	8-16
GENERAL CENSUS DATA CHARACTERIZING THE
WHOLESALE MARKET FOR GASOLINE:  1987
(SALES IN MILLIONS OF DOLLARS) 	   8-19
CONCENTRATION BY LARGEST FIRMS:  1987
(SICS 5171 AND 5172)	8-25
TRENDS IN CONCENTRATION BY LARGEST FIRMS:
1977-1987 (SIC 5171)	8-27
TRENDS IN FINANCIAL PROFITABILITY RATIOS:
1987, 1989, 1990 (SICs 5171 AND 5172)   .  .  .   8-28
ESTIMATES OF THE TOTAL NUMBER OF WHOLESALE
GASOLINE STORAGE FACILITIES:  1977-1990   .  .   8-30
FACILITY OWNERSHIP:  TERMINALS (PERCENTAGE
OF TOTAL)	8-31
LIQUID/GAS TANK TRUCK CHARACTERISTICS
IN 1982 AND 1987	8-35
RANKINGS OF MAJOR PETROLEUM PRODUCTS
PIPELINE COMPANIES 	   8-37
TRENDS IN CENSUS BUREAU-DEFINED SERVICE
STATIONS	8-38
REGIONAL AND NATIONAL MARKET SHARES BY
RETAIL OUTLET TYPE:  1987 AND 1989	8-40
TRENDS IN CONVENIENCE STORE GASOLINE
RETAILING	8-41
                               vii

-------
                    LIST OF TABLES  (Continued)
Table 8-18     CONCENTRATION BY LARGEST FIRMS: 1982-1987,
               SIC 5541-PUBLIC SERVICE STATIONS 	   8-46
Table 8-19     TRENDS IN FINANCIAL PROFITABILITY RATIOS:
               1987, 1989, 1990 (SIC 5541-GASOLINE SERVICE
               STATIONS)  	8-47
Table 8-20     1988 THROUGHPUT LEVELS AND PRICING MARGINS  .   8-50
Table 8-21     ESTIMATED DISTRIBUTION OF MODEL PLANTS FOR
               MOTOR GASOLINE BULK PLANTS, 1998	8-52
Table 8-22     ESTIMATED DISTRIBUTION OF MODEL PLANTS FOR
               BULK TERMINALS, 1998	8-53
Table 8-23     ESTIMATED DISTRIBUTION OF MODEL PLANTS FOR
               MOTOR GASOLINE SERVICE STATIONS, 1998  ...   8-54
Table 8-24     1998 FOR-HIRE GASOLINE TRUCKING FIRM
               CHARACTERISTICS  	   8-58
Table 8-25     MODEL PLANT THROUGHPUT BY FACILITY TYPE   .  .   8-59
Table 8-26     NUMBER OF FACILITIES  BY MODEL PLANT   ....   8-61
Table 8-27     ESTIMATED NUMBER OF NEW CAPACITY,
               REPLACEMENT CAPACITY, AND EXISTING
               FACILITIES	8-63
Table 8-28     ESTIMATED BASELINE YEAR PRICES
               AND QUANTITIES	8-67
Table 8-29     REGULATORY ALTERNATIVES IV, IV-Q, AND IV-M:
               MARGINAL FACILITY CHARACTERISTICS	8-73
Table 8-30     ESTIMATED EFFECTS OF  REGULATORY ALTERNATIVES
               IV, IV-Q, AND IV-M ON THE GASOLINE MARKETING
               INDUSTRY (QUANTITIES  IN BILLIONS OF LITERS;
               PRICES IN DOLLARS PER LITER)	8-74
Table 8-3OA    ESTIMATED EFFECTS OF  REGULATORY ALTERNATIVES
               IV, IV-Q, AND IV-M ON THE GASOLINE MARKETING
               INDUSTRY (QUANTITIES  IN BILLIONS OF GALLONS;
               PRICES IN DOLLARS PER GALLON)	8-75
Table 8-31     ESTIMATED EMPLOYMENT  IMPACTS  	   8-77
Table 8-32     ESTIMATED FIRM IMPACTS	8-79
Table 8-33     ESTIMATED CHANGES IN  ECONOMIC WELFARE
               ($106 1990  DOLLARS)    	8-83
Table 8-34     SBA DEFINITIONS OF SMALL BUSINESS AND
               CONCORDANCE WITH FIRM SIZE CATEGORIES FOR
               RELEVANT SECTORS OF THE GASOLINE DISTRIBUTION
               INDUSTRY	8-85

Table A-l      EVOLUTION OF THE BACKGROUND INFORMATION
               DOCUMENT	A-2

Table B-l      CROSS-INDEXED REFERENCE SYSTEM TO HIGHLIGHT
               ENVIRONMENTAL IMPACT  PORTIONS OF THE
               DOCUMENT	B-2

Table C-l      SUMMARY OF SOURCES OF DATA RECEIVED REGARDING
               GASOLINE COMPOSITION  	 C-3
Table C-2      EXAMPLE OF VAPOR COMPOSITION  CALCULATIONS FROM
               LIQUID DATA	C-6

                               viii

-------
                    LIST OF TABLES (Concluded)
Table C-3
Table C-4
Table C-5
Table

Table

Table

Table

Table

Table

Table
Table
Table
Table
Table
Table

Table
Table

Table

Table

Table

Table

Table
Table
D-l

D-2

D-3

D-4

D-5

D-6

D-7
D-8
D-9
D-10
D-ll
D-12

D-13
D-14

D-15

D-16

D-17

D-18

D-19
D-20
Table D-21


Table D-22

Table D-23

Table D-24
Table D-25
Table D-26

Table D-27
                                               Page

INDIVIDUAL SAMPLE HAP PROFILES  	  C-7
VAPOR PROFILE OF NORMAL GASOLINE  	   C-ll
VAPOR PROFILES USED IN ANALYSIS (HAP TO VOC
PERCENTAGE BY WEIGHT)  	   C-15

STATE REGULATORY COVERAGE FOR BULK GASOLINE
TERMINALS	D-5
STATE BULK TERMINAL THROUGHPUT BY LOADING
RACK CONTROL LEVEL	D-6
NATIONWIDE BULK TERMINAL LOADING RACK
BASELINE PARAMETERS BY CONTROL LEVEL	D-8
PIPELINE BREAKOUT STATION POPULATION BY STATE
SEPARATED BY STORAGE TANK CONTROL LEVEL   .  .   D-ll
STATE BULK TERMINAL THROUGHPUT BY STORAGE
TANK TYPE	D-14
BASELINE PARAMETERS FOR BULK TERMINAL
STORAGE TANKS  	   D-16
STATE REGULATORY COVERAGE FOR BULK PLANTS   .   D-17
BULK PLANT THROUGHPUT BY STATE  	   D-19
STATE BULK PLANT THROUGHOUT BY CONTROL LEVEL   D-21
BASELINE PARAMETERS FOR BULK PLANTS  ....   D-23
BASELINE PARAMETERS FOR TANK TRUCKS  ....   D-26
STATE SERVICE STATION THROUGHPUT BY
CONTROL LEVEL  	   D-28
BASELINE PARAMETERS FOR SERVICE STATIONS  .  .   D-30
STATE GASOLINE THROUGHPUT BY NONATTAINMENT
AREA CLASSIFICATION	D-33
HAP VAPOR PROFILES USED IN ANALYSIS AND
APPLICABILITY  	   D-36
BASELINE EMISSIONS FROM PIPELINE PUMPING
STATIONS	D-38
EMISSION FACTORS FOR PIPELINE BREAKOUT STATION
STORAGE TANKS  	   D-39
BASELINE EMISSIONS FROM PIPELINE BREAKOUT
STATIONS	D-40
BASELINE EMISSIONS FROM BULK TERMINALS .  .  .   D-42
BULK TERMINAL BASELINE LOADING RACK ANNUAL
THROUGHPUT BY AREA AND CONTROL LEVEL ....   D-43
BULK TERMINAL BASELINE STORAGE TANK
THROUGHPUT AND POPULATION BY AREA AND
CONTROL LEVEL  	   D-46
EMISSION FACTORS FOR BULK TERMINAL
STORAGE TANKS  	   D-48
BULK PLANT BASELINE ANNUAL THROUGHPUT BY
AREA AND CONTROL LEVEL	D-49
BULK PLANT EMISSION FACTORS  	   D-52
BASELINE EMISSIONS FROM BULK PLANTS  ....   D-53
SERVICE STATION BASELINE THROUGHPUT BY AREA
AND CONTROL LEVEL	D-54
BASELINE EMISSIONS FROM SERVICE STATIONS  .  .   D-56
                                IX

-------
                        1.0  SUMMARY
1.1  STATUTORY AUTHORITY
     National emission standards for hazardous air
pollutants (NESHAP) are established in accordance with
section 112(d) of the Clean Air Act, as amended in 1990.
Emission standards under section 112 apply to new and
existing sources of a substance that has been listed as a
hazardous air pollutant [section 112(b)].   This study
examines hazardous air pollutant (HAP) emission sources in
the gasoline distribution (Stage I) network of the petroleum
marketing source category which has been identified under
section 112(c) of the Act as presenting a threat of adverse
effects to human health or the environment.  The gasoline
distribution network consists of the following
subcategories, or facility types:
 Source Category
 Gasoline Distribution
   (Stage I)
Subcategorv
-Pipeline pumping stations
-Pipeline breakout stations
-Bulk terminals
-Bulk plants
-Service stations
1.2  REGULATORY ALTERNATIVES
     Six regulatory alternatives were developed by employing
various combinations of the available control techniques
utilized by facilities in the affected network.  Reflecting
increasing levels of emission reduction, these control
options range from requiring no new controls to imposing
very stringent standards at some facilities.  Chapter 5,
Section 5.2 provides a detailed discussion of these
alternatives.
                             1-1

-------
     In summary, Regulatory Alternative IV describes the
gasoline distribution network controlled under minimum
statutory requirements and represents a 4.6 percent
reduction from baseline emissions.  It provides for a leak
detection and repair (LDAR) program for equipment leaks at
new major source bulk terminals and pipeline breakout
stations.  Additionally, it provides for installation of
additional vapor control equipment (e.g., vapor processors
and primary and secondary storage tank seals) at all major
sources of these two facility types.   This alternative
provides the basis for incremental comparison of the other
regulatory alternatives.
     Regulatory Alternative IV-Q provides for an LDAR
program to be implemented at existing major source bulk
terminals and pipeline breakout stations.  These existing
major source sites would be monitored on a quarterly basis.
Implementation of this alternative would result in a 5.1
percent reduction in emissions from the baseline level.
     Implementation of Regulatory Alternative IV-M would
result in a 5.5 percent reduction in emissions by increasing
the frequency of leak detection and repair of equipment
components at existing major source bulk terminals and
pipeline breakout stations.  Monthly leak detection and
repair would be required for detection of equipment leaks at
these facilities.
     Regulatory Alternative III would increase the emission
reduction to 25 percent by requiring a quarterly LDAR
program for some sources and by requiring additional
equipment as well.  In addition to the controls required by
Alternative IV-Q, Regulatory Alternative III would require a
quarterly LDAR program for fugitive equipment leaks at area
source bulk terminals and pipeline breakout stations and
require additional equipment to be installed at these same
facilities as well.
     Implementation of Regulatory Alternative II would
improve control efficiency to 56 percent by requiring
controls at pipeline pumping stations, bulk plants, and
                             1-2

-------
service stations.  Installation of additional equipment
(e.g. vapor balance piping)  would be required at service
stations and bulk plants along with the implementation of a
quarterly LDAR program for equipment leaks at bulk plants
and pipeline pumping stations.
     Lastly, Regulatory Alternative I would effect a 57
percent control efficiency by requiring installation of
additional equipment at area source bulk terminals.
Installation of this equipment would be the only change from
controls specified in Alternative II.

1.3  ENVIRONMENTAL IMPACT
     Included in the evaluation of environmental impacts are
estimates of air quality, water, noise, and solid waste
impacts.  Table 1-1 summarizes the environmental impact
assessments for each regulatory alternative.
1.3.1  Air Quality Impact
     1.3.1.1  Existing Sources.  For the existing gasoline
distribution network (approximately 390,500 sources), the
total nationwide HAP emissions are estimated to be
approximately 45,800 megagrams per year (Mg/yr) at baseline.
Regulatory Alternative IV would reduce these emissions 4.4
percent to a total of 43,800 Mg/yr.  Alternative IV-Q would
reduce emissions by 5.0 percent, from 45,800 Mg/yr to 43,400
Mg/yr.  Alternative IV-M would reduce emissions to 43,300
Mg/yr, yielding a 5.5 percent reduction.  Alternative III
would yield a 27 percent reduction in HAP emissions to a
level of 33,400 Mg/yr.   Alternative II would reduce
emissions by 26,900 Mg/yr, to 18,900 Mg/yr (a 58.7 percent
reduction), and lastly, Alternative I would yield a 59
percent emission reduction to a total of 18,500 Mg/yr.
     1.3.1.2  New Sources.  For new sources through 1998,
total nationwide HAP emissions from gasoline distribution
facilities, approximately 13,100 total sources, are
estimated to be about 6,700 Mg/yr at baseline.  Regulatory
Alternative IV, IV-Q, or IV-M would reduce these emissions
to about 6,220 Mg/yr, a 6.6 percent reduction.  Alternative
                             1-3

-------
O 4-1
CO «J O
W
LTERNATIV
•^3

&
o

M
O
W
fa
o

W
EH
U
pi,
s
H

U
H
a
o
55
O
U
a
E-t
S
o
a
H
H

•
H
1
H

W
ij
PQ


Economic
Impact


Q) •*"*

•H *
2 B
H

>i4J
D> o
JH (0
C E
W H



jj
S-p
a °
•* (0
3*
o
W
•P Q
•* H


^
s&
B
H



0)
>
•H
4->
(0
f-i
Q)
rH

******
i-H rH H H CM CM
1 1 1 1 1 1




O O O O O O



******
******
rH i-H H CM n n




******
******
H rH H H H H
1 1 1 1 1 1

O O O O O O




******
******
i-H f>3 OJ O ^J* ^J*
+ + + + + +







a s H
H > > £ H H

QJ n
B (X
M S Q)
H rH
5 E-H
H 0> r<
1 H 0)
4-> 1 >
M O^ Q)
0 C M
x: q M
W M H

* * *
* *
*


.P
o
o< o
H 4-1 Q44->
0 B 0
0) (0 H (0
4-> rH O< CU
O ft B <1) E
«J .H H 4J H
fiji Ql (^
g 'H rH >4 Q)
H rH rH QJ O*
CP (0 TJ tl
O 
fl) T>
03 <
H- 1
I
1-4

-------
Ill would reduce emissions from 6,660 Mg/yr at baseline to
about 5,880 Mg/yr, an 11.8 percent reduction.   Alternative
II would reduce emissions to about 4,020 Mg/yr, a 40 percent
reduction.  Finally, Alternative I would reduce emissions by
about 2,780 Mg/yr to a total of 3,880 Mg/yr, a 42 percent
reduction through 1998.
1.3.2  Water. Solid Waste, and Energy Impacts for New and
       Existing Sources
     Since none of these alternatives would result in any
additional water discharges, there would be no negative
impact on water quality.  There is potential for a positive
benefit to water quality, however, due to decreased amounts
of organic materials entering drains, sewers,  and waste
water discharges because of better leak control.
     There would be no significant solid waste or noise
impact as a result of implementing any of the regulatory
alternatives.  Additionally since it is projected that many
additional facilities will use vapor recovery devices, there
will be energy benefits  (gasoline that would have evaporated
but is now recovered) gained from implementation of each of
the alternatives.  This benefit increases with the
stringency of the alternative because each successive
alternative requires additional control measures.

1.4  ECONOMIC IMPACT
     The impacts of the proposed standards were analyzed
(see Chapter 8) with regard to their effect on gasoline
price and consumption, facility closures, and employment.
While Alternatives IV, IV-Q, and IV-M require additional
controls only at bulk gasoline terminals and pipeline
breakout stations, facilities downstream from terminals and
breakout stations are affected by implementation of controls
due to higher gasoline wholesale prices and reduced enduse
demand, again due to higher prices.  The national average
base year increase in the price of retail motor gasoline as
a result of these alternatives is estimated at $0.001 per
gallon.  The national base year decline in gasoline
                             1-5

-------
consumption is estimated at less than 100 million gallons.
There is a limited number of facility closures projected to
result from the regulatory alternatives.  The base year
facility closure estimate is nearly 650, more than 90
percent of which are projected for the service station
sector.  While the number of service station closures is
estimated to be in the hundreds, it should be noted that a
total number of over 380,000 stations are projected in the
base year, so that the number of facilities closed
constitutes less than two tenths of one percent.
Furthermore, due to a consumption-spurred projection of
modest industry growth from 1993 to 1998, closures due to
implementation of controls may be more accurately
interpreted as reductions in new facility openings rather
than closures of existing facilities.  Employment reductions
due to reduced consumption and facility closure are
estimated at just over 1100 jobs, 70 percent of which are
estimated for the service station sector.  For the same
reasons given for facility closure, employment reductions
may be more accurately interpreted as reductions in industry
job opportunities rather than losses of existing jobs.
                             1-6

-------
                      2.0   INTRODUCTION

2.1  BACKGROUND AND AUTHORITY FOR STANDARDS
     According to industry estimates, more than 2.4 billion
pounds of toxic pollutants were emitted to the atmosphere in
1988 ("Implementation Strategy for the Clean Air Act
Amendments of 1990," EPA Office of Air and Radiation,
January 15, 1991).  These emissions may result in a variety
of adverse health effects, including cancer, reproductive
effects, birth defects, and respiratory illnesses.
Title III of the Clean Air Act Amendments (CAAA)  of 1990
provides the tools for controlling emissions of these
pollutants.  Emissions from both large and small facilities
that contribute to air toxics problems in urban and other
areas will be regulated.  The primary consideration in
establishing national emission standards must be
demonstrated technology.  Before NESHAP are proposed as
Federal regulations, air pollution prevention and control
methods are examined in detail with respect to their
feasibility, environmental impacts, and costs.  Various
control options based on different technologies and degrees
of efficiency are examined, and a determination is made
regarding whether the various control options apply to each
emission source or if dissimilarities exist among the
sources.  In most cases, regulatory alternatives are
subsequently developed that are then studied by EPA as a
prospective basis for a standard.  The alternatives are
investigated in terms of their impacts on the environment,
the economics and well-being of the industry, the national
economy, and energy and other impacts.  This document
summarizes the information obtained through these studies so
that interested persons will be able to evaluate the

                             2-1

-------
information considered by EPA in developing the proposed
standards.
     National emission standards for hazardous air
pollutants for new and existing sources are established
under section 112 of the Clean Air Act as amended in 1990
[42 U.S.C. 7401 et seq., as amended by PL 101-549,
November 15, 1990], hereinafter referred to as the Act.
Section 112 directs the EPA Administrator to promulgate
standards that "require the maximum degree of reduction in
emissions of the hazardous air pollutants subject to this
section (including a prohibition of ^nrh emissions, where
achievable) that the Administrator, taking into
consideration the cost of achieving such emission
reductions, and any non-air quality health and environmental
impacts and energy requirements, determines is achievable
...  ."  The Act allows the Administrator to set standards
that "distinguish among classes, types, and sizes of sources
within a category or subcategory."
     The Act differentiates between major sources and area
sources.  A major source is defined as "any stationary
source or group of stationary sources located within a
contiguous area and under common control that emits or has
the potential to emit considering controls, in the
aggregate, 10 tons per year or more of any hazardous air
pollutant or 25 tons per year or more of any combination of
hazardous air pollutants."  The Administrator, however, may
establish a lesser quantity cutoff to distinguish between
major and area sources.  The level of the cutoff is based on
the potency, persistence, or other characteristics or
factors of the air pollutant.  An area source is defined as
"any stationary source of hazardous air pollutants that is
not a major source."  For new sources, the amendments state
that the "maximum degree of reduction in emissions that is
deemed achievable for new sources in a category or
subcategory shall not be less stringent than the emission
control that is achieved in practice by the best controlled
similar source, as determined by the Administrator."
                             2-2

-------
Emission standards for existing sources "may be less
stringent than the standards for new sources in the same
category or subcategory but shall not be less stringent, and
may be more stringent than — (A) the average emission
limitation achieved by the best performing 12 percent of the
existing sources (for which the Administrator has emissions
information), excluding those sources that have, within
18 months before the emission standard is proposed or within
30 months before such standard is promulgated, whichever is
later, first achieved a level of emission rate or emission
reduction which complies, or would comply if the source is
not subject to such standard, with the lowest achievable
emission rate (as defined by section 171) applicable to the
source category and prevailing at the time, in the category
or subcategory for categories and subcategories with 30 or
more sources, or (B) the average emission limitation
achieved by the best performing five sources (for which the
Administrator has or could reasonably obtain emissions
information) in the category or subcategory for categories
or subcategories with fewer than 30 sources."
     The Federal standards are also known as "MACT"
standards and are based on the maximum achievable control
technology previously discussed.  The MACT standards may
apply to both major and area sources, although the existing
source standards may be less stringent than the new source
standards, within the constraints presented above.  The MACT
is considered to be the basis for the standard, but the
Administrator may promulgate more stringent standards, which
may have several advantages.  First, they may help achieve
long-term cost savings by avoiding the need for more
expensive retrofitting to meet possible future residual risk
standards, which may be more stringent (discussed in Section
2.6).  Second, Congress was clearly interested in providing
incentives for improving technology.  Finally, in the CAAA
of 1990, Congress gave EPA a clear mandate to reduce the
health and environmental risk of air toxics emissions as
quickly as possible.
                             2-3

-------
     For area sources, the Administrator may "elect to
promulgate standards or requirements applicable to sources
in such categories or subcategories which provide for the
use of generally available control technologies or
management practices by such sources to reduce emissions of
hazardous air pollutants."  These area source standards are
also known as "GACT"  (generally available control
technology) standards, although MACT may be applied at the
Administrator's discretion, as discussed previously.
The standards for hazardous air pollutants (HAPs), like the
new source performance standards (NSPS) for criteria
pollutants required by Section 111 of the Act (42 U.S.C.
7411), differ from other regulatory programs required by the
Act  (such as the new source review program and the
prevention of significant deterioration program) in that
NESHAP and NSPS are national in scope  (versus site-
specific) .  Congress intended for the NESHAP and NSPS
programs to provide a degree of uniformity to State
regulations to avoid situations where some States may
attract industries by relaxing standards relative to other
States.  States are free under section 116 of the Act to
establish standards more stringent than section 111 or 112
national standards.
     Although NESHAP are normally structured in terms of
numerical emission limits, alternative approaches are
sometimes necessary.  In some cases, physically measuring
emissions from a source may be impossible or at least
impracticable due to technological and economic limitations.
Section 112(h) of the Act allows the Administrator to
promulgate a design, equipment, work practice, or
operational standard, or combination thereof, in those cases
where it is not feasible to prescribe or enforce an
emissions standard.  For example, emissions of volatile
organic compounds  (many of which may be HAPs, such as
benzene) from storage vessels for volatile organic liquids
are greatest during tank filling.  The nature of the
emissions  (i.e., high concentrations for short periods
                             2-4

-------
during filling and low concentrations for longer periods
during storage) and the configuration of storage tanks make
direct emission measurement impractical.  Therefore, the
MACT or GACT standards may be based on equipment
specifications.  Under section 112(h)(3), the Act also
allows the use of alternative equivalent technological
systems: "If, after notice and opportunity for comment, the
owner or operator of any source establishes to the
satisfaction of the Administrator that an alternative means
of emission limitation" will reduce emissions of any air
pollutant at least as much as would be achieved under the
design, equipment, work practice, or operational standard,
the Administrator shall permit the use of the alternative
means.
       Efforts to achieve early environmental benefits are
encouraged in Title III.  For example, source owners and
operators are encouraged to use the section 112(i)(5)
provisions, which allow a 6-year compliance extension of the
MACT standard in exchange for the implementation of an early
emission reduction program.  The owner or operator of an
existing source must demonstrate a 90 percent emission
reduction of HAPs (or 95 percent if the HAPs are
particulates) and meet an alternative emission limitation,
established by permit, in lieu of the otherwise applicable
MACT standard.  This alternative limitation must reflect the
90 (95) percent reduction and is in effect for a period of
6 years from the compliance date for the otherwise
applicable standard.  The 90 (95) percent early emission
reduction must be achieved before the otherwise applicable
standard is first proposed, although the reduction may be
achieved after the standard's proposal (but before
January 1, 1994) if the source owner or operator makes an
enforceable commitment before the proposal of the standard
to achieve the reduction.  The source must meet several
criteria to qualify for the early reduction standard, and
section 112(i)(5)(A) provides that the State may require
additional reductions.
                             2-5

-------
2.2  SELECTION OF POLLUTANTS AND SOURCE CATEGORIES
     As amended in 1990,  the Act includes a list of
189 HAPs.  Petitions to add or delete pollutants from this
list may be submitted to EPA.  Using this list of
pollutants, EPA is to publish a list of source categories
(major and area sources)  for which emission standards will
be developed.  Within 2 years of enactment (November 1992) ,
EPA is to publish a schedule establishing dates for
promulgating these standards.  Petitions may also be
submitted to EPA to remove source categories from the list.
The schedule for standards for source categories will be
determined according to the following criteria:
     "(A) the known or anticipated adverse effects of such
pollutants on public health and the environment;
       (B) the quantity and location of emissions or
reasonably anticipated emissions of hazardous air pollutants
that each category or subcategory will emit; and
       (C) the efficiency of grouping categories or
subcategories according to the pollutants emitted, or the
processes or technologies used."
     After the source category has been chosen, the types of
facilities within the source category to which the standard
will apply must be determined.  A source category may have
several facilities that cause air pollution, and emissions
from these facilities may vary in magnitude and control
cost.  Economic studies of the source category and
applicable control technology may show that air pollution
control is better served by applying standards to the more
severe pollution sources.  For this reason, and because
there  is no adequately demonstrated system for controlling
emissions from certain facilities, standards often do not
apply to all facilities at a source.  For the same reasons,
the standards may not apply to all air pollutants emitted.
Thus, although a source category may be selected to be
covered by standards, the standards may not cover all
pollutants or facilities within that source category.
                             2-6

-------
2.3  PROCEDURE FOR DEVELOPMENT OF NESHAP
     Standards for major and area sources must (1)
realistically reflect MACT or GACT; (2) adequately consider
the cost, the non-air quality health and environmental
impacts, and the energy requirements of such control;
(3) apply to new and existing sources; and (4) meet these
conditions for all variations of industry operating
conditions anywhere in the country.
     The objective of the NESHAP program is to develop
standards to protect the public health by requiring
facilities to control emissions to the level achievable
according to the MACT or GACT guidelines.  The standard-
setting process involves three principal phases of activity:
(1) gathering information, (2) analyzing the information,
and (3) developing the standards.
     During the information-gathering phase,  industries are
questioned through telephone surveys,  letters of inquiry,
and plant visits by EPA representatives.  Information is
alsro gathered from other sources, such as a literature
search.  Based on the information acquired about the
industry, EPA selects certain plants at which emissions
tests are conducted to provide reliable data that
characterize the HAP emissions from well-controlled existing
facilities.
     In the second phase of a project, the information about
the industry, the pollutants emitted,  and the control
options are used in analytical studies.  Hypothetical "model
plants" are defined to provide a common basis for analysis.
The model plant definitions,  national pollutant emissions
data,  and existing State regulations governing emissions
from the source category are then used to establish
"regulatory alternatives."  These regulatory alternatives
may be different levels of emissions control or different
degrees of applicability, or both.
     The EPA conducts studies to determine the cost,
economic, environmental, and energy impacts of each
regulatory alternative.  From several alternatives, EPA
                             2-7

-------
selects the single most plausible regulatory alternative as
the basis for the NESHAP for the source category under
study.
     In the third phase of a project, the selected
regulatory alternative is translated into standards which,
in turn, are written in the form of a Federal regulation.
The Federal regulation limits emissions to the levels
indicated in the selected regulatory alternative.
     As early as is practical in each standard-setting
project, EPA representatives discuss the possibilities of a
standard and the form it might take with members of the
National Air Pollution Control Techniques Advisory
Committee, which is composed of representatives from
industry, environmental groups, and State and local air
pollution control agencies.  Other interested parties also
participate in these meetings.
     The information acquired in the project is summarized
in the background information document (BID).  The draft
BID, proposed standards, and a preamble explaining the
standards are widely circulated to the industry being
considered for control, environmental groups, other
government agencies, and offices within EPA.  Through this
extensive review process, the points of view of expert
reviewers are taken into consideration as changes are made
to the documentation.  A "proposal package" is assembled and
sent through the offices of EPA Assistant Administrators for
concurrence before the proposed standards are officially
endorsed by the EPA Administrator.  After being approved by
the EPA Administrator, the preamble and the proposed
regulation are published in the Federal Register.
     The public is invited to participate in the standard-
setting process as part of the Federal Register announcement
of the proposed regulation.  The EPA invites written
comments on the proposal and may also hold a public hearing
to discuss the proposed standards with interested parties.
All public comments are summarized and incorporated into a
                             2-8

-------
second volume of the BID.  All information reviewed and
generated in studies in support of the standards is
available to the public in a "docket" on file in Washington,
D.C.  Comments from the public are evaluated, and the
standards may be altered in response to the comments.
     The significant comments and EPA's position on the
issues raised are included in the preamble of a promulgation
package, which also contains the draft of the final
regulation.  The regulation is then subjected to another
round of internal EPA review and refinement until it is
approved by the EPA Administrator.  After the Administrator
signs the regulation, it is published as a "final rule" in
the Federal Register.

2.4  CONSIDERATION OF COSTS
     The prime objective of the cost analysis is to identify
the incremental economic impacts associated with compliance
with the standards based on each regulatory alternative
compared to baseline.  Other environmental regulatory costs
may be factored into the analysis wherever appropriate.  Air
pollutant emissions may cause water pollution problems, and
captured potential air pollutants may pose a solid waste
disposal problem.  The total environmental impact of an
emission source must, therefore, be analyzed and the costs
determined whenever possible.
     A thorough study of the profitability and price-setting
mechanisms of the industry is essential to the analysis so
that an accurate estimate of potential adverse economic
impacts can be made for proposed standards.  It is also
essential to know the capital requirements for pollution
control systems already placed on plants so that the
additional capital requirements necessitated by these
Federal standards can be placed in proper perspective.
Finally, it is necessary to assess the availability of
capital to provide the additional control equipment needed
to meet the standards.
                            2-9

-------
2.5  CONSIDERATION OF ENVIRONMENTAL IMPACTS
     Section 102(2)(C) of the National Environmental Policy
Act (NEPA) of 1969 requires Federal agencies to prepare
detailed environmental impact statements on proposals for
legislation and other major Federal actions significantly
affecting the quality of the human environment.  The
objective of the NEPA is to build into the decision-making
process of Federal agencies a careful consideration of all
environmental aspects of proposed actions.
     In a number of legal challenges to standards for
various industries, the United States Court of Appeals for
the District of Columbia Circuit has held that environmental
impact statements need not be prepared by EPA for proposed
actions under the Clean Air Act.  Essentially, the Court of
Appeals has determined that the best system of emissions
reduction requires the Administrator to take into account
counterproductive environmental effects of proposed
standards as well as economic costs to the industry.  On
this basis, therefore, the Courts established a narrow
exemption from the NEPA for EPA determinations.
     In addition to these judicial determinations, the
Energy Supply and Environmental Coordination Act  (ESECA) of
1974 (PL-93-319) specifically exempted proposed actions
under the Clean Air Act from NEPA requirements.  According
to Section 7(c)(l), "No action taken under the Clean Air Act
shall be deemed a major Federal action significantly
affecting the quality of the human environment within the
meaning of the National Environmental Policy Act of 1969"
(15 U.S.C. 793(c)(1)).
     Nevertheless, EPA has concluded that preparing
environmental impact statements could have beneficial
effects on certain regulatory actions.  Consequently,
although not legally required to do so by Section 102(2)(C)
of the NEPA, EPA has adopted a policy requiring that
environmental impact statements be prepared for various
regulatory actions, including NESHAP developed under
section 112 of the Act.  This voluntary preparation of
                            2-10

-------
environmental impact statements, however, in no way legally
subjects EPA to NEPA requirements.
     To implement this policy, a separate section included
in this document is devoted solely to an analysis of the
potential environmental impacts associated with the proposed
standards.  Both adverse and beneficial impacts in such
areas as air and water pollution, increased solid waste
disposal, and increased energy consumption are discussed.

2.6  RESIDUAL RISK STANDARDS
     Section 112 of the Act provides that 8 years after MACT
standards are established (except for those standards
established 2 years after enactment, which have 9 years),
standards to protect against the residual health and
environmental risks remaining must be promulgated, if
necessary.  The standards would be triggered if more than
one source in a category or subcategory exceeds a maximum
individual risk of cancer of 1 in 1 million.  These residual
risk regulations would be based on the concept of providing
an "ample margin of safety to protect public health."  The
Administrator may also consider whether a more stringent
standard is necessary to prevent—considering costs, energy,
safety, and other relevant factors—an adverse environmental
effect.  In the case of area sources controlled under GACT
standards, the Administrator is not required to conduct a
residual risk review.
                            2-11

-------
           3.0  PROCESSES AND POLLUTANT EMISSIONS

3.1  GENERAL
     The gasoline distribution network consists of 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
railcars, storage tanks, and service stations.  Crude
petroleum is shipped to refineries, which manufacture a wide
range of liquid petroleum products.  Finished gasoline is
then distributed in a complex system comprised of wholesale
and retail outlets.  The focus of this document is to assess
the impacts of distributing gasoline from gasoline storage
and loading operations at refineries to the loading of
storage tanks at gasoline dispensing facilities.  Other
sources, such as those associated with the production of
gasoline, vehicle refueling at service stations, and ship
and barge loading, are or will be covered in separate
documents.  The main elements in the distribution network
are depicted in Figure 3-1.
     Gasoline is delivered to bulk terminals from refineries
by way of pipeline, ship, or barge.  Large transport trucks
(30,000 to 38,000 liter or 8,000 to 10,000 gallon capacity)
deliver the gasoline to service stations or to intermediate
bulk storage facilities known as bulk plants.  The situation
also exists where gasoline is loaded into a railcar at one
terminal and transported to another terminal that does not
have access to a pipeline, or a waterway that could support
a ship or barge.
                             3-1

-------
                                                Imported or
                                              Domestic Crude
Imported Gasoline
                             Barge/
                            Pipeline/
                            Tanker
                          Bulk Terminal
                                            Bulk Plant
     Service Station
                          Commercial, Rural
                               Accounts
Consumer
Figure  3-1.  Gasoline  Distribution Facilities - United  States

                               3-2

-------
     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 dispensing facilities include the
familiar service stations, as well as commercial accounts
such as fleet services  (rental car agencies, private
companies, governmental agencies), parking garages, and
buses.  Another important consumer category consists of
small farms (approximately 2.7 million).
     This chapter discusses the sources of emissions at each
segment of the gasoline distribution chain, including
pipeline pumping stations, pipeline breakout stations, bulk
terminals, bulk plants, and service stations.  Section 3.2
discusses the factors influencing emissions, emission
factors, and volatile organic compound  (VOC) and HAP
emissions for typical facilities.  Section 3.3 then presents
the national 1998 baseline emissions for all industry
sectors.

3.2  FACILITIES AND THEIR EMISSIONS
     The pollutants emitted by each of the gasoline
distribution facilities are essentially the same.  However,
the operations that occur at each and the rates of emissions
to the atmosphere differ.  The emissions consist of a
mixture of VOC vapors and air.  The factors influencing
emissions, including gasoline composition, temperature,
                             3-3

-------
vapor pressure, and methods of loading gasoline are
discussed in Section 3.2.1.  Sections 3.2.2 through 3.2.5
present separate discussions of the operations at each
industry sector and of the associated emission rates.
3.2.1  Factors Influencing Emissions
     3.2.1.1  Hazardous Air Pollutant Content of Gasoline
Vapor.  As discussed in Section 2.2, the 1990 Clean Air Act
Amendments contain a list of 189 HAPs.  A comparison of
profiles of gasoline vapor with this HAP list reveals
several compounds common to both.  This section discusses
the HAPs found in traditional, or "normal", gasoline vapor
and how this is expected to change in response to
requirements contained in Title II of the Amendments.  This
section also presents vapor profiles that will be used in
evaluating HAP emissions from gasoline distribution sources
throughout this analysis.
     Motor gasoline is a complex organic mixture of varying
amounts of paraffins, olefins, and aromatics.  A study
conducted for the EPA which analyzed gasoline samples in the
northeastern United states in the early 1980's (Northeast
Corridor Study) reported liquid gasoline paraffin contents
ranging from 37-67 weight percent, olefins ranging from 0-12
weight percent, and aromatics ranging from 28-52 percent.
The average carbon number  for gasoline generally falls in
the C5-C7 range,  but gasoline  composition  can vary  widely.
     The National Institute for Petroleum and Energy
Research  (NIPER) reports gasoline composition trends semi-
annually.  For the winter  of  1991-92,2 the reported aromatic
volume percentage for unleaded gasolines ranged from
approximately three percent to almost 65 percent in the
samples analyzed, with the averages being 25.9 percent for
regular unleaded, 27.9 percent for mid-grade, and 30.3
percent  for premium.  Olefin  content ranged  from under one
to almost 69 percent, with the averages reported as  11.6
percent  for regular, 9.8 percent  for mid-grade, and  6.1
percent  for premium.
                             3-4

-------
     This variation in liquid composition causes the vapor
composition to vary a great deal.   The Northeast Corridor
Study indicated that paraffins made up from 76 to 98 percent
by weight of the vapors, 0 to 22 weight percent for olefins,
and 0.8 to 3.2 weight percent for aromatics.  The small
percentage of aromatics is due to the low volatility of
these compounds.  Conversely, the vapor profiles showed a
high percentage of paraffins due to the high volatility of
C4 and C5 paraffins.
     3.2.1.1.1  Normal gasoline.  In order to estimate HAP
emissions from sources in the gasoline distribution chain,
an investigation was conducted to identify and quantify the
HAPs in gasoline vapor.  A search was initiated to obtain
relevant data regarding gasoline vapor phase composition
during gasoline storage and transfer operations.  This
effort revealed that while a great deal of research was
being conducted related to the composition of tailpipe
emissions from automobiles, information related to the
composition of evaporative emissions from gasoline transfer
and storage operations was more limited.
     However, sufficient data were received to establish a
list of HAP compounds commonly present in gasoline vapor and
to provide an estimate of the quantity of these HAPs.  The
existence of benzene in gasoline vapors has been recognized
for a long time.  In addition, several other aromatic HAPs
were found in gasoline vapors.  These include toluene,
ethylbenzene, naphthalene, cumene, and all three
orientations of xylene  (para, meta, and ortho).
     As discussed above, gasoline vapors are made up largely
of paraffins.  Therefore, the existence of n-hexane is not
surprising.  Based on the data received, n-hexane is usually
the most prevalent HAP in gasoline vapor.  In addition,
2,2,4 trimethylpentane, or iso-octane, was found in gasoline
vapors.
     In order to quantify the HAP content of gasoline vapor,
the data were analyzed to determine the portion of the vapor
made up of HAPs.  For each vapor or liquid sample, the HAP
                             3-5

-------
weight percentage was calculated for individual as well as
total HAPs.
     The HAP contents were expressed as ratios by weight of
HAP to total VOC.  This was because VOC emissions from
gasoline distribution facilities have been studied and are
well documented, and HAP emissions from these sources could
be easily estimated by multiplying this HAP to VOC ratio by
the VOC mass emissions.
     The minimum, maximum, and arithmetic averages for the
HAP to VOC ratios calculated from the data are shown in
Table 3-1.  HAP emissions presented in this chapter and the
remainder of the document will be presented as total HAPs,
and not by individual HAP.  The arithmetic average ratio of
0.048 will be used throughout this document to represent the
total HAP to VOC ratio for normal gasoline.  A description
of the data and the analysis is contained in Appendix C
(Section C.I).
     3.2.1.1.2  Reformulated/oxygenated gasoline.  Title II
of the 1990 CAAA addresses emission standards for mobile
sources.  There are several elements in Title II that will
affect gasoline composition in the 1998 base year and, thus,
HAP emissions from gasoline storage and transfer operations.
     Section 219 of Title II amends the 1977 Clean Air Act
by adding Section 211.  Section 211(k) requires the
distribution of reformulated gasoline in those nine areas
having a 1980 population in excess of 250,000 and having the
highest ozone design values during the 1987-89 period.  All
other ozone nonattainment areas can "opt-in" to the program
regardless of 1980 population.  Beginning in 1995,
reformulated gasoline with the following limits must be sold
and marketed in these nonattainment areas:  l) benzene
content cannot exceed 1 percent; 2) no heavy metals can be
present; and 3) minimum oxygen content must be 2.0 percent.
Additionally, the more stringent of the Formula Standard
concerning aromatics  (level of 25 percent) or the
Performance Standards concerned with VOC or toxic emissions
                             3-6

-------
       TABLE 3-1.  VAPOR PROFILE OF NORMAL GASOLINE
                                  HAP TO VOC RATIO
                                (percentage by weight)
HAZARDOUS AIR POLLUTANT3
MINIMUM
ARITHMETIC
  AVERAGE
TOTAL HAPsb
                4.8
MAXIMUM
Hexane
Benzene
Toluene
2,2,4 Trimethylpentane
(iso-octane)
Xylenes
Ethylbenzene
0.3
0.2
0.4
0.03
0.05
0.03
1.6
0.9
1.3
0.8
0.5
0.1
4.4
2.2
4
2.6
1.5
0.5
              11
  Cumene and naphthalene were also identified in some of
  the data points in small quantities.  They are not shown
  as their addition does not significantly change the
  totals.

  The total HAP ratios shown in the table are not simply
  sums of the individual HAP percentages listed in the
  columns; rather, total HAPs were calculated for each
  individual sample in the data base. The values
  represented in the table reflect the maximum, minimum,
  and arithmetic average total HAPs of these samples.
                            3-7

-------
(15 percent reduction from emissions using a 1990 baseline
fuel)  shall also apply.  Concerning these final two
alternatives, it is most likely that in the future the
aromatic content of reformulated/oxygenated gasolines will
approach 25 percent.
     Also, section 211(m) requires the purchase and sale of
fuels with higher levels of alcohols or oxygenates in the
winter months in the areas exceeding the carbon monoxide
(CO) standard.  Beginning in 1992, these "oxygenated" fuels
must have at least 2.7 percent oxygen.
     The reformulated gasoline re^uij.eiu«nts will cause
reductions in the benzene and aromatic contents of the fuel
sold in those areas in the reformulated fuels program.
Since many of the HAPs in gasoline vapor are aromatic
compounds, this will reduca the total HAP content of the
gasoline liquid and vapors.  However, the addition of oxygen
containing compounds will cause a significant increase in
the HAP content.
     Methyl tert-butyl ether (MTBE), an oxygenate, is one of
several compounds that is expected to be added to gasoline
to increase its oxygen content.  Further, it has been
estimated and assumed in this report's analysis that MTBE
will make up at least 70 percent of the market of compounds
added to gasolines in the reformulated and oxygenated
programs in ozone nonattainment areas3.  MTBE is also listed
in the CAAA as a HAP.  Traditionally, MTBE has been used as
an octane booster in unleaded gasolines.  If the octane was
lower than expected, small allotments of MTBE would be added
to reach the desired octane level.  MTBE has many advantages
as an octane enhancer.   It has a high average blending
octane rating, dissolves easily in the refinery streams, and
will not precipitate out of solution when it comes into
contact with water.  Therefore, the quantity of normal
gasoline in the nation that contains some MTBE was large
prior to the implementation of section 211, although the
MTBE was present in only low percentages.  None of the data
                             3-8

-------
received for normal gasoline reported measurable levels of
MTBE.
     Other possible oxygenates are ethanol 113,  ethyl tert-
butyl ether (ETBE), and tertiary amyl methyl ether (TAME).
ETBE has a lower Reid Vapor Pressure (3-5 psi)  compared to
MTBE (8 psi), but its blending octane rating is higher.
However, there are limits on ETBE and the other blending
agents.  Ethanol 113 is not economical without government
subsidies and ETBE is similarly affected, as ethanol
feedstock is needed to produce ETBE.  Therefore, the amount
of ethanol and ETBE available will always be limited by
government subsidies.  The lack of isoamylene feedstock will
limit the use of TAME as well.  As a consequence, it is
expected that MTBE will be one of the most common oxygenates
used to meet the reformulated and oxygenated fuel oxygen
requirements.
     Widespread industry estimates indicate that it will
require approximately 15 volume percent of MTBE in liquid
gasoline to meet the 2.7 weight percent oxygen limit, and 11
volume percent to meet the 2.0 weight percent oxygen limit.
The moderate volatility of MTBE would cause high
concentrations in the vapor phase relative to the less
volatile aromatics.  In the search discussed above for
gasoline containing MTBE, vapor data and the corresponding
liquid composition were available for some samples.  Using
these samples, a relationship of liquid content of MTBE to
vapor content of MTBE was derived.  This MTBE ratio was
applied to the volume percents discussed to estimate the
MTBE to VOC percentage in the vapor.  Results of the
analysis showed that MTBE to VOC ratios were 8.8 weight
percent for the 11 volume percent liquid and 12 weight
percent for the 15 volume percent liquid.  A complete
discussion of this analysis is presented in Appendix C.
Consequently, it is expected that the inclusion of MTBE in
the liquid to meet the oxygen demands will increase the HAP
to VOC ratio in gasoline vapor from approximately 5 weight
                             3-9

-------
percent shown in Table 3-1 to near 16 percent (with the 15
percent MTBE gasoline).
     Because of these drastic differences in the HAP content
of gasoline vapor, the estimation of vapor phase composition
(HAP to VOC ratios) for several different fuel types was
considered necessary.  There will be four basic types of
fuels in use after full implementation of these programs.
These are 1) normal fuels (to be used in attainment areas
and those ozone nonattainment areas not opting into the
reformulated program), 2) oxygenated fuels (to be used in CO
nonattainment areas during the winter months), 3)
reformulated fuels (to be used in ozone nonattainment areas
in the reformulated program year round), and 4) reformulated
fuels with 2.7 percent oxygen, or reformulated/oxygenated
fuels  (to be used in  areas that are nonattainment for both
CO and ozone and require the reformulated fuels year round
and require oxygenated fuels in the winter months).
     Therefore, HAP to VOC ratios were developed for each of
these  fuel types.  The situation is further complicated
because two different ratios are required for the types
containing oxygenates (reformulated, oxygenated, and
reformulated/oxygenated) to account for MTBE.  One ratio
includes MTBE and the other uses one of the other, non-HAP
oxygenates.  This results in a total of seven different
HAP/vapor profiles.   The various profiles are shown in Table
3-2.   These profiles  are used throughout the analysis.
Following is a brief  discussion of the generation of these
profiles.  More discussion of the procedures is provided in
Appendix C  (Section C.2).
     Since these programs are not in effect at this time,
HAP to VOC ratios were theoretically developed using the
arithmetic average vapor profile for normal fuel shown in
Table  3-1.  For reformulated and reformulated/oxygenated
fuels, the benzene content in the vapor was calculated using
an equation from earlier EPA analyses4 based on a 1.0 weight
percent benzene content  in the liquid.  The other aromatic
compounds were reduced equally by an amount determined
                             3-10

-------
CO

H

CO


H!
55-H

H Q)


P*

W >i

CO J3


   0)

co en
W (Q


H C
Ji, QJ

o o
Pi >-!
PJ (1)
   Q,

PS
2
   U
   o
 ^ ^^

^
n
CQ


EH
       CD
            «*   -»  •-  K.  •*  •-       (*  •-  N-  >*  «—       oj

   •DO    »-'o^oc3c>

   4-f
   a




   o
   H-  UJ
   u  m
   °c  •-


            «-   O  «-  O  O  O  CO


       3







            «O   O  K>  CO  in  «-       eo

       o    »^   o  »^  o  o  c>
                                i
                                       fl>
                                       ca
                                                      g
                                                      a
                                                     "8

                                                      3
                                                      o


                                                     'i
                                                      o
                                                      L.
                                                      O.



                                                      a.
                                                      a
                      a

                      S
                            o   ~.  x
                            »-   CNJ  x
                                       >•  ui
                                       JZ  CD
E5

g  .

Oj M




3 "3
—• tA
a a
o o>


2-8


*S
                                                      ss

                                                      il
                                                      II
                                                      S oi
                                                      L. (_
                                                      u
                                                     22
                                                      g
                                                      UI
                                             3-11

-------
necessary to reduce total aromatics to a level of 25 percent
in the liquid.  The nonaromatic compounds in the liquid were
also reduced to account for the volume of oxygenate added.
     3.2.1.2  Temperature and Vapor Pressure.  Volatility
and temperature have major impacts on emissions from the
evaporation of gasoline.  Evaporation can be explained by
the kinetic-molecular model.  A liquid molecule near the
surface of the liquid can escape to the vapor phase whenever
it gains sufficient kinetic energy to overcome it's
attraction to other particles surrounding it in the liquid.
The weaker the attractive forces, the more readily
vaporization occurs, and the more "volatile" the liquid.
The rate of vaporization increases with increasing
temperature, as this increased temperature provides more
kinetic energy to the liquid, causing more molecules to
vaporize.
     Reid vapor pressure (RVP) is a standard industry
measure of fuel volatility and represents the vapor pressure
of the fuel at 100°F.  Although RVP is a measure of fuel
volatility at 100°F, the empirical emissions equations used
to calculate emissions in this analysis reflect actual
temperature conditions.
     The RVP of gasoline is adjusted through blending at the
refinery to account for temperature and pressure differences
across the country.  In the summer when warm temperatures
enhance volatilization, gasolines can be blended with a
lower RVP and still provide ample vaporization for
combustion in the vehicle engine.  Reducing RVP in the
summer, therefore, reduces emissions from gasoline transfers
without reducing vehicle performance.  Too high an RVP in
the summer can create excess volatilization in the engine,
causing vapor lock.  During the winter months when cold
temperatures  inhibit volatilization, gasolines can be
blended with a higher RVP to ensure sufficient volatiliza-
tion for engine start-up and operation.  This increase in
RVP when temperatures decrease, and decrease in RVP when
                            3-12

-------
temperatures increase, is an attempt to provide a uniform
fuel volatility for smooth engine performance all year.
     In order to reduce emissions, EPA has established
maximum volatility levels for gasoline sold during the
summertime months.  On March 22,  1989 (54 FR 11868),  EPA
published a final rule restricting gasoline volatility.
This initial rule is referred to as Phase I.  The EPA later
promulgated a second level (phase II) of more stringent
volatility controls on June 11,  1991 (55 FR 23658),
scheduled to take effect in the summer of 1992.  The second
phase of volatility controls set monthly RVP requirements
for each State based upon many factors including, for
example, meteorological conditions.  Under Phase II the
maximum allowable RVP of gasoline sold in northern states
was set at 9.0 psi and the maximum allowable RVP of gasoline
sold in southern States was set at 7.8 psi.  The summertime
RVP limitations promulgated are shown in Table 3-3 along
with RVP values for the remainder of the year.
     However, the CAAA of 1990 limited EPA's authority to
set gasoline volatility levels below 9.0 psi.  The 1990 CAAA
specify that EPA may set RVP limitations below 9.0 only for
ozone nonattainment areas and former ozone nonattainment
areas.  Therefore, on May 29, 1991 (56 FR 24242), EPA
proposed to change the volatility standards to eliminate the
volatility level requirements (9.0 psi)  for those areas
where EPA no longer had the authority to adopt such levels.
Specifically, EPA proposed that the RVP for areas designated
attainment for ozone be restricted to 9.0, even if
nonattainment areas in the State are restricted to 7.8.
     Attempts to locate data on the temperature of gasoline
in aboveground storage tanks were unsuccessful.  Therefore,
a temperature of 60°F was used in all emission factor
calculations for aboveground storage tanks and 60°F for
below ground storage tanks.  These are the temperatures used
in previous EPA analyses of gasoline distribution regulatory
strategies.5'6
                            3-13

-------
          <
          a:
          >
          <
                 §
                 tr
                            K»f\JCV-O^OO>»TOO>lAK>O(*CO«M^OO>iOOO«O^«*
                            »»o»-o»-rj»-»-»^o*-^<<>i^-'— »---»-»^^i>jfO»^»-«-
                             u^rvJKiK»***»*jr*»wKi-r-po**%i-^r^^^*ro^*»^**sjrK^****^»K1^**
B
a
s
CQ
W
s
H
cn
x
CQ
P*
n
 I
ro
W
CQ
              CJ
              UJ
                             m>dj>i|Ou^in
                         fMI/1OK>«Mf\J«»»»«»IMr>Jr-IMKlvTvrru-*IM.J--*.»»»>»rv|K»-»«VI«-'»>»
              a.
              UJ
              (A
                         eoiNjeoeoaaaoooeocoeou^oooocno'Ooeooooeocooooooo
                         ^•^^<•K.^^^«cxo«^*^^N•*~o>>o»ocxN.cxN•O'^^o^o»o«^*•^*•o^o^^-o»o»
                          eomeaeoeocooocoeocoinooooooooooaooocoeoooeooo
                          N-K^t^-t^>^S.O»O*^^*^*~O*&'OO^^CXIs-O*^O»O>'O»f^N-O»ON-^iO»
                          aiutcocoocooococoeomoooocoocoocoooocacooocooo
                          cotncoaococooocotooouioooocoocoocoooocoaioocooo
                          N-rnhJhJr^Nlo>O>h~r^h^'-O>oIcxi>N!<>KO>^O>O>O>Nlh^O^O>^O>O>
                          oraooo^ooooomooooooooooooooooooo
                          o-sre^o»o^o>'»o'>o.'<>-^c>o^cv<><>o^c>c^c>cx(>cxeKCK(>o^cxo^<>
                          ^lT>O'~'-'-fOfOtMr-^^-POKlKirMT-rMr-fOfOK|fOKV'-(MrJ>»>»~»ivKir>j»»^^.»^«\jfO«*»»'<>'>»'>*'
                          ir>oir>ir>-j-<\JoorJif\iAin(\JOoo(MPjinoooooiAfMOOvroo
                          K1«mcMMK>«»min«»Kito»-«»vi»mir>«»«»Kiininul\inint»!'*iA«r>K>inm
                          u>oi/\(MoooN
                                                                                                               Q  K 5  <  2
                                                                                                                                          S
                                                                                                                                          Of
                                                            u.  o  z •-  -«  •—  •—
                  uj         a.
    si     §l-  _
<3»ui_i<  —  iu~O<33
                                                                        3-14

-------
                                                                                                             •—       u
        !!   S
                           O>OOOOtXOOOOOOO.O-OOOOO

                                                                                              ~   "
                                                                                                             rg

                                                                                                             O
            ai     u-
                           inorvioorviorMoorviooomorMinooooooo
         _j  »—
 _       i  S
 S  2  2  2
                             8
                                 O
                             3   3
                                         *-   O  <
    S«—              UJ
    -  oe  »•»          «-•

s  >  s  =f      i
—      o  —      o

-------
     Using the RVP values in Table 3-3 (taking into account
those southern State attainment areas) and the State
gasoline throughputs (see Appendix D), a national weighted
average RVP was calculated,  as well as weighted average RVPs
for the winter season (November through February) and the
nonwinter season (March through October).  The rationale for
calculating RVP for these time periods is discussed in
Section 3.3 and Appendix D.   This annual weighted average
RVP is 11.4 psi, the winter season is 14.0, and the
nonwinter season 10.2.   These will be used throughout the
analysis to calculate emission factors.
     3.2.1.3  Methods of Loading Gasoline.  Many of the
operations under consideration in this study involve the
loading of gasoline into a storage vessel or tank.  The
method of loading can affect the emissions generated during
the gasoline transfer.   There are two basic methods of
loading, splash and submerged fill.  In the splash loading
method, the nozzle is inserted into the top of the tank.
Significant turbulence and vapor/liquid contact occur during
the splash loading operation, resulting in high levels of
vapor generation and loss.  If the turbulence is great
enough, liquid droplets will be entrained in the vented
vapors.
     The second method of loading is submerged fill.  This
category is further broken down into the submerged fill pipe
method and the bottom loading method.  In the submerged fill
pipe method, the fill pipe extends almost to the bottom of
the tank.  In the bottom loading method, a permanent fill
pipe is attached to the cargo tank bottom.  Most of the time
using the submerged fill pipe method and always using bottom
loading, the fill pipe is below the liquid surface level.
Liquid turbulence is controlled significantly during
submerged loading, resulting in much  lower vapor generation
than encountered during splash loading.
     Cargo carriers are sometimes designated to transport
only one product, and in such cases are practicing
"dedicated service".  Dedicated gasoline cargo carriers
                            3-16

-------
return to a loading terminal containing air fully or
partially saturated with vapor from the previous load.
Cargo tanks may also be "switch loaded" with various
products, such as diesel fuel, so that a nonvolatile product
being loaded may expel the vapors remaining from a previous
load of a volatile product such as gasoline.  These
circumstances vary with the type of cargo tank and with the
ownership of the carrier, the petroleum liquids being
transported, geographic location, and season of the year.
     One control measure for gasoline tank trucks is called
"vapor balance service", in which the cargo tank of the
truck retrieves the vapors displaced during product
unloading at bulk plants or service stations and transports
the vapors back to the loading terminal.  A truck whose
cargo tank is in vapor balance service normally is saturated
with organic vapors.  Therefore the presence of these vapors
at the start of submerged loading results in greater loading
losses than encountered during nonvapor balance, or
"normal", service.
     Emissions from loading gasoline were estimated using
the following expression:7
               LL = 12.46 SPM/T
where:
     LL  =  Loading loss,  lb/103 gal of gasoline  loaded
     M   =  Molecular weight of vapors ,  Ib/lb-mole
     P   =  True vapor pressure of liquid loaded,  psia
     T   =  Temperature of bulk liquid loaded,  "R ('F + 460)
     S   =  A saturation factor
The saturation factor, S, represents the expelled vapor's
fractional approach to saturation, and it accounts for the
variations observed in emission rates from the different
unloading and loading methods.  Table 3-4 lists the
saturation factors as found in AP-42.8
                            3-17

-------
     TABLE 3-4.
SATURATION (S)  FACTORS FOR CALCULATING
 GASOLINE LOADING LOSSES
   Cargo Carrier
         Mode of Operation
S Factor
 Tank trucks  and
   rail tank  cars
   Submerged loading:   dedicated
     normal service                  0.60
                    Submerged loading:   dedicated
                      vapor balance  service           1.00
                    Splash loading:   dedicated
                      normal  service                   1.45
                    Splash loading:   dedicated
                      vapor balance  service           1.00
Source:  AP-42, page 4.4-6.

     An examination of this equation and the saturation
factors in Table 3-4 indicates that the emissions from
submerged loading are approximately 40 percent of those for
splash filling.  The only variable that differentiates
splash from submerged loading is the saturation factor.  The
normal service saturation factors are 0.6 for submerged
loading and 1.45 for splash, which represents a 60 percent
increase.
3.2.2  Emissions from Pipeline Facilities
     As discussed in Chapter 8, there are 79,624 miles of
gasoline product pipeline in the United States.  Pipelines
transport approximately one half of the gasoline shipped in
the U.S.  The pipeline itself is only one component of the
product pipeline system.  Other major components of this
system include terminals, pumping stations, and breakout
stations.
                            3-18

-------
     Product is carried from refineries to terminals by the
pipeline,  often over great distances.   The pipeline is made
of sections of steel, welded together, and usually buried
underground.  At the refinery,  a pump sends the refined
product toward its destination.   Since this pump is not
strong enough to "push" the material the entire distance,
pumping stations are located along the pipeline to keep the
product flowing.  Occasionally,  flow may be interrupted and
the product pumped off of the pipeline into storage tanks.
These "breakout" stations usually occur at pumping stations.
     3.2.2.1  Pumping Stations.   Pumps carry product from
refineries to the pipeline, where a larger pump pushes the
product toward its destination.   In route to its
destination, product passes through numerous pumping
stations (approximately one every 30-50 miles)9,  where it is
pumped along its way.
     The centrifugal pump is the most widely used pump.
However, other types, such as the positive-displacement pump
and the reciprocating pump are also used at pipeline pumping
stations.
     Two generic types of sealing devices, packed and
mechanical, are used on pumps in the petroleum industry.
Packed seals can be used on both centrifugal and
reciprocating types of pumps.  A packed seal consists of a
cavity in which the pump casing is filled with special
packing material that is compressed with a packing gland to
form a seal around the shaft.  To prevent the buildup of
frictional heat between the seal and shaft, lubrication is
required.   A sufficient amount of either the gasoline being
pumped or another liquid that is injected must be allowed to
flow between the packing and the shaft to provide the
necessary lubrication.  Deterioration of this packing and/or
the shaft seal face after a period of usage can be expected
to eventually result in leakage of organic compounds to the
atmosphere.
     Mechanical seals are limited in application to pumps
with rotating shafts and can be further categorized as
                            3-19

-------
single and dual mechanical seals.   There are many variations
to the basic design of mechanical  seals, but all have a
lapped seal face between a stationary element and a rotating
seal ring.  In a single mechanical seal application, the
rotating-seal ring and stationary  element faces are lapped
to a very high degree of flatness  to maintain contact
throughout their entire mutual surface area.  As with pump
packing, mechanical seal faces must be lubricated to remove
frictional heat.  However, because of the seal's construc-
tion, much less lubrication is needed.  If the seal becomes
imperfect due to wear, the gasoline being pumped can leak
between the seal faces and be emitted to the atmosphere.
     In a dual mechanical seal application, two seals can be
arranged back-to-back or in tanden.  In the back-to-back
arrangement the two seals provide  a closed cavity between
them.  A barrier fluid is circulated through the cavity.
Because the barrier fluid surrounds the dual seal and
lubricates both sets of seal faces, the heat transfer and
seal life characteristics are much better than those of the
single seal.  In order for the seal to function, the barrier
fluid must be held at a pressure greater than the operating
pressure of the stuffing box.  As a result some barrier
fluid will leak across the seal faces.  Liguid leaking
across the inboard face will enter the stuffing box and mix
with the gasoline.  Barrier fluid going across the outboard
face will exit to the atmosphere.   Therefore, the barrier
fluid must be compatible with the petroleum liquid as well
as with the environment.
     In a tandem dual mechanical seal arrangement, the seals
face the same direction.  The secondary seal provides a
backup for the primary seal.  A seal  flush is used in the
stuffing box to remove the heat generated by friction.  As
with the back-to-back seal arrangement, the cavity between
the two tandem seals is filled with a barrier fluid.
However, the barrier fluid is maintained at a lower pressure
than the  fluid in the stuffing box.   Therefore, any leakage
will be from the stuffing box into the  seal cavity
                            3-20

-------
containing the barrier fluid.   Since this liquid is routed
to a closed reservoir, gasoline that has leaked into the
seal cavity will also be transferred to the reservoir.   At
the reservoir, the petroleum liquid could vaporize and be
emitted to the atmosphere.   To ensure that VOCs do not leak
from the reservoir, the reservoir can be vented to a control
device.
     There are also numerous valves at a pumping station.
The types of valves commonly used are globe, gate, plug,
ball, relief and check valves.  All except the relief valve
and check valve are activated by a valve stem, which may
have either a rotational or linear motion, depending on the
specific design.  This stem requires a seal to isolate the
process fluid inside the valve from the atmosphere.  The
possibility of a leak through this seal makes it a potential
source of VOC and HAP emissions.  Since check valves do not
have an external actuating mechanism in contact with process
fluids, they are not considered to be potential sources of
emissions.
     Pipeline pumping stations contain on the average
approximately 55 valves and 5 pumps.  Uncontrolled emissions
from an example pipeline pumping station are shown in Table
3-5.  These emissions were calculated using AP-42 emission
factors developed for light liquid components at petroleum
refineries of 0.26 kg/component/day for valves and 2.7
kg/component/day for pump seals.    A more  recent  study  has
provided evidence that emission factors for leaking
equipment components may be lower than those reported in
AP-4211; however, since these new data were limited to only
a few terminals, the data were deemed insufficient to
justify changes to the national emission factors and as
such, the refinery data were considered appropriate for this
analysis.
     3.2.2.2  Breakout Stations.  Pipelines often occur in
clusters of two or three pipes that carry product from the
same origin to the same destination.  At some point along
the path,  one, two, or all three of the lines branch off in
                            3-21

-------
      TABLE 3-5.
  UNCONTROLLED  EMISSIONS FROM EXAMPLE
     PIPELINE FACILITIES
Emission Source
PUMPING STATION6
Valves
Pumps
VOC Emission Emission Factor Units
Factor"

0.26 kg VOC/ valve/day
2.7 kg VOC/pump seal/day
Annual Emissions
(Mg/yr)
HAPb VOC

0.3
0.5

5.2
9.8
                 Total for Example Pumping Station
                               0.8
15.0
 BREAKOUT STATION

 Storage Tanks
Standing storage losses
Withdrawal losses
Fugitive Emissions"
Valves
Pumps

18.1
4.61 x 10"8

0.26
2.7
Total for Example
Mg VOC/yr/tank
Mg VOC/bbl

kg VOC/ valve/day
kg VOC/punp sea I /day
Breakout Station
3.5
0.1

1.1
0.8
5.5
72.4
0.4

23.7
17.7
114. 2
a  Emission factors for pumps and valves taken from AP-42,
   Section 9.1,  for light liquid components  at petroleum
   refineries.
   Table 3-7.
Storage tank emission  factors taken from
b  Calculated using the arithmetic average  HAP to VOC ratio
   for normal fuel in Table 3-1.

c  Assuming the example pumping station has 55 valves and 5
   pumps (2 pump seals per pump) operating  365 days/yr.

d  Assuming the example breakout station  has four
   "equivalent dedicated tanks" that are  external floating
   roof tanks with primary seals each having a capacity of
   8,000 m3 (50,000 bbls)  and an annual throughput of 1.2 x
   109 liters (315 x 106 gallons) which represents  150
   turnovers per year.

e  Assuming the example breakout station  has 250 valves and
   9 pumps (2 pump seals per pump) operating 365 days/yr.
                             3-22

-------
different directions.  When this occurs,  the throughput to
any one line is altered.  Storage tanks at breakout stations
are used in this situation to temporarily store the product
until compensation for the reduced flow can be made.  Also,
at times the diameter of a pipeline will be changed (reduced
or increased).  This also causes a change in the flow rates,
and breakout stations are needed to store product at these
locations.
     There are two major sources of emissions at breakout
stations.  These are the storage tanks and the pumps and
valves used to transport the gasoline.  Fugitive emissions
from pumps and valves are discussed above under pumping
stations.
     Many tanks in gasoline service have an external
floating roof to prevent the loss of product due to
evaporation and working losses.  Fixed-roof tanks, 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 sources of greatest 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.  These 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
                            3-23

-------
overflowing through the pressure vacuum valve.  Combined
breathing 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 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 changes in the liquid level, constitute the greatest
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 a pressure differential
around the floating roof.  As air flows into the annular
vapor space (ring-shaped space between the seal edge and the
tank wall) on the leeward side, an air-vapor mixture flows
out on the windward side.  Another source of standing-
storage loss is associated with roof fittings.  Roof
fittings can be a source of evaporative loss when they
require openings in the floating roof.  Typical roof
fittings include access hatches, unslotted guide-pole wells,
slotted guide-pole/sample wells, gauge-float wells, gauge-
hatch/sample wells, vacuum breakers, roof drains, roof legs,
and rim vents.12
                            3-24

-------
     Withdrawal loss is another source of emissions from a
floating roof tank.  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 the 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.  A practice that is becoming
more popular is the installation of geodesic dome covers
over external floating roof tanks.  These domes do not allow
air to flow directly over the floating roof and therefore
reduce emissions.
     Tables 3-6 and 3-7 present emission factors for storage
tanks.  These emission factors were calculated using the
emission factor equations contained in Section 4.4 of AP-42,
assuming 60°F and the national weighted average RVP of 11.4
as shown in Table 3-3.
     While a breakout station may contain a large number of
storage tanks, there will only be a select few that are used
for gasoline at any one time.  It is estimated that a
breakout station typically has four "equivalent dedicated
storage tanks" for gasoline.  That is, at any one time, only
four storage tanks are being filled with and storing
gasoline.  These facilities also contain approximately 250
valves and 9 pumps.
     Emissions for an example breakout station were shown in
Table 3-5.  It was assumed that the average throughput
                            3-25

-------
         TABLE 3-6.  STORAGE TANK EMISSION FACTORS
             FOR BULK TERMINAL  STORAGE TANKSa'b
Type of Emission
Fixed-Roof Uncontrolled
Breathing losses
Working losses
Internal Floating Roofc
Rim Seal losses
Fitting losses
Deck Seam losses
Working losses
voc
Emission
Factor
10.1
38.1
0.5
1.2
0.6
7.33 X 10'8
Units
Mg VOC/yr/tank
Mg VOC/yr/tank
Mg VOC/yr/tank
Mg VOC/yr/tank
Mg VOC/yr/tank
Mg VOC/bbl
throughput
 External Floatina Roof
    Standing Storage losses

       Primary seal

       Secondary seal6

    Working losses
    14.5

     7.0

4.61 X 10
-8
Mg VOC/yr/tank

Mg VOC/yr/tank

  Mg VOC/bbl
  throughput
a Emission factors calculated with equations  from  Section
  4.3 of AP-42 using the nationwide weighted  average RVP of
  11.4 and temperature of 60°F, as discussed  in  Section
  3.2.1.2.

b Assumes storage tanks at bulk terminals have a capacity
  of 2,680 m3 (16,750 bbl),  a diameter of 15.2 meters  (50
  feet), and a height of 14.6 meters  (48 feet).

c Assumes that  internal  floating  roof  is equipped with a
  liquid-mounted resilient seal (primary only).

d Assumes that  external  floating  roof  is equipped with a
  primary metallic shoe seal.

e Assumes that  external  floating  roof  tank is equipped with
  a shoe-mounted secondary seal.
                            3-26

-------
          TABLE 3-7.   STORAGE  TANK EMISSION  FACTORS
               FOR PIPELINE BREAKOUT STATION
                       STORAGE TANKSa'b
Type of Emission
Fixed-Roof Uncontrolled
Breathing losses
Working losses
Internal Floating Roofc
Rim Seal losses
Fitting losses
Deck Seam losses
Working losses
External Floatina Roof
VOC
Emission
Factor
30.4
472.4
1.2
1.3
2.6
7.33 X 1(T8
Units
Mg VOC/yr/tank
Mg VOC/yr/tank
Mg VOC/yr/tank
Mg VOC/yr/tank
Mg VOC/yr/tank
Mg VOC/bbl
throughput
    Standing Storage losses

       Primary seald               18.1

       Secondary seal6               8.5

    Working losses             4.61 x 10
-8
Mg VOC/yr/tank

Mg VOC/yr/tank

  Mg VOC/bbl
  throughput
a Emission  factors calculated with equations from Section
  4.3  of AP-42 using the nationwide weighted average RVP of
  11.4  and  temperature of  60 °F, as discussed in Section
  3.2.1.2.

b Assumes storage tanks at pipeline breakout stations have
  a capacity of 8,000 m3 (50,000 bbl),  a diameter of 30
  meters  (100 feet), and a height of 12 meters  (40  feet).

c Assumes  that  internal  floating roof is equipped with  a
  liquid-mounted resilient seal (primary only).

d Assumes  that  external  floating roof is equipped with  a
  primary metallic shoe seal.

e Assumes  that  external  floating roof is equipped with  a
  shoe-mounted secondary seal.
                            3-27

-------
for a breakout station storage tank is approximately 1.2 x
109 liters/year (315  x 106 gallons/year).
3.2.3  Bulk Terminals
  As noted above, bulk terminals receive gasoline from
refineries by way of pipeline, ship,  or barge.  Some
terminals are located at the refinery.  The product is
stored and then loaded into transport trucks that carry it
further down the distribution chain.   In a few situations,
gasoline is loaded at bulk terminals into railcars.   This
gasoline is usually carried to other terminals that do not
have access to a pipeline,  ship, or barge.
  There are three categories  of emission sources at bulk
terminals.  These are the emissions associated with the
loading of transport trucks or railcars (loading rack
emissions), storage tank emissions, and fugitive emissions
from leaking pumps and valves.
   3.2.3.1  Loading Rack Emissions.  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 and railcars.
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 distribution 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 ten 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 and  railcar loading
operations at terminals occur when the product being loaded
displaces the vapors  in the delivery tank and forces the
                            3-28

-------
vapors to the atmosphere.  Loading may be performed using
either splash, top submerged, or bottom loading methods.
Top loading involves loading of gasoline into the tank truck
compartment or railcar through the hatchway located on top
of either the truck tank or railcar using 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).
As discussed in Section 3.2.1.3, 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.
A long established 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 the
advantages cited for bottom loading include:  (1) improved
safety, (2) faster loading, and (3) reduced labor costs.
Loading rack emission factors and emissions at bulk
terminals are summarized in Table 3-8.
  3.2.3.2  Storage Tank Emissions.  Bulk terminals
typically have four or five aboveground storage tanks for
gasoline, each with a capacity ranging from 1,500 to 15,000
m3 (9,400 to  94,000  barrels).16  Table 3-8 also illustrates
the magnitude of emissions from a bulk terminal with four
storage tanks for gasoline, using the emission factors shown
in Table 3-6.
  3.2.3.3  Fugitive Emissions.  There are numerous pumps
and valves at bulk terminals that convey liquid gasoline and
gasoline vapors.  As discussed in Section 3.2.2.2 under
pipeline pumping stations, these components can be sources
of HAP emissions.  Table 3-8 also summarizes the magnitude
of the fugitive emissions from a bulk terminal with 150
valves and 10 pumps.
                            3-29

-------
TABLE 3-8.  UNCONTROLLED EMISSIONS FROM BULK TERMINALS
Emission Source
Loading Racks"
Submerged loading
Splash Fill
Storage Tanks'*
Fixed- roof
Working losses
Breathing losses
Internal Floating Roof
Working Losses
Breathing Losses
External Floating Roof
Working Losses
Primary Seal Losses
Secondary Seal Losses
Fugitive Emissions'
Valves
Pumps
VOC Emission
Factor*

658
1,590


38.1
10.1

7.33X10"8
2.3

4.61X10"8
14.5
7.0

0.26
2.7
Emissions
(Mg/yr)
Emission Factor Units HAPb VOC

mg VOC/ liter 11
mg VOC/ liter 27


Mg VOC/yr/tank 7
Mg VOC/yr/tank 2

Mg VOC/bbl throughput <1
Mg VOC/yr/tank 0.4

Mg VOC/bbl throughput <1
Mg VOC/yr/tank 4
Mg VOC/yr/tank 2

kg VOC/valve/day 1
kg VOC/pump sea I /day 1

230
556


152
40

<1
9

<1
72
34

15
20
Loading rack and storage tank factors are calculated
using the weighted average RVP of 11.4  (summer RVP =
10.2, winter RVP = 14) and temperature  of 60°F  (see
discussion in Section 3.2.1.2).  Fugitive emission
factors are from AP-42 section 9.1, and are those for
light liquid components at refineries.

