0) o Q- c UJ 5> 5 s-s 8" > o o ) O O o o o o o o o oo o o o co co r- oo cn LO CM «a- m eo po -^ us f— « 3 t. 3 C 01 i— C r— (O *— S- 3 S- 03 i— C 0) i— 3 «£ = I e •!— oa s- = «c S; 0) o . js c o i en > 4-> I OO C o UJ a. § x UJ o 9-25
-------
U.
o

CO
o
o
LU
E:
   00
UJ H- I
O C£J
a: ui
o i— *— •—
LL. 
tu j— o -a
   CO CM =3
U4     1 i—
>. >- U3 O
»-i cai co c
H- o en o
=5 —1
O UJ
   o:
01

 (U
I—*
X)
 «J
*,-*
c c
<4_ S 0
0 §£
in s_ t-
0 0 Z
o "-
e v»
UJ ^"*
o









tn
cj
u

3 Ul
0 t-
tn «
03 CO
o:>-
c o
I aj
o
£
C
UJ



VI
c

o1 •»-
OJ M- O
=3 O «U
CT OL
(U CA
U C
LL. ^<










_
4?
«
|t
e/>
O OO O O O
CO CO tn CM CM f>
CM to, CM i— t
'







§o o o o o
•-H CO O W CO
tO CM •"•* VO trt PO
i-H















C3 CO ^3 CO ^3 ^3
o t-. »r CM oo -H
CM m -«.-4








r— fl r- rtS
J- 3 J- 3
0) r— C 0) i— C
S- 3 «C I- 3 «5
(0 C 1 (SCI
3 C i- 3 C T-
C3* ^C CQ ^3*^C CQ
V>
S o tn
0) 4J 0
OJ O. t^-
IB H 4J

00 X g"
-(. -O X
s- a> uj
i— i fa N
Of- 0
 z
e>c
jj • . °
tO + «S CQ
»
*^
X








I/I
01
O)

-------
1




oo
~^
0
!"**
o.
o
_J
o
0
1—)
t-H
LU
CD
1— ,
CD

O

3Z
^D
^2

r— 1
i
r^
O
LU
I .
I_L»
1 t
LU
LU
00
13
1
-2
0
 "~"
3
M>
 «J Oi
• t— i y
™ ^"""i,
UJ *•%•
i -—
.>

0

LJ *-^«
Cfl O
3 3
CM
4J 1
= to
S — CO
_ sen
V ^r-l
•» ~— '
O IE
k- =)
5 °

«n
S I
0
CU 1
N ^O
n- CO
io^2
3 s:«—
C' 	 ^
O
cu


U)
o
*T ^*
i 3
CU CM
ce . i
C CO
o — en
££E
JL! ^J

8

LU — — 	
a:

*


•
sO
i
LU
_J
CO
*c










ooooo ooooo ooooo ooooo
tO~4LOCOCM i-ICOCMtOtO tOLO^HtOCO -OCOUOLOLO
CMCMCntOf^ f-ttOLOPOOO OCnLOPOCO O^"CM*-I«-H
CMOOCMCMCM i-4*-^»-I^H»-( CMCMCMCMCM »-4»-<^^»-4^^





OOOOO OOOOO OOOOO OOOOO
to**i — r^ro ~4coco^ro totOLoto^* o co CM »— t «•
CMCMCOLO^ «-HtO«9*fOCM OCn«CTCM»-* O"*CM*^O
CM CO CM CM CM •—! »~* •••* •"* *** CM CM CM CM CM i-< *H r- 1 «— < *— (








1 IOOO 1 IOOO IIOOO 1 IOOO
I iroLOCM I ir^r^o I itoLOLO i ILOCO*-H
1A 00 CM ** CM r*. -H CM to CM
CM






ooooo ,00000 ooooo ooooo
O LO o O LO OOOOO O O LO O LO O LO O LO O
^» to to CM o co to cn r*-» to co cn en cn co c? r^> *— < o o
cOLOcnencn tointovoto IO^LOLOLO CM*^CMCMCM
»H ^H i-l 1-1 — 4








§0000 ooooo ooooo ooooo
oooo ooooo o.oooo ooooo
CS C5 C3 C3 C3 CS C3 C3 C5 C3 C3 C? C3 C3 C3 C3 C3 C? C? C?
ooooo ooooo ooooo ooooo
r-r-encoco r~mr»cor» ^Htoco>-Hro OCOCMLOCO
cMcor^^-co vocoto^Hco coto*ertor-* o»-nr^.cocy>
CO <^to"r~l~ 00 CO ^- LO LO CM •-! CM CM CM CM .-1 ~4 <-H -H








, *c c r— >, z >— >,zc: i— >>
zinr— 10 »— ^ o i— «F- «i— ioi— o>—  U (OCi— CU •— O O lOCf—CU i— •!-> O IO C r- CU
Z T- i- EC(O*J Z CLt- E C OJ *J < i- •»- EC «C CUT- EC(O+J
+J +J •?<3l. S +•> -^  ••-> »-<3i. E •>-> t- <: 3 S-
1 CL CUCUCtCrO 1 CU CUCUCICIO 1 Q. CUCUCIC(d 1 CU CUCUCICIO
E u S.^ t- e s x s-ini-^-e:3 E i-in^-^-cs x s-in-^--r- = 3

MXCUI "•• CU 1 "-IXCUI •— CUI
uj.cc: cufc LU jz c Q) j= c
CU 1— i— O O O O CU N 1— i-HOOOO CU t— >-" O O O O - CO M t— « O O O O
O1 O O^ •*" Ol O O} «r"
IQZII raooii (ozii 10 (/) i. i
OO 00 OO 00































trt
c
o
^
u
3
^3
cu
ec
c
o
in ,
in
. I
in U
1 t
•*•> in
0 +»
3 in
•o o
cu o
06 *,
c c
O 0)
•t— E
in u

€ O
LU *4—
e
O LU
O
:» +

in in
in in
O O
o o
•0 XI
CO CO

•^ t—
« 10
3 3
c c:
c c
ie J=
9-27

-------
9.5  REFERENCES

1.   Presentation by Gary Uvagnino,  EPA  Region  IX, at EPA Technical
     Workshop "Control  of Volatile Organic  Compounds from Gasoline
     Storage and Transfer Facilities,"  Fresno, California.  December
     8-9, 1983.
                                   9-28

-------
   APPENDIX A




HISTORY/BACKGROUND
     A-l

-------

-------
                       APPENDIX A  -  HISTORY/BACKGROUND
  Date

 Late  1973


 6/11/73


 Early 1974


 3/8/74


 1974-75


 10/9/75


 11/75


 1976


 11/1/76



 1976-77

 5/13/77



6/8/77


8/77
                        Activity

 EPA  approves  San  Diego  and  S.F. Bay Areas  Stage  II  SIP
 Regulations.

 EPA  proposes  NSPS regulations for Petroleum Liquid  Storage
 Vessels.   Published in  Federal Register 38 FR 15406.

 EPA  promulgates Stage II and other gasoline marketing
 regulations for all or  part of 13 AQCR's.

 NSPS for Petroleum Liquid Storage Vessels promulgated.
 Published  in  Federal Register 39 FR 9308.

 San  Diego  and S.F. Bay  Area implements first generation
 Stage II technology.            •  •

 EPA  proposes  amendments to the Stage II regulations.
 Published  in  Federal Register   40 FR 47668.

 EPA  issues Criteria for Stage I Vapor Control Systems
 at Gasoline Service Stations.

 California implements Stage II programs in all California
 ozone impacted areas.

 EPA  proposes  additional  amendments to the Stage  II  regulations
 as a result of comments received on the October  9,  proposed
 amendments.   Published in Federal  Register 41 FR 48043.

 D.C. implements first generation Stage II technology.

 Final compliance deadlines for EPA-promulgated Stage II.
 Regulations indefinitely deferred.  Published in Federal
 Regi ster 42 FR 27674.                            	

 Benzene listed as Hazardous Air Pollutant under  Section 112.
 Published in Federal  Register 42 FR 29332.

 Clean Air Act Amendments of 1977

 Section 202(a)(5)  - Requires promulgation of fill pipe
 standards for new cars if Stage II required by EPA.

 Section 202(a)(6)  - Requires EPA study of Stage  II versus
Onboard Technology.

Section 324  - Precludes lessee from paying cost of procurement
and installation of Stage II.
                                     A-3

-------
Date

8/77 (cont.)



10/77



12/77
8/77-1/79


6/78



8/78
10/78



12/78



4/4/80


12/80
                     Activity

Section 325  -  Limits applicability of  any  EPA  Stage  II
.regulations  and provides  for  phase-in  schedule for small
independents.

.Control Techniques  Guideline  issued  for  Control of Hydro-
carbons1 from Tank Truck Gasoline  Loading Terminals.
EPA Publication No.  EPA-450/2-77-026.

Control Techniques  Guideline  issued  for  Control of Volatile
Organic Emissions from Storage of Petroleum Liquids  in
Fixed-Roof Tanks.   EPA Publication No. EPA-450/2-77-036.

Control Techniques  Guideline  issued  for  Control of Volatile
Organic Emissions from Bulk Gasoline Plants.   EPA Publication
No. EPA-450/2-77-035.

EPA reviewed feasibility/desirability  of Onboard versus
Stage  II  (no action  taken or  assessment  document released).

Standard  Support Environmental Impact  Statement for  Control
of Benzene from the  GasoTine  Marketing Industry (Draft Report).
Final  report never  completed.

Comments  received during  an open  NAPTAC  meeting, from industry,
environmental  groups  and  other interested  parties on a
draft  112 regulation  and  the  6/78 support  document for gasoline
bulk terminals and  plants and storage  tank loading at service
stations  (Stage I).   Regulation or document never proposed.

American  Petroleum  Institute  (API) releases study
demonstrating  the feasibility of  an Onboard (refueling)
control system.  API  Publication  No. 4306.

Control Techniques  Guideline  issued for  Control of Volatile
Organic Compound Leaks from Gasoline Tank  Trucks and Vapor
Collection Systems.   EPA  Publication No. EPA 450/2-78-051.

NSPS for  Petroleum  Liquid Storage Vessels  promulgated.
Published in Federal  Register 45  FR 23374.

EPA proposes NSPS regulations for Bulk Terminals.  Published
in Federal Register  45 FR 83126.  Also publishes Background
Information  tor Proposed  Standards Bulk  Gasoline Terminals.
EPA Publication No.   EPA-450/3-80-038a.

EPA proposes NESHAP  regulations for Benzene Storage.
Publishes Background  Information  for Proposed  Standards -
Benzene Emissions from Benzene Storage Tanks.
EPA Publication No.  EPA-450/3-80-034a.
                                  A-4

-------
Date

4/13/81




1982
12/82



5/83




7/14/83

8/83
1/84
6/84
7/25/84
                    Activity

Reflecting the Vice-President's announcement of a reduction
o-f EPA's regulatory burden on the Auto Industry, Federal
Register 46 FR 21628 states that Onboard controls~wiTl"
not be required.
Six States committed in
Stage II regulations.
their 1982 03 SIP's to consider
STAPPA and ALAPCO passed resolution to request EPA to
review Stage II and publish CTG.

EOF and NRDC filed notice of intent to file citizens suit
to compel EPA to enact NESHAP to control benzene from
gasoline marketing and other sources.

EPA starts assessment of control alternatives for the
control of volatile organic compound (VOC), benzene,
ethylene dibromide (EDB), and ethylene dichloride (EDO
emissions and risk from the gasoline marketing industry.

EDF and NRDC file suit.

NSPS for Bulk Terminals promulgated.  Published in Federal
Register 48 FR 37578.  EPA publishes Background Information
for Promulgated Standards - Bulk Gasoline Terminals.  EPA
Publication No.  EPA-450/3-80-038b.

Control Techniques Guideline Document for Control  of Volatile
Organic Compound Emissions from Volatile Organic Liquid
Storage in Floating and Fixed-Roof Tanks (Draft) released
for public comment.

EPA starts preparing support document entitled - Evaluation
of Air Pollution Control  Alternatives for the Gasoline
Marketing Industry.

EPA releases draft staff paper on "Estimation of Public
Health Risk From Exposure To Gasoline Vapors Via The
Gasoline Marketing System."

Science Advisory Board reviews June 1984 EPA staff report.
                                         A-5

-------

-------
         APPENDIX B



BASELINE EMISSIONS ANALYSIS
              B-l

-------

-------
                               APPENDIX B
                       BASELINE EMISSIONS ANALYSIS

      The  purpose  of  establishing  an  emission baseline  is to be able to
 estimate  the  impacts  of  reducing  emissions  from this baseline through
 the  implementation of additional  control measures.  The baseline
 emissions must take  into account  the level  of control  already in place
 in the base year  to  get  an accurate  assessment of the  impacts of the
 control alternatives.  The base year for the gasoline marketing source
 category  was  selected as 1982.  This year was selected because this was
 the  final implementation date for many State regulations concerning gasoline
 marketing sources and because the latest data available on facilities
 and  gasoline consumption at the time of the development of this
 document  was  representative of 1982.
      The  general approach for establishing the emission baseline was
 basically the same for each sector of the industry.  Data were obtained
 on the level of control already used by the States and emission factors
 were  selected to represent this level of control.  Uncontrolled areas
 were  defined and emission factors were selected to represent the type
 of loading or type of operations in those areas.   Emissions were
 calculated by multiplying the emission factors by the corresponding
 throughput for the controlled and uncontrolled areas.  Nationwide
 throughput, on a county by county basis, was obtained from EPA's
 National  Emissions Data Base (NEDS).   Although these data were dated
 1980, the total  throughput in the NEDS system was the same as our 1982
 reported  total.   Therefore, this data was considered representative of
 our 1982 base year.
 B.I  Control Techniques Guidelines
     Control Techniques Guideline (CTG)  documents have been prepared by
 EPA for every sector of the gasoline marketing industry (with the
 exception of automobile refueling).   The purpose  of these documents is
 to outline what EPA defines as reasonably available control  technology
 (RACT) for existing sources.   Table  B-l  summarizes the CTG recommended
 limits for each  of the industry  sectors.   Some of the recommendations
are in the form of emission limits and others are in the form of
recommended control  equipment to  be  installed.
                                  B-3

-------
                        Table B.I.   SUMMARY  OF  CT6  RECOMMENDATIONS
                                    FOR GASOLINE  MARKETING  SOURCES
 Facility

 Bulk Terminals
 Fixed-Roof
   Storage Tanks
(40,000 Gal.  Capacity)

 Bulk Plants
 Tank Trucks and
   Vapor Collection
   Systems
Limit truck loading emissions to
80 Mg VOC/liter from vapor processor

Retrofit with internal floating roof,
Cover or seal floating roof vents,
and visual  inspection requirements.

Equipment and work practice specifications
for submerged fill, balance system loading,
and storage tank pressure relief settings
Pass an annual
leak-tight test
which requires
having <3" H20
pressure change
when under
18" H20 pressure
or 6" H20 vacuum
No leaks
greater than
100 percent
L.E.L.9 when
monitored at
any time with
a portable com-
bustible gas
analyzer.
Vapor collec-
tion system
Back pressure
not to exceed
18" H20 when
measured at
the truck.
 Service Stations
   (Underground Tanks)
Design criteria for drop tube specifications,
vapor recovery requirements, tank truck inspections,
vent line restrictions, etc.
             a lower exposure limit
                                                 B-4

-------
     States with nonattainment areas for ozone are required to adopt
regulations consistent with these CT6 recommendations to provide for
attainment of the ambient standards.
     To accurately determine how the States implemented these regulations,
a recent summary report1 and contacts with States were used.  Table B-2
lists the States which had implemented requirements for bulk terminals
during the base year of 1982.  The States listed in the first column
require that all terminals within these boundaries achieve a level  of
control consistent with that of the CTG recommendation (80 mg/liter).
The second column includes States which require controls consistent
with the CTG only for areas within the States which do not meet the
ambient standard for ozone (nonattainment areas).  The third column
includes States which do not have any emission control regulations
pertaining to gasoline terminals.  A similar table was assembled for
bulk plants and service stations.
     In determining baseline regulatory coverage for the tank truck CTG
equivalent controls, two cases were considered:  trucks in "normal"
service and trucks in "collection" service (i.e. truck tanks equipped
with vapor collection equipment).  Normal  service pertains to
areas where no controls are required at the terminal  or bulk plant
loading racks.  "Collection" service pertains to loading when vapor
balance systems are employed.  For normal  service, there are no collec-
tion systems, therefore there can be no leakage of vapors from the
vapor recovery system or sealed truck tanks.  In these cases, the tank
truck baseline emissions are included in the bulk terminal  emission
estimates.  For areas where collection systems are used, the CTG
recommendations are to have vapor tight tank trucks.   Table B-5
indicates areas where collection systems are required at terminals
along with vapor tightness requirements for tank trucks.  All  areas
which require bulk plant vapor collection  systems also require tank
truck controls.
     Baseline emission levels were calculated for fixed-roof storage
tanks and external  floating-roof storage tanks for both breathing
                                  B-5

-------
    TABLE B-2.  STATE REGULATORY COVERAGE FOR BULK GASOLINE TERMINALS
Entire State
Consistent with
CTG Controls9
CTG Controls3
Nonattai nment
  Areas Only
No Control
Regulations'1
Alabama
California
Connecticut
District of Columbia
Georgia
Illinois
Louisiana
Massachusetts
Michigan
New Hampshire
New Jersey
Pennsylvania
Rhode Island
South Carolina
Tennessee
Wisconsin







Arkansas
Col orado
Del aware
Fl ori da
Indiana
Kansas
Kentucky
Maine
Maryland
Mi ssouri
Nevada5
- New Mexico
New York
North Carolina5
Ohio
Okl ahoma5
Oregon
Texas
Utah
Virginia
Vermont
Washington
West Virginia
Alaska
Arizona
Hawai i
Idaho
Iowa
Mi nnesota
Mississippi
Montana
Nebraska
North Dakota0
South Dakota
Wyomi ng
-










 aCTG  Controls  =  80 mg/liter  standard  or  lower, tank truck vapor-tight
  controls.

  Portion of State not covered  by  CTG  controls  is covered by  submerged
  fill  requirements.