Calculated using the arithmetic average HAP to VOC ratio
for normal fuel of 4.8 percent as derived in Table 3-1.

Assuming a throughput of 950,000 liters/day  (250,000
gallons/day) for 340 days/yr  (average annual throughput
of 35 x 107 liters).

Assuming four storage tanks, each having a capacity of
2,680 m^ (16,750 bbl) and a throughput of 950,000
liters/day (250,000 gallons/day) for 340 days/yr  (13
turnovers per year).

Assuming that bulk terminals typically  have  150 valves
and 10 pumps  (2 pump seals per pump).
                          3-30

-------
3.2.4  Bulk Plants
   Bulk gasoline plants are secondary distribution
facilities that receive gasoline from bulk terminals by
truck transports, store it in aboveground, fixed-roof
storage tanks, and subsequently dispense it via smaller
account trucks to local farms, businesses, and service
stations.  Bulk plants typically have 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.17  A bulk plant  is  defined 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.
   3.2.4.1  Storage Tank FillingEmissions.  Gasoline is
delivered to bulk plants in large tank trucks from bulk
terminals.  One source of emissions is during the filling of
the storage tank at the bulk plant.  The storage tanks at
bulk plants are almost always fixed-roof tanks.
Consequently, before the filling of the tank, the space
available for filling contains saturated gasoline vapors.
Emissions are generated when the incoming liquid forces
these vapors out the vent.  Due to the configuration of the
aboveground tanks, this loading is usually accomplished
using bottom loading.
   3.2.4.2  Loading Rack Emissions.  The methods of loading
gasoline 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.  In a 1976
survey of bulk plants, 75 percent used either top-submerged
filling or bottom filling and 25 percent used top splash
filling.18  These  bulk plants that  use top  splash  filling
are typically located in areas where no control is required.
Emissions from an example bulk plant with a daily throughput
                            3-31

-------
of 19,000 liters/day (5,000 gallons/day) are shown in Table
3-9.
   3.2.4.3  Storage Tank Emissions.  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 that combined are called "working losses".
Storage tank emissions are also estimated for an example
bulk plant with three storage fan^s in Table 3-9.
   3.2.4.4  Fugitive Emissions.  As with bulk terminals,
there are numerous pumps and valves at bulk plants that
convey liquid gasoline and gasoline vapors.  As discussed in
Section 3.2.2.2 under pipeline pumping stations, these
components can be sources of HAP emissions.  The estimated
emissions shown  in Table 3-9 are for an example plant that
has 50 valves and 4 pumps.
3.2.5  Service Stations
   The discussion on service station operations is divided
into three areas:  (1) the filling of the underground
storage tank, (2) automobile refueling, and 3) storage tank
emissions.  Although terminals and bulk plants also have two
distinct operations  (tank filling and truck loading), the
filling of the underground tank at the service station ends
the wholesale gasoline distribution 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.  Storage tank emissions occur due to storage
tank breathing during pressure and temperature changes and
the inbreathing  and subsequent outbreathing during storage
tank emptying.
    3.2.5.1  Storage Tank Filling Emissions.  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
                            3-32

-------
              TABLE 3-9.  UNCONTROLLED EMISSIONS
                  FROM AN EXAMPLE  BULK PLANT3
Annual Emissions
(Mg/yr)
Emission Source
Storage Tanks'1
Working Losses
Breathing Losses
VOC Emission
Factor"

432
203
Emission Factor
Units

mg VOC/ liter
mg VOC/ liter
HAPe

0.1
0.1
VOC

2.5
1.2
 Tank Truck Unloading/
   Storage Tank Filling         1,081        mg VOC/liter      0.3      6.2
 Loading Racks
   Submerged loading           738         mg VOC/liter      0.2      4.2
 Fugitive Emissions'
   Valves                  0.26       kg VOC/valve/day     0.2      3.9
   Pumps                    2.7      kg VOC/pump sea I/day    0.3	6.5

                        Total for an Example Bulk Plant      1.2      24,4


3  Assuming the example bulk plant  has a gasoline throughput
   of 19,000  liters/day (5,000 gallons/day), 3 storage
   tanks,  50  valves,  and 4 pumps, and operates 300  days/yr.

b  Storage tank filling (working  loss)  and breathing loss,
   emission factors calculated using  equations in Section
   4.4  of AP-42.   Loading rack emission factor calculated
   using the  AP-42 equation from  section 4.4 discussed in
   Section 3.2.2.2 of this document.   Fugitive emission
   factors taken  from Section 9.1 of  AP-42 for light liquid
   components at  refineries.  Nationwide weighted average
   RVP  of 11.4 and temperature of 60°F as discussed in
   Section 3.2.1.2.

c  Calculated using the arithmetic  average HAP to VOC ratio
   for  normal fuel of 4.8 percent as  derived in Table 3-1.

d  Assumes storage tank capacity  of 76 m3 (640 bbl).

e  Assuming the example bulk plant  has 50 valves and 4 pumps
   (2 pump seals  per pump).
                              3-33

-------
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.  Estimated emissions for an example 190,000
liters/months (50,000 galIons/month) service station are
shown in Table 3-10.
     3.2.5.2  Vehicle Refueling Emissions.  In addition to
service station tank loading losses, vehicle refueling
operations are considered to be a source of emissions.
Vehicle refueling emissions are attributable to vapor
displaced from the automobile tank by dispensed gasoline and
to spillage of fuel.  The major factors affecting the
quantity of emissions are gasoline temperature, auto tank
temperature, and gasoline RVP.  Table 3-10 illustrates the
uncontrolled emissions from an example gasoline service
station.  The refueling emission factors presented in Table
3-10 are from a technology guidance document for vehicle
refueling controls.20
     3.2.5.3  Storage Tank Breathing and Emptying Emissions.
Emissions have also been reported at service stations due to
storage tank emptying and breathing losses.  Breathing
losses are attributable to gasoline evaporation due to
barometric pressure and temperature changes.  Breathing
losses in fixed volume storage tanks are caused by vapor and
liquid expansion and contraction due to diurnal temperature
changes.  As temperatures increase, vapor volume increases,
pushing vapor out of the vent pipe  (out-breathing).  When
temperatures decrease, vapor volume decreases and air is
drawn into the tank  (in-breathing).  Breathing loss
emissions have traditionally been minimal at service
stations since storage tanks have generally been located
underground, insulated by the earth, with a very stable
temperature profile.  However, breathing losses from service
                            3-34

-------
        TABLE  3-10.   UNCONTROLLED EMISSIONS FROM AN
                   EXAMPLE  SERVICE STATION3
Emission Source
VOC Emission
  Factor1"
Emission Factor
   Units
                                                    Annual Emissions
                                                      
-------
station storage tanks are becoming more prevalent due to the
popularity of aboveground storage tanks and the installation
of vaulted underground storage tanks.  Aboveground storage
tanks are more susceptible to temperature and pressure
changes than underground tanks and thus are more likely to
experience both vapor growth and vapor shrinkage quite
similar to working and breathing losses for fixed-roof tanks
at bulk terminals which were discussed earlier in this
chapter (see Section 3.2.3.3).  Consequently/ the emission
factors cited in AP-42 and which appear in Table 3-8 may be
used to calculate emissions from these tanks even though
they are necessarily smaller than bulk terminal fixed-roof
storage tanks.  It is also reported that the double wall, or
vaulted underground storage tanks being installed to comply
with underground storage tank (UST) regulations are
susceptible to thermal effect and therefore breathing losses
as well.  However, these losses are reported to be
insignificant.21'22
     Emptying losses occur when gasoline is withdrawn from
the tank, allowing fresh air to enter.  This enhances
evaporation (i.e., vapor growth) and causes vapors to be
vented from the pipe as the saturated gasoline vapors tend
to occupy a larger volume than air.  The EPA's AP-42 cites
an average breathing emission rate of 120 milligrams per
liter of throughput.23
     The original source for this factor was an article in
the Journal of the Air Pollution Control Association
(November 1963) based on a study by the Air Pollution
Control District of Los Angeles County (LAAPCD) and was
entitled "Emissions from Underground Gasoline Storage
Tanks".2*  This article describes emptying losses as
follows:
          When an automobile is fueled, gasoline is
     pumped from the underground tank, causing air to
     be inhaled through the vent pipe, the volume being
     approximately equal to the volume of gasoline
     withdrawn.  The air then becomes saturated with
     gasoline vapors, tending to occupy a larger
     volume.  This, in turn, causes the vapor-air
                            3-36

-------
     mixture to exhaust from the underground tank until
     a pressure equilibrium is attained.
     The mg/1 emission factor listed in AP-42 was estimated
in this study by measuring air expelled from the vent pipe
after vehicle fueling.  Since the authors concluded that it
was impractical, in their study, to collect representative
vapor samples for analysis, they assumed a theoretical
gasoline vapor to air ratio of 40 percent.  Using these
data, an emission factor of one pound per thousand gallons
of throughput (approximately 120 mg/1) resulted.  While an
emission factor was calculated by the authors, they went on
to discuss complexities with estimating emissions.  The
study concluded:
          Factors affecting the breathing losses are
     complex and interrelated, depending on the service
     station operation, pumping rate, frequency of
     pumping, ratio of liquid surface to vapor volume,
     diffusion and mixing of air and gasoline vapors,
     vapor pressure and temperature of the gasoline,
     the volume and configuration of the tank, and the
     size and length of the vent pipe.  Because of
     these many variables involved, much more data from
     a number of representative retail stations would
     be necessary before an accurate determination of
     overall, basin-wide breathing losses could be
     made.
     Since the time of this original analysis, several
studies have been conducted to attempt to account for many
of these variables.  These range from studies that conclude
there are no VOC emptying losses to those reporting
emissions much higher than those predicted by the AP-42
emission factor.
     Dr. R.A. Nichols has studied this subject extensively
throughout the 1970's and 1980's.  In a 1987 paper on the
subject,25 his conclusion was that the model used in the
LAAPCO analysis ignored the effect of the vent line.  Dr.
Nichols states:
     As can be seen when air enters a nearly flat tank
     containing saturated vapors, as it layers, it is
     exposed to a large area for diffusion and quickly
     saturates....Consequently,  as the surface layer
     gains vapor, the lighter upper vapor free area is
     vented from the tank....if a tank being
                            3-37

-------
     continuously defueled is then held quiescent,  the
     roughly steady-state but unsaturated profiles  in
     the vapor space will slowly but continuously
     enrichen.  As the profiles enrichen, the amount of
     vapor in the vapor space will grow and this amount
     of vapor will be exhausted into the vent
     line....emissions will result.  However, since
     high turnover tanks subject to appreciable
     concentration profiles in the vapor space...are
     also subject to higher more uniformly frequent
     withdrawals and typically have fuel which is
     unsaturated with respect to air to a greater
     degree..., little vapor is expected to be vented.
          There is an additional effect which tends to
     mitigate venting....as saturated vapor moves up the
     vent pipe, it creates a sliaht pressure on the
     remaining vapor space.  Until the entire vent  pipe +
     1.5 gallons of vapor saturation is produced, virtually
     no vapors will be vented.
     Dr. Nichols indicates that vapor emissions could only
occur during periods of long refueling inactivity.   He
concludes that high fueling activity followed by long
periods of inactivity will lead to the highest (and possibly
the only) vapr>r venting emissions.  Thic paper did  not
provide any emission factor for these emissions.
     The California Air Resources Board  (GARB) conducted a
study in 1987 to estimate storage tank breathing losses.26
Emissions were measured at a low throughput  (15,000 gallons
per month per tank) station and a high throughput (50,000
gallons per month per tank) station.  The study found
different results for the two stations.  The emission factor
calculated for the low throughput station was 0.92  Ibs VOC
per 1,000 gallon throughput (110 mg/1), and 0.21 pounds per
1,000 gallon  (25 mg/1) for the high throughput station.
Observations made during the testing indicated that mass
emissions from the underground storage tanks appeared to
occur during periods when dispensing of product was the
lowest, that emissions were at a minimum during conditions
of near continuous fuelings, and that the highest mass
emissions occurred during intermittent vehicle fuelings
followed by relatively long periods of dispensing
inactivity.  The differences in emission factors at the high
                            3-38

-------
and low throughput stations are explained in these
observations.
     The National institute for Petroleum and Energy
Research (NIPER) conducted a study and reached conclusions
partially in agreement with those of both Dr. Nichols and
CARB.27  NIPER's study concluded that no vent losses would
occur if the dispensing frequency were high enough and that
vent losses would be markedly reduced if the height of the
vent was increased.  The rationale for the origin of
emissions agreed with the discussion provided in the
original LAAPCD study.  This was that emissions were due to
1) air induction through the vent; 2) dilution of the
hydrocarbon vapor in the tank; and 3) saturation of the
diluted vapor by evaporation of the liquid fuel, resulting
in increased pressure in the tank.  When this pressure was
greater than that exerted by the column of vapor in the
vent, emissions resulted.  The emissions measured for high
flow stations were 0.85 and 1.05 grams per gallon dispensed
(225 and 277 mg/1, respectively).
     A comparison of the CARB and NIPER studies shows that
the NIPER emission factors are much higher than those from
CARB.  Recognizing this discrepancy, CARB and NIPER met on
August 24, 1987 to discuss the differences.  The conclusion
reached at this meeting was that NIPER7s results should be
adjusted because the dispensing period during NIPER's tests
was not considered representative of the effective
dispensing period at a high volume station.  Adjustments
were made and it was determined that a more appropriate
emission factor for the NIPER data is 0.6 lbs/1,000 gallons
(72 mg/1) for a high throughput station.28
     In summary, these studies indicate that the emissions
from storage tank emptying are affected by several factors,
most notably the height of the vent pipe and the vehicle
fueling activity.  Additionally, for this analysis,
calculations of emissions are based on emission factors for
.underground storage tanks even though it is recognized that
there are above ground tanks in existence (the number of
                            3-39

-------
above ground tanks is very small in comparison to the number
of underground tanks).  Therefore, for the purposes of the
analysis in this document, it is believed that the AP-42
factor of 120 mg/1 for underground tanks represents an
emission factor that may be very conservative, but is not
unrealistic.

3.3  BASELINE EMISSIONS
     The baseline is defined as the quantity of emissions
expected in the "base year" in the absence of additional
regulation.  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 distribution ^cnrce
category was selected as 1998.  This year represents the
fifth year after the expected proposal of the regulation
when the selected regulation would be in full effect.  The
general approach for establishing the emission baseline was
basically the same for each sector of the industry.
     As discussed in Section 3.2.1.1.2, there are four basic
types of fuels that will be used.  These are normal,
reformulated, oxygenated, and reformulated/oxygenated.
During the winter months, all four types will be used while
only normal and reformulated will be required in the
remainder of the year.  The use of each of these fuels
depends on the ozone and CO area attainment designations as
well as area populations.  For purposes of this analysis, it
is assumed that all nonattainment areas would "opt-in" to
the program.  Consequently, it is estimated that these areas
would utilize approximately 42 percent of the total gasoline
consumed nationwide.  Due to the different types of fuels
that will be in use  in the base year, the parameters for
                            3-40

-------
calculating emissions (either gasoline throughput or
facility population) were separated according to location.
     For each State, data were obtained on the level of
control already in use.   The appropriate regulatory coverage
for each fuel type area in each State was determined and the
parameters for the area attributed to that control level.
Table 3-11 shows the baseline parameters by control level
for all industry sources.
     VOC emission factors were selected to represent the
level of control in both controlled and uncontrolled
situations.  VOC emissions were calculated by multiplying
the VOC emission factors by the corresponding throughput or
facility population.  HAP emissions were then estimated by
multiplying the VOC emissions by the appropriate HAP to VOC
ratio.
     The HAP and VOC emissions for the base year of 1998 are
presented in Table 3-12.  A complete description of the
baseline emissions analysis is provided in Appendix D.
                            3-41

-------
         TABLE 3-11.  1998 BASELINE PARAMETERS USED
                    IN EMISSIONS ANALYSIS
                                      Annual
                                     Gasoline     Number of
 Source Category/Control  Level      Throughput      Sources
	(106 liters)	

 PIPELINE FACILITIES
 Pipeline Pumping Stations
   Fugitive Emissions
      Uncontrolled                                1,989
 Pipeline Breakout Stations
   Fugitive Emissions
      Uncontrolled                                  270
   Storage Tanks8
   External Floating Roof
 Tanks
      Primary and Secondary           325,000       272
 Seals
      Primary Seals                   567,000       476
   Fixed Roof Tanks
      Internal Floating Roofs         105,000        88
	Uncontrolled	171,000	143
 8    These tank populations represent the "equivalent
      dedicated" storage tanks used for the emissions
      analysis (see Section 3.2.2.2).  The total storage
      tank population at breakout stations is estimated to
      be 2,227 external floating roof tanks (808 with
      primary and secondary seals and 1,419 with primary
      seals only) and 1,073 fixed-roof tanks (662 with
      internal floating roofs and 411 uncontrolled).
                            3-42

-------
                 TABLE 3-11.  (Continued)
                               Annual Gasoline
                                  Throughput
Source Category/control Level    (106 liters)     Number of
                                                  Sources
BULK TERMINALS
  Loading Racks
     80 mg/1 and 90% Control         115,000       265
     35 mg/1                         187,000       430
     10 mg/1                          13,000        29
     Submerged Fill                  123,000       282
     Splash Fill                       8,000        18
  Storage Tanks
  External Floating Roof
Tanks
     Primary and Secondary           134,000     1,802
Seals
     Primary Seals                   180,000     2,426
  Fixed Roof Tanks
     Internal Floating Roofs          95,000     2,732
     Uncontrolled                     37,000     1,072
  Tank Trucks
     Annual Vapor Tightness
     Testing                         317,000    31,169
     Uncontrolled           	 129,000    12,731
                           3-43

-------
                  TABLE  3-11.   (Concluded)
                                Annual Gasoline
                                   Throughput
 Source  Category/Control  Level     (106 liters)    Number of
	Sources

 BULK PLANTS
   Incoming Loads
     Vapor Balance                     52,600     5,661
     Uncontrolled                      34,700     6,936
   Outgoing Loads
     Vapor Balance                     48,800     4,488
     Submerged  Fill                    29,800     6,375
     Splash  Fill                        8,700     1,734
   Tank  Trucks
     Annual  Vapor Tightness
     Testing                          52,400     22,440
     Uncontrolled                      34,900     21,360
 SERVICE STATIONS
   Underground Tank Filling
     Vapor Balance/No                 156,100    135,146
 Exemption
     Vapor Balance/With               142,700    123,562
 Exemption
     Submerged  Fill                    75,800     66,476
     Splash  Fill                       71,400     62,566
                            3-44

-------
         TABLE 3-12.   1998  BASELINE EMISSIONS FROM
               GASOLINE DISTRIBUTION SOURCES
   Facility/Emission Source
Annual Emissions (Mg/yr)
    HAP           VOC
Pipeline Facilities
     Pumping Stations
     Breakout Stations
        Storage Tanks
        Fugitive Emissions

Bulk Terminals
     Storage Tanks
     Loading Racks
     Tank Truck Leakage
     Fugitive Emissions

Bulk Plants
     Storage Tank Filling
     Truck Loading
     Truck Leakage
     Fugitive Emissions


Service Stations (Stage I)
   2,370
 31,610
6,370
860
9,600
5,510
2,960
3,730
4,340
16,540
1,960
2,390
890
9,190
14,430
11,880
84,110
11,450
127,170
90,210
48,020
53,960
56,450
248,640
35,600
41,200
13,210
130,760
220,770
213,970
          TOTALS
  52,450
810,550
                           3-45

-------
3.4  REFERENCES
      1.  Northeast Corridor Regional Modeling Project -
          Determination of Organic Species Profiles for
          Gasoline Liquids and Vapors.  U.S. Environmental
          Protection Agency, Research Triangle Park, NC.
          Publication No. EPA-450/4-80-036a.  December 1980.

      2.  Dickson, C.L. and P.W. Woodward.  Motor Gasolines,
          Winter 1991-92.  National Institute for Petroleum
          and Energy Research.  Bartlesville, OK.  NIPER-175
          PPS.  June 1992. Page 30.

      3.  Telecon.  Shedd, S., U.S. Environmental Protection
          Agency to Stricter, B. , American Petroleum
          Institute.  February 20, 1992.  API view of
          reformulated gasolines and MTBE.

      4.  Evaluation of Air Pollution Regulatory Strategies
          for Gasoline Marketing Industry.  U.S. Environ-
          mental Protection Agency.  Research Triangle Park,
          NC and Ann Arbor, MI.  Publication No. EPA-450/4-
          84-0123.  July 1984.

      5.  Reference 4.
          Refueling Emission Regulations for Gasoline-Fueled
          Motor Vehicles — Volume I - Analysis of Gasoline
          Marketing Regulatory Strategies.  U.S. Environ-
          mental Protection Agency.  Research Triangle Park,
          NC and Ann Arbor, MI.  Publication No; EPA-450/3-
          87-OOla.  July 1987.

      7.  Compilation of Air Pollutant Emission Factors,
          Fourth Edition (AP-42) .  U.S. Environmental
          Protection Agency, Research Triangle Park, NC.
          Section 4.4, Transportation and Marketing of
          Petroleum Liquids.  September 1985.

      8 .  Reference 7 .

      9.  Memorandum.  Thompson, S., Pacific Environmental
          Services, Inc., to Shedd, S., U.S. Environmental
          Protection Agency, March 27, 1991.  Report of Trip
          to Plantation Pipeline, Greensboro, N.C.

     10.  AP-42.  U.S. Environmental Protection Agency,
          Research Triangle Park, NC.  Section 9.1,
          Petroleum Refining.  October 1980.

     11.  American Petroleum Institute.  Development of
          Fugitive Emission Factors and Emission Profiles
          for Petroleum Marketing Terminals, Interim Report.
          July 1992.

                            3-46

-------
12.  American Petroleum Institute.   Evaporative Loss
     From External Floating-Roof Tanks,  API Publication
     2517.  February 1989.

13.  Norton, Robert L.,  Pacific Environmental Services,
     Inc.  Evaluation of Vapor Leaks and Development of
     Monitoring Procedures for Gasoline Tank Trucks and
     Vapor Piping.  Prepared for U.S. Environmental
     Protection Agency.   Research Triangle Park, N.C.
     Publication No. EPA-450/3-79-018.  April 1979.

14.  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.
     Pages 3-15 and 3-17.

15.  Reference 7, p. 4.4-13.

16.  Reference 14.

17.  Pacific Environmental Services, Inc.  Study of
     Gasoline Vapor Emission Controls at Small Bulk
     Plants.  Prepared for U.S. Environmental
     Protection Agency,  Region VIII.  EPA Contract No.
     68-01-3156, Task No. 5.  October 1976.

18.  Reference 17.

19.  Technical Guidance - Stage II Vapor Recovery
     Systems for Control of Vehicle Refueling Emissions
     at Gasoline Marketing Facilities.  U.S.
     Environmental Protection Agency, Research Triangle
     Park, NC.  November 1991.

20.  Reference 19.

21.  Telecon.  Bowen, E., Pacific Environmental
     Services, Inc. with Bradt, R., Hirt Engineers.
     September 25, 1991.  Comments on preliminary draft
     Stage II technical guidance document.

22.  Letter.  Kunaniec,  K., Bay Area Air Quality
     Management District, to Shedd, S.,  U.S.
     Environmental Protection Agency.  July 31, 1991.
     Comments on preliminary draft Stage II technical
     guidance document.

23.  Reference 7, p. 4.4-7.

24.  Burlin, R., and A.  Fudiruch.  Air Pollution From
     Filling Underground Gasoline Storage Tanks.  Los
     Angeles Air Pollution Control District.  December
     1962.
                       3-47

-------
25.  Nichols, R.A.   Service Station Underground Tank
     Breathing Emissions.  R.A. Nichols Engineering.
     October 13,  1987.  Rev. February 16, 1988.

26.  Memorandum.   Simeroth, D.C., California Air
     Resources Board, to Cackette, T.,  California Air
     Resources Board, September 15, 1987.
     Determination of Mass Emissions from Underground
     Storage Tanks.

27.  National Institute for Petroleum and Energy
     Research.  Evaporative Losses from a Vented
     Underground Gasoline Storage Tank (with addendum
     discussing August 24, 1987 meeting with CARB).
     Undated.

28.  Reference 26.
                       3-48

-------
              4.0  EMISSION CONTROL TECHNIQUES

4.1  CONTROL TECHNIQUES
     This chapter describes available control techniques
that can be used to reduce emissions from sources in the
gasoline distribution network.  A large portion of the
gasoline distribution industry employs vapor control
technology that has been demonstrated, installed, and
operated at facilities for many years.  The control strategy
for storage tanks has been to reduce emissions 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 incorporate
submerged fill and to collect and transfer vapors back to
the bulk terminal vapor recovery unit (VRU) or thermal
oxidizer for treatment.  The control of fugitive emissions
from pumps and valves has been studied extensively for other
petroleum and chemical process industries but never
specifically applied to gasoline marketing sources through
EPA rules.  Controls for storage tanks,  bulk plants, bulk
terminals, and underground tank filling at service stations
are commonly referred to as Stage I.  Controlling emissions
as a result of vehicle refueling at service stations is
commonly referred to as Stage II, but is not included in
this source category effort.
     This chapter discusses techniques for controlling
emissions from each of the sources in the gasoline marketing
chain.  For each source or type of sources, the control
techniques discussion is followed by a section addressing
the technique effectiveness.  In most instances, this
discussion is in terms of effectiveness for controlling

                             4-1

-------
VOCs.  Since the focus of Title III is the control of HAPs,
the effectiveness of controlling HAPs is critical.  In all
instances except bulk gasoline terminal loading racks, the
effectiveness for HAPs should be comparable to that for VOC.
This is because all of these technologies involve the simple
capture and/or collection of the vapors (in the case of bulk
plants and service stations), the prevention of vapor
formation (in the case of floating roofs for storage tanks),
or the prevention of vapor leaks from equipment.  A
difference would not be expected in these methods for the
control of HAPs.  The section on bulk terminal vapor
processors contains a discussion specific to the control of
HAPs.
4.1.1  Submerged Fill
     One basic method of reducing vapors generated during
the loading of gasoline into tank trucks, aboveground
storage tanks, underground storage tanks, or any container
or vessel 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 turbulence.  This is compared to splash
loading where the transfer line outlet is at the top of the
tank (Figure 4-la).
     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 4-lb).  In the bottom.filling method, the
fixed fill pipe enters the tank truck from the bottom
 (Figure 4-lc).
     As discussed in Chapter 3 (Section 3.2.1.3), submerged
filling can reduce emissions by approximately 60 percent
compared to splash filling.
                             4-2

-------
          Vaoor Emissions
                 Gasoline
                  Vapors,
Vaoors
                          Gasoline
        Proouct
       A. TOD Sotasn Loading Metnoa
Fill Pine

—— Hatcn Cover
                                         Tank Truck Comoartment
          Vaoor Emissions
      Product
                            u
                            _c
                            "o
                            M
                            a
                            a
        8. Top Submerged Method
        Vapor Vent to
        Recovery Equipment
        or to AtfTKMphere
                         Httert
   C.. Bottom Loading Method
 Fiil Pipe

 —— Hatcn Cover
                                         Tank Truck Compartment
                                        Tank Truck Compartment
                                          Fill Pipe
Figure  4-1.   Gasoline  Tank  Truck  Loading Methods
                            4-3

-------
4.1.2  Loading Racks at Bulk Terminals
     4.1.2.1  Location and Applicability.   Bulk gasoline
terminals are the first key transfer points from refineries
to tank truck distribution.  Loading racks at terminals
allow the metered loading of products from bulk terminal
storage to large transport trucks.  Loading rack equipment
does not vary in type from small to large facilities;
instead, the number of loading positions increases.
     The control techniques described in this section are
applicable to all terminal loading racks.   In addition,
these controls have been used at terminals for many years
and the baseline analysis presented in Chapter 3 (see Table
3-10) estimates that approximately 70 percent of the bulk
terminals will have some type of vapor processor in place in
1990.
     4.1.2.2  Description of Control Techniques.  Emissions
resulting from outgoing transfer operations at terminals are
controlled by two main elements, a vapor processing system
(or vapor processor) and a vapor collection system.  A
simplified example of controls at bulk gasoline terminals is
shown in Figure 4-2.  The vapor collection system consists
of all the piping and components necessary to transfer the
air-vapor mixture from the loading rack and tank truck or
railcar to a vapor processor.  A properly designed vapor
collection system at the terminal should not result in
excessive backpressure at the tank truck or railcar during
loading and should have no vapor leakage during transfer.
It is also necessary that provisions be made in the vapor
collection system to prevent vapor displacement from one
loading position to another.  Check valves are typically
used for this purpose.
     There are three major types of vapor processors
commonly used at bulk terminals:  (1) carbon adsorbers,
(2) thermal oxidizers, and  (3) refrigeration condenser
systems.  All can be monitored for correct operation through
                             4-4

-------
                                              ID
                                             rH

                                              s
                                             -p

                                              o
                                              o

                                             
-------
use of hydrocarbon exhaust concentration or temperature
monitors in lieu of continuous emission monitors (CEMs) that
monitor specific pollutants in an emission stream.   However,
CEMs are used at industry facilities similar to bulk
gasoline terminals to measure break-through on carbon
adsorbers.  Carbon adsorption vapor recovery systems use
beds of activated carbon to remove gasoline vapors from the
air-vapor mixture.  These units generally consist of two
vertically positioned carbon beds and a carbon regeneration
system.  During gasoline tank truck loading activity, one
carbon bed is used for adsorption while the other bed is
being regenerated, usually by vacuum application accompanied
by an air purge.
     Figure 4-3 illustrates a simplified schematic of a
typical carbon adsorption system.  The vapors enter the
active carbon bed through the bottom and are dispersed
upward through the carbon.  Hydrocarbons are adsorbed on to
the carbon, and purified air exits to the atmosphere through
the top vent.  As hydrocarbons are being adsorbed in the on-
stream bed, the other carbon bed is being regenerated.
Regeneration occurs by applying a high vacuum to the carbon
bed using a liquid ring vacuum pump.  Near the end of the
regeneration cycle, an ambient air purge is introduced into
the carbon bed to enhance regeneration.  Hydrocarbon vapors
and condensed hydrocarbon liquids discharge from the vacuum
pump to a separator/absorber vessel.  The liquid collected
in the separator is returned to storage.  Non-condensed
vapors, along with a small quantity of air, flow to the base
of the packed absorber column and rise upward.  Liquid
gasoline from storage is pumped to the top of the column
and, as it cascades downward through the packing into the
separator, absorbs virtually all of the hydrocarbons from
the air/hydrocarbon mixture.  The small amount of
hydrocarbon vapor and air exiting the top of the absorber is
recycled to the carbon bed that is on-stream.  Two carbon
beds are used for continuous service.
                             4-6

-------
    AIR VENT   AIR VENT
GASOLINE
 SUPPLY
                                   RECYCLE
        ^CARBON  ^
        ADSORPTION
           BEOS
INLET VAPOR
                                          ABSORBER
                                         SEPARATOR
                           VACUUM PUMP
     GASOLINE
     RETURN
           Figure 4-3.   A Simplified Schematic of a Typical Carbon
                               Adsorption System
                 (Diagram  Courtesy of the John Zink Company)
                                       4-7

-------
     Manufacturers indicate that most carbon adsorber-
absorber systems on the market can meet the emission level
of 35 mg of hydrocarbon per liter of product loaded, as
specified in the regulations.  One manufacturer estimates
that a carbon adsorption/absorption system can recover
approximately 2 gallons per 1,000 gallons of gasoline loaded
at an average inlet hydrocarbon vapor concentration of 40
percent.1
     Manufacturers also report that they can provide vapor
recovery units using the same technology that will achieve
emission rates under 10 mg/1.  These more efficient units
are equipped with more activated carbon and greater vacuum
capacity to accomplish this additional emission reduction.2
     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 concentration 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.  Typical thermal
oxidation units include elevated flares, enclosed flares,
and temperature controlled combustors (including those
devices where only the combustion air is controlled).
     The elevated flare system typically contains a
combustion unit, special anti-flashback burner(s), automatic
ignition pilot with a continuous monitor, motor operated
vapor block valve(s), flame arrestor(s), an air-assist
blower, a liquid seal, piping, instrumentation and a master
control panel.  Figure 4-4 illustrates a simplified flow
diagram for an elevated flare system.  When not in use, the
vapor combustion system is in a standby mode with no pilot
flame, the vapor block valve is closed, and the air-assist

                             4-8

-------
                             t/t 1*1
      gs
C3

•-i i<:
o <_>
«s «s
o ex.
                                                    w

                                                    Q>
                                                    M
                                                    (0
                                                    •H
                                                    fa
                                                   4* _

                                                    r* °>
                                                    otJ
                                                   r1 3
                                                   fc o
                                                   •ou
                                                    0) B
                                                   H ^'

                                                   fr"
                                                   •Sa
                                                   w

                                                   <
                                                   0)
                                                   -H
                                                   fn
                4-9

-------
blower is off.  The start-up sequence begins with a short
air purge using the air-assist blower to purge the air
plenum of any combustibles prior to pilot ignition.  This
brief air purge is followed by automatic electronic ignition
of the pilot.  Pilot fuel of propane or natural gas is used.
     After the pilot ignition, product loading begins at the
loading rack and an air-vapor mixture begins to flow from
the transports being loaded to the vapor combustion system.
Flow through the vapor combustion system first consists of
the air-vapor mixture from the loading rack bubbling through
a liquid seal.  As soon as sufficient flow is attained, the
pressure monitoring controls automatically open the vapor
block valve allowing the air vapor mixture to flow through
the flame arrestor to the burner, where the combustible
vapors are ignited by the pilot and burned.  Only minimal
pilot fuel is needed.  The gasoline vapor air mixture
provides sufficient fuel to maintain combustion
temperatures.  The air assist blower provides partial
combustion air and mixing energy to the burner tips to
assure smokeless combustion.  As the loading operation is
completed, vapor flow to the combustion unit decreases.  The
pressure monitoring system closes the vapor block valve when
the vapor flow is insufficient to maintain minimum burner
velocity.  If no further loading occurs, the combustion unit
will shut down and return to the standby mode to await
automatic re-start as previously described.
     The enclosed flare operates similarly to the elevated
flare but has the advantage that the flame is totally
contained in a refractory-lined cylinder.  This can help to
minimize thermal radiation and noise.  Figure 4-5
illustrates a typical enclosed flare.
     The temperature controlled flare is generally used if
the combustion temperature has to be maintained at a minimum
temperature or if the waste vapor does not have sufficient
combustible content to maintain combustion.  This system has
                            4-10

-------
                                    §
                                    4J
                                    to
                                    >i
                                    W

                                    0)
                                    14
                                    A
                                    •O
                                    0)
                                    E*
                                    w°
                                    O
                                       0)
                                    BJ
                                       tfl

                                       0)
                                    I1-5
                                    •n8
                                    CO

                                    <
                                    0
4-11

-------
the same features as the enclosed flare with the addition of
automatic temperature control which is accomplished by the
application of quench air and supplemental fuel.  Combustion
air is controlled by dampers to ensure the proper oxygen
content and temperature.  This system also automatically
supplements the waste vapor, as needed, with assist gas
(normally natural gas or propane).  Figure 4-6 illustrates a
temperature controlled flare.
     Refrigeration condenser systems recover gasoline vapors
from the loading operation in the form of a liquid product.
In these systems, the air-vapor mixture from the loading
racks is routed to a condensation chamber and passed over a
series of cooling coils.  Temperatures in the condensation
section can be as low as -180°F  (-118°C).  The gasoline
vapors condense, with some water vapor in the air, and are
separated in a gasoline/water separator.
     In this unit, the vapor mixture is precooled to a water
vapor dew point of approximately 34°F  (1°C) to remove most
of the water vapor.  From the precooler unit, the vapor
enters the condenser where vapor with heavier molecular
weight is condensed and collected.  The design and use of
refrigeration direct expansion condensing coil heat
exchangers permits raising the refrigeration compressor
suction pressure.  This results  in increased capacity of the
unit at a constant condensing temperature.  At periodic
intervals, defrosting the finned surfaces may be required.
This is accomplished by circulation of a warm solution which
is stored in a separate reservoir.  Defrosting is normally
completed in 30 to 60 minutes, depending upon the amount of
frost collected on the finned surfaces.  The warm solution
temperature is maintained by heat reclamation from the
compressor equipment.  There are also multi-stage
refrigeration units that allow the vapor to be cooled to
even lower temperatures.  In these units, refrigerants are
used to cool other refrigerants  that in turn cool the vapor.
                            4-12

-------
                 VAPOR
        PUEL LIKE   LINE

                              Schematic Diagram Showing
                              Incorporation of Temperature
                              Controlled Flare
        VAPOR I
       3URNOU

        /APOR
     DISPOSAL
        RINB
       All OM9CKS
 SHELL
/
 INNER LI.HER
(•

^...JimWCTDJ

.UATCRPROOF HOUSING

  UK 8URNCX
                                            SEAL
                    ARXCSTDR
Figure 4-6.  A  Typical Temperature Controlled  Flare and
                  Simplified  Flow Diagram
       (Diagram  Courtesy of  the John  Zink Company)
                            4-13

-------
Figure 4-7 illustrates a simplified diagram of a
refrigeration condenser system.
     Controlling emissions from railcar loading racks is
very similar to control at truck racks.  The vapor
processors discussed above for truck loading racks are
suitable for controlling emissions from railcar loading.
     4.1.2.3  Effectiveness of Control Techniques.  Vapor
processors for controlling loading rack emissions at bulk
terminals have been in place for about 20 years for the
control of VOC.  The CTG level of control for ozone
nonattainment areas was set at 80 mg VOC/liter in 1977.3
Processors have not experienced difficulty meeting this
level.  In addition, the NSPS level of control for new,
modified, and reconstructed sources was set at 35 mg/liter
in 1983 (40 CFR 60, Subpart XX).  Control device
manufacturers have also not experienced difficulty designing
and manufacturing devices to meet this level.  In the Bay
Area and Sacramento Air Quality Management Districts of
California, the limit is set at 10 mg/liter.  While the
types of control devices that meet this level may be
limited, sources are able to comply with these limits for
VOC control.  Additionally, afterburners may be retrofitted
to existing vapor recovery units that can no longer meet
these specific emission levels.  These combustors are
somewhat different from flares in that they are designed to
destruct an air and hydrocarbon mixture, while flares are
designed to burn only hydrocarbons.  Several plants in
California have undergone this retrofitting operation
(Texaco, Arco, and Santa Fe pipeline) and now meet the
required emission limitations.4
     Table 4-1 contains a summary of test data obtained from
various State agencies including the California Air
Resources Board and the American Petroleum Institute, as
well as data previously gathered by the EPA.  The data are
presented in emission limitation order, from lowest to
                            4-14

-------
  K UJ
  Wff
  ^Lr-
  KZ
  ex
  uZ
	1

oQ
w>-

8**
P ui 5
||S

« w>
  IU
                                   g
                                   *J
                                   U
                                   >i
                                   w
                                   •P  O
                                   10  O
                                   fc
                                   o>  tr
                                   O O
                                   K C
                                     •H
                                   «J O>

                                   0-1 W
                                   O
                                      (0
                                   e-o
                                   (0 h
                                   ti id
                                   t^J*
                                   «J TJ
                                   •H W
                                   Q
                                       «
 I -H
t Q

tt) ^
 4-15

-------
TABLE 4-1.  SUMMARY  OF EMISSION TEST  DATA
FOR BULK GASOLINE TERMINAL VAPOR PROCESSORS
Control
Type
TO
CA
TO
CA
TO/VRU
TO
CA
TO/REF
CA
CA
REF
CA
CA
TO/CRA
CA
CA
CA
CA
CA
CA
CA
CRA
CA
CA
CA
CA
REF
CA
CA
VRUb
CA
Date of
Test
08/22/90
06/01/90
09/29/89
09/20/90
11/30/89
08/30/89
07/12/89
06/29/90
05/24/89
03/08/89
09/06/90
08/10/89
08/09/89
07/26/89
01/30/90
10/23/90
09/08/89
12/15/89
03/13/90
12/20/89
01/04/90
06/20/90
11/29/88
06/13/90
08/08/81
12/07/89
04/12/90
06/04/89
06/15/90
09/19/90
10/26/81
Allowable
Emissions
(mg/l>
10
10
10
10
10
10
10
10
10
10
10
10
10
35
35
35
35
35
35
35
35
35
35
35
35
35
35
35
35
35
35
Actual
Emissions
(n«8/ 1)
0.006
0.06
0.11
0.6
1.1
1.2
1.6
1.7
1.9
1.9
2.4
3.6
4 1.552"
0.12
0.33
0.45
0.5
0.7
0.75
0.9
1.1
1.6
1.6
1.8
1.97
2.1
2.6
2.6
2.9
2.9
3
Source
of
Data
2
2
2
2
2
2
2
2
2
2
2
2
2
2
1
1
1
4
1
4
4
1
1
1
1
4
2
1
1
1
1
                    4-16

-------
TABLE 4-1.   (Continued)
Control
Type
CA
TO
CA
CA
CA
CA
TO
CA
VRUb
CA
CA
CA
CA
CA
CA
CA
CA
CA
VRUb
CA
CA
CA
CA
TO
TO
CA
CA
CA
CA
TO
CA
CA
REF
Date of
Test
04/09/87
03/07/89
07/03/90
02/28/89
07/10/91
NA
09/11/89
06/28/90
06/26/90
05/20/87
02/27/91
03/01/91
05/16/91
03/10/88
02/12/89
10/11/89
07/25/90
06/25/90
07/25/90
03/07/89
06/22/89
06/20/90
09/15/89
07/29/87
03/22/91
05/17/91
02/07/90
06/08/90
12/16/88
10/24/90
05/10/91
06/29/90
09/21/89
Allowable
Emissions

-------
TABLE 4-1.   (Continued)
Control
Type
CA
CA
REF
CA
REF
TO
REF
REF
REF
CA
CA
TO
TO
CA
CA
CA
TO/COM
CA
CA
CA
CA
CA
CA
CA
CA
CA
CA
CA
CA
CA
CA
CA
CA
Date of
Test
06/21/89
07/11/90
03/28/90
05/05/89
06/29/88
07/20/89
03/02/90
03/25/87
05/11/88
12/05/89
07/20/90
12/16/80
01/20/81
09/17/80
09/22/80
02/04/81
05/14/80
01/22/81
02/02/81
02/06/81
10/01/80
"10/06/80
12/02/83
11/14/80
09/26/80
11/12/80
10/10/80
02/11/81
11/13/80
06/06/79
10/01/80
07/10/80
04/30/80
Allowable
Emissions
(ing/ 1)
35
35
35
35
35
35
35
35
35
35
60
80C
80C
80C
80°
80C
aoc
80°
80C
80°
80°
80C
80C
80C
80C
80C
80C
80C
80C
80C
80C
80C
80C
Actual
Emissions
(ing/ 1)
18
18.2
19.7
20.8
25.7
27
29.8
30
33.6
34
0.22
0.2
0.22
0.65
0.66
1.2
1.2
1.5
1.6
1.6
1.8
2.3
3.5
4.5
4.5
4.8
5
5.2
5.6
5.9
6.3
6.7
6.9
Source
of
Data
2
1
1
1
1
1
1
2
2
2
2
3
3
3
3
3
3
3
3
3
3
3
1
3
3
3
3
3
3
3
3
3
3
           4-18

-------
                             TABLE  4-1.     (Concluded)
Control
Type
CA
CA
CA
CA
VRUb
CA
CA
CA
CA
REF
REF
CA
REF
REF
TO
REF
CRA
CA
Date of
Test
01/08/81
12/09/80
06/28/90
05/22/80
07/11/91
10/03/80
09/29/80
10/02/80
05/26/89
05/30/80
03/26/81
07/31/90
02/20/81
11/07/90
10/31/84
12/19/89
04/25/84
10/31/89
Allowable
Emissions
(ing/ 1)
80C
80C
80
80C
80
80"
80C
80e
80
80C
80C
80
80°
80
80
80
80
108
Actual
Emissions
(ng/l)
7.5
7.7
7.8
7.9
8.4
11
15.6
17.9
21.2
21.9
22.6
30.9
41.8
46.6
60.5
69.6
69.8
0.18
Source
of
Data
3
3
1
3
1
3
3
3
1
3
3
1
3
1
1
4
1
2
Sources
Notes
            Test  reports obtained from requests made  to State Agencies.  Data obtained from
            Georgia,  Kansas, Kentucky, Maryland, New  Jersey, New Mexico, Oklahoma,  Tennessee,
            Texas, Virginia, Washington, and Wisconsin, October 1991.

            CARB  Bulk Gasoline Terminal Vapor Recovery System Certifications, October 23,  1990.

            Bulk  Gasoline Terminal Background Information Document, Volume II (EPA-450/3-80-038b),
            August 1983.
        4   American  Petroleum Institute study,  "Determining the Benzene Emission Factor  of
            Existing  Marketing Terminal Vapor Recovery Units," June 1990.
        (a)  Arithmetic average emission rate  for units subject to 10 mg/l  standard.

        (b)  Vapor recovery unit (VRU) type not specified.

        (c)  Allowable emissions not reported.  Assumed that allowable emissions  were equal to 80
             mg/l since most of the tests reported from Source 4 were performed prior to the
             proposal of the NSPS for bulk terminals (December 1980).
NA = Not available.
                                              4-19

-------
highest.  Also provided are the dates the tests were
performed, the vapor control system types (CA = carbon
adsorber, TO = thermal oxidizer, REF = refrigeration unit,
VRU = vapor recovery unit, CRA = compression/refrigeration/
absorption unit, COM = compression unit), and the emission
rate determined during the tests,  insufficient information
was available in the test data that were submitted to
determine the type of flare system tested (elevated,
enclosed or temperature controlled with or without a vapor
holder, etc.).  The test data indicate that control systems
of all three types discussed above easily meet the
appropriate emission limitations and that emission rates
less than 10 ing/liter can be achieved.
     As discussed in Appendix D, it is assumed that 94 per-
cent of uncontrolled loading at terminals occurs by
submerged fill and 6 percent by splash fill.  Using the
submerged fill  (658 mg/1) and splash fill (1,590 mg/1)
emission factors calculated from the national weighted
average RVP (11.4 psi) and the selected temperature (60°F),
the weighted average emission factor for uncontrolled
loading at terminals is calculated to be 715 mg/1.
Therefore, the levels of control discussed above represent
control efficiencies of total VOC of slightly less than 90
percent at 80 mg/liter, 95 percent at 35 mg/liter, and 99
percent at 10 mg/liter.
     The focus of this report is the control of HAPs.  It is
possible that these vapor processors could control HAPs at a
different percent reduction than total VOC.  Therefore, the
effectiveness of each of the three major types of control
devices is discussed below.
     Initially, the effectiveness of controlling HAPs
relative to total gasoline vapors can be considered from a
theoretical standpoint.  As discussed in Section 3.2.2.1,
the major part of gasoline vapors is made up of alkanes with
four or five carbon atoms.   However, most of the HAPs
contained in gasoline vapor are aromatic compounds.  There
                            4-20

-------
are several properties of aromatics that allow their control
effectiveness to be higher than for the alkanes.
     First, it would be expected that both carbon adsorption
and refrigeration/condensation type control systems would
control these aromatics to a level slightly greater than
that for total VOC.  This is because of the higher molecular
weights and lower boiling points and volatilities of the
aromatics.  Conversely, due to the increased bond strength
in aromatic compounds, incineration may control the more
volatile and lighter compounds- slightly better than the
aromatics.
     Specific tests have been conducted to determine the
control device efficiency for HAPs.  Several test reports
from the late 1970's and early 1980's were analyzed to
estimate benzene emissions from various types of vapor
processors.5  This analysis showed that carbon adsorption
and refrigeration systems significantly reduced VOC and
benzene in the vapor stream.
     In a report entitled "Determining the Benzene Emission
Factor of Existing Marketing Terminal Vapor Recovery Units",
dated June 4, 1990, AmTest, Inc.  (for API) described
emissions testing and liquid and vapor sample analyses for
five terminals in the Pacific Northwest.6  The intent of
this test program was to make a rapid determination of the
ability of existing vapor recovery units at bulk terminals
to meet the EPA proposed benzene emission standard (1989) of
0.2 mg/liter.  One control system was a refrigeration system
designed to meet the 80 mg/liter VOC standard and the other
four were carbon adsorption systems designed for the 35
mg/liter VOC standard.  Hydrocarbon emissions from the
adsorption systems ranged from 0.7 to 2.1 mg/liter, while
emissions from the refrigeration system were 69.6 mg/liter.
The average benzene concentration in both regular  (leaded)
and unleaded liquid gasolines was 2.2 percent, while the
concentration in super grade averaged 2.5 percent.  The
benzene emissions averaged less than 0.01 mg/liter, and the

                            4-21

-------
concentration in the system outlet vapors was less than 3
ppm.
     The report also summarized test results from an
independent study conducted by an API member company in
southeastern Pennsylvania.  This testing was conducted
November 14-17, 1989, on four systems described in the
report as charcoal, refrigeration, lean oil charcoal, and
compression.  Hydrocarbon emission rates were 11 to 14
mg/liter for the charcoal systems, and 45 and 152 ing/liter,
respectively, for the refrigeration and compression systems.
Control efficiency for benzene was well over 99 percent for
all systems except the compression type, which controlled
benzene at 72 percent.
     Inlet and outlet vapor samples were also analyzed for
toluene and xylene content.  Toluene control efficiencies
were approximately 99 percent for all systems except the
compression system, which controlled toluene at about 75
percent.  Xylene was controlled at 85 to 98 percent for the
three systems and at about 76 percent by the compression
system.
4.1.3  Storage Tanks at Terminals and Pipeline Facilities
     4.1.3.1  Locations and Applicability.  Gasoline storage
tanks are located at all of the gasoline marketing
facilities with the exception of pipeline pumping stations.
However, the type of storage tank varies considerably among
the gasoline storage and distribution facilities.  This
variation ranges from large external floating roof tanks
having capacities of up to 5 million gallons at pipeline
breakout stations and bulk terminals to underground storage
tanks with capacities of around 10,000 gallons at service
stations.
     The control techniques discussed in this section are
specifically related to the larger storage tanks at pipeline
breakout stations and bulk terminals.  Control techniques
for bulk plant and service station storage tanks are
discussed later in this chapter.
                            4-22

-------
     4.1.3.2  Description of Control Techniques.  Storage
tank emissions arise from breathing losses and from filling
and emptying losses (working losses).   There are two major
types of storage vessels, fixed-roof tanks and external
floating roof tanks.  Fixed roof tanks may have internal
floating roofs as well.  Each tank type has its own
associated emission rate.
     Storage tank control requirements for gasoline storage
tanks have been made by the EPA through control technique
documents.7'8  As discussed in Appendix D,  many States have
promulgated regulations in response to these CTGs for
storage tanks.  In addition, EPA has promulgated NSPS
regulations for petroleum storage tanks (40 CFR 60 Subparts
K, Ka, and Kb) that apply to gasoline storage tanks at
terminals and pipeline facilities.
     A fixed-roof tank is the original, traditional vessel
used 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:
        by the installation of an internal floating roof
        with rim seals; or
     •  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 at bulk terminals and
pipeline breakout stations are most readily controlled by
the installation of internal floating roofs.  An internal
floating roof, regardless of design, reduces the area of
exposed liquid surface to air in the tank.  Reducing the
area of exposed liquid surface, in turn, decreases the
evaporative losses which are the largest source of emissions
for this piece of equipment.  The presence of the floating
roof vapor barrier precludes direct contact between a large
portion of the liquid surface and the atmosphere, thus

                            4-23

-------
reducing emissions.  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 bolted, aluminum, internal
floating roof with a single vapor-mounted wiper seal.  The
four types of losses from this roof design are: (1) rim or
seal losses, (2) fitting losses, (3) deck seam losses, and
(4) withdrawal losses.  Rim or seal losses and fitting
losses constitute the largest percentage contribution to the
total loss from an internal floating roof tank.
     External floating roof tanks do not experience the
fitting losses or deck seam losses that occur with most
internal floating roof tanks. External floating roof tanks
are constructed almost exclusively of welded steel, thus
assuring 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 roof tanks and internal floating roofs is
that the external floating roof seal losses are believed to
be dominated by wind induced mechanisms.9  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.
     4.1.3.3  Effectiveness of Control Techniques.
Available emissions test data10  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 emission rate and thus a higher control
efficiency than a vapor-mounted seal.  A secondary seal,

                            4-24

-------
whether in conjunction with a liquid- or vapor-mounted
primary seal, provides an additional level of control.11
Table 4-2 shows these control efficiencies.
     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 effectiveness of
the various types of seals can be evaluated by analyzing the
seal factors.  These seal factors were developed on the
basis of emission tests conducted on a pilot scale tank.
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.  Table 4-3
presents these control efficiencies.
4.1.4  Tank Truck Leakage
     4.1.4.1  Locations and Applicability.  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 vapor-tight seal is realized when
bottom loading is used.  A 1979 survey12 covering
approximately 1,900 tank vehicles, or about 2 percent of the
gasoline tank truck population at that time, indicated that
22.8 percent of tank trucks had only top loading, while the
                            4-25

-------
       TABLE 4-2.  TANK SEAL CONTROL EFFICIENCIES -
              INTERNAL FLOATING ROOF TANKS3
    Tank  &  Seal Type
% Reduction
 From  Least
  Control
Incremental
% Reduction
Fixed-Roof Uncontrolled
"Least Control"

Internal Floating Roof

     Primary Seal only
     (Vapor-mounted)

     Primary Seal only
     (Liquid-mounted)

     Primary Seal
     (Vapor-mounted)
     w/Secondary Seal

     Primary Seal
     (Liquid-mounted)
     w/Secondary Seal
   93.5%


   94.9%


   95.1%



   95.5%
   93.5%


    1.4%


    0.2%



    0.4%
    Calculated with equations from Section 4.3 of AP-42
    using the nationwide weighted average RVP of 11.4 and a
    temperature of 60°F.
                            4-26

-------
       TABLE  4-3.   TANK SEAL CONTROL EFFICIENCIES -
               EXTERNAL FLOATING ROOF  TANKS3
     Tank & Seal Type
% Reduction
 From  Least
  Control
Incremental %
  Reduction
External Floating Roof

     Primary Seal only
     (Vapor-mounted)
     "least control"

     Primary Seal
     (Vapor-mounted)
     w/weather shield

     Primary Seal
     (Vapor-mounted)
     w/Rim-mounted
     secondary

     Primary Seal only
     (Mechanical)

     Primary Seal
     (Mechanical)
     w/Shoe-mounted
     secondary

     Primary Seal only
     (Liquid-mounted)

     Primary Seal
     (Liquid-mounted)
     w/weather shield

     Primary Seal
     (Mechanical)
     w/Rim-mounted
     secondary

     Primary Seal
     (Liquid-mounted)
     w/Rim-mounted
     secondary
   38.7%



   63.8%




   80.5%


   90.6%




   91.2%


   93.1%



   94.8%




   94.9%
    38.7%



    25.1%




    16.7%


    10.1%




    0.6%


    1.9%



    1.7%




    0.1%
    Calculated with equations from Section 4.3 of AP-42
    using the nationwide weighted average RVP of 11.4 and a
    temperature of 60°F.
                           4-27

-------
remaining 77.2 percent could be either top or bottom loaded.
Although no more recent definitive information is available,
the trend is toward more trucks using bottom loading, due to
State vapor recovery regulations and the advantages cited in
Section 3.2.3.1.
     Tank trucks become a separate source of emissions when
fugitive leakage occurs from the truck-mounted vapor
collection systems and truck compartment dome covers.  This
vapor leakage has been observed to be as high as 100
percent, with an average loss of 30 percent when no regular
leak testing and repair program was in effect."
     4.1.4.2  Description of Control Techniques.  There are
two basic control methods for reducing emissions from tank
truck leakage.  Vapor leakage can be minimized by ensuring
that the tank trucks are vapor tight or a vacuum can be
generated to draw the vapors from the tank truck to the
vapor processor.  Figure 4-8 illustrates the tank truck
vapor collection-equipment.
     There are two methods of ensuring vapor tightness for
trucks, both involving the periodic leak-testing of the
tanks.  The CTG for gasoline tank trucks recommends pressure
limits for an annual test on the tanks and their vapor
collection equipment.14   The  CTG  recommendations  for  vapor
tight tank trucks are that 1) the tank truck must pass an
annual leak-tight test that requires having less than 3" H2O
pressure change under 18" H20 pressure or 6" H20  vacuum;  2)
there will be no leaks greater than 100 percent of the lower
explosive limit (LEL) when monitored at any time with a
portable combustible gas analyzer; and 3) vapor collection
systems back pressure not exceed 18" H2O pressure when
measured at the truck.
     In addition to the CTG level, many districts in the
State of California require an annual leak-tight test with
less than 1" or 2" H20 pressure change rather than the CTG
                            4-28

-------
  OVERTURE
(VAPOR RETURN)
    RAIL
                                                   RUBBER BOOT
                                                      OR
                                                    ETAL COVER
                                                       VENT
                                                      VALVE
                                               OVERFILL SENSOR
                                               DOME LID SEAL
                                            BASE RING GASKET
  TANK SHELL
     Figure 4-8.   Tank Truck Vapor Collection
      Equipment for Bottom Loading Operations
                         4-29

-------
recommendation of 3" H2O.   In addition to this difference,
there are enforcement programs in California that actively
monitor trucks using portable gas analyzers or equivalent
methods. The combination of this more stringent test and
increased enforcement, results in a control level slightly
more effective than the CTG level.
     Recently, the U.S. Department of Transportation (DOT)
has also required an annual leak tightness test for cargo
tank trucks.  According to 49 CFR Part 180 §407 (c), the DOT
test requires all cargo tanks, except cryogenic tanks,  to
have an annual leakage test.  The test specifies that the
cargo tank should be pressurized to at least 80% of the
maximum allowable working pressure (MAWP), which is
approximately 2-3 psi for a typical gasoline tank truck.
Once pressurized, the cargo tank must maintain the test
pressure for at least 5 minutes.  Any valves or vents set at
a release pressure lower than the test pressure are either
rendered inoperative or capped off prior to testing.  Such
valves include the P-V vent under the dome plate assembly
and the vent valve which is connected to the overturn rail.
The DOT leakage test does not include a vacuum test as
specified in EPA's Method 27.  However, the DOT considers
EPA's Method 27 test an acceptable alternative.  The P-V
vents under the dome covers that are capped off during the
DOT test are potential emission points, thus Method 27
testing is needed to make certain that the tanks are vapor-
tight at loading (less than 14 inches of water) and
unloading (less than 6 inches of water) pressures.
     Vapor leakage can also be minimized through the use of
a vacuum assisted vapor collection system.  The system
employs a vacuum source in the vapor return line and
maintains a slight negative pressure at the tank truck
during loading.  The system is designed, through permissive
interlocking, to prevent loading from occurring unless an
adequate vacuum is created and maintained in the system.
This system is in use at a few bulk terminals  in Texas15'16'17

                            4-30

-------
and one of the systems has been operating for over 2 years.
At that terminal, the negative pressure is created at the
tank truck and in the vapor return line by means of a 15
horsepower (hp) blower.17  This system application  for truck
loading racks is relatively new technology and although it
is now employed at only a few terminals, apparently others
are planned.
     4.1.4.3  Effectiveness of Control Technicfues.   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.18  In June  1978  the
EPA conducted a series of vapor leak tests on 27 tank trucks
that were required to undergo an annual leak tightness
     1Q
test.   Tests  were conducted on the tank trucks before  any
maintenance was performed to establish the truck leakage
rate since the last certification.  Evaluation of these data
indicated that the average vapor 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
                                                7ft
atmosphere without reaching the vapor processor.
     The design of the vacuum assist system suggests that
tank truck leakage should be reduced nearly to zero.
Although leakage at the truck is reduced or eliminated, the
vacuum system introduces additional air into the vapor
collection system requiring additional processing by the
vapor processing system.  To the Agency's knowledge, the
systems that are in operation have not experienced any
significant problems either at the processor or at the tank
truck.  However, test data on this system are not yet
available for effectiveness analysis.   Additionally, these
                            4-31

-------
systems are not designed for use without a vapor processor;
therefore, they would not be appropriate at a bulk plant
where a processor is not in use.
4.1.5  Tank Truck Unloading and Loading at Bulk Plants
     4.1.5.1  Location and Applicability. Bulk plants are a
secondary facility in the gasoline distribution system and
are typically located in more rural areas.  Bulk plants have
fixed-roof tanks for storing gasoline and have loading racks
that do the sane job as those at terminals, only on a
smaller scale.  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 EPA developed CTG guidelines for
bulk plants in 197721 recommending  control  alternatives  of
1) submerged fill of outgoing tank trucks, 2) submerged fill
of outgoing tank trucks and vapor balance for incoming
transfer, and 3) submerged fill and vapor balance for
outgoing transfer and vapor balance for incoming transfer.
     4.1.5.2  Description ofControl Technigues.  The vapor
balance system consists of a pipeline between the vapor
spaces of the truck and the storage tank which essentially
creates a closed system allowing the vapor spaces of the
storage tank and the truck to balance with each other.
Figure 4-9 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).

                            4-32

-------
                                 <0
                                 iH
                                 O.

                                 O

                                 •H

                                 O
                                 (0

                                 o

                                 X
                                 10

                                 4J
                                 (C

                                 i
                                 M

                                 4)
                                 O
                                 c
                                 1C
                                 »H
                                 (0
                                 CQ
                                  0)
                                 •H
                                 EL,
4-33

-------
     4.1.5.3  Effectiveness of Control Techniques.   As
discussed earlier, submerged filling of tank trucks can
reduce vapor loss by almost 60 percent when compared to
splash loading.
     The balance system has proven to be effective in bulk
plant applications for both the delivery of gasoline by
transport trucks to the bulk plant and for the loading of
account trucks.  Based upon test data, controls on bulk
plant storage tanks can reduce filling and working/breathing
losses and tank truck loading losses by greater than 95
percent.22^3'24
     Based on the uncontrolled emission rates discussed in
Chapter 3  (see Table 3-9), an emission factor of
54.0 mg/liter was used to represent the balance system
control technology for tank filling losses based upon
95 percent control of the uncontrolled emissions
(1,081 mg/liter).  Emission factors for storage tank working
losses and tank truck loading losses were assumed to be 21.7
mg/liter and 49.0 mg/liter respectively, based upon 95
percent control of the respective uncontrolled emission
factors (tank working losses - 432 mg/liter, truck loading
losses (balance service) - 980 mg/liter).  High efficiencies
are achieved by maintaining the integrity of the storage
tanks, tank trucks, and associated vapor collection systems,
and ensuring that proper connections are made.
4.1.6  Service Stations
     4.1.6.1  Locationand Applicability.  Service stations
are numerous and located virtually everywhere.  Vapor
balance and submerged fill controls for service station
underground storage tanks were recommended in a CTG issued
by the EPA in the mid 197O's.25
     4.1.6.2  Description of Control Techniques.  Emissions
from underground tank filling operations at service stations
have been demonstrated to be reduced by the use of vapor
balance systems  (Stage I control).  In the service station
balance system, vapors which would normally be vented to the

                            4-34

-------
atmosphere are routed back to the delivery truck during
unloading through a vapor collection system.  The truck
transfers the vapors to the terminal or bulk plant for
ultimate treatment by the vapor processor at the terminal.
     Gasoline is loaded by gravity into the underground
storage tanks via a flexible hose.  Liquid gasoline
displaces a nearly equal volume of air partially saturated
with gasoline vapors.  The vapor is routed 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 tank 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 as
shown in Figure 4-10.  To prevent this excess vapor from
escaping into the atmosphere, a pressure-vacuum  (P-V) valve
may be installed on the vapor vent line.  Not only would the
P-V valve prevent leakage caused by vapor growth during
underground tank loading, but such a device would also
prevent breathing losses due to diurnal fluctuations in
temperature and barometric pressure.26'27
     4.1.6.3  Effectiveness of Control Technicrues.  The
effectiveness of the Stage I vapor balance system is
adversely affected by leaks.  Truck hatches must 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.28'29'30  Note that breathing and emptying losses are
not controlled by this method.  These two sources account
for 5 percent of total station losses.  However, by
installing a P-V vent some of this vapor loss can be
stopped.  According to one source, an average 90,000 gallon
per month facility will save 8.3 gallons of gasoline per

                            4-35

-------
                                     c
                                     o
                                     W


                                     8
                                     •H

                                     B
                                     (0
                                     (O

                                     <0
                                      R)
                                      to
                                      0)
                                      o
                                      c
                                      (0
                                      rH
                                      10
                                      (Q
                                       O
                                       CU
                                       to
                                       o
                                       H
                                       I
                                       0)
4-36

-------
month by installing P-V valves on service station storage
vents.31
     In order for the vapor balance system's performance to
be maintained at design efficiency levels, the following
objectives must be met:
     •   assure that the vapor return line will be connected
        during tank filling;
     •   assure that there are no significant leaks in the
        system or tank truck which reduce vacuum in the
        truck or otherwise inhibit vapor transfer;
     •   assure that the vapor return line and connectors are
        of sufficient size (minimum 3 inches in diameter)
        and sufficiently free of restrictions to allow
        transfer of vapor to the tank truck and achieve the
        desired recovery; and
        assure that gasoline is discharged below the
        gasoline surface in the storage tanks (submerged
        filling).
4.1.7  Fugitive Emissions
     4.1.7.1  Locations and Applicability.  Pumps, valves,
and other components capable of leaking and producing
fugitive HAP emissions are present at pipeline pumping
stations, pipeline breakout stations, bulk terminals, and
bulk plants.  The control techniques discussed in this
section could be applied at any of these facilities.  CTG
recommendations and NSPS and NESHAP regulations have been
developed to control fugitive emissions from pumps, valves,
and compressors in both liquid and vapor service, but not at
these specific facilities.
     4.1.7.2  Description of Control Techniques.  There are
basically two approaches to the control of fugitive
emissions from pumps, valves, and other components.  The
first entails a leak detection and repair program in which
fugitive sources are located and repaired at certain
intervals.  The second is a preventive approach whereby
potential fugitive sources are controlled either by
installing specified controls or leakless equipment.
                            4-37

-------
     Leak detection and repair programs use various
monitoring techniques in a leak detection program to
identify leaking equipment.  These methods include
individual component surveys, area surveys, and fixed point
monitoring systems.
     Each component is surveyed on a periodic basis.  There
are two common methods of conducting this survey.  These
include 1) leak detection by spraying each component with a
soap solution and observing bubble formation, and 2) leak
detection by measuring VOC concentration with a portable VOC
detector.  Another method is to perform visual inspections
of each component to detect the evidence of liquid leakage.
     The area survey entails walking through the area
measuring the ambient VOC concentration within a given
distance of all equipment located on ground and other
accessible levels.  This is conducted using a portable VOC
detection instrument utilizing a strip chart recorder.
Fixed point automatic hydrocarbon sampling and analysis
monitors can also be placed at various locations.  The
instruments may sample the ambient air intermittently or
continuously.   Elevated hydrocarbon concentrations indicate
one or more leaking components.
     The detection of a leak is only the first step in
reducing emissions from leaking equipment.  The emission
reduction depends on prompt and proper repair of the leak or
replacement of the component.
     An alternative approach to controlling fugitive
emissions from these components is to replace them with
leakless equipment.  There are various types of so-called
leakless equipment.  These include dual mechanical seal
pumps, sealless or canned-motor pumps, and closed-vent
systems with control devices.
     4.1.7.3  Effectiveness of Control Techniques.  The
control efficiency achieved by a leak detection and repair
program is dependent on several factors, with the most
critical being the inspection interval.  This interval is
                            4-38

-------
related to the type of equipment and service conditions, and
different intervals should be specified for different pieces
of equipment.  Monitoring may be scheduled on an annual,
quarterly, monthly, or even weekly basis.  Monitoring may
also be scheduled for a skip-period approach where less
frequent monitoring is allowed for components that achieve a
specified level of performance.  Estimated control
effectiveness for leak detection and repair programs for
pumps and valves is shown in Table 4-4.32
     The installation of improved shaft sealing mechanisms
can reduce emissions to a negligible level, and can be
eliminated entirely by installing sealless pumps.  Also, the
installation of closed-vent systems with control devices can
be expected to achieve efficiencies of greater than 90
percent.32
                            4-39

-------
TABLE 4-4.  ESTIMATED CONTROL EFFECTIVENESS FOR LEAK
 DETECTION AND REPAIR PROGRAMS FOR VALVES AND PUMPS
    Monitoring Interval
 Control Effectiveness
       (percent)


   Valves
Light Liquid    Pumps
   Monthly
     59
61
   Monthly/Quarterly
     46
   Quarterly
     44
33
   Source:   Reference 30.
                           4-40

-------
4.2  REFERENCES

     1.   Letter from Matthes, B.,  John Zink Company to
          Tedijanto, M.,  Pacific Environmental Services,
          Inc.  September 19, 1991.  John Zink vapor control
          equipment.

     2.   Telecon.  Tedijanto, M.,  Pacific Environmental
          Services, Inc., with Tuttle, N., John Zink
          Company.  October 7, 1991.  The achievable
          emission levels for John Zink carbon vapor
          recovery units.

     3.   Control of Hydrocarbons from Tank Truck Gasoline
          Loading Terminals.  U.S.  Environmental Protection
          Agency.  Research Triangle Park, NC.  Publication
          No. EPA-450/2-77-026.  October 1977.

     4.   Telecon.  Tedijanto, M.,  Pacific Environmental
          Services, Inc.  to Tuttle, N., John Zink Company,
          January 15, 1992.  Monitoring and control devices
          at gasoline loading terminals.

     5.   Pacific Environmental Services, Inc.  Description
          of Analysis Conducted to Estimate Impacts of
          Benzene Emissions from Stage I Gasoline Marketing
          Sources.  Prepared for U.S. Environmental
          Protection Agency.  Research Triangle Park, NC.
          August 1989.

     6.   AmTest, Inc.  Determining the Benzene Emission
          Factor of Existing Marketing Terminal vapor
          Recovery Units.  Redmond, WA.  June 4, 1990.

     7.   Control of Volatile Organic Emissions from Storage
          of Petroleum Liquids in Fixed-Roof Tanks.  U.S.
          Environmental Protection Agency.  Research
          Triangle Park,  NC.  Publication No. EPA-450/2-77-
          036.  December 1977.

     8.   Control of Volatile Organic Emissions Petroleum
          Liquids in  External Floating Roof Tanks.  U.S.
          Environmental Protection Agency.  Research
          Triangle Park,  NC.  Publication No. EPA-450/2-78-
          047.  December 1978.

     9.   American Petroleum Institute (API).  Evaporation
          Loss from Internal Floating-Roof Tanks.  API
          Publication 2519.  Third Edition.  June 1983.

     10.   Reference 7.
                            4-41

-------
11.   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.  September 30, 1991,
     pg. 4-13.

12.   Hang, J.C. and R.R. Sakaida, Pacific Environmental
     Services, Inc. (PES).  Survey of Gasoline Tank
     Trucks and Rail Cars.  Prepared for U.S.
     Environmental Protection Agency.  Research
     Triangle Park, N.C.  Publication No. EPA-450/3-79-
     004. March 1979.   p. 3-15.

13.   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.

14.   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 No. EPA-450/2-78-
     051.  December 1978.

15.   Air Permit for the Diamond Shamrock, Inc. bulk
     terminal, Laredo, TX.  Texas Air Control Board.
     June 8, 1992.

16.   Air Permit for Navajo Refining Company's bulk
     terminal, El Paso, TX.  Texas Air Control Board.
     June 16, 1992.

17.   Telecon.  LaFlam, G., Pacific Environmental
     Services, Inc. with Saitas, J., Texas Air Control
     Board.  November 30, 1991.   Permit information on
     bulk terminal vacuum assist system.

18.   Reference 13.

19.   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).

20.   Norton, R.L., Pacific Environmental Services, Inc.
     (PES) Evaluation of Vapor Leaks and Development of
     Monitoring Procedures for Gasoline Tank Trucks and
     Vapor Piping.  Prepared for U.S. Environmental
     Protection Agency.  Research Triangle Park, N.C.
     Publication No. EPA-450/3-79-018.  April  1979.
                       4-42

-------
21.  Control of Volatile Organic Emissions from Bulk
     Gasoline Plants.  U.S. Environmental Protection
     Agency.  Research Triangle Park,  NC.  Publication
     No. EPA-450/2-77-035.  December 1977.

22.  Pacific Environmental Services, Inc. Compliance
     Analysis of Small Bulk Plants.   Prepared for U.S.
     Environmental Protection Agency Region VIII.
     Denver, CO.  December 1976.

23.  Evaluation of Air Pollution Regulatory Strategies
     for Gasoline Marketing Industry.   U.S.
     Environmental Protection Agency,  Office of Air
     Quality Planning and Standards.  Washington, D.C.
     EPA-450/3-84-012a.  July 1984.

24.  California Air Resources Board.  California
     Perspective on Controlling Gasoline Evaporation
     Emissions.  Mobile Source Division, #SS-86-01.
     March 1986.

25.  Design Criteria for Stage I Vapor Control Systems,
     Gasoline Service Stations.  U.S.  Environmental
     Protection Agency.  Research Triangle Park, NC.
     November 1975.

26.  Facsimile from Kunaniec, K.,  Bay Area Air Quality
     Management District, to Norton, R.L., Pacific
     Environmental Services, Inc.   September 23, 1991.
     Effectiveness of pressure vacuum vents.

27.  Telecon from Kunaniec, K., Bay Area Air Quality
     Management District, to Norton, R.L., Pacific
     Environmental Services, Inc.   November 22, 1989.
     Vent line restrictors.

28.  Norton, R.L., R.R. Sakaida and M.M. Yamada,
     Pacific Environmental Services (PES).  Hydrocarbon
     Control Strategies for Gasoline Marketing
     Operations.  Prepared for U.S.  Environmental
     Protection Agency.  Research Triangle Park, NC.
     Publication No. EPA-450/3-78-017.  April 1978.

29.  Reference 22.

30.  Reference 24.

31.  Reference 2 8.

32.  Fugitive Emissions Sources of Organic Compounds —
     Additional Information on Emissions, Emission
     Reductions, and Costs.  U.S.  Environmental
     Protection Agency.  Research Triangle Park, NC.
     Publication No. EPA-450/3-82/010.  April 1982.


                       4-43

-------
        5.0  MODEL  PLANTS AND REGULATORY ALTERNATIVES

     This chapter presents a description of the model plants
used in the analysis to represent facility populations in
the United States in the 1998 base year.   These model plants
are used in the estimation of the impacts of implementating
the regulatory alternatives developed to reduce hazardous
air pollutant emissions.  Section 5.1 presents the model
plants for pipeline facilities, bulk terminals, bulk plants,
and service stations.  Section 5.2 discusses the regulatory
alternatives for each emission source.

5.1  MODEL PLANTS
     This section presents model plants for each of the
gasoline distribution industry sectors.  Varying sizes of
facilities within each source category were selected to
represent a cross-section of the total industry.  For each
source category, model plant characteristics are provided
with a description of the design parameters for each.  Also,
a riationwide profile using the model plants is presented by
distributing the total number of facilities across the
various model plants.
5.1.1  Pipeline Facilities
     The pipeline facility model plant parameters for
pipeline pumping stations and breakout stations are based on
information collected from industry representatives,1 and a
search of the literature.2'3'4
     5.1.1.1  Pumping Stations.  As discussed in Chapter 3,
pipeline facilities are a major element in the distribution
of gasoline between the refinery and the bulk terminal.  The
emissions at pipeline pumping stations are attributed solely
                             5-1

-------
to leaking pumps and valves.  The emission factors (Section
3.2.2.1) and control costs for these components (Section
7.1.2) are based on the number of components at the facility
and are not related to facility throughput.  Therefore, the
only parameters necessary to define for the model plants are
the number of pumps and valves at the facility that are in
gasoline service, and the operating schedule.  Any pump or
valve that will handle gasoline is considered to be in
gasoline service.  The pump or valve does not have to handle
gasoline on a continuous or dedicated basis to be considered
to be in gasoline service.  Therefore, any pump or valve at
a pumping station that periodically handles gasoline will be
considered in gasoline service.
     Pipelines may occur as single pipes or in clusters of
two or three pipes.  The smallest pipeline pumping station
model plant represents a single pipeline facility and has
two pumps and 25 valves.  As with all pipeline pumping
stations, the facility operates 24 hours a day, 365 days per
year.  The second model plant represents a facility with two
pipelines and has five pumps  (two of which operate on one
pipeline and three that operate on the other) and 50 valves.
The largest model plant represents a facility handling three
pipelines and has nine pumps  (three per pipeline) and 100
valves.  The model plant parameters for pipeline pumping
stations are shown in Table 5-1.
     The 1998 baseline estimate for the pipeline pumping
station population is 1,989 facilities  (as discussed in
Section 8.2).  Data reviewed  indicated that it was not
unique to have a facility handling one, two, or three
pipelines.  However, no specific information was available
to determine relative percentages of single, double, or
triple pipeline facilities.   Therefore, an equal
distribution of pumping stations across the three model
plants was assumed.
      5.1.1.2  Breakout Stations.  As noted above, pipelines
often occur in clusters.  At  some point along the path, one,
                             5-2

-------
TABLE 5-1.  PIPELINE PUMPING STATION MODEL PLANT PARAMETERS
                                  Model Plant Number

 Design Value	1	2	3_


 Number of Pipelines             1           23


 Number of Pumps*                2           59


 Number of Valves"              25          50        100


 Operating Schedule
  hrs/day                      24          24         24
  days/year                    365        365        365


 Percentage of  Total            33%        33%        33%
 Facilities

 Number of Facilities           663        663        663



9 In gasoline service.
                            5-3

-------
two, or all three of the lines branch off in different
directions.  When this occurs, the throughput to any one
line is altered.  Breakout stations are used at these points
to temporarily store gasoline or other products until
compensation can be made for the altered flow.  As discussed
in Section 8.2, the baseline population of facilities where
lines branch in different directions is estimated at 120
facilities.
     At times, the diameter of connected pipes in the
pipeline will be reduced or increased.  This causes a change
in product flow rate between the different sized pipes.
Breakout stations are again used to store gasoline in these
situations.  The baseline predicted population for this type
of facility is 150.  Combining both types of facilities
results in an estimated 270 total breakout stations in the
United States in the base year.
     These two situations dictate the sizes of the two model
plants used to develop pipeline breakout stations.  The
model plant to represent break-out stations that occur when
two or three pipelines split has 15 storage tanks, 35 pumps,
and 400 valves.  As discussed above, there are an estimated
120 of this type station, or 45 percent of the total.
     The model plant developed to represent breakout
stations where the throughput is affected by changes in
pipeline diameter includes 10 storage tanks, 20 pumps, and
250 valves.  This model plant represents approximately 150
facilities, or 55 percent of the total.
     It is important to note that products other than
gasoline are sent through pipelines and stored at breakout
stations.  Product is stored temporarily and the tanks may
not have product in them all the time.  Therefore, all
tanks, pumps, and valves are not in constant gasoline
service.
     Since the emission factors for storage tanks, pumps,
and valves are on a per-tank or per-coaponent basis in
constant gasoline service, utilizing the numbers of tanks
and components cited above would overstate emissions and
                             5-4

-------
emission reductions attributed to gasoline operations.
Consequently, adjustments were made to reflect the number of
tanks that are in gasoline service.  This was accomplished
by assuming a certain number of "equivalent dedicated tanks"
for gasoline service.  This does not signify that specific
tanks are dedicated to gasoline and never used for other
products.  Rather, the "equivalent dedicated tank" reflects
the equivalent number of tanks that would be in constant
year round gasoline service.  These equivalent tanks were
determined by multiplying the number of tanks by the percent
of time gasoline is stored.
     A fraction of the total number of pumps and valves at a
breakout station is associated with the pipeline itself and
functions in the same manner as those pumps and valves at
pumping stations; i.e., pumping product down the pipeline.
There is also another fraction of pumps and valves
associated with storage tanks.  For those associated with
storage tanks, the "equivalent dedicated" concept was again
applied.  The bases for the "equivalent" dedicated value
concept were observations made during a site visit to a
facility5 and subsequent  conversations with industry
representatives.  The parameters for pipeline breakout
station model plants are shown in Table 5-2.
     The tanks typically used at breakout facilities are
external floating roof tanks (76 percent of the total; see
Section D.I.2.1) with capacities ranging from 1,600 to
16,000 m3 (10,000 to  100,000 bbl).   The tank size assumed in
the analysis for gasoline storage tanks at breakout stations
was 8,000 m3 (50,000  bbl)  with a diameter of 30  meters (100
ft) and a height of 12 meters (40 ft).
5.1.2  Bulk Terminals
     5.1.2.1  Tank Truck Loading.  The bulk terminal source
category has been studied for over a decade by EPA.  Model
plants for bulk terminals were originally developed during
preparation of the bulk terminal CTG document and were
further investigated and conclusions documented in the
development of the new source performance standards (NSPS)
                             5-5

-------
           TABLE 5-2.   PIPELINE BREAKOUT STATION
                   MODEL PLANT  PARAMETERS
                                            Model Plant
                                               Number*
 Design Value
Breakout Station Information
Total Number of Storage Tanks
Total Number of Pumps
Total Number of Valves
Number of Storage Tanks
Storage Tank Volume
m3
bbl
Number of Turnovers/tank/yearb
Operating Schedule
hrs/day
days/year
Percentage of Total Facilities
Number of Facilities
10
20
250
10
8,000
50,000
150
24
365
55%
150
15
35
400
15
8,000
50,000
150
24
365
45%
120
 Parameters Used to Estimate  Emissions

      Number of "Equivalent Dedicated
      Storage Tanks" in Gasoline Service     4         5

      Number of "Equivalent Dedicated
      Pumps" for Storage Tanks in
      Gasoline Service                       3         4

      Number of Pumps Associated with        5         6
      Pipeline

      Number of "Equivalent Dedicated
      Valves" for Storage Tanks in         160       200
      Gasoline Service

      Number of Valves Associated with      50       100
	Pipeline	                  	

* Model Plant 1 represents those stations at pipeline
  branches and Model Plant 2  those stations at pipeline
  diameter changes.
b Turnovers per year based upon assuming three turnovers per
  week for 50 weeks per year.
                             5-6

-------
for bulk terminals (promulgated as 40 CFR Part 60, Subpart
XX).  In addition to the NSPS rulemaking, the same model
plant sizes were used in subsequent regulatory development
programs.6-7-8  During these regulatory development programs,
EPA received no significant comments citing problems with
these parameters.  Therefore, after evaluating the industry
in 1990, this document will continue to use these historical
model plant sizes.  However, while the parameters have
remained the same, the population and distribution of these
model plants were modified to reflect 1998 base year
conditions (see Chapter 8, Section 8.2).
     The data base for determination of the original 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, data from EPA's National Emissions Data System
(NEDS), 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.9
     5.1.2.2  Storage Tanks.  As discussed in a previous
bulk terminal model plant analysis,10 a typical terminal has
four or five aboveground storage tanks for gasoline, each
with a capacity ranging from 1,500 to 15,000 m3 (9,400 to
94,000 bbl).  Most tanks in gasoline service have a floating
roof to prevent the loss of product from tank "breathing and
working."  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 tanks are most readily
controlled by the installation of an internal floating roof.
A set of model plant parameters was 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),  a value 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
                             5-7

-------
as typical values for a tank of this capacity.11  In
addition, it was assumed that storage tanks at terminals
were subjected to 13 product turnovers per year (based on
previous analyses).12
     The model plant parameters are shown in Table 5-3.
This table also provides the 1998 base year characterization
of the bulk terminal industry as distributed across these
model plant sizes.
     5.1.2.3  Railcar Loading.  Information was sought from
industry representatives, literature, and trade associations
concerning railcar loading of gasoline.  Little information
was obtained; however, one facility that loaded gasoline
into railcars was visited.13   In addition, railcar loading
of chemicals was studied to determine the applicability of
filling technology.14  This information was used to develop
a single model plant based on the parameters at the single
gasoline loading facility, although it is estimated in the
model plant analysis that there will be 20 such facilities
in the base year.  The model, or typical, plant parameters
are described in Table 5-4.
     It is assumed that a terminal that loads gasoline into
railcars also has truck loading racks.  Therefore, no
separate storage tanks or pumps were attributed to railcar
loading racks, which avoided double counting emissions.  In
addition, it was assumed that the railcar loading racks were
located at a significant distance from the truck loading
racks and that separate vapor piping and vapor processing
equipment would be required.
     A very small portion of the total gasoline transported
is moved by rail and this occurs at only a few facilities.
As discussed in Section 8.2,  it is estimated that there are
20 terminals in the United States that load railcars.  Due
to the lack of information on additional facilities and the
small number of total estimated facilities, all are assumed
to be represented by the single model plant.
                             5-8

-------
vj




W
B
E^ Kl

if
K
PH
u
i *
rtj z

* I
>1 *
w 7; iv
a |
o I
£
rf
2
H
W
E-i
W
2 -
H

O
w

CO
n
1
in

CQ
1
I
°«
§O »* CM 13 O P *"*
O ^ w E r*» o
o a 5 (v o
» •» i. *> ««
Ii Is ™~
KT«-^ CO L.
O
fr
1—

II •" * II §1
§O tt ** fNJ W
5 SI
IA £ CD
—^ £» fc.
0
H-




§O M O* TJ ^* O <**
0 « E Ko
O O> O (M O

O O Af 4J CNJ vx
in IA E o
O- OJ £ CC
"fc
|
v



§1 ~ * li Si
- •* i- *-• •»*<)
O O At +•* f^J v>
00 O E O

0

|





?l E e
^•o g g
i*. s < .?
Jg * 5 1
M 1 1 1 *lf
SLa: -J +• *^"-&
t™* OI «) QJ "^ OJ
^w -s -s * -^^
•| «.!-??
I Hit
i- z * _J a.
^3 ^s O ^3 ^^ O ^3 ^K ^B
(MO K) O K) <* O
«\j - h^ » in »
K» CO OJ N- O* O
s



o***o o*%o P *"* "^
W S S C 5 R
^^






ru o %» o KI So

IV* • O " Kl" •
KI a cv in iv o
<^ v^ in
^





O <— O O<~-O O^sKI
rj « KI • o -
KI eo «- KI »- in
**^ *-* ^O
**




o
^"* 4**
Sffl
a
i c" fe «>
•*• *r- 0> C
s -g £ x =
« 2 SI -t 8
E b u a
•- g> "89 u
t- 3 »rf a.
X O Q L.
4-< ^ O> S JJ U O
3 ^* w *^> 10 .20)
ffitAC-n^ "poi (O
H.fc-5 o C__ * « _ M.
o"-1 o aOOf «- ** C
u^-i -i ^SS o O*>ri 0
•^0 «« & ^ en *X3 4 ^
j<^0> A< £--01 «° to E Z ^
fj^^^rf o fc-^^^' n s^v^a.
33 C V 0
u t. E »p- c
1- t- 3 *J — L.
^ .1 So |
i i g 8. S 1
h- 1- X O <3 Z
Kl O O O O O O
•- «\l «- ^ O »- Kl





Kl *• 00 O fc 1
«- •- K» a IM •*








Kl «O O in O Kl K.








Kl Kl O O O O (V
«- «- O. «- •* •-







i 1

c. a.
51 *
8u S - — Q.
•o L. *• x:
t- V V r+ •*• O>
2L S — — 1
O <» n 'a *u t.
« i *• « c a f
1- — ' C H- M- U. t-
$ g 1 r E - -
o •- 5 m — t « «
C E B- _ ?. — ** v> *J
i. u f « ^ 0 O o
3 V O B- > « — • h- 1-
*- *- W E 7; u- a
3m 4-» •»-»*•
H- H- •*-<*•> H- O O O
O O O O «->
4* **
C. &. L. i. ^ C C
Hi H- 1 i
z z z z •— a. a.
5-9

-------
    TABLE 5-4.  RAILCAR LOADING BULK GASOLINE TERMINAL
                   MODEL PLANT PARAMETERS
             Design Value                       Model Plant
	Parameter


 Throughput
      (million liters  per year)                          322
      (million gallons per year)                          85

 Number of Loading Arms                                   3

 Loading Method                                   Submerged
                                            (Top or Bottom)

 Pumping Rate/Loading  Arm
      (1pm)                                           3,800
      (gpm)                                           1,000

 Railcar Capacity
      (liters)                                       110,000
      (gallons)                                      29,000

 Number of Railcars Owned/Leased by                      30
 Facility8

 Maximum Instantaneous Loading Rate
      (1pm)                                          11,350
      (gpm)                                           3,000

 Number of Facilities                                     20

 Total Throughput
      (billion liters)                                  6.2
      (billion gallons)                                 1.6
8  It is assumed that all railcars are dedicated to gasoline
  service and owned/leased by their terminal owners.
                            5-10

-------
5.1.3  Bulk Plants
  As described in Section 3.2.4,  bulk gasoline plants are
secondary distribution facilities within the gasoline
distribution network.  Model bulk plant parameters were
developed and utilized in connection with earlier guidance15
and environmental impact studies.16'17*18  An analysis of the
conditions of the industry in 1990 indicates that these
basic parameters still adequately represent the industry,
with one exception.  Bulk plants that store and transport
aviation gasoline were not included in earlier EPA studies.
These facilities are generally located at airports, and
store and move gasoline by truck to aircraft located in
various parts of the air terminal.  Information obtained
from the National Air Transportation Association19  indicates
that the basic parameters described for gasoline bulk plants
are generally representative of these aviation gasoline
facilities, except that the estimated average throughput for
an aviation bulk plant (1,500 liters/day) is considerably
less than that designated for the smallest model bulk plant
(11,350 liters/day).  Therefore, an additional model plant
was added to represent aviation gasoline bulk plants.  All
of these model bulk plant parameters are shown in Table 5-5.
  As delineated in Table 5-5,  the typical bulk plant
facility includes tanks for storage of gasoline, loading
racks, and incoming and outgoing tank trucks  (account
trucks).  Regardless of throughput, it is assumed that all
bulk plants have the same numbers of tanks, loading racks,
and account trucks.20  Larger model plants simply load more
trucks per day than the smaller model plants.  The typical
bulk plant utilizes two relatively small aboveground 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 the top-splash
method or a top-entry submerged fill pipe.  Since the number
of pumps and valves is usually determined by the number of
storage tanks and loading racks, the estimated number of
                            5-11

-------





U)
K
w
i
B.
{•7
5
W
ft
j
w
Q
O
S
||
2
ft

w
55
0
1
If

1-3
CQ
IT)
1
in
w
»
IS













,_
.D
1

1
£
_
I























in









K>







CM

















1
C
O>
(A
O
OO O O IM •- CM O •* O O t-. O O i >
in ^7 o o C3 in C3 C3 ^3
-^r*T irTo" «-^K^
vi »- F. IM .- '
0 0
in 0
KIO
* •
^3 «~
OO O O CM »- f\) O *^ O O CM O O • *
oo in o o m o CM §2
N-*fC sfh."1 A? pTiw
t t

& 0
S"
§O O O IM «- fM O >» O O O Q O ' '
in CM o KI co o *<•
«»• O ,- O K| 5- O- •«
•- K» in -» CM O CM
I O
Sin
«0
in



OO OOCM^-CMO NT O II ii OOO
OOOIA o in oOh~
«- IM o ro «-
i
o





<~ w
B) L. tl 0)
s 1. 5 g
I-M ~-«Col ^»t-
«« >> u t- a i~ a
n * ^x to o o t) jc aoi
SO C T3 u- 5 >. — 4)X
o CD v^ m w w ^ *"Si ^ 10 *>» ^
3 «»«»-2i S«~£§^*^£S

ra X Q C > O u T3 > W Q. •- » — « _, — —o>o o ->cn
O-D-NS tt T> -N. »* o 1- 0 O — BO « B
i- v, oo
1- 1. O 3 L. O •»-•*- •*- O)H- « H- O 3 •- — s- 3 •— •—
t) — B. o — o o o c o to ti « o o.— -co a.— —
tl 4-> — • X *• -. "- Q. > C V f — — O £ — ' ->
o»—« OI---B L. i. b *J b E — • •«• b c a *^ •>- — b » ••• «s
a _ a> 5 — 01 «) u OJ a 41 3 « — V 5 5 E E w vgEE
in in



^ ?$




S 12
fi 1*1








CM •-






in IM
CM








8 5
= f
'i ^
u. t-
"S "c
o o
Percentage of 1
Percentage of 1
5-12

-------
these components is also constant for all model plants.
Therefore, the only difference among the model plants is the
volume of gasoline handled by each facility.
     Transport trucks supply bulk plants with gasoline from
bulk terminals, while account trucks are used to deliver
gasoline to bulk plant customers.  Bulk plants typically
average two account trucks.  These two trucks are usually
privately owned by the bulk plant owner.  While the basic
specifications of the model plants have remained constant,
the distribution of the bulk plant population across the
industry has been updated to reflect 1998 base year
conditions (see Chapter 8, Section 2).  This distribution is
also shown in Table 5-5.
5.1.4  Independent Tank Truck Facilities
     The trucking industry generally consists of two major
groups, private and for-hire.  Private carriers are defined
as those firms that transport their own goods in their own
trucks.  An example of a private carrier is an oil company
that uses its own tank trucks to move gasoline from its
terminals or bulk plants.  For-hire carriers transport
freight that belongs to others, renting out the hauling
services of their trucks.
     As discussed and documented in Section 8.2, it is
estimated that 81,300 tank trucks will be used for the
movement of motor vehicle gasoline in 1998.  This estimate
is based on an earlier EPA study of tank trucks21 and was
adjusted to reflect the expected 1998 base year population.
While adjustment of the population was necessary, no more
recent information was located concerning the category
distribution of tank trucks, either private or for-hire
(independent ownership).  This earlier study assumed that
about 31 percent of the gasoline tank trucks were used at
bulk terminals.  The remaining 69 percent were therefore
assumed to be associated with bulk plants.  However, there
has been a significant decrease in the percentage of
gasoline handled by bulk plants from the time period of the
1979 tank car study (27 percent) to the 1998 base year  (18
                            5-13

-------
percent).  To attribute the same fraction of tank trucks to
bulk plants probably overstates this portion greatly.
Therefore, the percentage of tank trucks estimated for the
1998 base year associated with bulk plants was decreased
from the 1979 study by a proportion equal to the decrease in
throughput for bulk plants (18/27).  Consequently, the
updated percentage of bulk plant trucks is estimated to be
46 percent of the total tank truck population.
     The remaining 54 percent of the total tank truck
population is attributed to bulk terminals, which represents
43,900 vehicles in 1998.  This number comprises 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, 37,400 vehicles,
are smaller tank trucks used primarily to transport motor
vehicle gasoline from bulk plants.
     As shown in Tables 5-3 and 5-5, parameters for the
model bulk terminals and bulk plants are predicated on the
fact that a certain number of tank trucks are owned by the
model plant owners.  Based on this information, it is
estimated that of the total number of terminal tank trucks,
7,200 are bulk terminal trucks and 18,800 of the total bulk
plant trucks are owned by the model plant owners.  The
remaining 36,700 bulk terminal trucks and 18,600 bulk plant
trucks are assumed to be "independents."  This information
is summarized in Table 5-6.
     In addition, there are account trucks associated with
aviation bulk plants not included in the earlier estimates.
As shown in Table 5-6, it is estimated that there are 6,400
of these vehicles.  It is also assumed that all of these
vehicles are privately owned.  Therefore, the total  1998
nationwide tank truck population  is projected to be  87,700.
5.1.5  Service stations
     Service stations, as defined in this document,  include
motor vehicle refueling operations that receive revenue from
                             5-14

-------
        TABLE  5-6.  CHARACTERIZATION OF NATIONWIDE
                   TANK TRUCK POPULATION
 Type/Owner of Tank Truck	Population	


 Total Nationwide Tank Trucks"                87,700


 Bulk Terminal Trucksb                        43,900

     Private                                 7,200

     For-Hire (Independent)                   36,700


 Bulk Plant Trucks                           43,800

      Private0                               18,800

      For-Hire (Independent)6                 18,600

      Aviation Bulk Plant Trucks'1             6,400


8 All trucks are assumed to have four compartments.

b 71 percent of the trucks assumed to have vapor collection
  equipment installed (see Appendix C).

c 60 percent of the trucks assumed to have vapor collection
  equipment installed (see Appendix C).

d Assumed no trucks have vapor collection or bottom loading
  equipment.
                            5-15

-------
either the sale of gasoline (public retail outlets) or that
service government, commercial, and industrial fleet
operations (private outlets), excluding agricultural
refueling operations.  As opposed to counts made by the U.S.
Census Bureau that include only those outlets that derive 50
percent or more of their dollar business from petroleum
products, miscellaneous retail outlets that were considered
service stations for this study include convenience stores,
mass merchandisers, marinas, parking garages, and others
that obtain less than 50 percent of their revenue from
gasoline sales.
     In addition to "public" outlets, there are a
significant number of "private" facilities included in this
subcategory.  These outlets are maintained by government,
commercial, and industrial consumers for their own fleet
operations.  Government agencies with central garages
typically consist of regional locations for the U.S. 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.  As noted previously, the agricultural
sector of private outlets which includes farms, nurseries,
and landscaping firms, etc. was not included in the study.
     As for bulk terminals and bulk plants, there have been
model plants developed for service stations in connection
with previous EPA studies.22'23'24  While recent data indicate
that facility distributions may be different in metropolitan
areas, the distribution used in previous EPA studies is
believed to be representative of the nationwide facility
distribution.25  The  service  station model plant category
parameters were originally derived from size ranges used by
the Bureau of the Census, total facilities reported for
197726 and  198227,  and the total consumption of gasoline
(excluding agricultural)  for each year.28
                             5-16

-------
     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 liters/month (10,000 gallons/month) .29-30  The
remaining 10 percent of private facilities which had
throughputs greater than these amounts were distributed
among model plants 3 through 6 in proportions representative
of the public service station distribution.
     The model plant parameters developed for EPA's 1984
model plant scenarios were basically well received by
industry during the associated comment period.  However,
there was one alteration made in the 1987 analysis document
in the service station model plant section that was based on
comments received from the industry.31  The pertinent
comments were related to the throughput amount of gasoline
at private stations; i.e., that the 5,000 gallons per month
average used in the 1984 document to represent approximately
190,000 private stations in model plant 1 overestimated the
nationwide throughput that would be exempted by a 10,000
gallon per month cutoff.  Therefore, model plant 1 was split
into two separate model plants with different average
throughputs.  These revised model plants and their design
parameters are retained in this analysis.
     Design characteristics for the six model plants are
presented in Table 5-7.  The 1998 base year nationwide
distribution discussed in Section 8.2 is also provided in
this table.  In addition to the private facilities that are
represented by the smallest model plant, this analysis also
includes 1,600 aviation facilities that fit the description
of service stations (i.e., private airplanes pull up to a
dispenser and fill their tanks).  The monthly throughput for
these aviation facilities places them in the model plant 1
category.  However, the average monthly throughput for these
aviation facilities is slightly higher than the 7,600 liters
indicated.
                            5-17

-------







1
EH
i
PH
EH
§
U
PH
•J
Q
i

o
H
EH
EH
W

W
U
H
W
w




1
in
W

CQ

£_(















L.

Z
4-*
C
CD
CX
1
X


















CO
"5
o
* tO ^ O»
e s ° fe
»- Kl «M
og £g |:| o.g
8 *~ A TT *^^ * *
iN» *•• CO O*
*~
°S K§ 8"8 §-§
5 ii » » <• »
CM <>O «M •» .» «-
•O in to O ^


om oo oino otoo
CM «- i %>• eo •* t~-
10 i m - - • •
«- in CM to en rw «-
O ** *O •""



oo inin oino o^o
• CM OCM OCMO OO
K) «- «*" cs" eo* I*-"
•* •»
\


O ^O CO O O O O ii
• K\ T- O  .
«- in O.




%oro om o»— o ooo
• «- i in v  —
CO " — • ~* C £ _•
•2 £ e o 'Z •— c
^ S 2lc s «t 5 ^
Sal fl s S'S'SI " 8*5 "Sl^
|-a jba | |||f | lillf;
inn £nn C 3 u +• Z fc 3 o *< ~ (5 <
1.OO OO (ii D.I.1.C M Q.l_C0t.l>O
?»-«- «'-'- to OVO.C oO4JOjr«-
p w ^- 5 w- ^- v' a. a c. >- - a.o.i/>a.»-^'
a : cv
tl £ U 4->
01 a — a
co 3 — * >
fc S -8 r
> £ a. a.
< i- r s
OJM 00 g
>S W Pw •" ^
s
g- 5
M
o
IM
g«> ^
Kl


i O Kl CO
§^ -
O
in



• o •* «-
g" -
in



00 Kl
Ifi
ID
m



o rv o o- -*
o K- in sf

^ «.
*~ O^






>c
o
^
M *^
Si -•
1 1 1
•" 2 ~* ? ••• 5 £ °
L "3 3 *> ,tj "3 o °
S 5- c *^ ^ 5- "- — •
M p £— — $ y *• «
Ck^~^ (i (LiL 34^
SB fi. 0
U. £ 4->
co co o o
"> 0 £ *«
< t- t- <~











2
§**

B>
|
03
01
1
«
IM
^
4^
|

8
'E
a>
(A
C
•^
CO
i
CO
t-
o
jc throughput
a
I

5-18

-------
5.2  REGULATORY ALTERNATIVES
     The purpose of this section is to describe and develop
regulatory alternatives from the emission source and control
information presented earlier in Chapters 3 and 4.  The
purpose of this development is the establishment of
alternatives to present the evaluation of the environmental,
energy, and cost impacts.
     In the formulation of regulatory alternatives for the
gasoline distribution industry, the determination of those
facilities that would be classified as "major" is paramount.
Using the emission factors and HAP to VOC ratios discussed
and documented in Chapter 3 (Section 3.2.1.1), the
uncontrolled emissions for normal, average type, and
reformulated gasoline at each model plant were calculated
and are presented in Table 5-8.  These uncontrolled annual
emissions as well as MTBE emissions from reformulated and
oxygenated  gasoline (presented in Table 5-9) were used to
make the major/area source estimations for each subcategory
facility.  These annual emissions were based upon model
plant average throughputs and a range of total HAP contents
from normal to reformulated gasoline (4.8 percent minimum to
16.3 percent for reformulated and oxygenated gasoline with
MTBE) as described in Tables 3-1 and 3-2.  To test for
individual HAP criteria, MTBE was chosen for analysis
because it makes up the greatest individual component
portion of the HAP vapor profile for reformulated and
oxygenated gasolines.  As shown in these tables (Tables 5-8
and 5-9), only bulk gasoline terminals and pipeline breakout
stations would be classified as encompassing major HAP
sources.  All of the other subcategories of the gasoline
distribution network would be considered area sources.
     Various combinations of control options were examined,
ranging from control of all emission sources at both major
and area facilities to control of only major source
facilities.  A cost effectiveness analysis was then
performed to eliminate the inferior options  (those with
                            5-19

-------
CO
55
O
H
W
CO
(X
J

IS

8
^
<
H
B
5S
W
EH
&

(4
W
Q

§
CO
 I
in
m

o

*— **
!8
_j ^ G.
Ill Jc «•"
i
CO

 h— Jir 3: v-

£ z
1

X °*

£ •*
	

o
o'
^_ *~
»»*
s *"*
°" > °
I_J d. •
Is -
eo








03

"

KJ
ro

.,


J



«— O f\ fO
«- ro >d o>
^ ^


r*- o K» ro

*-
fO O O 03

o ro «— in


•— w
(A C
(DUO
is |Ss
1 I 1 ;»«
8 I 5 HI
UJ *•»
3^1
UJ Q (D
2: 5 £
a o- m
(M eo ro to

v± CM in <\i
is •- o
«•*

CO CO «O CM
N- oo ro o
in R.
ro co«o rs.

ca ro
«- CM c> ro
rJ o in s!
«*• «— in



o o in sr
eo ^ ro o
cv ro


•o «- in r\j

IM PO •— N.



eo N- -O «-
*O N^ NJ O^
•- INJ


>O Kl OJ «-

•- in 10 «-
O fO *^ ^

in (M •- eo
to

o c
a o
•£ 'c 1 '«
^ CO UJ W
eo H-
O 01 S
-1 t> > IU
"> * 0«-
1 SSS'S
2 >— c/> u_ H.
UJ
1
o ro eo «-

^o^ro

K CM K) ^^-
— 1: E «
*D eo UJ «
CO t- •—
O HE
— j cy > uj
O) .p-
j* eg 4-> --
y L. ~- a
«2 ES§^
2 r- «/> u. h-
_j
a.
i
ro

o

CM
O

^.
O


0




^
£}



^_

o
V


,_
o
V


,.

?
,_
C)
V


I
i =
H !
to •-
Ul
u
ee
UJ
u
                                               5-20

-------
•8
 O
 c:
 o
u
CO
 I
in
CQ
                     g-   s
g
                      •
                                                                                     in S. co o

                                                                                     o! o «^ in
                                                                        K in« in

                                                                        »- C) «-^ IO
                                                                        co c\j *o >o

                                                                        o o o «-^
                                                                                          oo

                                                                                          o
                  oj

                  o
                                                      CM •* co •»

                                                         Sin in o*
                                                         •-    co
                                                      o  O


«- O «- Kl
                                                                        •O IM O vl

                                                                        o es o f^
•*

o
                                  u   w    a
                                  <   0    w


                                  Ul         M

                                  2   g    8
                                  _i   •-   ^
                                  IU   Q.    O
                                                                      gco to o
                                                                      J< _ .-
                                       IB*-—"    t   ^«**» —
                                       t. •>-  ej   S   u c. «-   u. t-   J

                                                 £                <
                                                 Ul                -I
      i           s
      o           «
      ~- »         o



  ||||        I

  l    f E   (A   -^
  i ft) IX til   5?   *•—
   S'v -   S   "•

  iSS-BS   5    c
   *-• 5 o   p    a
   in u- i-   Z   *-
                                                                          5-21

-------


CQ
S5
O
H
CQ
CQ
H
S
W
CU
<
X
•J
1
H
£>
r^
H
Q
a
H

c

X
prf
E
E-t
5
Q.
H4
ij
W
Q
O
g

o\
1
in
r_T
3
(Q
g







*o
H- CO
 •« ™ «*•
:• i £
* x
«25
!s -
-iO £ g
RS 1-5
5s i«^
o _. £ .«
gsr g 1
ass s §
uj 5. *
vi a -v
S>- M
X C
tort O O
 t— "
J.i"' 5 «
U ^ UJ *
P. S: BD CA
_J UJ ** |_ «"
* t _. E .«
ii S a
sl i
O- tt *-
2g
* CM

2 £ "
w < u, S
3 s!« «
_i*:.2
UJ E
o u

•
»— CO
X C
3"j.£
SSa
UJ 1
§UJ












r<^



O> inro h-  ft
•* «^ >» C5 »^ C3 «^ M O
IM «- v»






•o to in «x N- N. r\j KI rg rj
to r\j » K! in oo«-^fO o





O^ ^ to ^ f^J tft t^B *^ *^ *~ fO OO *~
•- raM oo«-«- o
»- r\j ro ^t y




oo y> K to -* h- •* in «- o 10 <* «-
o *«»•* c\jinto«- oo»-»- o
«- »- «M V



« 12 e» g iS
••-« oc •— w •~a S
CO C CD C  w « o o te w o
'i 'S •£ "S 'i '« -S c "E '5 e
UJIA •D«U1« •OBU1W —
^ Wl-.^ X»_M. Z|_._ _,
E g I «5£ 3*5i5 J«5J s S
ill re- s ^g'^- *s=- § "•
c_] AJ a U*~C9 ^ UUk^-Q UI-*^fD i_ ^
2 B 5 oaw 5 5oaw M 5oa*' < c
£ 5 M V3O « l- *- 3 O " uv3O * S
5> u.»- E i- in u. i- 2 i- -
uj 4J oe <
z n 3 uj _J iu
j .- M " °- H
s: 1 s a ^ 5
— 3 c- 3 3 uj
o. Q. co m m v>














r*
rvi
t
to

S
4->
<9
U
to
j;
4^
L.
O
**•


«
U)
1
o.
£
0}
s
4-»
•s
4-*
£
y
&
«J>
S
o>
IU
m


•
5-22

-------
higher costs for the same or lesser emission reductions).
The alternatives that remain are termed Alternatives IV-Q,
IV-M, III, II, and I.  Alternatives IV-Q and IV-M are
variations of Alternative IV.  Alternative IV-Q includes a
quarterly monitored leak detection and repair (LDAR) program
for equipment leaks (either pumps or valves) at major source
pipeline breakout stations and bulk terminals.  Alternative
IV-M specifies a more stringent monthly monitored LDAR
program for equipment leaks, as well as other equipment leak
requirements (same as requirements in 40 CFR 60 Subpart V)
at these same sources.  (There are additional provisions for
reducing the monitoring frequency of valves to quarterly).
     Alternative III includes control at all bulk terminals
and pipeline breakout stations.  Finally, the remaining two
alternative control levels  (II and I) require control of all
subcategory facilities within the network.  Tables 5-10
through 5-16 summarize the regulatory alternatives developed
for each industry sector.
                            5-23

-------

X
i

w
H ?
a -
1
<
REGULATORY
IV
w =
H "
^
H
<
W -
M
U
&<
H
CU
15 •—
U

o
r-i
10 «j J
i-3 * "o
ffl Si
g 1
Emission
source
1
*•»
SgS
J4I
±
<

5
**
w <
s "
o TO u
— o>5 * £"»« 3
Efjn-k
$
u a
0 C
1 £ ""
en <
> 1
gg^ g|_
V) <  <
< •*- o
3 § ? 1
±- -»5-2r*,«
i_ o c — xonow
I Illlslll
?
— • "P- «1
0 4J JC —
3 a *j ID
Ul O .- v «•»
— X TJ W 0
> •*- C H-
o —"c * C°
•- aa x a o«
TJ CD C *J U T)
0 -* c. 9 C «°
••- O 41 »*- E O O.
U 41 *• g — 0 W)
41 JC X 8 U 4> Z
a. u ui t. & w ^
c
a
El ol "5 a 4> a o
3|*'| fr t> L. ts +•
Q-lcnl u — i co 10 co
1 .?
4-» *I>
JO V V
2 i£
< i» <

$ 5
Is Is
<•- «
^ L-4vO.(Ac. fgu
w^4>o ^O 4^EL.
XCwS-gc." C3
•^25CiSl ItSS
> .?
4^ 4^
4j 41 ft 41
CO < CO <
41 41
So to g
I- •£-
flj 5> C) 4)
CO < * <
*. fe 3
o — i or o
o X S < "O — '
•*- t. L. g 0 C
o c. a _i a X
O4^— 'DE^V &- -^
c. •- a C •— a 41 co x o  u a co
. I C — t- t.«l — • •— > 41 41 41 41
T) C.4WQ.(AaU £OU4W1-U
iZw — s-wuiou) znataatn
?
ji a —
c a
a — s- 01 3
4^ a Q — ' ^N a
M- £ u S " '>
"O "" *^ fi CO O J*
t> jz 9 E a. — o
X *-i O •— CO L. 4>
•~ — — i- x 41 j:
u. X <*- Q.%^ o. u
ui — i
5-24

-------


X
1
~

i

f£ O-
1 *
SH
«
g s
§
§
W
W ~
H £

H
H
U

H ~

i"*
£**
CU
H
m
O
55
H _
EH
W
H
X
W

•
r~t *u

H o> J
I .E -J
go
a -i
§
E-I

° s
09 -
UJ




1

4J
CO 10
« C >


V
^
0) 0) *"*
"4- n
0 Ol
u c u
J= 40 >» 3
•— OI4J «_ 4-> O
E* •£ U U -0 * **
4V «3 CZ CO U
oi n ^e E o •— o
4J o c — o « •*->
x •- « u o> o> s
uj H- 4-1 Q- co en E

01
7;
n 10
10 E~
i 5
, J= ID >> CO C (-
t- IOC— >>IBIOO>COOIO
0) C— »C-T) l-OI'-'DCO
4-> U4-)O>COUJ=(-
Cr uj>»-4->a.cncoioco' — 'D-io
Ol
c:
•— "- W
10 4-> jr •—
3 «O 4-1 10
CO O •— Ol —
—• •— X T3 W O
> -.O
0 •— C fc-
— ns re >> 10 «9
"O CO C 4-» t- ~O
O Jt C- 
••- O O) **- E O d.
I. 01 4-> 0 -5 0 CO
01 -C X O U Ol Z
O- u UJ c- Q. en — •>

JkT
c
10
*> t—
10 C 4-> CO
Ol CZ O) 3 C 0)
CI O E O O Ol
••-I-'- B. co j* ••- «g
OJ4^ f- ^: (O 4-» C-
El 10 3 03 Ol 10 O
514-1 O" 01 t- 4-> 4->



1

10 If)
JO)""*
5

4)
-?
ir
o «
u- u u u
4,* El",i" "
1-M5 S-D-;^
•- 10 c •— c cu a
U. 4-> ••- "4- 10 CO E

Ol
~
CO 10
10 C «-
[-
S -2
Crt <

01
CO 10
«O C «-"

O) 0)
tt r~*
CO <

u
14-0 I- . —
O ••-> O CO
O >> (O *f~ 01
it- u %- E o
o j: ^_  10
^D t 4-t ^ CO 10 O 4^ 0)
— loc'— coicoo-ct-
U_4->-'-ct-fOcn— >*- CO
4-« 10 O •— —
c o 10 co
14- U C. 01 I—
o oi co u
O 4J OJ
c- c c >,<«
T3 4-> a CO
oi jz « E d.
X 4-> O -r- CO
•- •— •— «- z
U- X <4- Q.— '







4J
a
CO
i £
_) 3
3>°
£15
li
u
•S
>) g
III




a>
^i
V) 10

(0 U

i- O C
Q) """S V
1 s s
S-B fe



IO
3
W
Periodic v
checks


•M
L

&1
UJ — 1
5-25

-------
                  01
                  >
                  C- I
                ]) U

                Is
                                          HI ID
                                          •o c :
                                        •> a

                                        " Ei
                                        o «,
                II
                           in «
                           « c :
JO)
s
                                            I
                                            4J
                                          CO 10
I
w
0)
,. e
O 3
-V t-
cS o
¥
m 1
01
.*
in 10
10 C *— '

a> 01
§0)
•o o
c *-
J= « X 3
•— OI4-" C- 4-> O
10 £ •£ X IO « VI
i- 4J « c in u
01 re ^ g o — o
4-> O C ••- o nj '1-5
x •— 10 «- oi oi s
UJ <*- 4-» Q. Ifl 10 E
0)

4-*
in 10
10 C 1-1

OI 01
0 OI
st e
0 J= S3
O 4-> •— D> E •"-" O
u — « c -5 10 •>
1 * C -- 1-
Tb u 4-> D. ei t-
Ol -^ OI 10 t— O
X C 4J O 13 "0 i->
— 10 C •— C B S
u_ 4-> •— M- w w E
01
>
in 10
10 C •-.
i-
01 OI
CO
s.& w
loi!
» It- to
OI
£
in 10
10 C
u •—
V 01
1
X o to
t— i-> ai
JZ g 0
J*> o
•0 in
0)
.*
in 10
10 C
u <_•
m OI
                                                          to <
55
H
S

ta
E-« "


M

m


W
S5
i-l
1
if)

w •-
kJ ?

< ~~*
E-i g "o


c



Is
V) L
1 *



in
o •—
t- (- ^-
O 3 C7I
in oi to
•— in t. oi
•v» u ro 10 o
& -iTD «- 3
S c o o
in E 10 u- w

GO
a>
i- " >^
O 3 C3)
 "D s- 3
•0 C O O
10 E 10 14- co

.2

c
l_
— ^ o «
CO H- OI
0. — <—
Z 10 «-
-— CJ

•S. V^ ~-

IO O •-
O •-! O
8 |

8 •— K
•S I! i1
*i i
3 ai o
CD|I— _i





en
ID

i
10





Externa


OI

4-1
IO

,
IO
c
OI
4->
X
UJ
CO
a
t—
V
OI
10
u
o
4-1
CO


OI

••"
10
c
t-
Ol

<


o
o

c
4^
8
(t-




.c
4->

-g
10










01

•^
in 10 in
10 c: i-i a
u
| a, g

co«t en
u
*4- O 4-1
o -^> in
o x s ••-
M- U I- E CO
§J= m co
4-» '^ OI E 4-* IO IO
— 10 c -5 n) ai in
i * c — l- u oi E
•a i- *J a. M ~ >- 3
XC4-"O-aiOT33 O
— lOC'-COICO 10
U_4->-*-44-IOfiOIOIO >



C in
j^ 10 m
C 01
10 i — «+- in c
4J 10 O i— — 4->
EO 10 O J^
i- OI t— O)
§01 in o —
4J tjl 4-*


•O +j n en n
0) J= n E Q. 3
X 4-> £ •- CO C
U. X M- O->— ' <£

_^
^
t.
1—
1
(—


OI

•^
to
c

Ol
4^

"*


+j
(- OJ
O •*-> IO
*^> tt) W
m *— i_ a>
g g « U
U C U 3
O C O O
U- IO U- M






CO
a.
CO
X

4->
CO
0)
**



1
«
OI
—I





in
a

i

en


g
s
_>,
4J
i



_
10
3
CO
u


o
L-
01
a.


c
i
o-
UJ


01

"-
m
c

OI
4-1

<
a:

2
- >,
u •—
O U IO
i-l OI 0)
it 55
10
IO O^ »









en
o
01
-C
o



in
1
_l
                                              5-26

-------




^
»— 1



W
EH |r

H ~
W
S

JH
O
EH
^
D
w ~



.J

™»
S
H
•§

W
E-1
S
§
t?
ss
HI

CO
H •-
X
w

H
1 _
«" B $
W •£ -j
3 S u
rt« m •£
^ 1



c
O 01

in i-
V) 3
18
LU






m
a
i
CO




«

(-
0

10
C 3»
lj~
«t

01
4-J
ID
1*
CO
a
e
3
o
in
u
¥


01
44
IO

&- >— 1
5

«z


in i- *-~
Ol O in
o • — **- oi
t- -~. O
3 gl ^W C <-
(0 0) W O 1 O
in u o) u "O en
t- M 10 o 4J ai
O «- C « 10
•*~l *O t- 3 O IO 0)
gc o o o -c u
10 u- v> *—* a. m



0>

"D
10 in
a)
I- U
o t-
g O
IO
c
t-
i in — ^ o in
4-» IO CO U- 01
10 01 a. — f—
c t- co •— 3
O 

S-^ h- in
10 cj kn o •*-
10 4J >— - CO ** £3












Ol
U3
in n

l!~
to <

0)
1 .
in 10
* E2
U V
CO <
i*- in
§AI
•g s
£ <0 >> 3
£4! * fel * "
01 10 je e o •— o
4-> O C »- O <0 ••->
x i— 10 t- 01 01 a
UJ >*- 4-» Q. in W g


01
^
M m
10 C •— i

1-2

CO <




Ol
4-1
ID C
1- l-y
O) O)
E 4->
H) I —

in


0 oi T3 in o in
O X) M C 01 U- Ol
u, c > u in c u
«oc"->.ioooo i o
C-- X u-O"->«i-TJin
U +J 10 C a 4-> 0)
4-*oc->"u oio

g

s^ s~
• — 2 "O in o
«- c t-
-* ID >-,LJ
' — C (-
ID IO >t IO «C
C 4-> t- T5
L. 10 C CO
0) «*- E O Q-
4-> O — O CO
X O 1- 01 K
LU t- D. W -^
in

c
<0
1—

01
Dl
IO
o
CO



01
u
in a
" E2
SO)
44
""*

1
44
m n
* E2
1) a)
I ^
to «£
•4- «l


V^lff «!
o> .x ai « i— o
•*— _•
o> 01
E 4^

CO i re o "*- 01
It- <- 1- t. E M U
o Oi 10 en c i-
t-cc— ioa>"o7o
i **-•*- i- u 5- ^o in
"D 4-» a. in 10 4J oi
ai j: 10 •— c in 10
X4->O~DIO'OOIOO>
.^»-r-coicc>J;t-

T3
c
* "
10 » — n- in
44 10 O •— " -~
c o 10 u>
H- t- (- 01 t—
o oi in o
O 44 Ol
i- C C >,<*>
i — »- i_
"O 44 « CO
Ol £ « H Q-
X 44 O -^ CO
— -^ i— t- Z
Lu. k I*- O.^-*














fl>
«
i
CO



«
«
1
to

u
ID
>
«
3
C
C







U
IO
CO






in
IO




L-
O
O.
10
^
"i
c
c

in
in
Ol
c
4-1
-C
44

C
c





jj

>-
^
a
>—



ai
U
g
1
<

S
44
IB
C
(.
01
44
<



44
5-


01
4-1



S

<




01

a
c



in
Ol
4->
in
in
01
c
Ol
4-*


u>
C->

in
Ol







01
Ol
10
Ol
_J





»






>.
t-H



1
m
01
















"



"D
C
10
1-
o
"->
i
L.
o






























n
8

w.
•ft





















u
<-
g
n
a
01
(-
A3
















44
ID

O£
<
_l
^
I



»_*

£
I









»


i
IO
CO






in
10




o
_J
>>
t-
01
10
3


*J
m
u

TI
Ol
O-




^J
c
!
3
O"
UJ

VI
Ol
o
i
m
|

o



i







Ol
~
IO
c

a>
44

f




01
4J
Alterna




c

t>
o
••-9
i







1
U






in
1
_J












|





















«




01
u

s
o

a

















5-27

-------
§
                                                  10 C

                                                  tu tu ""
                                                         10  C
                                                            t- *
                                                         g«
H
 I
If)
i—     C
 10     O
.0 4-> —
    3 ->  CU
   2    O)
 IO —  X
:>  S   *J  O
 w  in ••-
                               u  E  x
                               c  a. «;
                               a -^
                              •—  3  O t—
                               re  o- c o
                              oo  a) — - —
                                                  v    cu o "
                                                  •<->    x   -1
                                                  in    oj >    •—    i
                                                  in -w    v -
                                                     c -D
           t> to t- in
           U E ,
                                 «
                         .       T3
                    O •-  C  1- ^
                    —  3  10  3 •—
                    •0  D" 4->  O  
                                           -  S1
                                           ^io
                                            10  
-------
                                                                                   >
                                                                                   <->
                                                                                CO 10
                                                                                a c
in
                                                                                i
                            1C    VI
                            •—    c

                            1C    O
                            J3 *J "-

                               3 4-»


                            fe.8  t
                            Q.*J  01
                             "«• t
                                                                    1
                                               a> 0} i- «  >>
                                               O § 10 0)  10
                                               c B.-O o ~o
                                               

                                        *S
                                        c. 10
                                        «3 01
              L in

              i-S
                                                                          5-29

-------
w
in
ffi
             IS
                                - -M


                            •5Jf
   k-  E
tl V  V
o E  x
C 5. 8) ^~
10 ••-    C5

•— 3  O I—
a cr c o
m o> -——'
 e^
 Ol 
>i   4) Q-
•> ** c W
   C -Q £
0) « 3 X
u E   5i
 O I  X
 c a. o) —•
 «J •»-    U3
•— 3  O I—   •— 3 +J •— t—
   So- c o    10 tr-^ •- o
   « ——   m u * >•- —-
Ol
•o c
c —
3 i—
en
Service
Station:
o *~
OIU-
•S-M
5^5
                                                               5-30

-------
5.3  REFERENCES

      1.   Memorandum.  Thompson,  S.  H.,  Pacific
          Environmental Services,  Inc.,  to Shedd, S. A.,
          U.S.  Environmental Protection Agency, Chemicals
          and Petroleum Branch.   March 27, 1991.  Trip
          Report for Plantation Pipeline,  Greensboro, NC.

      2.   American Petroleum Institute.   Introduction to the
          Oil Pipeline Industry.   Third edition.  Austin,
          TX.  1984.

      3.   Kennedy, J.L.  Oil and Gas Pipeline Fundamentals.
          Tulsa, Oklahoma, PennWell  Publishing Company.
          1984.

      4.   True, W. R., U.S. Gas Pipelines Improve
          Operations, Want to Expand.  Oil & Gas Journal.
          pp. 41, 44.  November 26,  1990.

      5.   Reference 1.

      6.   Evaluation of Air Pollution Regulatory Strategies
          for Gasoline Marketing Industry.  U.S.
          Environmental Protection Agency.  Research
          Triangle Park, NC.  Publication No. EPA-450/3-84-
          012a.  July 1984.

      7.   Draft Regulatory Impact Analysis:  Proposed
          Refueling Emission Regulations for Gasoline-Fueled
          Motor Vehicles - Volume I, Analysis of Gasoline
          Marketing Regulatory Strategies.  U.S.
          Environmental Protection Agency.  Research
          Triangle Park, NC.  Publication No. EPA-450/3-87-
          OOla.  July 1987.

      8.   Pacific Environmental Services,  Inc.  Description
          of Analysis Conducted to Estimated Impacts of
          Benzene Emissions from Stage I Gasoline Marketing
          Sources.  Prepared for U.S.  Environmental
          Protection Agency.  Research Triangle Park, NC.
          August 1989.

      9.   Bulk Gasoline Terminals -  Background Information
          for Proposed Standards.  U.S.  Environmental
          Protection Agency.  Office of Air Quality Planning
          and Standards.  Research Triangle Park, NC.
          Publication No. EPA-450/3-80-038a.  December 1980.

     10.   Reference 6.

     11.   Graver Tanks. Commodity Storage Tank Product
          Literature.  Undated.
                            5-31

-------
12.   Reference 6.

13.   Memorandum.  Norwood,  L.P.,  and Thompson,  S.H.,  to
     Shedd,  S.A.,  U.S.  Environmental Protection Agency,
     Chemicals and Petroleum Branch.  June 18,  1991.
     Trip Report for Mobil  Oil Gasoline Terminal,
     Albany, NY.

14.   Memorandum.  Norwood,  L.P.,  and Thompson,  S.H.,  to
     Shedd,  S.A.,  U.S.  Environmental Protection Agency,
     Chemicals and Petroleum Branch.  June 18,  1991.
     Trip Report for a chemical loading terminal.

15.   Control of Volatile Organic Emissions from Bulk
     Gasoline Plants.  U.S. Environmental Protection
     Agency.  Research Triangle Park, NC.  Publication
     No.  EPA-450/2-77-035.   December 1977.

16.   Reference 6.

17.   Reference 7.

18.   Reference 8.

19.   Memorandum.  Norton, B., Pacific Environmental
     Services to Colyer, R. and Steve Shedd, U.S.
     Environmental Protection Agency.  January 10,
     1990.  Trip Report to Piedmont Aviation Services.

20.   Pacific Environmental Services, Inc.  Study of
     Gasoline Vapor Emission Controls at Small Bulk
     Plants.  Report to U.S. Environmental Protection
     Agency, Region VIII, Denver, CO.  Contract No. 68-
     01-3156, Task Order No. 15.   October 1976.  p. 3-8
     through 3-14.

21.   Hang, J.C. and R.R. Sakaida.  Survey of Gasoline
     Tank Trucks and Rail Cars.  U.S. Environmental
     Protection Agency.  Research Triangle Park, NC.
     Publication Number EPA-450/3-79-004.  March 1979.

22.   Reference 6.

23.   Reference 7.

24.   Reference 8.

25.   Technical Guidance - Stage II Vapor Recovery
     Systems for Control of Vehicle Refueling Emissions
     at Gasoline Dispensing Facilities.  U.S.
     Environmental Protection Agency.  Research
     Triangle Park, NC.  November 1991.

26.   Lundberg Estimates.  National  Petroleum News,
     September 1983.

                       5-32

-------
27.   "Franchising in the Economy,  1981-1983".   U.S.
     Department of Commerce.   January 1983.

28.   National Petroleum News.   Factbook Issues.  Mid-
     June 1978-1983.

29.   U.S. Department of Commerce.   1977 Census of
     Retail Trade.

30.   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, NC.  Publication
     No. EPA-450/3-78-029.  July 1978.

31.   Evaluation of Air Pollution Regulatory Strategies
     for Gasoline Marketing Industry - Response to
     Public Comments.  U.S. Environmental Protection
     Agency.  Research Triangle Park, NC.  Publication
     No. EPA-450/3-84-012C.  July 1987.
                       5-33

-------
            6.0   ENVIRONMENTAL AND ENERGY IMPACTS

     The purpose of this chapter is to discuss the
environmental and energy impacts associated with the
gasoline distribution regulatory alternatives presented in
Chapter 5, Section 5.2.  Although most of the discussion
will be concerned with the methodology used to generate the
quantitative analysis of air pollution emission impacts, an
analysis of other environmental and energy impacts of the
regulatory strategies is also included.

6.1  AIR POLLUTION EMISSION IMPACTS
     Estimates of the HAP and VOC emission reductions that
could be achieved under each of the regulatory alternatives
were made and are discussed in this section.  The potential
emission reductions achievable in the base year (1998) were
calculated for each industry sector.
6.1.1  Methodology
     Methods used for calculating emission reductions for
all sectors of the industry were basically the same.  As
discussed in Chapter 3 (Section 3.3) and Appendix C, the
nationwide gasoline throughput and/or facility population
were apportioned to categories representing the 1998
baseline control level.  Nationwide baseline parameters
(throughput or facility population)  were presented by
control level for all emission sources in Table 3-11.  These
parameters were then multiplied by the appropriate emission
factors to estimate baseline VOC emissions.  HAP emissions
were calculated by applying HAP to VOC ratios.  (Differences
between the HAP percent reduction and the VOC percent
reduction come about due to differences in vapor pressures

                             6-1

-------
and consequent evaporation rates in the individual compounds
that make up each chemical population).
     In order to estimate the air pollution impacts of the
regulatory alternatives, the facilities that would be
affected by each of the alternatives were identified.  Then
the control level associated with each alternative was
chosen, and its associated controlled emission factor
multiplied by facility throughput was used to estimate the
VOC emissions that would occur under that particular
alternative.  For example, the nationwide throughput at bulk
terminal loading racks was divided into six categories:
those having controls at  (1) 80 mg VOC/liter, (2) 35 mg/1,
(3) 10 mg/1, and (4) 5 mg/1; and uncontrolled loading racks
that utilize (5) splash or  (6) submerged loading.  The
baseline emissions were calculated by multiplying the
throughput for each of these control levels by the emission
factor for that level.  The emission reductions were
determined by subtracting the emissions calculated for each
alternative from the baseline emissions.  Emission
reductions would occur from all of the baseline control
level groups except those already at levels specified by
each particular alternative.
     Numbers of "new" facilities in each subcategory were
estimated based on industry sector growth, facility trends,
and estimated equipment life as discussed in Section 8.2.5.
Table 8-27 provides a detailed listing of new, replacement,
and existing facilities in the gasoline distribution
network.  For purposes of this analysis, a replacement
facility is one that will be built or rebuilt during the
period from 1993 to 1998  for replacement of worn-out or
obsolete equipment.  Furthermore, it is assumed that one-
half of these replacement facilities will qualify as
"existing" while the other half will be classified as  "new"
units.
     The HAP emission reductions were determined by
multiplying the VOC emission level and resulting emission
                             6-2

-------
reduction by the appropriate HAP to VOC ratio.  As discussed
in Chapter 3 and Appendix C, there are seven area HAP/VOC
scenarios that show varying total HAP vapor contents.  This
analysis is discussed in Appendix C, page C-14, and is
summarized in Table 6-1.  Gasoline throughput and facility
populations were analyzed separately so that the appropriate
profile could be utilized.  This discussion appears in
Appendix D.  As an example, the VOC emission reductions
achieved in an area expected to utilize normal gasoline were
multiplied by the normal total HAP to VOC ratio, 4.8
percent, while those VOC reductions in an area expected to
use reformulated gasoline were multiplied by profiles
representing reformulated gasoline  (assuming 70 percent with
MTBE at 12.9 percent, and 30 percent without MTBE at 4.2
percent).
6.1.2  Emission Reductions By Subcatecrory
     The air pollution impacts will be discussed for each
subcategory in the gasoline distribution network in the
following paragraphs.  For each subcategory, the baseline
emission level will be defined along with the regulatory
alternatives and their effect on emissions for each type of
area.  Baseline emissions and regulatory alternative
emission reductions are shown in Tables 6-2, 6-3, and 6-4.
Table 6-2 shows emission reductions at existing facilities,
Table 6-3 delineates emission reductions at new facilities,
and Table 6-4 provides a summary for all facilities.
6.1.3  Pipeline Pumping Stations.
     Emissions from pumping stations consist entirely of
fugitive emissions from leaking pumps and valves.  As shown
in Table 3-11, it was assumed that all emissions at pipeline
pumping stations were uncontrolled at the baseline and that
there are 1,989 facilities.  Furthermore, it can be seen
from an examination of Table 8-27 that 27.9 percent of these
stations will be new (555 facilities) and 72.1 percent will
qualify as existing (1,434 facilities).  The number of
facilities times the estimated model plant emissions, as
                             6-3

-------
TABLE 6-1.  SUMMARY OF HAP VAPOR PROFILES USED IN ANALYSIS2
Description of Fuel
      Type
Applicable Areas
 for Fuel Types
 Total HAP
to VOC ratio
 (percent by
  weight)b
Typical, or "Normal"
 Gasoline
  Ozone and CO
   attainment
     4.8
Reformulated
Gasoline
     with MTBE

     without MTBE
      Ozone
  nonatta inment
                         12.9

                          4.2
Oxygenated Gasoline


     with MTBE

     without MTBE
       CO
  nonattainment
                          16.3

                           4.4
Reformulated and
Oxygenated Gasoline

     with MTBE

     without MTBE
  CO and Ozone
  nonatta inment
                          16.0

                           4.2
     Data collected  from various  sources used to calculate
     normal gasoline vapor profiles which were adjusted  to
     represent possible compositions  of reformulated  and
     oxygenated gasolines.

     As calculated in vapor  profiles  and shown in Table  3-2.
                            6-4

-------






„
1
8
£
CO*
M
cj *^

Z Q)
*"" ?
c« £
£ S
"s.
•g
a
X
1









X

g
— * X
1
E x
Cl
< Jt
" fc
1
M

£ «-
•"• «^
> *
4-*
E
eu X
4-1
CL
Z L.
X
•x
S

X
§ £
O C>
ID
c
fc X
4-*
CL
- x
0)

1 H
gj^
^ L.
UJ D>
—i a.
LU <
(A X
CO

















3

o
Kl
co"

«O
Kl

-O


•O
Kl
O
Kl
eo"


-o

s
-o


8
"I
C\J
CM


o
"
U>
O IA
t- 05 CO
2 S3
*" »*, >-
Se» c 3
'E i °

5 S> 1L ex
ex > B" m
UJ 4»» UJ ert
5 »§ 2§
_i St. _, -
UJ U. H- UJ 1—
a. o. <
et cT w

0
p
§
Kf
5
0
in
5

e
*«^
1C

^.
o

o
in"

5
5
tc


5

0
e
in

g
CO


o
CM
ro
VO



CD
C
CO
01
O)
CD
O
V
in


o
Kl
o
^~
"*
s
o

3

e
•—
**

o


o


o
o
-*


o

o


o
o"


1


J2

O _J
CO 4J
•M S
'i g en
uj a. — i
ill
4v ee
•— g ui
U. H- ^

i

iP
a
CM
5
eo
i
r>T
iP

i

3

^_
eo

1
CM"

§
S
-»


S!

e
•*
CM"

§
Kl
-*


1
CM"



.3
u
CD
CL
1
— 1



3!
_
S
CM
in
S
0
CM
2

g
-O
CM

eo
in

o
CM"

3
§
in


00
in

O
co
*
CM

O
K)
§


1
>*



£
c
t-
L.
o
4->
(A


•O

g
3
o
K
o

JO

0
^
 a
en 4->
UI C
u D.
.P. UJ
l!
U. H-










Sis.
*O

| |

a ra

in CM
eo o
>£? Kl"
C\J CM


S3 S

CM r^
-* CM
^T ,-T

0 0
in in
in KI
o" in
Kl Kl


co in
«-^ CM"

g
= g
E 1
^ 3
c
a j<
52 t 2
5 «?*-«,
2 i- .* .*
•• £ SS
W (/> i— o:

i








ON- «-
& CO

O O O
Kl O»
•^ o"
•4- in

OS-  Kl
CM" co"

° fe 5
o o o
K) ot
^* in


° ft S

0 8 8
O Kl
CM" eo"

00 0
P » &


1 1 1
IV O


g, w
c
a ' 1-
.* iB 2 ?
2 $ B E»
^ > K- e. en
t- •—  3 u.
ce
UJ
en


0
o
S

1
CM*


O
Kl
in
in




0
VO"
PJ

O
in
in




(M
S."
CM
o
o"
C

g
in





_J
h-
o
H-
f
CD
h-
6-5

-------
•a
Q>
•o
O



I

ii $ x
"* ? '£
Sg, 1 *
«/> • ~
"* C ^
£ 41 O.
IU JJ *
s *.
1 £
a
O)
5 >*

^
> X
O)
>
i
L.
£ **
<
c.
X
o>
z





K . „ . „ ,
^? ^5 ^9 ^9 ^? ^3
5- P- fC >- S IM
in «» KI in ^ w ^ CO ^
CM «-
§
in poo
in KI S S.
«- in CM


^- in »O ^*
ra T-
§ Sg§

g i I t i
5 2 S S S |
? 5 OT o a o
en — uc E — i
a.iu < a ai _i a a ^ uj
Z 4^iu^ *^- — uj -n i. ^ — -^ _i
— 21 2. — ° 02.*- « o c S 2. °-
— i 5 O" _i *j 5 &• o 4-> o 3 fr
u u.ui iu (/> u.iu ^ ^ irt »— u-iu w
O. CL _l J
CL CL eo m

IM


s
m

§
in
KI





I
r>j




P_
»-
in
KI




o
8.
« g
8 § =
oe H- —
? 8 g =
•- « o ^
1 « «5 1
« « «J « g «» •-
h-OO^^-g«
4^ • a 3 O" •— C ui
V> t- 1- U. UJ > => —I
flc B
» *
                                              6-6

-------
CO
2
O
H
B
  !
S

H
CQ

g






is
il
« E
§ 8,
Cj 2

i s
&
1
en





















H
= £
I 1
*»
£ x
X
«
= 8 *
> *
I

^
* 0.

11 L.
1
-
GJ ^

..- *
a
c
u
±! >s
^
a.
S
1_
X
-v
D>
X
O (->
— o
(/) ^>
- 2
lug
2 ^
S 1
«/l *
s













s
o
N
M

-O
Kl


S
(NJ
*
O

Kl



-O
Kl


5
OJ


0
00
to"


1


§10
z
S t 2
h- O V)
U) "- W
 c; m
UJ 4-» O- UJ
S ft 5
UJ U. UJ UJ
O. 0.
CL a!
- 5 * = ,
00 0 O 0
-j o in
O "- Kl O O
-j- <\1 rsi
O O OOP
KI 55 K
o «j K5 ° K
o o o o o
fNJ *- •&
•f o in
rJ

O «- MOO
•** fSJ fsJ


0 S S ° |
o «- tn o «-
•^ S. 
c i -j 1/1 o c
a uj _i a a *
I- *•• < DC t- U
a; c z 3
«? *J Q. * .C " *~
O 5) 5 f BJ O C
4-> 3 CT O ** O
to u. UJ ^ ^ W »—
^
g
3
o
s
Kl
S
5
o
<0

Kl



S
in
Kl
0
tc
k,
*o



M
sff


0
r\i



o
*
tn
^

o

"
g
u
H-
1
~- 0)
ai jc
to a
UJ
-S
91
U. LU







SK O K «-
sD K» CO
O O O O O
'. 8. s. ".
*^ Kl *O Kl

in (M o f- O Kl N-


O O G O 0
-* *- 00 CX
f^ t^ ^ ^


O O O O O
*o m «Q N" «-
o cO cO ID in
in" in *- co" -C*


o o o o o
r\j K» •* Jo *>

*"
J2 e ?
o S •-
g £ * =
" SP * 8 "-
^ c en c
— • •— a 5 •*
•- -O J< ••- « c
u. o a at M a
o at at o  Z
C E -> O -D
i- u o *" v P 5
v> 5 3 41 C < o
I-O>I-I.>SI_L.
zeni-i-.^gwDi
< a v a. t.
a o'c'c'5'5(j^
w a o S o- ~ c
^^-l«U-UJ>^

m u>

o
o*

1

o
Kl
in
~*





rj

I

Nj-







1


M
O
Kl
in
o"
o

1


-------
•a
0)
•a
U
§
U
n
 I
vo







i
is

i__ *^
11
is,
o) 5
55
S5
*
I
c.
>.
o.
r












1
! *
£ x
5
1
X
8
Q|
V «.
> £
O)
>
I
«j X
1
X
D)
1



>. U

*™ s^
1 *
*-*
E
•j
^
a.
o
I




o •*

o o
in
o «*•

o in
0 •*


o o
in

O •*

o in



0 «*

0 0
in



o >»


0 Kl
» § s §
g i 2 i
S I« s l«
to * t- v> *
a « • Jj «e ,co S
CL UJ < « UJ
•i § "" o> i i
uj ** a. uj a ** D.
3 58- 3 S S8-
U ^ UJ UJ C/J ^ UJ
a. ex
a! CL.
ro o CM
CM CM
0 0 O
0 S
•- CM
M O •"
(M CM
S ° 8
R ° RJ


000
o -o
•- CM

RI ° CM

3 ° §



Kl O CM
CM CM
O O O

O *O
«-^ CM"

Kl O «-
CM CM

S ° 8

m
» « 8
^ ^ _i
in o c
_i « • ^
< oc ,- sj
t a> * £.
s * s *-
UJ -D L. .*
- 1 s g
W _J Crt 1-
»
>o

o
in
in
CM

"
1
•O


o
in
in
CM"

S

8
CM





O
in
in
CM"

•O
*~

§
CM
S- « «*-
— « -5 £ ~- «
!! i ! s «
iS a ^ ^ uj
*-* *•• y u *-»
?| 2 g | 2 ||
i^5" 3 "^^£ —
58- °- 2 S i Sg-
U. UJ W (/> ^ >- U. UJ
3


o
CM
•o


1




o
CM



O





0

CM
*




1
|
1 |
g | |
UJ ™
W I-
                                           6-8

-------
 I
VD
CO

_
—
$
I
£



=
j.
I i
Tu fc
il 1
if
UJ
" S
- ?
11
111 * «
•C S
O) .£
w 0
c
f.
t>



"o
g
u
1
"3
CD










g 9
(J f*"- ^

to "*


0 §
z in"
SO 0
f*fc ^J
u ""* NJ- in
§ - e*



1 Kr?
z in



0 00
 ;£ |«V





§o o
< K.s
3: in

o
«- O 0
*** o?'"~"



CL ro ro oS
a: r>j" -o"

§,|
uj Z o. ui a Z 5
2 •-•:; Z «-•- 0-
•^ w 3 »^ o en uj
— r 3 O" — i f 3
UJ U. UJ Ul tO U.
Q. Q.
O. CL
o o o o
CO r^ *j" sf

•* in S


o op o
IM (\J ^
§es°
(NJ *O •— NO
CM f\J O* «"•
**• in CM



r\j oo to ND
(\J (M «-



o o o o
CO N- Ni" "•*
jo '•o *~ ^o
in CNJ o^ «-
** IA fNJ





0000
N- in o* *Q
>o eo ro ^
IM* CM* •-*

r\j «— >o in
O Od O* N*
eo o ro*^T
** o ts\ m



§000
in K. ro


Vi-
C* W
SC CO
5 >**
j^ **~ ID
CD M a>
„ HfiJ
< Bf >- 0 W V
s ?*tll £
& — a 4J 3. <
UJ "D L. J< •» O" -J
V- § OC §>UJ C,
—1 i^
CO CD
O
*J-
ft
0


I
m
o o o o o
^T f^ «•" K> «0
Kfc 3 C fe
«- «*


Np 00 O^ O ^^
-"-" M' » R



o o o o o
^ O in «—  »» «- ro «- •-
«- rj co


o o o o o o
%O C^ CK O* co in
tx ro co «- S >»

•- in
% i S
a i. —
ac ••- — •
O) t) (A u.
C O) C IA
•g J^ •—  -1 V) h-

o i- i- > B- »- i. <
a £ 5 i- O
L.j*jrf*»-cr uj u i—
o c c c> uj u -n
w a a S » c iu
C/) 1— 1— U- > 2 — 1
ae. a>
" 2
                                        6-9

-------





(Concluded)
•
1
vo
CQ









|
"S
co  in*
1 r


J2 j»
8 8
F F
i i
EL D.
8- I
UJ IU
co § ^ i
5*1*
< S 2 i
& •£ w .5
CO H- CO
O co 3 to to
(L uj < a uj
••" O) *^
uj «-> tu a *<
2 "S Z o'a
-i 5-J4-5
U U. IU CO U.
a. a.
ix a
to in S «- •*
CM in in o eo
SES§ !
10 in to in K-

C K CM in ' K
«M*
o o o o o
to in to in to
»
 u c 'e c 'E 3 *s
< cSiS'o111 "uu1" J- 3 «°
Z 3!>M 3 3 U < O — 1
— g> 41 i- > i- «uc.> P u <
tf"-0}'£ >•
*~ O 2 0 S" °" 4J (0 0 5 " C IU
^ _i cn i- u. K to >- t- u. > 3 -/
—1 ^ tt CC
CD m to »-
6-10

-------
discussed in Section 3.2.2, were used to calculate baseline
emissions.  The baseline emission levels for leaking pumps
and valves at pipeline pumping stations are shown in Tables
6-2 and 6-3.
     Regulatory Alternatives I and II specify an LDAR
program for pipeline pumping stations.  As discussed in
Chapter 4 (Table 4-2), it is estimated that a quarterly leak
detection and repair program will reduce emissions from
leaking valves by 44 percent and from leaking pumps by 33
percent.  These efficiencies were applied to all baseline
emissions from area source pipeline pumping stations to
estimate the VOC emission reductions shown in Tables 6-2 and
6-3.
6.1.4  Pipeline Breakout Stations.
     The emissions at pipeline breakout stations consist of
those from tanks used for the storage of gasoline and
fugitive emissions from pumps and valves.  As discussed for
pipeline pumping stations, it is assumed that fugitive
emissions are uncontrolled at the baseline.  The baseline
emissions and regulatory alternative reductions of fugitive
emissions from these equipment leaks were calculated by
multiplying the number of equipment components estimated in
the model plant analysis by the component emission factors
that were shown in Table 3-5.  The resulting emission
reductions for Alternatives I, II, and III (quarterly LDAR
at new and existing area sources and existing major source
facilities,  monthly LDAR at new major sources) are 340 Mg
HAP/yr and 4,540 Mg VOC/yr.  It was estimated that 7.4
percent of pipeline breakout stations are major source
facilities (92.6 percent will be area source sites) and that
9.3 percent will be classified as being "new" (consequently,
90.7 percent will be existing) in the base year of 1998 (see
Table 8-27).
     The baseline assumptions for breakout station storage
tanks were 143 uncontrolled fixed-roof tanks, 88 fixed-roof
tanks with internal floating roofs, 476 external floating
                            6-11

-------
roof tanks with primary seals, and 272 external floating
roof tanks with primary and secondary seals (see Table 3-
10).  Baseline emissions from breakout station storage tanks
were calculated by multiplying the number of dedicated
storage tanks by the throughput estimated in the model plant
analysis.
     Regulatory Alternatives I, II, and III for storage
tanks require that all fixed-roof tanks be equipped with an
internal floating roof with primary seals and that all
external floating roof tanks be fitted with secondary seals.
The installation of an internal floating roof on a
previously uncontrolled fixed-roof tank would result in VOC
emission reductions of 95 percent, as shown in Table 4-2.
Upgrading external floating roof storage tanks with primary
seals to secondary seals would result in emission reductions
of 50 percent, using factors from the same table.
Therefore, the emission reductions attributable to
Alternatives I, II, and III are the 95 percent reduction
achieved for the installation of an internal floating roof
for the 143 uncontrolled fixed-roof tanks and the 50 percent
reductions achieved with the addition of a secondary seal
for the 476 storage tanks with only primary seals.  This
results in an overall emission reduction from breakout
station storage tanks utilizing the controls specified by
Alternatives I, II, or III of 90 percent.
     Regulatory Alternatives IV, IV-Q, and IV-M require that
fixed-roof tanks at major sources be equipped with internal
floating roofs and that external floating roof tanks  (again
at major sources) be fitted with secondary seals.
Consequently, the emission reductions associated with these
alternatives would result from the addition of internal
floating roofs on the estimated 11 uncontrolled fixed-roof
tanks and the installation of secondary seals on the
estimated 35 external floating roof tanks associated with
major sources.  This results in an overall emission
reduction of 4 percent.  Emission reductions at new
                            6-12

-------
facilities will be zero since the storage tank NSPS already
requires the same control levels.
     6.1.5  Bulk Terminals
     The emission points at bulk terminals consist of truck
or railcar loading racks, storage tanks, tank truck leakage,
and fugitive emissions from leaking pumps and valves.  As
can be seen from Table 8-27, 28 percent of the bulk
terminals (287 facilities) will be classified as new in the
base year of 1998, while 72 percent of these sources (737
facilities)  will be classified as existing sources.  Each is
addressed separately in this section.
     6.1.5.1  Loading racks.  The levels of control at
loading racks range from uncontrolled loading racks  (splash
or submerged fill) to those loading racks with vapor
collection and processing systems that meet or surpass an
emission limitation of 10 milligrams of VOC emitted per
liter of gasoline loaded  (mg/1).  Using the control levels
for the consumption rates shown in Table 3-11, the baseline
emissions were calculated by associating each throughput
with the number of estimated facilities.
     Regulatory Alternative I requires that loading racks at
new major source bulk terminals lower emissions to 5 mg/1
and that area bulk terminal racks and loading racks at
existing major sources lower emissions to 10 mg/1.
Therefore, the uncontrolled emissions from existing truck
loading sources would be reduced from the uncontrolled level
to 10 mg/1 (nearly a 99 percent reduction for splash and
submerged fill operations) and other existing sources would
need to reduce their emissions an incremental amount as
well.  This amounts to an 87 percent reduction for sources
operating at 80 mg/1 and a 29 percent reduction for sources
operating at 35 mg/1.  To obtain the emission reduction
gained by implementing the 5 mg/1 standard at new major
source facilities, the entire baseline throughput  (446
billion liters) was multiplied by 5 mg/1 to obtain emissions
if all facilities were regulated at 5 mg/1.  To obtain the
                            6-13

-------
emission level at new major sources, the resulting number
was multiplied by the estimated percentage of major sources
(27 percent) and by the estimated number of new sources (28
percent).   The resulting emission level for this alternative
was estimated to be 292 Mg HAP/yr and 2,642 Mg VOC/yr.  This
results in an overall emission reduction from bulk terminal
loading racks of about 90 percent.
     Similarly, Regulatory Alternatives II and III require
that area source loading racks meet 35 mg/1 and major
sources meet the same levels as Alternative I (5 mg/1 at new
facilities, 10 mg/1 at existing sources).   Alternatives IV,
IV-Q, and IV-M propose to regulate major source bulk
terminal loading racks only, and these must meet 5 mg/1 for
new facilities and 10 mg/1 for existing ones.  Emissions for
these alternatives were calculated in a manner similar to
the others.  Emission reductions for these alternatives
would be about 25 percent.
     6.1.5.2  Storage Tanks.  The baseline emissions from
storage tanks at bulk terminals were calculated in basically
the same manner as discussed for breakout station storage
tanks.  Baseline storage tank population was separated by
tank type for the analysis.  The storage tank population has
been characterized previously in Table 3-11.
     The emission reductions attributable to Alternatives I,
II, and III for the installation of an internal floating
roof on the 1,072 uncontrolled fixed-roof tanks and the
reductions achieved with the addition of a secondary seal on
the 2,426 storage tanks with only primary seals are 2,850 Mg
HAP/yr  (54 percent reduction) and 52,670 Mg VOC/yr (58
percent reduction).  The emission reductions attributable to
Alternatives IV, IV-Q, and IV-M, which require  (at major
source  facilities only) installation of internal floating
roofs on fixed-roof tanks and addition of secondary seals on
external floating roof tanks with only primary seals, are
approximately 15 percent.
                            6-14

-------
     6.1.5.3  Tank Truck Leakage.  The baseline regulatory
levels for controlling leakage from tank trucks during
gasoline loading include leak-tight inspection programs,
usually required annually.  The baseline emissions from the
446 billion liters loaded into tank trucks and railcars are
3,730 Mg HAP/yr and 53,950 Mg VOC/yr.  The baseline
assumptions were that approximately 317 billion liters were
loaded into trucks regulated by the annual leak tightness
program and 129 billion liters were loaded uncontrolled.
     Regulatory Alternatives I, II, and III require that a
vacuum assist vapor collection system be installed at each
new major source terminal (existing major sources and all
area sources would be required to implement annual vapor
tightness testing).   It is estimated that implementation of
vacuum assist loading would affect approximately 3,300
trucks at new major source facilities.  This number is
derived from a calculation based on facility population
characteristics (28 percent of bulk terminals are "new" and
27 percent of those are estimated to be major sources).  The
vacuum assist system, as discussed in Section 4.1.4.3, is
expected to reduce tank truck leakage emissions at the
loading racks nearly to zero (estimated 98 percent
reduction).  Therefore, the emission reductions for these
regulatory alternatives entail reducing tank truck leakage
VOC emissions at new major source facilities to 2 percent of
the previous levels.  Under these alternatives, trucks
loading at all other bulk terminals (approximately 40,600)
would have to undergo annual leak tightness testing
according to EPA Method 27.
     Regulatory Alternatives IV, IV-Q, and IV-M require that
the same vacuum assist system be installed at new major
source bulk terminals, and also require annual vapor
tightness testing, as specified above, of trucks and
railcars that load at new and existing major source
facilities.  It is estimated that these alternatives would
affect approximately 8,500 trucks (72 percent of facilities

                            6-15

-------
are classified as existing and 27 percent of those will be
classified as major sources).
     6.1.5.4  Fugitive Emissions.  The fugitive emissions at
bulk terminals occur from leaking pumps and valves that are
components of the piping that transfers gasoline and
gasoline vapors.  The baseline emissions (4,340 Hg HAP/yr
and 56,460 Mg VOC/yr) were calculated on a per-component
basis and as such 330 Mg HAP and 4,290 Mg VOC are attributed
to new major sources, 840 Mg HAP and 10,940 Mg VOC to
existing major sources, 890 Mg HAP and 11,500 Mg VOC to new
area sources, and 2,280 Mg HAP and 29,700 Mg VOC to existing
area sources.  The levels of control for the regulatory
alternatives for fugitive emission reductions at bulk
terminals are the same as those discussed for pipeline
breakout facilities.
6.1.6  Bulk Plants
     There are four sources of emissions at bulk plants.
Emissions occur during the filling of the storage tanks,
during the loading of tank trucks at loading racks, from
tank truck leakage during loading, and as fugitive emissions
from leaking pumps and valves.  Under existing criteria,
there are no major source bulk plants; all qualify as area
sources.  As can be calculated from data in Table 8-27, 14.2
percent  (approximately 1,790 facilities) of these sites
qualify as new and 85.8 percent  (10,800 facilities) fall
into the existing site category.
     6.1.6.1  Storage Tank Filling.  The current control
method for bulk plant storage tank filling consists of vapor
balance piping that transfers gasoline vapors from the
storage tank to the tank truck unloading gasoline.  As
discussed in Section 4.1.5.3, this technology has been
demonstrated to reduce emissions by 95 percent.
Approximately 45 percent of the estimated 25,200 storage
tank loading facilities  (approximately 3,600 new and 21,600
existing as calculated using the data in Table 8-27) use
this method.  The remaining 55 percent are uncontrolled.
                            6-16

-------
Baseline emissions were calculated by multiplying throughput
identified in Table 3-11 by these facility populations.
     Alternatives I and II would require implementation of
the above mentioned vapor balance system at area source bulk
plants (both new and existing).  As a result, emissions
under these alternatives are reduced approximately 85
percent from baseline.
     6.1.6.2  Tank Truck Loading Racks.  As discussed in
Section 4.1.5, the control technology for loading racks at
bulk plants consists of the installation of vapor balance
piping that transfers gasoline vapors from the tank truck
being loaded back to the storage tank.  This technology has
been demonstrated to achieve a 95 percent reduction in VOC
emissions.  The baseline analysis assumes that approximately
49 billion liters is loaded into trucks using vapor balance
methods, 30 billion liters using submerged fill, and almost
9 billion liters using splash fill (Table 3-10).
     Regulatory Alternatives I and II require that new and
existing area source bulk plants install vapor balance
piping on their loading racks, but allow a 15,000 liters/day
(4,000 gallon/day) exemption.  Submerged fill is required
for plants with throughputs below this level.  Therefore,
emission reductions calculated for these alternatives would
arise from plants with previously uncontrolled throughputs
(an estimated 14 percent of the total of 12,600 facilities,
or 1,750 loading sites).  Throughputs associated with this
segment of the population were multiplied by the controlled
emission factor to obtain emission quantities.  This results
in an overall emission reduction from tank truck loading at
bulk plants of about 65 percent.
     6.1.6.3  Tank Truck Leakage.  None of the presented
alternatives requires additional controls or control
procedures for tank trucks loading at area source bulk
plants.  As a result, none of the alternatives would yield
an emission reduction for this emission point.  Baseline
leadage emissions are 890 Mg HAP/yr and 13,220 Mg VOC/yr.
                            6-17

-------
     6.1.6.4  Fugitive Emissions from Equipment Leaks.  The
fugitive emissions at bulk plants occur from leaking pumps
and valves that transport gasoline and gasoline vapors.
Baseline emissions of 9,190 Mg HAP/yr and 130,757 Mg VOC/yr
were calculated on a per-component basis.  Alternatives I
and II specify the implementation of a quarterly LDAR
program at both new and existing facilities.  This level of
control is the same as that specified for area source bulk
terminals.
6.1.7  Service Stations (Storage Tank Filling)
     The emissions from service stations considered in this
regulatory development result during the filling of the
storage tank, which is typically underground.  The control
technique used to reduce emissions from this operation is
vapor balance.  The vapors being forced out of the storage
tank by the incoming liquid gasoline are collected and
returned to the tank truck.  This has been demonstrated to
reduce VOC emissions by at least 95 percent.  The baseline
assumptions for service stations are that approximately 289
billion liters are loaded into service station storage tanks
using vapor balance, about 86 billion liters loaded using
submerged fill, and the remaining 71 billion liters loaded
using splash fill (Table 3-11).  As can be calculated after
an examination of Table 8-27, the majority of this
throughput can be attributed to existing service stations
(97.3 percent).  It is estimated that only a minor amount
(2.7 percent) will be attributed to new service stations in
the base year of the analysis.
     Regulatory Alternatives I and II require the
installation of vapor balance systems nationwide  (all
service stations meet area source criteria), but each
contains an exemption for stations with throughputs less
than 10,000 gallons/month  (about 7 percent of the
throughput, see Table 5-7).  Submerged fill will be required
for stations with throughputs below this level.  Therefore,
the emission reductions for these alternatives would  come
                             6-18

-------
entirely from previously uncontrolled areas (approximately 9
percent of the 387,750 stations or approximately 35,000
service stations).  This results in an overall emission
reduction for each of these alternatives of a little more
than 75 percent.