 CNorth Dakota  has no nonattainment areas for ozone but  entire State
  covered by submerged fill regulations.

  Approximately 90 percent of total  throughput  is loaded by submerged
  fill.
                                    B-6

-------
  (storage)  losses  and  working  losses.  Most States regulate emissions
  from storage  tanks  in their State  Implementation Plans  (SIPs) with CTG
  recommended controls.2  Based upon the historic pattern shown in a
  previous storage  tank  survey,3 it was assumed that the storage tanks
  located in nonattainament States or nonattainment areas of States would
  be controlled by  external floating roofs.  This survey also indicated
  that the majority of  floating roofs had a mechanical  type primary seal.
  This  survey further estimated that 10 percent of the storage tanks
  would be new tanks subject to New Source Performance Standards (NSPS)
  and would be controlled by external floating  roofs  with primary  and
  secondary seals.  The storage tanks located  in attainment States or
 attainment areas of States were  assumed  to be 10 percent new tanks
 controlled by  external floating  roofs with primary  and secondary seals;
 63 percent would be external  floating-roof tanks with  primary metallic
 shoe seals  and 27  percent would  be  fixed-roof tanks.4

 B.2   CALCULATION OF  BASELINE EMISSION  LEVEL
      Once the  extent of regulatory  coverage was  established,  the
 methodology used was to  determine the  base year  gasoline throughput for
 each  of the gasoline marketing facility operations in  the  regulated and
 nonregulated areas.  An  emission factor corresponding  to the regulatory
 coverage, loading  method, type of storage used, etc., was  then selected
 and emissions were calculated by multiplying the corresponding throughput
 by the corresponding emission factor.  Table B-6 summarizes the baseline
 emissions calculated for the gasoline marketing industry in base year
 1982.  The following sections describe the methodology for each of the
 industry sectors.
 B.2.1  Bulk Terminals
     Given the  regulatory coverage in Table B-2, entire States or areas
within States  were divided into controlled and uncontrolled areas on a
county basis.   Gasoline throughput representative of 1982 consumption
was obtained on a nationwide,  countyby-county  basis,  from the  EPA area
source computer file (National  Emissions  Data  System).
                                   B-7

-------
          Table B-3.  STATE REGULATORY COVERAGE FOR BULK PLANTS
Entire State
Consistent with
CTG Controls9
CTG Controls3 in
  NonAttai nment
   Areas Only
No Control
Regulations0
Alabama
Connecticut
Illinois
Louisiana
Massachusetts
Michigan
New Jersey
Pennsylvania
Rhode Island
South Carolina
Tennessee
Wisconsin










California0
Col orado
Delaware
Indiana
Kentucky
Mary! and"
Missouri
Nevada
New York
North Carolina
Ohio
Oregon
• Texas
Utahb
Virginia
Washington






Alaska
Arizona
Arkansas0
Florida0
Georgia
Hawaii
Idaho
Iowa
Kansas
Maine
Minnesota
Mississippi
Montana
Nebraska
New Hampshire
New Mexico
North Dakota
Oklahoma
South Dakota
Vermont
West Virginia
Wyomi ng
 aCTG  recommendations  include  submerged fill and pressure relief setting
  for  storage  tanks, and  balance  system loading for the loading racks.

 bSubmerged fill  required at loading racks in some portion of State where
  no CTG  equivalent controls required.

 cTypical  facilities were assumed to have 25 percent splash fill and
  75 percent submerged fill at loading racks unless otherwise specified.

 dCTG  equivalent controls in nonattainment areas on storage tanks only.
                                    B-8

-------
        Table B-4.  STATE REGULATORY COVERAGE FOR SERVICE STATIONS
       	(UNDERGROUND TANK FILLING)
 Entire State
 Consistent with
 CTG Controls3
•CTG  Controls3  in
   NonAttai nment
   Areas  Only
No Control
Regulations0
 Alabama
 California
 Connecticut
 District of Columbia
 Illinois
 Louisiana
 Massachusetts
 Minnesota
 New Jersey
 Pennsylvania
 Rhode  Island
 South  Carolina
 Tennessee
 Wisconsin
   Colorado
   Delaware
   Florida
   Indiana
   Kentucky
   Maine
   Maryland
   Mi ssouri
   Nevadab
   New Mexico
   New York
   North Carolina13
   Ohio
   Oregon
   Texas
   Utah
   Virginia
   Washington
  Alaska
  Arizona
  Arkansas
  Georgi a
  Hawaii
  Idaho
  Iowa
  Kansas
  Minnesota
  Mississippi
  Montana
  Nebraska
  New Hampshire
  North Dakota
  Oklahoma13
  South Dakota
  Vermont
  West Virginia
  Wyomi ng
aCTG Controls specifications include submerged fill  of storage tanks,  vapor
 balance between truck and tank, and a leak free truck and vapor transfer
 system.

bportion of State not covered by CTG controls is covered by submerged
 fill requirements.                                               3

typical facilities were assumed to have 50 percent  splash and 50 percent
 submerged filling of storage tanks.
                                  B-9

-------
               Table B-5.  SUMMARY OF REGULATORY COVERAGE
                       FOR VAPOR-TIGHT TANK TRUCK
                       REQUIREMENTS AT TERMINALS
                       USED IN BASELINE ANALYSIS
         All Areas Where Vapor
        Collection Systems Required
           Consistent with CTG
             Recommendations
  No CTG
 Recommended
 Controls in
 Vapor Collection
   Areas
Arkansas
Cal i form' a
Colorado
Del aware
District of Columbia
Florida
Georgi a
Illinois.
indi ana
Loui sisana
Maryland^
Michigan
Mi ssouri
Nevada
New Hampshire
New Jersey
New Mexico
North Carolina,
Ohio
Oklahoma
Oregon
Pennsylvania
Rhode Island
Texas
Utah
Washington
Wi sconsi n
Al abama
Connecticut
Kansas
Maine
Massachusetts
New York-
South Carolina
Tennessee
V,ermoat
Vi rgi ni a
West Virginia
                                   B-1.0

-------
      Emission  factors were  then selected to represent the different
 levels  of  control   already  installed at terminal loading racks.  For
 areas where  vapor  recovery  equipment was installed, three emission
 factors  were used.  The CTG recommended limit for vapor recovery
 controls was 80 milligrams  of hydrocarbons per liter of gasoline loaded
 (mg/liter).  However, a recent EPA study indicated that approximately
 60  percent of  the  systems installed to meet these requirements were
 operating  at or below an emission rate of 35 mg/liter.5  Therefore,
 unless otherwise specified, an emission factor of 35 mg/liter was used
 for 60 percent of  the throughput in areas where terminal  loading rack
 controls were  required and 80 mg/liter was used for 40 percent of the
 throughput.  Some  States, however, specified a 90 percent reduction
 regulation rather  than the 80 mg/liter CTG recommended emission limit.
 In these areas, an emission factor of 96 mg/liter was used.   This
 represented  a 90 percent reduction in EPA's AP-42 emission factor for
 loading  trucks in  collection service.6
      Emission factors associated with tank truck leakage were also used
 to estimate  emissions in areas where vapor recovery was installed.  EPA
 estimates indicated that tank truck leakage averages about 30 percent in
 areas where  there are no vapor tightness testing and maintenance
 requirements and can be reduced to about 10 percent with  vapor tight-
 ness  regulations.8  An emission factor of 288  mg/liter was used to
 represent areas without vapor tightness requirements (30  percent of the
960 mg/liter factor for truck loading in "balance"  service)  and 96
mg/liter to  represent areas with vapor tightness regulations (10 per-
cent  of 960 mg/liter).
     For areas with no vapor recovery regulations,  two emission factors
were  used:   600 mg/liter for submerged fill  and 1440 mg/liter for splash
fill.9  Some States required submerged fill  in all  areas  not controlled
by vapor recovery.   In these cases the 600 mg/liter factor was associated
with the corresponding area throughput.   If there were no  specific
requirements for submerged  fill,  it was assumed that 90 percent of the
                                   B-ll

-------
                           Table B-6.   SUMMARY  OF  BASELINE  EMISSIONS
                                FOR GASOLINE MARKETING FACILITIES
                                        FOR BASE YEAR 1982
                                           Emissions,  Mg/yr
Facility
Bulk Terminals
Storage Tanks
Bulk Plants
Service Stations
- Underground tanks
- Automobile refueling
Total
VOC or
Gasoline
Vapors
140,000
56,000
208,000

222,000
407,000
1,033,000
Benzene
840
340
1,250

1,330
2,690
6,450
EDBa
3
1
5

5
10
24
EDCa
30
15
50

50
100-
245
a
 Applies to leaded gasoline only.   Leaded gasoline consisted  of  approximately
 48 percent of total  consumption in 1982.7
                                             B-12

-------
 loading took  place with submerged fill and 10 percent was loaded with
 splash fill.10
     Table B-7 contains the emission factors and throughputs used to
 calculate the baseline emissions for bulk terminals.  These calcula-
 tions result  in the baseline value for gasoline vapor emissions.
 Reference in  this document to volatile organic compound (VOC) emissions
 (photochemically reactive hydrocarbon emissions) are considered
 equivalent to the gasoline vapor emission from terminals since the
 amount of non-photochemically reactive compounds (methane, ethane) in
 the emissions are very small.  Factors representing the fraction of
 benzene, EDB, and EDC in the gasoline vapor emissions were used.  These
 factors were calculated based upon liquid temperatures and vapor pres-
 sures of the compounds.  For determining the emissions of these
 compounds, the following factors  (calculated at approximately 60°F),
 multiplied by the gasoline vapor emissions, were used:  benzene -
 0.0060, EDB - 0.000046, EDC 0.00047.  Since EDB and EDC appear only in
 leaded gasoline, these factors were applied only to the emissions
 associated with the loading of leaded gasoline.
 B.2.2  Storage Tanks
     As stated earlier in Section B.I, it was assumed that 10 percent
 of all  tanks were considered to be new tanks and would have both primary
 and secondary seals installed on floating roofs.  The remaining 90
 percent of the tanks in the attainment areas were assumed to be 70
 percent floating-roof tanks with primary seals only (63 percent of all
 tanks in attainment areas)  and 30 percent fixed-roof tanks (27 percent.
 of all  tanks in attainment areas).  In nonattainment areas, if there
was a State regulation specific to grimary and secondary seals, then it
was assumed that all  tanks required primary and secondary seals.
Conversely, if no applicable State regulation could be found, then it
was assumed that only the 10 percent of tanks considered to be new
 tanks would have both primary and secondary seals installed on
floating roofs.   The remaining 90 percent were assumed to have
primary seals  only.
     Emission  factors were calculated using the latest information from
EPA and the latest American Petroleum Institute equations.   The major
                                  B-13

-------


£
re

ro
(U
0

_
^

 1—4 m
ro 10 CM CM 10 in ro
ro CM 10 CM 1-4

«a- en 10
i i-i co cn
1 «S- — 1 CM
t •



CM CO 10 10
i in cn co <0
1 (O CM CM

ro

E



CO
CO
CO
  L CM
  I CO
   cn
••» o;
-i «:
UJ LU
 i
cn
.a
 ra





in
j
10
o,
c
a
to
o
_J

,

c
0
o
cu
to

CO
tn o

-^ ro
CO 10
CM 10

CM




C

V,
•f-
X

i— 4
o
CM C
to — *
t-*


10 r- * o
O^ ("•» O
cn 10 ch
r-T oT i—T
»-i »-4 ro
CM in o,
i^» *^



. — to
~~ •!-> O.

Is 5

J2 "re a
10 10
E «o c:
ro ^^ O
ja w, N
to to •r-

*£.*£.*£.
CO r-4 >H
in ro

IO 10 CTt
ro 10 CM
o ^r CM

CM —<




in en
r-4 ro
i-4








r"-. ro en i — »-4
^< ro co in CM
CM i-l
•"•



r-t en ro co ro «a- CM
*-H o 10 in o »-H CM
CO —4


r— «f OO CMCO ro t-~O 10
oco CMCM co«a- cn cncM r^
r*^co O T— * rocM I-H cor*- I-H
Oin m«s- in--4 «-4 ror^ co
r*-*a- CMCO »^r^ r-4 CMin CM
r*» in men r-.r*- cn ^fio o
ro co CM CM CM «3- 1-1
ro
 +J r—
^s ^ - - (Q _/-\
at Q in O E
(O t= 4-> U- 3
r— IOC 103 SOi—
3 t- 10 -O O 
-------


0>
en
to
to
J
o
?
C
to

uj_js_| cn-HcocMOf «-r^ io
S £ -J> 2 'J=!lor~0 »a-in CM
f o w " «^co-nin«d- loco ^j-
oo i— E 1 P- co —4 CM in" co" CM" CM"
o
cn
€
CO
CO
CM

ii i in *$•
II 1 LO O^
II 1 CM ^O
•r— 1

O
i
10
CTl

(-- tn i i 10 in i
O l*» 1 1 U) CM 1
Ol O 1 1 1^ CM 1
CO <-H

                          ^
                          cn
                          o

                          to
                          00
                          UJ CM
                            CO
                           I 00
                           : <
                           : ca
                         -a
                          a>
                          O
                         O
                          i
                         ca
                          a

                         .a




u
10
a:
o>
c
•5
o
_J
i
 E
t^ 0 0
te
||
h- o in
CJ C— 3 PO
a i.
Q-  CO IO ^*
in CM 10 CM in cn o 10 co
-HO co co co vo o cn co
cnin P^ co cn «3- cn co CM
-H^J- VO ~H IO CM COCO CO
co cn o •— 4 i— 4 p>* P^ CM in
-a~~- — *j ~- ^ T
i- O "O £ •*-» 5 o -— .
tO4-» 2 c S4-> j^4^>4->
3oj-i-><->c:u> i-oiE
O— to 4J -t-> 	 tO' — • C
i- — to -U r- JJ
CO *-~ to (/> O 4-*
to to •— ~ «^- to (O
•o — .^- o c ~
•7- cn .^. o c: to
o o s to ^ -5 s
i —  W tO +•>
cn tjj - _ - ^ flj
3' — i- • — .
O (O
O >> CD
.21
in o
tO 3
C/l 4->
c c
to Ol
_
                                                            B-15

-------



1
10
 K-S:
0
en
B
CO
CO
CM

u

i
to
en
O co CM co co csj r*. .-4 f** ***
ui «• "i CM i? rC co ^1 S S
-^.co °» oiroocno oS
-H — < OJCMtnCVJ^J- -H"_"
.

8 SIS i i
CM



•r* •-* co i i en r^ i i
**• co --« t i ou o i i
1*- UO LO 1 1 ^-t CO 1 |
"
                   •o o

                   0.?
                                                                 oo
                                                                 vo
                                                                 O
                                                                                     m    o «r
                                                                                     vo    in o
                                                                                     co    —-a-
S£
   CO
UJ O\
CO

gi
A£

^
ec
T3
 01

 C
           cc

           i1
                C3 C VO
           i—   I— o en
           «   oo
                   J§
                   =110
                   C CM
                   o ro
                   o — <
                  10



                   Sen.
                a =
                >— o <
                o o <
CQ
 (U
 IQ
2^
<=Zm
^ o m
o on
*J*wi
=3 t-
QL (U
JS +j
cn*»-
!o
J=O
1— O
C\J

CO

in
in
CM
01
in
r--
in
1—1


CO
r*.
ft
in
r-t
^~
r*.


CO
«T
CT»
in
to
CO
ff
co r*-
CSJ 10

en *f co
<— • o ^r
^r vo en
co r^ «"
r*» ^ cr*
r*. co «-i
^- ^ co


o^
i-H
in
co"
en
in
•-H


vo «s-
m o
•^ *r
O CO
CO TT
p-« r-
CM
                            •=      en
                            -O      Q «-~
                                              •S      3
                     S
                                    VI U       i- U
                                    O -*-»      < +3
                                    ^ Q)         O
                                    •0	       C —
                                                     JS   3
                                                                            >> c

                                                                            r- i.
                             i   "5
                                                      O    Ol    in    in    3
                                                      re   •»—    
-------
                   !s*
                 et
                 1 — o
                               CO


                               CM
                     in

                     in

                     en"
 OO UJ
 «s >-
 CD
-a
 
jQ

 (O
                     E
     cs c
—   >- O 10
IB   o o en
                 C3 C
                 i— o o
                 u o co
                    O   .
                    i- en
     i— e> in
     o o co
                                                                                o

                                                                                vo
                                OCMCM
                                  O-l
                                  «a-

                                  CM
                        CM
                        I~.

                        CM
                    3 i-

                    CL V

                   f 4-1
                    O
                    i. 0
                   -C O
                  m o -H
                  10 in -)
                            O   -^
                            o   en
           CMfO   f-H
           in rt   co
           ^i- ^-   in
                  in
                  en
                  inioro
                  voooen
                  CM en in
o
no
in
co
ir>
eo
                                                          en
                                                          «a-
cnia
cSS
                                                                        O       rt    ^v
                                   U
                                   4->
                                       IB  -,->->    -^ ^c
                                                      *-      -4->    trt    r—  c_>
                                              >>    O  IB      «i-    o.   O  OJ

                                              s   ^^   ^°    =    ^^
                                              i-    X       i.
                                              OJ    CO       O
                                             ^3   S      >•            JZ
                                                            O
                                       g
                   s
                   ai
                                                                    IB      —
                                                                   4->
                                                                    O
                                                                   ^
                                                                                     O

                                                                                     -C
                                                                                     o
                                           B-17

-------
              \-
                             —I     CM
                    I
  to
                    II
                   :§
O)
o
CO

 cu
                   :S
                                                        i—I   i-* If)   VO CO   <-H LD
                              P-»    O GO
                                                        f-4   f- i—t   (O CM
                                               —   2    E     ~
                                                         
-------








1
!

!






i.


en


z
0

C/1 •— *
i— CM
E CO
uj en

UJ


_J UJ
CO 21
_J 	 1

3
i—
O

O

1
ca
0)
o

1—








Ul _J i-
s •?
1— O CT1
(/I f— E
U
1 ^

Q- CM
O>
_J
1 ° ~
- t
~ vo
h~
en
M
(Q O


CO ^H








O
 O
CM cn ^H co o o
<& *r or-* m co
1^3 CM ,-t —* CM CO
ir> ro CM *f m co
r"» CM to co



U *~^
•—  > 3:

CO
*

r-
I-H






0











CO
t— 1











S




CO



in co
ro CM
en ro
o m
in to
CM




•"7
CJ



c 40
40
VI
H ^O
co.ul
CO CM
CO tO
en cn




—
40

c


Wisconsin
Bn
;i
CMJ!
g
1
!
1
1
1



1
t
1

!
O
cn










CM
i




















cn
CO
CO






40

4J
40
cn
c:
5

en «r
t-H CO
o









'o



o
o



en

I~
CO
0)


fl


c
OJ


ct
o
c
- - j | |












cn
c




<*-

V) • JZ
•^ 40 c/>

ui 0) <^
aj <*- a.
o a -c
40 3
S 2^3
 C
0 0
*— o -~
r— OJ Wl
•^ t_ in
t_ s
at a.


!gTB
V) CJ C
L. en
C CJ —
3 VI
JZ 4O -t—

3 40 40
O 10 3
t_ a.
JZ CJ JZ
40 ui 31
O 3
CU JZ O
C 40 U.
	 	 C 4^
O -f-
en c
D 40
^— o c
(O C_ 1)
jo u O
o a. t-
40 0)
o a.
"o CT o
en 31
•T3 O C

C C
a) -a T-
a> ID 5
ei. s t-
U u Q}
t— CJ 1 —












en
c c.
•~ 0
t- a.

3 >
cr
CU *^
cn u u
•»- 3
V) U
CJ rtj 4O

•^ L. j>4
> fl C
t. (T3
0) C 40
JZ CJ
CU 40 U
(J -f- CJ
= 2JZ
2 d) *
(Q U cn

> CU
C U i-
•f- CU «J

J2 
    40 o 0) o i-~ C 0) O> -r- t. en 0.40 vi a. JZ "a» 3* 40 O cn (_ •f~ JZ r— 4O t. JZ cn -a 3 .-a ' a. en 3 O JZ w en JZ 3 4O O 3 >*, 0 40 O C 3 OJ O t- O a> §CJ JZ 40 = V) 3 •r- VI >>JZ C 3 Ul CJ a> oi t- 40 o ja 3 .<- a. •i,^ 3 Q. O JZ t. en JZ 3 40 O ai jz J= 40 4O cn 40 C 40 ai a. 40 •3 fj 40 OJ JZ o -3 C 5 CJ cs t— 0 CJ JZ 4O OJ O JZ 40 40 40 -a u c CJ TO T-J J3 40 3 3 cn a. v» cn a> 3 •»- o 40 C_ C JZ 3 40 O •— C ^- ZJ £-5 4-> = 3 3 a. cn JZ en cu 3 JZ O 40 JZ VI 4O C 0 ••- 4O rt5 • 40 CJ ! = S V) tO V) -r- C 40 JZ . CU 3 40 • v) a. u "en^ a. 3 a> o c- t. t_ ai = 40 C - O) ^ CJ 40 S 40 rQ QJ ai 40 t- O) 40 'ra 40 C 4O §o t_ 40 O JJ *«- = JZ C o > 40 • en ?O 40 C C 40 •p- S 3. tn O) JZ a) t_ 31 -0-^3 3 O H3^ I t_ 40 a> B-19

-------
emissions from a fixed-roof tank are breathing and working losses.
Breathing and working losses can be estimated from emission equations
developed for.EPA Publication No. AP-42.H  The equations used in
estimating emission rates from fixed-roof tanks are:
          1.02 x ID"5 M,
                           0.68 D1.73 H0.51 T0.5 F
                           14.7 - P
     -w
1.09 x 10~8 MvPVNKnKc
(2-1)
                                                                  (2-2)
where,
     LB = breathing loss per tank (Mg/yr)
     Lyj = working loss per tank (Mg/yr)
     Mv = molecular weight of product vapor (Ib/lb mole)
     P  = true vapor pressure of product (psia) •
     D  = tank diameter (ft)
                                       o
     H  = average vapor space height (ft); use an assumed value
            of one-half the tank height
     T  = average diurnal  temperature change in °F; assume 20°F as
            typical value
     Fp s paint factor (dimension!ess)
     C  = tank diameter factor (dimension!ess )
            for diameter equal to or > 30 feet, C = 1
     Kc s product factor (dimensionless) =1.0 for volatile organic
            liquids (VOL)
     V  = tank capacity (bbl)
     N  = number of turnovers per year (dimensionless)
Several assumptions were made in order to calculate emission factors
on a per tank basis, for both breathing and working losses, from a typical
fixed-roof tank storing gasoline.  The following assumptions were used:
     Mv = 66 Ib/lb mole (for gasoline)
     P  =5.2 psia (for gasoline)
     D  =  50 feet
     H  = 48/2 = 24 feet
     T  = 20°F
     Fp = 1.0 (for a white tank)
     C  = 1.0
                                  B-20

-------
     Kc = 1.0 (for VOL)
Therefore, after substituting into equation 2-1,
LR = 1.02 x 10~5 (66) /    5.2
                     I  	:	
                     I    14.7 - 5.2
    =8.8 Mg/yr
                           0.68 (50)1-73 (24)°-51  (20)°-5(
Using equation 2-2 and these additional assumptions:
     N = 13 turnovers per year
     V = 16,750 bbl tank capacity

 Lw = 1.09 x lO'8 (66) ( 5.2) (16,750) (13) (1) (1)
    - 34.2  Mg/yr
In summary, the VOC emission factors for a typical  fixed-roof tank storing
gasoline are 8.8 Mg/yr from breathing losses and 34.2 Mg/yr from working
losses.
     Standing-storage losses and withdrawal losses are the major sources
of emissions from external floating-roof storage tanks.  From the
equations presented below it is possible to estimate both the withdrawal
loss and the standing-storage or roof seal loss from an external floating-
roof tank.  These equations are taken from AP-42.
      SE
4.28 x 10'4 QCWL/D
Ks Vn P*DMVKC/2205
                                                             (2-3)
                                                             (2-4)
where ,
     LSE
     Q
     C
     V
     N
withdrawal loss (Mg/yr)
standing-storage or seal loss (Mg/yr)
Product average throughput (bbl/yr)
product withdrawal shell clingage factor (bbl/103 ft2)
density of product (Ib/gal)
tank diameter (ft)
seal factor (dimensionless)
average windspeed (mph); 10 mph assumed average windspeed
seal windspeed exponent (dimensionless)
                                  B-21

-------
     P*  = vapor pressure function (dimensionless)
     Mv  = molecular weight of product vapor (Ib/lb mole)
     Kc  = product factor (dimensionless) =1.0 for VOL
For the purposes of calculating the external floating-roof emission
factors, several additional  assumptions were made as follows:
     Q   = the value for product throughput varied from State to State
            (bbl/yr)
     C   = 0.0015 (for light rust)
     W|_  - 5.6 Ib/gal (for gasoline)
     Kg  s 1.2 (for a metallic shoe with primary seal) and 0.8 (for a
             metallic shoe with secondary seal) •
     V   ~ 10 mph
     N   = 1.5 (for a metallic shoe with primary seal) and 1.2 (for a
             metallic shoe with secondary seal)
     P*  = (P/PA)/C1 + (1 - P/PA)°-5]2
           where,
                   PA  = average atmospheric pressure = 14.7 psia
                   P   = true vapor pressure at average actual organic
                           liquid storage temperature =5.6 psia
           Therefore,
                   P*  - (5.6/14.7)/[l + (1 - 5.6/14.7)0-5]2 = 0.10871
     Therefore, after substituting into equation 2-3 and 2-4,
     -W
= 4.28 x ID'4 Q (0.0015) (5.6J/78
= 36 x 10-7 Q/78 Mg/yr
= (1.2) (10)1-5 (0.10871) (78) (66) (D/2205
=9.6 Mg/yr for a metallic shoe with primary seal
= (0.8) (10)1-2 (0.10871) (78) (66) (D/2205
=3.2 Mg/yr for a metallic shoe with secondary seal
In summary the VOC emission factors for a typical  external  floating-
roof tank storing gasoline are 36 x 10-7 Q/-/Q Mg/yr from withdrawal
losses, 9.6 Mg/yr from storage or seal losses on a tank with a
metallic shoe primary seal and 3.2 Mg/yr from storage or seal  losses
on a tank with a metallic shoe secondary seal.
                                  B-22

-------
     Since some of the emission factors for fixed-roof and external
floating-roof seal losses are expressed on a per tank basis, it was
necessary to convert the State gasoline throughput to reflect the number
of tanks per State.  The number of tanks per State was calculated using
the following relationship:

     State throughput (bbl/year) = State capacity (bbl) x number of
                                                          turnovers/year
Therefore, State capacity (bbl)  =   State Throughput (bbl)
                                    Number of Turnovers/year
Storage tank capacities of 36,000 bbl. and 16,750 bbl. were assumed
for external floating-roof storage tanks and fixed-roof storage tanks,
respectively.
        Number of Tanks/State
State Capacity (bbl)
                                     Storage Tank Capacity (bbl)
This method was used to calculate the number of fixed-roof and floating-
roof tanks in each State.  These tank numbers were then applied to the
appropriate emission factor to obtain VOC emissions (in megagrams per
year) for fixed-roof working and breathing losses and external floating-
roof primary and secondary seal losses.
     The VOC emission factors for external floating-roof withdrawal
losses are expressed in terms of the throughput of gasoline for each
storage tank and were simply applied to the gasoline throughput for
each State.  Table B-8 contains the complete results from the calcula-
tions of VOC emissions from fixed-roof and external floating-roof
storage tanks.  Emissions for benzene, EDB, and EDC were calculated
in the same manner as for bulk terminals.
B.2.3  Bulk Plants
     The approach to calculating baseline emissions from bulk plants
was essentially the same as for terminals.  The emission factors for
uncontrolled bulk plants also came from AP-42.  Uncontrolled emissions
from the storage tanks were estimated using the following factors:
breathing losses - 600 mg/liter, filling losses - 1150 mg/liter, and
                                   B-23

-------
TABLE  B-8.   STORAGE  TANK BASELINE  EMISSIONS  (Mg/yr)
 STATE
                    ATT
                 THROUGHPUT
              (lOOO'a liters)
      NA
   THROUGHPUT
(1000's  liters)
     ATT
   THROUGHPUT
ClOOO's barrels)
Alabana
Alaska
Arizona
Arkansas
California
Colorado
Connecticut
Delaware
District of Columbia
Florida
Georgia
Hawaii
Idaho
Illinois
Indiana
Iowa
Kansas
Kentucky
Louisiana
Maine
Maryland
Massachusetts
Michigan
Minnesota
Mississippi
Missouri
Montana
Nebraska
Nevada
New Hampshire
New Jersey
New Mexico
New York
North Carolina
North Dakota
Ohio
Oklahoma
Oregon
Pennsylvania
Rhode Island
South Carolina
South Dakota
Tennessee
Texas
Utah
Vernont
Virginia
Washington
West Virginia
Wisconsin
Wyoming
NATIONWIDE TOTAL
0
519, S77



2.715,332
0
423,897 .
0
8,191, 527
0
1,188,202
1,698,519
0
7,39O,O8O
5,328,341


0
547,954
1,649,208
0
0
7,415,178
4,365,948
4.778,419
1,598,519

265,654
0
0
2,264,452
7,494,581
10,096,147
1,241,544
1,805,332
4,672,230
2.024,229
O
0
0
1,396,124
O
16,454,602
998,100
565,232
7,210,118
3,525,005
2,570,935

1,317,589
111,712,675
7,217,995
0

545,884

2,771.248
4,911,193
657,720
1,028,116
9,450,263
10,678,632
• 0
0
17,246,996
2,339,667
0
885,274

7,721,294
1.372,066
5,536,032

15,759,255
0
0
4.544,547
O

1,529,614
1,530,400
11,858,791
632,413
12,959,957
569,124
O
16,299,183
1,718,663
2,843,932
17,725,227
1,356,552
5,766,603
0
8,913,653
12,744,871
1,516,506
324,950
2,417,832
3,388,070
615,328

0
197,377,851
O
3,269
14,824
24.9O3
67,597
17,079
• 0
2,666
0
51,523
0
7,474
1O,6S3
0
46,482
33,514
23,568
36,990
0
3,447
10,373
0
0
46,640
27,461
30,055
10,054
13,400
1,671
0
0
14,243
47,140
63,503
7,8O9
11,355
29 , 388
12,732
0
0
0
8,781
0
103,497
6,278
3,555
, 45,350
22,172
16,171
49,609
8,287
933,545
                         B-24

-------
 TABLE B-8.   STORAGE  TANK BASELINE EMISSIONS (Mg/yr)
                           (continued)
       STATE
Alabama
Alaska
Arizona
Arkansas
California
Colorado
Connecti cut
Delaware
District of Columbia
Florida
Georgia
Hawaii
Idaho
Illinois
Indiana
Iowa
Kansas
Kentucky
Louisiana
Maine
Maryland
Massachusetts
Michigan
Minnesota
Mississippi
Missouri
Montana
Nebraska
Nevada
New Hampshire
New Jersey
New Mexico
New York
North Carolina
North Dakota
Ohio
Oklahoma
Oregon
Pennsylvania
Rhode Island
South Carolina
South Dakota
Tennessee
Texas
Utah
Vermont
Virginia
Washington
West Virginia
Wisconsin
Wyoming

NATIONWIDE TOTAL
NA
THROUGHPUT NO .
ClOOO's barrels)
45 , 4OO
0
16,200
3,434
193,517
17,431
30,891
4,137
6,467
59,441
67,167 "-
0
0
108,481
14,716
O
5,568
3,872
48,566
8,630
34,821
5,401
99,123
O
0
28,584
0
4,708
9,621
9,626
74,59O
3,978
81,516
3,58O
0
102,519
10,810
17.888
111,489
8,532
36,271
0
56,065
80,163
9,539
2,044
15 , 2O8
21,310
3,870
0
0
ATT
OF TANKS
O
8
37
62
169
43
0
7
0
129
0
19
27
0
116
84
59
92
0
9
26
0
0
116
69
75
25
33
4
O
O
36
118
159
20
28
73
32
0
0
0
22
0
258
16
9
113
55
40
• 124
21
ATT
NO. OF TANKS
FIXED
O
2
10
17
46
12
O
2
0
35
0
5
7
0
31
23
16
25
0
2
7
0
0
31
19
20
7
9
1
0
O
10
32
43
5
8
20
9
O
0
0
6
0
70
4
2
31
15
11
33
6
1,465,171
                     2331
                                     629
                                B-25

-------
TABLE B-8.   STORAGE  TANK BASELINE EMISSIONS  (Mg/yr)
                         (continued)
STATE
Alabama
Alaska
Arizona
Arkansas
California
Colorado
Connecticut
Delaware
District of Columbia
Florida
Georgia
Hawaii
Idaho
Illinois
Indiana
Iowa
Kansas
Kentucky
Louisiana
Maine
Maryland
Massachusetts
Michigan
Minnesota
Mississippi
Missouri
Montana
Nebraska
Nevada
New Hampshire
New Jersey
New Mexico
New York
North Carolina
North Dakota
Ohio
Oklahoma
Oregon
Pennsylvania
Rhode Island
South Carolina
South Dakota
Tennessee
Texas
Utah
Vermont
Virginia
Washington
Uest Virginia
Wisconsin
Wyoming
ATT
NO. OF TANKS
FLOATING
PRIMARY SEAL
0
5
23
39
106
27
0
4
0
SI
o
12
17
0
73
53
37
58
0
5
16
0
O
73
43
47
16
21
3
0
0
22
74
10O
12
IS
46
20
O
O
0
14
O
163
10
6
71
35
25
78
13
ATT
NO. OF TAMKS
FLOATING
SECONDARY SEAL
O
1
4
6
17
4
0
1
0
13
0
2
3
0
12
8
6
9
O
1
3
0
0
12
7
8
3
3
0
0
o
4
12
16
2
3
7
3
O
0
0
2
O
26
2
1
11
6
4
12
2
NA
NO. OF TANKS
97
0
35
7
413
37
66
9
14
127
144
O
0
232
31
0
12
8
104
18
74
12
212
O
0
61
0
10
21
21
159
8
174
8
0
219
23
38
238
IS
78
0
120
171
20
4
32
46
8
0
0
NATIONWIDE TOTAL
                             1469
                                            233
                                                          3131
                             B-26

-------
TABLE B-8.  STORAGE TANK BASELINE EMISSIONS (Mg/yr)
                        (continued)
STATE
54 OF
W/
Alabama
Alaska
Arizona
Arkansas
California
Colorado
Connecticut
Delaware
District of Columbia
Florida
Georgia
Hawaii
Idaho
Illinois
Indiana
Iowa
Kansas
Kentucky
Louisiana
Maine
Maryland
Massachusetts
Michigan
Minnesota
Mississippi
Missouri
Montana
Nebraska
Nevada
New Hampshire
New Jersey
New Mexico
New York
North Carolina
North Dakota
Ohio
Oklahoma
Oregon
Pennsylvania
Rhode Island
South Carolina
South Dakota
Tennessee
Texas
Utah
Vermont
Virginia
Washington
West Virginia
Wisconsin
Wyoming
NA NA NA
FLOATING x OF FLOATING NO. OF TANKS
PRIMARY W/ SECONDARY FLOATING
SEAL SEAL PRIMARY SEAL
O.9
0.0
0.9
O.O
O.O
0.0
O.9
O.O
0.0
0.0
0.0
O.O
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.9
0.0
0.9
0.0
O.O
0.0
0.0
O.O
0.9
0.9
0.9
O.O
O.9
0.9
O.O'
0.0
0.9
0.9
0.0
0.0
O.O
0.0
0.0
0.0
0.0
O.O
0.9
0.0
0.0
0.9
O.O
0.0
0.1
0.0
0.1
1.0
1.0
1.0
0.1
1.0
l.O
1.0
• 1.0
O.O
0.0
l.O
l.O
0.0
1.0
l.O
l.O
0.1
l.O
O.I
•l.O
O.O
0.0
1.0
O.O
0.1
0.1
. O.I
1.0
0.1
0.1
1.0
0.0
0.1
0.1
1.0
1.0
1.0
l.O
0.0
1.0
1.0
l.O
0.1
1.0
1.0
0.1
0.0
0.0
87
0
31
0
0
0
59
0
o
0
0
0
0
0
0
0
o
0
0
17
o
10
0
o
0
0
o
9
19
19
0
&
157
0
O
197
21
0
O
a
O
0
0
6
0
4
0
0
7
0
O
NATIONWIDE TOTAL
                                                          645
                            B-27