6.2  WATER POLLUTION IMPACTS
     The overall impact of the alternatives on water
resources is negligible.  None of the emission control
technologies creates a significant water discharge.  Only if
refrigeration systems, which cool and condense the vapors
from the loading operation for liquid recovery, are used for
bulk terminal control, would a potential water pollution
impact be created.  In a refrigeration system the vapor-air
mixture collected at the loading rack is cooled to very low
temperatures (as low as -180°F).  Along with the gasoline
vapors, moisture in the air is condensed.  The amount
condensed is dependent upon the humidity of the entering
process stream flow.  As a consequence, a small amount of a
liquid gasoline-water mixture is generated.  This mixture is
then passed 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 since gasoline is
essentially insoluble in water1.

6.3  SOLID WASTE IMPACTS
     The only solid waste that may be generated by any of
the control systems being evaluated would be spent activated
carbon used in a bulk terminal carbon adsorption system.
For this scenario, the assumption would be that the carbon
could not be reactivated and would have to be discarded
after its useful life.  Table 6-5 summarizes calculations of
this potential solid waste impact.  This analysis assumes
that approximately one-third of the terminals requiring
control would choose carbon adsorption.  This estimate is

                            6-19

-------
slightly higher than the estimated national average of
emissions processed at bulk terminals using vapor recovery
devices (25 percent) but this impact analysis is intended to
be conservative.  Consequently,  the average annual solid
waste impact is averaged over the 10-year life of the
carbon, which results in a total environmental impact of 260
tons per year or an average of 0.73 ton per terminal.  To
put this impact in perspective,  the average person generates
almost 2 Mg of solid waste per year2 (10 pounds per day,  365
days per year = 1.6 Mg per year).  Therefore, this solid
waste impact could be considered negligible.

6.4  ENERGY IMPACTS
     Energy impacts for the regulatory alternatives were
estimated in the form of gallons of gasoline saved.  Energy
savings were derived by determining the liquid gasoline
equivalent of the emission reductions presented in Table
6-5.  Liquid gasoline is saved from equipment leaks and
storage tanks since less product is allowed to evaporate and
escape.  Gasoline is recovered at terminals when carbon
adsorption or refrigeration systems are used to control
emissions.  Gasoline is recovered, or not lost to
evaporation, at bulk plants where vapor recovery is used on
outgoing loads.  When gasoline is pumped from storage to
fill the trucks, vapors are returned to the tank, thereby
reducing evaporation and saving gasoline.
     Table 6-6 summarizes the liquid gasoline saved.  For
bulk terminals, it was assumed that 25 percent of the
emission reductions would be processed using recovery
devices (carbon adsorption, refrigeration).  Although these
control devices use energy for their operation, the amount
is relatively small and has been subtracted from the gross
savings at bulk terminals shown in Table 6-6.  Savings
ranged from 68 million gallons per year for underground
storage tank filling at existing service stations under
                            6-20

-------
       TABLE 6-5.  ESTIMATED SOLID WASTE IMPACTS FROM
         CARBON DISPOSAL AT BULK GASOLINE TERMINALS
                                                     Annual
  Bulk Terminal        Carbon        Regulated8    Solid Waste ,
   Model Plant       Capacity0,      Facilities         Mg
  	Ibs	


        1             10,000           123             56


        2             14,000            69             44


        3             18,000            84             69


        4             25,000            30             34
      Total                           306            203
a Regulated facilities determined by assuming 30 percent of
  all facilities require control.  Number of facilities by
  model plant determined by using 30 percent of facilities
  presented in Table 5-3.

b Annual solid waste impact determined by assuming one
  third of all facilities will use carbon adsorption and
  carbon must be disposed of after end of useful life  (10
  years).  Annual solid waste impact averaged over 10 years
  life.

c Reference 3.
                            6-21

-------




g
i
fa
S"
W
CO CO
W
W >
55 H
H EH
O 55
CO K
< H
O EH
N™*' I-H
to*
fe5 0
H K
> EH
CO O
u
O 55
CJ O
W H
W D
CQ
EH H
W K

CO
Q H
W D
EH
rtj W
S a
H H
CO O
W CO
rtj
O
•
vo
1
vo
w
03
rtl
E"1






•J «*"x

2 °3
*•*
* -rj
* ;i
§

^
S *~ •
•5 ii
ta- S **" w
ty t/J
,,_
O
C
l_ ^

3 Iu
e oi
UJ E
g


8^.v
— a *•
8 ITS

— u (0 °
>
o
c
o ."
< QC U
X
Si
i
















eo eo
a-



in in
h  (. 0 >
O. ••- OP O) —
4j eg *j
(U .— «J I- .-
CO) COO)
— rf U. — »  IA
W eg co Ji uj
-* oe H- o
§3 U
5> O I- >
— C 0) 1- •-
E '-o t*Z
* S 15^

"3
CD

co



1
~


>» t> C> K> fO
*^ ° 5! S cS
^




§11 1 §
«-^sT eo* cT h-*
Kt (M <« P. 0
«- >»



^» o o ro in
r\j o o- co r^
^ «- «- ^ o





o o o o o
o o o o o
^ N. «- Kl O
«- N. eo K» o
K) f\l -* K- O
^ in







^ C
o ~-
c?£ S
•— O) 0) W) u.
— c 01 c
— * •*• a o \/
.^ -D J< •- C
u. CD a CA a
O O in l-
J^ «J — 1 ••- 0)
MVI-l-> C0t._l
4JO>t-h-«- *-> O) «
CO *-<  t. »-
^ o"c^'5 «»-n °
Q. *j a co 3 u c
V> t- *- u- "-3 UJ
* > _l
— . (.CD
3 «) «t
co in t-


u
i
§
4->
8
4^
1
&
*
i
i
^j
1
L.
S
V)
1
L.
ft)

O
w"
U
o
T3
fc. X
«J ^
.~ X
_ l_
CO 0
K. «
Kl «-
M
CD ^
•** o
fe "
0 0
^ o
L>
£ |
.^ o
Ji *-
•* o
no ~
c §
*- s.
"fe
II X
xi
M CA
| §
si

«9 .O
6-22

-------















•o
Q)
•o
3
rH
U
c
o
CJ

t
VO
1
VO
w
03
<
EH


















c,*2
•- -V
T !•!«
** 
$
1 :
*"* T
v

-o
c *~ "t.
13 c/T ^*
O Cfi ^*
i If 5
CO
.£
I
1 1-
uj x
CJ
§


^
g~v
•i i-
*^ M ,£ *0
> « *£ 0
^ (Q
(Jj W
.i
4_j
CD
c

o
S "2
^ DC U
. •t'
UJ X

§








CM CM
CM O



§o
o
in
in


CM «-
CM O






o o
o o
in




r\i
(M O
CM O









§0
in
•O_
in



wt
v> S z 6
g E 2 c
•- C co c w
co o 5 M
••-CO 1— ••- 0
Sen j* 3 en 4>
_« <9 O  _1
t"" "E -i < "c 'i *-•
u ui a u c
uj — £ ui o> — •—
z *-> a. 2 a *j 3
!j n '5 j o 'ra uj
uj 5 &• uj v 3
O. U. UJ Q. CO U.

a. a.

CM^-^>0 ^
•^ 



§§§§ g
•* IA K* CM in
CM in in oT eo"
^ •— •*


CM •- •-; -O O
»^ *O CM CM CO






O O O O O
O O O O O
st in KI ^o N-
CM in in t ^
CV CO ••- D> fl» CO U-
0)CB) ->CO)CCO
fDOJ* — • ••- CQ O J^ J^
J^ ••- to «P- "O J^ ••- CD C
jj cacocu ivoaaiu coo
coinvv)_i ocuco_i z h-
.*.* _i •— ^_j_j»- o
-J 0 C E *^ C £•(-> — "O
< a CD j^ uj c cgjtfjxujc •- c co
zaci-u a> coi-uy  a. z 4> i- c. > a. cnt, »-
O£ C O) 1— •»• — < O) t- 1- — — U> O
LU •- 03 4J 3 _l O <-'2. UJ U h-
t«'Ot.j^*^-D~ a.i-jj^^»-cT ocv
aoccntu o c c DILU ~ "6 uj
« O 4-- CO 3 V 4V (0 CD 3 > C — 1
_j _i co t— u. _i co i- »- u. oca m
^ 3 ui <
m oo en t—


«
8
1
§
4-*
4^
1
t.
1
g
*^
y
^K
c
s
*w
U)
i
4)
^r
o
W
4)
O
"?
*^ s
fc «
•t^ >*
g i
hi «-
"co -Q
O) _


01 ^
"* o

• c
° 8
0) CO
4-> g
"* '«
** .-
O) ^_
'o 4J
C £
H
£.0
II X
Gallons/yr
Assumes onl


6-23

-------
Alternatives I and II to 0.02 million gallons per year
savings for equipment leaks at new breakout stations under
Alternative IV.

6.5  OTHER ENVIRONMENTAL IMPACTS
     Other potential environmental impacts include noise
impacts.  The relative impacts of the regulatory
alternatives on this environmental concern are expected to
be insignificant.  An EPA test4 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.
     If incinerators/combustors/flares are utilized to
control loading rack emissions at bulk terminals, the
combustion of the gasoline vapor will create secondary air
emissions of other compounds, specifically particulate, SOX,
and NOX.  Assuming a worst-case situation that one third of
all terminals install a destruction device that burns the
gasoline vapor, the estimated particulate, SOX,  and NOX
emissions are shown in Table 6-7.  These estimates were
calculated using AP-42 emission factors for natural gas
fired boilers of 3.0 Ib/million ft3,  0.6 Ib/million ft3,  and
100 Ib/million ft3, for particulate,  SOX,  and  NOX,
respectively.  Consequently, the total impact would apply
under Alternatives I, II, and III, but only 27 percent of
total impacts would apply (27 percent of sources are major
sources) if Alternative IV, IV-Q, or IV-M were implemented.
                            6-24

-------
TABLE 6-7. ESTIMATED PARTICULATE, NOX,  AND  SOXEMISSIONS
     FROM INCINERATION AT BULK GASOLINE TERMINALS
Bulk Terminal Regulated8
Model Plant Facilities
1 123
2 69
3 84
4 30
Railcar 19
Total 325
Annual Emissions (Mg/year)
Particulate NOX SOX
0.8 28.4 0.2
1.2 39.4 0.3
2.9 96.5 0.5
2.1 68.9 0.5
0.3 9.4 0.1
7.3 242.6 1.5
  Regulated  facilities determined by assuming  30
  percent of all facilities require control.   Numbers
  of  facilities by model plant determined by using  30
  percent of facilities presented in Table 5-3.

  Calculated using emission factors for natural gas-
  fired boilers less than  10 mmBTu/hr.
                          6-25

-------
6.6  REFERENCES

 .1.  The Merck Index.  Eleventh edition.  Merck & Co., Inc.
     1989.  p. 4269.

 2.  "Procedures for the Preparation of Emission Inventories
     for Carbon Monoxide and Precursors of ozone."  U.S.
     Environmental Protection Agency.  Research Triangle
     Park, NC.  Publication No. EPA-450/4-91-016.  March
     1991.

 3.  Telecon.  Haves,  T., Pacific Environmental Services,
     Inc., to Keller,  D., IT McGill, Inc.  February 26,
     1991.  Carbon adsorber cost estimates.

 4.  Betz Environmental Engineers, Inc.   Gasoline Vapor
     Recovery Efficiency Testing at Bulk Transfer Terminals
     Performed at Pasco-Denver Products Terminal.  Prepared
     for U.S. Environmental Protection Agency.  Research
     Triangle Park, NC.  Contract No. 68-02-1407.  Project
     No. 76-GAS-17.  September 1976.
                            6-26

-------
                     7.0  CONTROL COSTS

     This chapter presents a discussion of the costs of
 implementing HAP and VOC emissions control at gasoline
 distribution facilities.  Using the model plant parameters
 previously described in Chapter 5, costs have been developed
 for each of the six regulatory alternative arrays.  Section
 7.1 presents model plant costs for each facility type to be
 regulated: pipeline facilities, bulk terminals, bulk plants,
 and service stations.  Costs associated with storage tanks
 and leak detection and repair programs are discussed
 separately since they will be incurred at facilities in more
 than one category.  Section 7.2 presents an analysis of the
 control costs for each of the regulatory alternatives.
 Tabular costs are provided along with a discussion of the
 sources of data and the assumptions used in deriving the
 costs.

 7.1  MODEL PLANT COSTS
 7.1.1  Storage Tanks
     This section addresses the cost of controls for storage
 tanks present at pipeline breakout stations and bulk
 terminals.   Storage tank control technigues have been
 discussed in Section 4.1.3 and include the installation of
 internal floating roofs on fixed-roof storage tanks and the
 addition of secondary seals on external floating roof
 storage tanks.
     The annual costs associated with installation of an
 internal floating roof within an existing fixed-roof tank
structure were derived from costs developed in previous EPA
studies for the third quarter of 1991.1  The  capital  costs
are based on a model tank with a capacity of 2,680 m3 and a

                            7-1

-------
diameter of 15.2 m for bulk terminals, and a capacity of
8,000 m3 and a diameter of 30  m for pipeline breakout
stations, and are summarized in Table 7-1.  According to
estimates from vendors2,  degassing and cleaning costs for
tanks at terminals and breakout stations, shown in Table
7-1, as well as the floating roof tanks detailed in Table
7-2, are approximately $9,000 and $13,000, respectively.
The waste disposal cost averages approximately $3,000 for
all the tanks.  The roof and seal costs were based on
figures and formulas given in the draft 1991 floating and
fixed-roof tank CTG.  The deck fitting costs also were taken
from the CTG.  The annualized costs for maintenance; taxes,
insurance, and general and administrative charges; and
inspections were estimated using the same percentages as
presented in the draft 1991 CTG.  A recovery credit was
calculated to reflect the amount of gasoline that would no
longer be lost through evaporation, breathing loss, etc.
after this control measure was implemented.  Note that the
price per liter of gasoline used to calculate recovery
credits is different at bulk terminals than at pipeline
breakout stations.  This is due to the fact that some
federal tax is actually collected at the bulk terminal, thus
raising the price slightly.  Additionally, the concept of
equivalent dedicated storage tanks (number in use as opposed
to the total number at the facility) was used to calculate
emissions as presented in the tables.  However, the recovery
credits should be distributed among the actual number of
tanks at each model plant.  Since there are a different
number of storage tanks and dedicated storage tanks at each
model plant, the recovery credits calculated for Tables 7-1
and 7-2 are presented as weighted averages.  The combined
annualized "costs" result in a net annual savings  (recovery
credit - annualized cost) of $13,540 at bulk terminals and
$66,080 at pipeline breakout stations.  Emission reduction
(storage tank emission factors from Tables 3-6 and  3-7 times
                             7-2

-------
        TABLE 7-1.    COSTS  OF  INSTALLING  A  BOLTED  INTERNAL
            FLOATING ROOF ON  AN EXISTING FIXED-ROOF TANK
                        (THIRD  QUARTER  1990  DOLLARS)

  BULK TERMINALS                                  BREAKOUT STATIONS

  Assumptions:    Tank Capacity    « 2,680 M*         Assumptions:    Tank Capacity*   8,000 m3
                        Tank Diameter   • 15.2 n                        Tank Diameter*  30 m
                        Tank Height     « 14.6 •                        Tank Height *   12 m
                        Emission Reduction «                            Emission Reduction
                        45.9 Mg                                        • 497 Mg
                                                         BULK TERMINAL
BREAKOUT STATION
  Capital Cost I Installation


Degassing, Cleaning, I Waste Disposal*
Roof with Liquid-Mounted Sealb
Controlled Deck Fittings'"
Total Capital Cost
$9,000
$19,900
$200
$29,100
$13.000
$41,550
$200
$54,750
Annual i zed Costs ($/yr)




Net
Cost
Maintenance (5X)b
Taxes, Insurance. G&A (4X)b
Inspections (1Xr
Annual Capital Charges (11.76X, 20 yrs. a
10X)
Total Annuali zed Cost
Product Recovery Credit
Annualized Cost ($/yr)"
Effectiveness ($/Mg)
$1 ,460
$1,160
$290
$3,420
$6,330
$19,870C
($13,540)
($295)
$2.740
$2,190
$550
$6,440
$11,920
$78,000"
($66,080)
($133)
'  Based on vendor estimations of $6,000 - $11,000 for degassing and cleaning,  and about $3,000 for
   waste disposal.3

b  Reference 1.

c  Based on a calculation which subtracts losses from internal  floating roof tanks from
   uncontrolled josses at fixed-roof tanks and a cost of gasoline at bulk terminals of
   $0.290/1iter.4

d  Based on the same loss calculation as specified in footnote  "c" and $0.285/1iter of gasoline at
   a breakout station.5

'  Net  annualized cost (savings).
                                          7-3

-------
         TABLE  7-2.    COSTS OF  INSTALLING  A  SECONDARY  SEAL
             ON  AN EXISTING EXTERNAL  FLOATING  ROOF TANK
                        (THIRD  QUARTER  1990  DOLLARS)
 BULK TERMINALS

 Assumptions:
Tank Capacity    * 2,680 m3
Tank Diameter    * 23.8 m
Tank Height     * 12 m
Emission Reduction  * 7.5 Mg
BREAKOUT STATIONS

Assumptions:     Tank Capacity^  8,000 m3
               Tank Diameter^  30 m
               Tank Height *   12 m
               Emission Reduction *  9.6 Mg
                                                         BULK TERMINAL
                                                          BREAKOUT STATION
 Capital Cost & Installation
Degassing, Cleaning. I Waste Disposal*
Secondary Seal Cost
Controlled Deck Fittings'*
Total Capital Cost
Annualized Costs (S/yr)
Maintenance <5%)b
Taxes, Insurance. G&A (4%}b
Inspections (1%)
Annual Capital Charges (11.76%, 20 yrs. 3 10%)
Total Annualized Cost
Product Recovery Credit
Net Annualized Cost ($/yr)
Cost Effectiveness ($/Mg)
$9,000
$13,200
$680
$22,880

$1,140
$920
$230
$3,730
$6,020
$3.250C
$2.770
$370
$13,000
$16,960
$680
$30,640

$1,530
$1,230-
$310
$4,990
$8,060
$1,510d
$6,550
$682
3   Based on Vendor estimations of $6,000 - $11,000 for degassing and cleaning and about $3,000 for
   waste disposal.

   Reference 1.

   Based on a calculation which subtracts secondary seal losses on an external floating roof tank
   from primary seal losses on an external floating roof tank and a cost of  gasoline at bulk
   terminals of $0.290/liter.4

d   Based on the same loss calculation as specified in footnote "c" and $0.285/liter of gasoline at
   a breakout station.5
                                          7-4

-------
control efficiencies from Tables 4-2 and 4-3) and overall
cost effectiveness  (annualized cost divided by emission
reduction) reflect  this same trend.  As discussed previously
for installation of seals on a fixed-roof tank, the net
annual cost to install a secondary seal on an external
floating roof tank  (annualized cost - recovery credit) at a
pipeline breakout station is $8,060 and at a bulk terminal
is $6,020.  Emission reduction and cost effectiveness were
calculated in the same manner as noted for fixed-roof tanks.
7.1.2  Leak Detection and Repair
     As discussed in Chapter 3, leaking pumps and valves are
sources of emissions at pipeline facilities, bulk terminals,
and bulk plants.  Vapor leakage from tank trucks will be
discussed later.  The basic control technology discussed in
Chapter 4, Section  4.1.7, involves LDAR programs with
varying frequencies of inspections.  Tables 7-3 and 7-4
present model plant costs as well as cost effectiveness for
quarterly and monthly LDAR as implemented at pipeline
pumping and breakout stations, bulk terminals, and bulk
plants.  Table 7-5  provides costs per monitoring event.
Capital costs do not appear in the tables as there are none
assumed to be associated with the implementation of LDAR (no
equipment purchase, only annual monitoring and maintenance
costs).
     According to an estimate by a company providing this
service9,  a technician  can monitor approximately 300-600
components (i.e., pumps and valves) per day.  Model plant 2
for pipeline breakout stations has 470 components;
therefore, this analysis assumes that all monitoring can be
performed in one day for all model plants.  According to
another company's estimate, the minimum charge for a
technician to perform LDAR is $600/day.  The model plants
for the pipeline pumping stations have the fewest number of
components, so this analysis assumes that a technician can
monitor two facilities in one day for $600 or monitor one
facility for $300.  Extra charges for repair cost are
                             7-5

-------






K
H
W
K
a
o
H
EH
O
EH
W
D


<
(4
(-5
JH
. *i
K
W
EH
K
S
 O •-'
at


^s
4-> c"
«) —
O L.
u o
4-*
"c 'c
4-> O
0 X
1- >»
4A

-c O)
« C
§ T
U 0
4-i
> o
a x
£ ^
^^

™4J ?
« .„
O i_
(J 0
""" *C
1*
QC M



WO
If
'("
.?s
L. "r
o c
•t^ X
§ $












IMT3TJ T*7 T7TJ"O^ tf*
"Q V V V V WVVV -*
i- L. *- <- UL.UU
U 0 U 0 UUUU

P! * s. * ^ 8 S 8 S fS
"^SM inlC ro ^ ^ •« •-

^^»K> ^ep KI p- o- «-
h-Nin ^«> » 4 o KI co «5
m^o* *« «*»» *o
»^fOM Io2 S3





•o >O *^? ' ' i i i i i
co co co ii i i i i i






o o o o
rt in in r-1 --. in in
in in ° • • ° ° • • m
•^ PJ r»- *^ *^ PJ rvj








ooo oo o o o o o
ooo oo oooo o
KI KI KI -OO -O -O >O O -O





.,
? i «
'i- « ? "
3 c « g .£ c £
°-°.-f\)Ki to ° •- ni ^«-r»Ki>» ^X
0 ~ a. o- a '^xx uifc0-0- a^-
—* (/) '^  in
4-* 3 •
• g "O PJ
O 4J M _ •
U « O 




'N
^ 10 43
7-6

-------





tf
H
W
PS
0
§

O
H
E-i
U
W
w
o
a
w

^4
a

2
O
£

&
O
E-i
8
ij
g
2
•5
^«

.
*^
ri.

PT*J

h~i
cn
E-<






(A
W V)
ifi
C **-
Emission
Reduction
(Mg/yr)

"g «-• L.
§8£
CUM

X
^ *^ ^N
§0$



w?
cn •—
O L.
u o
si
O X
tl

^
*** c
o 'C
U 0
CJ 'c
> 0
a i
^ ^s
^
o T
0 o
i- £
— Q
||
^^


t/i ^^
O Ol
cn i-
C 0
Z •—
o c
• z

o vt
X >-<








fe * H § £
VX *^

*o 3 S ^j S
5 S i$ g S
^ v— Kl

o> ^ S P
in *o ** •* o
o « IM  o ^
-" M* «0" 0»" ~




0 0
r- •-  ~ in
4} tti
.- 2-
o- £



•0 ^ M 0 O

3 Jo ~  *> K-

*r S ~ Q •*
•o o «» •- •*
•* ^ ** » »


S*S 5 "5 Z
•*• N- ** to
>Q r^ *? to tv




0 0
o o in in „,
*O "O





1 1 1 1 1
1 . 1 . 1




o o
° ° * ™
•^ *^ cv rj







o o o o o
o o o o o





(fl
1 3
'i II
i; a! a. a. CL cv i
" x x x x ^ £
•" ^ *
3 *"





1
L.
CO
i
U
1
1
—
1
3
%
8

1
8
y
"o
tft
a
^S
+J
i

4-*

Cfl
'^
X
'i
^
«!«;
rvj to ^
«j o o
u — ' in
CD >.r\j
1 	 > CM
CJ o *o
«) 2 CB
C? (A
s1-^
cn O o
o cv in
• S "O  X
O U W —
0 CA O «l
01 o «-*
tl g
•^ o v *^
. *. TJ > X
ID C ID O
O) CJ U I.
CB * "3 °
C C t-
SO O O
— ««-
o cJ »^ o>
|S3.S
"6 CD < *-•


^
^ B XI
7-7

-------
 TABLE 7-5.  SUMMARY OF LEAK DETECTION AND REPAIR NET COSTS
                    PER MONITORING EVENT
                 (THIRD QUARTER 1990 DOLLARS)
          Model Plant3
                                   Cost  ($/component)

                                  Quarterly    Monthly
LDAR"
LDAR"
 Pipeline Pumping Stations

 Model Plant 1
   2 pumpsc,  25 valves

 Model Plant 2
   5 pumpsc,  50 valves

 Model Plant 3
   9 pumpsc,  100 valves
   4.97
  (2.84)
  (5.62)
   8.79
   1.34
  (1.58)
 Pipeline Breakout Stations

 Model Plant 1
   20 pumpsc,  250 valves

 Model Plant 2
   35 pumpsc,  400 valves
  (2.94)
  (2.92)
  (0.46)
  (0.81)
 Bulk Terminals

 Model Plant 1
   10 pumpsc,  90  valves

 Model Plant 2
   10 pumpsc,  115 valves

 Model Plant 3
   10 pumpsc,  130 valves

 Model Plant 4
   10 pumpsc,  160 valves

 Bulk Plants

 Model Plants 1-5
   4 pumpsc,  50 valves
3.78)
3.85)
3.88)
3.93)
.36
( -06)
(.25)
(.52)
   2.97
   6.39
a Model plants and parameters  from Table  5-1.
b ( ) Indicates a negative cost or net  savings,
c Assuming two pump seals per  pump.
                             7-8

-------
estimated at $2.5O/component.  An extra charge for travel is
added to the costs at pipeline pumping stations due to their
often remote locations.  The total cost for monitoring
includes extra repair and travel cost.  Since quarterly LDAR
occurs four times a year, the "Total Cost" per monitoring is
multiplied by four to obtain the "Annual Total Cost" for
quarterly LDAR in Table 7-3.  Similarly, monthly LDAR occurs
12 times a year.   As a result, the "Total Cost" per
monitoring is multiplied by 12 to obtain the "Annual Total
Cost" for monthly LDAR.  (Costs can be scaled back or scaled
up accordingly, for components that are allowed to drop back
to a quarterly monitoring period or for those that must be
monitored monthly for a time.)
     Annual baseline emissions were calculated for each
model plant by multiplying the leakage rates for pumps and
valves (see Table 3-5) by the number of pumps and valves at
the model plant over the annual operating schedule.  Annual
emission reductions were calculated using the efficiencies
associated with quarterly and monthly LDAR as shown in Table
4-4.  The emission reductions were used to calculate a
product recovery credit to reflect the amount of gasoline
that would no longer be lost through evaporation or leaking
at the pumps or valves.  The "Annual Cost Effectiveness" was
calculated by dividing the difference between the "Annual
Total Cost" and the "Recovery Credit" by the "Emission
Reduction."  In several model plants, implementation of
quarterly or monthly LDAR results in a net savings or
negative cost, due to the recovery credit.  This occurs
primarily at the model plants which have the most pumps and
valves.  Since these model plants have a greater emission
reduction when LDAR is applied, they also have a greater
recovery credit.
  7.1.3  Bulk Terminals
     7.1.3.1  Truck loading racks.  Capital expenditures and
annualized costs for the control of emissions from bulk
gasoline terminal loading operations were estimated for the
four model plant sizes presented in Section 5.1.2.  Three
                             7-9

-------
types of vapor processing systems have been included in the
analysis: carbon adsorption (CA), thermal oxidation (TO),
and refrigeration (REF) systems.  Based on conversations
with terminal operators and control equipment manufacturers,
these are the most common types of systems in use today.
Varying estimates were prepared based on assumed processor
outlet emissions (35 mg/liter, 10 mg/liter, and 5 mg/1) and
whether the installed system was a new unit or, in the case
of thermal oxidizers, an add-on system.  The costs presented
include capital investment, annualized costs, and cost
effectiveness for each type of control device for four
different throughput levels.  Table 7-6 presents the
estimated costs for a new unit designed to meet a 35
mg/liter outlet emission limit; Table 7-7 provides cost
estimates for a control device designed to meet a 10
mg/liter limit; and Table 7-13 gives cost estimates for a
new unit designed to meet a 5 mg/1 standard.  Tables 7-8
through 7-14 present costs associated with upgrading
existing terminal loading racks to limits imposed by the
alternatives developed in this analysis.  Table 7-8 details
costs for upgrade of uncontrolled facilities to a 35 mg/1
standard; Table 7-9 provides costs for converting existing
80 mg/1 units to meet a 35 mg/1 standard; Table 7-10 shows
costs of upgrading uncontrolled facilities to a 10 mg/1
emission limit; Table 7-11 gives costs for retrofit of 80
mg/1 units that will allow them to meet a 10 mg/1 standard;
Table 7-12 presents costs for upgrading 35 mg/1 units to 10
mg/1; Table 7-13 provides costs for upgrading 35 mg/1 units
to meet a 5 mg/1 limit; and Table 7-14 shows costs for
retrofit of 10 mg/1 units such that they will meet a 5 mg/1
standard.  Finally, Table 7-15 presents the costs of adding
on a thermal oxidizer to an existing system in order to
obtain improved emission control (from 35 mg/1 to 10 mg/1).
Manufacturers were contacted and previous EPA cost
information was reviewed to obtain the purchase costs
                            7-10

-------
H


D
 IT)

 PI
       CO
       a
       8
 u>
 EH    O
 co    en
 O    ai
 O    H

 W    K
 O    W
U    P
5=    a
M
D    a


§    I
      EH
P
CQ
VD
 I
       W
       O
a
CQ
                   g
                                 3  I
                        g|       2   £
                       SI      2   S3
                                     a
                   I
                       o
                       >-
                                5    ffi
                                M
                                eo
      •eco

in^}>oiO   •-        r\j   •*»   KI

Kl*C*O<\)fw>fO      fl
              M   in    •-   eo   N-   %i      K-
                            osj^-omKi
     *~        rv       rg   ^»   ^   pj
                                                           in   eo   N-
                                                                                           rvi     fo
                                                      *~;        *
                                                      rJ   >o   >o
                 J=


                  8

                 "5
                  CA
                  a
                       o
                       M

                       £

                       o   ^
                       8.
purchase cos


installation


ing Cost
                                                               _•
                                                                fe   s
                                                                              I!
                                                                         D.   X   W   „
                                                                 .
                                                  O.UXO
                                                                                           O)      O)
                                                                                          I      X


                                                                                          *      *.
                                                                                          —'      (A
                                                                                           o      g


                                                                                          I      >
                                                                                          U      4-*
                                                                                          o      SJ
                                                7-11

-------








EH
H
r-t
"S.
e
0
H --%
to
S K
g 5
S
1 S
w
EH O
W ff>
O ffi
U H

O W
U D
s a
H
Q 0
(^ f^
S S
< CK
2 0
H
C^ Q
EH s
o
8
§
cc
Kl




1
—
O
O
o


^









v*
§
O
o
s






>J
ra
•8
i
§
O
o
CO
ro








i rough put
f.
1
jj.
o
U>
CD
U
K
u.i ru rO
ml l£ O»
act  - r\j
< ro o
<-> OJ CM
V
(A
O
W C
o 5
(_ U •*-
O 4->
M 0) 0
« Vl —
tt CD —
o 4^ £ eg
O c u *-*
-III
M •" -
> « => o
o
W

Q
 <0



CO -O CO
0 ^ NO

N- in eo
NO NO ro NO


0- eo
f\i r\j NO NO

ro NO co
-J- ^~ NO




ro m co
«- N! ro NO




*- eo



•o


" o 2S
M 
u
«
fO
'Q.
n





ISl
*

o


£
•o
S

oo
00




0-
CO



•o
s

CO
CO


N
CO


ro





CO




NO


|
•l
(
3
8
i
tn






Kt
IA

O

«
IA
ru
s


o




rsi
CO
fXJ



NO
IO


o


NO
ro


in





o




-*



"**-
^
u
U

»
1
"o
C/l
(D





ry
S

5

5
1
CO
0


s




£
CO
in




ro

eo
N."
rs.


LA
f\J


S




•-
m




eo






S
o
I
"ffl
^


*
*
OH
K

4

0-
4

i

^
i




^
s



rv
ro
ro

N.
ro
ro


rs.
ro


(M




f^
ro




FN.
rO

X
i
o
u

§
u
n
M
O





                          s
7-12

-------


S3
O
H
W
1

O
u
us
1
0
2
O
<
S
H

H
^ i
^ j
o
in °
n o
a s
ICONTROLLED '
RD QUARTER 1
H
D-
1
S 6
o g
u 2
U g
^ O
K
Z ^
H
O
S

^
W
§
U
*
c
ffi




CO
1
r-
S
ca
rtl
fH







U.I (MCOfV CO «-'•<) ^ CO Q 0 v
§
M
0 to in co <6 in Kl -f
1" -S« - "-OSES K
CO


CM «0 
| g «, ^ 3, ^^~;o
O
S
*~ CO O>> CO O« 4Ti
VI f£ 5 S °~  m •*- Ii u J5 £ cj (fl •
£« _ ' w « « x « ;: (_ c u
ffl — ' U. o >* — «*— Xio — W^j
fok ^3 c § c **• S u c -^ o — • «SU
^ &. »- O ... I- C4>*-»4,CI """P
C.J? Jn*-'*-'-K(oOOi C i- ,ir -^- C»^ *

— +* O O <
O — D ^
w Q c -Q *->
as a c 3c»
O U < W 2
7-13

Os U"
""

"?
CM •-



•C
•* Kl
Ot O
CM C\J


to
N. CM
3 C

SO X)

Kl
3 S

*>
CM CO
•O
f\j N-
M in



X X
*^
TJ* **

— Cft
0 S
L. C
1 =
3 ^
(_ Ul
2 0 W
£§ S






-------
M
 e
in
 6     _
0     S
co     _i!
&<     j
w
M*«
S
M
Q
W
s
§
H
CO
£

w     g
o     S,
u     0
v     Q
M     «
        w
        Q
        Z
        <
        M
                           a
                           SI
Sii
                     I
                     s
21
                           Si
apor
                                                       N
                                            I         R
           S    i
                                           §
                           Ul         GO
                                             «CO%OPJCON<«^

                                             «~          IA   o   r\j   oa    y   «o   ^    N    o*   m       *~




«O               4    CO         infM^OCMv-





      Kl         Ift    CO    *      N
                             (M   rM               ^   r\i      ^

                                                                 «\r









Kt               >OeON-iArMKifMir»
                 •-          fU   03   Or           IA
                                                                N."


           «-          oD    o^   K>   oo    m    tn   eo
                                                             in
                                                                                                              in
                                           .
                                            C     u
                                            =X
                                                       IUO.U3CO         o»-C3
                                                                                   re
                                                                                   o
                                                                                   
-------



a
o
H

O
CJ

H
B
a
iH
6 «"
2 5
»J
8 S
Q 0
iJ ffi
•J rH
O
dj p$
H W
a EH
0 tf
0 <
5j »3
D &

1 Q
rt
Cfl H

W H
o
CJ En
o
o en
< o
w g
O W
a P
M O
D K
< EH
S ~
-
^"J
g
H
S
W
EH
J«5
1
D
to
O
H
3
»
<
EH












-
Ikl IM f*"! CVJ Kl ^ NO *~ 63 00 •<*• *O
wl -O O* in ^» r- oOS-^OK^S
1
« in co » N r\
O O[ N- «O (\J CO »- fO 'O O" Q ** O -^
O >— | K» »- in *~ ^ So n -3 *-
fM CO 4> «- m N-
ul p^iAirt KI tnnJin^Ki
OJ (NJ 00 CM in IM

•O - N-
acl M KI 4) N rw
i
"* irt CO K» «- «* N»
| a s o | - R « - ; s s o 8
™^"*^2~5
(M ro »o *- rsj s^

03 ^ CO fM C> fvJ >0 f^
s fe a | s ^-ag^21


ff
TD
*x
— !^ u-\ eo •- >J if\
§O O- »- Ch vfl NO Kl >O* fO 0 -J* 0 N-'
1— «— OKI «— Kl-^fl O
r- r- SO ^- f\J
o
&•


O^ oo N- Kl 00 -O r\j
< Nfl&-O> (\I rsJ^NQN-CQKlKl'O
u ?i^!2 " ^^r^^o



o eo ^ KI so ^ rxj
° oo o -o KI *— -o rvi 1/1 o r1- «—
J*- «— N.CO >j- «— oj *o -^ in N-
a KI ry xj *- *-

(D
•c
2i Kl lA CO *O O> NO
•c eorvj*D ^ N! KI NO oo •— i/i o iX
gOoo-rj «-orj^*
S
00
fO

- eo ON ry -o ^ ro
» oorxjNO (> r\jNONOKi«~o-r^rvt
*J fC S 0- rj»£Kim^£


1 s §
*S c « E1 S | '-s
O O ftl . -> *D t.
U O— C E (_4jWUO
O *^ O u) W On«>3
M OJ (D -^ ^j O «U J_ N" w '
g^^€£>".-^«&c g'°'w-ct"S
"Six ^3 c Oc'""*^1" c-^tj"* <6cu


uQ ^*-'<4-'^QUoJ5 C t_>r."~ V_32OCQUJQ.UZO L?h—U(g
•*- "ffl — *-> C
— 4-» CD O <
o — 3 *-•
(ft Q r -D «-*
(DOC DO)
O o < tri z

O in
oj rj

ON
NO N-
fU K)
KI f\»
ON
PO •-

^
S 3

>r
s s
Kl •*
S S

N.
Kl v*




fv.
— " "NO

Kl NO




K.
*— O




N-
(\i O
Kl <>-
*~ »
»—

^
IM P^.
2 g
•-"


t.
IM »V
*- O
^T

S f
— • V)
o |
i 1
0 «
o 8
>_ •*-
(_ uj
— • >•
(D «%. 4^

O O O
I- > 0





7-15

-------
H
2
e

o
g

o
oo

fa
o
§
o
&
D
M
H
W

8
O


K


U


M

D
U
03
W


03
        w
        K
        o
        ffl
        05
        W


        1
        D
        O'

        o
        K
        Ct,
        O
        a
        z.

        (A

        o
        a
                                                         o
                         *        fc   B
                                                    in
                    I
                         SI
                         a
                                                *O    CO    -O   Kl   IM   —         *t

                                                *-*O»-O^**O*   o*      *o










                                                in    oo    O;   nj   «-         CM   **



                                                                **   *~         •£           Jn









                                          «M             ^   <>      h^










                                                *Q   fO         *-   y>   O    CO   CM

                                                •^«o>ocot>inkn      co
                                                V         >*   ^-   fNJ   #-    -O   x*      2

                                                                «-              *"           ^l








                                                in   oo    ro   O;   co              CM

                                                Ki^^-incoosJin      *>f
                                                           in   fo              o<   •*•      r\j


                                                                                             CM







                                          co         co-ocooKsOh.ru

                                          MsO'OAJ'OcOO'OOm      in



                                                                                             f\J







                                                so   eo    rvj   h*   -o   eo    oo   
                                                                               co    rj    co
                                                                         co

                                                                         •o
                                                                                                    S
                                                                               w   ^
                                                                               ti   >*
                                    I    i
Unit install


Operating Cos
                                                               £          O    m   o»    3
                                                                         Si    L   O>    tf)     •

                                                              -^o5g.fe£««

                                                    i:  -.    fr^    01°'S?^    °
                                                    £§,"-£.££-«;?"

                                                    i:^i^S£2«-'8
                                                    8   £   1    .£    fe    5   a    S    S    .2
                                                    «'•<-    to    ra    O-   ti   ra    iD<6«^
                                                    uj   a.   u    X   o         ut-ts
                                                                                                         5       .-

                                                                                                         E       I
                                                                                                         +*       it
                                                                                                         g      i
                                                                                                         U       *•*
                                                                                                                  il
                                                    7-16

-------


H
r-4
X.
e

If)
1
o
H
§
^
g1

in ^
O $
2
! |_J
hJ
"&> ^
n °
H
° «
M W
1 §
O ^
DJ ^
D Q
1 J_j
W r^
8 o
«£ s
s <
CO
g §
§ ~
s

>fi
s

H
W
*
D
CQ
n
_j
1
r-
S
CO
<
EH










4
g
C
«•





I
~*
I
g
°i









I
—
o
0
£







(C
o
§
o
00
Kl






•g
0
Jc
*~
tf
C
~o
If
fO
a
KJ-O in •ccoovcjuico^ca
s 55 B.,,.,a. 5
O in CO fO N- O» »
^OKl ^-F. ONT*> 'O Kl M

h- N co *-Kicor^oo
jfl CNJ p K m COO ro^OKfOOOIMO «-
^~ K» in**«- p*"^2
IA


m oo eo»- ^6 •* 10 <*
^i ... „ ^ * ^_ * * „ ^ ^ ^ _J ^
Oi o^ "J1 fv -vj" QO r\) •& *~ "*t
fvi ftj •- IA

^-KI ^ *°. *°m CNi°:rJ. ^
Si 3& " - ^MKI 0; g
o"

•- u-»w-4-iAr^ inh-
O xj fO *O O Kl'O'OMOOO^ O
1— xjrsj rvj h^^^—O1 Kl
o"

r^Kl ^» 0 OJ^CJ«>fV N.
ru (V •- Kl


r^co <^ WD eo ro<— N>N-O>
1 ^^ *' ^^EGSRi-'S"!
o
N. in CO *0 CO O-
-*J_. KtKl «— CO Kl-OOOO-OO-fO F1*
'"* Kl •— OJ ^T ^O O*
*~ «- *- eo
sT


0>0 - OSCh^fM^ 0
t*' (M CJ *~ •- S
to
g
*" *"* X X Z
'l § I . .5 .^ i 1 ? 1
U u«- £ i_£Tfl«-O I-C
O *^« £ 0«W3 ^¥
2,lllir l^-ll-^s1 1
^S2C?'-Si'"C'-o--"'£U *-t "^
fe » 7; i *• S S B ".^ 8 ? "^ • -v ™
« * 'E 'E * — ~ o 'ra 8. 'S & 5 S 3 o S O
> ^33 !^UO!LJXO ^^ *J fr— t3 "IB ^ > u

"a — • ^ S
*- B 2


U < (/} Z
7-18

-------









EH
H
H

in
g
i-H

0 -~
«-( W
K
6k itf
O iJ

W Q
D D
S 0
&i &
D H
1 tt

W EH
EH K
O D
O O
X Q
O K
rtf H
p-j Jj^
EH
O
H O
Q
< w
S |
tJ W
< !b
2 O
H K
2 ^-"
w
f"1
S
g
£
w
J
ffl
EH













uil Klin •* £ >$ 3 M ** 3
eel in -» •- «\

i
IA co ro K »
O *-[ «2 »O w fC O>fr* <
S
10
N- ra co *- ro oo «
O ** K
<| AJCO 00 rO*PO OO fO'O'OtOOOO
OK-I^OJ rj ro-j'*- o>^
o »— «—
o
&•


r«- ro ^f o* co^~AJO"^-i^
ulcOvfAj roeOAV-j-
AJ AJ <~*


h-CO O ^)QO Kl^Kl«-
** 1 lAO N! «-"OKlON-OO
lt| *0 •" ^-ro^rONO
auj ro ro ^ —

1
^ (s. in eo eO eo
o -°oj roro «-oo ro^ooooog
si
v| c a s ~ - - s s s - a



« ^ JJ
*! t» S •*
to o f^1 *-* jj
, S - u x> „ .-
t; c c g = g ?
O 5 flv — -X • L.

M AlB^ U .Q j_ O) W •
8W — M ajjctOait-CU
« — • o > —q;— X«3 —  c o> ° u « o
^^OirWoOffll-WC^ g^O
QtLg7Cc>"O) C •— y — * flJ C
Xt.{* ^4-'*JO*-''D(*^(/l— 'X
£» CC C 4;—-.- (Q (13 Q.S«J««> —
> — 3r>£uja.uio u i- u re
? <= •£ ?
•S "5 - 2 £
— *•* c o <

(fl a c .0 *^
a a c 3 A'
O U -C V) 2

**
AJ
S

in
AJ
5Q

in
-0 g
o
N
Kl Kl
S
*
AJ
S
Kl


AJ
K) 4

5~

•O
- «
to
JS
AJ

*0
*- Kl
S
S



-0
*— ro
^O
°^
an

O
O IO
ro
00
AJ


o 8
O **-
>_. **-
_^ t- UJ
** O V)
25 S






7-19

-------
TABLE 7-15.  BULK TERMINAL LOADING RACK COSTS - THERMAL
                    OXIDIZER ADD-ON

       (THOUSANDS OF THIRD QUARTER 1990 DOLLARS)
Model Plant
Vapor Processor
Capital Investment
Unit purchase costd
Unit installation Cost*
Annual Ooerating Costs ($/vr)
Electricity
Pilot gas'
Carbon replacement8
Maintenance*1
Operating labor1
Subtotal (Direct Operating Costs}
Capital charges' (16.3X)
Taxes and Insurance (4X)
Gasoline rec. credit1*
Net Annualizcd Cost
Total VOC Controlled, Hg
VOC/yrm
Cost Effectiveness, S/Mg
VOC
1
12!

35
29.8
1
2.0

3.5
6.8
13.3
10.4
2.6
0.0
26.3
1.0
26,300
2
J&

35
29.8
1
3.4

3.5
6.8
14.7
10.4
2.6
0.0
27.7
8.0
3,463
3
is!

35
29.8
1
6.3

3.5
6.8
17.6
10.4
2.6
0.0
30.6
16.0
1.912
4
is!

35
29.8
1
8.3

3.5
6.8
19.6
10.4
2.6
0.0
32.6
32.0
1,019
                          7-20

-------
FOOTNOTES FOR TABLES 7-6 THROUGH 7-15

a Carbon adsorption unit.

b Thermal oxidation unit - enclosed flame.

c Refrigeration unit.

d Costs for MP1, MP2, and MP3 are based on same units for CA
  system.  Differences are due to the amount of carbon in
  each system.

e Estimated at 85 percent of control unit cost.

f Estimated that 50 percent TO units used propane and 50
  percent used natural gas; price of propane was $1.03 per
  gallon and pilot burner was estimated to burn 2 gallons
  per hour.  Burning an equivalent amount of natural gas was
  estimated at $0.80.  Final estimate is the average cost
  for propane and natural gas.

8 Estimated activated carbon replacement period is 10 years,
  at $2.09 per pound carbon cost.  Estimated carbon in each
  unit:

  MPl - 10,000 Ibs.
  MP2 - 14,000 Ibs.
  MP3 - 18,000 Ibs.
  MP4 - 25,000 Ibs.

h Telecon with John F.  Jordan Co. (Reference 22).

1  Daily system inspections at 1 hour per day.  Labor rate is
  $20/hr.

J  Total capital investment x (capital recovery factor +
  0.04),  where interest rate = 10 percent, equipment
  economic life = 10 years (0.163 capital recovery factor).

k Amount recovered per year,  at $0.342 per liter assuming a
  density of 0.67 kg/liter.

1  Calculated assuming baseline uncontrolled loading (see
  Table 3-11); i.e., 94 percent times the submerged loading
  factor, 658 mg/1, and 6 percent times the splash loading
  factor, 1,590 mg/1 (see Table 3-8).  These factors are
  based on an RVP of 11.4 psi and 60°F, as discussed in
  Section 3.2.1.2 of Chapter 3.  Emission reductions are the
  difference between this weighted average factor, 713 mg/1,
  and each controlled level,  multiplied by the model plant
  throughput.
                            7-21

-------
m Assuming existing control device meets 35 mg/1 emission
  limit and VOC controlled calculated using emission
  reduction factor of 25 mg/1 (35 mg/1 to 10 mg/1).
                            7-22

-------
presented  in these tables  for  carbon adsorption,10'11  thermal
oxidation,12'13  and refrigeration type14 vapor control
systems.
     For the carbon adsorption system, one manufacturer
stated that essentially the same unit could be designed to
handle the throughputs of  the  first three model plants.  The
only difference in these systems would be the amount of
activated  carbon  needed for each system.15  This  same
manufacturer estimated the amount of carbon for a  10 mg/1
unit for MP1 at 10,000 Ibs., MP2 at 14,000 Ibs., and MP3 at
18,000 Ibs.16  MP4 would  require a larger design  to handle
the throughput, and a separate estimate was provided for
this system.   The price of carbon is estimated at  $2.09 per
pound, and the carbon is assumed to have a working life of
10 years.17  These sources  also indicated that retrofitting
a carbon adsorption system to  comply with lower emission
limits increases  the capacity  of the system by at  least 20
percent; and feasibility studies indicate that in  most
cases, installation of a new unit is more cost-effective.
Therefore, retrofit was not considered to be an option for
carbon systems.
     Similarly, for thermal oxidation systems, the same unit
could be designed to handle the throughputs of MP2 and MP3,
and the unit price estimate for those two systems  is the
same.  Installation costs  were assumed to be 85 percent of
the unit purchase cost, which  is consistent with the
findings in earlier EPA studies.19'20
     Annual operating costs include electricity to power
compressors, pumps,  and blowers, routine maintenance and
operating  labor (daily inspections), pilot gas for the
thermal oxidizers, and activated carbon replacement  for the
carbon units.  Operating labor consists of a routine 1-hour
inspection per day at a labor  rate of $20 per hour.  For
carbon systems, the estimated  maintenance cost is  $6,000 per
year, including parts and  labor.  The annual cost  for
thermal oxidation units is $3,500, while refrigeration units

                            7-23

-------
are approximately $11,600 yearly.21  Thermal oxidizers
require a pilot fuel source and, based on conversations with
manufacturers, it is estimated that half use propane and the
other half use natural gas.22  The current cost for propane
is approximately $1.03 per gallon.23  control systems are
assumed to burn about 2 gallons per hour.  The cost of
burning a comparable amount of natural gas is about $0.80.
The estimate in the tables is the average of these two
figures.
     Other costs include capital charges, administration,
taxes and insurance, and the gasoline recovery credit.
Capital charges are assumed to be 16.3 percent of the
capital investment, while administration, taxes,  and insur-
ance charges are 4 percent of capital investment.  The gaso-
line recovery credit is the amount recovered per year at
$0.342 per liter (see Chapter 8), assuming a density of
0.67 kg/liter.  The total VOC controlled is the difference
between the uncontrolled and the controlled emission level.
The cost effectiveness is defined as the total net
annualized cost divided by the total emissions controlled
per year ($/Mg VOC controlled).
     7.1.3.2  Railcar loading racks.   Table 7-16 presents
costs of installation and operation of three vapor control
systems, all achieving an emission rate of 10 mg/liter for a
railcar loading operation.  Based on observations of a
railcar loading facility,24  it was concluded that railcar
loading occurs at a rack with similar operating
characteristics to that of model plant 2 for tank trucks.
The yearly throughput for the railcar loading rack model
plant is estimated at 85 million gal/yr with a maximum
instantaneous loading rate of 3,000 gal/min.
     7.1.3.3  Tank Truck Leakage.  As discussed  in Section
4.1.4, there are two basic options for controlling vapor
emissions from tank trucks during loading.  These include
installation of a vacuum assist vapor collection system at
                            7-24

-------
TABLE 7-16.  RAILCAR VAPOR  CONTROL COSTS FOR 10 ing/1
             (THIRD QUARTER  1990  DOLLARS)


Cost Item
Capital Investment
Equip Purchased
Equip Installed
Rack Converted
Railear Converted
Total Capital
Annual Costs ($/yr)
Electricity
Propane
Carbon
Replacement
Maintenance
Operating Labor
Tank Test
Taxes, Insurance,
and Admin. (4%)
Total
Recovery Credit
Capital Recovery (16.3%)
Net Annual] zed Cost
Total VOC Controlled, Mg
VOC/yr
Cost Effectiveness, $/Mg
VOC
Carbon
Adsorption
(1,000 $)

246
209
639
21
1,115

12
-

3
6
7
-

45
73
130
182
125
332

377

Thermal
Oxidation
(1,000 $)

106
90
639
21
856

6
3

-
4
7
-

34
54
0
139
194
332

585


Refrigeration
(1,000 $)

387
329
639
21
1,376

11
•

-
16
7
-

55
89
130
224
279
332

841

                         7-25

-------
the loading rack and implementation of a periodic vapor
tightness testing program for the trucks.  The total costs
to design, purchase, and install a vacuum assisted system
were estimated by Fina Oil and Chemical Company to be
approximately $320,000.^   (These costs may differ markedly
from what another facility would have to spend for a similar
system, due in part to engineering resource expense involved
for site specific parameters and refining of the system.)
The estimated breakdown of costs is as follows:
     Equipment
          blower/motor                  $25,000
          control valves/actuators       40,000
          air compressor/drier           15,000
          PLC modules  (computer)         18,000
          electrical equipment           15,000
     Contractors
          design                         60,000
          installation                  120,000
          facility refinements           27,000

     Contacts with various tank truck manufacturers
indicated that, on average, the cost to install vapor
collection equipment on bottom loading tank trucks is $3,500
per truck. '   Also, any gasoline tank trucks or railcars
operating at bulk terminals affected by the proposed
regulation will be required to have annual vapor tightness
testing performed using the EPA Method 27 test found in  40
CFR 60, Appendix A.  Method 27 contains both pressure and
vacuum tests to be performed on the cargo tank.  The annual
DOT test, which consists of only a pressure test, considers
the pressure portion of the EPA Method 27 test as an
acceptable alternative test.  Contacts with various vendors
that perform these tests indicated that the DOT test costs
approximately $200 for a 4-compartment tank truck, while the
complete Method 27 test costs approximately $350.  As a
result of this proposed regulation, tank truck owners who
were paying $200 per year for a tank truck inspection would
                            7-26

-------
now have to pay $350 per year.  Consequently, the cost
impact of this proposed regulation is the difference between
these two costs, or $150 per year per cargo tank (tank truck
            00
or railcar).
7.1.4  Bulk Plants
     In order to obtain up-to-date cost estimates for
retrofitting bulk plants, a wide variety of organizations
was contacted.  These included petroleum marketers trade
organizations, oil companies, State environmental agencies
that have recently adopted Stage I regulations, bulk plant
owners, and installation contractors.  Information
received29'30'31 showed that the costs of installing controls
at a bulk plant are very close to the costs presented in the
Draft Regulatory Impact Analysis: Proposed Refueling
Emission Regulations for Gasoline-Fueled Motor Vehicles,
July 1987 report.  Since the costs from 1987 provided
detailed cost breakdowns, the costs given in Tables 7-17 and
7-18 are from the 1987 report updated to 1990 dollars, using
the CE Index.32
7.1.5  Service Stations
     The same organizations contacted concerning bulk plant
control costs were contacted to obtain current information
regarding service station Stage I costs.  In addition,
several service station owners were contacted.
     Additionally, industrial contractors were asked to
provide cost estimates for retrofitting service stations
with Stage I vapor recovery equipment.  Several of these
contractors responded with estimated costs.33'34'35  Based on
these estimates and an analysis of catalogued costs, the
average capital cost given for retrofitting a service
station with a coaxial system is approximately $1,524.36
Also, the contractor estimated cost for installation of a
dual point system ranged from $800 to $3,500 per tank, with
an average of $2,323.37   Since facilities  examined  in  this
analysis typically have three tanks, costs would be $6,969
                            7-27

-------
       TABLE  7-17.   AVERAGE CONTROL COSTS  FOR  BULK PLANTS
                                  (NO  EXEMPTIONS)
                        (THIRD  QUARTER  1990  DOLLARS)
 Model Plant No.                    1           2.34            5

 Throughput (liters/day)           1,500       11,400      24,600       47,300        64,400

 Weighted Average Top & Bottom
  Loading Costs

 Balance Incoming & Outgoing
Loads on Uncontrolled Plants'
Capital Costs"10
Annual 0 & M (3%)
Capital Charges (13.1%)
Taxes, Ins. (4%)
Recovery Creditd
Net Annual ized Cost ($/yr)
Emission Reduction (Mg/yr)
Cost Effectiveness ($/Mg)
Balance Outgoing Loads on
Plants with Incoming Load
Balanced
Capital Costsb'°
Annual 0 & H (3%)
Capital Charges (13.1%)
Taxes, Ins. (4%)
Recovery Credit
Net Annual ized Cost ($/yr)
Emission Reduction (Mg/yr)
Cost Effectiveness ($/Mg)

31,208
936
4,088
1,248
200
6,073
<1
6,073

23,227
697
3,043
929
200
4,469
<1
4,469

31,208
936
4,088
1,248
1,512
4,761
3
1,587

23,227
697
3,043
929
1,512
3,157
3
1,052

31,208
936
4,088
1,248
3,277
2,996
7
428

23,227
697
3,043
929
3,277
1,392
7
199

31,208
936
4,088
1,248
6,301
28
14
2

23,227
697
3,043
929
6,301
(1,632)
14
6-'

31,208
936
4,088
1,248
8,572
(2,300)
19
_e

23,227
697
3,043
929
8,572
(3,904)
19
_e
a  Includes the cost of retrofitting two account trucks for use in vapor balance service.

b  Top Load Cost - $21,310 (91%), Bottom Load Cost  - $42,610 (9%), Incoming Load Cost - $7,981.

c  References 2 and 19.

d  Recovery credits are based on a control efficiency of 95 percent on outgoing loads from a balance
   system (or storage tank emptying losses), and a  product cost of $0.30 per liter.

'  Cost effectiveness not calculated because net annualized cost is a negative quantity (cost
   credit).
                                          7-28

-------
      TABLE  7-18.    ESTIMATED CONTROL COSTS  FOR BULK  PLANTS
                           (EXEMPT  <  4,000  GAL/DAY)
                         (THIRD QUARTER  1990  DOLLARS)
 Model Plant No.                    12345

 Throughput (liters/day)            1,500      11,400      24,600       47,300      64,400

 Weighted Average Top & Bottom
   Loading Costs

 Balance Incoming Loads and
Install Outaoino Submerged
Fill on Uncontrolled Plants
with < 4.000 aal day"
Capital Costs*1'
Annual 0 & M (3X>
Capital Charges (13.1%)
Taxes, Ins. (4X)
Recovery Credit*5
Net Annuali zed Cost ($/yr)
Emission Reduction (Mg/yr)
Cost Effectiveness (S/Mg)
Balance Outgoing Submerged
Fill on Plants with Incoming
Load Balanced < 4.000
gal /day3
Capital Costs'"
Annual 0 & H (3X)
Capital Charges (13. 1X)
Taxes, Ins. (AX)
Recovery Credit*1
Net Annuali zed Cost (S/yr)
Emission Reduction (Mg/yr)
Cost Effectiveness ($/Mg)
0
0
0
0
0
0
0
0


0
0
0
0
0
0
0
0
4,270
278
1,214
371
1,313d
550
4.4
1,587


1,308
39
171
52
358
(96)
1.2
_r
31.208
936
4,088
1,248
3,277*
2,996
7
428


23,227
697
3,043
929
3,277
1,392
7
199
31,208
936
4,088
1,248
6,301e
(28)
14
2


23,227
697
3,043
929
4,358
311
14
22
31.208
936
4,088
1,248
8.572e
(2,300)
19
_f


23,227
697
3,043
929
5,970
(1,301)
19
j
3  Includes the cost of retrofitting two account trucks for use  in vapor balance service.

b  Top Load Cost - $21,310 (91X). Bottom Load Cost - $42,616 (9X), Incoming Load Cost - $7,981.

c  References 2 and 19.

d  Recovery credit based on control efficiency of 58X for conversion from top splash loading to
   submerged fill.

"  Recovery credits are based on a control  efficiency of 95 percent on outgoing loads front a balance
   system (or storage tank emptying losses), and a product cost  of $0.30 per liter.

f  Cost effectiveness not calculated because net annualized cost is a negative quantity (cost
   credit).
                                          7-29

-------
per station.  More recently acquired information has
reinforced these results.38
     Information on the owner preference of a coaxial versus
a dual point system was not available, although each system
has its advantages (coaxial - low cost, dual point - ability
to drop two products at the same time).  For purposes of
cost estimation, an average of the dual point and coaxial
costs was used.  There is no vapor recovery credit
associated with service stations due to the fact that no
vapor recovery devices are used and if vapor balance piping
is used, vapors are returned to the truck tank for recovery
or process at other subcategory facilities in the network.
Table 7-19 provides a comprehensive analysis of the costs
associated with the service station subcategory.

7.2  COST ANALYSIS OF REGULATORY ALTERNATIVES
     The costs of control for each facility emission
source's control option(s) were calculated by multiplying
the facility number or gasoline throughput shown in Tables
3-11 and 8-27 by the appropriate model plant costs.  The
model plant costs used in the calculations are those
discussed previously in Section 7.1.  Cost effectiveness
ratios  ($/Mg HAP, $/Mg VOC) were calculated by dividing the
control option net annualized cost by the HAP or VOC
emission reductions achieved under each control option as
discussed in Chapter 6.  The capital and annualized control
costs, HAP and VOC emission reductions, and cost
effectiveness estimated for each control option at both new
and existing pipeline facilities, bulk terminals, bulk
plants, and service stations are presented in the following
tables:  Tables 7-20 and 7-21 for pipeline facilities,
Tables 7-22 and 7-23 for bulk terminals, Tables 7-24 and 7-
25 for bulk plants, and Table 7-26 for service stations.
                            7-30

-------
      TABLE 7-19.  SERVICE STATION STAGE I CAPITAL AND
                 ANNUALIZED COST ESTIMATES3'11
                 (THIRD QUARTER 1990 DOLLARS)
 Capital Cost and Installation0
 Annual!zed Costs f$/vr)
      Maintenance (3%)
      Taxes, Insurance, and G&A (4%)
      Capital Charges5 (0.131)
      Annualized Cost
      Recovery Credit
      Net Annualized Cost
$4,250

   127
   170
   557
   854
    NA
   854
Throughput


MP1 ( 7,600
MP2 ( 23,000
MP3 ( 76,000
MP4 (132,000
MP5 (246,000
MP6 (700,000



1/mo. )
1/mo. )
1/mo. )
1/mo.)
1/mo. )
1/mo. )
Emission
Reductions

0.138 Mg/yr
0.407 Mg/yr
1.343 Mg/yr
2.341 Mg/yr
4.347 Mg/yr
12.370 Mg/yr
Cost
Effectiveness
($/Mg VOC)
6,188
2,098
636
365
196
69
a Since the number of underground storage tanks at service
  stations does 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.
b Capital charges are based on a 10 percent interest rate
  and eguipment life of 15 years.
c Average of rounded costs for coaxial ($1,500) and dual
  point ($7,000)  systems.  References 25, 26, 28, 33, 34.
                            7-31

-------
O
eg
 I
CQ
I
s|>
z>
g
z
III
gj»*
<
^
f
Ul
3»*
S^S
l-o


*C *"~

ȣ ^
2°£
M|&

2*
" 2 C
« r X
£ £J •v
^S z
|B~
Ul
5 v>
uJ§ ^
<2 w -^
2£5>
|S


£
§5
U. _l
O C
cc
Ul t~
0. 3
1- C
oin So oooo
v*in co«~ coo^co^
f^ ^- CJ«M ~- K i? o <^1 ^3 o c^ in P*-


Kl «- »- «-
ro v^ ^
A w «~ % i*-i  "*
g g ,, p g g 5 S
0 K> CX « ^
oo oro oomKi
rorsj rsiry N-»~K»C\I
o >o S- •»» *«t K>
in



o o o
£ P{ g
,
I * g 'E'S: « «
•^ « o w A' u' K o i a: i c
a o"'* IAQ.X ^^-S0^^
4-- _ix«> • • -z -'*•£-'£*•
f/j «•« ^»^C •^OO*-^
X&.4-I « 4J— ^o C 10 -^ C — ^ «
4) O D 4J -X X C 0 A) O 3 (0 O O :
_i x o L. c uj — i _i i o x. x. x a
a. CD a
C u C 41 4J
•^ E tf~ Dl E

01 •- u u. —
.- 8- £• 2 8-
IX Ul "a. Vt 111
 I

i

 4-t

I
                                                                                                                      X

                                                                                                                      1
                                                                                                                      (A
                                                                                                                      S
                                                                                                                      S
                                                                                                            en
                                                                                                            ui
5   .    T>

 s-s  .  s
 O  41 _^  0
    •» • -c
_     4) It
 «  X «/>
4->  C     (0

•5.-S iT'o
 •T c a  c.
 U  O E  v

 o  8T  g
Z  «/> O.  
-------
W
EH
W
W
O

g
2
O
u

w
Q
H
O
H
H
E-i
H
^
H
U
w
55
H
CM
H
CQ
O)
£
Eg
Oi
u
2

^
I


1
1
|
1


iii


< **

ii°
o ^^
Z tt


2§
w £ £
« C_> X,
* 3 O)
|£~
*-
§2
u £
is
U. _J
°g
& 0
1- O

e I



o o
in -o
(M ^



§ §
in l>-

1 !
Kl ~


O O
0 00
Kl «-


0 0


*
? « i fe
'5. ID -c **
•^ 1
! »
O- Ul

-° § g
*"


.a o o
•- in
«- «M



•° § §
in o

in KI
«- Kl


•° § §


-° "o "o


JO X «
0 0
*O s»


jo o o
ir\ Kl



o o
•o oo





s
4-f
" ~
*-• Sr "
i/i 5 oe
S 2
1 S £ ""
aw a JB o
* £ w I *
S 0 |
"— O) £
— to a.
U 1. —
. £ 3
Q. (/> Ul

i
CM
w


g
CM



|

i
in
C


o
0


t


8


O











oc
o
.J
X
Ouarterl


> «\

0
^^


o
^-



i

1
^


in


o


X


Kl













V
1
i











2
6
1
|
o
H-
"o
z
§
u
1
§
c.
o
"£
t_
Q>

2 **
L. °
'i "S
*" 0
ii
c u
° B>
£ V
III
JC ^ ~*
.*" - fe
C C/9 4^
ft fc
"8*2.
.2 *• B
<• '^> >
" -8 fe
*•**'?
° 'i^ i"
!!•
« .0 u



-------
O]
M
 I
I
to
!§
UJ
i
s
I
1
X °-
¥
l
€
V)
l«t
-J £
- Ǥ
CJ ^
. Q. 0
*S£
ilk
IP
UJ
UJ * t
52 <« ^
5 u> °
|£5


FACILITY/
OPTION
U. _l
° i
>- C
1- U
o o o o •«.
CM h. N» — y
co >- <- eg 2


o o o o o

1 § § i -
IT! ,J CN» rn
§000 o
ooo eo
CO Kl O O
' ^— ^^
go o o o
ooo o
M Kl O Kl >O
j* ^ in" c> in
o o o o o
o o o o o
o o o o o
KT »^ cb" M ro"
fO ^3 W~ *O >O
CM ?• ^-
ro c\j «- in co
c^ o oo cv in
o o o o o
o o o h- in
in in CM o co
c\j <\i rvi c\T
° 1
*o o-
«M -*


CL Q. V
o in o
«», a co co
g> ai o a> » x>
< < < in a
O "^ ^^ ^x t 1
CA«— O O O (0 — * — *
Ji «-»-«- Jt <0 JO
U <-> i i i C C C
a o L. L. ucout.
DC _O O O 1- 4) 41
o) — ' eg co ro 41 x c
C
vi a. >.
DO) C
SS a
I-1






















34

-------






.-*
elude
c
o
u^


CNJ
OJ
1



W
, ~\
«

t^

























!
*8
if
s
r
K
I
IBS
<*"*«—
w

< *

a o
u 3





o. o
X «£ LU
X <*:



g
W § T
"^ ^
*i£
S*"

UI
ril£
crt 5 .C
m «° 51

f^* £

z






25
O 1-
fife
U. _l
UJ H-
£g
>- 0
888 ill §
°. *~. m. "~ c *~ c

!-§?- 'I- I
o o o o o o o
>o in o» CM cvj tv ^
CM «- K) ^- *^
o o o o o o o
o o o o o
o c\j Ch ro ro
cu rsj KI ^ *^
§O O *"" O JO C5
o o *- o o
«o in in **
*o & to «-"
in rvj ^

o o o '«'«•«' "°_
o o o o o o o
o *- «-

sj- fM fU
in ro ro
(\j





«— in P- o K ^o o
co ro ^o ro «— «—




§00 o o o o
0 *- 000 0
-41 o csj O^ «— in ro
rJ" ^-" «-" T-*



1 §
M K\






01
4^
L.