-------
TABLE B-8.  STORAGE  TANK BASELINE EMISSIONS (Mg/yr)
                         (continued)
STATE NA ATT ATT
NO. OF TANKS STRG LOSS STRG ' LOSS
FLOATING PRIM SEAL SEC SEAL
SECONDARY SEAL CMg/yr> CMg/yr)
Alabama
Alaska
Arizona
Arkansas
California
Colorado
Connecticut
Delaware
District of Columbia
Florida
Georgia
Hawaii
Idaho
Illinois
Indiana
Iowa
Kansas
Kentucky
Louisiana
Maine
Maryland
Massachusetts
Michigan
Minnesota
Mississippi
Missouri
Montana
Nebraska
Nevada
New Hampshire
New Jersey
New Mexico
New York
North Carolina
North Dakota
Ohio
Oklahoma
Oregon
Pennsylvania
Rhode Island
South Carolina
South Dakota
Tennessee
Texas
Utah
Verwont
Virginia
Washington
West Virginia
Wisconsin
Uyoning
10
O
3
,7
413
37
7
9
14
127
144
O
0
232
31
0
12
8
1O4
2
74
1
212
0
0
61
0
1
2
2
159
1
17
8
0
22
2
38
238
18
78
0
120
171
20
0
32
46
1
0
0
0.0
49.5
224.5
377.2
1023.9
258.7
0.0
40.4
0.0
.780 . 5
0.0
113.2
161. a
0.0
704.1
507.7
357.0
560.3
0.0
52.2
157.1
0.0
0.0
706.5
416.0
455.3
152.3
2O3.0
25.3
0.0 .
O.O
215.7
714.1
961.9
118.3
172.0
445.2
192.9
0.0
O.O
0.0
133.0
0.0
1567.7
95.1
53.9
687.0
335.8
244.9
751.5
125.5
0.0
2.6
11.9
20. 0
54.3
13.7
0.0
2.1
0.0
41.4
0.0
6.0
8.6
O.O
37.3
26.9
"•' 18.9
29.7
0.0
2.8
8.3
O.O
O.O
37.5
22.1
24.1
8.1
10.8
1.3
0.0
0.0
11.4
37.9
51.0
6.3
9.1
23.6
1O.2
0.0
O.O
0.0
7.1
0.0
83.1
5.0
2.9
36.4
17.8
13.0
39.9
6.7
  NATIONWIDE TOTAL
                              2486
                             B-28

-------
TABLE B-8.  STORAGE TANK BASELINE EMISSIONS (Mg/yr)
                        (continued)
STATE ATT ATT ATT
STRG LOSS WRKG LOSS WRKG LOSS
FIXED ROOF FLOAT ROOF FIXED ROOF
 
Alabama
Alaska
Arizona
Arkansas
California
Colorado
Connecticut
Delaware
District of Columbia
Florida
Georgia
Hawaii
Idaho
Illinois
Indiana
Iowa
Kansas
Kentucky
Louisiana ,
Maine
Maryland
Massachusetts
Michigan
Minnesota
Mississippi
Missouri
Montana
Nebraska
Nevada
New Hampshire
New Jersey
New Mexico
New York
North Carolina
North Dakota
Ohio
Oklahoma
Oregon
Pennsylvania
Rhode Island
South Carolina
South Dakota
Tennessee
Texas
Utah
Vermont
Virginia
Washington
West Virginia
Wisconsin
Wyoming
0.0
1.9.4
87.8
147.5
4OO.5
101.2
0.0
IS. 8
O.O
305.2
0.0
44.3
£3.3
O.O
275.4
198.6
139. &
219.1
0.0
20.4
61.5
O.O
O.O
276.3
162.7
178.1
59.6
79.4
9.9
O.O
O.O
84.4
279.3
376.2
46.3
67.3
174.1
75.4
0.0
0.0
0.0
52. 0
O.O
613.2
37.2
21.1
268.7
131.4
95.8
293.9
49.1
0.0
0.1
0.6
1.0
2.7
0.7
0.0
0.1
O.O
.2.O
O.O
O.3
0.4
O.O
1.8
1.3
0.9
1.5
O.O
O.I
0.4
O.O
O.O
1.8
1.1
1.2
0.4
0.5
O.I
0.0
0.0
0.6
1.9
2.5
O.3
0.4
1.2
0.5
0.0
O.O
0.0
0.3
0.0
4.1
0.2
0.1
1.8
0.9
0.6
2.0
0.3
O.O
75.4
342. 0
574.5
1559.3
394.0
O.O
61.5
0.0
1188.6
O.O
172.4
246.4
- o.o
1072.3
773.1
543.7
853.3
O.O
79.5
239.3
0.0
O.O
1O75.9
633.5
693.3
231.9
3O9.1
38.5
0.0
O.O
328.6
1087.4
1464.9
18O.1
261.9
677.9
293.7
0.0
0.0
0.0
202.6
O.O
2387.5
144.8
82.0
1046.2
511.5
373.0
1144.4
191.2
                            B-29

-------
TABLE B-8.  STORAGE TANK BASELINE EMISSIONS (Mg/yr)
                        (continued)
STATE NA MA NA
STRG LOSS STRG LOSS WRKG LOSS
PRIM SEAL SEC SEAL FLOAT ROOF
 CMg/yr) 
-------
       TABLE B-8.  STORAGE TANK BASELINE  EMISSIONS  (Mg/yr)
                                (concluded)
STATE
Alabama
Alaaka
Arizona
Arkansas
California
Colorado
Connect, i cut.
Delaware
District of Columbia
Florida
Georgia
Hawaii
Idaho
Illinois
Indiana
Iowa
Kansas
Kentucky
Louisiana
Maine
Maryland
Massachusetts
Michigan
Minnesota
Mississippi
Missouri
Montana
Nebraska
Hevada
New Hampshire
New Jersey
New Mexico
New York
North Carolina
North Dakota
Ohio
Oklahoma
Oregon
Pennsy 1 van i a
Rhode Island
South Carolina
South Dakota
Tennessee
Texas
Utah
Vermont
Virginia
Washington
Weat Virginia
Wisconsin
Wyoming
TOTAL CNTRLD . CNTRLD FIXED ROOF
STATE STRG LOSS WRKG LOSS EMISSION
LOSSES FIXED ROOF FIXED ROOF REDUCTIONS
(Mg/yr) (Mg/yr) (Mg/yr) 
-------
draining losses - 460 mg/liter.   Uncontrolled emissions from loading of
tank trucks were separated into submerged loading  (600 mg/liter) and
splash loading (1440 mg/liter)..  Unless submerged  loading was specified
by a State, uncontrolled bulk plants were assumed  to practice submerged
loading in 75 percent of the cases and splash loading in 25 percent of
the cases.
     Based upon EPA test data, it was assumed that controls on bulk
plant storage tanks would reduce filling losses by 95 percent, and
draining losses and tank truck loading losses by 90 percent.12
These controls had no affect on breathing losses.  Tank truck loading
losses were estimated using 90 percent reduction of emissions from
trucks in "balance" service (96 mg/liter).
     Gasoline throughput data was obtained from the National Emissions
Data System for the year 1982, on a State-by-State and county-by-county
basis.  However, only a portion of the total  gasoline consumed within a
State passes through a bulk plant.  The percentage of total  gasoline
throughput that is loaded at bulk plants is calculated for each State
by the following relationship:
     Percent of gasoline throughput at bulk plants for State A in 1977 =
           Total  gasoline throughput at bulk  plants for State A in 1977
                Total  gasoline throughput for State A in 1977
The 1977 Bureau of Census provided the gasoline throughput data which was
required to complete this calculation for all  of the States.13  This
percentage was then applied to the 1982 gasoline throughput  for each
State and county,  to derive the gasoline throughput loaded at bulk
plants in 1982.
     Table B-9 contains the complete calculations for the  baseline
emission estimate.   Benzene,  EDB,  and EDC baseline emissions were
calculated in the  same manner as they were for bulk terminals.
B.2.4  Service Stations
     Again,  the  approach to calculating baseline emissions from  service
stations was  the  same  as that for  bulk  terminals and  bulk  plants.   All
                                  B-32

-------
         :   p    §
         i   is|

         IS
         'O
       ste  us.S
         is  ^i
         S  |o
         jo  A o
            cn^>i

            lli
            —
            a
             O
 —      ore  — si
 "»5    5£  ~ai
 2S    su  rs!
.-"    l||    -

            f-.
              "
os    5 5*2 = "Is'
 |     o o 5 j< o «>|
CQ    "S-ISSI
                                              «

                                          03Cj
                                       u> •   10

                                       09    CM
          3    SS '   2  3  S
                •           -
                             — -51
 OJ
.a
 ns

                                             3 iS
                                            J=1C
all

§   i   s
                                                                      $1
   1^
   581
                                                    !    » S
                                                    !u»W
                                                                     *5
                                                                  {   M> •]
                                                                  ,1  is
                                                                  -
'     ™
!    =sl   a
   -i-  '
                                                                Sf  =,-1
                                                  5g^o.a-

                                                  '3 111
                    (MC3>OU)  ^IP^i  O—  -•«»  O%^  a3-»»O
-------

                                           £   3
      >=  S!
                                        s   s
                 •r in
  i  V!
  !S i|   S   s
  3 |ll  -
  I

  ills   *s

  •Piss
si
              3  s  a
         III   "
                                ^ at   n
                                                         a.^-)  —• r^ m CTI  •«• *
                                                                        5S  -S  S-S2 S?SS2S  §S
                                                                                         Sra —• e
                                                                                         ui —• c
                                                             ;"l sS  SssS=s  IS  SSSS
                                                                                            en   jo
                                                             —   I
(C
             I
                                                   B-34

-------

                                                a    a
   i3     -i

   is     i!
         sl
       ill    ;
           i  is
           P  a*"'

           I  "A
           1=  51
                         =     a
       M   -S
   ;«   a «rl
 — u*   c)
 SIS   r Bl

 a    =3«   S—   S — — t
 a.    in «r e

 CTT3 SI "CT>*—|  coo   ^f^c^e
*ac  a    tori   IOP^<*»C
                                 O i— -o *
                                   ^ o «
 d!

 c
•r^
-M

 O



  *
•01



 a
r—
J2
 as
        SO)>~.
        33
                                                                                                              g    8
                                                                                                                                                  S
                                                                                  SI
                                                                                                                                      S   S
                                                                            •    =-;
                                                                                 -
                                                                            Is
                                                                                        S  S    M
                                                                                        O  jo    —
                                                                                   '£    -i
                                                                                   iit  5fj    S   S     2
                                                                                   if!   =g|    I-   '
                                                                                   o S>  u. ^-1
                                                                                   ill     -
                                                                                        —t  CM    O
                                                                                                                   «.r»   ?»J —     ^in   ^ — •   (-» —   «—   >--,o
                                                                                                                   MO   mm     O^   ,-MCy   m — • CM .^
                                                                                                                   M r-i   10 —     3 vo   — r-   — to LO ia   -".^i ^.IA

                                                                                                                   o —   \0r^.     tOO   O O   ."-. ^ to CO   O f*l 33 «j*
                                                                                                                   Mf^   'J)C\i     -^ ID   3%tO   ^fflior^.   S JTt  *°
                                                                                                                                       -                   "
                                                                               O) 3 3!
                                                                               V O C3I
                                                                                   !!  i   -
                                                                       B-35

-------
        Ill
       1


       I   J
 Is

 li
          581
          a Si
        sft  s
       ,.s  a"
      !;» =si  s
       i ^=!
       " r-
                                                       3,
                              \



                              £

      1  112  =-^=  MV3IM  =^>tv

     ^—^•ilr  lsS5  35?s  -Z~
                                          =S? 2333  =5
                                         r»eM(-i--. ™ f^in_i«)_  J\*^
 tu   -:
 o

 o
o
         sK,
         •s-ss:;   as   g  s   s  s
           i.     ^ ft   ~*  -*   —  —
cn
 i
CQ

01

ja

£
                                                           2
            ^. =
                                   B-36

-------
Footnotes for Table B-9
 Twenty-five percent of gasoline throughput was assumed to be loaded by
 splash fill.

 Seventy-five percent of gasoline throughput was assumed to be loaded
 by submerged fi11.

'storage tank filling losses were assumed to be 95 percent controlled.

 Storage tank draining losses were assumed to be 90 percent controlled.
a           -  •                           •
"State regulations require 77 percent vapor control on storage tank filling
 and draining losses and tank truck losses.

 The percentage of gasoline throughput that is loaded at bulk plants is
 calculated for each State by the following relationship:

     Percent of gasoline throughput at bulk plants for State A in 1977 =

     Total gasoline throughput at bulk plants for State A in 1977
             Total gasoline throughput for State A in 1977

 The 1977 Bureau of Census provided the gasoline throughput data which was
 required to complete this calculation for all of the States.

g
 Storage tank filling and draining losses are 95 percent controlled as
 required by the Bay Area regulation.
 i
State regulations require bottom or submerged fill  for bulk plants with
throughput <_ 4,000 gal /day.

State regulation requires vapor balance and an emission limit of
80 mg/liter for bulk plants with a throughput >4,000 gal/day.
Therefore 58 percent of throughput was assumed to be loaded at plants
with a throughput >4,000 gal/day.

Represents the total gasoline throughput and the percent of gasoline
throughput loaded at bulk plants for the county listed.  The total
State throughput is obtained as the sum of the county throughput and
the throughput shown for the remainder of the State.  The total State
throughput from bulk plants is obtained as the sum of the county
throughput attributed to bulk plants and the throughput from bulk plants
shown for the remainder of the State.

The designation "county (etc.)" represents the total gasoline throughput
and the percent gasoline throughput loaded at bulk plants for all
counties subject to the CTG requirements.  The total State throughput
is obtained as the sum of the county throughput and the throughput shown
for the remainder of the State.  The total State throughput  from bulk
plants is obtained as the sum of the county throughput attributed to
bulk plants and the throughput from bulk plants shown for the remainder
of the State.
                                   B-37

-------
gasoline, with the exception of agricultural accounts, was assumed to
pass through service stations, which includes retail  outlets and private
outlets (see Section 4.2.1).
     Emission factors for emissions associated with the service station
underground storage tanks were obtained from AP-42 and were all  based
on gasoline throughput.  Uncontrolled underground tank filling would be
performed either by splash filling (1380 mg/liter) or submerged filling
(880 mg/liter).  Unless otherwise specified, it was assumed that 50
percent of the service stations practiced submerged fill  and 50 percent
practiced splash fill.14  Where underground tank filling  was controlled
by a balance system, an emission factor of 40 mg/liter was used.
Breathing losses were estimated using a factor of 120 mg/liter.
     The majority of all service station automobile refueling is
uncontrolled.  The AP-42 factor used for the operation was 1080 mg/liter.
Data from the State of California indicates that systems  installed to
control automobile refueling are at least 95 percent efficient.15  The
emission factor used for controlled automobile refueling, therefore,
was 54 mg/liter.  Automobile refueling controls are installed only in
portions of California and in Washington, D.C.  An emission factor for
spillage (84 mg/liter) was also used.  This was used regardless of
whether refueling controls were used.
     Table B-10 contains the calculations used for determining baseline
emissions from service stations.  The factors used to estimate benzene,
EDB, and EDC, emissions from the service station underground tanks were
the same as for bulk terminals.  The factors calculated for automobile
refueling were higher, however, because of the higher temperature associ-
ated with automobile fuel  tank (tank temperature assumed  as 70°F).  The
factors used for automobile refueling were as follows:  benzene - 0.0066,
EDB - 0.000052, and EDC -   0.00053.
                                   B-38

-------



























*z
s
E
^S.
o
CO
00
•i:
UJ
UJ
3£
_J
UJ

^
CD

CD
t— l CM
_J CO
UJ O\
3 *^
U.
UJ Ql
OS «C
LU

lu
«r co
O CO
g —

13
•^

%
0
_I

•z
o
1—
<
CO

UJ
U
>
o:
UJ
co
O
r™
1


OJ
(O
f—
















UJ _J i-
< "*"-^
1 — o en
00 t- E
 rf)
CO
¥
t- J3
0) -O C
3 flj QJ ^
**~ r— • E ^*,
o> f— o» en
01 ° re E
O +J r2 «3-
•*-> c CLLO
3 o  f—
ttl C7
O E
10
i— O
Q.CO

a

UJ _J i.

i— o^>


„,_
•r- en
.c E

10 O
(U CM

en m
e
•o

J ^
c If
t— ^— O
+•> CLCO
to to co
CO
a
u
> 
"cn^
3
O (/>
jfg
s






>^

c
3
O
C_3


(O
4->
CO
§

CO

10
O

















LO
r-.

r-*



LO
«— 1
t-H




to
to
CO


























en

CM



cn
^

CM
^






•o
2
4-1
CO

(TJ
E
ta
-Q
n3

<
LO O
O «3-
i& r^

to

^ <•
^~ f^


















tO CM
LO CO

LO



SLO
to

to




CM CM*-*

LO 9

to

z


CO ^
* c
CO •»-

tn
*x
 i-
rt3 fO
— S
ta
to c:
JK; o
W N
fO «^-

< <
LO
CM
O

LO

r-. to

CO
















CM O
r-. en

^*



cn
LO
CM
LO




CM U9

^«






CM
0
to

CM







Oi CD
LO CO














O CO
!^ CO
,O LO

r^. to
co"






,ii
in
re
3
in D_
fO
in
. c
rt3


<:
o
10
•o

£

^ co •— i LO en

r^ r-. CM to ^H







<- r— i , C 3 CO
•i- (O (O O
C CQ CO CO i-
i- OJ
O .C
U- . 4J
i- O

• ro
0
CO
*r>

to

CO CO
CM CO
CM CM
















CO CO
co cn
cn cn

CM CM



$
CO
CO




to co
CM CO
CO CO






CO









LO
en
•—i








ft

•— i



CM CO
CO ^
CO CM
LO ^

r^ i —
CM CM

1 —
U
4->

•»"


"•a
•3
£
t/i


4->
3
CJ

ZJ
U
0)
c
c:
o
CJ
B-39

-------
             Ul _J i
             v-< :
                                                            .—     \o
                                  CO «3-   (-.
                                              §
               *   !
                  J
                  -
                                  coo   ^   -*   -^
                              co   en ^f   •—•   ro
                              CM   CO CO   CO   "3"
                              —t   crt «H   CM   -H
        O     r- C
                si1

                      CO r^   —4
                                              CVJ   S
T3
 O)
                :§
                             «   CO 01   O   -<   —
O
                             •r-      U
                             •i
QQ


 O)
                      (O     1—
                                         £   £
                                         2-   "5
                              (A   O     O3:
                             i—   i~     (U   rO
                             o   u.     cs   =r
                                             B-40

-------
        F—   -O C
         OJ    QJ OJ ^
             ^ OJ cn
              O o E
              U (15
                0.0
                (AC
                S"
                                   CM CM   ~-t Ch   CO   O
                                   O\ CO   CO r-   O   CM


                                      1-4   —« LO   o*   r*.
                       ^M      CO
                                           CM     _4
                                                  ^1   ^4   O
                                      ^4   ^*4 \Q   O
                                                                 LO   tO _4 f
         B

         5
         o   •   »— <
 g*-
 =3 LU
 «s
 O C3
-H   CO
CM   O
cn   «—i

—4   CM
•a
 O)
                       r*- LO   -*f
                       d LO   a\
                       CM LO   CM
                                                                               co   o c
                                                                               er>   en
                                                                               to   —i r
4-J
 £=
 O
                                                                        •*- to   .—.     • —
  I
CD
 
-------
        PiS-1
         3   U •*-
 5

 £
 §—
 S- CM
 _J O3
 S2

 s§

 e-
 3K
 §
                                                            -H      CO
 en en



dec'
                        J Cn O
                        j r*- in
en   to   Oco   o o en   coco   ^j-   cMcn
CM   en   en tn   ro o co   ^••er   o   into
^~t   en   ~H     to in m   co     •—'   »•* co
           I
                                                            —t    - If)
                                                  en-< ^-   ^^\
         g
         5
         O      r- <
                           in     1-4   co

                           «-H     in"   ^-
•o
 O)
 ZJ
fc.
vt

«
S«* o
0 -H
to vo co
to co en
cu en in

8
0

^
en
CO
in
CO
£
tn 2

to co
CM tO
CM
— < CM in
CO CO CM

en m o
^r en o
r-. in r*-
^- CM

en to
o in
O
»— i
^- CM 00
^ CO CO

^* o en
CM CO CM
f-t ^t to
 e
 o
o
  I
ca
 (O
            •w   X
               irt
          J£f—   i.   —
          i. cn   t-

          ^S o   vi
          o o   a.

        3        tt


        i        :
                                      —•      o
                                      «      i—
                                                            T- *—   (O
                                                                   O     F—
                                              B-42

-------
                                      :  g   -
                  e><
                  as en

                                                                    ^J -«
                                                                    co r»-
                                                                     -
                 •5.S
                  too
                         «O IO   \O .-<   CO

                         SS   SS   S
  -  - -,  <~< IO    CO *-t
  1  E   to O    V V
 j    \  ^CM    c^ro   ^


 t**l
                                                 CM   CO   IX)
                                                 os   vo       *  r.

                        V ^   CM CM
                                          O CO


                                          T** ^-t
 O

O
  1

CQ
JQ
 ,

                                       W    (U   ,C
                                       C   -O   4J
                                             B-43

-------
i










1?
'^
I
01

S
13





^
Jg*
£
•HM
C/)
g
I/}

5
UJ
s:
-j cn
LU K
I< a
CO «
o
S?7 c
uj en •**


Of wC /)
— * it* «.
p >- a»
fr— U
= UJ •-
•C to >
C^ CO 0)
g— «
g




«•!
g

1
>—
V)
.

CJ

•5
cc
UJ
CO
1
c

!Jj
c
C3

«— •
*
O
ca
(U
r—
(ft
t—
i


UJ r t tn 1 t£)

< Q "S *~i
C/) J— jj-| ,-4*


Ini
-<
XI CO
SCO
(O




a>-Z
c ^
f^O
Q *T
CO




in
fO) CM O
4J rotn
•»- CM Crt
§tn «a-
tn to CM
u- to co
i— §>

^^

j

cu
c
0
•o
c
0)
>s *•*
o u
U J->
3; I

4^ CU
CO >•

f^^
o
CM
•-4


to co
0 0
to CM









CO — <

r»- CM

o
o

oT




in o
to cn
CO CM





in
cn_






CO

CO





^
en








CO CM
»H ro
-4 CO
CM *r
r»- CM



*^

o!
c
o
cn
c
10 

i —

o
CO



cn co
CM CM








r— 01
o in
co to
ro ro

Ol


^T




ro r~
CM O





CM

CM





_
U)
in
*





vo
CO
•~4







gs
0 0
in co
CM CO
to co
CO CO





^^
S
cu
if
c re
cn
•»-
.e
5
X

o

r-
ro


to CM
— i in
CM








r- in
r- vo

CM

^
CO

CO




cn «f
o r»
CO





«T in
p~ CM






_-,
ro r-
-. CM
-"















in co
ro CM
cn co
o in
in to
CM





•"^
10 i
•*— TO
C C
f- 4J
cn 3
i- 0.


I/)
cu
2

o *r
co PO
»-* Lfl
cn i-<


to ^o — i

CM *r ^i









IO O" CM
»-< CO «^
CO L(1 1-4

-4

CM (O
** «




CM ^r co
LO cn in
CO LO »-<





en
O
o%






0
g






r-- CO
»H cn
~H rt







coco cn
*~4 to CO
co in in
ro CM r~
ro in ~*
o» cn co
CM *r ^





^.
-2i ?
C 4^
g 1^
C i- -—
i- ca
c/i cn
c c
O *i—
U S
in o
tc >»
2 2B

CM rx.

tfi cr»
cn ^o
^ o
«er 

o


CO
2=
LU
0 ^
CO LU
«— « oo
S 
3
CL
JZ
en
|
JE
4->
CU
c
o
m
TO
cn

4->
ro
jr
4-)
*
T3 l/l
aj 
0 10
01 "S
*— 4.J
'o E
4J 'Z
C to
O 01
u
V)
>,f-

l&
i_ aj
c. ."
5 U
U f-

01 u_
I- 0)


c/> o
01 t.
co c
0 0
•— o
c .

=3 £_
«- 10
Ol Q.
•O
Ol C
— re
tj
•i- 1C
-C CU
cu <-
5» 10
























CU
t_
.01
.-§

t/}
s
(_
TO
c
i

o
«sr

c

TO
4-1
cn

>t


TO
u
)J^
u
CU
CU
cn

cu •
u ^
TO CU
•Q .f«
CU J_)
4-> U
.^2
F— ^
(/) VI
c "*"


£- CU
3 TO
CT4J
0) CO
O C
4-> 'i-
0 U
E TO
OX?
in o
•r~ Q.
S TO
UJ >
0





























O
4J
J_>
C
CU
^
cu
a.
CO

£

"O
ai
at
TO
cu
£. •
u in
CU J-»
•^3 c
3
(U O
t. 0
cu o
3E TO
m i—
c re
o c.
*•" 3
(/) 4->
in i—
f- 3


(_
cu cn
C TO
t— o
Q) ^J
in


cn..-
c: -^
t— (U
r— L.
a c-
4-> O
u c
'S o
<- u
cu u
CO (O
•a
B-44

-------
B.  References
1.
2.

3.
4.
5.



6.



7.


8.
9.

10.

11.



12.
13.
14,
 GCA  Corporation.   Status  Summary of  State Group  I'VOC RACT Regulations
 as of March  10,  1980.   Report to U.S. Environmental Protection
 Agency.   Research  Triangle  Park, N.C.  Contract  No. 68-02-2607, Task
 No.  44.   May 1980.   168 p.

 Reference 1.

 Peterson,  P.R. et  al.,  Evaluation of Hydrocarbon Emissions from
 Petroleum Liquid Storage.   U.S. Environmental Protection Agency.
 Research  Triangle  Park, N.C.  Publication No. EPA-450/3-78-012.
 March 1978.

 Pacific Environmental Services, Inc.  Estimated  Nationwide Petroleum
 Storage Tank  VOC Emissions  for the Years 1983 and 1988.  Report to
 TRW  Environmental  Engineering Division, Research Triangle Park, N.C.
 Contract  No.  M23399JL3M.  April 5, 1983.

 Bulk Gasoline Terminals - Background Information for Promulgated
 Standards.   U.S. Environmental Protection Agency.  Research Triangle
 Park, N.C.   Publication No. EPA-450/3-80-038b.   August 1983.

 Transportation and Marketing of Petroleum Liquids.  In:  Compilation
 of Air Pollutant Emission Factors.  U.S. Environmental Protection
 Agency.   Research Triangle  Park, N.C.  July 1979.

 U.S. Environmental Protection Agency.  Federal Register.  Vol. 47
 No. 210.   October 29, 1982.  p. 49323. 	

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

 Reference  6.

 Reference  1,  p. 3-24.

 Storage of Organic Liquids.  In:  Compilation of Air Pollutant
 Emission Factors.  U.S.. Environmental Protection Agency.  Research
 Triangle Park, N.C.  April  1981.

 Pacific Environmental Services, Inc.  Compliance Analysis of Small
 Bulk Plants.  Report to U.S. Environmental  Protection Agency, Region
 VIII.  Denver, Colorado.  Contract No. 68-01-3156, Task 17.  December
 1976.

 U.S. Department of Commerce.  1977 Census of Wholesale Trade-Volume
 I, Subject Statistics.  Petroleum Bulk Stations and Terminals.
August 1981.

 Standard Support Environmental Impact Statement  for Control of
Benzene From the Gasoline Marketing Industry.  U.S. Environmental
                                  B-45

-------
     Protection Agency.  Research Triangle Park, N.C.  (Draft)   June
     1978; p. 2-17.

15.  Memorandum from Norton, R.L., Pacific Environmental  Services,
     Inc., to Shedd, S.A., Environmental  Protection Agency.   December
     20, 1983.  Trip Report to California Air Resources Board.
                                  B-46

-------
          APPENDIX C



ASSESSMENT OF ONBOARD CONTROLS
           C-l

-------

-------
                                          EPA-AA-SDSB-84-01
                   Technical  Report
                The  Feasibility,  Cost,

              and Cost Effectiveness of

                Onboard Vapor Control
                  Glenn W.  Passavant
                      March  1984
                        NOTICE

Technical Reports  do not  necessarily represent  final  EPA
decisions  or  positions.    They  are  intended  to  present
technical  analysis   of   issues   using   data  which   are
currently available.   The  purpose  in  the release  of  such
reports  is   to   facilitate  the   exchange   of  technical
information   and   to   inform   the  public   of  technical
developments  which  may form  the   basis  for  a  final  EPA
decision, position or regulatory action.

       Standards Development and  Support Branch
         Emission Control Technology Division
               Office of Mobile Sources
             Office of Air and Radiation
        U.  S. Environmental Protection Agency
                          C-3

-------
                       Table of Contents
                                                            Page
I.    Introduction	1
II.   Technological Feasibility  	  1
III.  In-Use Performance of Onboard Systems  	  7
IV.   In-Use Emission Control Effectiveness  	 10
V.    Costs of Onboard Vapor Recovery  	 13
VI.   Cost Effectiveness		 21
VII.  Leadtime Requirements	24
VIII. Onboard Control Versus Time  	 27
IX.
Conclusions	 33
References
Appendix A: "Recommendation    on    Feasibility    for
            Refueling Loss Control," February 1980.
                                                   . .  35
                                                   Onboard
Appendix B: "LDV and LOT Operation and Usage Characteristics"
Appendix C: Tables  from  "Manufacturing  Costs  and  Automotive
            Retail  Price  Equivalent  of  Onboard Vapor  Recovery
            System for Gasoline - Filling Vapors"
                               C-4

-------
 I.
Introduction
      This  report   updates   the   previous   analysis   of   the
 technological feasibility, in-use effectiveness,  cost,  and  cost
 effectiveness  of   an  onboard   vapor   recovery  system   for
 controlling  refueling  emissions  from   gasoline-fueled   motor
 vehicles.   The  last  report  in   this  area  is  dated  February
 1980.   In  that  report  it was  concluded  that  onboard  vapor
 recovery  was  feasible for light-duty vehicles  (LDVs).   However,
 some  question remained about the feasibility for  other  types  of
 gasoline-fueled  motor  vehicles  and  the  cost  effectiveness  of
 controlling  refueling  vapors  through   the  use  of  an  onboard
 system.

     Therefore,   this  report  addresses   the  feasibility   of
 control  for  other  gasoline-fueled  motor  vehicles  (light-duty
 trucks_ (LDTs) and heavy-duty  gasoline-fueled vehicles  (HDGVs))
 in  addition to LDVs, and  also examines those factors related  to
 cost effectiveness.   The  feasibility  is examined  for HDGVs, but
 the  cost and emission .reduction impacts a-re  not  determined.
 However,  cost-effectiveness values similar  to those calculated
 for LDVs and  LDTs  would be  expected.

     This report  begins with  a discussion of the  technological
 feasibility of onboard control,  and this  will  be  followed by  a
 calculation  of  the  in-use  effectiveness  of  onboard  control
 systems.  After  reviewing  and updating the  previous estimates
 of  the  costs  of control,  the  cost effectiveness  of an onboard
 strategy  will be calculated.   In addition, a  fifth section  of
 the  report  estimates  the  leadtime   necessary  to  implement
 onboard  controls,  and  the  last section  estimates  the  time
 required  for   an  onboard   strategy  to achieve   control  of   a
 substantial portion  of in-use refueling  emissions.   A summary
 of the overall conclusions closes the report.

 II.  Technological Feasibility
     A.
     Introduction
     The bulk  of  the experimental work  in the  area  of onboard
vapor  recovery has  been  performed  by  the  American  Petroleum
Institute   (API)  and   their   contractors,   Exxon,  Mobil,  and
Atlantic   Richfield    (ARGO).    They   completed   a   vehicle
demonstration  of  onboard vapor recovery in  October  of 1978[1].
The  results   of  that   study   strongly   suggest  that  onboard
controls  are  feasible  and  effective  in  controlling  gasoline
refueling  losses  from  low-  to  mid-mileage  LDVs,  with only  a
negligible impact on a  vehicle's  ability to  comply with current
exhaust or evaporative emission standards.
                               C-5

-------
                              -2-
     Following  the  release  of  the  API  work  in  1978,  EPA
solicited comments  from the  motor  vehicle  industry concerning
the cost  and  technological feasibility of  onboard  controls for
LDVs  and  LDTs.   These comments  were   incorporated  in  EPA's
analysis  of  the  API  vehicle  demonstration  program  _   (This
report   is   presented   in  Appendix   A,   "Recommendation  on
Feasibility for Onboard Refueling Loss Control,"  dated February
1980.) [2]   The   judgment  that   onboard  vapor   recovery  is
technologically feasible  for  gasoline-fueled  motor  vehicles is
based  largely  on  this  analysis  of  the  API  work  and  the
technological  feasibility  comments   submitted  by  the  motor
vehicle  industry.   In  the  remainder   of  this  section,  the
information leading  to the conclusion that onboard  control is
feasible  for   LDVs  is   reviewed,   and   the   feasibility  of
controlling LDTs and HDGVs is discussed.

     B.    Review of LDV Feasibility

     1.    New System Performance

     The  onboard  contol effectiveness  of new  systems  is  based
on  the  results  of the API vehicle  demonstration  program.   This
program  consisted of  SHED tests of. the  entire system minus the
filipipe  seal  (the fillpipe  was  plugged)  and  bench SHED  tests
of  the  ARCO  rotary  fillpipe  seal.   These  tests  showed  that
refueling emission  control efficiency ranged  from  98.2 to  99.3
percent  for both  total  HC  and benzene, with an average value of
98.9  percent.[3]    Based  on  these  results,  a  control   system
efficiency   of   at  least   98  percent   is   judged   to  be
representative  of   potential  new  vehicle   control  for  the
canister/modified fillpipe and  seal  system evaluated by API.  A
diagram  of  the system  evaluated by  API  is  given  in Figure A-l
of  Appendix A.
     2.
Mechanical Durability
     EPA's  1980  report summarized  the  ARCO API durability  data
on  the  nozzle/fillpipe seal  effectiveness.   These data were  of
necessity   derived  from  an   accelerated   test  program,   and
therefore concerns about  seal durability over time could  not  be
addressed.   After  completion of the original work for API,  ARCO
installed a fillpipe cone seal  (Figure  A-6  of Appendix A)  in  a
company  vehicle  and monitored seal effectiveness over 26  months
and  approximately  54,000  road miles near their  Harvey,  Illinois
facility.    During the  26  months,  the  seal  was   exposed  to
environmental    extremes    representative    of   most   of   the
continential United States.