D> tfi 0£ Q Of
10 — — • <_JX in— o 3 coocoo
«- > CA < g^1 a xzi-j

C 3
10 CT
t— UJ
n *% r-











.

c •
'> i
to to
tf) 1-
o»
«-> o
O I-

ffi C-
.2 i
&>
«-> L.
8-0
" »
ID C
c -°
C 4-*
• $
V ~ •
SO X
T3 —
ID C
O .* O
ft) ID
•* i 8 .
-p o t- u •
» 07
« ffl C — > O
IA — O 4J
. S 2-fr X5
— U B) ft 6 3
O • > O  — • •— o a u — •
co « 4-» jr s- co
«> o — TS. u
X to o a *•>
k *4- 4-» W C *^
fx <•- — — v o
i- u> a. o e z
a o i- S
E 4-* U 4^ U II
(P Z O O O C O
in a. u z o — z
« ^ u *Q « ^- M




-------
at
*"" o
UJ P
35*
g

^
1
CO 	
EH
co S
o *
J il
(^ t_J
W z
O
« &
EH S
55 01
O €
u «
W
D 	
H ^
O z ° g
H < ° —
EH
55
rn -1 S»
£ li§
S < 8 °
55 3 C
H
U
C_i a. o
" X ^ UJ
3= QC
ID
CQ z
o 2

w £o£r
s Jl§
•
n
V *M
5§E
S i « o>
CQ o. S E
EH =






>;
— g
•P»4 ^-

f 5
O O
oc
UJ 1—


H- (_


§ § 1 s g s -°
Ki1 «\T -^ *** ** ™
*~

§ 0 g g g g „
>O cx in >o *O N-
ro fv iv ro ro fv



°§ ^ g § | § -°
s" s* s fe « 5


i i 1 § § § "°
CO -O O N- t«- «-
in *^ *^ in in «^




go o o o o xt
o o o o o
Ki vj- r- o to
M O- N. Kl Kl •
KI eo KI sj- «* co
in Kl Kl «•» «-


»«»«>«»« H »« XI
»O ^ (\J Kl Kl «—
«5 h- r^ rj tM  l-
L. U U <
< O < < v»
win >s «- ~»vo«
ji in in in «- .x

oiDcaut-i-a
tt _O O _O ^O t-
o) — * CD—' eoacoo;
c
— a


o «->
_i i/>


2 ^cj o xi
M Z S
in" ^"


§ 0 g „
O O CO
~ ~ ^"



II g -°
a" •»* ft


o o o
ro c\j fo




g g g -°
^~ v— >O
O in «^


o o o xi
§o
0
Sro
IM


XX X XI
r- J\j T-

0 O OX)
Kl CO C\J





o


c.
r+ O
_j *«^
— CD


t> V 4J
Ol « (A
(0 .- .»- _.
ji « vi a
a 10 in 2
^l < < C


1 1 11 i i
H* > ^ O W 4



a

7-36

-------












-5"
T>
l
r~i
u
c
o
u
""'


,
f)
CM
1
^
w
en
E-i



























f
*" y
QJ V
x *
uj
a:
u
z
1
1
|i
S
£

&

en
._..!..
^
J 4A
a: 8 °


_j ^

M I/! °
Q^ g) O
5 u 3

-«-„-*.
ao
X tt


w I r
«r >.
£ 2 £>

^~ fy
*


IU
5 to
slj
™ 12 x
£5~








J;
Bs
ct
§IU ^-


1 § I °


° i § °
III0
CM CNJ IM



O O O O
O O
00 Kl
	

O ^} O O
5 S



« 0) « 10
oooo



.____ 	 _.

>~e K x ^f
§K\ P^ f*
•* ro »-


oooo
fO f^J "^ O
N- in NJ" t\J







o
 4-> O 1_ O
c c < a
V O X 3 X
E X Z O £

'5
cr
UJ




1
1
5
o
H-
.2
i!

8
H-
g
(A
eo
1
g
«T
o
H-
CJ
CJ
x;
. •-
i ~6
L. -M
»s
h.^
c- ^

Q. "ol
£1
g S
._ (/>
^ o
«.!:
4-» 2
•S S"
L.
SJC
U
£*
3 Q. R
•g z o
4J O U

'3 w in
O O i
« c u. •
S 3 .2. TJ
3 ID **
« « X «
8 S § 3
0 •- I. —
4-> M- O
_ — o
co -• v
•- 'o S o
Q. 0 1 C
5 "^ u II
0 O C 0
Z Z — Z






































(0
V
u.

-------
cm
§


o
&
H

E-i
W
H
X
€
i §
Ul
i
Ul
ee S X
1-
!
|1
1
Ul
£^|
> *» =
— l V»
£5°
< o E
«-> c
»t < S
x oc
m o *"*
V) >* (
40 J^ >.
Ul
5 «>
ill

TYPE OF FACILITY/
CONTROL OPTION
"O O
•-*

•O CM

O O
e s
<0* JO

§ §
r^ in
•o" r-.*
0 O
•O in
O>* CM
§ §
K 0
•* ?\
in •-
CO CO
0 0
o o
S o
«-* r\T

Storage Tank Loading
Vapor balance
Tank Truck Loading
Vapor balance-
no exemptions
o o o o o o o
in o o oo o co
• » » « •
•O in 00 Kl •-

O O O O 0 O 0
in o o o o «- oo
CM O> «- fM 05 N. «-
Ki in in Kl
§0 o o o o o
O O O O O Q
in o ^ N. •— «
^* ** K- ^1^ CM CM
co eo r>- -o IM

o o o o o o o
o o o o o o o
*O CO O f> N- «— *O
~»" in" N-" o* r^" o rj
co co r^ -o -«-
o o o o o o o
§00 o o o o
Kl «- CM »
in «-^ S-* c\T o co i^
o o o o o « %
0 00 00 00
C\J *- O O O

*- ** cxj
§ §
S co_
N"
IA
a «i
Ji U
Z 2 £
* TJ 0> ««
o ci S c <
0 0> 0 ±- 2 Q.
C CO— to-* ^^^ <-J
O^ O C9O  03O.I CO £4^
8.&3«f28Xf-ZC/) a.
1 '1
1- Ul
7-38
 "a.
 o


I

 V
 u
 0)
 4->

-S
                                                                                                                                          JC
                                                                                                                                          4-1


                                                                                                                                          '5



                                                                                                                                          "8
                                                                                                                                           Q.

                                                                                                                                           IB

                                                                                                                                           U

-------






g
CQ
O
O
i
s
EH
O
0
w
Q
M


^~f
O
H
S3
W
EH
|
CQ
1

•
IT)

u
g
5 ^ y
UJ ^ g
o>
3B <
1
I**
>5r*



_ > 5
S *• o


^ *~
w

t *"*
5; V»
*~ *•*
X tf.

«/> 2 C
12 *- >•
£=»
& S N-
X


Ul

111
» 12 i
< B ^
z






•5s
ii
U. _J
°8
Ul t—
o. z
>- o
t- u
o oo oo oo oo
•o oin °z: ocvi into
KI «-CM oeo •*\o in «-
*«'*' W-'K? >/^ i\Ti\r
«- eo «- K- «- r\j
o oo oo oo oo
o oo oo oo oo
N. in>o mco st co «— ^o
*O N- -^ mm r\to ^ fvj




too oo oo oo
*-o* eo»n OJKI ^rj

*^ r\T oJ* c\T eo •-


o §§ °° °° coo
^ ru r\j
CD COO ~»CM 5(M OKI

^o oo oo oo oo
KI eo«- >CKI into coco
f\l CM CM ^ sj-







§00 o
•J- M «-
CM Kl «- Kl



VI
JrT
U
a ai
u
en co
0 C " g
? ? 1 C ? 0) tK 0
•V-CUDIO oovcg c <^
•DUCU O-^O>O ••- O^
coC'^c c 4-1 m — _. •o-- _i_».
«J— CO— 'C— »E"tD*J3 O3 ^Xl-
mogoroC>>->3t-ZCO- 1- LJ
7-39






















i
1
a.
1
L.
1
1
4-*
u

^
•o
ID

—
4-1
'5
1
CO
u
o
Ul
CO
a
«
4-*
u
"5
4^
1
o
z

«




-------



CO
El
W
8
A
A
o
o
0
pq
Q
H

jjj
O
H
EH
rff
g
g
O
H
EH
<
EH
W


W
U
H
£
MH
w
w
SS
W
S
Q

O
H
H
X
H


•
VO
CN
1
W
PQ
g











1
z "
Is

If?

1
SS
1*
|


u
(S
5 r Q'
UJ ^. 5
«



N




-j i?
< *•

§Q
°
O ^


a. o



2-
w ° r
•2 i- x
V U X.
•"g 5
la

z
— z ^
UJ 7~ v
JJ eo v
« ~ i

^




Jj
H-
5°
u ~

u. o
u. -J
CK
U t-
Q. Z

t- O
g
h^-
in

o
cS

8



Q
O
o
>0~
^™



o
o
o
rsi
in
*~



o


eo*
tc



00


o
fO
in
o"




e
°i
o
^



vt
g
t-
*" "c c
UJ O O
u t-
*> -a a
oc c £
Hi 5 C
c/> o x
U 0) C
u a> c
Z i. -^ O
h- "O — ^
(A C ••"

X
UJ
o
K>


O
M

in



o
Q
O*
in




eo"





o
o
o
hn"
j*



S


8
eo













1
4-»
O
QJ

•£
•5





g 5
o «~


0 O
Jo ^-

te M



o o
o o
^ ^»
in - a
S -5 S- »
c e x
ui 5 u 41
u o x
x 1. 0) tl f
i? ?-^ o •-

-------
7.2.1  Pipeline Facilities
     For equipment leaks at pumping and breakout stations,
alternative control techniques are based on EPA's LDAR
modelfor monthly and quarterly monitoring.  The costs
associated with monitoring pumps and valves in light liquid
service have been described in Section 7.1.2 and are assumed
to apply at these facilities.  The total component
populations (10,600 pumps and about 116,000 valves for
pumping stations and 85,500 valves and 7,200 pumps for
breakout stations) were multiplied by their appropriate
associated costs to estimate the annual totals.  These
component totals can be arrived at through an analysis of
the data presented in Tables 5-1 and 5-2.  Additionally,
further component breakdowns can be calculated by applying
new/existing and major/area ratios to the above totals.
     At pipeline pumping stations, it was estimated from
data in Table 8-27 that 72.1 percent of the facilities would
be classified as "existing" in the base year of 1998 (27.9
percent would therefore be "new") and all pipeline pumping
stations are area sources.  Under Alternatives I and II, a
quarterly LDAR program is required at all of these
facilities.  The remainder of the alternatives do not
require LDAR.
     At pipeline breakout stations, 90.7 percent were
estimated to be existing in the base year (9.3 percent would
be classified as "new" as shown in Table 8-27) and it was
further estimated that 7.4 percent of these sources would be
classified as major sources of HAP emissions  (92.6 percent
would be area sources).   Based on this analysis, at pipeline
pumping stations, approximately 6,530 pumps and 77,500
valves would be found at existing sources, while 670 pumps
and 7,950 valves would be located at new sources.  Further
breakdowns for valves are as follows:  590 major source new,
7,360 area source new, 5,740 major source existing, and
71,810 area source existing.  The analysis of number of
pumps follows similarly with the following results:  50
major source new, 620 area source new, 480 major source
                            7-41

-------
existing, and 6,050 at area source existing sites.
Alternative IV requires a monthly LDAR program at new major
source sites (590 valves and 50 pumps).   Alternative IV-Q
requires a quarterly LDAR program for the equipment at
existing major source sites as well (5,740 valves and 480
pumps).  Alternative IV-M requires that monthly LDAR be
implemented at these sites.  Alternatives I, II, and III
provide for implementation of area source control in
addition to the major source control as specified in
Alternative IV-Q.  These alternatives all require quarterly
LDAR for all area source facilities (approximately 79,200
valves and 6,700 pumps).
     Alternatives I, II, and III for storage tanks at
breakout stations require the retrofit of all fixed-roof
tanks with an internal floating roof and require
installation of secondary seals on internal floating roof
tanks as well.  Therefore, under Regulatory Alternatives I,
II, and III the cost of retrofitting internal and external
floating decks can be applied to the entire uncontrolled
fixed-roof tank population (143) and internal floating roof
tanks with only primary seals (476).
     Alternatives IV, IV-Q, and IV-M require that controls
be implemented at major source facilities only.
Consequently, these controls would apply to 11 fixed-roof
and 35 internal floating roof tanks.
7.2.2  Bulk Terminals •
     7.2.2.1  Truck Loading Racks.  Alternative I requires
new major source terminal loading racks to meet an emission
limit of 5 mg/liter, while all other terminals are required
to meet a 10 mg/1 limit (existing major and area  sources
would be allowed to phase-in controls).  Of the 1,024
facilities (see Table 3-11), it is estimated that there are
76 sites that fall into the new major source category
(27 percent of the total number of loading racks  are major
sources and 28 percent of those are classified as new  [see
Table 8-27]).  Of these 76, it was further determined that 2
of these new source facilities were designed to meet the 10
                            7-42

-------
mg/1 standard and 74 were designed to meet the 35 mg/1 NSPS
standard.  Therefore, all 76 sources must upgrade to the 5
mg/1 limit.  Tables 7-13 and 7-14 provide the necessary cost
information for this category.
     The remaining 948 sites must all meet the 10 mg/1
emission limit specified by this alternative.  Two hundred
of these sources are classified as existing major
(approximately 19 percent of the total number of facilities
[72 percent are existing, 27 percent are classified as
major]), 207 are new area sources (28 percent are new and 73
percent are area), and 541 fall in the existing area
category (approximately 53 percent of the total population).
Using the facility numbers and the percentages from Appendix
D, Table D-3, it was determined that 485 of these facilities
must upgrade their level of control to meet this standard
(194 from 80 mg/1 to 10 mg/1, and 291 from 35 mg/1 to
10 mg/1) and that 213 of the previously uncontrolled sources
must undergo rack conversions besides.  Tables 7-10, 7-11,
and 7-12 provide this cost information.
     Alternatives II and III require the same levels of
control at major sources as under Alternative I  (phase-in
controls at existing major sources).  However, at both new
and existing area sources, each of these alternatives allows
an emission rate of 35 mg/1  (again with phase-in control).
Since all new sources must meet the NSPS standard of 35
mg/1, none of the new area sources was required to modify
its loading racks.  However, of the 541 existing area
sources, 151 will be required to upgrade from 80 mg/1 to the
35 mg/1 limit, and 131 previously uncontrolled facilities
must undergo rack conversion as well.  Cost data for these
categories are provided in Tables 7-8 and 7-9.
     Alternatives IV, IV-Q and IV-M require control at major
sources only, and at the same levels previously specified
(5 mg/1 at new sources, 10 mg/1 at existing sources).  As
previously stated, the cost data are contained in Tables
7-10, 7-11, 7-12, 7-13, and 7-14.
                            7-43

-------
     For railcar loading,  it was assumed that none of the
facilities can meet either a 5 mg/1 or a 10 mg/1 level.  As
a consequence, all facilities with railcar loading racks
would need rack conversions.  Therefore, the costs in Table
7-14 were applied to all 20 railcar loading racks and added
to the overall cost for terminal loading racks.
     7.2.2.2  Storage Tanks.  Alternatives I, II, and III
for storage tanks require the conversion of all 1,072
uncontrolled fixed-roof tanks to internal floating roof
tanks with phase-in allowed at area sources  (incurring those
costs in Table 7-1).  Also, all 2,426 external floating roof
tanks with only primary seals would be required to install
secondary seals (phase-in at area sources),  incurring the
costs in Table 7-2.  Alternatives IV, IV-Q, and IV-M require
storage tank control at major source facilities only.
Consequently, the number of fixed-roof and internal floating
roof tanks requiring control would be reduced to 289 and
655, respectively  (27 percent of all tanks are located at
major source sites).  Table 7-23 shows that  there are no
costs associated with implementation of these controls for
new sources.  This is due to the fact that the storage tank
NSPS already requires these controls for new sources.
     7.2.2.3  Tank Truck Leakage.  For tank  truck vapor
leakage, Alternatives I, II, and III require the
installation of a vacuum assist vapor collection system at
new major sources  (estimated to be a total of 76 sources  (27
percent major and  28 percent of those are new as has been
calculated from Table 8-27)) and mandate annual vapor
tightness testing  at all bulk terminal facilities.
Consequently, the  cost of installation of a  vacuum assist
system  (see Section 7.1.3.3) involved with these
alternatives would be incurred by 76 bulk terminals,
excluding the very few that already have this system.  The
estimated cost of  annual truck testing  is $150 per truck
plus downtime.   This cost was applied to the  12,731
uncontrolled bulk  terminal tank trucks.
                            7-44

-------
     Alternatives IV, IV-Q,  and IV-M require controls at
major sources only and,  as such,  the number of tank trucks
requiring annual vapor tightness testing would be reduced to
3,437 (27 percent of the previously uncontrolled tank truck
population).
     7.2.2.4  Equipment Leaks.   The costs for controlling
equipment leaks were calculated in the same manner as those
discussed for pipeline facilities.  The control option
programs (quarterly and monthly LDAR) are the same and the
component inspection costs are also the same as have been
discussed for pipeline facilities.  It is assumed that there
are approximately 10,000 pumps and 116,000 valves at bulk
terminals  (component populations summed across model plant
facility numbers as presented in Table 5-3).  Of this
number,  it is estimated that approximately 800 pumps and
9,000 valves will be found at the 76 new major source
terminals and would therefore require monthly LDAR.  The
remaining equipment components (those found at existing
major source and all area source terminals) would be subject
to a quarterly LDAR program.  All of these components are
considered to be uncontrolled at the baseline and, as a
consequence, they would incur the total costs.
7.2.3  Bulk Plants
     For incoming loads (from tank trucks into storage
tanks),  Alternatives I and II require all bulk plants to
install a vapor balance system.  Implementing costs for
these alternatives would therefore apply to the 13,857
facilities that were uncontrolled at the baseline, using the
costs in Table 7-17.  The remaining alternatives require no
controls for storage tank filling and bulk plants would
therefore incur no costs under these alternatives.
     For outgoing loads, Alternatives I and II again require
all bulk plants to utilize a balance system, but with an
exemption.  These alternatives require all bulk plants with
a daily gasoline throughput greater than 15,000 liters
(4,000 gallons) to install a vapor balance'system and all
bulk plants with a throughput of 15,000 liters (4,000
                            7-45

-------
gallons) per day or less to install submerged fill
equipment.
     It was estimated in Table 5-5 that approximately 48
percent of the facilities have daily throughputs less than
15,000 liters (4,000 gallons)  per day.  Applying this
percentage to the baseline breakdown presented in Appendix
D, Table D-10, it was calculated that 1,082 facilities of
the 2,256 currently in areas with exemptions would therefore
continue to be exempt.  Also,  48 percent of the remaining
3,826 motor gasoline terminals (1,836) and all 3,200
aviation gasoline bulk plants would be exempt.
Consequently, under these alternatives, it was estimated
that 5,036 of the newly subject facilities (1,836 + 3,200)
would be exempt, and that 1,990 would be required to install
vapor balance.  The costs of implementation of these
controls were taken from Table 7-18.
     Alternatives III, IV, IV-Q, and IV-M require no
additional controls on outgoing loads.  Likewise, none of
the alternatives includes controls for tank trucks loading
at bulk plants.  Consequently, there are no costs associated
with tank trucks for any of these alternatives.
     The costs for controlling equipment leaks were
calculated as have been previously described for pipeline
facilities and bulk terminals, and were added to the overall
costs of Alternatives I and II.  These calculations were
based on the assumption that there are 100,800 pumps and
629,900 valves at bulk plants nationwide.  All of these
components were again considered to be uncontrolled at the
baseline and as a result would incur the total control
costs.
7.2.4  Service Stations
     Alternatives I and II require the installation of a
vapor balance system for all facilities with throughputs
greater than 38,000 liters (10,000 gallons) per month.  As
shown in Appendix D, Table D-13, 123,562 stations are
currently in areas with a 38,000 liter  (10,000 gallon) per
month exemption.  Also, Table 5-7 indicates that
                            7-46

-------
approximately 58 percent of all service stations (public and
private) have throughputs less than 38,000 liters/month
(10,000 gallons per month).  Therefore, 71,666 facilities in
these areas would continue to be exempt under this
alternative.  Of the remaining 129,042 facilities without
vapor balance, it is assumed that 58 percent of the motor
gasoline stations (74,844) and all of the aviation gasoline
stations (1,620) would have throughputs less than 38,000
liters/month  (10,000 gallons per month).  This leaves a
total number of 104,474 stations, approximately 2,800 new
and 101,650 existing (the service station population is
characterized as 2.7 percent new and 97.3 percent existing
as shown in Table 8-27), that would need to install vapor
balance systems to comply with Alternative I or II.  Costs
for each of these alternatives were calculated by
multiplying this number by the costs in Table 7-19.
7.2.5  Summary of National Alternative Impacts
     Table 7-27 presents an overall summary for each of the
regulatory alternatives developed and analyzed for this
study.  Note that Alternatives IV, IV-Q, and IV-M are
variations on the same theme in that all of these
alternatives propose controls for major sources only.  The
remaining alternatives propose controls for area sources as
well as major sites, hence the break line in the center of
Table 7-27.
     Of the negative increments appearing in the table, both
favor Alternative IV-Q over Alternative III (both are
calculated in increments from Alternative IV).  These
increments fall under the headings of HAP cost effectiveness
and VOC cost effectiveness.  In this analysis, the smaller
the number, the greater the cost effectiveness of the
alternative.  In this regard, Alternative IV-Q is not only
very cost-effective, it provides a net cost benefit over
Alternative IV while providing a greater emission reduction.
     Table 7-27 presents the alternatives discussed earlier
(including 5 mg/1 for new facilities).  Table 7-28 has been
                            7-47

-------
added to show the impacts of having both new and exisiting
bulk terminal loading rack controls at 10 mg TOC/1.
                            7-48

-------
H
i!
  0)
  Q)
  g
  in
r»
w
»
g
1
1
0> 1

~o S !
IB r* \
8, ° ^M *.
f U 0 M
O S * «
C > H-
- £

5
M $ w
•&»
H-
IU
Is,
11*
U, | »
O w ++
§ *



w


** ** f
E U H- X
U O. U M
U < ft) v
C 5C M-
UJ





0 s 5
a., O ^t
K S-

UJ

4-r
S ~
0 8
— o
a o





4^
o y
"a o
"5. C
a

2
f o "5 (o
C •*• V) O
O 4-»
u y
* -P «
a. S *•>
K O
1—

1 g ~
w — <-
'^ " 'S


< ae

* 1


a
c

Cl

<




^s
ess
e M ~
i •*
i
i
i
i
i
1 s ° i
i
i
i
i
111
1 8 fe 8
[ in •» «M
i
i



i
u

*s
1 O O O
1 O O O
1 • • %<
: R *
1
1
i
1
1
1

1 O O O


|j
Ij



II
II O O O
n o o o
ij IM_ in co^
ii O^ -^ "-^
II f- * •-
II w
II
II
II
II
II
II
II O O O
i: s 2 g
II ......
II Nj CO M
ii to 55 K
i K- -o tsi
1!
n
u
U
II M X M
ii 
^3 *O f^







ill






§ g §
CM CM CM
O 0 0
fO fO fO




XXX
«» K) Kt


XXX
in in m


I {2 i
CvT CM* (vT







x a


M |S




















i
i
i
i
i
i
i
|
i
i
i
i
i
i
i
i
i
i
!

i
t

i
i
i
i
i »
•>
u
1 L.
i 3
: $
\ c
! V
1 I
i
n ae
i s
| ^
a **
n &>
1 5
s «
s »
IS >
i $
M •«-
tl +•*
u 5

II u
!! £
—
H <
ii
u *

II O
II 1
8 2














































±
*
0
£

a
«
S
2

i
•£
x

5




*
^^


i

X

S
                           7-49

-------
o
H
. i
— CO 1

*•» 4-» C
Mil
|K|B
Ul
** V *•>
•Is
U W V
Si
IU
"5 o f+
V X
u * w
M-

_, *
a «2
** ** £
w ** **»
t- a. y t»
i? z «»- *""
MM H-
Ul


0
§8! —
> o>
u -- x
a. u M

* «C
IU
— - -

8 ~
u yj


| c


Ig

5 §
•»* *P"
& ^

^

S -8 2
1-33
" i -
s. i s
ac o
J*i



1 1 ~



^j
g

«

.^
<
1
i

gas

i § 1


§i S
. « -I
| § 1 S
i
i
i
i

i
s
: § § §
1 CM fi O>
i R •** w
!
i
!
i
i
1 O O O
IOOO
1 -O CM {£
4 • » %rf
1 CM CM

II
4
II
II
II
11
II — - _ «.* «„, .«. —

ii e o o
80 o o
CM in to

«<>«»•-
M fC 55 •-
N «•*
N
II
U
n
H
f S 8 §
a "i -. «
!! a S R
jj K * CM
8
u
n
H
S S K R
« «  K» «»

1 1 i
!S *" 5 |
H
N
|
S
1
•^ |l
« « «. 1
CM <-  M

*-•-•- II
N
S

N
H
§ § § S
0 » 03 Jj
>» NT •* II
(M CM CM U
- - - 8
S
H
II
N
M M M N
2 C C jj
in in in M
*
1
CM CM CM U
II
II
II



M
X I > \

** ^** II
H
II
II
































»




4^

<
I

o
1

1»*

































s
u
L.
I
L.
O
I
4^
C

5
s
JE
I
j;
4*
"5
>
§

4-*
§
tl
«rf

<
1

X
1
>

                             7-50

-------
7.3  REFERENCES

   1.  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,  NC.  Draft.   October 1993.

   2.  Memorandum.  Johnson,  T., Pacific Environmental
       Services, Inc.,  to Shedd, S.,  EPA:CPB and  Mathias,
       S.,  EPA:SDB.   June 19,  1992.  Storage tank costs.

   3.  Reference 2.

   4.  Reference 2.

   5.  Reference 2.

   6.  Reference 2.

   7 .  Reference 2.

   8 .  Reference 2.

   9.  Memorandum.  Johnson,  T., Pacific Environmental
       Services, Inc.,  to Shedd, S.,  EPA:CPB and  Mathias,
       S.,  EPA:SDB.   June 22,  1992.  Leak detection and
       repair costs.

  10.  Telecon.  Hawes, T.,  Pacific Environmental Services,
       Inc.,  with Keller, D.,  IT McGill.  February 26, 1991,
       Control equipment cost estimates.

  11.  Telecon.  Hawes, T.,  Pacific Environmental Services,
       Inc. with Tuttle,  N.,  John Zinc Company.   February
       21,  1991.  Control equipment cost estimates.

  12.  Telecon.  Hawes, T.,  Pacific Environmental Services,
       Inc. to Shotts,  K., IT McGill.  February  26, 1991.
       Control equipment cost estimates.

  13.  Memorandum.  American Petroleum Institute  to Wyatt,
       S.,  U.S. Environmental Protection Agency,  Research
       Triangle Park,  NC.  December 19,  1991. Comments on
       1991 Draft Gasoline Marketing Industry (Stage I) -
       Background Information for Proposed Standards.

  14.  Telecon.  Hawes, T.,  Pacific Environmental Services,
       Inc.,  to Waldrop,  R.,  Edwards Engineering.  February
       25,  1991.  Control equipment cost estimates.

  15.  Reference 10.

  16.  Reference 10.
                            7-51

-------
17.  Telecon.   Hawes,  T.,  Pacific  Environmental  Services,
     Inc.  to  Keller, D.,  IT McGill.  February  28,  1991.
     Carbon adsorber costs.

18.  Telecon.   Hawes,  T.,  Pacific  Environmental  Services,
     Inc., to Keller,  D.,  IT  McGill.  March  5, 1991.
     Retrofitting  carbon adsorption units.

19.  Draft Regulatory  Impact  Analysis:  Proposed Refueling
     Emission Regulations for Gasoline-Fueled  Motor
     Vehicles — Volume I - Analysis of Gasoline Marketing
     Regulatory Strategies.   U.S.  Environmental  Protection
     Agency.   Research Triangle Park, NC,  and  Ann Arbor,
     MI.   Publication  No.  EPA-450/3-87-001a.   July 1987.

20.  Evaluation of Air Pollution Regulatory  Strategies  for
     Gasoline Marketing Industry - Response  to Public
     Comments.  U.S. Environmental Protection  Agency.
     Research Triangle Park,  NC.   Publication  No.  EPA-
     450/3-84-012C.  July 1987.

21.  Memorandum.   Dautenhahn, P.,  Pacific  Environmental
     Services,  Inc., to Norwood, P., Pacific Environmental
     Services,  Inc.  January  17, 1992.  Costs  associated
     with  various  vapor recovery units.

22.  Telecon.   Hawes,  T.,  Pacific  Environmental  Services,
     Inc., to Jordon,  J.,  John Jordan Co.  March 4, 1991.
     Maintenance costs of vapor recovery units.

23.  Telecon.   Hawes,  T.,  Pacific  Environmental  Services,
     Inc., to Moore, G.,  Public Service Company.  March
     11,  1991.  Natural gas costs.

24.  Memorandum.   Norwood, P. and  Thompson,  S. ,  Pacific
     Environmental Services,  Inc., to Shedd, S., EPA/CPB.
     June  18,  1991.  Trip report to Mobil  Oil  railcar
     loading  terminal.

25.  Memorandum.   LaFlam,  G., Pacific Environmental
     Services,  Inc., to Shedd, S., U.S. Environmental
     Protection Agency,  Research Triangle  Park,  NC.
     January  11, 1991.  Report of  trip to  Fina Oil and
     Chemical Company's tank  truck loading terminal,  Port
     Arthur,  Texas.

26.  Telecon.   Hawes,  T., Pacific  Environmental  Services,
     Inc.  to  Surdriff, A., R.W. McCollum Co.   February  22,
     1991.  Tank truck conversion  costs.

27.  Telecon.   Thompson,  S.,  Pacific Environmental
     Services,  Inc.  to Olsen, T.,  Penske Tank.  February
     26,  1991.  Loading rack  conversion costs.
                          7-52

-------
28.  Memorandum.  Johnson, T., Pacific Environmental
    Services,  Inc., to Shedd, S., EPArCPB.  August 6,
    1992.  Tank truck vapor tightness testing.

29.  Cost Estimates  from Exxon Corporation.  Internal
    memorandum to Exxon bulk plant owners.  March 14,
    1991.

30.  Telecon.   Norwood, P., Pacific Environmental
    Services,  Inc., to Alsopp, c., Jones and Frank, Inc.
    March  20,  1991.  Stage I service station and bulk
    plant  control costs.

31.  Telecon.   Norwood, P., Pacific Environmental
    Services,  Inc., to Wilkins, J., Kubat Equipment Co.
    March  19,  1991.  Stage I service station and bulk
    plant  control costs.

32.  Reference  19.

33.  Reference  29.

34.  Reference  30.

35.  Reference  31.

36.  Memorandum.  Norton, B., Pacific Environmental
    Services,  Inc., to Shedd, S., EPA/CPB.  December 28,
    1989.  Service  station control costs.

37.  Reference  36.

38.  Letter from Akin, C., Service Service Stations, to
    Norwood, P., Pacific Environmental Services, Inc.
    February 26, 1991.  Service station Stage I costs.
                          7-53

-------
                            CHAPTER 8
                    ECONOMIC  IMPACT ANALYSIS

8.1   PROFILE OF THE U.S. GASOLINE DISTRIBUTION INDUSTRY
      This chapter profiles  elements of the U.S. gasoline
distribution industry most  affected by the proposed regulation.
This  industry includes:
      • bulk terminals,
      • bulk plants,
      • service stations  (both public and private) ,
      • railroad tank cars,
      • pipelines, and
      • tank trucks.
      Because motor gasoline constitutes approximately  99 percent
of all gasoline consumed in the United States, the vast majority
of available gasoline industry data pertains to motor  gasoline-
related operations.

8.1.1  Description Of The U.S. Gasoline Distribution Industry

      Gasoline is the major  petroleum product produced  from crude
oil at refineries.  A small   quantity,  less than one percent in
1987, is produced from natural gas liquids at gas processing
plants.1  Finished gasoline accounted for approximately 47
percent of the volume of total finished petroleum products
supplied.  The next largest petroleum product supplied in 1990
was distillate fuel oil, accounting for 20 percent of the total
volume of petroleum products.2  Table 8-1 displays trends in
U.S. gasoline production and distribution.

      Figure 8-1 depicts the flow and storage of gasoline through
the U.S. distribution system.  Gasoline is distributed from
approximately 224 refineries owned by about 115 companies.^
                               8-1

-------
2
O
O
a<
Q
a
§
M
9-
CJ
D
Q
O
(X
Cd
O CN
CO  —
<  CO
O  K
    CU
  •  E-
CO  M
2  2
M  O
E-"  M
oc;
 i
oo
CO
o
w
(0
o

I-l
o
u
o
              01  i
           _,  C  D-

           sill
           **  10  o
              O U
           C  V  I
           oca
           •** -^  E c
           JJ t-H  3 O
           (0  O  (0 -H
           •-<  in  C a->
           >  *  o
           < o u
           E  C
           3  O
         .           ~
     o g1  w  -  u  2
     1-1 ro  c  jj «t-i  9
     0)    O        =
                 >,TJ
                 .u  a>
                -^  e
 tJ ^H
  0) (0
 •0 >->
.  10 o
1  0) E-> C  M
 •-i    ro  c
  C «-< 3  O
 O O 00
        •g
                u
       .,  0)
       u  g
          §

    Hgc

    |I3
    s ^
                                   oo   fl>  *»•  TT  «T
                                       (-SJ  r-<  fN  rH
                                               m  m  r^i  rt  r^  *T  T
                                               ^T^rnromnmr^
             O\  *H  r*l  O3
             rH  04  CM  fN
                       r-
                       VO
                       fN
                                           (N
                                   m
                                   o
                                   10
                                   0
                                       cooocDCDooooooajoDoor-r-r-r-t^r-
                                                                                                      m

                                                                                                      jj

                                                                                                      O
                                                                                          u
                                                                                          in
                                                                                           0)

                                                                                          JJ

                                                                                           C
                                                                                           o
                                                                                 a

                                                                                 3
                                                                                 CO

                                                                                 O
                                                                                 o


                                                                                 o

                                                                                •D
                                                                                 0>
                                                                                 U3
                                                                                 to
                                                                                                      ro
                                                                                                      10
                                                                                                      a;
                                                                                           o>


                                                                                           01
                                                                                                      O
                                                                                                      in
                                                                                                      (B
                                                                                                      Di

                                                                                                      c
                                                                                                      o
                                                                                                      10

                                                                                                      i-i
                                                                                                      o
                                                                                                      10
                                                                                                      •D
                                                        8-2

-------
P
R
O
D
U
C
T
I
O
N
D
I
S
T
R
I
B
U
T
I
O
N
                     Domescic
                   Natural Gas
            Gas Plant
                                 Imported
                                 Gasoline
                       Imported
                      Crude Oil
    Domestic
   Crude Oil
                                                                  X
                             U.S.  Petroleum
                                Refinery
Other  Natural
GAS Products
   Total  Gasoline
(100% of  throughput)
Other  Petroleum
    Products
                                                            Transportation from
                                                             Production  Level: •
1
Exports b
(1% of throughput)
Areas of Proposed
controls:
n Storage
Activity
0 Trans-
portation
Activity
Proposed
regulation adds
additional
controls, except
for barge/tanker
transportation
which is covered
by a different
regulation

i
^S,

S, ' pipeline 67% J
Tanke r / ba r g e^^%
Truck 10%
Railroad <1%

Wholesale Storage
(level one)
Bulk Terminal
(99% of total throughput) c

Wholesale Storage
(level two)Bulk "•
Plant(18% of Bulk _
Terminal
throughput) *

^
j i
\—
*••
,•••
Retail Motor Fuel Outlets
t Other Consuming Sectors '
(e.g., industrial use)
(99% of total throughput)
1
(82% of Bulk
/ Terminal
/ throughput)
J f 	 "" •>
Wholesale
^ ^ Transportation1*

16% Pipeline
^— - 77% Truck
1% Railroad
V, J


5% Tanker /barge
                                  Final Gasoline Distribution
        (e.g., gasoline pumped  from underground storage tank to  automobile gas tank)
     Figures frcra 1977 Comnodicy Transportation Survey for 'gasoline and ]ec fuels,
     pipeline shipments would be regulated by the proposed standard.
                                                                      only
     Assumed all exports are taken frcre the refinery-level and that none go through terminals.
   < Assigned all refinery shipments other Chan exports go through a terminal (i.e., there
     arc no refinery-to bulk-plant shipments).
   d Transportation node figures from an unpublished Bureau of Census source for shipments
     from SIC S171 (this source's data are used to estunate mode of transport for all
     wholesale shipments of gasoline; non-truck transportation apparently results fron
     terminal-to terminal shipments).
   • The percentage has varied over tine; the  18 percent figure represents the estimate
     for 1998 (see Section 8.2.3 for description of estimation procedure).
   C For detail on consuming sectors, see Table 8-2.
                Figure  8-1.     The  U.S.  Gasoline System
                                           8-3

-------
Most gasoline goes first to one of over 1,000 large bulk
terminals, located generally along a pipeline or on the
coastline of a navigable body of water, where companies can take
barge or tanker delivery.  Most of these bulk terminals are
owned by refiners.  A significant, but declining, proportion of
gasoline is transported by truck from the bulk terminal to
another storage facility, the bulk plant, which is generally
smaller than the terminal and nearer the final customer.  Bulk
plants are located in areas with smaller volume requirements
that do not justify the additional investment required for a
bulk terminal.  EPA defines a bulk gasoline terminal as having
gasoline throughput of at least 75,700 liters (20,000 gallons)
per day; bulk plants have an average throughput of less than
75,700 liters per day.

      Increasingly, gasoline bypasses the bulk plant and is
shipped directly to service stations because of the construction
of large-volume retail outlets and the use of more efficient
truck carriers.5  Gasoline wholesalers often distribute
additional petroleum products, especially home heating oil, and
may also operate retail gasoline outlets.  Gasoline is
transported through the wholesale distribution chain by railroad
cars, tank trucks, pipelines, and barges and tankers (two forms
of water transport covered by a separate EPA regulation).

     The gasoline distribution industry consists of three broad
entities:
     • "major" oil companies,
     • independent marketers with refineries,  and
     • all other entities,  which include distributors (jobbers!
       and retailers.

     Major oil companies, such as Exxon, Shell,  and Texaco,
account for a large percentage of total refinery capacity.
Major companies are vertically integrated;  that  is,  besides
                               8-4

-------
gasoline and other petroleum product production, they own
wholesale distribution facilities and retail outlets.
Independent marketers with refineries are similar to major oil
companies in that they are vertically integrated and have
refinery capacity.  However, independent refiners hold a much
smaller percentage of the market.  The remainder of the gasoline
industry comprises independent wholesale distributors (jobbers)
and retailers that do not own refinery capacity.  Some of these
smaller firms specialize in one phase of the industry such as
providing transportation services.  These firms obtain gasoline
from the major and independent oil companies.

8.1.2  Complexities and Problems Affecting the Industry Profile.
     Two major problems arise in attempting to'profile the
gasoline distribution industry:
     • general deficiencies in the available data and
     • the complexities involved in defining and characterizing
       ownership of industry establishments given the presence
       of significant industry vertical and horizontal
       integration.

     8.1.2.1  Data Deficiencies.  Most of the available industry
data comes from three major sources:  previous EPA reports,  the
U.S. Department of Commerce's Bureau of the Census,  and various
petroleum industry associations  such as the American Petroleum
Institute (API).   Unfortunately, data from these three sources
are often collected using different definitions.  For example,
the Census Bureau data on public service stations,  Standard
Industrial Classification (SIC)  5541--Gasoline Service Stations,
only describe stations that receive at least 50 percent  of their
revenue from sales of gasoline and automotive lubricants.

     A significant shortcoming of much of the available data is
the lack of specific data for gasoline distribution  activities;
most of the data  that have been  identified are provided  for
                               8-5

-------
total petroleum products.  For example, data are only provided
for petroleum product employment; data are not available for
employment in gasoline operations only.

     Inconsistent use of terminology in industry data also
causes problems.  For example, the term "jobber" may refer to
any petroleum product wholesaler, to wholesalers of fuel oil
exclusively, or to petroleum product wholesalers with bulk
plants, depending on the source.

     8.1.2.2  Industry Integration.  Many firms in the industry
are  also involved in other lines of business;  they not only
market other petroleum products, but have diversified into
businesses as dissimilar as real estate and lobster
distribution.6  Unfortunately, detailed data for differentiating
gasoline distribution from other activities are not available.

8.1.3  Data Characterizing the Gasoline Distribution Industry
     8.1.3.1  Gasoline Production and Consumption.  Table 8-1
shows that motor gasoline production peaked in 1978 at over 430
billion liters.  In 1982, production reached its lowest level
since 1974, at nearly 380 billion liters.  With increased demand
due to economic growth and falling gasoline prices, the level of
gasoline produced has recently increased to near 1978 levels.

     Table 8-2 presents consumption of gasoline by end-use
sector for the years 1982,  1987, and 1989.   These data show that
the private and commercial transportation sector accounted for
approximately 95 percent of total gasoline consumed in each
year.

     8.1.3.2  Prices and Margins.  Table 8-3 presents nominal
and real (in 1990 dollars)  retail motor gasoline prices
                               8-6

-------

II

o\
00
r*
r .
C*
H
J
t,
O
to
Q
^y
CO
I
H

2
M
fT»
V t
00
(TV
t-H
r--
oo
en
rH



01
M
D
rH
10
*r4
o
IH
Comme
•D
c
m
0)
JU
to
•ft
Ll
o.


(N H
00
o\
rH II
II
H
U
2 H
r^ II
•J
O






>.
10
Dl
•^
X
t
o
z











Q
^
S
Marine
i
o
t!
3-
O
U
I-H r-H
(TJ (13
•rt -^
ii-g K
s s i
•D e
C O
M U

c
c
i-
(T
•M
rt

1
Is
fe5
<

N
ro
3
Dl
..- 1
3;


CO

2 en
0 =>
t j
s
1
(N
CD
3
CD
•2
S



o
•I-*
•— i
»Q
2
oJ > a
5 = "
™ D "H
w 0 C
W U 3
•g

Federal

•o
•si
n
U



M
10
H
r~ °°
LT o
o>
^ 0
s s
•T ^

s s
0



rH «>
ro •-'
0



2 r~
ro
"*" .
rn" °


s s
•» m
o


LO
O M3
c\ p*.
T «cr
 O rH
IN U (0
co l-i 01 u
Cr, Q) O> O
TH CL (0 H
(N ^ ro
r-l &\
^0 ^
 O
O CT. rH
^5T *T




O oj
r* oj cc
r-t ^£ ^
* • *
r-4 04 O
«-l r-l

i ^ s
Oft 0 OD

i § i
s 1 s
*r ^ •«•

i
ij
C «-<
01 O rH
r- u (0 e*
oo ^ A) jj co
en o> 01 o o\
ft O. A) EH rH
^>
rH

O
o\
o
1-1

ID
01
o



n
*T
0



*H
ro
o



8-7

-------
        TABLE 8-3.  TRENDS IN RETAIL MOTOR GASOLINE PRICES
           (IN CENTS  PER GALLON,  INCLUDING TAXES)10'  13-
Year
1990
1989
1988
1987
1986
1985
1984
1983
1982
1981
1980
1979
1978
Nominal
Leaded Regular
115.0
99.8
89.9
89.7
85.7
111.5
112.9
115.7
122.2
131.1
119.1
85.7
62.6

Reala
Unleaded Leaded Regular
Regular
117.0
102.1
94.6
94.8
92.7
120.2
121.2
124.1
129.6
137.8
124.5
90.3
67.0
TRENDS IN RETAIL MOTOR GASOLINE
PER LITER, INCLUDING
Year
1990
1989
1988
1987
1986
1985
1984
1983
1982
1981
1980
1979
1978
Nominal
Leaded Regular
30.4
26.4
23.8
23.7
22.6
29.5
29.8
30.6
32.3
34.6
31.5
22.6
16.5

115.0
103.9
97.5
100.5
98.9
132.2
137.8
146.4
160.7
183.4
182.7 '
143.4
114.0

Unleaded
Regular
117.0
106.3
102.6
106.2
107.0
142.5
148.0
157.1
170.4
192.8
191.0
151.1
122.0
PRICES (IN CENTS
TAXES)
Reala
Unleaded Leaded Regular
Regular
30.9
27.0
25.0
25.0
24.5
31.8
32.0
32.8
34.2
36.4
32.9
23.9
17.7
30.4
27.5
25.7
26.5
26.1
34.9
36.4
38.7
42.5
48.5
48.3
37.9
30.1

Unleaded
Regular
30.9
28.1
27.1
28.1
28.3
37.7
39.1
41.5
45.0
50.9
50.5
39.9
32.2
aln 1990 prices,  (adjusted by GNP implicit price deflator)
                                 8-8

-------
(including gasoline taxes)  for regular leaded and unleaded
gasoline over the period 1978-1989.  In real terms,  the price of
motor gasoline declined each year during the period 1982-1988.
The Persian Gulf crisis caused much Of the large price increase
between 1989 and 1990.

     Gasoline producers distribute their products through both
direct and indirect channels.  Each channel represents about
half the volume sold in the United States.12  Direct supply
means that the refiner retains ownership of the gasoline
throughout the wholesale distribution process.  Directly
supplied gasoline is delivered to retail stations at "dealer
tank wagon" prices.  In the indirect method, distributors buy
gasoline from refiners at terminal prices (discounted from the
tank-wagon price) .  They may then deliver it to. other
distributors and to their own or other retail"outlets,  hoping to
cover costs and make a profit on the spread between terminal and
resale prices.  Distributors using the indirect method are
referred to as "jobbers.1  All the major oil companies use both
forms of wholesale distribution depending on whether refiners
believe that their costs of distribution would be less than the
jobber discount.

     By using both forms of distribution, refiners can reduce
their investment and operating costs, and can compare the costs
of directly supplied and distributor-supplied product.   This
serves as a check on the economic efficiency of refiners'
distribution systems.12  Refiners usually choose direct
distribution in densely populated areas where station
representation is good; jobbers are used to distribute gasoline
to areas where the refiners'  stations are few and widely
dispersed.13

     Table 8-4 presents estimates of average margins at each
point in the gasoline distribution chain.  These margins
represent the total dollar value per liter added to the cost of
gasoline by each sector in the distribution chain to cover that
                               8-9

-------
      TABLE 8-4.   ESTIMATES OF MARGINS AT VARIOUS  POINTS
             IN THE GASOLINE DISTRIBUTION CHAIN10'11

                                   Margin                  Margin
         Sector                  ($/gallon)               ($/liter)
  Pipeline                           0.030                   0.008
  Bulk Terminal                      0.020                   0.005
  Truck Transportation                0.025                   0.007
  Bulk Plant                         0.020                   0.005
Tofcal Wholesale                      0.095                   0.025
  Service Station                    0.05                   0.013
Total Retail                         0.05                   0.013
                                8-10

-------
sector's costs and profit.  Other data compiled by EIA support
these estimates.14-18

      8.1.3.3  Margins  and  Product Differentiation.    Attempts at
product differentiation in retail trade have centered on
extensive advertising  campaigns extolling the virtues of various
additive packages to protect engine parts, give better mileage,
or reduce tailpipe emissions.  As a result of similar attempts
at differentiation during the years before the Organization of
Petroleum Exporting Countries  (OPEC) price hike, a majority of
customers paid 2 or 3  cents a gallon more for major brand
gasolines than for independent brands.19  However, some analysts
in the industry believe that little "brand loyalty" now exists
because of  the unprecedented price increases resulting from the
gasoline shortages of  the last two decades.  The theory is that
these increases have convinced consumers that "gasoline is
gasoline" and should be bought on the basis of price  rather than
brand.

      The market share  of  "regular" and "mid-grade" gasolines,
which have  lower retail margins than "premium" high octane
gasoline, has also been affected by price increases.  As a
result of precipitous  increases in retail gasoline prices during
the Persian Gulf crisis, consumers have recently switched to
cheaper, lower octane  gasolines.  The percentage of premium
gasoline to total gasoline sold by refineries dropped from 24
percent to  16 percent  between October 1989 and  October 1990.20
During the  1982-1989 period, the market share of premium-grade
gasolines had increased substantially,  despite the difference •
between average retail prices of premium and regular grades,
which averaged approximately $0.04 per liter ($0.15 per
gallon).21

     The stability of  prices within any marketing territory has
depended on the presence or absence of aggressive independent
marketers.22  These independent marketers pioneered the building
                              8-11

-------
of retail outlets with large  storage capacity.  This enabled

them to bypass bulk plants and resulted in lower costs.  They
also lowered margin requirements with direct-operated units, and
further reduced per-gallon operating costs with high-volume

retail outlets.

     8.1.3.4  Total Industry Employment:  and Sales   .  Employment

data for the U.S. gasoline distribution industry in 1989 are

available on the following:

     • pipeline transportation of petroleum products, excluding
       natural gas—17,825 employees

     • wholesale services for petroleum products—201,957
       employees

     • retailing activities at "traditional" gasoline service
       stations--622,799 employees.23   (Not included in this
       estimate is the number employed at  "non-traditional"
       service stations such as convenience stores.)


     By contrast,  1982 petroleum product  employment in these

sectors was approximately 34,842 less than in 1989.
Approximately 20,514 people were employed in product pipelines
and in product wholesaling activities.  Service stations

employed 561,172 in 1982, and it is the only sector that

increased employment in 1989.

     The Petroleum Marketers Association  of America's  (PMAA's)
1990 Marketer  Profile Survey  estimates 12,500 to 14,000

independent petroleum marketers nationwide in 1990.  PMAA's
current estimates represent a decline from an estimated 21,000

at the beginning of the 1980s:

       Continued declines in the number of marketers is no
       longer attributable to shrinking markets, as was the case
       during the early 1980s, when the highest rate of industry
       exits occurred.  A PMAA long-range study committee
       estimated that roughly half of the present total will
       make it to the year 2000.  In more recent years, factors
       external to the market have exerted a greater influence
       on competitive conditions; government regulation in the
                              8-12

-------
       environmental arena has had a particularly marked impact
                                                      O A
       on the nation's petroleum marketing businesses."^

National Petroleum News (NPN) estimates that the vast majority
of jobbers are small jobbers located in small rural areas away
from the large highly competitive markets that the majors and
large chains fight over:
       Two current situations seem to favor those small jobbers
       still in business:  the contraction of the 1980s has
       reduced competition in their small markets, providing in
       some cases for higher prof it-margins; and the gallonage
       potential, generally speaking, is insufficient to attract
       either major or chain direct-retail operations."

Also, NPN estimates that many small jobber's retail outlets are
debt-free and that some larger but debt-burdened chains could
have difficulty covering the cost of underground storage tank
and vapor recovery regulations.

     Only independent petroleum marketers are represented in the
1990 Marketer Profile Survey.  Therefore, absolute values from
the survey only apply to that segment of the marketing industry.
However, figures from the survey can be used to illustrate
trends for the industry as a whole.  Table 8-5 shows employment
data using PMAA's total independent petroleum marketing
employment estimates for 1985, 1987, and 1989.  The 12 percent
increase in employment between 1987 and 1989 is consistent with
an industry trend toward larger businesses.  Much of this gain
in employment has been due to an increase in part-time
employment.

     The Bureau of Labor Statistics1 (BLS) Monthly Labor Review
provides estimates of projected employment for wholesale trade
in petroleum and petroleum products and gasoline service station
retail trade.  BLS estimates that wholesale trade will lose
approximately 2,000 workers  (or an annual rate of change in
employment of -1.0 percent) in petroleum and petroleum products
over the period 1988-2000.  For gasoline service stations,  BLS
projects an increase of 74,000 workers  over that same time frame
for an annual rate increase of 0.9 percent.2^
                              8-13.

-------
WV£>
O   o\

^  CD
a.
   o\
in
 i
oo

u










7!
*r
JJ
*"



















O)
E
*r4
JJ
1
JJ
r<
10
Oi














Q>

"*H
JJ
IH

I











in
00
i— i






p-
CO
en
f_i




en
00
en
rH






in
CO
cn
rH






-^
CO
Cn
rH





en
oo
rH





in
co
rH




f^
OO
en
 m p- in ^3" m «H
OO VO ^* VO p* VO P*
^* r*l (N ro OO ro 00
p- cn rH en vo en vo
VO m rH m rH CO rH




vo  p- 0 I-H
O (N VO » rH
p- p- O ro p- CN ro
C] rH rH ^1 rH ^> rH
rH




VO fr, VO ft ^ p* ^*
O CO ^* j r-t
CN




m rH vo oo cn *H ro
(N rH O VO O CD VD
oo oo in  OO rH O
 00 00
o\ oo oo vo in oo 
T> «* rH O (N O rH
» « « % » * »
oo in o oo in o r-t
rH p- rH rO rH *» rH




0) 0)
U U
C C
0) ^ «
oi c -r< m oi c
U 0 C r< U 0> 0>
•H -H o> o> n Q) -H jj u t-i
> JJ >h 0> > > C -H 0)
fclO CO —< -H W-H >4J Si
Q\ ^J Q ^j |^ Ij Q) (0 H^ 1 J
COCO UCO CO Q COS O O
o
p-
in
0
m
*N



CN
0
cn
»
rH
IN

rH
rH
0
rH
•*»
(N




m
0\
p-
«
rH





O

vo
^
0




rH
VO
(N
VO



P-
P-
oo
in
*H


i 0)
rH U 01
C 0 JC
0 JC JJ
(0
M JJ
O u O
JJ C C
O 0)
oi E -
w S —
O»rH JH
•r* g jQ
jjj ov E
0> 0)
.* >, E
>H U
§O O M
Dl O U
o> o o>
E ~L* * AJ
3 10 rH 0)
0) (J rH J^
*~* JM
0 Xt >i 10
rl 0 rH £
JJ -l-i 
0) M (1, D.
Q. iO "0
0) di — jj
-DE C
c o a a>
•H U -H T3
JC C
a> i-i vt v
f. o i-t a
XI TJ
>j -a E c
0 C (U -H
«-i 10 £
vw
(ft 10 W 0
0) 'D -
JJ C < i-i
 «c a>
•H tl S i
JJ 3
V) JJ Oi C
o> c c
01 •<-< i— i
0)  oi

o a. oi
J= O 01 JJ
« JJ IH 10

01 TJ W -H
01 01 01 JJ
IH JJ JJ (A
3 C (8 4)
Di 0) E
•rH M -H O
M-t 0) JJ O
rH M 0
jj a o> -
c ^
01 0) Q) rH
E )H rH 1
rS C «J 0
0 o
rH Q « in
tJJ JJ .
(0 <0 
-------
     Total sales  for the gasoline distribution industry were
estimated from 1987 Census data.  These data provide a range of
total gasoline sales between $173 and $200 billion.  The $173
figure is the sum of gasoline sales by the dominant wholesale
SICs 5171 — Petroleum Bulk Stations and Terminals and 5172--
Petroleum and Petroleum Products Wholesalers, except Bulk
Stations and Terminals, and the predominant retail SIC 5541--
Gasoline Service  Stations).   In 1987, service stations without
payroll had total sales from all sources of revenue of
approximately $2.8 billion.   According to the National
Association of Convenience Stores, gasoline sales at convenience
stores totaled $20.5 billion in- 1987.  Convenience stores which
have revenues from gasoline sales equaling at least 50 percent
of their total sales, are included in-the Census.  Determining
how many of these convenience stores are already included in the
Census figures is not possible.

     8.1.3.5  Ownership and Concentration.  Table 8-6 presents
concentration ratios for 1970-1987 for total wholesale and
retail gasoline sales.  This table shows that concentration in
gasoline sales decreased slightly during this period.

8.1.4  Wholesale Gasoline Distribution
     The wholesale gasoline distribution sector involves
intermediate storage and transportation of gasoline.

     8.1.4.1  Wholesale Distribution and Sales.   The U.S.
Department of Commerce's Bureau of the Census collects data on
wholesale petroleum product  sales using the SIC  system.
According to the Census'  1987 Census of Wholesale Trade--
Commodity Line Sales—United States, 11 different four-digit
wholesale SICs had sales of  petroleum products in 1987.
                              8-15

-------
TABLE 8-6.  CONCENTRATION RATIOS FOR GASOLINE SALES
            (PERCENTAGE OF  U.S.  TOTAL)27

Top 4 firms
Top 8 firms
Top 15 firms
Top 20 firms
Top 30 firms
1987
28.9
48.7
65.0
70.2
76.4
1986
29.5
49.6
66.4
70.5
76.6
1985
29.8
50.3
67.7
71.8
71.2
1980
28.5
49.5
66.3
'72.1
77.9
1975
29.5
50.3
68.6
74.7
—
1970
30.7
54.6
74.9
80.0
—
                        8-16

-------
However,  96 percent of total petroleum product wholesale sales
were by SICs 5171--Petroleum Bulk Stations and Terminals and
5172--Petroleum and Petroleum Products Wholesalers, except Bulk
Stations  and Terminals.  SIC 5172 comprises truck jobbers,
packaged  and bottled petroleum products distributors, and others
marketing petroleum and its products wholesale, but without bulk
liquid storage facilities.

     Figure 8-2 and Table 8-7 present generalized sales data  for
petroleum products and gasoline available from the Census.
Sales of  petroleum products in 1987 were approximately $188
billion dollars, with SICs 5171 and 5172 accounting for
approximately $181 billion of that total.  Detailed data
available from the Census in 1987 show that motor gasoline sales
totaled $97.8 billion in these two SICs.  Aviation gasoline
sales from these two SICs amounted to approximately $750,000.

     8.1.4.2  Employment.  No figures were identified for
employment in wholesale marketing activities specifically for
gasoline.  However, the data available for petroleum products
show that 201,957 people were employed in wholesale activities
as of January 1, 1989 (down from approximately 226,000 from
January 1982).29,30

     8.1.4.3  Economic Agents and Relationships. Industry
analysts  often refer to three categories of firms in the
gasoline  production and distribution industry.  The "major oil
companies" (most often referred to as Amoco,  Atlantic Richfield,
Chevron,  Exxon,  Mobil,  Shell,  and Texaco)  and "semi-major oil
companies" (often defined as American Petrofina,  Ashland
Petroleum, Citgo,  Conoco,Crown Central Petroleum,  Diamond
Shamrock, Kerr-McGee Refining,  Marathon Oil,  Murphy,  Phillips
Petroleum, Standard Oil [now BP-America),  Sun,  Tenneco Oil
[acquired by Amoco in 1987],  and Union Oil of California)  own a
large percentage of refining capacity and have vertically
                              8-17

-------
              Total Petroleum Product Sales = S188  Billion

                     4.0%
                                                    D SIC 5171


                                                    E3 SIC 5172


                                                    • Other SICs
             Share of Gasoline  Sales  from SICs 5171 and 5172
                                                    D SIC  5171


                                                    O SIC  5172
        Share of Gasoline  Establishments from SICs  5171 and 5172
                                                       SIC  5171


                                                    Q SIC  5172
          Figure 8-2.   SIC 5171  and  5172 Characteristics28

Note: SIC 5171--Petroleum Bulk Stations  and Terminals
      SIC 5172—Petroleum and Petroleum  Products,  except Bulk Stations and
      Terminals
                                   8-18

-------









fr
w
X,

•?
§

w
$
CO
WHOLE
w

L^

O
RIZIN
U
£
U
CJ
£
&*
<
Q

CENSUS

I_J
3
oi
w
O


,
p*.
|
CO
u
rJ
1

























CO

J
s
ClH
O

CO
2
ILLIO:
s

2
M
SALES
*-**

(0
r-
00
en
iH
• •

2
M
rJ
0
<
o

f£
o
[IH












1





0)
C
•H
rH
O
(0
It)
o

0
•rH
JJ
<0
• rH
£
**•

JJ VM
c o -Ij
0) •"
S « o
i M. y
w O> tl
0) iO tr>
a












U-l
0 I
-< -s
1*
5 u
_J
2







0)
c
r Gasoli
o
u
o
S


i
u «-i
C 0 ^
m 'w
" ® 0
r< 01 P
4J
o .
n "S
HI "8
1 »
| U
^
z



rH
«
| 1
o
£




^4
_ ° '
« * -8
Mi
2
z








tA
01
rH
|Q
CO






0)
41
rH
A)
M



tfl
.shmenti
•r^
rH




Sales



Sales




ishments
•f-i
rH



«
01
p^
«)
CO




0]
JJ
c
0)
1
Cfl
•^
rH


o
M
CO




*J" (N ro
O O O







vo o\ in
o *» in
VO rH t—





•*f O **
in o in
O rn ro
rH rH





in IN CM
in oj "3«



TJ O ^f
rH f- 00
r- o r~
vo I-H r~
r- CN a\




o rr TJ<
r- r- •«*
00 rn (^
o oa r*>
rH rH




in o\ •*»
in rH r~
VO (N CO
a» in •*#
m o\ n
rH (S



m o vo
in r-  rt p*

r- t^ ^ 5
rH t-l *? 25
in m ^ ;2
£
c^




,fc

u
O
M-l
T3
01

JJ
C
3
0
U
(J
«
fN
C~
rH
in

T3
c
«J
rH
r~
rH
in
la
O
M
CO

r-
00
<7l
^
C
M
«





























CU
rH
ja
a
i-H
•H
§
r*
a
jj
o
M
>rH

Q)
C
•»*
i-H
O
M
a
o>

rH
O
H-l
r-t
•0
JJ
O
JJ

cu
1

-------
integrated operations from the refinery down to the retail
service station level.  Independents,  also known as "jobbers,"
can be vertically integrated but often are integrated to a
lesser degree than the majors or semi-majors.

     Census data indicate that refining companies have the
largest share of wholesale gasoline sales  (approximately 55
percent in 1987), although the majority of establishments
involved in wholesale gasoline (80 percent in 1987) are owned by
companies that do not refine gasoline.

     These economic entities are related to one another in a
myriad of ways.  For example, refiners typically operate bulk
terminals with salaried personnel.  Most bulk plants,  however,
are operated by independent wholesalers.  Some bulk plants are
operated by cooperative associations or by the refiners
themselves using employees/agents who work on. a salary or
commission basis.  Cooperative associations own a small number
of bulk plants.  These serve mostly farmers, and available data
are limited.

     Historical data are available for bulk plants and terminals
(SIC 5171) describing recent trends in wholesale gasoline
establishment ownership and sales.  Figure 8-3 reveals that non-
refinery firms' shares of total wholesale gasoline sales and
total wholesale gasoline establishments increased between 1977-
87.

     Establishment and firm  size and  concentration.   Data  from
the 1987 Census of Wholesale Trade—Establishment and Firm Size
on the size of establishments and firms classified in SICs 5171
and 5172 pertain to all company activities, not just gasoline
sales.  Because gasoline sales are a large percentage of their
total sales, these data are assumed to be representative of
gasoline wholesalers.  Figures 8-4 and 8-5 show that refiner-
owned establishments were substantially larger and more numerous
than non-refiner-owned establishments.  On average,  refiner-
                              8-20

-------
 30.3
0.5
                                    34.9
                           69.2
                                      0.7
                                                                64.4
             1977
                                                  1987
                Percentage  of Total Gasoline Sales
62.5
                           35.9
                          1.6
                                       85.7
                                                        13.3
                                                              1.0
              1977
                                                   1987
              Percentage of Total Gasoline Establishments
     Refiners   [") Agents, Brokers,  and
                   Commission Merchants
Non-Refiners
   Figure  8-3.  Wholesale Gasoline Establishment Ownership
                and Sales Trends:  SIC  517131'32
                               8-21

-------
            $50 -r
           $40 ••
   Sales    $30 "
(millions of
  Dollars)   $20 +
            $10 ••
             $0
                                                             $49
                        $42
                   Refiner-Owned
                  Establishments
              Non-Refiner-
                  Owned
              Establishments
Agents,  Brokers,
 and Commission
   Merchants
          Figure 8-4.
 1987 Sales  Per Establishment:  for
SICs  5171 and  517228
                                8-22

-------
  Number of
Establishments
    Owned
20 .
s
10 •
•
•
27.4



1.2
7.8

1.1

[3 Refiner-Owned
Firms
B Non-Refiner-
Owned Firms

                    SIC  5171
SIC 5172
     Figure 8-5.  Refiner  vs.  Non-refiner Firm  Ownership
               of Wholesaling Establishments28
                             8-23

-------
owned establishments have substantially greater sales than non-
refiner owned establishments.  The 1987 Establishment and Firm
Size data presented on Table 8-8 show concentration by the
largest firms in the two SIC industries.  This table shows that
concentration is higher in refiner-owned firms than non-refiner-
owned firms.  Table 8-9 provides data characterizing trends in
SIC 5171 concentration between 1977 and 1987.  These data show
that overall concentration declined between 1977 and 1987 in the
overall bulk station/terminal market.

     Financial  ratios.  Financial  data  and  ratios  are  available
from Dun and Bradstreet's Industry Norms and Kev Business
Ratios.  This source presents "common-size" balance sheet and
income statement data along with key business ratios on
solvency, efficiency, and profitability.

     Table  8-10 shows three commonly used profitability  ratios
for SICs 5171 and  5172 in 1987, 1989, and 1990.  Financial
analysts tend to look increasingly to the return on net  worth as
an absolute measure of a firm's profitability.  The consensus
among financial analysts is that a return of at least 10 percent
is required to  provide dividends plus adequate funds for future
growth.37

8.1.5  Storage  Facilitv-Specific Data
     The EPA defines bulk plants and bulk terminals using
gasoline throughput.  Bulk plants have gasoline throughput of
75,700 liters  (20,000 gallons) per day or less; bulk terminals
have throughput of greater than 75,700 liters per day.   "Bulk
Station" is a Bureau of the Census term for bulk plant.
Throughput  is not  the determining  factor used by the Census for
separating  bulk stations from bulk terminals.  Instead,  the
Census uses a combination of storage capacity and method of
incoming product transportation to identify these facilities.
Although most other sources use the term bulk plant rather than
                               8-24

-------
TABLE 8-8.
CONCENTRATION BY  LARGEST FIRMS:
   SICs 5171,  AND 517228
1987,
SIC
5171
4 largest firms
8 largest firms
20 largest firms
50 largest firms
Mon- Re finer -Owned
4 largest firms
8 largest firms
20 largest firms
50 largest firms
Refiner-Owned
4 largest firms
8 largest firms
20 largest firms
50 largest firms
Number of
Estab-
lishments
12,353
341
692
1,327
1,587
10,400
31
58
104
185
1,781
340
688
1,316
1,715
Sales
% of
Total
100.0
2.8
5.6
10. -7
12.8
84.2
0.3
0.6
1.0
1.8
14.4
19.1
38.6
73.9
96.3
Amount
($106)
139,655
23,655
42,082
72,841
90,329
62,954
6,913
10,575
16,134
(W)
75,219
23,654
42,035
67,971
74,976
% of
Total
100.0
16.9
30.1
52.2
64.7
45.1
11.0
16.8
25.6
(W)
53.9
31.4
55.9
90.4
99.7
Paid
Employment,
March 12, 1987
Number
135,923
4,552
8,487
15,385
21,222
114,667
1,672
2,342
4,231
(W)
19,227
4,551
8,424
14,108
18,530
% of
Total
100.0
3.3
6.2
11.3
15.6
34.4
1.5
2.0
3.7
(W)
14.1
23.7
43.8
73.4
96.4
                                                 (continued)
                        8-25

-------
        TABLE 8-8.   CONCENTRATION BY  LARGEST  FIRMS:
                  SICs  5171,  AND 5172  (CONTINUED)28
                                     1987,
                                                              Paid
                                                           Employment,
SIC
Number of
Estab-
lishments
% of
Total
sax*
Amount
(S106)
es ruaicri J.^
% Of
Total Number
. , iyo /
% of
Total
                       4,373
        100.0
        95,219  100.0
                 39,265   100.0
4 largest firms
8 largest firms
20 largest firms
50 largest firms
   58
  112
  289
  429
1.3
2.6
6:6
9.8
27,224
39,600
55,380
70,227
28.6
41.6
58.2
73.8
1,378
1,945
3,506
5,989
 3.5
 5.0
 8.9
15.3
Non-Refiner-Owned
3,701
                                 34.6
        61,945   65.1
                  34,106    86.9
4 largest firms
8 largest firms
20 largest firms
50 largest firms
   27
   34
   57
  140
0.7
0.9
1.5
3.8
17,251
24,901
35,074
44,496
27.8
40.2
56.6
71.8
                             830
                          1,111
                          2,167
                          3,813
                            2.4
                            3.3
                            6.4
                           11.2
Refiner-Owned
                         438
         10.0
        17,473    18.4
                  4, 048
                                             10.3
4 largest firms
8 largest firms
20 largest firms
50 largest firms
  103     23.5     11,510   65.9        952    23.5
  238     54.3     14,589.  83.5      1,716    42.4
  325     74.2     16,803   96.2      3,408    84.2
  431     98.4     17,469  100.0      4,032    99.6
(W)--Withheld to avoid disclosing data  for individual companies;  data are
     included  in broader kind-of-business totals.
SIC 5171--Petroleum Bulk Stations and Terminals.
SIC 5172--Petroleum and Petroleum Products Wholesalers,  except Bulk
     Stations  and Terminals.
                                   8-26

-------
      TABLE 8-9.   TRENDS IN CONCENTRATION  BY LARGEST  FIRMS:
                     1977-1987 (SIC  5171)28,33




SIC
5171;
4 largest firms
8 largest firms
20 largest firms
50 largest firms
Non-Ref iner-Owned :
4 largest firms
8 largest firms
20 largest firms
50 largest firms
Re finer -Owned:
4 largest firms
8 largest firms
20 largest firms
50 largest firms
1977
Percentage of
Total
Establish-
ments
100.0
7.6
20.1
27.9
31.8
64.4
0.5
0.7
1.1
3.1
34.1
22.2
58.9
84.7
95.3
1987
Percentage of
Total
Establish-
ments
100.0
2.8
5.6
10.7
12.8
34.2
0.3
0.6
1.0
1.8
14.4
19.1
38.6
73.9
96.3

1977
Percentage
of Total
Sales
100.0
28.7
45.5
61.4
69.1
35.7
8.2
12.0
18.9
25.7
63.9
44.9
71.2
94.0
99.0

1987
Percentage
of Total
Sales
100.0
16.9
30.1
52.2
64.7
45.1
11.0
16.8
25.6
(W)
53.9
31.4
55.9
90.4
99.7
(W)--Withheld to avoid disclosing data for individual companies; data are
included in broader kind-of-business  totals.
                                 8-27

-------





o
CTv
a\
iH

*
cn
CO
a\
r-{
^
r-
CO

r-l
co
O
M
£•«
<
oi
vo
iST
^^
M 2L
HTo
5ROFITAB]
AND 517:
rH
^ <>
< ^
M ^
Z w
< y
g 5
a,
Z
M
C/)
Q

^\
CM
a:
H



o
r-l
1
oo

TABLE








u
rC
AJ
r-l
o
g
AJ
0)
2
C
O
g
J_l
3
AJ
O>
K





A
10

01
to
CO
*e£
c
o

c
rH
3
AJ
01
tt



ID

rH
id
W
c
o
c
rH
3
AJ
0)
K




r-
00
en
rH


cn
00
en
rH




O
en
en
rH


t-
00
en
rH





cn
00
cn
rH

O
cn
cn
rH


r-
00
cn
rH


cn
00
en
rH





0
o\
cn
f^





O
rH
•>H
AJ
rH
<
s



o
M
M


cn co o r- cn un
o o «* m 
o m r^j IN \o (N






cn in *a» rH oo in
00 »* rH O Tjl rH
rH

^3" 
cn in rH o in r-t
rH


cn ^j in in vo vo
 0 04 0) O
D 2 J ID S J



rH N
r- r»
»H rH
in tf»

























•
AJ
•rH
VM
0
rH
a
c
Oi
c
•r<
.X

CO AJ
AJ «
•H O
IM -H
2 •§
0. M
« J2


















—
xJ
c
0)

AJ
in
<1>
>
•rH

10
ockholder
AJ
in
>,
4J
•«H
&
01
CO
«
Vj
o>

O

<4-l
O

u
3
AJ

-------
bulk station, it is prudent to only compare the total number of
facilities between the different sources.