     ARCO  tested  the seal  effectiveness  by measuring  the  HC
concentration  at  the fillpipe/nozzle  interface  each  time  the
                                C-6

-------
                              — 3 —
vehicle  was   refueled.    At  periodic   intervals   the   seal
effectiveness was  checked by measuring  the leak  rate  past the
seal using a specially designed nozzle  to  pressurize the system
with  nitrogen.   The  pressure  check  tests were  performed  at
system pressures corresponding  to 5, 10,  15,  and 20  inches  of
water, at which the seal  effectiveness  was still 99  percent.   A
pressure of  4-5  inches of  water  is  typical  for  a  normal  fuel
fill.  Therefore,  the  ARCO  data  indicate  excellent  sealing
capability over  time.  Overall,  seal effectiveness  was better
than 99 percent after two years of  service using unleaded fuel,
and 99 percent effective  after an additional  11,000  miles using
a high concentration  (20 percent)  methanol/gasoline blend.[4]

     The  ARCO  in-use data  suggest  that  effective,  durable,
low-cost  fillpipe  seals   (rotary  seals  or  cone  seals)   are
feasible for LDVs  for in-use service to at least 50,000 miles
over a  two-year  period.  The  available data  is not conclusive
as  to  which  type   of   seal   is  preferable.   The  important
question, which  has.  not  been fully  addressed,  is  whether the
fillpipe seal will be effective throughout the full useful life
of  a  vehicle.    Remaining  effective  implies  no  significant
deterioration, contraction,  or  expansion  problems under normal
environmental conditions  such that  the seal fails to achieve a
leak-free connection with the fuel  nozzle,  or  the nozzle cannot
be  inserted  through   the  fillpipe seal  at all.    At  this time,
durability data do not  exist out to  the full  average life  of a
typical  LDV,  100,000 miles  (10  years),  or  120,000  miles  (11
years)  for  LDTs.  However, the  durability data  up  to 65,000
miles  indicate  no reason why  the  seal would  not  continue  to
perform  over  its  full  life.  Given this  durability  data  to
65,000 miles, the relative  simplicity of the  system design, and
the  nature  of   its   use,   it is  reasonable  to  project  that
full-life performance should occur.
     3.
Effect on Gaseous and Evaporative Emissions
     The  work  conducted  by  API   indicated   that  purging  the
refueling   vapors   had   no   significant   effect  on  exhaust
emissions.  However,  it  should be  cautioned that the  tests were
conducted  on  1978  and  earlier  model  year  vehicles  which  had
emission  levels  higher  than  today's  new vehicles.   Also, test
procedures  for  measuring refueling emissions  have  not yet been
fully  developed  and   it should  be  recognized  that  the test
procedure  requirements  for  purging the  vapor  recovery canister
could impact exhaust  and evaporative  emissions.  However,  it  is
expected  that  through  proper design  of  the  onboard control
systems    (taking    into    consideration    appropriate   purge
requirements),  increases  in  exhaust  or evaporative  emissions
can be avoided.
                              C-7

-------
     c.
                              -4-
Feasibility for LDTs and HDGVs
     Even   though   the   API   feasibility   evaluation   project
involved  only  LDVs,  onboard   control  technology  should  be
directly and fully adaptable  to  LDTs,,   LDV and LOT fuel systems
are  practically identical,  and both  use  similar hardware  to
comply with  the 2.0  g/test evaporative  emission standard.  The
primary difference between LDVs and  LDTs  is  in  the  fuel tank
specifications. . Analysis  of  1984  certification information and
discussions  with  the manufacturers  indicate  that, on  average,
LDT fuel tanks  are about  25 percent  larger than LDV fuel tanks,
and  about  20 percent  of LDTs  use dual  fuel  tanks.    A larger
volume fuel  tank would require  more  charcoal in the canister to
accommodate  the increased volume of refueling vapors,  and LDTs
using  dual  fuel tanks  may require  a  separate  onboard control
system  for  each  tank.   However,  neither of  these differences
has  an  effect on  the   conceptual   design  or   technological
feasibility of  an onboard control system.

     The  above  considerations   apply  equally  to  many of  the
smaller HDGVs  (those  less than  14,000  Ibs gross vehicle weight
(GVW)).   Approximately  65  to   70  percent of  all  HDGVs  are
essentially  LDT derivatives.[5]   These  HDGVs  are essentially
the  same as their  parent LDTs  in  their  basic  chassis,  body and
powertrain  designs,   but  have   been   classified  as,  HDGVs  for
purposes of emission control because  their GVW,  frontal  area,
or  curb weight are  just  above  the  LDT/HDGV  cut-off points.
Although these  characteristics  would  have  an  effect  on exhaust
emissions,  they would  have no  effect  on  the  ability  to comply
with an  onboard vapor recovery  requirement.   The  key  parameter
which  influences  feasibility  is  fuel  tank  volume.    Most  of
these  smaller HDGVs have  fuel tank sizes similar to the heavier
LDTs,  so the  onboard  systems   used  on  LDTs  could be applied
directly  to the smaller  HDGVs.   For  those smaller  HDGVs with
larger  fuel tanks,  larger charcoal  canister  volumes  could be
utilized.

     The application  of an onboard  control requirement to many
of  the larger,  heavier GVW HDGVs  is  somewhat  more complicated.
HDGVs  in this  group  are sold  in  many different  configurations
with  different fuel  tank sizes and  locations.   Also,  many of
these  HDGVs are  sold  initially as  incomplete  vehicles  by the
primary  manufacturer  to a  secondary manufacturer.  In the most
common case, the primary  manufacturer produces the chassis and
the  secondary  manufacturer adds a payload-carrying device.  In
                               C-8

-------
                              -5-
some  cases,  the overall  vehicle  fuel capacity  is increased by
the  secondary manufacturer.   In  these  cases  a  problem might
arise  because the  primary manufacturer  would have  to  certify
the onboard  control  system before it was  sold  to the  consumer,
but  the secondary  manufacturer  could  affect the  integrity of
the  system.    Many  of  these  problems  are   similar   to those
encountered   and.  resolved  in  the   recent   HDGV  evaporative
emissions  final   rule,   which   suggests   that   implementation
problems  can  be   solved.   Also,  for  the foreseeable  future,
these heavier  GVW  HDGVs  will be using leaded  fuel and will not
have  a  filler neck  restrictor.   Thus,   the  onboard  control
system  for  these  HDGVs  would  require the additional hardware
associated with  the filler  neck  restrictor  already  present on
vehicles using unleaded  fuel if  they  were  to  use fillpipe seals
similar to those used on LDVs and LDTs.

     Although  application  of an onboard control  requirement to
the heavier  HDGVs  is not as straightforward  as  for the  lighter
weight  HDGVs  and  may be more  costly, there does  not  appear to
be  any  technological  reason  why onboard  control would  not be
feasible for  the  heavier GVW HDGVs.   One  possible approach for
applying an  onboard  control requirement to HDGVs  if  costs were
excessive, would  be to  require  control for  the  lighter weight
HDGVs  (under  14,000  Ibs  GVW)  and defer control  for the  heavier
weight HDGVs  (over 14,000 Ibs GVW).
     D.
Safety Considerations
     In  addition   to  concerns   about  the   performance  and
durability  of  onboard control   systems  for   LDVs,  LDTS,  and
HDGVs,  there are  some  potential  safety considerations  which
require evaluation.   If  a blockage of  some  type  occurs  in the
line  from  the  fuel  tank to  the  charcoal  canister,  pressure
buildups within  the  system may lead  to damage  of the fillpipe
seal  and  possibly  a  spurt  of gasoline  from  the  fuel   inlet.
Second, if the  automatic  shut-off of the gasoline nozzle fails
to operate properly,  an  overfill  of the  tank  could occur which
also could result in  damage  to the fillpipe  seal and a spurt of
fuel.  Third, there  is also the  possibility that  a failure of
the vapor/liquid separator and rollover check  valve in the  line
from the fuel tank  to the canister and an  improperly operating
automatic  gasoline   nozzle  shut-off   could   lead  to  a   tank
overfill  and  fuel   flowing  up   the  line  and   poisoning  the
canister.
                               C-9

-------
                               -6-
     These  problems could  likely be  resolved with  a  pressure
relief valve which  would  vent vapor  or gasoline overpressure  to
the  environment  should  problems  occur.   However,   prototype
pressure  relief  systems  have  not   been  fully   developed  and
failure  modes  have  not  yet  been   adequately  identified  and
evaluated.   Also,  increasing the diameter of the  vapor line
from the  fuel  tank  to the charcoal canister  and  increasing the
vapor  flow  capacity  of  the  vapor/liquid separator   and  the
rollover  check  valve  should  decrease  overpressure   problems.
Thus while  some  questions relative to  the  safety  of an onboard
system remain  unanswered, any problems  should be solvable with
direct engineering  effort.
     E.
Summary
     The work  conducted  through API and  later  by ARCO  suggests
that onboard  control is technologically  feasible for LDVs, and
evidence  is  that in-use durability of these  systems should be
excellent.   Due to  the fundamental  similarities  between LDVs
and LDTs, onboard control  should also  be  feasible for LDTs.  In
fact, the onboard systems  would likely be nearly identical with
the possible exception of charcoal  canister size.

     The demonstrated  onboard  technology  also appears adaptable
to HDGVs.  For those HDGVs  of  less than 14,000  Ibs GVW  (65-70
percent  of  all  HDGVs),  the  application  of  onboard technology
would in  all likelihood essentially be accomplished through an
extension of  LDT systems.   The only major difference  might be
larger canister  sizes  to accommodate the  larger  fuel tanks used
on some of these HDGVs.   Onboard systems for  the heavier HDGVs
(those whose GVW exceeds  14,000  Ibs)  would  be  somewhat more
complicated and costly, but nevertheless appear practicable.

     It should be possible to minimize  any  effect of an onboard
vapor recovery  requirement on  exhaust  and  evaporative  emission
levels through the proper design of the onboard system.

     There are  some  potential  safety  considerations which must
be identified,  evaluated,  and  resolved.  However,  it is  likely
that these  can  be  adequately  addressed  through  the use  of  a
pressure relief valve within the fuel delivery system.

     Although  control  systems  could be  applied  to HDGVs.,  we
have not quantified  the costs,  benefits,  and  cost effectiveness
for these vehicles  at  this  time.   HDGVs  comprise only  about  3
percent of the gasoline-fueled  vehicles produced  each  year and
represent on the  order  of  5  percent of annual  nationwide total
gasoline   consumption.[6]     The   remainder   of    this   paper
concentrates on the costs, benefits, and  cost effectiveness for
LDVs and LDTs.
                              C-10

-------
                              -7-
III. In-Use Performance of Onboard Systems
     A.
Introduction
     Losses in  the  effectiveness of  in-use  onboard systems can
occur  through  two  mechanisms:   tampering  or  deterioration  of
the   efficiency  of   the   system.     Tampering   occurs   when
individuals  purposely  disable  part  or  all  of  the  onboard
control system.   Tampering could occur  with the  fillpipe seal
and   the   charcoal   canister   and    related   hoses.     System
deterioration  occurs  when  control  efficiency  of the  onboard
system  declines  with  mileage  and/or  time.    Either  mechanism
renders   the   onboard  vapor   recovery  system   partially  or
completely   ineffective.    The  projected   effects   of   these
mechanisms on onboard system performance are discussed below.

     B'.    Tampering

     1.    Fillpipe Seal Tampering

     It is possible  that  fillpipe  seals could be  subject  to
tampering  similar  to  that   reported  for   the tampering with
fillpipe  restrictors  in  vehicles  using  unleaded fuel,   since
violation  of the  leaded fuel restrictor  would  also destroy the
vapor recovery  seal.  Fillpipe  tampering  data  is available from
the  National   Enforcement   Investigations   Center   (NEIC).[7]
These data show substantial differences  in  fillpipe restrictor
tampering  in  areas  which  have  inspection/maintenance   (I/M)
programs  versus  non-I/M areas and different levels of  fillpipe
tampering  for  LDVs.  and  LDTs.  (See Figure B-l  of  Appendix B.)
A  linear   regression  of  this  fillpipe  tampering data  versus
mileage for 1982 produces the following results:
      LDVs:
      LDTs:
         I/M Areas:
         Non-I/M Areas:

         I/M Areas:
         Non-I/M Areas:
TAMP = -1.43 + 1.14(M)
TAMP = -0.78 + 1.65(M)

TAMP = 3.55 + 1.14(M)
TAMP = 10.6 + 1.65(M)
Where:
      TAMP = Tampering   incidence   expressed   in   percent   at   a
             particular  vehicle mileage.

         M = Mileage/10,000 miles.

     It  should  be noted that  the  tampering increase rates  (the
change  in  tampering  incidence  with mileage)  for  LDVs and  LDTs
are  the  same.   This  was taken to  be  the  case because the  size
of the LDT  sample was too small (323 LDTs  versus  1,999 LDVs)  to
                               C-ll

-------
                               -8-
allow meaningful  rates to be.determined.  However, the  LDT  data
were  used  to  derive  a  mean  tampering level  for LDTs at  the
average  LDT mileage, with the  LDV tampering rate being applied
to that  single point.

     However, these  tampering  rates are conservatively  high  for
the  late  1980's  and  beyond  when  an  onboard  vapor  recovery
requirement  might   be   implemented.    The   primary   motive   for
fillpipe  tampering  is  to  permit  the  use  of  somewhat   less
expensive leaded  fuel  in LDVs  and LDTs designed to use  unleaded
fuel.  The  tampering rate itself depends on the availability  of
leaded  fuel, the  leaded  to  unleaded  fuel  price differential,
and  the  actual difficulty and"  other  effects  of  the tampering
process  itself.   The  tampering  rates given above  are  based  on
data  gathered  in the  Summer  of  1982,  when  leaded  fuel  was
readily  available at  a  differential  of  about five  cents  per
gallon and  fillpipe  tampering was  a  relatively simple  process,
usually  with no  effect  on  the  integrity  of  the  filler   neck
itself.

     However, in  the late 1980's and  beyond,  leaded fuel  will
be generally less available  due  to  lower  overall  demand,  and
with  less demand  it  is  possible  that  the leaded  to  unleaded
fuel  price  differential would  decrease.   In  addition,  there
were only a handful  of I/M programs  in place  in 1982 when  this
data was  gathered.  As  more  I/M programs  are  implemented  over
the  next few  years  tampering  should  decrease.   Perhaps   most
importantly,   the   onboard    control   requirement    could    be
implemented  with  a  certification performance  standard  such  as
the   parameter   adjustment   requirement  for   carburetors    on
gasoline-fueled  vehicles.   This  requirement   would   force  the
design of filler  neck  restrictors and  fillpipe seals which are
a  more  integral  part  of  the  fillpipe,  thus  reducing  the
accessibility  and  success  of   tampering.   Therefore,  it   is
reasonable  to  project  that  fillpipe  tampering  will  decrease
markedly  by the  later 1980's.   After  briefly  considering  the
rate  of  tampering  with  charcoal  canisters  and   hoses,  a
composite tampering  rate will be determined for  LDVs and  LDTs
if  fillpipe tampering  is  reduced by  50  percent due   to  the
reasons discussed above.
     2.
Charcoal Canister and Hose Tampering
     Tampering   with   the   charcoal   canister   and   related
connecting  hoses would  also destroy  the  effectiveness  of  an
onboard  vapor  recovery  system.    Since  the  control  approach
expected  by  EPA   assumes  an   integrated  onboard/evaporative
emissions control system,  currently available  data on tampering
with  evaporative emission systems  (canisters/hoses)  would  be
directly  applicable  to onboard  controls  as well.   Evaporative
                               C-12

-------
                               -9-
 emission  control  system tampering rates are also  available  from
 the   National  Enforcement  Investigations  Center   (NEIC)   for
 1982. [7]   The tampering rates are  different  for LDVs and LDTs,
 but  not  different  in  I/M  versus  non-I/M  areas,   since-  the
 evaporative  emissions  control  system  is  normally  not  checked
 during  I/M.    (See   Figure  B-2   of  Appendix  B.)   A   linear
 regression   of   this   most  recent   (1982)   NEIC   evaporative
 emissions  control system tampering  data  provides the  following
 results:
     LDVs;

     LDTs:
TAMP = -0.55 + .360(M)

TAMP =  2.85 + .360(M)
TAMP  and  . M  are  as  described   previously  above,  and   the
explanation  regarding  the derivation  tampering  incidence  and
tampering rates for LDVs and LDTs  is also  applicable.

     3.    Composite Tampering Rates

     It  would be  convenient  to  have  composite  tampering  rate
equations  for LDVs and LDTs  for  computing  the in-use emission
reductions  expected  from  an  onboard  vapor   recovery   system.
Since   tampering   with   either    the   fillpipe  seal   or   the
canister/hoses would  disable  the  vehicle's  onboard system,  the
slopes  and  intercepts of the  different  tampering  equations
given   above  could   simply   be   added   for  LDVs   and   LDTs
respectively.  However,  this would overstate  the  total   effect
of tampering,  because  some vehicle owners  tamper  with both  the
fillpipe and  the  charcoal canister and  hoses.   Since disabling
either  would eliminate  the effectiveness of  onboard control,
just adding the equations would lead to some double counting.

     To  determine  the degree   of overlap   in tampering,   the
National Enforcement  Investigations Center data discussed  above
was analyzed.  After  the overlap  tampering was  accounted  for, a
linear  regression  of  the  combined data sets was  conducted  and
the following regression equations were obtained:
       LDVs:
       LDTs:
   I/M Areas:
   Non-I/M Areas:

   I/M Areas:
   Non-I/M Areas:
TAMP = -1.47 + 1.442(M)
TAMP = -1.52 + 2.114(M)

TAMP =  6.42 + 1.442(M)
TAMP = 13.67 + 2.114(M)
     The  tampering  levels   in  I/M  and  non-I/M  areas  can  be
weighted  (40  percent I/M and  60  percent  non-I/M)  according to
the fractions of  the U.S.  population residing  in  the two types
of  areas.   The  results  of  this  tampering  rate  weighting  are
shown below.
                               C-13

-------
                              -10-
     LDVs; TAMP = -1.5 + 1.8452(M)

     LDTst TAMP = 10.77 + 1.8452(M)

     The  weighted  composite  tampering  rate  equations  given
above  are  based  on  current  tampering  rate  data.   As  was
discussed above,  it is likely  that tampering  will  decrease in
the future,  thus  improving the overall  in-use effectiveness of
an onboard  vapor  recovery program.  To  estimate  this potential
decrease  in  tampering,   the  portion  of  the  above  composite
tampering rates associated with the fillpipe  will  be reduced by
50 percent.   The portion  of  the  composite  tampering  rate  due
solely  to  tampering  with  the  fillpipe was  taken  to be  the
difference  between  the composite  rate  and  the  tampering  rate
associated with  the evaporative HC  control  system.   The  result
of decreasing this difference by 50 percent is shown below.

     LDVs: TAMP = -1.0 + 1.1026 (M)

     LDTS: TAMP - 6.81 + 1.1026 (M)

     These   projected   weighted   composite   tampering    rate
equations  will  be   used   in  calculating  the  in-use  emission
reductions for onboard vapor  recovery  control.  These equations
will be  taken as applicable  to 1988 and later model  year  LDVs
and LDTs.
     C.
Deterioration
     As was  discussed above ,in  Section II.B. 2.,  the  ARCO seal
durability  data  show no  deterioration  of  the  onboard  vapor
recovery system effectiveness  with  mileage.   This is consistent
with historical EPA certification  information  which  shows that
the  efficiency  of  evaporative emission control  systems  (which
are similar to vapor  recovery  systems),  do not deteriorate with
mileage.   Limited  in-use  testing  of  LDV  and  LOT  evaporative
emission  systems   shows  that  these  systems  do  function  as
designed.  At  the  same  time,  some  small  loss  of effectiveness
with  mileage,  on  the  order  of  a  few  percent, would  appear
reasonable  due  to  contamination of  the  charcoal,  channeling,
aging,  leaks,   etc.   With   no  data,   it   is   not  possible  to
estimate  this  loss  quantitatively.   However,  the  tampering
rates of the previous section .appear  large  enough to overwhelm
any  expected  loss  in efficiency due  to  deterioration.  Thus,
the  losses in  system effectiveness  due  to  tampering  will  be
taken to include  any  losses due to deterioration.

IV.  In-Use Emission  Control Effectiveness

     An  estimate  of  the   annual  or   lifetime  HC   emission
reduction  potential  of  vapor-controlled  LDVs  and  LDTs   is  a
                               C-14

-------
                              -11-
function  of  annual  or  lifetime  mileage,  the  vehicle's  fuel
economy,   the  uncontrolled  refueling   emission  factor,   the
control  system  effectiveness,  and  an  adjustment  factor  that
accounts   for   a   loss   of  effectiveness    in-use,    (i.e.,
tampering).   This  relationship on an  annual  basis is expressed
below:
           HC =  (VMT)(EF)(NSEFF)(NTAMP)
                           MPG
Where:
     HC =


    VMT »

    MPG =

     EF =
  NSEFF
Average  annual HC  emission  reduction  per  vehicle,
grams.

Average annual mileage, miles.                  '""

Average in-use fuel economy, miles per gallon.

The uncontrolled  refueling loss  emission  factor, or
4.54 g of HC per gallon of dispensed gasoline.

Onboard  control  system  efficiency of  new vehicles,
or 0.98.
  NTAMP =  An  adjustment  factor  which  discounts   for   in-use
           tampering.  NTAMP equals  (1-TAMP) for any  given year.

     Estimates of  in-use  (over  the  road)   fuel  economy for new
LDVs   and   LDTs   are   based   on   projections   of    fleetwide
improvements for 1988 and  later  years.   Annual vehicle miles of
travel  estimates  are  those  used  in the  EPA  emission factors
program.[8]   This   information   is  contained   in   detail  in
Appendix B.  The new vehicle control 'system  efficiency of 0.98
and the range  of  tampering rates as a  function of mileage were
discussed above.

     The uncontrolled refueling  loss  emission factor  (4.-54  g
HC/gal) is based on  recent work  conducted  by the California Air
Resources  Board  (CARB).[9]   This  figure  is   11  percent  larger
than the emission  factor contained  in  the  EPA emissions  factor
document  (AP-42).[10]    The  CARB emission  factor  was selected
over the AP-42 emission  factor  for three  reasons.   First, the
EPA emission  factor  document  expressed uncertainty  about it's
emission factor  value.   Second,  the  CARB factor  is  based  on
data  at least  5  years  more  recent  than  the AP-42 • emission
factor.  And third,  an  increase in  the emission  factor  can be
explained  by  the   steady  increase   in   gasoline   volatility
(expressed as Reid Vapor Pressure)  over the past 10  years.[11]
In  fact,   information  recently  submitted  to  EPA   by General
                               C-15

-------
                              -12-
Motors indicates that.4.54  g/gal may be  conservatively low for
today's commercial gasolines.[12]

     This  equation  and  these factors may  be used  to estimate
the  lifetime  emission  reductions  for LDVs  and  LDTs.   For any
given  model  year,   the  only  variable   would   be  the  fuel
consumption.   The  remaining  factors  for   each  year  of  the
vehicle  life  can  be  determined  and summed  to  get  a  single
factor representative  for the entire vehicle average  lifetime.
Using an average lifetime of 100,000  miles  for LDVs and 120,000
miles for  LDTs,  and assuming  that the  tampering  occurs  at the
midpoint  in  each  year,  the   lifetime  HC   reduction  can  be
calculated for any model  year.
     HC (tons) =
                                      AL
                    (NSEFF)(EF)
                  ( 453 .6 ) ( 2 , 0 0 0 ) ( MPG
                          (VMT
NTAMP )
Working  through  the   mathematics   of  this  calculation,   the
following  equations have  been  determined  for  calculating  the
lifetime  tons  of  HC  emission  reductions  for  LDVs  and  LDTs.
These  are  based on  an average  lifetime (AL) of  100,000  miles
for LDVs and 120,000 miles for LDTs.

     LDVS:  HC * .4683
                  MPG
     LDTS:  HC =
.5095
 MPG
With  these  equations, the  average lifetime  in-use  HC emission
reductions  from onboard vapor  recovery for  any  model year LDV
or  LOT  can   be   determined  using  the  in-use  fuel   economy
estimates  in  Table B-3  of Appendix  B.  For  example,  for  1988
model  year  vehicles, LDV  and  LOT  reductions  of   0.0178  and
0.0264 tons respectively per vehicle,,  would occur.   These  will
be  used  in a  later  portion  of the  analysis  to  calculate the
cost effectiveness.

     In  addition  to  computing  the annual  or lifetime emission
reductions  on  a   per   vehicle  basis,  the   nationwide   annual
reduction  in  the   overall  HC  emission inventory  can  also be
estimated.    Determining   the   reduction   in   the   annual  HC
inventory  for  any  given  year  is  a  relatively  straightforward
calculation   involving   the  annual  -gasoline  consumption  of
vehicles  employing onboard  controls,  the  emission  factor, and
the  in-use control efficiency  of those vehicles.   The  annual
                               C-16

-------
                              -13-
gasoline  consumption  of  controlled  vehicles  is a  function of
their  total  registrations,  fuel  economy,  and the  annual miles
of  travel.    These  can  be  expressed  mathematically as  shown
below.
     IR = (EF)(NSEFF)
         (453.6) (2,000)
                          REGxzSRxZNTAMPxZVMTxz
z=.
                                   MPG
                                           xz
                           REGyz SRyzNTAMPyzVMTyz
                                  MPG
                                     yz
The variables  are  the  same  as  identified  above, and  as noted
below:

     x   = LDVs

     y   = LDTs

     z   = time (years)

     IR  = annual HC inventory reduction (tons)

     REG = new registrations of  gasoline-fueled  LDVs or LDTs in
           each year z=l,n

     SR  = new vehicle survival  rate  of  gasoline-fueled LDVs or
           LDTs in each year

The values for these variables are given in Appendix B.

     Working  through  the  calculations  above,  the   following
annual inventory reductions  are projected  from  all  in-use LDVs
and LDTs with effective vapor recovery systems.
      1988
     41,200
              Annual  Reductions  (tons)

              1989         1990         1995
                                 2000
             77,500
      108,400
213,300
257,500
One can see that  as  a greater portion of  the  LDV and LDT fleet
employs  onboard  control,   the  annual  reduction  in  refueling
emissions becomes substantial.
V.
Costs of Onboard Vapor Recovery
     Two  new  sources  of  information  on  the  costs  of onboard
vapor  recovery   hardware   have  become   available   since  the
                               C-17

-------
                              -14-
preparafcion  of  the  last estimates  shown  in  Appendix  A.   The
first is a June 1983  draft  report entitled "Manufacturing Costs
and Retail  Price  Equivalent  of  On-Board Vapor  Recovery System
For Gasoline-Filling  Vapors,n prepared by  LeRoy Lindgren under
contract to  API.   The  second is  a  January 1984  cost estimate
presented by  API  in their  final  report on  the  cost comparison
for Stage  II  versus  onboard control  of  refueling emissions.
The information  contained  in Lindgren's  report was  one input
used by  API  in their most  recent cost estimates  for onboard.
No updated cost .estimates from the auto industry were available
for this analysis.

     In this  section  of the  report,  the  Lindgren hardware cost
estimates will  be reviewed  and  discussed  first.   This  will be
followed  by  a   discussion  of   the  onboard  cost  estimates
developed by  API  and an update  of the estimate  of the  cost of
an onboard  vapor  recovery  system for a  current technology LDV
or LDT.  This section  will  close  with a discussion of the sales
impact of an onboard control  requirement.
     A.
Lindgren Report
     The Lindgren  report to  API provides  an estimate  of both
the manufacturer  (or vendor)  cost  and retail  price equivalent
(or customer  cost)  of  a  complete onboard  control  system.  The
estimates   of  these   two    costs   are   $12.95   and   $29.85,
respectively.   Tables  1  through 6  of Appendix  C  (taken from
Lindgren's draft report) contain the bases for these costs.

     Lindgren   estimated   hardware   costs   for    the    system
demonstrated by API  in  1978.   This system  is  shown in Appendix
A/ Figure  A-l.  This  system  was  a  fillpipe  seal, additional
charcoal canister,  and  separate plumbing  for  the evaporative
emissions and  onboard  recovery systems.  Lindgren  attempted  to
update  these  designs for  changes  in  LDV  engine  and  emission
control  technology  which have  occurred  since  1978.   However,
this  was  not  done  properly  in   every  case,  and costs  for
components   already   on   current   technology  vehicles   were
attributed  to the cost of  an  onboard  vapor  recovery  system.
For example,  costs  were  included  for a  leaded  fuel restrictor
and modifications related to the electronic  control unit, both
which would be present on current vehicles.

     Although  most of  Lindgren's component  manufacturing  costs
appear  reasonable,  there are two  other  major  deficiencies  in
the analysis.   First,   arithmetic  errors were made  in  several
places  in  the analysis,  and  an error  was  made  in calculating
the costs  after corporate  and  dealer markups  were added.   A
markup  factor  of  2.3 was used instead of  1.8 as  specified  by
Lindgren  in  his  report.   As  was  mentioned  above,  Lindgren
                               C-18

-------
                              -15-
estimated a  customer cost o-f  $29.85.   Correcting  these errors
and applying  Lindgren's 1.8 standard  markup factor  brings the
customer  cost  down  to  $24.06.   Second,  based  on  previous
analyses  of  Lindgren's  cost  methodology,  standard  absorbed
overhead and  profit  absorption rates  appear  to be used  at the
corporate  and dealer  levels,  rather   than incremental  rates.
This   results   in   a   substantial   overestimation   of   the
contribution  of  overhead and  profit  to onboard  control costs.
As will be discussed below,  it is believed  that  an incremental
approach  to  corporate  and  dealer  overhead  and  profits  is
appropriate for emission controls, resulting  in a markup factor
of 1.27 rather than  1.8.  Using this  incremental markup factor
brings Lindgren's estimate to $17.72.

     Thus,  Lindgren's  cost  estimates  cannot  be  used  directly
here,  but will have  to  be  modified  to  include  only the costs of
components  incremental  to those  already on current  technology
LDVs and to  more  accurately  reflect  appropriate  corporate and
dealer markups.  This process'is described  in  the next section,
after  a review of the cost estimates released by API.
     B.
API Cost Estimates
     In  their  recent  final  report  comparing  Stage  II  and
onboard costs,  API  presented  their  updated cost  estimates for
onboard     controls.[13]       API     did      not     present
component-by-component  cost  estimates,   but   only  a  fleetwide
average  cost  of  $13.43.   This  estimate  included  different
canister sizes for LDVs and  LDTs  and the  need  for two canisters
on   some   vehicles.    The  fleetwide   average   estimate   was
calculated using a cost of  $12.07  for  LDVs or  lighter LDTs with
one  canister,  $14.47  for  LDVs  or   lighter  LDTs  with  two
canisters, and  $20.87 for  all heavier LDTs.   These  costs were
then  weighted   70   percent,   20   percent,   and   10  percent,
respectively, representing  the projected  portions  of the  total
vehicle population.   When system  development  and certification
costs are  added,  this cost  rises  to $15.26 per  vehicle  ( 1983
dollars).

     The  API  cost  estimates  did  not include a  retail  markup
because they were  not  certain  about  what  markup  figure was
appropriate  or  how  the  vehicle  manufacturer  or  dealer   might
choose  to  absorb or  pass on  costs.  If  the  markup  factor  of
1.27 is applied,  a  fleet average cost of  $19.38  per  vehicle is
obtained.   In  the section  which  follows  directly,   it  will  be
seen  that the  marked  up API  figure  is  in  the  range  of the
updated estimate of this  report.
                               C-19

-------
                              -16-
     C.    Updated Estimate

     1.    Hardware Costs

     The  cost  estimates  here  are   based   on  an  integrated
evaporative emissions  and  onboard  control system  as  opposed to
the  separate  systems  on  which  the  Lindgren  and  API  cost
estimates  are  based.   This  design  is  expected  to  be  the
approach  preferred  by  the   manufacturers   because   it  makes
optimal use of  limited underhood space,  simplifies the design,
and reduces cost.   The key feature  of this  design is  that one
large  charcoal  canister  can be  used  for  evaporative emissions
and onboard control rather than two separate canisters.

     The first step in developing the  updated  cost estimate was
to decide  on what components would make up  the  system.   The
components selected were  mostly  the same  as those used  in the
API  demonstration program  and  priced  by  Lindgren.   However,
there  are  several   important  differences.   First,   as  was
mentioned  above,  an  integrated  onboard/  evaporative emissions
control   system   was   assumed,   thus   eliminating   obvious
redundancies , between  the two  systems.  Second,  the  cost  of  a
pressure relief valve  was  included  which  might be necessary as
discussed  previously  in  Section   II.   D.    And,  third,  the
components which  are  present  on current vehicles  but  were not
present on the 1978  vehicles  used  by Lindgren  were excluded.
Once the  components  of the  system  were  determined,  vendor and
retail price  equivalent  cost  estimates  were  developed  using,
and  in some  cases modifying,  the  manufacturing  cost estimates
provided by Lindgren.

     The expected  components  and their  costs  are summarized in
Table  1.   The vendor  costs include  material,  direct  labor, and
direct overhead and have  been  multiplied by a  factor  of  Ii4 to
account for indirect  overhead  and  profit  at the  vendor levels.
The 1.4 factor  for vendor  allocation and  profit  was  taken from
Lindgren's  methodology  and   represents  a   standard  absorbed
overhead rate and  rate of  return for this industry.  These full
rates  are  appropriate  here  because  the  production  of  the
emission control equipment is  the primary business activity for
the vendor and is not incremental in nature.

     These  vendor  costs  were  then   multiplied  by  1.27  to
estimate  the  corresponding retail price equivalent,  accounting
for corporate and  dealer overhead and  profit.   Lindgren applied
a  factor   of  2.3  to  account  for  these  factors, though  this
appears to be an error, since his  own  methodology specifies  a
factor  of 1.8.   The  1.8  factor  appears,  again,   to  include  a
standard   absorbed  overhead   rate   for  both  manufacturer  and
dealer and standard profit margins  for both.  These figures are
                               C-20

-------
                                -17-
                             Table 1
              Onboard Vapor Control Hardware Costs
              	(1983 dollars)
 Component  or Assembly
 Charcoal Canister LDV/(LDT)
 Purge Control Valve
 Liquid Vapor Separator
 Fillpipe Seal
 Pressure Relief Valve
 Hoses/Tubing
 Miscellaneous Hardware
Vehicle Assembly
Systems Engineering/Certification
                                           Incremental Costs
Vendor
$3.99/(7.83)
0.74
0.71
1.12
0.44
1.90
0.40
—
--
Retail Price
$5.07/(9.94)
0.94
0.91
1.42
0.56
2.41
0.51
IsOO
0.50
LDV Totals:
LDT Totals:
Vendor
Vendor
 $9.30  Retail $13.32
$13.42  Retail $18.19
                              C-21

-------
                              -18-
 not appropriate here  because  adding emission  control  equipment
 is   only  incremental  to  the  primary  business  of  assembling
 automobiles,  and overhead  and applied  assets  are  not  entirely
 variable  with  respect  to  vendor  cost,  but  have  significant
 fixed  components.   This is  particularly true  for the  dealer,
 who would   experience  almost  no  effect  due   to the   added
 equipment.   The  1.27  factor  is  the  result of  an  incremental
 analysis  of corporate and dealer  overhead and profit which  was
 performed as  part  of  a  recent  EPA  mobile  source regulatory
 analysis  for  LDVs  and  LDTs.[14]

     The  size  of  the  carbon  canister  in  Table  1   is  that
 associated  with a fuel  tank which would give an  in-use  driving
 range  of  about  300  miles.   Using  the  in-use  fuel   economy
 projections  of Appendix B  (Table  B-3)  for  1985-90, LDVs  would
 require an  average fuel tank  size of  10-13  gallons, LDTs  would
 require  an average  fuel  tank  size of  14-18  gallons.   To  be
 conservative,  in each case,  the  higher  end  of  the  ranges  in
 fuel tank sizes was  used to  size  the canisters.

     As shown in Table  1,  an onboard vapor  recovery system  is
 expected  to carry  a  consumer cost of $13.32 for LDVs and  $18.19
 for  LDTs.  Those LDTs using dual-fuel  tanks   (approximately  20
 percent)  may  require  two separate onboard control systems  for  a
 total cost  of $36.38.   This  is  a conservative assumption  since
 costs  could  likely  be  reduced   by  using  one  large  charcoal
 canister  rather than two separate  canisters.

     A  fleetwide  estimate   for   all   LDVs   and  LDTs  can   be
 determined by  sales  weighting  the costs  given above.  Using  the
 projected  sales  for  1988  from  Appendix  B  (Table B-3),   and
 assuming  20 percent  of LDTs  have dual-fuel tanks, the fleetwide
 average cost  is calculated  to be  $15.08  as shown  below.   For
 future  calculations  this   cost   will  be  rounded  to  $15   per
 vehicle.