     8.1.5.1  Bulk Terminals.  Table 8-11 shows that  the number
of gasoline bulk terminals operating in 1990 is only  three-
quarters the number  operating in 1977.  Table 8-12 shows time-
series data on ownership of bulk terminals by major/semi -major
oil companies versus all other entities.

     8.1.5.2  Bulk Plants and Bulk Stations.  Bulk plants
receive approximately one-fifth of the total volume of gasoline
that moves through the U.S. gasoline system.  Figure  8-6 shows a
5 percent decline in the percentage of motor gasoline passing
through bulk stations between 1977 and 1987.

     Table 8-11, which showed bulk terminal estimates, also
shows the estimated number of bulk plants for several years over
the period 1977-1990.  Non-Census sources of bulk plant data
include PMAA's Marketer Profile Survey.   Independent marketers
reported to PMAA a 26 percent drop in average storage capacity
from 616,955 liters in 1987 to 454,200 liters in 1989.  PMAA
believes that the capacity decline is related to selective
scrapping of older tanks that do not warrant upgrading or
investment, rather than closure of entire facilities.  An April
1989 study by the National Petroleum Council found that total
bulk plant storage capacity declined from 65 million to 50
million barrels between 1983 and 1988. 47
8.1.6  Gasoline Transportati
                            on
     Pipelines move the greatest volume of gasoline the greatest
distance through the distribution system.   Although,  published
data are not available on the total volume of gasoline that
moves by pipeline,  related data have been identified.   Figure 8-
7  presents data on the relative proportions of petroleum
products moved by various transportation modes in 1974 and 1989.
                              8-29

-------
TABLE  8-11.  ESTIMATES OF THE TOTAL NUMBER OF WHOLESALE
         GASOLINE STORAGE FACILITIES:   1977-1990
  Year             Bulk  Plants       Bulk Terminals         Total
  1990              11,00038             1,33539            12,335
  198740            15,000       .        1,500              16,500
  198241            15,000               1,500              16,500
  197741            17,850               1,751              19,601
                            8-30

-------
            TABLE 8-12.   FACILITY OWNERSHIP:   TERMINALS
                      (PERCENTAGE OF TOTAL)42'44
                                  1990 Bulk     1987 Bulk     1982 Bulk
       Segment Category3           Terminals     Terminals     Terminals

 Major + Semi-Major                  70             79             79

 Independent/Other                   30             21             20


aSee Section  8.1.4.3 for list of companies that fall under each category.
                                8-31

-------
 Percentage of
 Bulk Terminal
Motor Gasoline
  Throughput
Passing Through
 Bulk Stations
                 50 -
                 40 ••
30 ..
                 20 -.
                 10 ••
27
                                    21
                              1977
                                    1987
       Figure 8-6.   Trend in Bulk Station Throughput45'46
                                8-32

-------
              Railroads
             Railroads
       Trucks
 Water
Carriers
     Trucks
                          Pipeline
 Water
Carriers
                          Pipeline
               1989
              1974
  Figure 8-7.   Transportation of  Petroleum Products,  1974-1989
             (Relative proportion of total  ton-miles
                  shipped by various modes)48,49
                               8-33

-------
     Data on  the transportation of gasoline through the
marketing chain show that shipments further upstream in the
chain  (closer to the refinery) are mostly made by pipeline or
water  carrier; shipments further downstream in the chain
predominantly move by truck.

     8.1.6.1  Trucking.  Gasoline  trucking  firms can be
separated into three categories:   (1) 'private carriage," major
oil companies owning gasoline transport vehicles;  (2)  "common
carriage,"  firms providing transportation services to major oil
companies;  and (3) "jobber entities," independent firms
transporting petroleum products, but are also involved in some
other  aspect  of the petroleum marketing business such as owning
bulk plants or service stations.  Data on trucking
characteristics are available from the U.S. Census1 Truck
Inventory and Use Survey for two relevant categories:  petroleum
shipments and tank trucks  (liquids or gases). Table 8-13
displays the Census data characterizing the liquid/gas tank
truck  fleet in both 1982 and 1987.  The median age of tank
trucks was  8-9 years in 1987, compared to 7-8 years in 1982.

     Both  the PMAA's Marketer Profile Survey and an unpublished
1983 Census study conclude that the primary means of moving
gasoline from terminal to bulk plant to customer was by truck.
The number  of transport trucks owned by independent marketers
rose from 14,593 in 1987 to 19,630 in 1989; the per-marketer
average increased from 1.4 transports in 1987 to 1.8 in 1989.50
PMAA's survey also found that independent marketer use of common
carriers continued to increase in 1989, but that most marketers
continue to transport the bulk of their own sales volume.47

     8.1.6.2  Pipelines.   Most of the available data  for
pipeline movement includes all petroleum products and crude oil.
The Federal Energy Regulatory Commission (FERC) requires common
carrier, interstate pipelines to file annual reports on total
                               8-34

-------
        TABLE 8-13.
LIQUID/GAS TANK TRUCK CHARACTERISTICS
     IN 1982  AND 1987
                                                   1987
                                          1982
Total Number
                            213,000
   241,600
                                              Percentage of Percentage of
                                                  Total         Total
Manor Use
 Retail Trade
 For-hire Transportation
 Wholesale Trade
 Others
                             24
                             16
                             15
                             45
   25
   16
   14
   44
Range of Operation
  Local
  Short-range  (<200 miles)
  Off-road
  Long-range  (>200 miles)

Model Year
  Approximate median
Operator Classification:
 Not for-hire
 For-hire
   Motor carrier
   Owner/operator
Operator Q\ass ^fricat ion:
 For-hire jurisdiction
  (continuedL
                             63
                             22
                              8
                              7
                         1978/1979
                             34
                             16
                             12
                              4
    65
    19
    11
     5
1974/1975
   83
   17
   14
    3
Interstate
Intrastate
Local
Products Carried;
Petroleum
Chemicals
Others
Truck Fleet Size:
1
2 to 5
6 to 19
20 or more
46
41
12

56
15
29

16
25
34
26
53
30
16

71
20
10

18
23
28
30
                                  8-35

-------
petroleum products deliveries and total product pipeline
mileage.  In 1989 these companies comprised 79,624 miles of
products pipeline and 4.85 billion barrels (771 billion liters)
of petroleum product deliveries.  These figures represent
declines from 1988, which showed 80,264 miles of products
pipeline and 4.97 billion barrels (790 billion liters) of
products deliveries.

     Table 8-14 displays data on the top 10 pipeline companies
in 1988 for two categories:  petroleum product deliveries and
products trunkline mileage owned and operated.  Pipelines are
joint ventures involving several (usually large and well-
integrated) companies.

     The FERC does collect limited data characterizing
profitability in the overall liquids pipeline industry.  In
1989, for only the second time since figures have been
collected, net income as a percentage of operating income
declined from the previous year from 36.5 percent in 1988 to
34.2 percent.  In 1978 net income was 21.9 percent of operating
income.

8.1.7 Gasoline  Distribution  Industry:  Retail and Consuming
       Sectors
     8.1.7.1  Industry Employment and Sales.  There is no
comprehensive source of employment data for gasoline retailing.
The Bureau of the Census collects data only for payroll gasoline
service stations that receive 50 percent or more of their
revenue from automotive fuels or lubricants.   Table 8-15
displays historical Census data on the number of stations,  total
sales, and employment in gasoline service stations that fit the
Census definition.  In addition to the 701,690 people employed
by service stations, at least an additional 22,432 were employed
in the non-payroll stations counted by the Census in 1987.   The
                              8-36

-------
TABLE 8-14.
RANKINGS  OF MAJOR PETROLEUM  PRODUCTS
  PIPELINE  COMPANIES51
 The Top  10 Liquid Pipelines in Product  Deliveries--1988
                                      Product
                                     Deliveries
                                    (thousand of
                                      Product
                                    Deliveries
                                    (thousand of
Company
Colonial Pipeline Co.
Santa Fe Pacific Pipeline Partners LP
Buckeye Pipeline Co. LP
Chevron Pipeline Co.
Marathon Pipeline Co.
Phillips Pipeline Co.
Plantation Pipeline Co.
Explorer Pipeline Co.
Williams Pipeline Co.
Mid-America Pipeline Co.
The Top 10 Liquid Pipelines in Miles of Products
1988
Company
Mid-America Pipeline Co.
Williams Pipeline Co.
Colonial Pipeline Co.
Phillips Pipeline Co.
Chevron Pipeline Co.
Texas Eastern Products Pipeline
Buckeye Pipeline Co. LP
Santa Fe Pacific Pipeline Partners LP
Plantation Pipeline Co.
ARCO Pipeline Co.
bbl)
635,620
315.300
284,536
247,955
238,129
222,775
189,000
174,143
173,576
162,909
Pipeline
Mileage
8,082
6,775
5,274
4,192
3,385
3,373
3,289
3,174
3,146
2,831
liters)
101,044,511
50,123,241
45,232,688
39,417,406
37,866,367
35,414,542
30, 045,330
27,683,513
27,593,377
25,897,644
Owned/Operated--











                         8-37

-------







(Tl
U
CO.
§
M
fr>
<
EH
cn
u
U
M
>
ai
Da
to
Q
DJ
Z
M
Du
U
Q
I
D
<
U
as
D
CQ

to
E3
to
Z
[-0
U

Z
t— (

to
Q
Z
U
juf
QG
C_l
C^


•
m
r-t
|
CO
U
J
OQ
<
C-i
t.















,
>• xJ
°§
I1
Lu
Jj 1
C 1 0> TJ
01 C 01 01
O fl W C
X.4 W •*•* 3
01 Cu £ O
w a o
01
10 i
W 4J J
C S"D
4) ^ 0)
o 5 c
u §• 3
ai 5 O
°-5
^
Ij 
O) "—"
CX
c:
to o
0) ---I
1— 1 JJ
fl fl
U) JJ
W

i j u > u
>-t ^ (J «J 
.X 01 C 01 0)
j_i u  n
U "° " tJ 0) 0)
ai x u a,
Q. Q a.

_Q
«, ai w
o J{ §
•H O
^ *^
* LJ i_i
« W U
2 <" fl
CO jj
W



u
<0
V
>!


00 n us o
vo r~ co CP> ,..
VO '-D OJ O ^
* «. * » .^
r~ M ^r — < *-
•a1 r- o o
r~ vo i-o r»
CO O O *g* *-*
OA O rsl ^^ rH
C^ CO CD OO CO



(N O O ^o O
O O CO CO CO
(N CM f-H -H i-l



(N
«• vo n CD if)
C— Cf\ CO CM ^
f"~ & CT\ CO ^
o cfi n cr\ C^
^T r-t n ^o
r-i r-i r- r~ ^

r^- CD ^^ C**
r~ CO rH fM



o co o o u">
CO U5 O O •*
CO •«? CN) rs r-^
^ % « H »
t-H vo '-o &\ un
n LTl O CD r-t
r-l r-l


O tj\ O ON O

o o r^4 r^ rvj




O r-l O T-l O
• . . . •
O CA CO CO CD
(N i— 1 i— * *— 1 «-t




01 co o o
0
JJ E s
to o <*
fl U "-1
0) iJJ
T! O
j_i ai «-" •
fl u 0
fl (0 CTi
rj, (_H j_J CT
C 3 *O r—
-,< o T3 '
> r-l CO
.rt fl W CO
a; o o) cr>
0 ^ -
nj o> *O
u rji W >
c e
10 fl O
.j -c • £
* ° 2 8
jj 01 fl ^
3 01 0)
0 fl S 0)
Jj J=
r-t C J= •"
•rH 01 JJ
fl o g c
JJ U XI -H
0) 0)
,-t a c o
-t C
W ' o "-"
iO en 10 10
CJ\ <0 "H
m . ^ ai JC
c w co
O JJ O — ' c
• M C ij U) T3
jj iO 31-t
fl (J p- •£ ^
JJ -H CO
10 1-t en C W
-Q ) -rH •
01 3 C
U r-l C <0 O
t-o ° o^
U C . T3 -H fl
(u c i-t o) J-1 ~*^
W fl CO <0 O
ID o) fl JJ O
ai -i >, J3 n w
c oi 10
-rH 3 U-t T3 "-1 <
r-l « O 01 O
O jj ul
W W T3 fl -^ C
fl > C E «J -H
0) •H 3) .r-l XI W
jj jj e -H
W 0 u U) 3 .C
01 g fl 01 C O
CO C
•H J-> W 10 •* fl
UJ 3 c C fl r-J
01 Q O O 0> tu
13 .-t .H Q.
g u JJ --I
30 fl fl U fl
fl 14 jj jj O C
01 «M W co «-" o
lJ •-•
3(/l iu u-t J^ J-J
m oi o o g g
W C L, ,J J3 rJ
3 01 oi a) JJ o>
W > XI XI U •£
CO" E 6 « C
01 U 3 3 Cu M
U Z Z
a xi u















































•
01
1— 1
XI

-------
U.S. has approximately 70,000 convenience stores, of which about
65 percent of them sell gasoline.52

     8.1.7.2  Retail Motor Outlets and End Users. Retailing of
gasoline takes place at traditional gasoline service stations,
car washes, automobile dealers,  and convenience grocery and
liquor stores.  Retail motor outlets provide a wide array of
product and service mixes to consumers.  MPSI Americas, Inc.,
divides the retail motor outlet population into four major
categories: conventional stations, pumpers, convenience stores,
and other.  Conventional service stations have service bays for
automobile maintenance and repairs.  The other three categories
do not have service bays.  Pumpers are large-volume self-service
sellers providing few, if any,  of the traditional service
station services.  Convenience stores are differentiated from
the other three types by the larger amount of floor space
provided for the display of food and other convenience items.
The "other" category includes outlets of any type that have
other facilities, such as a car wash, or a quick oil change and
tune-up facility.

     Table 8-16 shows the 1987 and 1989 market share breakdowns
of the number of outlets and gasoline volume by retail outlet
type and U.S. region.  One obvious trend that the data show is
that average store volumes are increasing,  which corroborates
the Census data presented earlier.  The data also show that
service stations and "others" have decreased in market share in
both numbers of stations and volume,  while pumpers and
convenience stores have increased in market share in numbers and
volume.

     Table 8-17 shows some of the trends in convenience store  •
retailing of gasoline.  Convenience store gasoline sales have
increased from approximately $20.6 billion in 1987 to $27.1
billion in 1989.  Various end users of gasoline,  including
industry,  commercial and government fleets,  agriculture,
                              8-39

-------


•<*
in
»
ro
m

oo
cr>
t-H

2
r»
00
o\
^^


(0
u
Oi
>l
rj
t^
£
j
EH
3
O
J
M
<
C_,
~
U
(X
><
S
CO
w
s
X
CO

a
«
«!

J
s
s
o
HH
S
*•*«
s

1
^
i
8
M
O
u
Pi
VO
r-l
1
00
TABLE



in
S
JZ
JJ
o



to
0)
u
o
jj
w
CJ



01
>-,
0>
a
e
•3
a.




r*1*
CO
~


ON
CO
0


r^
CO
ON




o\
CO
O"\

r~
CO
ON


ON
CO
CN



en
c
o
•f4
4J
flj
JJ
M

01
ervic
CO



rH
(C
JJ
O
EH



1—
CO
ON



CN
CO
ON
rH


r~
CO
CN
r-l



ON
CO
ON
r-l









vo ON
" ° rS S
- - ^ „-
— « VO

fN  J2!
r-l in - ™
vo m T
*> Z
in ^
o o £ r-.
00"*
0 0 - ON
r-l r-l 2 r-l
fN
VO ^
0 0 £ £
d 0 * ^
22*3
10 c^

rH
(0 _
Ol -J
<~f ^^
01 01
(Q E E
iJ 0) 3 3
0) E rH rH
rH 3 O O
•U rH > >
3 0
0 > >, >,
rH rH
u-l u-l r? r"
O O J-) J-1
to! *> *> o o
|ss"a
i u u oi o» S
t a) ai > 5 S
g a. o- < ,< ^
a x
0> CO
ON ^ g ro
2 ^ ON rs
r» r~

r- oo
^ « S S
rH ^ CO ON
rH VO
™ £
vo o ° oo
» « vo" £
•«• ;_,


^ 5
rn 00 g 0
CD vo * m
TT
T r~
* - ? 2
s s s i
rH ^T
r-4 ro
^ ^ P o^

'* ^ r~ r-
" « 0 0
r-* -«T

™ s
o vo in S
ON CM . -
<-i n ON C?
1/1 S



0 ON S g
*«' c-- ""-,,"
r-i fN ON ?:
^ PJ
. r1
0 0 £ vo
0 0 ^ "*1
0 0 ' U1
rH r-i r; r-
"^ fN
VO *>
0 0 « fN
• iij vo
0 0 M
0 0 ^ CO
r-l r-l CO
^ (N

rH
(0 -~
Ol J
— ~—
Ol Ol
W E E
.U 01 3 3
0) E rH rH
rH 3 O O
tJ rH > >
3 O
0 > > >,
i~-* rH
vw >" j: jz
O O AJ J-l
c c
4J iJ O O
c c s z ,.
fl) 0> ^
o y • • c
r-l U Ol 01 $
01 iv > > -Q
o. ft- < < §
05
vo m
vo ^. r-
• vo m
f^ - - -
rH " CD O
rH (—

VO VO
co r» ef>
• . rH f-
rH ^ CO 00
rH VO
Ln «
fN fN ON jj
o in . _~
m rH CD g
fN 2
«~H

" !Q
^ ° S ^
fN VO . "
n -i oo g
^ "
° °.'< <
0 ^ ^ ^
r-i m •*• ^-

m co
« - S ?

s 5 s s
rH rO

in ^
^ « S 5^
S S f," -
•" S



r ^ i ?
•a- r-i . '
fN  rH > >
3 O
0 > >, >,
VM UH jc jc
O O JJ jj
C C
IJ 4J O O
C C X S d
o»  > M
0. 0. < < «
2
r- n
w ^ ° 2
03 ™ 0 Ov"
,-^ r~

r- vo
^ ° S 5
00 ™ o ON"
rN r»
o £
- i s *
r- "• uJ e*
***  S S

* - S S
rH «T

0 <*
ON rH 0 ^
r-) in - '
in TT vo
m ^



VO ^
co fN g jq
S S in" -
* S
_ rr
o o iQ S
0 0 ^ 1
o o _," m
rH rH ^ VO
^ S
. r*1
° ° S -H
0 0 « °1
o o * r-
rH rH „ 0

*— 4
(Q — •
cn -J
-™- •_»
0) 0>
n E E
JJ 0) 3 3
0) E rH rH
rH 3 0 0
iJ rH > >
3 O
0 > >, >,
^H ^^
vu <4-i x jr
O O iJ J-l j
C C »
4J iJ O O V'
c c s s: -
0) 0) —
o o • • _
u u 01 oi ^
ai o) > > £
0, Q. < < £
VJ
vo ro
in _ in rn
• ~ vo vo
rH ' »
rH ^ 00 0
rH f~

rH n
™ « S P
2 ^ rT vo'
rH VO
o £
- o, « "
" " - S
r> *J
t— 4

=° S
r - s 5
^ co * co
^ fN
n J_J
VO rH
- o 2 S
53^*
^ ^ rH f\)
rH T3«
CO ON
- <>» S ?

S S 2 S
rH »3>

r-l rt
r- ON co S
• • CTt
ro 0 - ^
"S S



* £
oo r- vo JJ
^ 2 ^
^ S



*
•o
03
e
cn
03
•>
^
jj
a
§ c
2 0
ij If^
s *J
a
is
w
.^ c
•o g
01 U
j_i
U r-l
.*: r-1
JH «J
03 >->
s v
>Jl 0
-1 o
CU 4J
5 >,
s?
'Jj k-1
03 «
01 U
*J ^^
t/i 'tJ
i s
C r.
a/ w
£s
"•> fTl
3 u
. j-j
rn O
" c
= 0
Q; T3
f-g
^ T ^ i
002;^ 3
• n II ^
S § S vo-I S «
rH rH "^ TT II ^ U
^ M U "> '2
H C 3
o o ^ SB ^ >
O4 .«. ai *^ Jl_
oo°° - S
o o 0 m o -c
^ rH i; vo
fN

rH
ID —
2 d

01 Ol
<0 E E
iJ 01 3 3
0) E rH rH
rH 3 O O
JJ rH > >
3 O
O > >1 >1
rH rH II
**H I4H JC U^ U
O 0 JJ J->fl
C C
jj AJ o on
c c z sg
01 01 1
u u • -B
u M 01 oiH
01 0) > >l
o. a. <  c
IH 3
JJ OJ °
5«0
i jj O
< (0 J-1
z^
8-40

-------







in
in
in
O
2
M
J

w !2
'c ®
£ £
6 u
w
0)



c
•H
0
05
C3
t
c
o


H)
c
•H
(-H
0
w
o

01
c
•rt
'o
V)
US
o







a>
O)
fl
c
0)
u
0)
a.












•— i
a


c
ja
M



* 0)
ui d W
2 fl o)
U i—l
t- C fl
.5 « w

fl ,? °
s cu

"J 8)
w en w
» fl 0)
"* U t— 1
Margin
Percen
of Sa

* ^
Q* d)
Xj
w o 2
QJ JJ
fl "
CO


r*
•" S^~
rH " 0
O — <-i
(fl JJ 
« M ~
O
2»S
SS?
M-S
S^g


to
*4-J 4)
0 Vi
0
O M
2 i
0

t.
10
O O
0 O
(N TJ"
 VO





rH &\
o in







in \o
<* 0
t-i


p» in
&\ o



in o
^JD O
oo vo

CTv O
(N

•** in


o o
0 0
o in
o t^
in to

CN r>
00 00

-------
aviation, and marine users, buy from the wholesale gasoline

market.  In 1989, less than 3 percent of gasoline was consumed

by these sectors.  Except for aviation gasoline facilities, no

recent data are available for these "bulk-users" of gasoline,

other than the data presented in Table 8-2 on the amount of

gasoline consumed.

     8.1.7.3  Economic Agents.  As with the wholesale sector, a

myriad of participants and relationships exist at the retail

level.  Retailers of gasoline may be single-site dealers,

operators of retail chains, jobbers, small refiners, or large,

integrated oil companies.

           Combinations of  ownership may also occur.
           For example, a landowner may  lease property
           to an  oil company which then builds a
           station and subleases the property to a
           dealer.  Also, a third party may lease a
           station to a wholesaler who in turn
           subleases to a dealer or operates the
           station directly.   These are but a few of
           the more common  combinations of
           ownership.57


     Service station operation methods are also diverse.   The
operator of a retail outlet is typically an independent

entrepreneur operating one or more outlets.

           The retail outlet operator is usually not
           an employee of an oil company;  refiners
           typically operate terminals with salaried
           personnel, but contract with independent
           wholesalers and  retailers to operate bulk
           plants and retail gasoline outlets.57


     Many wholesalers own the land, buildings,  and storage tanks

at their bulk plants,  and many also own retail outlets,  which •

the wholesalers operate  directly or lease to dealers.

     8.1.7.4  Number of  Retail Establishments.   Figure 8-8

presents estimates from  Lundberg Survey, Inc.,  of the total
                              8-42

-------
Thousands
of Outlets
           250 T
           200 +
           150 f
           100 +
            sot
                     210.9
                      1982
190.9
1985

Year
                                                         175
1990
        Figure 8-8.  Estimated Nuinber of  Retail Gasoline
                 Outlets--1982,  1985, 199058'59
                              8-43

-------
number of retail gasoline outlets for selected years.  The
Lundberg Letter estimated 210,900 outlets for 1982 and 190,900

outlets for 1985.  API and Lundberg Survey, Inc. independently

estimated the current number of retail gasoline outlets to be
175,000.  A recent article from NPN estimates the total number

of retail outlets at 210,000.60


     A series of gasoline distribution changes have  led to the

decline in the number of stations over the past two  decades:

      • Changing consumer preferences and station cost increases
       have altered the economic scale of gasoline retailing.
       As a result, the market requires fewer gasoline stations
       to service demand.

      • Gasoline demand growth has dropped substantially below
       the levels of the 1960s and early 1970s..  As  a result,
       the widespread retail gasoline distribution systems of
       many refiners, built in the expectation of strong growth,
       no longer seem likely to afford attractive returns on
       investment.

      • Refiners have attempted to improve their levels of
       profitability and have moved to focus their resources in
       their most profitable business activities.  As a
       consequence, many refiners have sold or closed stations,
       sometimes in groups containing all the stations owned by
       a particular refiner in a multistate region.61

      8.1.7.5  Presentation of_Census Data.  The 1?87 Census of

Retail Trade's Merchandise Line Sales provides data  on sales of

"automotive fuels."  These data show that nearly 93  percent

(over $81 billion) of automotive fuel sales at the retail level

are  from gasoline service stations.  The Census data show eight

other detailed SIC industries that retail gasoline;  however,

only one, grocery stores, has more than 2 percent of all

automotive fuel sales.  Data available from the National
Association of Convenience Stores 1990 State of the  Convenience

gtore Industry show that gasoline sales alone at convenience

stores in 1990 totalled $27.1 billion (total industry sales were

$67.7 billion).55  These 1990 figures show that gasoline sales
                               8-44

-------
made up 40 percent of total convenience store sales, up from 34
percent in 1987.62

     8.1.7.6  Establishment and Firm 5|ze.  Table 8-15 shows
total sales per Census-defined service station increasing from
approximately $140,000 in 1972 to over $1 million in 1990.
Other Census data presented in that table show that service
stations owned and operated by oil companies represented a
slightly smaller share of both total sales and total stations in
1990 than in 1972.

     8.1.7.7  Ownership and Concentration.  Table 8-18 shows
recent trends in concentration for public service stations with
payroll.  These data show increased concentrations between 1982
and 1987 by the largest firms.  Because these figures do not
include non-payroll stations,  they overrepresent the total
market shares of the largest firms in the industry.

     8.1.7.8  Financial Ratios.  Financial data and ratios for
gasoline service stations are also available from Dun and
Bradstreet's Industry Norms and Kev Business Ratios and Robert
Morris Associates' Annual Statement Studies.  "Common-size"
balance sheet and income statement data are presented along with
key business ratios on solvency,  efficiency, and profitability.
Table 8-19 shows three commonly used financial ratios for SIC
5541.  For 1990, the median return on net worth was  15.3,  or
about 50 percent higher than the wholesale median firms '  return
on net worth.
                              8-45

-------
     TABLE  8-18.   CONCENTRATION BY LARGEST FIRMS,  1982-1987:
              SIC 5541--PUBLIC  SERVICE STATIONS63'64
Category
4 Largest
Firms
8 Largest
Firms
20 Largest
Firms
50 Largest
Firms
Total
1982
Percentage of
Establish-
ments
3.3
5.4
8.9
12.8
116,188
1987
Percentage of
Establish-
ments
3.9
6.4
11.2
16.0
114,748
1982 Percentage
of Total Sales
6.4
10.3
17.5
24.4
$94,718,664
1987
Percentage of
Total Sales
7.1
11.0
18.5
25.1
$101,997,440
Note:  Data are only for service  stations with payroll.
                               8-46

-------
      TABLE 8-19.   TRENDS  IN FINANCIAL  PROFITABILITY RATIOS:
                1987,  1989,  1990  SIC  5541--GASOLINE
                        SERVICE STATIONS34'35
Return on

Quartile
Upper
Median
Lower

1990
4.5
2.0
0.5
Sales*
1989
4.5
1.9
0.6

1987
4.9
2.4
0.8
Return on Assets'3

1990
16.5
7.5
1.7

1989
15.7
6.8
2.2

1987
17.6
8.3
2.8
Return

1990
35.9
15.3
4.1
on Net

1989
32.7
13.3
5.3
Worth0

1987
41,1
15.9
5.5
aProfits  earned per dollar of sales.
^Indicates how well a firm has used its assets for making a profit.
GMeasures the rate of return on owner's equity (stockholder's investment).
                                 8-47

-------
8.2  ESTIMATES OF BASELINE YEAR CONDITIONS
     The economic impact analysis represents conditions  in  the
fifth year after promulgation of the regulation, or calendar
year 1998.  To determine the changes due to the regulation,
baseline prices and quantities must first be estimated.  The
baseline is defined as those quantities and prices that would be
expected in 1998 in the absence of the regulation.

8.2.1  Baseline Estimate of Gasoline Consumption
     Estimating gasoline consumption in the baseline year is
difficult because of the instability of crude oil supplies and
the many institutional and technical changes occurring during
this decade.  The Department of Energy's Energy Information
Administration (EIA) has made long-term forecasts of future
gasoline prices and consumption.65  m its consumption forecast,
EIA allows for both increases and decreases in the demand for
gasoline due to growth in the nation's incomes and population
and to improved fuel efficiency and penetration of the
transportation fuels market by alternatives to gasoline.  EIA
calculates gasoline consumption projections for four scenarios:
low oil price, high economic growth, high oil price,  and
"reference."

     Under these scenarios, projections for the annual
percentage growth rate in gasoline consumption between 1989 and
2010 range from approximately 0.1 to 1.1 percent.  The
"reference" scenario represents a mid-range estimate of  .5
percent per year.  Applying the reference case's growth rate to
1989 consumption of 426.7 billion liters (112.7 billion
gallons)2 yields an estimate of baseline 1998 gasoline
consumption of 446.3 billion liters (117.9 billion gallons).
Nearly all of this,  approximately 444.7 billion liters,  is
motor gasoline; only 1.6 billion liters are aviation gasoline.
                              8-48

-------
8.2.2  Baseline Estimates of Gasoline Price and Margins
     Gasoline prices at the retail level have varied a great
deal during the 1980s,  as previously shown in Table 8-3.  EIA
has forecast that over the period 1989-2010, the real price of
gasoline (i.e., price with effect of inflation removed) should
increase 43 percent, an annual percentage growth rate of 1.7
percent.  Applying this 1.7 growth rate to the July 1990 price
(adjusted for the 1990 federal tax increase) yields an estimated
price of $.357 per liter ($1.35/gallon) of gasoline in 1998.

     Wholesale and retail pricing margins are volatile and no
forecasts of future wholesale or retail margins have been
located.  Most qualitative discussions of gasoline margins in
the future have predicted tighter margins in the short run due
to the cost of complying with environmental regulations
(especially underground storage tank regulations).  Ultimately,
however, the margins must cover all costs of production and will
probably increase in absolute terms.  In the absence of
additional quantitative data or estimates,  however,  the margins
developed in Section 8.1.3.2 are assumed to be representative of
the margins for gasoline in the baseline year.

     Table 8-20 displays the estimated 1998 throughput levels
and pricing margins for the key points in the U.S. gasoline
distribution system.  Data were not developed for  particular
entities in the marketing chain if they were unnecessary for the
impact analysis.

8.2.3  Estimation of Baseline Year (1998)  Parameters
     Regulatory and economic forces have brought about
significant changes in gasoline distribution and marketing over
the last twenty years.   For example,  the number of bulk plants
declined 57 percent between 1972 and 1982.66  Therefore,
estimating the number and distribution of facilities within an
                              8-49

-------












co
Z
M
O
iV
l-u
^5
2J

U
2
H
U
M

Qj


Q
^y
*jl
**i

CO
J
to

w
h—l

g
r»
Ukl
X
o

o
Lt*
in
E-i

CO

CO
cn
r— (



*
o
fxj
1
00
to
J
CO

C_I
C^





c 4>
rn ^"*
VH 'H
S ^
^


















*« *
10
>H
0)
jj
•H
rH

C
0
.iH
rH
rH
• H
,£J
*~*
JJ
3
a
01
3
0
IH
r?
E"


















0)
rH 5
* '£

0 °

o






01
c

rH
0
Ifl
(0
o

c
0
•H
JJ
 cn CM cn vo vo vo ch ^»
• • • • . . • • • •
VO«* | rHVOrHOO (Nt~O
^* ^i ^i co co co in ^
«* TJI ^1 fl •*













VO *» T» JNJ O (N
1 1 1 1 1 • ...
rH rH rH O O O












C^ CN] 
o c
C i— I ^J
•H *H <0 rH
«M «J C CU
« C -H 0
a -H g .* c
E C rH 0
e *-• c a> 3 -H
0 fl tt) EH (Q JJ
rJ C fi C
H-l -r4 g AJ g JJ 01
E E O C O M 0 JJ
>! «COr4«Jr< -HIOrH
W D) C 0) M Vt_l rH <4H 4> rH > (0
t> JJ -H £H VM Oi U J2 -H JJ
C r4 rH J^ AJ -H 3 r< O
•^ 0 « A4 rH O X O >&Qi6H
M-l Ot Ol iH -H 3r-4 3M
vVX"H3«Jti3UiD
a:ua>aKE-iAEHC/)
rH
id
>l C
3 'e
[_t jj
ca oi
jj
n
3 0
tl) JJ
c
01 rH
u id
c
Q) -H
-C S
JJ M
0)
LU JJ
0
JJ
M 3 C
•IH id 01
WO) W
rH 3 r<
« a a.
C 0)
id r*i u
CO
>. cn w
H rH JJ
0 C
JJ TJ 01
a ii s
rH J2 Q,
3 W -H
D) -H JS
01 rH • 01
IH ,Q rH
3 -H rH
r4 Ol fl -H
O C )H «
•4-13 r<
Ol
>, c c .
IH O "-i
id in 01
W T3 3
CO 0) -H
0) W rH —
O fl r~
C in o
'O 3
JJ 4> O r4
O JJ rH JJ
C 
W -H E J2
•fH JJ O
Ul JH O
JJ 41 vu Ol
••H
rH U) >
0) Id JJ rH
m c c <-H
3 "4 0) (d
C E E 3
O H O. JJ
0) 01 -ft C
(0 >
•o e oi
Q) O ^
JJ 1-1 O W
§*AJ. Jj
JJ C
•rt r-l C 01
4J -r< 0) 6
(o « u a
0) H W -H
4> jC
- Value was not
a Throughput by
showing 1.4 p
b Assumed all s
shipments) .


jj
'u 0>
0 IH
a e
• *4 m •
id co 10
C X
01 O C
IH -H fl
iO JJ ^
Q
• JJ T3
M W C
JJ 3
COO
« 0 k
rH-rt O>
a> M
IH 0)
.* 0) •O
-H (0 C
3 • 3
CO
« •- E
W O
0) U

w id
3 rH >,
O Q,rH
rH U
-HOC)
0 JJ 01

•H -r«
W 13
fl 0)
c. in
T3 -H 0)
O> rH C
c 0 id
•rt (fl rH
**H *d pt
4* CT
•0 0
0) JJ
41 W C
>H C -H
fl 0)
a oi
U! W C
£"<"H
IH "O rH
0) 0
jj 0 
•H (H |Q
JJ -H
•H 3 a
»H 0* g
u * a
VU JJ JJ
id  0)
HJ-H -H
(rt jj jj
OI-H -H
c For aviation
storage facil
airport facil

8-50

-------
industry sector is challenging.  No projections are publicly
available, but historical data illustrate some of the trends.

     The general method used to estimate the baseline number and
distribution of facilities involved the following three steps:
     1.  Estimate the total number of baseline facilities in an
         industry sector by regressing historic facility data
         against time.
     2.  Estimate the number of facilities by facility size
         category in each industry sector using historic sales
         and capacity data while controlling to baseline levels
         of consumption.
     3.  Reconcile the differences in estimates of the total
         number of facilities made in steps 1 and 2 while
         maintaining the relative distribution of facilities by
         size estimated in step 2.

The Economic Impact Analysis contains a detailed description of
the data and procedures used to complete steps 1 and 2 above for
each industry sector.^7

     Tables 8-21, 8-22, and 8-23 present the results of the
initial estimation (step 2) of facility populations and
distribution of model plants within facility categories for the
baseline year.  Values in these tables have been rounded because
these numbers are projections.

8.2.4  Final Estimates of the Number of Facilities in the
       Baseline Year.

     Initial estimates of the total number of facilities in 1998
were adjusted to account for the throughput distributions and
for total estimated 1998 consumption.   The number of bulk
terminal facilities calculated from the Census-derived model
plant distribution and estimated 1998  throughput is
approximately 1,020.   This  figure is comparable  to the estimate
of 1,174 terminals in 1998  derived from the regression estimate
of Step 1.
                              8-51

-------
'O
GO u-i
a\ ^
S T, ° w
o\ *n Uf
-• * s,"
- 2" JJ
g * .H
cn .6 "f,
& * »o
2 w o J{
< w n £
J ,?
04 fl<

«
i-3
D 0)
CQ JJ 
« -5 ,, «
0 JJ £ 2
«f.
w a, gi o
| « 2 c
*? Jl j 4G O
i_j |i ••"* d* ._j
/\ 11 X ~
8 3
•J ft 'a
w a £
Q <
O
211
u
Cu ^
0 ° «
«» 3
2 w S-
O fl o.
x** II i i W
WII *"^ "1
II r1 •"*
C-* 1 ? O
tr* II fl) y
3 *£
CQ II U £
M a» ^
or ^
HM II
C^ II
" II
01
M l|
O II r-t
II 0) rH
Q •§ * «
S i c a"?
E-j * 5 jc ti
£C «  «
§ h§3
S «§£"
9 |SH

||
• II
T"*^ II
/"X3
\S
1
00 ^JQ
0) ^
U -D £
1 «3
H







ft o rs r-
m rji O)





O O O O
O O O O
CM CO r-t kO
(N ro OJ











O O O O
O O O O
cri o us r—
C\ CO <7\ rH
M (N r-l













m in r— in
*— * n r^> i-H







O O O O
o o o o
o o o o
in o o in
o co , en o
"S" n t-i f*>
**> r» ^* o%
ft ft









f r*> *f in







o
o
in





0
o
«
0\











o
o
(N
•.
o\
r»








.
•o

C ft 3
•HO. 0
fl -H C U
0 JJ 0 &
UJ fl -H JJ
fl 3 JJ
O) E fl Ol
fl C
U - 3 -H
0 *D O W
U * ft CO
— jl drf ^ri
O JJ " ^
E l< O ft
0
rH ft M 0
fl X -H C
JJ 
U .C  0
3 <^
XI JJ fl C
fl 0
0) £ C -H
C JJ -H JJ
•5 a
fi >" w e
0 0 JJ 3
W C W
fl JJ fl C
01 C fi O
0) ft 0
k 0
0 *< J«! U
., (J^ p_^ .^
O ft 3 JJ
E -Q w
0  JJ
3 JJ JJ
0 -H W C •
^4 ft 0 0 CO
JS JJ O CK
JJ C « ^ CM
0 £ 0 fi
0 -H -H Q,
C ft X C
•H fl O 00 -H
fl -H kl ft
O fl ft (0
ce ft e x
0 O» -H fl
JJ (^ 0 JJ 3
f» o> W J3
CO fl O
a, C JJ 0
CT> •« C C C
fl 0 « -H
CO fl
«w 0 fc C 0
0 -H 0 -H (0
JJ ft «
0 ft w 0>
jj E w JJ
fl 3 x; rH M
g 01 JJ 3 0
•H C 10 JJ
jj O >i 0 0
U 0 J3 fc E
u
«










.
01
4J
u
0
ft
14
•rH
fl
JJ
fl

u
JJ
c
fl
r-*
ft
X.
r-4
Q

(11
c
•rH
r-4
O
U]
fl
O)
c
0
••H
Aj
0
-1-*
J,
fl

i™H
rH
fl

(fl
J_l
C
0)
w
0
u
ft
0
u
(/]
JJ
g
fl
i-H
ft
^
rH
3
J3
M
o
M-l

rH

jj
c
nj
ft
ft
i-l
0
•o
1
jf
8-52

-------




00
^3^
Q^
to
2
M
oi
E-

^M
v-3
D
CQ
OS
O


-~
r ^
"

<3
^
J
O<
to
Q
O
JS

CL.
O

2
0
E"
CO
M
OS
EH
en
M
Q

Q
U
EH
2
M
EH
U
CN
(N
OO

U

EH






u
«j 3

01 0
AJ ^

U _.
U ^J
01 1
XJ ^^
3 W
a w
-C 4)
Dl AJ
§ M
M
rj i4j
£-> o

73 01
0) C
•u O
E rH
•
•H U AJ
X 0) "H
IH g -rt
a. 3 o
a 2 id
< h


*J O
0) 0> «
Dl .Q 'H
 4J ^*
Oi O
EH



Average Model
Plant Annual
Throughput
(liters)



c
n

CU

**4
m
a




01 r- rH 0







o o o o
o o o o
C? ^O CO ^*
» •. * *
LT1 C^" C^ ^ "
rH r-t










O O O O
rH n CO O
•"3" (N (N rH








o n r~ o







0000
o o o o
o o o o
o o o o
o o o o

C 01
0 4J

(V -H
(A AJ
id W
J3 0)
73 W
41 rH
AJ id
id c
.H 'g

(fi 01
0) ^

(A 
-------

(T,
CO
?
o
EH
s

oi
o
Cl4

W

2
«t
O.
_
u
Q
O
&

r
O
o

f
a
ffl
M
OS
£_,
U)
i— 1
Q

Q
TIMATE

a


ro
1

CO

u

CQ

g

i "° i
C O ^ ^ Dl
n *-" £ ^ 3
iL rr\ O '""* l Q*
0) re ^"* *^ .n
n. "I £
a- M H


•O , 01
 re r?
cu

lt J
i O I
X 0) — i »
O iJ Ll rH W
u re oj -H o>
a e J3 u — «
Q.TH e re *j
< 3 b.


1 * r- C ^,
X 01 u •§, O ^J

l_to
01 2 3 u
U 0) " O 3
1-1 D>£ 1-1 a
01 W ^ JS
a. E-I


i _
^. .^ _,.
S^ = 33^0!
u >o « 2 2 3 *
21 "^ * C w Q* i

^ rH -_-
*^




rH U
01 C
*83
S a.




»T (N rH CO T O
rH rH (N • CD O t—
rH rH


0 I O 0 O O
O i O O O O
r"- r- rH rH C^
r-4 rH rH




O i O • O O O
0 ( 0 0 O O
fN UD «T O r-l
C7\ CD [ — ^3* r-l
03
rH



O rH 1/1 O\ UH

O 1 ^T ro rH O
Ov i







rH O U"t 1/1 CO rH
V C*> CM  f> CN O


0 0 0 0 O 0
<•£> 0 O O O O
co o-j r— oo oo

Cn O CO ^ CD
t-t


V r-t rH OJ •»


_ O O O
O O O o O O
0 0 0 o 0 0

u o
O
J= 01
u as
'S S
01 Q
C 1C
0 >

XJ **-
re O
u
0) LJ
CJ
Oi re
O C
^> ^_

QJ O
01 —
E
HI O
c c
—i O
— i O
O Ul
W
re tt)
01 £
C
. o
-—) U
ftj O
• r^ Q,
> OJ

O CO
o ' r-
vD 00 CT
* ^ i-H
c-t 0\
.
re >H .
u O u
re **H c

01 Oi
3 *-* 01
M re rH
c e u
(1) -rH u
U *-• "H
01 J
C (JJ
i-i 0) Q
U-l )-l
re <
T!
0) 01 (/)
Q, IH .-i
O 0)
rH J-* re
01 **"* &J
> —> ra
Oi c -o
O C
C -
•a o u
01
01 ••-• 01
•H 3
S, 0. 0>
iJ X 4J
01 re
o 2 >
iJ O -rH
VH u
c .c a
o <•>

u •—( o
*rH *0
•D 3 01
•0 C 0
(TJ C U
m 3
c o
I-H W
a A

o
*H tJ

O T3
Li QJ Q
>, U l-l
re re
a E 01
— « 0)
L, u OI
O M • ra
ItH 01 W t->
— C C
p* O 01
CO i-t '*H O
en 3 «-> u
.C *J Pj
IH O< M
O 3 U
Li J-* Q«
re jc re jc
iJ u > O>
re —"3
73 C ^ O
01 -i J=
(U • u -C u
^-* ^— re Ot
rrj rH i_j 3 u
Ul O (.1 O C
r-c L, „,
01 a' -C •-<
3 0) u *-> a
0) 0 —
C C > Ol rH
UJ 0) Li C QJ
fj Li 0) -rH TJ
01 01 01 O
e «-" 01 £
o OJ —i re
LI a: re Qi 01
OJ o 0» .IH

d; 01 "H —I
u — ai -i a
re c o a
— i c — i oi re
3 o i_i re
U -H CJ O) >
rH iJ re .fl
re 3 L. x-i
O J2 1-1 O T3

C IH 3 i-> jj
o i-i 01 c re
.-. U) 3 ^
<-> — < >. O 3
3 T3 XI E U
fj fO r*<
•rH C TJ «
Li O 01 0) u
JJ '"H J_l ^
0) iJ nj JJ 0)
._ re £ .,H
•0 u — >i
yi [ > _'» ^j
^
a->
0)
i
c
h-

c
c
i.
 .-H -rH
0 ^ iJ r-< JJ
OJ CO Ql OJ 3
OJ 01 c E TJ Q-
O) O 3 O -C
m u .ri U) g ov
Ll "H *J C 3
0) rH (TJ O >, O
> & fcl U ^ Ll
re 3 01 J=
a. o> u <->
C O C 3
rrj < .r< --H Q, C
0. ^H rH j; O
g u .a o 01 -^
O 3 W 3 u
IH 01 Q| (TJ O*8
Iu-i 3 OI Li JJ
o j-: j; w

4J 01 O *J rH U
re -H u o IB •»<
rH a ^ £ u >
3 J-> O ^
0 (TJ rH jj fl)
rH jj ra 01
(0 T3 3 4J lu
CJ C Q,O O O
re c~. w -H
C CT C -J
o • s »H o J3
•HO) O O -H 3
jj c t, ,j a
3 _0 £ £ 3 ^

W (0 ^j!) lj ^ i
JJ J-> (fl 01 i_> O
01 o) j_> XI M u
'»H O •'H
a &- Q
o -O »
8-54

-------
     The throughput-derived estimate of the number of public
stations in 1998 is approximately 175,000. while the double-log
regression estimate is approximately 145,000 public stations.
There is a significant difference between the two projections.
The 175,000 throughput-derived figure is used in this analysis
because this represents a conservative estimate of the public
service station population.67  Use of this estimate will
therefore tend to overestimate the costs of the regulation.

     Over the past two decades, the percentage of terminal
throughput that passes through bulk plants has declined
significantly (see Figure 8-6).  Because this trend is expected
to continue into the near future,  the percentage of terminal
throughput passing through bulk plants in 1998 is estimated
using the Census-derived distribution of model plants and the
number of facilities estimated by the double-log regression of
the number of bulk plants.  A percentage of the terminal
throughput figure was selected that most closely approximated
the 9,227 bulk plants calculated from the regression (an 18
percentage throughput figure yields approximately 9,400 bulk
plants in 1998) .

     Twenty railcar-loading terminals are estimated for the
baseline year based on estimated 1998 throughput.  Applying 1983
data representing the percentage of total shipments from SIC
5171 that go by rail (1.4 percent)68 to total estimated terminal
throughput in 1998 (441.9 billion liters), results in an
estimate of 6.2 billion liters of gasoline moved by rail in the
baseline year.  The number of railcar-loading terminals was then
estimated based on one identified railcar model plant.6^
Throughput for that plant was divided by 1998 estimated total
railcar throughput to estimate 20 railcar loading terminals in
1998.  Because only one model railcar plant represents  this
small sector of the gasoline marketing system,  a model  plant
distribution is not required.
                              8-55

-------
     Delivery of gasoline in 1998 is expected to take place
using an estimated 87,700 tank trucks.  (Of these, 81,300 trucks
deliver to bulk terminals and motor gasoline bulk plants; only
6,400 trucks deliver aviation gasoline.  The 81,300 estimate is
derived from a two-stage process.  First, data available on the
number of gasoline tank trucks (not including aviation gasoline
trucks used at airports) from a 1979 report70 were updated to
1987 using the 1977 to 1987 ratio of total "liquid/gas tank
trucks" available from the Bureau of the Census
(236,000:213,000).71  This calculation results in an estimated
76,400 tank trucks used in gasoline service in 1987.  Next, the
ratio of 1987 gasoline tank trucks to total 1987 gasoline
consumption was calculated and applied to 1998 estimated total
gasoline consumption.  This method results in an estimated
81,300 tank trucks used in gasoline delivery in 1998.

     The distribution of these 81,300 tank trucks between
private and common carriers and between bulk terminals and bulk
plants is discussed in Section 5.1.4.  The 1979 report
characterizing gasoline tank trucks does not account for trucks
used by airports for delivery of aviation gasoline into
airplanes.  An additional 6,400 tank trucks are estimated to
deliver aviation gasoline into planes at airports based on the
1990 number of aviation gasoline bulk plants (3,200)72 and an
estimate of two tank trucks per aviation bulk plant.7^

     In addition to tank trucks owned by terminals and bulk
plants, for-hire, or common carrier trucking companies transport
gasoline.  Section 5.1.4 discusses how the total number of for-
hire tank trucks transporting gasoline in 1998 is estimated.   A
previously developed for-hire model firm characterization was
used to develop the distribution of for-hire trucks between
various size trucking firms.7^  This distribution provides a
relationship between the number of trucks owned by firms and the
number of people employed by those firms.  The 1987 Census of
                              8-56

-------
Wholesale Trade* contains firm-level data characterizing
employment and sales.  The employment data from the Census for
SIC 5172--Petroleum and Petroleum Wholesalers, except Bulk
Stations and Terminals were matched with the data from the
previously developed characterization to provide distributions
of the number of for-hire gasoline trucking firms with
particular fleet sizes and the distribution of throughput by
truck fleet size.  For-hire trucks used at terminals were
estimated using Census data for "manufacturer sales branches,"
and data for "merchant wholesalers" were used to characterize
trucks at bulk plants.  The estimated distribution of for-hire
gasoline trucking firms for 1998 is provided in Table 8-24.

     The number of pipeline pumping stations in 1998 is
estimated at 1,990.  This estimate is derived, from total
products pipeline mileage (150,000)75 and an estimate that a
pumping station occurs about every 40 miles.7^  The number of
pipeline break-out stations (270,  of which 150 are located at
points where the diameter of the pipe changes and 120 are
located at pipeline branching areas) are estimated from a map
displaying U.S. petroleum products' pipelines.77  Because no
data were available to trend these estimates to 1998,  the number
of these facilities is held constant between 1990 and 1998.  For
economic impact analysis purposes,  pipeline facility throughput
was apportioned across model plants based on the number of pipes
for pumping stations and the number of storage tank "equivalent
dedicated pumps" for break-out stations (see Tables 5-1 and
5-2) .

     Tables 8-25 and 8-26 display the final model plant
throughput and model plant distributions estimated for each
gasoline distribution entity in 1998.

8.2.5  New. Replacement.  and Existing  Capacity
     The baseline conditions imply that changes in the industry
sectors' capacity will occur over the  period 1993-1998;  industry
                              8-57

-------









^*
p^

*
rH
^
00
fN
CO


M
£-i
co
H
EH
O
*^*
(X

re
o
A?

M
U>

O
y*
M
x
o
^)
CS
E-
U
s
M
O
CO

O
to
oi
M
X
1
ce
o
Cl4
00
(^
en


CM
1
CO
W
2
*"















.* t;
r-l 5
3 *
*~ p-4












minal
>-l

£-













































O m
O fH
VO •fl'
» . rH
oo en
^^ *^1 CN










in
O r^
O Tf
j^T

VO VO O

















u
o
c
«J --.

« 01
•H -H 0
J3 i-H -H
>4 C <0
o o a
h -H
<4J rH U
0 -H 3
« "* ^
^ JJ JJ
111
Dl Dl
-« 3 3
COO
JJ M M
& £ g

«*> <*>
o o in o o f^ o o
rH| . m O rH i-l O tfl
(OJ OO ^ C"~ 
OOOVO O\^>
OJ CNin
o o a," vo 5 JJ o g 3 o
(N m >X> m
m



^A ^A

•H
t*<
WT- VT-
^^ voroo> co^oo
__,„_._ inftn mr~vo
 rH C^* in OJ
M S )H f^voro VfOin
vo vo ro i-i








•Q
.
18 U
£ rH .
^ K3 «
•H E C W
Du S — t .U W 0)
•H e c ^ .*
rHCl^Cfl U U
0) 0) rH 3 3

2 *O M M *"^ ^^
o o o « g w e
0) o> -n 0) -H
ft r4 C C 4J f*] 4J ^
IV 0 O -H -H
tO ft -H -H rH rH rH rH
Jrf 4J 1^* Q) Q)
UM33 C13 CTJ
3 o) ja ja oo oo
M 0) -H -H -H X -H S
t« >i M ^1 iH i-l
04-IU rH *4 0) rH H-l W
J^ rH 05 W -rl O 6 -HOE
C ft -rt -H g U _E G
t« W 0) CU .J « fa
•n
0
s
JJ JJ JJ A _ JJ .Q
 lH D> ^
W«33 23rH$ ^S1"*^
H (3






.
W
01
£
u

JU
u
JQ W
in oi
01 rH
rH (0
• « to
 (Q 01
S E
OJ (^ f^
•O p. r~
0 rH rH
S in in

COO
OD M rH
0) M M
3
u >-i i-i
0> O 0
T3 0) 0>
0> N N
T3 -H -H
•»« W W
•rl iJ iJ
n c c
0> 0>
j>. E e
coo
4) rH r-t
> a. a
01 0)
•2 U I-l
Q) 0)
tfi a. a.
0)
oi in u
>i 01 0)
O rH rH
rH «J HJ
ft n u
§ E E
LJ LJ
rH *M U-I
_
i> 0 0
o c e
rH 0 0
•H -H
033
>1 -H .H
U (H »-l
Q JJ 4J
Di n w
nt >»4 >»4
Vy 'rl "M
4J irj irj
10
U § §
(0
3 T3 TJ
U) 41 41
C w n
4) 0 «
O CO 0)
•a ja u
8-58

-------











U
d.
EH
EH
M
j
(— {
u
f£
iL,
03
D
CU
X
O
^
o
rn
EH
EH
3
_3
dl
J

Q
O
J2

i/l
i
03
U
00
EH










C
(TJ
,_4
a,
01
T)
S
J3
T3
0
C
0
(0
0
a
0)
0)
en
c
«
K
AJ
3
a
x:
CT
o
£




AJ
C
rH
3
0 .C
o ^
s o
(1) *H
01 "^
,
&
•^
10
0
rH
rH
5







I
W
V
AJ
•iH
J




>,
Ifj
°
W
0
rH
fl
O


>,
(Tj
Q
U)

0
-H



jj
g
^H
a
0
T3
s
0
a
^
^
AJ
•H
•H
U

O
0
o
il
ii
0
o
CM




O
O
O
**
0 in
o
°, •-"
^ o
r~ —
r-
^




o o
o o
0 0
0 0
o in
rH (N



o o
o o
o o
0 O
oo in
m en









Jj rH (N
1 * «
•H * «
^ o- a,
* iH rH
=•00
w *O *O
1 * *

o
o
0
11
M"
0
o
•**




o
o
0
i1-
i rj
o ^]
JT *
TjT ^
in




o o
0 0
0 0
o o
0 0
in o
rH



0 0
o o
0 0
o o
O 0
CF\ CO









p^ *fl*
AJ AJ
C C
O< O<
rH rH
0 0

Ji £




0
in
VO
V






in
*
VO
I
0
00
CM
O
n
o
o
in
(N
^

o
o
m




AJ
c
«J
a
0
1


0
o
0
o
 c
AJ O










0 O
O 0
O 0
1 1
* o o
Ty O O
o o
in* o*

0 0
o o
o o
o
o co in
O ft O\
. 1 1
a\ o o
rH O O
V 0 0
OV CO
r^ ro
o o o
o o o
o o o

£ 1 1 1
0) * * *


o
o
o
«
0
in
i
o
o
°
in
o
o
0
*
CO
1
0
o
o
in

o
o
0
in"
n

o
o
o
m
rH


AJ
C
*
a.
0
I


o
°
o
o
fH

'
5~
o
o
in
0
o
o

?•
o
0
o
Cf>
^
o
o
o
in
VO

o
0
o
VO
(M

in
AJ
C
«
Oi
0
I




o
o
o
*
~r
^
A





g
5


o
o
o
in
CO
^

O
O
o
o
o
c-

VO
AJ
C
XJ
a.
0
i
•n
0
3
C
•H
cont
*—»































8-59

-------







§
D
2
i— I
EH
E
o

u
a,
>-.
EH
s*
JJ
C
<0
CU
I—I
a)
•o
0
s
£
jj.
•o
0)
.L)
c
0)
(0
(U
)-l
a
fit
w
a:
0)
01
fl
a:
1 *•*
£ II a
i— i II -C
3 *
Ml 7S
u 2
< £
Ct, EH

S_,
r^1
CQ

THROUGHPUT
Model Plant
oughput
g r SL5
j
Cu
•J
U
3
0
S


in
«N
CO
w
J
CQ
<
EH






STP
1-1
2
<





§«
W «<• 1
.-4 C H
-.2 S
•H '•J «
• H — >*
*3
\«/







c ^
° 2 j-
-s|
•^ .^ ^
2 j








Million
Gallons/
Year

Million
Liters/
Year






AJ
*
ffl *
O







C v.
•2 " u
-2|
•|H *r« ^
a j








Billion
Gallons/
Year

Billion
Liters/
Year



















I
1
•

iii it
iii ii









i"ii ii
iii ii


.
JS
JJ
c
"^
U)
o
r^
I—I
«J
cn
o
o
•>»
(N
£
JJ
C
O
\
m
tj
d)
i_j
•H
r- 1

1 o



in i-H \o oo in
rH fO »a« O 00
rH > o ** vo
n r^ »--( \& 1/1
^ CO n rH ^»
rH r-l rH





• •
• c
C I
"• id
^ jj
K U5
JJ
M jj
i 9
, iH CJ m ^ r-l  '" a> a)
(U *D *O *O 1« *0 'O
S £ £ £ 1 £ £
<=M CM
o
(N
CM
hroughput of
jj
average
/I)
IV
«3
X
(0
01
•r<
JJ
•f-l
rH
-H
0
fl
<4-l

V
c
•H
rH
O
M
«J
D)
O
•H
JJ

-------












_
g
<
r-J
o,
u
Q
i

^*
CQ
CO
W

M
EH
M
%J
n
o

r*

o
OS
u
CQ
s
3
B
VJ3
l~^J
1
oo
W
(Q
EH





























w
VD rHrOOOVOr*
rH rOONCM VOOinrHr--
rH 00 rH r- 00
CO (N rH ro
1 III O O O 1 O
1 III 00010
» n o»
1 1 O O O O O 1 O
1 1 O O O O O 1 O
VO VO rH O rH rH
VO ^ CM VO
ro ro n


O 1 O O O O O 1 O
O 1 O O O O O 1 O
rH rH rH 00 TJ« Tj" OO
fN c>j o r** ro o
in "31 in




O 1 O 0 O O O 1 O
00 1 O O O O O 1 O
 ) CO (N VO VO •«•
ro ro o\ O\ rH *-H
00 00 O\
rH rH rH



A
•-;
cc

0 « t
S 5
^ « a « 3 i
(M o e o c ° -S
._ » -»4 ~ -H CO ^
«1 .. •? |H 1! <-* <0 m
c «iOO IB 2 5 „ 0 c
•E ^cS #3«.HC - i
E goo ww>^2 « "
1 J^K'^ ttJr*'C'3'IjrH *•*
^ *o«j .^ 0 cu ai  r\
m"






















o
in
(N
*.
rH




O O
in (N
f^ rH
rH


0 0
in in
OJ rH
rH




a
c
c
J.
1C
JJ
C/l J.
e Pumpipg
e Break-oi
c c
• H -r-
rH r-1
oi a
C C
• H -H
CM a
                   •- c
8-61

-------
growth implies that new capacity and new facilities will be
constructed.  At the same time, existing facilities will close
as their equipment wears out and becomes obsolete.  EPA has
estimated the number of new, replacement, and existing
facilities for 1998 based on industry sector growth, facility
trends, and estimated equipment life.*59  A new facility is one
that has been built to handle the increased output required of
the industry over the impact period.  A replacement facility is
one that has been built or rebuilt during the period to replace
worn-out or obsolete equipment.  An existing facility is one
that was operating in 1993 and continues to operate in 1998.
The resulting estimates are shown in Table 8-27.   These
estimates provide a context for evaluating the economic impacts
discussed in Section 8.3.
                              8-62

-------
   TABLE  8-27.  ESTIMATED NUMBER OF NEW CAPACITY,  REPLACEMENT
                CAPACITY,  AND EXISTING FACILITIES
Sector
Pipeline Break-out
Stations
Pipeline Pumping
Stations
Bulk Terminals
(loading racks)
Bulk Terminals
(storage tanks)
Bulk Terminal Trucks
Bulk Plants
(loading racks)
Bulk Plants
(storage tanks)
Bulk Plant Trucks
Service Stations
New
Capacity
10

80

40

40

1,690
0

0

0
9,540
Replacement
Capacity
30

960

490

110

14,070
3,580

570

12,440
40,740
Existing
230

960

490

880

28,140
9,020

12,030

31,360
337,450
Total
270

1,990

1,020

1,020

43,900
12,600

12,600

43,800
387,730
Note:  Figures may not add due to rounding.
                                8-63

-------
8.3  ESTIMATION OF ECONOMIC AND FINANCIAL IMPACTS
     Gasoline distribution in the United States represents  a
vertically integrated system that consists of several individual
markets.  Each market is affected by the supply and demand
forces of interlinked markets.  For example, refined gasoline
combined with pipeline services provides "delivered gasoline" to
the delivered gasoline market.

     The cost of the additional equipment and services at
several points in the distribution chain, creates incentives for
producers and consumers in related markets to simultaneously
adjust their production and consumption of gasoline marketing
services.  To evaluate the economic impacts requires an economic
model that can estimate the price and quantity changes on all
the distribution markets affected directly or indirectly by the
regulation.

8.3.1     Market Interaction Model Summary

     Figure  8-9 illustrates the key markets modeled to represent
the gasoline distribution system.  These particular markets are
key for two  reasons:  they represent the different stages of the
gasoline marketing system, and they reflect production
activities that were considered for direct regulation during
standard development.  Markets in the model were also chosen to
represent the major sectors involved in the marketing of
gasoline in  the U.S.  The market interaction model assumes that
all refinery gasoline moves by pipeline.  This assumption may
overstate market impacts because it prohibits substitution of
other possible modes of transportation.  Combining delivered
gasoline and terminal equipment produces terminal storage
services.  Terminal storage services can, in turn,  either be
combined with terminal transportation services to provide
retail-commercial gasoline for "large volume" (large throughput)
outlets or gasoline for storage in bulk plants.   The gasoline
                              8-64

-------
                                                        JJ
                                                         u

                                                        -3
                                                         o
                                                         i-l
                                                        04

                                                        •o
                                                        C
                                                        u
                                                        o
                                                        JJ
                                                        o
                                                        C
                                                        0
                                                       JJ
                                                       3
                                                       0>
                                                       C
                                                      •H
                                                      1—I
                                                       o
                                                       W
                                                       «
                                                      o
                                                      I
                                                     00
   0 U)
   jj iq

   u
   01
   ¥-1
   o> uf
   ^  o
   0  Q
  JJ  C)

   0) "
  M
      0)
  J= i
  U JJ
  «
  01 v-i
     0
  tl

 -§.2
  C TJ
  3 C
     0)
  -.  C
 T3  -H
 a
    C
 C  0
•-<  -H
    JJ
 n  a
 u  -H
 o>  n
JJ  U
jj  M
ai  o>
J T3
                                                               s
8-65

-------
from bulk terminals to be stored at bulk plants can be combined
with bulk plant equipment to provide bulk plant storage
services.  Combining these services with bulk plant
transportation services provides retail-commercial gasoline for
small volume  (small throughput) outlets.

     These markets are represented mathematically as a system of
thirty six linear equations based on Hicks' and Muth's work on
specification of theoretically correct systems of demand and
supply equations in linear form.78-79  The coefficients of these
equations represent the responsiveness of key product or service
supply and demand schedules to shifts in the corresponding
demand and supply, respectively.  The variables of the model are
proportionate changes in equilibrium prices and quantities of
the markets modeled and the "right hand side" variables are the
proportionate changes in market supply associated with the
additional cost of meeting the requirements of the regulation.
By specifying the supply shifts associated with the regulations,
the model can be solved to find associated changes in price and
quantity in all markets represented by the model.  Applying
these changes to baseline levels of price and quantity provides
estimates of the market impacts of a proposed regulation.  A
detailed description of the model's structure and data is
provided in the Economic Impact Analysis report.67

     8.3.1.1  Estimation of Baseline Year Values and Model
Parameters.  Table 8-28 presents the estimated prices and
quantities for the baseline year of analysis.  As discussed in
Section 8.2, baseline estimates of prices and quantities are
forecasts and are subject to the usual forecasting
uncertainties.  Baseline year prices for each sector are
estimated from the projected average retail price of gasoline in
1998 in 1990 price terms ($0.357 per liter; see Section 8.2.2
for the derivation of this price).   Price margins for each
sector are estimated in Section 8.1.3.2 from industry sources.
                              8-66

-------
TABLE 8-28.  ESTIMATED BASELINE YEAR PRICES AND  QUANTITIES

Commodity

Refined Gasoline
Other Pipeline Inputs
Delivered Gasoline
Other Inputs at Terminals
Terminal Storage Services
Terminal Storage Services — Input to
Wholesale Gasoline from Terminal
Terminal Storage Services — Input to
Gasoline from Terminal to Bulk Plant
Transportation Services from the Terminal
Transportation Services from the Terminal--
Input to Wholesale Gasoline from Terminal
Transportation Services from the Terminal--
Input to Gasoline from Terminal to
Bulk Plant
Wholesale Gasoline from Terminal
Gasoline from Terminal to Bulk Plant
Other Inputs at Bulk Plants
Bulk Plant Storage Services
Transportation Services from the Bulk Plant
Wholesale Gasoline from the Bulk Plant
Other Low Volume Service Station Inputs
Low Volume Station Gasoline
Other High Volume Service Station Inputs
High Volume Station Gasoline
Quantity
(in billions
of liters)
441.8
441.8
441.8
441.8
441.8

362.3

79.5
441.8

362.3


79.5
' 362.3
79.5
79.5
79.5
79.5
79.5
79.5
79.5
362.3
362.3
Price
(in
S/liter)
0.322
0.008
0.330
0.005
0.335

0.335

0.335
0.007

0.007


0.007
0.342
0.342
0.005
0.347
0.007
0.354
0.013
0.367
0.013
0.355
Percentage of
Commodity Market Shares
Terminal Transportation Services--Input to
Wholesale Gasoline from Terminal
Terminal Transportation Services --Input to
Gasoline from Terminal to Bulk Plant
Terminal Storage Service- -Input to Wholesale
Gasoline from Terminal to Bulk Plant
Terminal Storage Service--Input to Gasoline
from Terminal to Bulk Plant
Total Volume









82

18

82

18
                           8-67

-------
These margins are subtracted from the retail price of gasoline
in 1998 (in 1990 dollars)  to compute the price of gasoline as it
leaves each sector.  Because federal and state gasoline taxes
are assessed at several different points in the system but
primarily at the refinery (typically for federal taxes), no
attempt was made to net taxes out with the other operating
margins.  Industry quantities for 1998 are estimated based on
total projected gasoline consumption, calculated in Section
8.2.1, and on historical trends in shares for each of the
industry sectors.  The model requires certain "elasticity"
parameters to represent the conditions and interrelationships in
the U.S. gasoline market.   For example, it is necessary to
develop an estimate of how responsive gasoline consumers are to
changes in the price of gasoline.  That is, for. a given price
change, what is the effect on the quantity of "gasoline consumed?
This relationship is called the own-price elasticity of demand.
The Economic Impact Analysis report presents the estimated
values for these parameters.^7  The parameter values were
selected to represent nonvolatile economic relationships.  For
example, it is assumed that producers are severely limited in
their ability to alter the mix of each product's inputs  (i.e.,
the elasticities of substitution are very small).

      8.3.1.2  Impacts of Regulatory Supply Shifts.  Shifts in
market supply due to the proposed regulations will initially
take place at three points in the gasoline distribution
industry.  These supply shifts are estimated based on the
control costs presented in Chapter 7 for regulatory alternatives
IV, IV Q, and IV M.  These are the regulatory alternatives
examined in this economic analysis because they control major
emission sources only.  The correct control costs to use depends
on the level of control consistent with the regulatory
alternative and the "marginal" facility being controlled.
                              8-68

-------
     The marginal  facility is that establishment whose
production costs  (including a "normal" profit) equal the price
that consumers are willing to pay for the last unit of gasoline
consumed.  Thus, the marginal facility provides the supply at
the point where the supply and demand schedules intersect. This
is depicted in Figure 8-10 for a hypothetical supply and demand
schedule for the market for Other Inputs at Terminals.  Before
regulation, the supply of these services is S° and the demand is
D°.  S° is a short run supply schedule (existing firms will
produce so long as they cover their fixed costs),  but it also
reflects the willingness of new firms to enter the market and
provide additional capacity at price pO.  The new firms comprise
the marginal firms in this market over this period.  If existing
firms attempted to raise the price higher than pO,  new firms
will enter the market and bid away the business of existing
firms.  Such market conditions are particularly likely in
"transition" industries characterized by technical or
institutional changes that affect the long run cost of
production.80  jn  this setting, then, the economic impact will
depend on the minimum control cost needed to meet the regulation
required of new firms.

     The imposition of the regulation will cause facilities'
production costs to rise equal to the additional cost of
complying with the regulation.  The market impact of the
regulation is depicted in Figure 8-10 by a new supply curve such
as S1.  Holding post-regulatory demand constant, the new price
and quantity for retail gasoline is determined by the
intersection of the post-regulatory supply function,  S1,  and the
demand function D^.  Given the perspective that the marginal
firm is best represented by new firms, this analysis bases the
relevant shift from S° to S1 in this analysis on the cost of
control at new facilities.   To emphasize that this  is likely to
be different from the control costs of existing facilities,  we
show the downward sloping segment of the new supply schedule as
                              8-69

-------
Price
  P0
Ql   Q0
                                           Quantity
 Figure 8-10.   Hypothetical Bulk Terminal  Services
                Other Inputs Market
                        8-70

-------
having a different slope from S° .  This highlights the fact that
the costs of regulation imposed on existing firms will vary with
such circumstances as facility size, initial level of control,
etc.  A corollary observation is that regulation will impose
distributional impacts  (net financial gains or losses) on firms
that are distinct from the market impacts identified in this
section of the analysis.

     8.3.1.3  Estimation of Marginal Facility Cost.  As
described in the industry profile, there are a wide variety of
plant sizes in the gasoline distribution industry.  Theory
indicates that this is due to the fact that demand for wholesale
and retail gasoline distribution varies considerably over space
and/or that the cost of production varies considerably with
distance.  In both cases, this means that the.markets for most
gasoline distribution services are  "local."  Trends toward
larger production facilities were identified in Section 8.1, but
most markets are still geographically circumscribed,  especially
in  the later stages of distribution.

     Selecting a supply shift for marginal bulk terminal
facilities in the market interaction model should therefore
reflect the diversity of local markets.  These range from larger
metropolitan markets served by large capacity facilities to
small rural markets served by small facilities.  Consequently,
EPA estimates the shift in the supply price of new bulk terminal
facilities as the weighted average of the cost of compliance of
all the relevant model plants.  The weights are based on the
amount of throughput attributed to each of the bulk terminal  .
model plant size categories in xhe baseline.

     Similarly, the supply shift in bulk terminal transportation
inputs due to required monthly truck leak testing and repair at
new plants is based on the weighted average of cost of these
tests to the different model plants.  The costs for each model
plant varied in proportion to the number of trucks that served
                              8-71

-------
that plant  (the weights included a 40 percent allowance for new
plants in non-attainment areas where Control Technology Guidance
already specified monthly leak testing of gasoline trucks).  The
supply shift  for pipeline breakout stations is also based on the
weighted average cost of monthly leak detection and repair at
new model plants.

     Table  8-29 describes each affected sector's marginal
facility and  the estimated increased cost per liter of
throughput  represented by that marginal facility.  The cost
shift for pipelines is negative because recovery credits
anticipated from leak reduction are greater than the cost of the
monthly inspection and repair.

     Costs  associated with required control at -existing plants
or in sectors where only existing plants are affected by the
regulation  are not included in this table because new plants are
marginal facilities  (see the discussion in Section 8.3.1.2).  As
discussed below, existing plant costs are reflected in the
economic welfare effects of the regulation but they are not
expected to have any significant influence on the market
impacts.

8.3.2     Market Adjustments

     The marginal  facility cost increases per liter of output
from Table  8-29 were entered into the model and solved for
estimated market changes in price and quantity.   The effects of
the supply  shifts for regulatory alternatives IV, IVQ,  and IVM
on all markets are shown in Table 8-30 and 8-30A.  This table
shows that  the estimated market impacts of the proposed
regulation  will be relatively small, because the additional
costs imposed are relatively small and buffered as they are
passed through the market in the form of price and quantity
changes.  These estimates apply to all the regulatory
                              8-72

-------
      TABLE  8-29.   REGULATORY ALTERNATIVES IV,  IVQ, AND IVM:
                  MARGINAL FACILITY CHARACTERISTICS
                                                      Cost
Facility      Marginal Facility                     Per Liter
Type                                                   ($)
Pipelines     Weighted average cost of
              leak detection and repair
              at new model plants               -9.77818  x 10~7a

Bulk          Weighted average cost of
Terminals     vacuum assist at new model
              plants.                            4.9047185 x 10"4

Bulk          Weighted average cost
Terminal      of leak detection and
Transpor-     repair at new model plants.          7.2,x 10"^
tation
                                          •

a For pipelines, the credits for detection  and repair are  greater than the
  costs resulting in a negative cost per liter.
                                  8-73

-------



2
i

^
M
Q
< Ct,
^O

O CO
>0
J

^> U^
L!J '""'
^ CQ
CO 2
> M —
M oi
EH CO CO
< CO EH
2 M M
Oi EH rJ
CO M
EH EH Qi
J 2 W
<"3 n
rQ^ ul*
"s_l 5- v r rt
>" Oi CO
oi >— K
o <

S 5 J
>-H EH O
3 to Q
0 D
W Q 2
^** ^ •"">
£ CO
O o W
CO^M

EH £-1 C£
u ca a-
ft] K^
^ cd •-
&u < en
WZ_J ,.,,
S 0£
w
Q CO EH
CO 2 Ht
EH t_j ."]
t rH 1— H
* O
M CO
EH <
w 8
W w
. ^
i o
00^
W
ca





a>
u
u £
•H p,
rH
0
01 >.
AJ
•U -rl
M AJ
a 5
<* 0
Oi


0)
o
OJ Wi
c cu
c
It
JC >i
CJ AJ
c
3
o


J 0)
u u
U -H
Lj
, w
(/) Ol
fi) ^
Dl
C n ?*t
0 1 AJ

0 0) ^
3 C
-a 3
a


i -5
01 u
± £


01
C w
r *
« "5
w c
m rj
m "
3
o











e
M
§
O
o) co o in in in
O4 O f> O f> fr>
ro o f» o n f>
oooooo



in in in in in o
^^ *"^ T^ ^^ ^H o^
•q- •q- 5 •* •» rt



O tH O 00 rH rH
O O O O\ O O
1 1 1





oooooo
1 1 1 1 1 1



oooooo
oooooo
oooooo
oooooo
1 1 1




m n it fi n o
oooooo
1 1 1 1 1 1



rs co o in in in
m o m o ro ft
oooooo


1

00 00 CD CO CO f*>
,H rH rH rH rH «
^ 



O
tt
jj
1
• H
a i
«H m m a
rH rH 0) •
O 0 U O
V G *^ ^4
(d ^ «^ ^ fct
Ll ft C LJ U
** w B W M
3 M « •
a*> « w 01
C • fr*
O I**! el di A
c -n AJ D> en
rH C • rl (4
0 -H ps m o o
(0 rH 4J ^} ^>

o
1



r-
o
o
o




CO
,_,
^}l
^*




«
•C
§
M
IM
0)
0)
*rt
• «
5fi
a -n

fM US
Ol XI
s|
£
rH
•i-l
o> r** o*j CN in IN r^ ^J1 f*> r^ ro 1/1 u
O O TJ ^Ji O«»O Ul rH VOrHm
o o ro f*> Of^O ro o rto^fo >,
O O O O OOO O O OOO • -H
0) jj
O) C
C 0
o in o in in in in in in moo JS tr
rs) cr\  « u 0
•rl
>H 0)
a u
0 0 r-l rH OlHrH i-H O rHOtH'Q^
O O O O OOO O O OOO v>ri
II 1 1 ,, c
O *rl
0
i-l 0)
0)
0 Ol
	 	 s*
o o o o ooo o o ooo 2-i-
1 • 1 1 1 III 1 1 1 1 1 •£ u
01
rl U
01 01
o o o o ooo-o o ooo AJJ
o o o o ooo o o ooo 0
o o o o ooo. o o ooo Mw w
O O O O OOO O O OOO O>O AJ
II 1 1 C AJ -rt
•rl U -H
•O 01 -^
C U v0
rl f*J
m o m o ooo o o ororo g^ .
0 0 0 0 OOO 0 0 000 ^ '||J °.
i i i i 111 i i i ' • o S
AJ 0) M
)H C
« o
3 > oj inr»t> *» fn r>fnin>t3 -H
O O •» ^J> O^O in rH VOrHin "O fl
o o ro m orto f*> o fioo Oc CTI
O O O O OOO O O OOO 01 O
NO) 0
f— 4 • •
01 0 W r-l
0 in ai 
m in ro in ininin in in mro«*> O.3a>
Q, 0) AJ
0 T>> CJl -H
C rH
M >i 01 0
it AJ 0 AJ .c in
«MrH 0) ^ C 3 60UCO
0)0»0 jSrHO Oi UXf-
OOC.* AJ3-H C «-HO>-
• H -rl -rl rH CQ AJ rH 0) "O rt
> > e 3 £ "J » a> c AJ
U rl C 0) (A«O 0> AJ • C a C-HOIII
« 0> 0) Aj«in£ C/l a O -H 0 ^
MMW EH 0 COM-l AJ ^l-HfHjSTJriC
0 AJ  0 M 0 U C0
000rH^~-Hr-l> 01 Olflrl-H
AJ w AJ a. o> AJ g 3 » M « w a o o o, -o rH
u « ri c c C en 01 0) c o -H ^ _wa)
QrHOr<-r««« *W -H g ^>*i ®«AJ»
Q, O Q.rH rH rH EH AJ M -H E AJ (H « CTI -H U 0)
MJCU3OCL 0OC O 3 0 0> *J 0 AJ 4) AJ
C S C « W S **OAJW -• 4JC001 Aj.HrH0
rHMOr^M Aj'lJSo > ^O)* So«M

10 t* 1J ft AJ 3 *W 3 A) lO ^ ^ 6 S U E W C
c o) ja ad^ruv z m B 3 3 ME o
•rlrHAJrHAJrHl4)AJC0U --4 AJ O AJ 3 rH rH 4) O 0) -rl
E03030CCCHHOJ^0CiJ3r-IOO fcOjaw
CcOiCQ.no-H0 o, a r-i w 0 a p ?> > n
W-HC-HCBCrHrHtl BJ30»rHrlC>_— " W
ttSrHSr-lrH— OO.«JXC«rHOUfl»rHX^ 9 >
E C o w rC •-! 0 o £• > o>JJ AJ c
0> Q) -C 0 AjflU rC AJ X'-CjTSiS 0
EH gn S O O«fn S O J K « Z CJ
8-74

-------
2
 i
 .0
M O

   M

 » rJ

> r-

M M

   CQ
   coQ
DJ S
   o
Pi

O
U

Du

fc
Cd
   U
M O
EH CO
Cd
   W
 i  O
CO
m
«<
EH
u
0
a.

TJ U
W rH
41 O
01 Ou
C
O
00
CO
CM
0)
C
-H
rH
4)

pa
0)
u
>
c
0)
u
a.
>
AJ
C
It
o
4)
U
a
>
c

AJ
c
COMMODITY Qu
 vo t~- r» o
VO VJO «O <43 VO in rH
0 rH O 00 rH rH H
o o o 
Terminal Transportation Services -
Input to Wholesale Gas
0
o
0
rH
fN
O
o
o
1
o
o
o
o
o
o
1
o
o
o
rH
tN
Terminal Transportation Services -
Input to Bulk Plants
n
 t-

r* c\ o\
Low Volume Station Gasoline
High Volume Service Station Inputs
High Volume Station Gasoline
o>
JJ.
rH
• H
AJ
• -rt
41 JJ
01 C
C (C
«J 3
JC CT
U
•o
o> e
u fl
•H
U 01
a u
• r4
0 0.
Jt C
O -n
rH M
01
(0 CT
c
C (0
« j:
.C U
4)
01 4>
AJ 3
0
V-l U)
Ol 0
C AJ
•H U
•0 (j)
C IA
3
0 -H
l-l -r4
lrt
0 AJ
JJ 0>
u
01
3 CM
•a
•o
o c
1-1 «
0)
N 4>
W (0
 .C U
a 3 "H
a r-
IS V &
>, Ol 01
§JJ O)

O .*
01 jQ U
O ITJ
•H c e
t-l -H
a c
01 -H
a a>
O* **H *O

AJ -H AJ
C TJ U
a> o o)
O g rH

0) O 0)
3< CJ M
„
01
AJ
O
z





























u
0>
JJ
— t
\
vo
OJ
«
o


II

c
0

rH
<8
D>

o
o
•


»
Ul
^,
(1)

•H
(— *
in
00
r-

rn

II

0
rH
rH
HJ
0>

rH


• •
4)
JJ
<0

C
0
•H
ft
0)

£
0
0
                                             8-75

-------
alternatives (IV, IVQ,  and IVM)  since differences among them
only affect controls required of existing plants.

     The biggest price change will occur in the cost of other
inputs to bulk terminal storage (9.8 percent).  Since these
other inputs constitute only a small share of costs, however,
bulk terminal storage services are estimated to increase in
price by only one tenth of a percent.  While the rounding
convention of the table obscures some differences in the change
in quantity estimated for the proposed regulation, these are all
in the neighborhood of one tenth of one percent for each
industry sector.  This amounts to a reduction in consumption of
roughly 300 million liters of gasoline per year.  Thus,  while
the relative changes in gasoline distribution markets are
estimated to be small,  the market is so large that some of the
absolute market effects are non-trivial.

8.3.3  Employment  Imnacts.
     If percentage changes in output due to the regulation are
assumed to be perfectly reflected in percentage changes in
employment, roughly 1,100 jobs will be lost from estimated
baseline employment in the gasoline marketing sectors considered
here.  These results are put into perspective in Table 8-31.
Nearly 80 percent of the jobs lost will be in the service
station sectors due to the reduction in gasoline consumption
occasioned by the rise in the retail price of gasoline.   These
jobs, however,  constitute only five one-hundredths of a  percent
of baseline employment in the low volume service station sector
and seven one-hundredths of a percent in the high volume service
station sector.  These job losses are also a very small
percentage of the baseline job increases projected for most of
these sectors in the five year period following proposal action,
1993-1998: just under 3 percent of increased employment  in the
                              8-76

-------














CO
EH
U
<
cu
S
Ht
EH
2
U
2
>H
O
J
cu
S
03
Q
Cd
EH
M
EH
CD
U



i-H
CO
1
oo
w
r4
PQ
£











in
C
o
•tH
JJ
o
•0
01
C£
JJ
c
0)
£
o
rH
I





i
•H _
•H§
n

"£-0
° 01
_- Jj
** 0
» o-
0


rH
0 0
JJ JJ
O C
JJ 01
"" >1
0 0
rH
of a
£
tfl 01
0





jj en
C C
0) O
2 -iH
>i JJ
O U
—t 3
an
S 0)
c3 «

^
c
O)
e
->*
0
cl
s

13
JJ
0
H

M
o
4J
U
a>
0}
c
0
•iH
JJ
X)
.^
>-(
JJ
ta
•^
Q















r-
o
0







fN
r~








o
o
0
V0
o
rH




M
0)
•iH
M
Refine















t*
o
o







CM
rH








O
o
00
r-
rH






m

a\

rH








00
o
o







en
V£>








O
in
r~
rH
cn





n
rH
0
•H
U
0)
EH
J^
rH
S















r-
o
o







rH
in








o
*3"
(M
in
r-



Xt
c
inal Transportatio
«
e*
.*
rH
S



(J

1
1








in
o
o







co









o
o
in
^o
rH





n
jj
c
0
rH
CU
X
rH
m















m
o
o







oo









o
o
in
vo
*— t





t Transportation1-*
c
0
rH
0,
Jf
rH
S



in
o

(N








in
0
o








£.
O>
•H
«

























«*
rH
*-H
>-H






O
'Jl
c^
fNJ
n
vo
rH





Total
o>
X
JJ

U)
•H

0)
c
0
•H
JJ
0
rH
S
01
U
01
r"
«JM
JJ

>H
Q)
JJ
H -|
0
T3
C
0
01
t,
M
0
U-l
<3)
X!
>.
JJ
• H
rH
0
&
U
C
•H
01
r^
f
C
0
U
0>
en
0
jj
c
0)
U
}-l
01
a
0>
JJ
JJ
0
J=
s the assumption t
jj
u
0)
rH

e
•^
percentage change
ra
0
same
o
lo
           o
8-77

-------
high volume service station sector and just over 2 percent in
the low volume service station sector.

     For bulk terminals, the job losses constitute just  under
two percent of anticipated job growth.  With the exception of
the bulk plant sectors, where sixteen jobs are expected  to be
lost over the analysis period,  the projected job losses  due to
the regulation are more accurately interpreted as reductions in
job opportunities rather than terminations of existing jobs.

     Loss of jobs also imposes some displacement or transaction
costs on the economy.  An examination of these costs showed
that, in a statistical sense, workers would be willing to accept
wage reductions equivalent to roughly $57,000 for an increase in
job security equal to the statistical equivalent of one  job.81
Since most of the job reductions estimated here are changes in
job opportunities, rather than actual losses in jobs,  it is not
clear that the estimated job displacement costs apply to any but
the bulk plant and bulk plant transportation jobs.  For  these
two sectors, job displacement costs estimated by the imputed
value of job security are less than one million dollars.

8.3.4      Facility and Firm  Impacts

     8.3.4.1  Facility Closure Estimates.  Although the
reductions in quantity reflected in the market interaction model
results discussed in Section 8.3.2 are not large in percentage
terms, the scale of activity in the gasoline marketing industry
makes them noteworthy.  The quantity changes may reflect changes
in output of existing facilities, closure of facilities, or
both. Assuming in the extreme that all the quantity changes
occur as a result of closing existing facilities or never
opening new facilities, plant closure due to the regulation can
be estimated.  Further assuming that the smallest model plants
in each sector are most vulnerable to closure, this analysis
estimates the plant closures listed in Table 8-32.
                              8-78

-------
                 TABLE  8-32.   ESTIMATED FIRM IMPACTS

Distribution
Sector
Refineries
Pipelines
Bulk Terminals
Bulk Term. Transportation^.
Bulk Plants
Bulk Plant Transportation*3.
Low Vol. Service Station
High Vol. Service Station
Total
Facilities
1998


1020
12600
279650
108100
Potential
Plant
Closuresa
N/A
N/A
3
15
12
12
440
165
% Reduction
in new
facilities15


6.57
	 c
25.64
2.11
Total
647
Note: Potential plant closure  figures  are  not  applicable  for  refineries and
      pipelines because it  is  assumed  that these  types 'of  facilities do not
      close,  but rather reduce capacity  or capacity utilization or postpone
      addition of new capacity.

a Potential plant closures  are the absolute change in quantity of
  throughput divided by throughput of  the  smallest model  plant.

b Percentage reduction in new  facilities is facility closures
  percentage of anticipated facility growth.

c No growth anticipated for bulk plants.

d Assumed for-hire firm for Bulk Terminal  Transportation  and  captive for
  Bulk Plant Transportation because they have  the smallest throughput
  (this creates a worst-case scenario).
           as a
                                   8-79

-------
     The total estimated number of closures is 647.  of all
closures, more than 90 percent are in the service station
sector.  In this sector, 72 percent of closures are among Low
Volume Service Stations, while the remaining 28 percent are
among.High Volume Service Stations.  While the number of
facility closures among service stations is in the hundreds, it
should be kept in mind that the total number of stations in the
country is over 380,000 and that the number of facilities closed
constitutes less than one percent.  While there are 647 total
plant closures estimated across all sectors, the projected plant
closures due to the regulation are more accurately interpreted
as reductions in new facility openings rather than closures of
existing facilities.  Plant closures for refineries and
pipelines are not applicable because it is assumed that these
types of facilities do not close, but rather reduce capacity or
capacity utilization, or postpone the addition of new capacity.

      8.3.4.2  Firm  Impacts and Financial Health.  The EPA
includes estimates of firm-level financial impacts in many of
the economic impact analyses of its regulations.  Identification
of the firm-level impacts for the "gasoline distribution
industry" involves two aspects:  the size of the financial
impacts and whether these impacts threaten the existence of
firms in the industry.  Chapter 7 presents cost data at the
facility or establishment level using model plants for selected
regulatory options for the pipeline, bulk terminal, and bulk
terminal transportation sectors of the industry.

      These data show that the cost of all the regulatory
alternatives are relatively small when compared to current costs
of production or current prices per liter.  These data also show
that  small model plants will experience higher costs of control
per unit of throughput than large model plants.  These facility
or model plant costs can be combined with firm level
descriptions and financial information to provide estimates of
                              8-80

-------
the firm level financial impacts of the proposed regulations.
Such impact estimates are reported in the Economic  Impact

Analyses report.^


     Estimating firm  financial impact estimates  involved the

following sequence of activities:

     1. Characterize  "model  firms" based on available data on
         firm  size  and facility ownership in each industry
        sector.  This characterization  concluded with estimation
        of model firm sales.

     2. Construct  pro-forma  balance sheets  and income statements
        for model  firms based on Dun and Bradstreet financial
        ratios for each industry sector.  Three  sets of  ratios
        were  used, each set  representative  of firms in either
        above average, average, or below average  financial
        health.

     3. Compute compliance costs for each model  firm based on
        the control costs of facilities estimated  to be  owned by
        each  of the model firms and the cost of  capital  based on
        industry sector and  firm financial  health.

     4. Revise the model firms pro forma balance sheets  and
        income statements based upon the estimated  compliance
        costs for  firms.  Model firms with  below average
        financial  health were treated as  financing purchases
        out of cash reserves.

     5. Use the revised balance sheets and  income statements  to
        compute new financial ratios for model firms  and assess
        the impact of  the regulation on these ratios.  Ratios
        used  were  the  liquidity, activity,   leverage,  and
        profitability ratios.

     This financial analysis reported in the Economic  impact

Analysis report was conducted using the most stringent

regulatory alternative, Regulatory Alternative I,  as a basis for

estimating firm compliance  costs.   In addition,  the analysis

assumed that each model plant would have the highest possible .

control costs i.e., existing plants with the lowest initial

level of control.   Under these extreme  conditions,  small model

firms with below-average financial  health still has enough cash

in their pro-forma balance  sheet  to cover the cost  of control.
                              8-81

-------
At the same time, the financial ratios of model firms were
hardly affected by the compliance costs.

     No average or above average firms' ratios fell in  the  range
of the less financially healthy firms' ratios after the
regulation.  Regulatory alternatives IV, IV-Q, and IV-M are
substantially less stringent than Regulatory Alternative I and
would result in considerably lower control costs.  Consequently,
even firms in below average financial health are expected to be
able to cover the costs of complying with this regulation and
firms in average or better financial health will not suffer
serious financial affects.

8.3.5  Economic  Welfare Changes
     The results of the market impact model can be used to
improve estimates of the costs of the regulation so that they
more closely correspond to economic welfare measures.   Even
though the impact of the regulation directly affects only
certain gasoline distribution markets, the interaction among the
markets transmits these changes to upstream and downstream
markets.  The cumulative welfare impact, as well as the
distributional effect of this regulation on consumers and
producers, can be measured in the two "final" markets:  High
Volume Service Stations and Low Volume Service Stations.82

     For this analysis, measures of producers and consumers
surplus are used to approximate the theoretically correct
willingness-to-pay measures of welfare change.  If the income
effects of the regulation are small, this approximation is quite
good.83  The Economic Impact Analysis report provides a more
detailed discussion of the theory and procedures used to
estimate these economic welfare and distribution estimates.67

     Table 8-33 presents estimates of changes in producer and
consumer surplus and economic welfare based on the quantity and
                              8-82

-------
   TABLE 8-33.   ESTIMATED CHANGES IN  ECONOMIC
           WELFARE ($106  1990 DOLLARS)
                   ALT IV      ALT IV-Q    ALT IV-M
Transfers
Consumer Surplus
   High Volume      -134.4        -134.4      -134.4
   Low Volume        -29.2         -29.2       -29.2
      Total         -163.6        -163.6      -163.6
Producer Surplus
      Total          145.3         145.8       145.4
Net Welfare Change
      Costs          -18.3         -17.8       -18.2
                        8-83

-------
price changes of the market interaction model and the facility
costs estimated in Chapter 7.  All consumers lose some surplus
(bear some cost) due to the increase in price and decrease in
quantity of gasoline associated with the regulation.  Although
the price and quantity changes are themselves relatively small,
the estimated loss amounts to about $163 million a year.  The
magnitude substantially exceeds aggregate control cost estimates
because of the huge volume of gasoline across which the price
increases apply.  At the same time, some producers lose  (those
with high compliance and production costs) while others benefit
from the higher prices more than they are damaged by the costs
of compliance.  On net, producers gain an estimated surplus of
about $145 million per year. These estimates of producer surplus
vary slightly across the three regulatory alternatives because
the real resource costs borne by existing firms change with the
alternatives.

     The net difference in surplus changes is the economic
welfare cost of the regulation after market adjustments.  This
figure is estimated to be roughly $18 million per year and
varies slightly between regulatory alternatives IV, IVQ, and
IVM.  Note that this estimate does not reflect the environmental
and health benefits that the regulation yields.  Judging the
merit of the regulation on grounds of economic efficiency is
possible only if one weighs these economic welfare costs against
the benefits they produce.

8.3.6  Small  Business  Impacts
     The Economic  Impact Analysis^? develops estimates of the
size distribution of firms in different segments of the gasoline
distribution industry based on the number of establishments
owned and assignment of model plant combinations to the firms
owning multiple plants.  As shown on Table 8-34, when the Small
Business Administration's definition of small business is
                               8-84

-------

C-a
M
en
5 E-«
M cn
u< 3
a: §
E- M
M
S2
lORDANCE
ITRIBUTIO:
<*~> cn
2 M
0 Q
W
^ rH
^* ^-3
en cn
cn <
U C5
M cy
en EC

rf^ ^^
ti,
J °

< tn
2 a;
eng
Ci- CJ
0 U
cn
en
0 2
M <
M [rj
2 J
H CO
Cl4 Qi
Q (£
S§2
CO CO
H-f
• «
'a' O
00 E^
**c
W CJ
§
^



E-
r. 3
SRCENTAGE 01
AL THROUGH?1
o^ fc



lit
o v)
ERCENTAGE
TOTAL FIRM
a.
















§
VH;
< f*
« 2
CO g
a,
u
Q

















i
d
CO
Q
§
"~*




Q
REPRESENTE



U
C>3
M
CO

2
h
E-
co
u
O
§
j





h.
O

















v>
PRESENTED B
U
cc




BY THESE



s
E"
1-1
S

n
s
a,
2:
8





j
co




U
r-l
CO











HESE ESTAB-
LISHMENTS
EH





ESTAB-
LISHMENTS



0
2 a
o rz
^ . t/1
H r ..
M I
H M
&• D
Q «
< H
ca d
CO g
u w
X
E"1


co
rn
WJ
U
M
frt
wJ
CQ




§
O







«
g
O
U
M

-






VO
in



CO
g
w
s
CO
y ^

§
co
U

1
r-l



CO
U
U
>H
o
a, •

o
o
V

t—
in




to
2
M
"
3
D
0)

a»






o\



co
2
y
s
co
j
CQ
<
E-
co
U

,
r-4



co
U
U
>t
o
vJ
a,
§
o
o
V

r-l
rH
in





CO
i
E!
3
§

o

o
a
Oi
s
0
o
V

iH
tn



Z
0
M
g
11
CO »J
2 a

0)






































2
H
1
cc
§

i


tn






ffl



CO
W
§
CO
n
J
1
co
w

(
rH



u
2 P
O 2
fH LQ
j W
M C6
in <
• ^
^ 2
^ §

r-l
tn
in




U
O
M
>
£§
CO O
M
CJ £•«
M <
J E-
CQ M
2
o »
r^" §
5g
T? rH
3
« C
g 5
.^ **
^"^ ^
0)
o *"
5 a,

* j:
rrt O>
^ j;
fl) ^t
(-» J-* .
ij
t— <
^ •£
5 -5
2 w
VH _i
0 ^
Q.r2
si
H3 B
^)

J-
§ w
•al
"3 '^
C *
10 -H
^
0 w
iJ
irt

-------
applied to these firms, the majority of firms are classified as
small businesses in every industry segment examined.  The
percentage of firms classified as small ranges from  56 percent
for bulk terminals to 99 percent for public service  stations.

     This striking result occurs in part because of  the way in
which these data were compiled: the firm size categories were
coarse and the data did not allow for vertical or horizontal
integration of firms.  Finer, more complete data would probably
result in a substantial reduction in the number of firms
classified as small in each sector of the gasoline distribution
industry.  Even so, the evidence compiled in Table 8-34, when
added to the information on industry organization compiled in
Section 8.1, suggest that there are a substantial number of
small firms distributing gasoline that will be affected by the
regulation either directly or indirectly through increases in
the cost of gasoline or reductions in gasoline consumption.

     At the same time, however, there is little to suggest that
any of the regulatory alternatives under consideration would
result in financial impacts that would significantly or
differentially stress the affected small businesses.  This
conclusion is based on three considerations:
      • First, the sectors that are being directly regulated are
       the same sectors that are characterized by larger firms
       and vertical integration back through gasoline
       production: pipelines, bulk terminals, and bulk terminal
       transportation.  Bulk plants,  bulk plant transportation,
       and service stations are not affected directly by the
       regulation because they are not major emissions sources.
      • Second, for all but the smallest facilities in directly
       affected industry segments,  the costs of control
       associated with any of these alternatives are a minute
       fraction of production costs.   More importantly,  small
       scale facilities are likely to be serving small or
       specialized markets.  This makes it unlikely that the
       differential in unit cost of control estimated between
       the smallest and largest model plants of an industry
       sector will seriously affect the competitive position of
                              8-86

-------
       small firms,  even assuming that the small firms own small
       facilities.

     • Finally,  the examination of firm financial impacts
       performed using pro forma balance sheets showed that even
       small firms  in poor financial condition could fund
       estimated control costs with cash balances and that
       financial ratio of small firms were not significantly
       impacted by  the regulation.  The available data,  while
       admittedly limiting the precision of the analysis,
       nevertheless suggest that only firms that are
       exceptionally vulnerable financially will be threatened
       by the cost  of these controls.  This threat appears to
       depend more  on the financial condition of the firm that
       on its size.


     While EPA expects that this regulation will slightly slow

growth in facilities and jobs in most sectors and that,  in the

bulk plant and bulk plant transportation sectors, the closure of

some existing firms will be hastened, small firms in the

gasoline distribution industry would not be differentially

affected by these regulations because of their size alone.
                              8-87

-------
 8.4   REFERENCES
    U.S.  Department  of Energy,  Energy Information
    Administration.  Petroleum Supply Annual,  1987.   Volume 1
    May  19,  1988.

    U.S.  Department  of Energy,  Energy Information
    Administration.   Petroleum Supply Monthly,  January 1991.
    January  28,  1991.

    U.S.  Department  of Energy,  Energy Information
    Administration.   State Energy Data Report  1960-1988.   1990.

    Temple,  Barker & Sloane,  Inc.   Gasoline Marketing  in  the
    1980s:   Structure,  Practices,  and Public Policy.   (Study
    sponsored by API).  May 1988.   p.  1.

    deChazeau, M. G.  and  Kahn,  A.  E.   Integration and
    Competition  in the Petroleum Industry.  Port  Washington,  New
    York, Kennikat Press.   1973.   p.  371.

    Petroleum Marketers Association of  America.   1990  Marketer
    Profile  Survey.   1990.   p.  13.

    U.S.  Department  of Transportation,  Federal Highway
    Administration 1983.   Highway  Statistics, 1982.

    U.S.  Department  of Transportation,  Federal Highway
    Administration 1988.   Highway  Statistics, 1987.

    U.S.  Department  of Transportation,  Federal Highway
    Administration 1990.   Highway  Statistics,  1989.

    Telecon.  Thompson, Sam,  Pacific Environmental Services, with
    Mercer,Charlie,   Ewing Oil, March 4, 1991.   Stage I Gasoline
    Marketing Sector Margins.

11. Telecon.  Thompson, Sam, Pacific Environmental Services,
    Inc., with Childers, Grady, Plantation Pipeline.  March 6,
    1991.  Pipeline Margin.

12. Reference 4,   p.  12.

13. Measday,  w.  S.   The Petroleum Industry,  The Structure  of
    American  Industry, Fifth Edition.  New York,  Macmillan
    Publishing.   1977.  p. 142.
10
                              8-88

-------
14. U.S.  Department; of Energy,  Energy Information
    Administration.  Performance Profiles of Major Energy
    Producers,  1989.  January 1991.

15. U.S.  Department of Energy,  Energy Information
    Administration.  Performance Profiles of Major Energy
    Producers,  1987.  1988.

16. U.S.  Department of Energy,  Energy Information
    Administration.  Performance Profiles of Major Energy
    Producers,  1985.  1986.

17. U.S.  Department of Energy,  Energy Information
    Administration.  Performance Profiles of Major Energy
    Producers,  1983.  1984.

18. U.S.  Department of Energy,  Energy Information
    Administration.  Performance Profiles of Major Energy
    Producers,  1982.  1983.

19. Reference 13,  p. 153.

20. U.S.  Department of Energy,  Energy Information
    Administration.  Petroleum Marketing Monthly.  January 1991.
    p. 129.

21. National Petroleum News.  Belated API Study Denies Motorists
    Are Being Conned by Premium Grade Advertising. National
    Petroleum News, March 1991.  p.  25.

22. Reference 13,  p. 153.

23. National Petroleum News.  1990 Factbook Issue.  National
    Petroleum News,  June 1990.  p.  52.

26. Reference 6,  p. 2.

25. National Petroleum News.  Survey Indicates Jobbers,
    Especially Small Ones,  Doing Better  Than Expected.  National
    Petroleum News,  May 1990.   p. 39.

26. U.S.  Department of Labor,  Bureau of  Labor Statistics.
    Outlook 2000:   Industry Output and Employment.  Monthly
    Labor Review.   November 1989.   p.  36.

27. Reference 23,  p. 149.

28. U.S.  Department of Commerce,  Bureau  of the Census.  1987
    Census of Wholesale Trade—Establishment and Firm Size.
    February 1990.

29. Reference 23,  p. 52.
                              8-89

-------
30.  National Petroleum News.   1983 Factbook Issue.  National
    Petroleum News,   June 1983.   p. 56.

31.  U.S.  Department of Commerce,  Bureau of the Census.  1987
    Census of Wholesale Trade--Commodity Line Sales.  July 1990.

32.  U.S.  Department of Commerce,  Bureau of the Census.  1977
    Census of Wholesale Trade—Commodity Line Sales.  1980.

33.  U.S.  Department of Commerce,  Bureau of the Census.  1977
    Census of Wholesale Trade--Establishment and Firm Size.
    June 1980.

34.  Duns Analytical Services.  Industry Norms and Key Business
    Ratios, 1990-1991.  Dun and Bradstreet Credit Services.
    1991.

35.  Duns Analytical Services.  Industry Norms and Key Business
    Ratios, 1989-1990.  Dun and Bradstreet Credit Services.
    1990.

36.  Duns Analytical Services.  Industry Norms and Key Business
    Ratios, 1987-1988.  Dun and Bradstreet Credit Services.
    1988.

37,  Duns Analytical Services.  Industry Norms and Key Business
    Ratios, 1979-1980.  Dun and Bradstreet Credit Services.
    1980.  p. 2.

38.  Telecon.  Thompson, Sam,  Pacific Environmental Services,
    Inc.  with Faulkner, Barbara,  Petroleum Marketers Association
    of America, February 6, 1991.  Number of Bulk Plants.

39.  Stalsby/Wilson Associates, Inc.  Petroleum Terminal
    Encyclopedia, Fifth Edition.   Houston, Stalsby/Wilson Press.
    1990.

40.  Norton, R.J. and L.P. Norwood.  (Pacific Environmental
    Services, Inc.)  Description of Analysis Conducted to
    Estimate Impacts of Benzene Emissions from Stage I Gasoline
    Marke'ting Sources.  Prepared for U.S. Environmental
    Protection Agency.  Research Triangle Park, NC.  August
    1989.

41.  U.S. Environmental Protection Agency.  Evaluation of Air
    Pollution Regulatory Strategies for Gasoline Marketing
    Industry.  Research Triangle Park,  NC.  EPA-450/3-84-012a.
    July 1984.

42,  Reference 23, pp. 36-43.
                              8-90

-------
43. National Petroleum News.  1987 Factbook Issue.  National
    Petroleum News.  June 1987.   pp.  40-46.

44. National Petroleum News.  1982 Factbook Issue.  National
    Petroleum News.  June 1982.   pp.  38-44.

45. U.S.  Department of Commerce,  Bureau of the Census.  1987
    Census of Wholesale Trade—Miscellaneous Subjects.  March
    1991.

46. U.S.  Department of Commerce,  Bureau of the Census.  1977
    Census of Wholesale Trade—Miscellaneous Subjects.  1980.

47. Reference 6,  p. 12.

48. Association of Oil Pipelines,  various years.   Shifts in
    Petroleum Transportation.

49. Eno Foundation for Transportation,  Inc.  various years.
    Transportation in America.

50. Reference 6,  p. 11.

51. Watts, J.  Dimensions of the 500  Leading U.S.  Energy
    Pipeline Companies:  The 9th P&GJ Report.   Pipeline and  Gas
    Journal  £-.21-36.  August 1989.

52. Reference 6,  p. 8.

53. Reference 23,  p.  130.

54. National Petroleum News.  1989 Factbook Issue.  National
    Petroleum News  June 1989.  p. 128.

55. Reference 23,  p.  141.

56. National Petroleum News.  The Rural  American Market:   How
    Are Marketers  Coping?  National Petroleum News  November
    1990.

57. American Petroleum Institute.  Gasoline Marketing in the
    United States  Today,  Second  Edition.   API  Publication Number
    1593.   September 1986.   pp.  8-9.

58. Reference 57,  p.  11.

59. Telecon.  Bollman, Andy, Research Triangle Institute,  with
    Keene, Bill, Lundberg Survey, Inc. April 19, 1991.   Number
    of Public Service Stations.
                              8-91

-------
60. Shaner,  R.  J.   Counting Procedure Shows How Retail Outlet
    Population Is  Greater than Suspected.  National Petroleum
    News,   April 1991.

61. Reference 4, p. 38.

62. Reference 23,  p. 142.

63. U.S. Department of Commerce,  Bureau of the Census.  1987
    Census of Retail Trade—Establishment and Firm Size.
    January 1990.

64. U.S. Department of Commerce,  Bureau of the Census.  1982
    Census of Retail Trade—Establishment and Firm Size.  1985,
65. U.S. Department of Energy,  Energy Information
    Administration.  Annual Energy Outlook, 1990.
January 1990.
66. Abcede, A.  Bulk Plants Continue Decline Amid New
    Regulations,  Poor Economics.   National Petroleum News.
    August 1986.

67. U.S. Environmental Protection Agency.  Economic Impact and
    Preliminary Regulatory Flexibility Analysis for Air Quality
    Standards Proposed for the Gasoline Distribution Industry,
    (September, 1992).

68. U.S. Department of Commerce,  Bureau of the Census.
    Unpublished 1983 survey of shipments by SIC by
    transportation mode.

69. Memorandum, from Bollman,  A.  to Mathias S.  U.S.
    Environmental Protection Agency,  Office of Air.  January 28,
    1992.  Revised Estimates of New Capacity, New Preplacement
    Capacity and Existing Facilities.

70. Hang, J.C. and R.R. Sakaida.   Survey of Gasoline Tank Trucks
    and Rail Cars.  U.S.  Environmental Protection Agency.
    Research Triangle Park, NC.  Publication No. EPA-450/3-79-
    004.  March 1979.

71. U.S. Department of Commerce,  Bureau of the Census.   Census
    of Transportation:  Truck Inventory and Use Survey,  1977 and
    1987.

72. Telecon.  Bollman, Andy/ Research Triangle Institute, with
    Cebula, Andy/ National Air Transportation Association, July
    29, 1991.  Number of Aviation Gasoline Facilities.

73. Telecon.  Cebula, Andy, National Air Transportation
    Association,  with Norwood, Phil,  Pacific Environmental
                              8-92

-------
    Services,  July 25,  1991.   Number of Tank Trucks per Aviation
    Gasoline Bulk Plant.

74. U.S.  Environmental Protection Agency, Bulk Gasoline
    Terminals Background Information for Proposed Standards.
    Publication No. EPA-450/3-80-038a.   December 1980.

75. Telecon.  Thompson,  S.H.,  Pacific Environmental Services,
    Inc.,  with McCauley,  V.,  U.S. Department of Transportation.
    March 12,  1991.  Product  Pipelines.

76. Memorandum, from Thompson,  S.H., to Shedd, S.A.,  U.S.
    Environmental Protection  Agency/ Chemicals and Petroleum
    Branch.   March 27,  1991.   Trip Report for Plantation
    Pipeline,  Greensboro,  NC.

77. Products Pipelines of the United States and Canada.  Tulsa,
    PennWell Publishing Company..   1988.

78. Hicks,  J.R.  The Theory of Wages.  New York,  Peter  Smith.
    1948.

79. Muth,  R.F. the Derived Demand Curve for a Productive Factor
    and the Industry Supply Curve.  Oxford Economic Papers.   16:
    221-234.  1964.

80. Nicholson, Walter.   Intermediate Microeconomics and Its
    Application,  2nd ed.  The  Dryden Press. Chicago, IL.  1979.
    pp. 292-293.

81. Anderson,  D.W. and Chandran,  R.V.  Market Estimates of
    Worker Dislocation Costs.   Economics Letters 2A-  381-384.
    1987

82. Just,  Richard E.,  Darrell L.  Hueth, and Andrew Schmitz.
    1982.   Applied Welfare Economics and Public Policy.
    Englewood Cliffs:   Prentice-Hall, Inc.

83. Willig,  Robert D.,   1976.   Consumer's Surplus Without
    Apology.  American Economic Review.  66(4): 597-98.
                              8-93

-------
                            APPENDIX A
         EVOLUTION OF  THE BACKGROUND INFORMATION DOCUMENT

     The purpose of this study was to develop a basis for
supporting proposed national emission standards for hazardous air
pollutants (NESHAP) for the gasoline distribution (Stage I)
network.  To accomplish the objectives of this program, technical
data were acquired on the following aspects of this industry:
(1) facility types and emission sources, (2) the release of HAP
and VOC emissions into the atmosphere by these sources, and  (3)
the types and costs of demonstrated emission control
technologies.  The bulk of the information was gathered from the
following sources:
     1.   Technical literature;
     2.   State, regional, and local air pollution control
agencies;
     3.   Plant visits;
     4.   Industry representatives; and
     5.   Equipment vendors.
     Significant events relating to the evolution of the
background information document are recorded in chronological
order in Table A-l.
                               A-l

-------
   TABLE  A-l.    EVOLUTION OF  THE  BACKGROUND  INFORMATION  DOCUMENT
Date
                     Company, Consultant, or Agency/Location
                                           Nature of Action
3/8/74
U.S. Environmental Protection Agency
11/1/76 to 6/1/77    U.S. Environmental Protection Agency
Promulgated NSPS for  New Petroleum
   Liquid Storage Tanks (40 CFR Part 60
   Subpart K, 39 FR 9317).

Section 114 letters sent to oil
   companies regarding specific bulk
   terminals.
6/8/77
U.S. Environmental Protection Agency
Benzene is listed as  a hazardous air
   pollutant  (HAP) under section 112 of
   the Clean  Air  Act.
10/77
U.S. Environmental Protection Agency
Bulk Gasoline Terminal Control
   Techniques Guideline issued (Control
   of Hydrocarbons  from Tank Truck
   Gasoline Loading Terminals.  EPA
   Publication No.  EPA-450/2-77-026).
12/77
U.S. Environmental Protection Agency
Fixed-Roof Tank Control Techniques
   Guideline issued  (Control of
   Volatile Organic  Emissions from
   Storage of Petroleum Liquids in
   Fixed-Roof Tanks.  EPA Publication
   No.  EPA-450/2-77-036).
12/77
U.S. Environmental Protection Agency
Bulk Gasoline Plant  Control Techniques
   Guideline issued  (Control of
   Volatile Organic  Emissions from Bulk
   Gasoline Plants.  EPA Publication
   No.  EPA-450/2-77-035).
6/78
U.S Environmental Protection Agency
1978
12/78
National  Air  Pollution Control Techniques
Advisory  Committee (NAPCTAC)

U.S. Environmental Protection Agency
Petroleum Refinery Equipment Leak
   Control  Techniques Guideline issued
   (Control of Volatile Organic
   Compound Leaks  from Petroleum
   Refinery Equipment.  EPA Publication
   No.  EPA-450/2-78-036).

Review of draft  Stage I Benzene
   Package.

External Floating  Roof Tank Control
   Techniques Guideline issued (Control
   of Volatile Organic Emissions from
   Petroleum Liquid Storage in External
   Floating Roof Tanks.  EPA
   Publication No.  EPA-450/2-78-047).
                                                 A-2

-------
                                 TABLE  A-l.    (Continued)
Date
                      Company,  Consultant,  or Agency/Location    Nature of Action
 12/78
U.S. Environmental Protection Agency
4/4/80
12/17/80
U.S. Environmental Protection Agency
U.S. Environmental Protection Agency
8/18/83
U.S. Environmental Protection Agency
5/30/84
6/84
8/8/84
2/7/87

4/8/87
7/87
U.S. Environmental Protection Agency



U.S. Environmental Protection Agency





U.S. Environmental Protection Agency




Natural Resources Defense Council

U.S. Environmental Protection Agency




U.S. Environmental Protection Agency
9/14/89
U.S. Environmental  Protection Agency
Tank Truck/Vapor Collection System
   Control Techniques Guideline
   issued (Control of Volatile
   Organic Compound Leaks from
   Gasoline Tank Trucks and Vapor
   Collection Systems.  EPA
   Publication No. EPA-450/2-78-051).

Promulgated additional NSPS for New
   Petroleum Liquid Storage Vessels
   (40 CFR 60 Subpart Ka, 45 FR
   23379).

Proposed NSPS for new Bulk Gasoline
   Terminals (40 CFR 60 Subpart XX,
   45 FR 83126) and issued draft
   background information document
   (EPA Publication No. EPA-450/3-80-
   038a).

Promulgated NSPS for new Bulk
   Gasoline Terminals (40 CFR 60
   Subpart XX, 48 FR 37590) and
   issued final background
   information document (EPA
   Publication No. EPA-450/3-80-
   038b).

Promulgated NSPS for Equipment Leaks
   of V(3C at Petroleum Refineries (40
   CFR 60 Subpart GGG, 49 FR 22606).

Draft For Risk Exposure issued
   (Estimation of the Public Health
   Risk from Exposure to Gasoline
   Vapor via the Gasoline Marketing
   System).

Issuance of Evaluation of Air
   Pollution Regulatory Strategies
   for Gasoline Marketing Industry
   (EPA-450/3-84-012a).

NRDC lawsuit.

Promulgated additional NSPS for New
   Petroleum Liquid Storage Vessels
   (40 CFR 60 Subpart Kb, 52 FR
   11428).

Issuance of "Draft Regulatory Impact
   Analysis: Proposed Refueling
   Emission Regulation for Gasoline-
   Fueled Motor Vehicles • Volume I:
   Analysis of Gasoline Marketing
   Regulatory Strategies."  EPA-
   450/3-87-OOIa.

Proposed Gasoline Marketing Benzene
   Standards (54 FR 38083).
                                                 A-3

-------
                                TABLE A-l.     (Continued)
Date
                      Company,  Consultant,  or Agency/Location    Nature of Action
12/20/90
Piedmont Aviation Services,
Uinston-Selem, NC
Plant visit to gather background
   information concerning airplane
   fueling and gasoline throughput.
3/7/90
U.S. Environmental Protection Agency
Withdrew Gasoline Marketing Benzene
   Standards <45 FR 8292).
11/15/90
U.S. Environmental Protection Agency
Additional compounds in gasoline
   listed as HAPs (1990 CAAA).
12/18/90
Fina Oil & Chemical Co.,
Port Arthur, TX
Plant visit to gather background
   information concerning vacuum
   assist technology for tank truck
   loading at terminals.
1/17/91
2/4/91
2/21/91
Puget Sound Air Pollution Control
Agency, Seattle, UA
                      New Jersey State Department of
                      Environmental  Protection, Trenton, NJ
American Petroleum Institute (API),
Washington,  DC
Plantation Pipe Line,
Gas torn a, NC
Letter requesting performance test
   reports for vapor control  systems
   at bulk gasoline terminals.

Letter requesting performance test
   reports for vapor control  systems
   at bulk gasoline terminals.

Letter requesting information
   concerning the composition of
   gasoline vapors.

Plant visit to gather background
   information concerning operations
   at pipeline pumping stations.
2/22/91
2/25/91
2/26/91
2/26/91
4/22/91
4/23/91
Service Distributing Company,  Inc.,
Albemarle, NC
Braswell Equipment Co.,
Wilson, NC
Arnold Equipment Co.,
Greensboro, NC
                      Southern Pump and Tank Co.,
                      Raleigh,  NC
Braswell Equipment Co.,
Wilson, NC
Mobil Oil Corporation,
Albany, NY


Powell Duffryn Terminals,  Inc.,
Bayonne, NJ
Letter requesting cost information
   concerning installing and
   retrofitting Stage I  vapor
   recovery at service stations.

Letter requesting information
   concerning bulk gasoline plant and
   service station costs.

Letter requesting information
   concerning bulk gasoline plant and
   service station costs.

Letter requesting information
   concerning bulk gasoline plant
   and service station costs.

Letter requesting information
   concerning bulk gasoline plant and
   service station costs.

Plant visit to gather background
   information concerning railcar
   loading operations.

Plant visit.
                                                 A-4

-------
                                TABLE  A-l.    (Concluded)
Date
Company. Consultant,  or  Agency/Location    Nature of Action
6/21/91
9/19/91
9/30/91
11/91


7/16/92




9/92

11/17/92

2/18/93
U.S. Environmental Protection Agency
Maryland Department of Environment,
Baltimore, MD
U.S. Environmental Protection Agency
Industry members,  selected equipment
vendors and consultants

U.S. Environmental Protection Agency
NAPCTAC

U.S. EPA/NAPCTAC,  Durham, NC

U.S. EPA/API,  Durham,  NC
Federal Register notice  announcing
   availability of preliminary draft
   list of categories  of major and area
   sources of HAPs (56 FR 28548).

Letter requesting information
   concerning bulk gasoline plant and
   service station costs.

Floating and Fixed-Roof  Tank Control
   Techniques issued (Control of
   Volatile Organic Compound Emissions
   from Volatile Organic Liquid Storage
   in Floating and Fixed-Roof Tanks.
   Draft.)

Mailed draft BID Chapters 3-8.2 and
   Appendices B & C.

Federal Register notice  publishing
   initial list of categories of major
   and area sources of HAPs (57 FR
   31576).

Received draft BID for comment.

NAPCTAC meeting.

Meeting to discuss issues and comments
  from NAPCTAC meeting.
                                                A-5

-------
                           APPENDIX B
          INDEX TO ENVIRONMENTAL IMPACT CONSIDERATIONS

     This appendix consists of a reference system which is cross-
indexed with the October 21,  1974,  Federal Register (39 FR 37419)
containing the Agency guidelines concerning the preparation of
environmental impact statements.  This index can be used to
identify sections of this document which contain data and
information germane to any portion of the Federal Register
guidelines.
                               B-l

-------
     TABLE  B-l.  CROSS-INDEXED REFERENCE  SYSTEM TO HIGHLIGHT
           ENVIRONMENTAL IMPACT PORTIONS OF  THE DOCUMENT
Agency guidelines  for preparing
regulatory action  for environmental
impact statements  (39 FR 37419)
Location within the background
information document
1.    BACKGROUND AND SUMMARY OF
      REGULATORY ALTERNATIVES

      Summary of regulatory
      alternatives
The regulatory alternatives  from
which standards will  be chosen  for
proposal are summarized in Chapter  1,
Section 1.2.
      Statutory basis  for proposing
      standards
The statutory basis for proposing
standards is summarized in  Chapter  1,
Section 1.1.
      Relationship  to other
      regulatory agency actions
The relationships between  EPA  and
other regulatory agency actions  are
discussed in Chapters  3, 7,  and  8.
      Industries affected  by the
      regulatory alternatives
A discussion of the industries
affected by the regulatory
alternatives is presented  in  Chapter
3, Section 3.1.  Further details
covering the business and economic
nature of the industry are  presented
in Chapters 6,  7,  and 8.
      Specific  processes  affected by
      the regulatory alternatives
The specific processes  and  facilities
affected by the regulatory
alternatives are summarized in
Chapter 1, Section 1.1.
A detailed technical  discussion  of
the processes affected  by the
regulatory alternatives  is  present in
Chapter 4, Section 4.1.
                                   B-2

-------
                        TABLE B-l.    (Concluded)
Agency guidelines for preparing
regulatory action for environmental
impact statements (39 FR 37419)
Location within the background
information document
2.    REGULATORY ALTERNATIVES

      Control  techniques
The alternative control techniques
are discussed in Chapter 4.
      Regulatory alternatives
The various regulatory alternatives
are defined in Chapter 5, Section
5.2.  A summary of the major
alternatives considered is included
in Chapter 1, Section 1.2.
3.    ENVIRONMENTAL IMPACT OF THE
      REGULATORY ALTERNATIVES

      Primary impacts directly
      attributable to the regulatory
      alternatives
The primary impacts on mass emissions
and ambient air quality due to the
alternative control systems are
discussed in Chapter 6, Section 6.1.
A matrix summarizing the
environmental  impacts is included in
Chapter 1.
      Secondary or induced impacts
Secondary impacts for the various
regulatory alternatives are discussed
in Chapter 6, Sections 6.2, 6.3,  6.4,
6.5, and 6.6.
4.     OTHER CONSIDERATIONS
A summary of the potential  adverse
environmental impacts associated with
the regulatory alternatives is
included in Chapter 1,  Section 1.3,
and Chapter 6.  Potential  socio-
economic and inflationary  impacts are
discussed in Chapter 8,  Section 8.3.
                                    B-3

-------
                         APPENDIX C
       CALCULATION OF HAP VAPOR PROFILES FOR GASOLINE

     The purpose of this appendix is to present the
methodology and results of the analysis to estimate the
hazardous air pollutants (HAPs) in gasoline vapor.  This
appendix consists of two sections.  The first section
contains the information resulting from a search conducted
to obtain data related to the composition of gasoline vapor,
that was specific enough to allow the identification and
quantification of those HAPs contained on the 1990 Clean Air
Act Amendments list.  Section C.I discusses the information
obtained from this search as well as the mathematical
procedures used to develop a "typical" HAP vapor profile for
normal gasoline.
     Requirements in Title II of the 1990 CAAA will lead to
the fuel composition being changed in many areas of the
country.  These programs are not yet in effect, so it was
difficult to obtain any actual data related to the
composition of gasoline vapors from reformulated or
oxygenated gasoline.  Therefore, adjustments were made to
the normal gasoline profile to attempt to represent vapor
compositions of possible reformulated or oxygenated
gasoline.  The methodology used to modify the normal profile
forms the basis for the second section of this appendix and
is discussed in Section C.2.

C.I  NORMAL GASOLINE
     To locate information on gasoline vapor composition,
literature searches were conducted and trade organizations,
research organizations, regulatory agencies, and large and

                          C-l

-------
small oil companies were contacted.  Overall,  over 100
sources were contacted to attempt to obtain information on
this subject.  These included the American Petroleum
Institute (API), Western States Petroleum Association
(WSPA), the National Institute for Petroleum and Energy
Research (NIPER), the Coordinating Research Council (CRC),
the Society of Automotive Engineers (SAE), the Motor
Vehicles Manufacturers Association (MVMA), all the major oil
companies, the California Air Resources Board, and many
others.
     Information obtained during this search indicated that
a great deal of research was being conducted related to the
composition of tailpipe emissions from automobiles.
However, information related to the composition of
evaporative emissions from gasoline transfer and storage
operations was limited.
     A total of forty nine analyses of gasoline vapor were
located that contained speciation of sufficient detail to
identify the CAAA HAPs.  These came from a variety of the
sources listed above.  In addition, EPA obtained a number of
compositional analyses of liquid gasoline.  Table C-l
summarizes the sources of the test data received.
     For each vapor sample, the individual HAPs were
identified and their weight percentage relative to the total
VOC weight was noted or calculated (in cases where the
fraction was reported as a volume or mole percent).  In
addition, the sum of all of the weight percentages of the
HAPs was determined.
     For the liquid samples, Raoult's law was used to
estimate the vapor phase composition.  Raoult's law
describes the relationship between the partial pressure  of a
component in the gas phase and the mole  fraction of that
component in the liquid phase.  Raoult's  law  is expressed as
follows:
                    PA = VAP = XAP*A
                          C-2

-------
            TABLE  C-l.   SUMMARY  OF  SOURCES OF  DATA
            RECEIVED REGARDING GASOLINE COMPOSITION
Data
 ID
Source of Test Data
Number
   of
Samples
Form
 of
Data
 A    Memorandum, from Knapp, K.T., EPA
      AEERL, to Durham, J., EPA OAQPS,
      regarding speciation of components in
      gasoline with data attached.  August
      1, 1990.

 B    Furey, R.L. and B.E. Nagel,
      Composition of Vapor Emitted from a
      Vehicle Gasoline Tank During
      Refueling.  GM Research Laboratories,
      Warren, MI.(Presented at SAE
      International Congress and
      Exposition, Detroit Michigan)

 C    Sisby, J.E., S. Tejada, W. Rau, J.
      Lang, and J. Duncan.  Volatile
      Organic Compound Emissions from 46
      In-Use Passenger Cars.  (Reprinted
      from Environmental Science and
      Technology, May 1987)

 D    Letter, from Woodward, P., National
      Institute for Petroleum and Energy
      Research, to Norwood, P., Pacific
      Environmental Services, Inc.,
      regarding composition of gasoline
      with data.  January 10, 1991

 E    Haider, C., G. Van Gorp, N. Hatoum,
      and T. Warne.  Gasoline Vapor
      Exposures.  Part I.  Characterization
      of Workplace Exposures.  American
      Industrial Hygiene Association,
      47(3):164-172 (1986).

 F    Appendix to Northeast Corridor
      Regional Modeling Project -
      Determination of Organic Species
      Profiles for Gasoline Liquids and
      Vapors - Sampling and Analysis Data
      Sheets, EPA-450/4-80-036b.  U.S.
      Environmental Protection Agency,
      Research Triangle Park, NC.  December
      1980.
                                         liquid
                                         vapor
                                         vapor
                                         liquid
                                         vapor
                                 20
           vapor
                              C-3

-------
                   TABLE  C-l.   (Concluded)
Data
 ID
Source of Test Data
Number
   of
Samples
Form
 of
Data
      Information Obtained From Braddock,
      J., EPArAEERL regarding vapor
      composition of refueling emissions.
                                  14
          vapor
  H    Environ Corporation, Arlington, VA.
      Summary Report on Individual
      Exposures to Gasoline.  Prepared for
      Gasoline Exposure Workshop Planning
      Group.  November 28, 1990.

  I    Passenger Car Hydrocarbon Emissions
      Speciation.  EPA-600/2-80-085.  U.S.
      Environmental Protection Agency,
      Research Triangle Park, NC.  May
      1980.
                                         vapor
                                         vapor
                 TOTAL NUMBER OF DATA POINTS
                                  49
                              C-4

-------
where, p*A is the vapor pressure of pure liquid A at temperature
T and yA is the mole fraction of A in the  gas phase.   Raoult's
law is an approximation that is generally valid when the mole
fraction  of compound A in the liquid is approximately close to
one and when the mixture is made up of similar substances, such
as straight chain hydrocarbons of similar molecular weights.
Gasoline  was assumed to meet the second criteria based on general
compositional data.
     An example of the calculational procedure used to estimate
vapor HAP composition from liquid composition is shown in Table
C-2.  All  non-HAP components were grouped according to the number
of carbons.  All compounds within each carbon number were assumed
to have the vapor pressure and molecular weight of certain
compounds  selected as representative for the carbon number.
Those compounds selected are shown in parenthesis in Table C-2.
     The  weight fraction for each HAP was identified in the
liquid data, and the weight fractions for each carbon number
(excluding HAPs) totalled.  The mole fraction of each HAP and
carbon number group were calculated.  The vapor pressure was then
estimated  using the Antoine equation (a. common vapor pressure
estimation technique) at 25 degrees F for each HAP or carbon
number group.
     Using the liquid mole fraction and the vapor pressure, and
assuming  one atmosphere total pressure the mole fraction in the
vapor phase was calculated using Raoult's law.  This was
converted  to mass fraction, after which the HAP to total VOC mass
ratio was  calculated.
     After the individual and total HAP weight fractions were
calculated for each individual sample, the data were combined and
summarized.  The results of all of the individual samples are
shown in Table C-3.  Also, Table C-4 presents the summary of the
data for normal gasoline.  The table shows the maximum and
minimum percentage for each HAP and for total HAPs.  The
arithmetic average was also taken for each of these situations.
                               C-5

-------
    TABLE C-2.   EXAMPLE OF VAPOR COMPOSITION
           CALCULATIONS FROM LIQUID DATA


CHEMICAL/CLASS
Hexane
Benzene
Toluene
2,2,4 trtmethylpentane
Xylene
Ethyl benxene
Naphthalene
Methane t
HTBE
TOTAL HAPS
c3 (propane)
c4 (n- butane)
c5 (iso-pentane)
c6 (2 methyl pentane)
c7 (2 methyl hexane)
cB (iso-octane)
c9 (1 meth-3 eth benz)
clO n-decane
c11 (n-undecane)
c12 (n-dodeeane)
TOTAL VOC

wt frac
in liq
1.8
1.31
6.19
3.02
6.33
1.27
0.67
0
0
20.59
0.02
4.83
H.85
11.45
8.5
6.53
12.45
9.74
6.13
0.82
95.91

moles in
liquid
0.021
0.017
0.067
0.026
0.060
0.012
0.005
0.000
0.000

0.000
0.086
0.212
0.136
0.087
0.058
0.099
0.070
0.040
0.005
1.001
liquid
mole frac
Xa
0.021
0.017
0.067
0.026
0.060
0.012
0.005
0.000
0.000
0.208
0.000
0.086
0.212
0.136
0.087
0.058
0.099
0.069
0.040
0.005
1
vapor
•ole frac
Ya
0.0027
0.0013
0.0015
0.0011
0.0003
0.0001
0.0000
0.0000
0.0000

0.0033
0.1513
0.1347
0.0251
0.0043
0.0023
0.0002
0.0001
0.0000
0.0000


wt frac
in vap
0.231
0.103
0.137
0.121
0.030
0.009
0.000
0.000
0.000

0.145
8.475
9.429
2.105
0.425
0.262
0.025
0.008
0.001
0.000
21.508

HAP/VOC
in vap
0.0108
0.0048
0.0064
0.0056
0.0014
0.0004
0.0000
0.0000
0.0000
0.0294











other gasoline formulations may contain methanol or MTBE
                       C-6

-------
ft

I
en
B
H
>
H
Q
2
 I
U

w
J
ffi
<
EH
H
II
II
n
u
u

n
u
tt
M
ti
u
ii
H
1)
II
It
tl
II
u
H

H
M
H
H
tl
II
H
H

H
H
U
U
H
tt
H
H

II
II
M
II
tt
tl

I
U

II
II
II
II
II
II
II
U
II
tl
II
II
II
II
II
tl
II
tl
II
u
II
II
u
1
1
1
u
tl

M
II
1)
II
H
It
II
II
||
II
II
tl
II
II
M
II
It
II
II
II
II
U
II
H
II
B
II
II
H
II
M
II
It
M
II
II
II
U
II
II
II
tt
II
II
II
u
II
If
II
II
jl
II

CO
W




_
h*-





s*
UJ



W
UJ




CM
UJ




_
UJ






1— CM

13

i


m
O —
D
^
oc

8
> rg
O.
«t




^



  S- «-
CM O O O O
o o o o o
o o o o o



O 0 O 0
h* fM IM «—
«- CM CM •-
o o o o

o o o o


g! S 5 E £
»- 0 — 0 0
o o o o o
o e e e o



e e o o o
»  n o to
•- ^ CM O •>»
~ IM CM IM O
O O O O O

O O O O O


CM *> •— *• CM O

^ «- r>a •• S O
o o o o o o
0 O O O 0 0


XI -O CM •• O tf\
O O CM O O O
O O O O O O
o o o o o o


O r^ CM •- •- K>
»/> «* *r m *• o
o o o o o o
o o o o o o
0 O O 0 0 0
o o e o
— r- o M
•— o •- o
0000

0 0 O 0



o m in o


o o o o

0 O O 0



o ^ O -» KI >n

«- o o o o o
o o o o a o
0 O O 0 O 0



to (o •» « •» »»
o «r «o trt v o
^ o o o o o
o o o o o o
o o o o o o


^^
I
C M
2 S
i 1

a. x 8"
< 'Sea
v «> « o
I »-6 J!

•I tl *< M £ «
dec «i j; v
cinsrc— •-£

K C — (M*— jr'S.Em
titio - >. •< o 3 i—
S
s
o



•o
i
o'



o~

o

o


•0
I
o



§
^
o



s
§
o'



o


o

o





o
o


in
o
o
o


fM

fM
O
O
O

o

o



o

-o
o

o



o

fM
o
o



^
o
fM
0
o



I/I
o.
I
o








H
II
U
II
II
y
u
n
it
u
it
n
n
u
u
II
II
u
II
II
H
II
H
H
It
U
It
II
II
H
N
H
H
II
II
H
II
If
M
II
tt
fl
II
II
II
II
II
II
H
II
tt
II
II
11
II













































1
1
1
1
i










                                         C-7

-------
                                   tn  o> -o  to

                                   fe  s i  g
                                   a  o o  e

                                   d  d d  d
                                   •— cvi  «j  o-  in  KI
                                   o. KI  in  >o  —  o
                                   o o  «-  o  o  o
                                   a e  o  o  o  o

                                   o o  o  o  o  o



                                   o» «3  «n  CM
                                   03 «»  
                                  «o oj  »..  w> S

                                  5 §  §  5 §
 0)
 3
 C
•H

4-)
 C
 O
o
f>
 I
o
CO
    0  0  0
                      o

                      d
•*  Kl  K.
*•*  K.  CO

•-  O  O
000


odd
—  o
o  o
o  o

d  d
                    u
                    o
                    :>
»""  «*T  ^y  O

°  8  =  §

o  d  o'  d
II
II
li
M
it
I
I
1
!
H
II
II
II
t
1
t
,
1
1
I
I
H
II
II
II
II
II
II
H
II
II
II
II
II
II
II
II
II
II
II
II
II
U
II
II

II
II
II
II
II
II
II
II
II
n
u
*. 1
o

rvj
-O
IM
•0 0
*"" d

o«
(M
in O
t*. •
O

5
*^ O

**" d

S
rj
Kl O
^ •
O

o
*•»
in


a
S
s


•1
Jf
5
)g
o
3C
O
O
o

o-
rg
o
0
d

in
CO
o
o
d






m
*?
o
o
o










g
5



en
IM
0
O

o
m
O
o
O

l«o
O-
o
o
d

CO
r*.
*"
o
d

o
r*.
O
o
O










S
£

3

O
o o
o o
o o






o
o
o

s
o
o
d

CM in
fSI 0
0 O
0 0
0 0

*•»
1
C M
a C
- X
£ =
s. >.
x §•
j: ti L.
U U t> §*
E N c u 	
— C Cl —
L. t, — ^
«• 
•»
O
o


in
c.
«
z
«J
«
5
^










l
i
i
i
1
l







ll
li
n
n
it
ll
ll
ll
U
II
II
II
M
II
ti
II
ll
II
II
ll
li
It
II
II
II
II
II
II
II

H
H
||
H
H
II
II
H
II
II
H
                                                                  C-8

-------
                                O •£>
                                f\» w-
                                o o
                                          o>  (M O»  •»
                                          in  (V ^  o
                                O O  O O  O
                         O
                         13
o in
fM •••
o o

o o
^  O
o  o

o  o
                                v •*  o   o
                                m •»  m  o>  £  «M

















^

G-
•c
X
































f>.
43




«
°





>n
u


^
u



o





r\j
13



^
u


C
ro



0;


00


N.
**•



a



«n
w






o
o




o
e





o






o


o
CO
o
o
o


g

o


f\J
O
O
0



CO
o
0
0
_
IN)
g
0
^
Kl
O
o
o
S
o
o
o
a
o
o
o












I
N
c

o
e



R
g
a



e

g
o


5
o
0
0


i
0

o


eo
•o
0
0
0



•o
o
o
o
»•!

o
0
M

g
o
R
§
o
£
o
a
o












%>
|

^
0
o



e
S
o



in

0
o


in
§
0


o
o

o


IM
o
o
0



Irt
g
0
in
o
0
o




K.
o
o
o
»
o
o
o







a
1
X
j:

^
IM

0
O



g
o
o
o




-------
                                            2«P  «-      f-  «M  •- •-
                                            V*  *—      O  f\J  O O
                                        »-  »  CM      »-  O  O O
                                IW      O  O  O      O  O  O 0
T!
 0)
T3
 U
 C
 o
 u
  I
 U
 OQ
 <
 EH

•>-
5
i

>-
a
Q

ac
§
*>•.
a.
























o o o a a a e
^C fc X ^ *rt **
7- O O O O O O
O O O 0 0 O

nj •— N. u"i o
O> eo -» co ^
*- O IM O O
— o a o o o

a o o o o
is* ~r •- o ~a -a  (^ (> O O O -
f ^^ C
i u ti
e M c
C tl
i








3
0
>^
CO

*-
o
»o
o

^*
o
^
•o

•—
o
1
V

c
&
r-

CL
o
If

0
CD
s
o


CO
-o
o

o
CO
fN.
o
CO
CO

^
o
^


^
o
^
cc
o
rsi
o
VI
a.
^


^
o




                                                   St>  «-  «  j2  "•  w
                                                   c      »i     je   n
                                               tl  tl  >»  C  —  ~   C
                                               M3    >4>>.f|ltU
                                               C-rM--.cQ.Em
                                               tlO    •XwsSr-
                                               CD>-(S1XU2UZ
                                                                           C-10

-------
       TABLE  C-4.   VAPOR  PROFILE OF NORMAL GASOLINE



HAZARDOUS AIR POLLUTANT8
Hexane
Benzene
Toluene
2,2,4 Trimethylpentane
(iso-octane)
Xylenes
Ethylbenzene
TOTAL HAPSb
HAP
TO VOC RATIO
(percentage by weight)

MINIMUM
0.3
0.2
0.4
0.03
0.05
0.03
2.0
ARITHMETIC
AVERAGE MAXIMUM
1.6 4.4
0.9 2.2
1.3 4.0
0.8 2.6
0.5 1.5
0.1 0.5
4.8 11.0
Cumene and naphthalene were also identified in some of
the data points in small quantities.  They are not shown
as their addition does not significantly change the
analysis.

The total  HAP ratios shown in the table are not simply
sums of the individual HAPs.   Total HAPs were calculated
for each individual sample in the data base and the values
represented in the table reflect the maximum, minimum, and
arithmetic average total HAPs of these samples.
                           C-ll

-------
C.2  REFORMULATED AND OXYGENATED GASOLINES
     Title II of the 1990 Clean Air Act Amendments addresses
emission standards for mobile sources.  There are several
elements in Title II that will affect gasoline composition
in the 1998 base year, and thus affect HAP emissions from
gasoline storage and transfer operations.
     Section 219 of Title II amends the 1977 CAA by adding
Section 211(k).  This section requires reformulated gasoline
in nonattainment areas with a 1980 population greater than
250,000 (a total of nine cities with the worst ozone
problems).  All other ozone nonattainment areas can "opt-in"
to the program regardless of 1980 population.  Beginning in
1995, "reformulated" gasoline must be sold and marketed in
these nonattainment areas with the following limits:
1) benzene content cannot exceed 1 percent, 2) no heavy
metals present, and 3) minimum oxygen content of 2.0
percent.  Additionally the more stringent of the Formula
Standard concerning aromatics (level of 25 percent or the
Performance Standards concerned with VOC or toxic emissions
(15 percent reduction from emissions using a 1990 baseline
fuel) shall also apply.
     Section 211(m) requires the purchasing and selling of
fuels with higher levels of alcohols or oxygenates in the
winter months in the areas exceeding the CO standard.
Beginning in 1992, these "oxygenated" fuels must have at
least 2.7 percent oxygen.
     The reformulated gasoline requirements will cause
reductions in the benzene and aromatic contents of the fuel
sold in these areas classified as nonattainment.  Since many
of the HAPs in gasoline vapor are aromatic compounds, this
alone would reduce the total HAP content of the gasoline
liquid and vapors.  However, the addition of oxygen
containing compounds to both reformulated and oxygenated
gasoline will significantly increase the HAP content, all
other things being equal.  Therefore, these measures will
alter the HAP content, but in opposite directions.
                          C-12

-------
     Methyl tert-butyl ether, or MTBE, is a major source of
oxygen that will be added to gasoline by the petroleum
industry to meet these requirements.  MTBE is also listed in
the CAAA as a HAP.  Traditionally, MTBE has been used as an
octane booster in unleaded gasolines.  If the octane was
lower than expected, small allotments of MTBE would be added
to reach the desired octane level.  MTBE has many advantages
as an octane enhancer.  It has a high average blending
octane rating, dissolves easily in the refinery streams, and
will not precipitate out of solution when it comes into
contact with water.  Therefore, the quantity of gasoline in
the nation which contains some MTBE is quite large, although
the MTBE content is very low.  If fact, none of the data
received for normal gasoline showed measurable levels of
MTBE.  There were four samples that contained MTBE but these
were intentionally spiked during laboratory analyses to
estimate reformulated gasoline percentages.
     It is expected that MTBE will be the most common
oxygenate used to meet the oxygen requirements.  Other
octane boosters/ oxygenates in use are ethanol 113, ethyl
tert-butyl ether (ETBE), and tertiary amyl methyl ether
(TAME).  ETBE has a lower RVP (3-5) compared to MTBE (8) and
its blending octane rating is also higher.  However, there
are limits on ETBE and the other blending agents which will
keep MTBE in the forefront.  Ethanol 113 is not economical
without government subsidies and ETBE is similarly affected
since ethanol feedstock is needed to produce ETBE.  There-
fore, the amount of ethanol and ETBE available will always
be limited by government subsidies.  The lack of isoamylene
feedstock will limit the use of TAME as well.
     It requires approximately 15 volume percent of MTBE in
liquid gasoline to meet the 2.7 weight percent oxygen limit,
and 11 volume percent to meet the 2.0 weight percent oxygen
limit.  The effects of these large percentages in liquid
gasoline are significant.  The moderate volatility of MTBE
would cause high concentrations in the vapor phase relative
                          C-13

-------
to the less volatile aromatics.  It is therefore expected
that the inclusion of MTBE in these percentages may increase
the HAP/VOC ratio in gasoline vapor from approximately
5 weight percent to near 15 percent, with liquid
concentrations of MTBE in the 15 percent range.
     The drastic differences in the HAP content of gasoline
vapor (depending on the type of fuel) necessitate the
estimation of vapor phase composition (HAP to VOC ratios)
for several different scenarios.  There will be four basic
types of fuels in use after full implementation of these
programs.  These are 1) normal fuels (ozone and CO
attainment areas and those ozone nonattainment areas not
opting into the reformulated program), 2) oxygenated fuels
(CO nonattainment areas), 3) reformulated fuels (ozone
nonattainment areas in the reformulated program),  and
4) reformulated fuels with 2.7 percent oxygen, or
reformulated and oxygenated (CO and ozone nonattainment
areas that are in the reformulated program).
     Therefore, HAP to VOC ratios were developed for each of
these fuels.  The situation is further complicated by the
fact that two different ratios are required for
reformulated, oxygenated, and reformulated/oxygenated fuels
to account for MTBE.  This results in a total of seven
different HAP vapor profiles as shown in Table C-5.  As
discussed in Section 3.3 on baseline emissions, these
profiles are used throughput the analysis.
     Since these programs are not in effect at this time,
HAP to VOC ratios were mathematically developed using the
arithmetic average vapor profile for normal fuel as the
starting point.  For reformulated fuel, the benzene content
in the vapor was calculated based on a 1.0 percent content
in the liquid.  This was calculated using the equation from
EPA7s 1984 study, "Evaluation of Air Pollution Regulatory
Strategies for Gasoline Marketing Industry", EPA-450/3-84-
012a (page 2-5).  This equation coupled with the VOC
emission rate equation predicted that the vapor phase
                          C-14

-------




w
H
W
>*
J
*-»
S3 -P
< ,C
tP
3 -H
H OJ
Q *
u)
W >i
W(*i
M
E>
0)
CO &
W_j
(0
J -p
H C
FK 0)
O U
OS >-i
di 0)
a
a
o u
a< o
< >
>: ^
o
4J
.
in (X
1 f£
1 r^
U K
v^-*
fr3
3
CQ
_**
£











'«
« *
II y
$ 5
1 a
§
2
- ^
! *
i *
Of "-
a


1 >u
s
x
_ 0
UJ ^ ^
II LU 4-*
- s
II — ' O
0 0)
< ^ ^
3 o e
I *** r~
u. Z
0 -=
"J .^
& *
£


1 UJ
g
^ s
• ^

1 £
0
"£ S
" £

i



_^
5



ft
^





i
: •* •* «- K •* «- «M
> «~ o •- o o o .*
*

«*^^N.^^O> ^
«-o»-ooo^ *-





•* ^ «- N. >± ^- ^
«- o •- o o o »»



>trx»-^>j^O' M
«-OT-O'C>C>^ ^2





-»•»<- K ^j ^- pj
«- 0 ^ C3 0 0 vf





^^^•N.*^*— hw ^
•- 0 v- O 0 0 CO JiJ



•O O- Ki co in ^ eo
«- ° «- C3 0 0 ^

8
CD
I
X
~ c
U ci
EN co
T g 6
4, ^ g •- « 0> =
C O OJ vt C — _i
-
o
f
2
41
»4»
**-
0
a
L.
o
8.
(D
>
V
c
'^
o
v>
to
"s
^ 
°s
a •":
4v f>
si
0)
u
i
C/)
C-15

-------
benzene to total VOC ratio would be 0.44 percent by weight.
This value was used for the vapor phase benzene content of
all reformulated and reformulate/oxygenated gasolines.
As stated above, the total aromatic content must also be
reduced for reformulated gasolines to 25 weight percent in
the liquid.  To determine the extent of reduction necessary,
a baseline aromatic content of liquid data was calculated
using data from the 1990 Motor Vehicle Manufacturers
Association (MVMA) National Fuel Survey.  The arithmetic
average aromatic content for all fuels over all times of the
year was 28.7 percent.  Using this as representative of the
average aromatic composition of gasoline, the percent
reduction needed to meet the 25 percent level was calculated
to be about 13 percent.  Therefore, all of the aromatic HAPS
(except benzene) would be reduced by this percentage.  The
resulting HAP to VOC weight percentages for toluene  (1.1 %) ,
ethyl benzene (0.1 %), and xylenes (0.4 percent) were held
constant for all reformulated or reformulated/ oxygenated
fuels.
     As discussed in Chapter 3, data were received for
gasolines containing MTBE.  For some of these samples, vapor
data and the corresponding liquid composition were
available.  Using these sample results, a ratio of liquid
content to vapor content was derived.  This ratio was then
used (at the liquid percentages of 11 and 15 percent MTBE
levels) to estimate the MTBE to VOC percentage in the vapor.
These estimates of MTBE to VOC ratios were 8.8 weight
percent for the 11 volume percent liquid and 12 weight
percent for the 15 volume percent liquid.
     The addition of these large amounts of MTBE would force
a reduction in the relative percentages of other compounds
simply due to the volume that would be occupied by the MTBE
in the liquid.  Therefore, to account for this fact, the
nonaromatic HAPs  (hexane and 2,2,4 trimethylpentane) were
reduced by 11 percent.  In order to simplify the analysis,
                         C-16

-------
it was also assumed that these same reductions would also
occur if other oxygenates were used instead of MTBE.
     The oxygenated fuel profiles were similarly developed.
When approximately 15 percent MTBE (or other oxygenate) was
added to the profile, all other components were reduced by
15 percent.  For those reformulated/oxygenated gasoline, the
benzene and aromatic levels were the same as discussed
above, and 15 percent oxygenate was used instead of 11
percent.
                         C-17

-------
                         APPENDIX D
                 BASELINE  EMISSIONS ANALYSIS

     The purpose of establishing an emissions 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.  As noted in Chapter 3, the base year
for the gasoline marketing source category was selected as
1998.
     Generally, the approach for establishing the emissions
baseline was the same for each sector of the industry.  An
important factor in the determination of baseline emissions
is the level of control that would be in effect in the
absence of any hazardous air pollution regulation.
     Due to the various types of gasolines that will be in
use in the 1998 base year, it was necessary to divide the
parameters used to estimate emissions (source population and
gasoline throughput) into groups according to the type of
fuel expected to be used.   This breakdown was made using
nonattainment area designations since this is the
determining factor for the type of fuel.
     To aid in the presentation of the above mentioned
factors, this appendix is separated into three sections.
Section D.I discusses the baseline regulatory coverage for
all States.  Section D.2 follows with a description of the
separation of gasoline throughput and source population by
nonattainment area, and Section D.3 presents the baseline
emissions calculations for the various industry sectors.
                          D-l

-------
D.I  Regulatory Coverage
     There are two basic control levels in effect in the
United States for gasoline marketing sources.  Control
techniques guideline (CTG) documents have been prepared for
bulk terminals, bulk plants, service stations (underground
tank filling), tank trucks, and storage tanks.  Also, new
source performance standards (NSPS) are applicable for new
or reconstructed bulk terminal loading racks and large
storage tanks such as those at terminals and pipeline
breakout stations.
     The purpose of the CTG documents is to outline what the
EPA defines as the presumptive norm for reasonably available
control technology (RACT) for existing sources.  Some of the
recommendations are in the form of emission limits and
others are in the form of recommended control equipment to
be installed.  States with nonattainment areas for ozone are
required to adopt regulations consistent with these CTG
recommendations to provide for attainment of the national
ambient air quality standards (NAAQS).  The NSPS are
national standards regulating new or reconstructed sources
of criteria pollutants, including ozone (VOC sources).
     To estimate how the States have implemented the CTG
recommendations, State regulations were reviewed for Stage I
gasoline marketing sources.  The results of this survey were
used to estimate the affected gasoline throughput on a
State-by-State basis.  In instances where regulations
covered an entire State, it was assumed that all throughput
for the State was covered by the regulation.  Base year 1998
State gasoline throughputs were determined as follows.  The
State and national 1990 gasoline throughputs were obtained
from the 1991 National Petroleum News  (NPN) Factbook issue.
The ratio of the 1998 national throughput discussed in
Section 8.1 to the 1990 national throughput from NPN was
determined and multiplied by the 1990 throughputs for each
State to obtain 1998 State gasoline throughput.
     However, many States have regulations that cover only
ozone nonattainment areas.  For these States, the counties
                          D-2

-------
that were covered were determined and the percentage of
county throughput to State throughput was calculated using
1985 NEDS gasoline consumption.  While these throughputs may
not be applicable to the base year 1998, it was assumed that
the relative county to State throughput percentages were
acceptable approximations.  Estimates were made regarding
the percentage of the throughput and/or source population
affected by NSPS regulations.
     The following paragraphs address the CTG and NSPS
control levels and the penetration of standards throughout
the nation.  The areas discussed are bulk terminal loading
racks, storage tanks, bulk plants, tank trucks, and service
stations (storage tank filling).  While there are
regulations for similar applications for the control of
fugitive emissions from leaking pumps and valves, there are
no regulations that specifically address these components
for pipeline facilities, bulk terminals, and bulk plants
(although a few bulk terminals apparently practice leak
detection and repair).  Therefore, for the purposes of this
analysis, it is assumed that all fugitive emissions at
gasoline marketing sources are uncontrolled.
D.I.I  Bulk Terminal Loading Racks
     There is both a CTG and an NSPS regulation for loading
racks at bulk terminals.  The recommended CTG level of
control is 80 mg VOC/liter of gasoline loaded.  This limit
is based on submerged fill and vapor recovery/control
systems.  The CTG also recommends that no leaks be allowed
in the vapor collection system during operation.  The NSPS
level is similar, except that the numerical limit is 35 mg
total organic compounds (TOC)/liter.  State regulations were
reviewed to determine the requirements for bulk terminals.
Table D-l lists the States that have implemented
requirements for bulk terminals.  The States listed in the
first column require that all terminals within their
boundaries achieve a level of control consistent with the
CTG (80 mg/1).   The second column includes States that
require controls consistent with the CTG only for areas
                          D-3

-------
within the States that do not meet the ozone NAAQS
(nonattairunent areas).
     An earlier study indicated that approximately 60
percent of the systems installed for the purpose of meeting
the 80 mg/1 limit routinely operate at the NSPS level of 35
mg/1.  In conversations with equipment manufacturers in
1991, it was determined that control devices are no longer
manufactured to meet 80 mg/1, but are typically designed to
meet 35 mg/1.  Therefore, unless otherwise specified, it was
assumed that 60 percent of the terminals in the controlled
areas listed in Table D-l are operating at 35 mg/1, with the
remainder operating at 80 mg/1 (or 90 percent control in one
instance).  This 60 percent includes those new or
reconstructed terminals that are required to meet the NSPS
level.  In addition, two districts in California (Bay Area
and Sacramento) have loading rack emission limitations
equivalent to 10 mg/1.  Test data indicate that many
terminals are operating at levels considerably below 10 mg/1
(see Section 4.1.2.3).
     Therefore, there are four basic control levels.  These
are 10 mg/1, 35 mg/1, 80 mg/1, and uncontrolled.  The
uncontrolled sources may be further divided into those
loading with submerged fill and with splash fill.  As
discussed in the 1987 Response to Public Comments document,
it is believed that 94 percent of uncontrolled terminals
load using submerged fill and 6 percent by splash fill.
These percentages were also used in this analysis.  State
gasoline throughput by control level is shown in Table D-2.
Also, Table D-3 presents nationwide parameters by control
level used in the baseline emissions analysis.
     It was assumed that the breakdown of the bulk terminal
population would be parallel to throughput.  Therefore, the
terminal population by control level shown in Table D-3 was
calculated by multiplying the percentage of throughput in
that control level category by the total nationwide terminal
population.
                          D-4

-------
            TABLE D-l.   STATE REGULATORY COVERAGE
                 FOR BULK GASOLINE TERMINALS
 Entire State Consistent
 With CTG Controls*
                         CTG Controls*
                         Nonatta iranent
                         Areas Only
No Control
Regulations'1
 Alabama
 California
 Connecticut
 District of Columbia
 Illinois
 Kentucky
 Louisiana
 Maine
 Massachusetts
 Michigan
 New Hampshire
 New Jersey
 North Carolina
 Pennsylvania
 Rhode Island
 South Carolina
 Tennessee
 Wisconsin
                         Arkansas
                         Colorado
                         Delaware
                         Florida
                         Georgia
                         Indiana
                         Kansas
                         Maryland
                         Missouri
                         Nevadab
                         New Mexico
                         New York
                         Ohio
                         Oklahoma13
                         Oregon
                         Texas
                         Utah
                         Virginia
                         Vermont
                         Washington
                         West Virginia
Alaska
Arizona
Hawaii
Idaho
Iowa
Minnesota
Mississippi
Montana
Nebraska
North Dakota0
South Dakota
Wyoming
a
b
CTG Controls =80 mg/liter standard or lower.
Portion of State not covered by CTG controls  is covered
by submerged fill requirements.
North Dakota has no nonattainment areas for ozone, but
the entire State is covered by submerged  fill
requirements.
Approximately 94 percent of total throughput  is loaded
by submerged fill.
                          D-5

-------
TABLE D-2.  STATE BULK TERMINAL THROUGHPUT BY
          LOADING RACK CONTROL LEVEL3
             (1,000 gallons/year)
STATE
ALABAMA
ALASKA
AIIZONA
ARKANSAS
CALIFORNIA
COLORADO
CONNECTICUT
DELAWARE
DISTRICT OF COL.
FLORIDA
GEORGIA
HAWAII
IDAHO
ILLINOIS
INDIANA
I QUA
KANSAS
KENTUCKY
LOUISIANA
MAINE
MARYLAND
MASSACHUSETTS
MICHIGAN
MINNESOTA
MISSISSIPPI
MISSOURI
MONTANA
NEBRASKA
NEVADA
NEW HAMPSHIRE
BO ng/l 90
858.258
0
390,520
9.053
4,038,743
338,180
585,145
140,460
0
1,181,764
622.024
0
0
2,114,729
490,485
0
111,405
749,042
819,406
160,852
755,437
985,152
979.093
0
10,241
572,469
0
0
0
146,601
X control
0
0
0
0
0
0
0
0
71,155
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
35 «0/l
1,287,387
27,739
657,992
139.262
6,058,115
579,290
877,717
210,690
106,733
2,105,803
1,138,936
39,339
49,751
3,172,093
885,944
139,287
265,854
1,123,562
1,229,109
262,931
1,162,575
1,477,728
1,666,167
210,227
140,811
994,106
44,963
80,497
65,956
234,871
10 MB/I
0
0
0
0
3,365.619
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
UNCONTROLLED
0
249.652
649,906
1.131,139
0
648,179
0
0
0
2,998,412
1,853,104
354,050
447,756
0
1,351,945
1,253,582
' 888.711
0
0
194,878
264,777
0
1.777,741
1,892.045
1,129,045
1,218,620
404,667
724,472
593,608
134,728
                   D-6

-------
           TABLE D-2.  (Concluded)
  STATE
80 mg/l  90 X control  35 "8/1
10 MB/1  UNOMTROUfD
NEU JERSEY
NEW MEXICO
NEU 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

1,435,664
0
1,664,553
1.350,866
0
1,690,480
110,902
221,246
1,916,045
154,234
654,910
0
1,057,880
1,683,407
155,837
0
1,225,531
46,777
90,751
859,352
0
30,377,488
26X
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
71.155
01
2,153,497
82,107
2.699,889
2.026,298
35,639
2.696,532
311.912
414,836
2.874,067
231,351
982,364
39.858
1.586,820
3,000,737
269,103
29,410
1,838,296
292,325
197,961
1,289,027
26,523
49,513,986
42X
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
3,365,619
3X
0
738,965
1,827,538
0
320,747
1.447,300
1.310,030
746,705
0
0
0
358,720
0
4,280.640
318.131
264,686
0
1,999,501
556,513
0
238,705
34,569,200
29X
The control  levels represent the emission level.
As an example,  it is assumed that 49,513,986
thousand gallons per year of gasoline passes
through terminals emitting VOCs at approximately
35 mg/liter  of  throughput.
                    D-7

-------
       TABLE D-3.  NATIONWIDE BULK TERMINAL LOADING RACK
              BASELINE PARAMETERS BY CONTROL  LEVEL
       Control Level
  Annual    Percent of
Throughput     Total      Number of
  liters)
Throughput  Facilities
10 mg VOC/liter
35 mg VOC/liter
80 mg VOC/liter and 90
percent control
Submerged filling only
Splash filling
 13,000
187,000
115,000

123,000
  8,000
    3%
   42%
   26%

   27%
    2%
 29
430
265

282
 18
                              D-8

-------
D.I.2  Storage Tanks
     There are CTG documents for petroleum liquid storage in
fixed-roof tanks and external floating roof tanks, and NSPS
regulations covering fixed-roof and external floating roof
petroleum liquid storage tanks.  The CTGs recommend the
installation of internal floating roofs on fixed-roof tanks
and a continuous primary seal on external floating roofs.
There are several NSPS standards (Subparts K, Ka, and Kb)
for storage tanks with varying control level requirements.
However, in order to simplify this analysis, it was assumed
that the NSPS level of control of storage tanks was internal
floating roofs for fixed-roof tanks, and primary and
secondary seals for external floating roof tanks.  A review
of State regulations revealed that most States regulate
emissions from storage tanks in their State implementation
plans (SIPs) with CTG recommended controls.  Based on
information contained in an earlier tank survey and the
results of this review of State regulations, the following
assumptions were made.
     In attainment areas with no storage tank regulations,
10 percent of the tanks would be external floating roof
tanks subject to NSPS and have primary and secondary seals,
with an additional 47 percent having external floating roofs
with primary seals.  The remaining 43 percent were assumed
to be fixed-roof tanks, with 16 percent having internal
floating roofs and the remaining 27 percent having no
controls.
     Many areas require the CTG level of control for fixed-
roof tanks and primary seals on external floating roof
tanks.  For these areas, it was assumed that 78 percent of
the tanks were external floating roof tanks, with 10 percent
subject to NSPS and having secondary seals in addition to
the primary seals and the remaining 68 percent being
external floating roof tanks with primary seals.  The
remaining 22 percent were assumed to be fixed-roof tanks
with internal floating roofs.
                          D-9

-------
     Finally, there are areas where both primary and
secondary seals are required.  For these areas,  it was
assumed that 75 percent of these tanks were external
floating roof tanks and 25 percent fixed-roof tanks with
internal floating roofs.
     Working losses for both fixed-roof and external
floating roof storage tanks are a function of gasoline
throughput, and not the storage tank population.  Storage
tank throughputs were estimated for each of the control
levels.  However, these throughputs were arrived at in
different fashions for bulk terminal storage tanks and
pipeline breakout station storage tanks.  The following
describes in more detail how the storage tank populations
and throughputs were derived.
     D.I.2.1  Pipeline Breakout Station Storage Tanks.
     As discussed in Chapter 8, the total nationwide
population of breakout stations was estimated by counting
observances of pipeline branches and diameter changes across
the country.  These branches and diameter changes were noted
by State.  The total number of breakout stations by State
was then placed in the appropriate control level as
discussed above.  This is shown in Table D-4.  Assuming an
average of four "equivalent dedicated storage tanks"  (see
Chapter 5) per breakout station, the nationwide breakout
station storage tank total (for emissions purposes) was
calculated by control level.  This calculated to a total of
748 external floating roof tanks, with 476 having primary
seals and 272 having primary and secondary seals.  It was
also estimated that there were 231 fixed-roof tanks, with 88
having internal floating roofs and 143 being uncontrolled.
     The throughput by control level was calculated assuming
that each tank had a storage capacity of 50,000 bbls with
150 turnovers per year, for an annual throughput of
315,000,000 gallons.  This individual tank throughput was
multiplied by the number of tanks in each control level to
give the throughput.
                          D-10

-------
TABLE D-4.  PIPELINE  BREAKOUT STATION POPULATION BY STATE
         SEPARATED BY STORAGE TANK  CONTROL LEVEL3
STATE
ALABAMA
ALASKA
ARIZONA
ARKANSAS
CALIFORNIA
COLORADO
CONNECTICUT
DELAWARE
DISTRICT OF COL.
FLORIDA
GEORGIA
HAWAII
IDAHO
ILLINOIS
INDIANA
I QUA
KANSAS
KENTUCKY
LOUISIANA
MAINE
MARYLAND
MASSACHUSETTS
MICHIGAN
MINNESOTA
MISSISSIPPI
MISSOURI
MONTANA
NEBRASKA
NEVADA
NEW HAMPSHIRE
STORAGE TANK CONTROL LEVEL
Total Nunber Primry Seal Secondary Seal
of Stations Areas Areas
I 4
0
10
3 3
10 10
2 2
1 1
0
0 4
4 3
8 3
0
3 3
17 17
11 11
11
15 1
0
13 13
0
3 3
3 2
7 7
11 11
2
10
4
4
2 2
0
Uncontrolled


10






1
5




11
10




1


2
10
4
4


                         D-ll

-------
              TABLE D-4.  (Concluded)
STATE
NEW JERSEY
NEW MEXICO
NEW YORK
NORTN CAROLINA
NORTH DAKOTA
OHIO
OKLAHOMA
OREGON
PENNSYLVANIA
RHODE ISLAND
SOUTH CAROLINA
SOUTH DAKOTA
TENNESSEE
TEXAS
UTAH
VERMONT
VIRGINIA
WASHINGTON
WEST VIRGINIA
WISCONSIN
WYOMING
STORAGE TANK CONTROL LEVEL
Total Nutfcer Primary Seal Secondary Seat
of Stations Area* Areat Uncontrolled
2 2
4 4
8
4 4
2
13 5
7 3
4 1
17 17
0
0
7
2 2
27 32
2 2
0
9 1
8
0
1 1
2
 NATIONWIDE TOTALS
277
                             83
62
132
                           30. OX
                  22. 4X
         47. TH
The storage tank control  levels  shown in the column
heading are defined as follows:
-   Primary seal areas are those areas that require
primary seals only on external floating roof tanks and
internal floating roofs on fixed-roof tanks.
    Secondary seal areas are those areas that reguire
primary and secondary seals on external floating roof
tanks  and internal floating roofs on fixed-roof tanks.
-   Uncontrolled areas are those areas that do not have
any storage tank emission control regulations.
                      D-12

-------
     D.I.2.2  Bu^.k Terminal Storage Tanks.  The bulk
terminal storage tank population and throughput were arrived
at in a different manner from the breakout station
parameters discussed above.  The initial step was to divide
each State's gasoline throughput into the various control
levels applicable to the particular State.  State gasoline
throughput by control level for bulk terminal storage tanks
is shown in Table D-5.  The number of tanks per State was
calculated the same for each control level using the
following relationship:
State capacity (bbl)     =    State Throughput fbbll
                              Number of Turnovers/year

Number of Tanks/State    =       State Capacity fbbl)	
                              Storage Tank Capacity  (bbl)
Storage tank capacities of 36,000 bbl and 16,750 bbl were
assumed for floating roof and fixed-roof storage tanks,
respectively, and 13 turnovers per year per tank.  Baseline
parameters for bulk terminal storage tanks are presented in
Table D-6.
D.I.3  Bulk Plants
     The CTG for bulk plants contains recommended control
alternatives of 1) submerged fill of outgoing tank trucks,
2) submerged fill of outgoing tank trucks and vapor balance
for incoming transfer, and 3) submerged fill and vapor
balance for outgoing and incoming transfer.  The CTG
discusses exemptions from vapor balance on outgoing loads at
bulk plants with daily throughputs of less than 4,000
gallons.
     A review of all State regulations was also conducted to
determine the regulatory coverage for bulk plants.  States
commonly responded to the recommended CTG alternatives by
selecting Alternative 3 as the control level.  However, some
State regulations include an exemption from vapor balance
for those plants with daily throughputs less than 4,000
gallons, requiring only submerged fill on outgoing
transfers.  Table D-7 shows a summary of State bulk plant
                         D-13

-------
TABLE D-5.  STATE  BULK TERMINAL THROUGHPUT
           BY STORAGE TANK TYPE3
STATE
ALABAMA
ALASKA
ARIZONA
ARKANSAS
CALIFORNIA
COLORADO
CONNECTICUT
DELAWARE
DISTRICT OF COL.
FLORIDA
GEORGIA
HAWAII
IDAHO
ILLINOIS
INDIANA
IOWA
KANSAS
KENTUCKY
LOUISIANA
MAINE
MARYLAND
MASSACHUSETTS
MICHIGAN
MINNESOTA
MISSISSIPPI
MISSOURI
MONTANA
NEBRASKA
NEVADA
PRIMARY
SEALS
34,484
3.1Z1
23,773
0
0
8,042
23,510
5,643
0
85,476
47.522
6,322
7,996
0
34,762
15,670
11,539
30,095
0
9,9;3
31,172
39,582
71,064
35,787
14,401
21,871
5,058
9,056
10,600
THROUGHPUT IT TANK TYPE BY STATE
(10*3 BBL/yr)
SECONDARY FIXED WITH UNCONTROLLED
SEALS INTERNAL FIXED
5,109
660
4,044
22,847
240,401
16,895
3,483
836
3,177
14,967
8,605
937
1,185
94,408
6,496
3,316
6,733
4,459
36,581
1,473
5,197
5,864
10.531
5,005
3.048
19,649
1,071
1,917
1,570
11,495
1,040
7,925
7,616
80,134
7,745
7,837
1,881
1,059
28,492
15,841
2,107
2,665
31,469
11,587
5,223
5,277
10,032
12,194
3,314
10,391
13,194
23,695
11,262
4,800
12,297
1,686
3,019
3,533
0
1.783
4,695
0
0
4,595
0
0
0
20,731
14,081
0
0
0
12,116
8,954
6,594
0
0
0
5,211
0
0
0
8,229
12,498
2,890
5,175
0
                 D-14

-------
              TABLE D-5.  (Concluded)
                         THROUGHPUT BY TANK TYPE BY STATE

                              (t
-------
         TABLE D-6.  BASELINE PARAMETERS FOR BULK
                  TERMINAL STORAGE TANKS
     Control Level
 Annual   Percent  Number  Percent
Thruput     of       of       of
  (106    Thruput  Tanks    Tanks
 bbls)
External Floatincr Roof
Tanks
     with Primary
     Seals

     with Primary and
     Secondary Seals
Fixed-Roof Tanks

     with Internal
     Floating Roofs

     Uncontrolled
1,135
  843
  595

  234
40%   2,426     57%



30%   If802     £3%


      4,228    100%





21%   2,732     72%


 8%   1.072     28%


      3,804 	100%
                         D-16

-------
   TABLE D-7.   STATE REGULATORY COVERAGE FOR BULK PLANTS
 Entire State Consistent
 With CTG Controls*
CTG Controls*
Nonatta inment
Areas Only
No Control
Regulations13
 Alabama
 California0
 Connecticut
 District of Columbia
 Illinois
 Kentucky0
 Louisiana0
 Massachusetts
 Michigan
 New Jersey
 North Carolina0
 Pennsylvania0
 Rhode Island0
 South Carolina0
 Tennessee
 Virginia0
 Wisconsin
Arkansas
Colorado
Delaware0
Georgia
Indiana0
Maryland0
Missouri0
Nevada
New York0
Ohio
Oregon
Texas0
Utah0
Washington
Alaska
Arizona
Florida
Hawaii
Idaho
Iowa
Kansas
Maine
Minnesota
Mississippii
Montana
Nebrasksa
New Hampshire
New Mexico
North Dakota
Oklahoma
South Dakota
Vermont
West Virginia
Wyoming
*CTG recommendations include the use of vapor balance,
 submerged fill, and pressure relief settings for storage
 tanks, and vapor balance for the loading racks.
bLoadings assumed to be 25 percent splash fill and 75
 percent submerged fill at loading racks, unless otherwise
 specified.
Regulations require vapor balance on all outgoing
 transfers.  All other areas with CTG regulations exempt
 plants with daily throughputs less than 4,000 gallons/day
 from installing vapor balance equipment.
                         D-17

-------
regulations in a manner similar to the bulk terminal table
shown earlier.
     Bulk plants are intermediate storage and distribution
facilities.  Therefore, all of the gasoline throughput for
an area does not pass through a bulk plant.  In order to
estimate emissions from bulk plants, the throughput that
travels through bulk plants was a necessary parameter.
Information contained in the 1987 Census of Wholesale Trade
was used to estimate the bulk plant throughput on an
individual State basis.  The State throughput for bulk
stations contained in the Census information was divided by
the total State throughput to obtain an estimate of the
percentage for bulk plants.  These percentages were applied
to the estimated 1998 State throughput to calculate baseline
bulk plant throughput.  This is shown in Table D-8.
     This throughput was then separated by State by control
level.  The four basic control levels were 1) vapor balance
on incoming and outgoing loading operations with no
exemptions, 2) vapor balance on incoming and outgoing
loading operations with submerged fill requirements for bulk
plants with throughputs less than 4,000 gallons per day, 3)
vapor balance on incoming loads with submerged fill only on
outgoing loads, and 4) no controls.  The throughput by State
by control level is shown in Table D-9.  The uncontrolled
throughput was further divided into splash and submerged
fill.  It was assumed that 75 percent of the uncontrolled
plants load using submerged fill and 25 percent using splash
fill.  Table D-10 presents national parameters used in the
baseline emissions analysis for bulk plants.
     The populations in Table D-10 were basically derived
using the throughput breakdowns by control level and
applying those to the bulk plant population provided in
Section 8.2.  This was done except in the instance of
aviation bulk plants.  All of these were assumed to be
uncontrolled with the percentage loading by submerged fill
the same as for motor gasoline.
                          D-18

-------
TABLE D-8.  BULK PLANT THROUGHPUT BY STATE
            (1,000 gallons/year)
STATE
ALABAMA
ALASKA
ARIZONA
ARKANSAS
CALIFORNIA
COLORADO
CONNECTICUT
DELAWARE
DISTRICT OF COL.
FLORIDA
GEORGIA
HAWAII
IDAHO
ILLINOIS
INDIANA
I QUA
KANSAS
KENTUCKY
LOUISIANA
MAINE
MARYLAND
MASSACHUSETTS
MICHIGAN
MINNESOTA
MISSISSIPPI
MISSOURI
MONTANA
NEBRASKA
NEVADA
NEW HAMPSHIRE
NEW JERSEY
1998
TOTAL X
THROUGHPUT P
2, US, 645
277.391
1,698,418
1,279,454
13,462,477
1,565,650
1,462.862
351,150
177,888
6,285,978
3,614,063
393,389
497,506
5,286,822
2.728,374
1,392.869
1,265,970
1,872,604
2,048,515
618,660
2,182,788
2,462.880
4,423,002
2,102,272
1,280,097
2,785,195
449,630
804,969
659,565
516,200
3,589,161
; THRU i
LANTS 1
23X
19X
24X
33X
18*
42X
6X
68X
18X
12X
30X
3X
37X
18X
21X
36X
53X
28X
37X
25X
10X
9X
12X
24X
43X
30X
18X
56X
4X
66X
5X
ttJLK PLANT
HMUGHPUT
493,498
52.704
407,620
422,220
2,423,246
657,573
87,772
238,782
32,020
754,317
1,084,219
11,802
184,077
951,628
572,959
501,433
670,964
524,329
757,951
154,665
218,279
221,659
530,760
504,545
550,442
835,559
80,933
450,783
26,383
340,692
179,458
                 D-19

-------
TABLE D-8.  (Concluded)
STATE
NEW MEXICO
MEW 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
1998
TOTAL X THRU BULK PLANT
THROUGHPUT PLAHTS THROUGHPUT
821,073
6,191 ,9T9
3,377,164
356,386
5,834,312
1,732,844
1,382.787
4,790,112
385,586
1.637.274
398.577
2,644,699
8,964,784
743,071
294,095
3,063,827
2.338,598
845,225
2,148,379
265,228
117,897,448
37X
7X
26X
31X
ax
41X
25X
13X
3X
18X
18X
18X
17X
18X
52X
13X
15X
34X
21X
43X
20£
303,797
433,439
878,063
110,480
466,745
710.466
345.697
622.715
11,568
294,709
71.744
476,046
1.524,013
133,753
152.929
398.297
350,790
287.377
451,160
114,048
23,061,106
        D-20

-------
TABLE D-9.  STATE BULK PLANT THROUGHPUT BY CONTROL LEVEL"
                   (1,000 gallons/year)
STATE
ALABAMA
ALASKA
ARIZONA
ARKANSAS
CALIFORNIA
COLORADO
CONNECTICUT
DELAWARE
DISTRICT OF COL.
FLORIDA
GEORGIA
HAWAII
IDAHO
ILLINOIS
INDIANA
IOUA
KANSAS
KENTUCKY
LOUISIANA
MAINE
MARYLAND
MASSACHUSETTS
MICHIGAN
MINNESOTA
MISSISSIPPI
MISSOURI
MONTANA
NEBRASKA
NEVADA
NEU HAMPSHIRE
VAPOR BALANCE
NO EXEMPTIONS
0
0
0
0
2,423,246
0
0
238.782
32,020
0
0
0
0
0
257,505
0
0
524,329
757.951
0
188,859
0
0
0
0
429,352
0
0
0
0
VAPOR BALANCE
WITH EXEMPTIONS
493,498
0
234,312
7.469
0
355,089
87,772
0
0
354,529
466,518
0
0
951,628
0
0
147,612
0
0
100,532
0
0
293,728
0
11,009
0
0
0
0
241,891
VAPOR BALANCE IN
SUBMERG FILL OUT
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
221,659
0
0
0
0
0
0
0
0
UNCONTROLLED
0
52.704
173,308
414,751
0
302,484
0
0
0
399,788
617,701
11,802
184,077
0
315,454
501,433
523,352
0
0
54,133
29,420
0
237,032
504,545
539,433
406,207
80,933
450,783
26,383
98,801
                        D-21

-------
               TABLE D-9.  (Concluded)
STATE
NEW JERSEY
NEW HEX ICO
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

VAPOR BALANCE
NO EXEMPTIONS
0
0
291,297
878,063
0
0
0
0
622,715
11,568
294,709
0
0
715.448
70,127
0
398,297
0
0
0
0
8,134,266
35%
VAPOR BALANCE
UITH EXEMPTIONS
179.458
0
0
0
0
338,096
113,675
138.279
0
0
0
0
476,046
0
0
0
0
17,539
77,138
451,160
0
5,536,979
24 X
VAPOR BALANCE IN
SUBMERG FILL OUT
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
221,659
1%
UNCONTROLLED
0
303.797
142,142
0
110,480
128.649
596.792
207,418
0
0
0
71,744
0
808,565
63,626
152,929
0
333.250
210,238
0
114,048
9,168,201
40!L
a VAPOR BALANCE NO EXEMPTIONS refers to those areas that
  have regulations requiring vapor balance on the
  incoming side for all bulk plants, regardless of
  throughput.  VAPOR BALANCE WITH EXEMPTIONS refers to
  those areas that require vapor balance on the incoming
  side for all bulk plants, and vapor balance on the
  outgoing side for all plants with daily throughputs
  below this level.  VAPOR BALANCE IN SUBMERG FILL OUT
  denotes the areas that require vapor balance on
  incoming loads, but only submerged fill on outgoing
  loads.  UNCONTROLLED refers to those areas without any
  emission regulations covering bulk plants.
                       D-22

-------
        TABLE D-10.   BASELINE PARAMETERS  FOR BULK PLANTS
       Control Level
  Annual
Throughput
   (106
  liters)
Percent of
  Total      Number of
Throughput   Facilities
Vapor balance incoming and    30,791
outgoing load, no
exemptions


Vapor balance incoming and    20,960
outgoing load, submerged
fill on outgoing loads at
plants
< 4,000 gal/day
                 35%
                 24%
               3,315
               2,256
Vapor balance incoming,
submerged fill outgoing
    839
      1%
   90
Submerged fill incoming
and outgoing
     Motor vehicle
     gasoline
     Aviation gasoline
 26,029
     30%
5,202

2,802
2,400
Submerged fill incoming
and splash fill outgoing
  Motor vehicle gasoline
  Aviation gasoline
  8,676
     10%
1,734


  934
  800
                              D-23

-------
D.I.4  Tank Trucks
     In determining baseline regulatory coverage for tank
trucks, two cases were considered:  trucks in "normal"
service and trucks in "collection" service (i.e., trucks
equipped with vapor collection equipment).  Normal service
pertains to areas where no controls (or only submerged fill)
are required at the terminal or bulk plant.  In this
situation there are no collection systems; therefore, there
can be no leakage of vapors from the vapor collection system
or the truck tank.  "Collection" service pertains to loading
when vapor balance systems are employed.  For areas where
vapor balance systems are used, the CTG recommendation is to
have vapor-tight tank trucks.  The CTG recommendations for
vapor-tight tank trucks are that 1) the tank truck must pass
an annual leak-tight test that requires it to have less than
3" H20 pressure change under 18" H2O pressure or 6" H2O
vacuum, 2) it have no leaks greater than 100 percent of the
lower explosive limit (LEL) when monitored at any time with
a portable combustible gas analyzer, and 3) the vapor
collection system backpressure not exceed 18" H20 when
measured at the truck.
     In addition to the CTG level, many districts in the
State of California require an annual vapor tightness test
with less than 1" or 2" H2O pressure change rather than the
CTG recommendation of 3" H20.  In addition to this
difference, there are enforcement programs in California
that actively monitor trucks using portable gas analyzers or
equivalent methods.  The combination of this more stringent
test and increased enforcement results in a control level
slightly more effective than the CTG level.
     It was assumed in this analysis that all areas
requiring vapor collection and control at terminal loading
racks require that tank trucks be vapor-tight.  It was also
assumed that all areas requiring vapor balance for the
outgoing truck loading racks at bulk plants require that
bulk tank trucks be vapor-tight.
                          D-24

-------
     Emissions from tank truck leakage are calculated using
gasoline throughput.  Therefore, gasoline throughput was
separated into controlled and uncontrolled at bulk terminals
and bulk plants to calculate tank truck leakage emissions.
For both terminals and plants, the throughput in California
was separated into an "enhanced" truck tightness category.
     As discussed in Chapter 8, Section 8.2, the population
of tank trucks may be divided into two groups within the
overall categories of bulk plant trucks and bulk terminal
trucks.  These are private (owned by terminal or plant
owner) and for-hire.  In addition, bulk plant private trucks
may be broken down into motor vehicle gasoline trucks and
aviation gasoline trucks.  In order to estimate the number
of these trucks that already had controls installed, the
throughput percentages discussed above for bulk terminals
and bulk plants were applied to the populations of tank
trucks to estimate the number controlled and uncontrolled
(except for aviation gasoline trucks, which were all assumed
to be uncontrolled).
     Table D-ll shows the baseline gasoline throughput
percentages and populations by control level for tank
trucks.  While this represents the baseline conditions, only
the throughput is used in the emissions analysis.
D.I.5  Service Stations
     The approach for determining the regulatory coverage
for service stations was similar to that for bulk terminal
loading racks and bulk plants.  All gasoline, with the
exception of agricultural accounts, was assumed to pass
through service stations (including public and private
outlets).  The service station design criteria document
contains emission limits in terms of equipment
specifications.  Recommended controls are submerged fill of
storage tanks, vapor balance between truck and tank, and a
leak-free truck and vapor transfer system.  There are no
exemptions noted in the design criteria document.
                         D-25

-------
       TABLE  D-ll.   BASELINE PARAMETERS FOR TANK TRUCKS
                                      Percent of
                                        Total       Number of
Control Level                         Throughput      Trucks
Bulk Terminal Tank Trucks

Enhanced leak tightness                   11%          5,079
Annual leak tightness                     60%          26,090
Uncontrolled                              29%          12,731
Bulk Plant Tank Trucks

Enhanced leak tightness                   11%          4,818
Annual leak tightness                     49%          17,622
Uncontrolled                              40%          21,360

  Motor vehicle gasoline                              14,960
  Aviation gasoline                                   6,400
                              D-26

-------
     State regulations were also reviewed to determine the
regulatory coverage for storage tank filling at service
stations.  Although the design criteria document does not
contain exemptions, there are various exemption levels
contained in the state regulations.  Many of these
regulations contain exemptions with respect to tank size,
which exempts most agricultural accounts.  Other regulations
specifically exempt agricultural dispensing facilities.
Some States exempt dispensing facilities according to
monthly throughput, with the common exemption level being
38,000 liters (10,000 gallons) per month.
     For the purposes of this analysis, there were three
basic control levels selected.  These are 1) vapor balancing
with no exemptions, 2) vapor balancing with a 38,000 liters
(10,000 gallons) per month exemption, and 3) uncontrolled.
Control level 1 includes areas with no exemptions as well as
the areas with exemptions for very small tanks.  This
exemption affects very few public and private facilities
except for agricultural accounts.  Also, as with bulk
terminals and bulk plants, the uncontrolled stations are
divided into submerged and splash fill.  Unless otherwise
noted, uncontrolled throughput was split 50/50 between
submerged and splash fill.  It was assumed that all aviation
service station type facilities were uncontrolled and
operated with the same split between submerged and splash as
stated above.
     Gasoline throughput by State by control level is shown
in Table D-12.  Baseline population and throughput for
service stations is summarized in Table D-13.

D.2  BASELINE ANALYSIS OF FUEL TYPES
     As discussed in Chapter 3 and Appendix C, there are
four basic fuel types that are expected to be in use in the
base year of 1998.  These are 1) normal, 2) reformulated,
3) oxygenated, and 4) a combination of oxygenated and
reformulated.  Since HAP emissions are calculated by
multiplying the VOC emissions by a HAP to VOC ratio, the
                          D-27

-------
TABLE D-12.  STATE SERVICE STATION
   THROUGHPUT BY CONTROL  LEVEL*
       (1,000 gallons/year)
STATE
ALABAMA
ALASKA
ARIZONA
ARKANSAS
CALIFORNIA
COLORADO
CONNECTICUT
DELAWARE
DISTRICT OF COL.
FLORIDA
GEORGIA
HAWAII
IDAHO
ILLINOIS
INDIANA
IOWA
KANSAS
KENTUCKY
LOUISIANA
MAINE
MARYLAND
MASSACHUSETTS
MICHIGAN
MINNESOTA
MISSISSIPPI
MISSOURI
MONTANA
NEBRASKA
NEVADA
NEW HAMPSHIRE
NO EXEMPTIONS
2.U5.645
0
0
0
13.462,477
0
0
0
177,888
0
0
0
0
0
1,226,213
0
0
1,872,604
2,048,515
618,660
0
2,462,880
0
0
0
0
0
0
0
0
WITH EXEMPTIONS
0
0
0
22,634
0
845,451
1,462,862
351.150
0
2.954.410
1.555.059
0
0
5,286,822
0
0
278,513
0
0
0
1.888,592
0
4,423.002
0
25,602
1,431,173
0
0
0
366,502
SUBMERGED FILL
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
Q
0
0
0
0
0
0
0
UNCONTROLLED
0
277,391
1.698,418
1.256,821
0
720, 199
0
0
0
3,331.569
2.059,004
393,389
497,506
0
1,502,161
1,392,869
987,457
0
0
0
294,196
0
Q
2,102,272
1.254,495
1.354,023
449,630
804,969
659.565
149.698
              D-28

-------
              TABLE D-12.   (Concluded)
  STATE
NO EXEMPTIONS  WITH EXEMPTIONS
SUBMERGED FILL
UNCONTROLLED
NEU JERSEY
NEW MEXICO
NEU YORK
NORTH CAROLINA

OHIO
OKLAHOMA
OREGON
PENNSYLVANIA
RHODE ISLAND
SOUTH CAROLINA
SOUTH DAKOTA
TENNESSEE
TEXAS
UTAH
VERMONT
VIRGINIA
WASHINGTON
WEST VIRGINIA
WISCONSIN
WYOMING
NATIONWIDE

0
0
4,161,382
3,377,164
0
4,226,201
0
553.115
0
385,586
0
0
0
A, 208, 518
389,592
0
0
0
0
0
0
41,316,439
35%
3,589,161
0
0
0
0
0
277,255
0
4.790,112
0
392,946
0
2.644,699
0
0
0
3,063,827
116,930
245,115
2,148,379
0
38,160.196
33%
0
0
0
0
0
0
1.455.589
0
0
0
0
0
0
0
0
0
0
0
0
0
0
1,455.589
1%
0
821,073
2.030.598
0
356,386
1,606,112
0
829,672
0
0
1,244,328
398,577
0
4,756,266
353.479
294,095
0
2,221,668
600,110
0
265,228
36,965,224
31%
NO EXEMPTIONS  indicates those areas where the  service
station regulations  do not contain exemptions  related
to throughput  (i.e.,  38,000 liters/month or 10,000
gallons/month).   WITH EXEMPTIONS refers to those  areas
that do not have  exemptions based on this throughput.
SUBMERGED FILL refers to areas that require only
submerged filling of storage tanks.  UNCONTROLLED
indicates those areas without Stage I service  station
regulations.
                     D-29

-------
     TABLE D-13.  BASELINE PARAMETERS FOR SERVICE STATIONS
Control Level
Vapor balance with no exemptions
Vapor balance with submerged fill
Percent of
Total ,
Throughput
35%
32%
Number of
Stations
135,146
123,562
for stations with less than 10,000
gal/month throughput

Submerged fill                            17%         33,621
  Motor gasoline                                      32,821
  Aviation gasoline                                      800

Splash fill                               16%         30,970
  Motor gasoline                                      30,170
  Aviation gasoline          	   	  	800
                              D-30

-------
parameters used to calculate VOC emissions discussed in
Section D.I must be separated according to fuel type.  The
major criterion for this breakdown is the attainment
designation.
     Nine ozone nonattainment areas will be required to
utilize reformulated gasoline throughout the year and all
other ozone nonattainment areas may opt into this program.
Also, all CO nonattainment areas will be required to
distribute oxygenated gasoline during the winter months.
     For this baseline emissions analysis, several
assumptions were necessary.  First, the areas that will opt
into the reformulated gasoline program are not known at this
time.  It was assumed that all moderate and above ozone
nonattainment areas will opt in and utilize reformulated
gasoline.  Another separation was by time of year.  The year
was divided into the winter season (November - February) and
the nonwinter season (March - October).  The rationale for
this breakdown is that the oxygenated fuel requirements for
CO nonattainment areas apply only in the winter period,
which will affect the types of fuels used in this time
period without affecting the remainder of the year.
     Exceedances of the ambient CO standard occur during
different months, depending on the geographical location.
Therefore, the use of oxygenated fuels is not always
required during the same months for all CO nonattainment
areas.  However, in order to simplify the analysis, it was
assumed that all oxygenated fuel throughput occurs during
the months of November through February.  These are the most
common months for exceedances.
     Based on 1990 throughput as reported in the 1991
National Petroleum News Factbook, it is estimated that
approximately 68 percent of the gasoline throughput occurs
in the eight nonwinter months (March - October).  During
these months, there will be two types of fuels in use.
These are reformulated and normal gasoline.  The areas
assumed to use reformulated fuel in this analysis are
                          D-31

-------
moderate and above ozone nonattairunent areas.  All other
areas will utilize normal fuels.
     For the winter, there are a greater number of fuels
that will be used.  In areas that are moderate and above
ozone nonattainment areas and nonattainment for CO, the fuel
used will be reformulated/oxygenated (i.e., reformulated
with the higher oxygen content).  Areas nonattainment for
CO, but not also moderate or above for ozone, will utilize
oxygenated fuels.  Moderate and above ozone nonattainment
areas that are not also CO nonattainment areas will utilize
reformulated gasoline.
     In response to these situations, the percentage of
gasoline throughput for four nonattainment scenarios was
determined.  For the nonwinter period, the only necessary
breakdown was the throughput for moderate and above ozone
nonattainment areas.  In the winter, throughput percentages
were determined for moderate and above ozone nonattainment
areas that are also CO nonattainment areas, moderate and
above ozone nonattainment areas that are not also CO
nonattainment areas, and CO nonattainment areas that are not
also moderate or above ozone nonattainment areas.  These
percentages were determined using preliminary estimates of
nonattainment area designations based on 1987-89 design
values and 1988-90 design values for a few areas and the
1985 NEDS gasoline consumption report.  Table D-14 shows the
percentages of throughput by State for these nonattainment
area (and resulting fuel type) designations.
     The regulatory coverage was then applied by State for
each attainment area designation in the analysis.  An
emission factor corresponding to the regulatory coverage,
loading method, type of storage used, etc., was selected and
VOC emissions were calculated by multiplying the
corresponding throughput by the corresponding emission
factor.  The winter RVP, 14.0 psi, and nonwinter RVP,
10.2 psi, as discussed in Chapter 3, were used to calculate
separate VOC emission factors for each time period.  The
resulting VOC emissions were multiplied by the total HAP to
                          D-32

-------
TABLE D-14.
STATE GASOLINE THROUGHPUT BY NONATTAINMENT
     AREA CLASSIFICATION
STATE
ALABAMA
ALASKA
MIZONA
ARKANSAS
CALIFORNIA
COLORADO
CONNECTICUT
DELAWARE
DISTRICT OF COL.
FLORIDA
6EOR6IA
NAUAII
IDAHO
ILLINOIS
INDIANA
I DMA
KANSAS
KENTUCKY
LOUISIANA
MAINE
MARYLAND
MASSACHUSETTS
MICHIGAN
MINNESOTA
MISSISSIPPI
MISSOURI
MONTANA
NEMASKA
NEVADA
NEW HAMPSHIRE
NEW JERSEY
PERCENT PE
>MOD OZONE CO
NONAmiN DON
OX
OX
S7X
OX
94X
OX
100X
7TX
100X
31X
40X
OX
OX
61X
19X
OX
ox
26X
KX
58X
87X
100X
55X
OX
ox
s;x
ox
ox
ox
65X
96X
•CENT PER
ft >NOD CO
ATTAIN NONA
OX
OX
57X
OX
B2X
OX
86X
59X
100X
ox
Z3X
OX
ox
37X
1ZX
OX
OX
OX
OX
OX
87X
100X
39X
OX
ox
26X
OX
OX
OX
61X
97X
CENT
ONLY
ATTAIN
OX
62X
1?X
OX
\X
71X
OX
OX
OX
ox
ox
ox
ox
3TX
OX
OX
OX
OX
ox
ox
ox
ox
ox
55X
2X
OX
28X
OX
48X
OX
OX
                        D-33

-------
TABLE D-14.   (Concluded)
STATE
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
PERCENT PERCENT PERCENT
>MQO OZONE CO t >MOO CO ONLY
NONATTAIN HONATTAIN NONATTAIN
OX
49X
28X
OX
SOX
OX
OX
t>n
100X
ox
ox
16X
45X
45X
OX
13X
OX
an
35X
ox
43%
ox
49X
28X
OX
20X
OX
OX
OX
OX
OX
OX
ox
2X
OX
ox
ox
ox
ox
ox
ox
28%
2M
sx
4X
OX
IX
OX
OX
OX
ox
ox
ox
ox
ox
ox
ox
ox
ox
ox
ox
ox
5^
        D-34

-------
VOC ratio for the appropriate fuel type to obtain the total
HAP emissions.  These HAP to VOC ratios and the
corresponding attainment area situation where they were used
is summarized in Table D-15.  The following sections
describe the methodology for each of the industry sectors.

D.3  BASELINE EMISSIONS FOR INDIVIDUAL SUBCATEGORIES
     In this section, baseline emissions are presented for
the individual source subcategories within the gasoline
marketing chain.  For each subcategory, the breakdown of
parameters into the different attainment designations is
presented by control level.  The VOC emission factors used
to calculate VOC emissions are discussed, and baseline HAP
and VOC emissions are presented.
D.3.1  Pipeline Facilities
     D.3.l.l  Pipeline Pumping Stations.  Emissions from
pipeline pumping stations are attributed to fugitive
emissions from pumps and valves.  The emission factors used
for pumps and valves were taken from AP-42, Section 9.1.3
for light liquid components at refineries, 0.26 kg/valve/day
and 2.7 kg/pump seal/day.  All pipeline pumping stations are
assumed to be uncontrolled (i.e., not routinely monitoring
for liquid and vapor leaks) in the 1998 base year.  As
discussed in Chapter 8, it is estimated that at the baseline
there are 1,989 pumping stations in the United States.
Using the model plant distribution shown in Table 5-1, this
converts to a total component population of 10,600 pumps and
116,080 valves.  The nationwide VOC emissions were
calculated using these component populations.
     The types and quantity of gasoline traveling through a
pipeline will mirror the nationwide consumption.  Therefore,
the VOC emissions were separated by time of year (68 percent
during nonwinter and 32 percent during winter) and by fuel
type according to the attainment area designations shown in
Table D-14.  For example, it was assumed that about
43 percent of the nationwide throughput is in moderate and
above ozone nonattainment areas.  Therefore, 43 percent of
                         D-35

-------
H


§
H
J
CU
W
H
W
H

Q
M


§

W
U
A
H
in
H
j
Q
m

g











m-4
mw T
'"j
^rt
o
*_J
•1 1
rH
&














u
>
5}
o •w ^
•^ O 4J 4J
CL'H £ •£
JT} 4J QJ ty>
S (0 U "H
* P« M Q)

^ Qi
10 ij!
4-)
O
EH



0
04

E-i

rH
0)
e

e 10 4J 10
i c c c
3 o -H o
w c s c






CO

















«
rH
10
1
O

s

j^
O

- 0
rH C
(0 -H
O_j
r*(
•H O
04 to
?•( <0
EH O




0 0
^ ^
O O
•Q J3
<0 (0 O
to

O O (0
0 04J
•P -P O
10 10 C

0000
•O C73 C
O O O O
fiSfi§8
to to
« M 10 rl M
0 O 0 0 O
tj (*J I. |»J Uj
10 10
4J 4J4J
rH CrH C C
rH 0 rH 0 0
< g< | |

•H -H -H
•• io •• (0 (0
h 4J r( 4J *J
0 4-> 0 4-> -P
e « +j 10 10
i c c c c
3 O-H o o
w c s c c






ON ^
w +
H ^













id
0

•H
H
o
to
<0
0 H

•0 E-i
0 w fe
4^ W
10 EH 4->
H fe ^
j3 O
g 5 5
M 4^ 4"*
0 -H -H
tH > £
0


4J
O
C

to
(0
0
r«
(0

4J
C 0
i§
C N
-H O
(0
US
  CQ
10 
0 (0
« O






^
H
CQ
t^
E
C
•H
10
4J
c

4J (0

*O 0) *
Q) ^^ Q)
B >i-P
3 x^lo
to o c
to --. 0
lO *O t3^
0 >1
tO 4J X
•H <0 O
rH
•H B 0
* O 4J
tO 
4-> W
O A
ti (^ 1 i
O 0 -H
bu lj ^C


«
                                D-36

-------
the nonwinter VOC emissions were multiplied by the
reformulated vapor profiles to estimate HAP emissions.  The
baseline emissions from pipeline pumping stations are shown
in Table D-16.
     D.3.1.2  Pipeline Breakout ?tations.  There are two
sources of emissions at pipeline breakout stations.  These
are fugitive emissions from leaking pumps and valves and
emissions from gasoline storage.
     The fugitive emissions were calculated based on the
model plant information discussed in Chapter 5.  The smaller
station was assumed to have 8 "equivalent" pumps and 210
"equivalent" valves.  The larger model plant was assumed to
have 10 equivalent pumps and 300 equivalent valves.  Using
the distribution of facilities by model plant in Chapter 5,
a total nationwide component population of 69,389 equivalent
valves and 2,465 pumps was estimated.  These were multiplied
by the emission factors discussed above for pipeline pumping
stations to determine nationwide baseline VOC emissions.  It
was also assumed that throughput for breakout stations is a
representation of the nationwide throughput.  Therefore, the
VOC emissions were separated by the percentages for the time
of year and attainment area, and multiplied by the
corresponding HAP to VOC ratios to estimate baseline HAP
emissions.
     Emissions from storage tanks were calculated using the
storage tank populations and throughputs by control level
discussed in Section D.I.2.1 and multiplying these by the
VOC emission factors.  These VOC emission factors were
derived assuming an RVP of 10.2 psi for summer and 14.0 psi
for winter, and are presented in Table D-17.  The HAP
emissions were calculated using nationwide percentages of
throughput as discussed above.  Table D-18 presents baseline
storage tank and fugitive emissions from pipeline breakout
stations.
D.3.2  Bulk Terminals
     There are three basic sources of emissions at bulk
terminals.  These are loading rack emissions (which include
                         D-37

-------
TABLE D-16.  BASELINE EMISSIONS FROM
      PIPELINE  PUMPING STATIONS
Baseline
Emissions
Existing
New
TOTAL
Fugitive Emissions
(Mg/yr)
HAP
1,710
660
2,370
VOC
22,800
8,810
31,610
              D-38

-------
TABLE D-17.  EMISSION FACTORS FOR PIPELINE BREAKOUT STATION
                       STORAGE TANKS"-b
Type of Emission
         VOC
      Emission
       Factor
NonWinter   Winter
             Units
Fixed-Roof
Uncontrolled
   Breathing losses
   Working losses
 27.0
431.3
 37.7     Mg VOC/yr/tank
559.6    Mg VOC/yr/tank
Internal Floating Roofc
   Rim Seal losses
   Fitting losses
   Deck Seam losses
   Working losses

External Floating Roof
   Standing Storage
   losses
      Primary seald
      Secondary seal6
   Working losses
  1.0
  1.1
  2.3
 1.5
 1.6
 3.3
     7.33 x 10*
               8
 15.8         23.1
  7.4         10.8
     4.61 X 10'8
Mg VOC/yr/tank
Mg VOC/yr/tank
Mg VOC/yr/tank
  Mg VOC/bbl
  throughput
         Mg VOC/yr/tank
         Mg VOC/yr/tank
           Mg VOC/bbl
           throughput
   Emission factors calculated with equations  from Section 4.3 of
   AP-42 using a nonwinter RVP of 10.2 psi,  a  winter RVP of 14.0
   psi,  and a temperature of 60 *F,  as discussed in Section 3.2.1.2
   Assumes storage tanks at pipeline breakout  stations have a
   capacity of 8,000 m3  (50,000 bbl),  a diameter of  30 meters  (100
   feet),  and a height of 12 meters (40 feet).
   Assumes that internal floating roof is  equipped with a liquid-
   mounted resilient seal (primary only).
   Assumes that external floating roof is  equipped with a primary
   metallic shoe seal.
   Assumes that external floating roof is  equipped with a shoe-
   mounted secondary seal.
                            D-39

-------
TABLE D-18.  BASELINE EMISSIONS FROM
     PIPELINE BREAKOUT STATIONS
Baseline
Emissions
Existing
New
TOTAL
Storage Tank
Emissions (Mg/yr)
HAP
6,320
60
6,370
voc
83,370
740
84,110
Fugitive Emissions
(Mg/yr)
HAP
780
80
860
VOC
10,410
1,030
11,450
              D-40

-------
tank truck leakage at facilities controlled by vapor
collection), storage tank emissions, and fugitive emissions
from leaking pumps and valves.  Baseline HAP and VOC
emissions from bulk terminals are shown in Table D-19.  Each
will be addressed in the following subsections.
  D.3.2.1  Loading Rack Emissions.  The national baseline
control levels shown in Table D-3 were separated according
to the nonattainment designations shown in Table D-14.  It
was assumed that all throughput for ozone nonattainment
areas was controlled at the control level for that
particular State or part of that State.  For example, it was
estimated that 67 percent of the gasoline throughput
occurred at terminals subject to New York's 80 mg/1
standard.  It was also estimated that 49 percent of New
York's throughput occurred in moderate or above ozone
nonattainment areas.  This 49 percent of the State
throughput was assumed to all be subject to the 80 mg/1
standard and control levels set as discussed in Section D.I.
Using this approach, throughput was divided into the various
attainment designations according to control level.  Table
D-20 shows this breakdown that represents the baseline.
  Emission factors were selected for each control  level and
applied to the throughput.  The 80, 35, and 10 mg/1 emission
factors did not change from nonwinter to winter.  The
calculated emission factors for submerged fill are 667 mg/1
for the nonwinter and 860 mg/1 for the winter.  Those for
splash fill are 1,611 mg/1 for the nonwinter and 2,079 mg/1
for the winter.  Using these emission factors, the VOC
emissions for each attainment class were calculated and the
HAP emissions estimated using the appropriate emission
factors.
  Tank truck leakage emissions are  also attributed to the
loading rack since they occur in the rack area while the
truck is loading.  As noted previously, it was assumed that
all throughput controlled for loading racks was subject to
leak-tight tank truck requirements.  The three basic control
levels are annual leak tightness inspections, enhanced
                          D-41

-------
a
g

I
PM
o
H
W
W
H
S
W
S5
H
J
W
CQ
cn
H
 I
O
CQ

<3 w
^ O J^
®'m \
rt-S ^
o 6 "^"'
i) W
CO


0) (0
•H O 17
VJ-H >i
*rH (0 "^x-
&* U) &1
3-H g
bu € 
'i C) *^ «£
c j_q g ^^
g H

|.
*Js
r" W xx
.^ W &i
fr^ >r^ ^
Irt E ^"^
O



o
o
>


a,
^


Q
O
£*



Pk
ss
o
o
>



cu
^

u
o
>



cu
^
SH

0) g
<-* '-'
^•3
S.s
* e
CD

o
tH
o
o
CO
o
H
o\
«
o
"*
^
o
•*

O
CO
H
0
00
H


O

CO

-------
      TABLE D-20.  BULK TERMINAL BASELINE LOADING RACK
        ANNUAL THROUGHPUT BY AREA AND CONTROL LEVEL

             Area/Control Level                Throughput
	(10° liters)8
 NONWINTER
 Moderate  and above ozone NA areas
 80 mg/1                                          48,600
 35 mg/1                                          22,300
 10 mg/1                                          55,400
 5 mg/1                                            5,400
 uncontrolled                                          0
 All other areas
 80 mg/1                                          30,600
 35 mg/1                                          14,000
 10 mg/1                                          34,900
 5 mg/1                                            3,500
 uncontrolled                                     88,900

 WINTER
 Moderate  and above ozone nonattainment
 areas not also CO nonattainment
 80 mg/1                                           8,300
 35 mg/1                                           3,800
 10 mg/1                                           9,400
 5 mg/1                                              940
 uncontrolled                                          0
                         D-43

-------
                  TABLE D-20.   (Concluded)
             Area/ Control Level                Throughput
_ (106 liters)3

 Moderate and above ozone nonattainment
 areas that are also CO nonattainment


 80  mg/1                                          14,600

 35  mg/1                                           6,650

 10  mg/1                                          16,650

 5 mg/1                                            1,700

 uncontrolled                                          0


 CO  nonattainment areas that are not
moderate or above ozone nonattainment
areas
80 mg/1
35 mg/1
10 mg/1
5 mg/1
uncontrolled
Attainment areas
80 mg/1
35 mg/1
10 mg/1
5 mg/1
uncontrolled


1,100
500
1,300
130
4,100

13,200
6,100
15,100
1,500
37,800
  The throughputs shown in this table reflect estimated
  actual emitting levels of loading racks at bulk
  terminals, which are often better than the 80, 35, or
  10 ing/1 regulatory limits in effect at the terminals
  (see Section D.I.I).
                          D-44

-------
leak tightness inspections, and uncontrolled.
     For the uncontrolled case, the emissions would all be
attributed to the loading rack.  For the annual leak
tightness inspections, the emission factors were calculated
to be 111 mg/1 for the nonwinter season and 143 mg/1 for the
winter.  The enhanced leak tightness testing emission
factors are 27.8 mg/1 for nonwinter and 35.8 mg/1 for
winter.
     D.3.2.2  Storage Tank Emissions.  The baseline bulk
terminal storage tank populations and throughputs shown in
Table D-6 were divided according to attainment area
designation in the same fashion as discussed above for
terminal loading racks.  This breakdown of bulk terminal
storage tank parameters is shown in Table D-21.  The VOC
emissions were then calculated using the emission factors
shown in Table D-22 for each attainment designation and the
proper HAP to VOC ratios applied to estimate HAP emissions.
     D.3.2.3  Fugitive Emissions.  Since it was considered
that fugitive emissions from leaking pumps and valves were
uncontrolled at the baseline, it was not necessary to break
down the number of components by control level by attainment
area.  Rather, the total nationwide number of components was
calculated (115,750 valves and 10,240 pumps) and the same
emission factors discussed above under pipeline pumping
stations were applied to obtain baseline nationwide VOC
emissions.  These VOC emissions were assigned to the various
attainment areas using the same proportions as the bulk
terminal loading rack throughput and multiplied by the
proper HAP to VOC ratio to estimate baseline HAP emissions.
D.3.3  Bulk Plants
     The baseline bulk plant throughputs and populations
shown in Table D-10 were divided according to attainment
area designation in the same fashion as discussed above for
terminal loading racks.  This breakdown of bulk plant
parameters is shown in Table D-23.  The VOC emissions were
                          D-45

-------
      TABLE D-21.   BULK TERMINAL BASELINE STORAGE TANK
    THROUGHPUT AND POPULATION BY AREA AND CONTROL LEVEL
   Area/Control Level        Population        Throughput
	(f of Tanks)	(106 bbl/yr)
 NONWINTER
 Moderate  and above
 ozone NA  areas
 External                         657               307
 floater/primary
 seals only
 External                         694               325
 floater/primary and
 secondary seals
 Fixed-roof with                  899               196
 internal  floater
 Fixed-roof uncontrolled          0                 0
 All  other areas
 External                         992               464
 floater/primary
 seals only
 External                         531               249
 floater/primary  and
 secondary seals
 Fixed-roof with                  959               209
 internal  floater
 Fixed-roof uncontrolled         729               159
 WINTER
 Moderate  and above
 ozone nonattainment
 areas not also CO
 nonatta inment
 External                         115                54
 floater/primary
 seals only
 External                         115                54
 floater/primary  and
 secondary seals
 Fixed-roof with                  153                33
 internal  floater
 Fixed-roof uncontrolled          0                 0
                          D-46

-------
                  TABLE D-21.   (Concluded)
       Area/Control Level          Population    Throughput
	(#  of  Tanks)   (106 bbl/yr)

 Moderate  and above ozone
 nonattainment areas that are
 also CO nonattainment

 External  floater/primary               194            91
 seals only

 External  floater/primary               212            99
 and secondary seals

 Fixed-roof with  internal               270            59
 floater

 Fixed-roof uncontrolled                  0             0

 CO nonattainment that  are not
 moderate  or above ozone
 nonattainroent areas

 External  floater/primary                28            13
 seals only

 External  floater/primary                44            21
 and secondary seals

 Fixed-roof with  internal                49            11
 floater

 Fixed-roof uncontrolled                  3             l

 Attainment areas

 External  floater/primary               439           205
 seals only

 External  floater/primary               206            96
 and secondary seals

 Fixed-roof with  internal               403            88
 floater

 Fixed-roof uncontrolled                340            74
                         D-47

-------
             TABLE D-22.   EMISSION FACTORS FOR
               BULK TERMINAL STORAGE TANKSa'b
 Type of Emission
       VOC
     Emission
      Factor
Nonwinter  Winter
           Units
 Fixed-Roof
 Uncontrol1ed
    Breathing losses       8.9
    Working losses        34.8
            12.5
            45.1
 Internal Floating Roofc
    Rim Seal losses        0.5
    Fitting losses         1.1
    Deck Seam losses       0.6
    Working losses           7.33  x 10

 External Floating Roof
    Standing Storage
    losses
       Primary seald
       Secondary seal6
    Working losses           4.61  x 10
             0.6
             1.4
             0.7
             -8
       Mg VOC/yr/tank
       Mg VOC/yr/tank


       Mg VOC/yr/tank
       Mg VOC/yr/tank
       Mg VOC/yr/tank
         Mg VOC/bbl
         throughput
 12.7
  6.1
18.5
8.9
-8
Mg VOC/yr/tank
Mg VOC/yr/tank
  Mg VOC/bbl
  throughput
a Emission factors calculated with equations  from  Section
  4.3 of AP-42 using a nonwinter RVP of  10.2  psi,  a winter
  RVP of 14.0 psi, and a temperature of  60eF, as discussed
  in Section 3.2.1.2.
b Assumes storage tanks at bulk terminals have  a capacity
  of 2,680 m3 (16,750 bbl) ,  a diameter of 15.2 meters (50
  feet), and a height of 14.6 meters (48 feet).
c Assumes that internal floating roof is equipped  with a
  liquid-mounted resilient seal (primary only).
d Assumes that external floating roof is equipped  with a
  primary metallic shoe seal.
e Assumes that external floating roof tank  is equipped with
  a shoe-mounted secondary seal.
                          D-48

-------
   TABLE D-23.  BULK PLANT BASELINE ANNUAL THROUGHPUT BY
                  AREA AND  CONTROL  LEVEL
                                             Throughput
	Area/Control Level	(106  liters)

 NONWINTER

 Moderate and above ozone NA areas

   vapor balance incoming/vapor                   12,584
      balance outgoing with no
      exemptions

   vapor balance incoming/vapor                    7,450
      balance outgoing with 4,000
      gallon/day exemption

   vapor balance incoming with                       571
      submerged fill outgoing

   uncontrolled                                        0

 All other areas

   vapor balance incoming/vapor                    8,354
      balance outgoing with no
      exemptions

   vapor balance incoming/vapor                    6,802
      balance outgoing with 4,000
      gallon/day exemption

   vapor balance incoming with                         0
      submerged fill outgoing

   uncontrolled                                   23,600

 WINTER

 Moderate or above ozonejonattainment
 areas not also CO nonattainment

   vapor balance incoming/vapor                    3,786
      balance outgoing with no
      exemptions

   vapor balance incoming/vapor                    1,927
      balance outgoing with 4,000
      gallon/day exemption

   vapor balance incoming with                       268
      submerged fill outgoing

   uncontrolled                                        0
                         D-49

-------
                  TABLE  D-23.   (Concluded)
           Area/Control Level                Throughput
	(106 liters)

 Moderate  and above ozone nonattainment
 areas  that are also CO nonattainment

   vapor balance incoming/vapor                    2,136
     balance outgoing with no
     exemptions

   vapor balance incoming/vapor                    1,579
     balance outgoing with 4,000
     gallon/day exemptions

   vapor balance incoming with                        0
     submerged fill outgoing

   uncontrolled                                       0

 CO nonattainment areas that are not
 moderate  or above ozone nonattainment
 areas

   vapor balance incoming/vapor                       63
     balance outgoing with no
     exemptions

   vapor balance incoming/vapor                      423
     balance outgoing with 4,000
     gallon/day exemptions

   vapor balance incoming with                        0
     submerged fill outgoing

   uncontrolled                                    1,768

 Attainment areas

   vapor balance incoming/vapor                    3,868
     balance outgoing with no
     exemptions

   vapor balance incoming/vapor                    2,778
     balance outgoing with 4,000
     gallon/day exemptions

   vapor balance incoming with                         0
     submerged fill outgoing

   uncontrolled                                    9,338
                         D-50

-------
then calculated for each attainment designation using the
emission factors shown in Table D-24 and the proper HAP to
VOC ratios applied to estimate HAP emissions.  Baseline bulk
plant emissions are shown in Table D-25.
D.3.4  Service Stations
  Service station baseline emissions were calculated in a
manner very similar to bulk plants.  The baseline service
station throughputs shown in Table D-13 were divided
according to attainment area designation in the same fashion
as discussed above for terminal loading racks.  This
breakdown of service station throughput is shown in Table
D-26.  The VOC emissions were then calculated for each
attainment designation using the emission factors calculated
and the proper HAP to VOC ratios were applied to estimate
HAP emissions.  The VOC emission factors are 970 mg/1 and
1,254 mg/1 for nonwinter and winter submerged fill,
respectively.  The splash fill factors are 1,526 mg/1 and
1,972 mg/1 for nonwinter and winter, respectively.  Baseline
service station emissions from storage tank filling are
shown in Table D-27.
                         D-51

-------
         TABLE D-24.  BULK PLANT EMISSION FACTORS


                                      VOC Emission
                                         Factor
                                        (rag/liter)

Type of Emission	Nonwinter	Winter

Tank Truck Unloading
(Incoming Loads)

  Storage tank  filling
     uncontrolled vapor               977          1,260
     balance                           49             63

Tank Truck Loading  (Outgoing
Loads)

  Storage tank  draining
     uncontrolled vapor               391            504
     balance                           20             25

  Tank  truck  filling
     splash filing                  1,611          2,079
     submerged filling                667            860
     vapor balance                     56             72

Storage Tank  Breathing                179            259
                         D-52

-------
in
CM
i
o
CQ
( fA
Q/ JB4

ti

0
SO)
&1
^1
.* <0
§ (S
EH

>
3 §
CT>"H
"«rf "•"
w £3
I"
rage Tank
lissions
n ^
Jj W
en







1
£.

en
o IT
•H >,
Mv.
^^
W l>
"B *



17
E


TT
£
*— •


U
>
&,

u
o



CM
j£

U
>

p-i
g
u
o

PM

CO
O
O
rj
H
o
CT\
n
CM
o
o
VD
in
0
VO
CT\
H



<
g
^
                                D-53

-------
    TABLE D-26.  SERVICE STATION BASELINE THROUGHPUT BY
                  AREA AND CONTROL LEVEL
         Area/Control  Level             Throughput
	(106 liters)	

 NONWINTER
 Moderate and  Above Ozone NA Areas
    vapor balance with no                    73,501
      exemptions
    vapor balance with 10,000                55,681
      gallon/month exemption
    submerged  fill                                0
    uncontrolled                                  0
 All Other Areas
    vapor balance with no                    32,850
      exemptions
    vapor balance with 10,000                42,546
      gallon/month exemption
    submerged  fill                            3,747
    uncontrolled                             95,151

 WINTER
 Moderate or above ozone nonattainment  areas  not  also CO
 nonattainment
    vapor balance with no                    23,414
      exemptions
    vapor balance with 10,000                14,988
      gallon/month exemption
    submerged  fill                                0
    uncontrolled                                  0
                          D-54

-------
                  TABLE D-26.  (Concluded)
         Area/Control  Level             Throughput
	(106 liters)	
 Moderate and  above ozone  nonattainment areas that are  also
 CO nonattainment
    vapor balance with no                    11,174
      exemptions
    vapor balance with 10,000                11,215
      gallon/month exemption
    submerged  fill                                0
    uncontrolled                                  0

 CO nonattainment areas that are not moderate or  above
 ozone nonattainment areas
    vapor balance with no                       273
      exemptions
    vapor balance with 10,000                 2,350
      gallon/month exemption
    submerged  fill                                0
    uncontrolled                              6,657
 Attainment Areas
    vapor balance with no                    15,186
      exemptions
    vapor balance with 10,000                17,671
      gallon/month exemption
    submerged  fill                            1,763
    uncontrolled                             38,120
                         D-55

-------
TABLE D-27.  BASELINE EMISSIONS FROM
           SERVICE  STATIONS
Baseline
Emissions
Existing
New
TOTAL
Underground Tank
Filling Emissions
(Mg/yr)
HAP
10,970
920
11,880
VOC
197,460
16,510
213,970
              D-56

-------
TECHNICAL REPORT DATA
(Please read Instructions on the reverse before completing)
1. REPORT NO.
EPA-453/R-94-002a
4. TITLE AND SUBTITLE
Gasoline Distribution I
Background Information

ndustry (Stage I) -
for Proposed Standards
7. AUTHOR(S)
9. PERFORMING ORGANIZATION NAME AND i
Office of Air Quality P
US Environmental Protec
Research Triangle Park,
VDDRESS
lanning and Standards
tion Agency
North Carolina 27711
12. SPONSORING AGENCY NAME AND ADDRESS
Office of Air and Radiation
US Environmental Protection Agency
Washington, DC 20460
3. RECIPIENTS ACCESSION NO.
5. REPORT DATE
January 1994
6. PERFORMING ORGANIZATION CODE
8. PERFORMING ORGANIZATION REPORT NO.
10. PROGRAM ELEMENT NO.
11. CONTRACT/GRANT NO.
68-D1-0116
13. TYPE OF REPORT AND PERIOD COVERED
Interium Final
14. SPONSORING AGENCY CODE
EPA/200/04
15. SUPPLEMENTARY NOTES
16. ABSTRACT
National emission standards for hazardous air pollutants (NESHAP) are
being proposed for the gasoline distribution industry under authority of
Section 112 (d) of the Clean Air Act as amended in 1990. This background
information document provides technical information and analyses used in
the development of the proposed NESHAP. The alternatives analyzed are
to limit emissions of hazardous air pollutants (HAPs) from existing and
new Stage I gasoline distribution facilities. Gasoline vapor emissions
contain about ten of the listed HAPs. Stage I sources include bulk
gasoline terminals and plants, pipeline facilities, and underground
storage tanks at service stations. Emissions of HAP ' s from these
facilities occur during gasoline tank truck and railcar loading,
gasoline storage, and from vapor leaks from tank trucks, pumps, valves,
flanges and other equipment in gasoline service.
17.
a. DESCRIPTORS
Air Pollution
Volatile Organic
Compounds Hazardous
Air Pollutants -
Gasoline Bulk
Terminals Bulk Plants
Pipelines Service
Stations
18. DISTRIBUTION STATEMENT
Unlimited
KEY WORDS AND DOCUMENT ANALYSIS
b. IDENTIKIEKS/OPEN ENDED TERMS
Air Pollution Control
19. SECURITY CLASS (n. **-,,
Unclassified
20. SKCUKJTY CLASS ,n, f.f,i
Unclassified
c. COSAT1 Kidd/Groop
13 b
21. NO. OF PAGES
407
22. PRICE
EPA Fonu 2220-1 (Rev. 4-77) PREVIOUS EDITION IS OBSOLETE

-------
-------





H
M
i^
B
o
S
S
s
in
n
fa
O
w
o
3
s
D
*
CO
fH
W
O
U
«
U
3
• H
55
H
Q
S
t-3
2:

iW
£H
g
»-^
05
*
rj
H
1


U
(0
?j
*•




















tr\
%
2
j
8
o
Ch
H
K
U
S
1
O
M

O
w
1
w
s
as
t^























IM « «  ^ ^ ^
il s I •* - o s s 2 RJ
1
o *O in to o* CM *- CM
OOj £ £ £vO ^^S52°ZJ
Kt
r\j co N. s.
<| t ^ IA IA*S~^^
« -O «O K X) Kl
U.I Kpt *^ «-^ 4) O ^O CO K CO
lul -| «- o KI in KI *
6
.J





00 ^O CO M K -O tfs
u.1 r*»o- o *->oeoKi'-
lul «O(\l »- ^- rj*-fM 1^
«l Kl M ^ »-

1
— N. ir* «o o* O ^3 m vO Kl IA CO O N-
>-| P 0 T- Kl Kl K-
o
s
o- oo h- r>j r\i in so
U| *jT O *~  KI »— o
JJjI^N- «-OJO^rust

1
_„, KI in oO ^o tn ^
§•*_[ cO N. ^O -» CO

< K[ 0 (M f^ ^ *-
I-)l f\J (M *—




^ H j? «
to 2 K1 'S M
„ S . 5 « , .r
«v ^/ ._ *• u T3
« c c g ** e s
OO C _ £ — o>£ o Jsjoiw.
^8 iSzlgx 2*«Jg.!5£gw
g8^-S2S.«-B,8.g8-^_«-g
ftisfig-sa^sf-^sgu
3 ^ O- — ,^ L. Cu^ai — •— Pa— t-.t- a}ra.xcn.._
f« cC C fc — .- CO ffl S. S « « » —
>iiI3OXujaoxotj*-o a
•£ "c — • £ c
•z ** s ° *
S '5 g £
S SI 5 S £
7-17
Kl
fM in
" S.

K.
fM f\J

K S

3 1
CM
* 8
in



J2 S
s."

V.
CO -;
*
t\i
ca in
S

^-
« g
Kt
V

rvj
" 1
^?

(V
« |
CO

CM

to
°L
*»
Kl

§
a, »
r^*
t .
.— w
l_ C
I 1
O H-
>_ H-
l- Ul
J-1 tj CA
o o o
t— > O