 (10.582M)  ($13.32) +  (2.768M) ( (.8) ($18.19) +  (.2) ($36.38))=$15.08
                           13.35M

     This estimate of  $15.08  is  comparable to  API's estimate  of
 $19.38  after   application   of  the  1.27   markup  factor   and
 Lindgren's estimate  of $17.72 after corrections  and  using  the
 1.27  incremental  markup  factor.   The  main  reason  for   the
difference between this  estimate  and those developed by API and
Lindgren    is    because     EPA    assumed     an    integrated
onboard/evaporative emission control approach as  opposed  to two
 separate systems.

     2.     Differences   Between   Past   and  Current EPA  Cost
           Projections"———

     EPA's February  1980  report  projected a fleet average cost
 of  $19.70.   When inflated  to 1983  dollars,  this  cost  becomes

                              C-22

-------
                              -19-
$26  per  vehicle.   There  are  three  reasons  for   the  overall
decrease  of $11  between this  cost  estimate and  the previous
estimate.

     First,  the  1980  projection  used  a   1.8  retail  markup
factor.   As discussed  above,  it  is  now  believed   that  a  1.27
markup  factor  is more  appropriate;  this change  alone accounts
for approximately 70 percent of the cost difference.

     The  second  reason is related  to changes  in system mixes.
The 1980  projection assumed  higher costs  in some  cases  due to
the use of two  canisters  rather than  one  larger  canister,  or
due   to    a    manufacturer    not    choosing   an    integrated
onboard/evaporative emissions  control approach.   This accounts
for another 23 percent of the difference.

     Third,   there   have  been  changes   in   the   components
anticipated to make up  an onboard control  system and  the prices
for some components (notably the  fillpipe  seal).  This accounts
for the remaining 7 percent of the cost difference.
     3.
Fuel Economy Impacts
     As was stated in  the  previous  EPA report (Appendix A), the
implementation of  an  onboard  vapor recovery  requirement would
not  be  expected  to   impact   LDV   or  LOT  fuel  consumption.
Hydrocarbons  retained  by  the  onboard canister  represent about
0.1 to  0.2  percent of  vehicular  fuel consumption.   The  use of
this fuel by  the engine could  thus  be expected to decrease fuel
consumption  by  this  amount.   However,  the  additional  fuel
needed  to transport  the added  weight  of the  onboard system is
also in this  range.  Thus,  no  net  change in fuel consumption is
expected.
     4.
Overall Cost Estimate
     As  discussed in  the previous  two  sections,  the  updated
LDV/LDT  cost  estimate  is about  $15 per  vehicle and  there is
adequate explanation as  to why this cost  is  well below that of
February  1980.   However,  there  are still  reasons , to  believe
that  the  total  cost   of onboard   control   could  be  somewhat
greater than $15 per vehicle.

     One,  this  figure  includes  primarily   hardware  cost  and
excludes   any   costs   associated   with   possible   fuel    tank
modifications,  modifications  to  the  vapor  line  and  rollover
check  valve  between   the  fuel  tank  and  the  vapor  canister,
modifications  to  make  the  fillpipe more  tamper-resistant,  and
general    packaging     costs     to    fit    the    integrated
onboard/evaporative emissions  control  system into the vehicle.
                               C-23

-------
                              -20-
Also, there  may be  some  cost related  to manufacturer-specific
electronic control unit  (ECU)  modifications.   For example, some
manufacturers may  desire  to use  specific  canister purge cycles
which may  require  reprogramming or modification  of their ECUs.
Finally, since  the actual  pressure  relief valve discussed above
has not been  identified,  there is some  uncertainty in the cost
for that component.

     Two,  except  for the  allowance of  a  dual system for LDTs
with two  fuel  tanks, the  system considered  herein is somewhat
ideal.     Completely   integrated   onboard/evaporative  emission
control systems are  assumed in every case, and  this simply may
not be  possible.   For reasons of canister production economies
of  scale,  underhood  packaging  restrictions,  or   for  unique
vehicle models, manufacturers may  choose a  non-integrated two
canister  system similar  to those  considered by  API.   As was
discussed  above,   a  non-integrated  system  would  increase  the
costs over those shown in  Table 1.

     Three, in  the final  analysis,  the  actual canister size and
purging system  will  depend on  the details of the  test procedure
implemented   to  measure   compliance  with   an   onboard  vapor
recovery   requirement.     Factors  such   as   the   degree   of
interaction between  the evaporative emissions  and onboard test
procedures and  whether  the  charcoal  canister would  have to be
purged  during the exhaust emissions  test will  affect the size
of  the charcoal  canister and  the complexity  of  the   purging
system.   These  in  turn would  affect the  overall  cost  of the
onboard system.

     To account for  these  and  other potential costs, a range of
$15-25  per vehicle will be used  rather  than  the  single  cost of
$15 per vehicle.   While  the final cost is expected to be.closer
to $15  rather than $25, the  use  of $25  as  an upper  limit will
allow  the sensitivity of  any  subsequent  decisions to this cost
to be addressed.
     D.
Impact On Sales of LDVs and LDTs
     An  average  purchase  price  increase  of  $15  to  $25   is
 expected  to  have no discernible  impact, on the sales of  LDVs  or
 LDTs  and,  therefore,   no  effect  on  the  profitability  of  the
 companies  comprising  the  regulated  industry.  The  "own  price
 elasticity  of  demand"  for  LDVs  and  LDTs   (that  ignoring  any
 crossover  purchases in  other  vehicle classes)  is  approximately
 -1.0,  which  means  that for each 1  percent  increase  in  price,
 sales  drop 1 percent.   With the  price of  an average new  LDV or
 LDT  now exceeding  $10,000,  a  $15  to  $25  first price  increase
 would  be predicted  to decrease  sales  by  no  more  than  0.15  to
 0.25  percent.   However,  there  is   some  question  whether  the
                                  C-24

-------
                              -21-
 elasticity  of demand is even  meaningful  in measuring the  sales
 impact  of a  $15  to  $25 increase.  Such  an increase would  tend
 to  be lost in the annual  price  increases occurring at the  time
 of  model  year introduction.

      Furthermore,  onboard controls  are not  expected  to  affect
 operating and maintenance  costs, nor  significantly affect  the
 owner's   experience  of  refueling.   Thus,  there  should  be  no
 non-economic    resistance   which   will    affect   sales    or
 satisfaction.   In  the  long   term,  an  onboard  vapor  recovery
 requirement  should have no  perceptible impact on  the  sales  or
 profitability of  either  the manufacturers or  dealers.

 VI.   Cost Effectiveness

      The   cost   effectiveness  of  onboard   control   can   be
 calculated  using  the  LDV and  LDT  in-use  emission  reduction
 equations developed  in Section IV and  the  range  in the average
 costs of  control  calculated in Section V.   The in-use emission
 reduction varies  with each model year's vehicles  depending  on
 the fuel  economy,  and the  average cost varies somewhat based  on
 relative  sales of LDVs  and LDTs.   The 1988  model  year  will  be
 used  here,  since  it  is  possibly  the first model  year  in which
 an onboard requirement could be implemented.

     Referring to  Appendix B   (Table B-3) ,  the  1988  LDV and  LDT
 fuel  economies are  26.30  and 19.28 mpg  respectively,  and  the
 sales are 10.582  and 2.768 million,  respectively.   Using these
 fuel  economy  figures,  the  lifetime  reduction  for  LDVs is  0.0178
 tons  and  for  LDTs  the lifetime reduction  is 0.0264  tons.  Sales
 weighting  these   figures,  the fleetwide  average  lifetime  tons
 reduction  is  0.0196  tons.    Dividing  these  figures   into   the
 range of  fleet average  weighted  cost of  $15-25  per  vehicle,
 yields an average lifetime cost-effectiveness  value of $766  to
 $1,277 per  ton.   As  shown in Table 2,  this cost-effectiveness
 value falls   in  the  range of values   for  other   mobile   source
 related HC control strategies, though nearer  the end.

     On an  annual  basis,   the cost  effectiveness  is  somewhat
 larger.    Using  a  10-year vehicle life  for  LDVs and LDTs,  a  10
percent  discount  rate,   and  assuming  payment   in  mid-year,
annualization  of  the $15 to  25  lifetime  cost  yields  an annual
cost of $2.34  to  3.90.   Assuming  annual mileage is constant  for
 the   ten   years,   the   0.0196   fleet-weighted   lifetime    tons
 reduction converts  to 0.00196  tons annual  emission reduction.
The annual  cost  effectiveness is  then about  $1,194-1,990   per
 ton.  The simplifying  assumption  of   constant annual  mileage
 results in a  slight  overestimation  of  this  figure, since annual
mileage is  higher early in  the   vehicle's  life.   Nevertheless,
 this provides  a valuable additional way of looking  at  the  cost
                               C-25

-------
                               -22-
                            Table 2
            Cost Effectiveness of Mobile Source HC
              Control Strategies  (1983 $/ton)[1]
    Control  Strategy
Cost Effectiveness
     HDGV Evaporative Control
     HDGE Useful  Life
     LDT  Useful Life
     LDT  Statutory Standard
     HDDE Statutory Standard
     HDDS Useful  Life
     Interim High-Altitude Standards
     Onboard Vapor Recovery (with evap.  benefits)
     LDV  Statutory Standards
     Motorcycle Standards
     Onboard Vapor Recovery (w/o evap.  benefits)
     I/M
     Auto Coatings
     Transit Improvements
              $112
          $100-200
              $406
              $207
              $319
              $323
              $416
          $435-725
              $508
              $616
        $766-1,277
              $943
            $1,301
           $15,767
[1]   Short ton
                               C-26

-------
                               -23-
 effectiveness  of an  onboard  requirement,  especially  when  the
 cost effectiveness  of  an onboard requirement  is  compared to HC
 strategies  where the  cost  effectiveness  is  calculated  on an
 annual basis.

      This estimate  of the cost  effectiveness of  onboard 'vapor
 recovery_  only  considers  the emission  reductions  derived  from
 eliminating  refueling  losses.   However,  preliminary  data  from
 EPA s   emission   factors   program   indicates   that    in-use
 evaporative  emissions   appear   to  significantly   exceed  the
 lu3?^3.^6 ??  SKandaKd.*K ThlS  °CCUrS Pri^rily  because  in-usJ
 fuels typically  have  higher  volatility than  the  fuel specified
 for  certification   testing   and,  therefore,  produce  larger
 amounts of   evaporative  HC  which  cannot  be  adsorbed  by  the
 current charcoal canisters.  Preliminary estimates  of the level
 of these  excess evaporative  emissions can  be made, using  the
 f^anrfYrre^ly  availablue.  fr°™  EPA's  evaporative  emission
 factors testing program which is now in progress.   This program
 involves   evaporative    emission    testing   using    Indolene
 (certification)   and   commercial   fuel   in   carbureted   and
 fuel-inuected^ vehicles.   Based  on  preliminary data  from this
 program,  it  is  estimated that LDVs  have  evaporative emissions
 n^f  r,fn£e °,f • °'23  to °'44  g/mi Usin9  commercial  fuel  and
 0.16  to 0.24 g/mi using  certification  fuel, yielding an excess
 t?nAe  T^ ^.°-07  t°.°-20 g/mi.   A best  eatimate  at ?hiS
 time  based on this  preliminary data is evaporative  emissions of
 0.33   g/rai   using   commercial   fuel   and   0.20   g/mi   using
 TrSiJ1^!11*?"6!'  f?r an excess  of  °*13  9/mi-  Although datl
 is  not  available for LDTs, one would expect results in  the same
 ranges  since LDV and  LDT evaporative  control  systems  are very
 similar.   Simply multiplying the  best  estimate of  these  excess
 ?Ta?°^nJ1Ve Jeinissions  by  the  average lifetime for LDVs  and LDTs
 (100,000  and  120,000  miles,  respectively)   and  converting  to
  2S «yi«eAds  llfetime excess  emissions  of 0.0143  tons  for  LDVs
 and  0.0172  tons  for  LDTs.    The  fleet-weighted   LDV/LDT  per
 vehicle  excess  would be  0.0149  tons  of HC  lifetime  or  0.0015
 tons annually.

     Since   refueling   only  occasionally   coincides  with   the
 occurrence   of   evaporative  emissions,  the   larger   charcoal
 canister   associated  with  an  integrated  onboard/evaporative
 emission   control   system  could  also  control   these   excess
 evaporative emissions at  little or  no  extra  cost.   Adding  these
 benefits  to  those  from  onboard  control  improves  the   cost
effectiveness  by  approximately  43  percent.   if  all  excess
evaporative  HC   emissions  were  controlled,   the   lifetime  cost
effectiveness of  onboard  control  would become  $435  to  $725  per
ton  and the  annual  cost  effectiveness  would  become  $678  to
$1130 per ton.
                               C-27

-------
                              -24-
     As indicated  above,  the in-use  evaporative  emissions data
is  preliminary  as all  testing  has  not  been  completed.   As
additional data become available,  it  will be  possible to make a
firmer  estimate   of  excess   in-use   evaporative   emissions.
However,   regardless   of   the  magnitude   of   excess    in-use
evaporative emissions, the onboard  control  system does have the
potential   to   control   a   large   portion  of   these    excess
emissions.   This  additional  HC  control,  if credited  towards
onboard control,  would improve  its cost  effectiveness.   While
not  central to- the  issue  of  controlling refueling emissions
through onboard  vapor recovery, this  potential  for  control  of
excess  in-use  evaporative  emissions  provides  an  additional
perspective  on  the  value   of  implementing  an  onboard  vapor
recovery requirement.

VII. Leadtime Requirements

     If an onboard  vapor  recovery  requirement were implemented,
it  is  estimated  that  approximately 24 months of  leadtime would
be necessary before the systems could be required on production
LDVs  and  LDTs  once   a   rule   is   promulgated.   This  leadtime
estimate  is based  on engineering judgment,  and on leadtimes
necessary  in  similar, previous EPA  rulemakings.   These include
the  original  1978   6.0  g/test  LDV/LDT   evaporative  emission
standard which was implemented in just one year,  the 1985 HDGV
evaporative  emission   standard  which  will  be  implemented with
two  years  of   leadtime,   and  the  1981  2.0  g/test  LDV/LDT
evaporative emission  standard which was  also  implemented  in two
years.  The  two-year  leadtime estimate  to  implement an onboard
vapor  recovery program is based on  the following considerations.

     A  program  to comply  with  an  onboard   requirement  would
first    include    approximately    six    months    for    the
vendors/manufacturers  to  develop  and optimize working prototype
systems  applicable to  all  of  their  different  vehicle models.
Next,  initial verification  of the  fillpipe  seal  and pressure
relief  valve durability  could be  conducted  in  two  months  or
less   under   laboratory  conditions.    However,   purge   system
optimization  and optimization  and proveout  of  the   integrated
onboard  vapor  recovery/evaporative   emissions   control   system
would  require some  vehicle testing,  as would  verification  of
the  efficiency and durability  of  the fillpipe seal and pressure
relief  valve.   This  vehicle  testing would require  four   to six
months, based  on manufacturer estimates  for  similar  in-vehicle
testing  programs.   Thus,  prototype  testing  and   proveout  is
estimated  to take  12  to 14 months  to  complete.

     Although  many of the components of  an onboard system would
be   "off-the-shelf"   or    readily   fabricated   from  existing
production tooling, some  tooling  changes would be necessary for
some  components,  such as larger  charcoal  canisters.  However,
                               C-28

-------
                              -25-
 the critical  items  in  terms  of  production  tooling  are  the
 fillpipe seal and pressure  relief  valve.   If the  fillpipe  seal
 and^ pressure  relief  valve  used  are  some  form  of  currently
 available  component,  then   only   the  question  of   capacity
 exists.   Capacity  is  necessary to  meet  long   term  demand  in
 excess   of  13  million  units  per  year.   If.  the  vendors  and
 manufacturers   ultimately  settle  on  prototype   designs   which
 would  require  significant  tooling  changes  or  completely  new
 production,  or if current  production capacity is  insufficient,
 then longer  tooling  leadtimes may be  required.

     In  any  event,  commitments  leading  to  production tooling
 changes  could  probably  be  made after the  initial  laboratory
 verification  of  the fillpipe  seal  and  pressure  relief   valve
 durability.   If  vendors/manufacturers  are  able  to  use   seals
 similar  to  those  used in the 1978 vehicle demonstration program
 and  an  acceptable  pressure  relief  valve  is  available,   then
 total  tooling  leadtimes  of  three  or  four  months   would  be
 necessary.   If fillpipe  seals  and  pressure  relief valves  must
 be  procured from  modified  tooling,  then  leadtimes  of- six  to
 eight months  are  reasonable.   If new tooling must  be developed,
 then leadtimes  for tooling  will require approximately  12 months
 or  perhaps  longer.   Thus,  the  range for tooling  leadtimes  is
 three months  to one  year or more,  depending  on  the  source  of
 the fillpipe seals  and  pressure  relief  valve.   Assembly  line
 tooling  changes  would  be  handled  during  normal model   year
 changeover, and thus would have  no  effect  on this estimate.

     Finally,  some  time  would  be   required  to  allow  for  the
 normal  EPA  certification  process.    -It  normally  requires  a
 manufacturer  10  to  12  months  to  certify  its   entire product
 line.[15]

     Given  these  estimates   of the  leadtime   necessary   for
 development,  laboratory  testing,  in-vehicle  testing,  tooling,
 and certification, Figure 1  shows how these different  estimates
were put together  to  arrive  at a  leadtime estimate of   two
years.    The  critical   path  on  this  figure  is   6 months   for
development, 2 months for laboratory  testing,  4  to 6 months  for
 in-vehicle  testing,   and  10  to  12  months  for  certification.
Presuming  that  tooling  commitments  can  be  made  after   the
laboratory testing is concluded,  tooling  is  not  a critical  path
even if  the fillpipe  seal  and  pressure  relief  valve  required
new  tooling.    Tooling  would  only   become   a   concern   if
commitments  were   delayed   until   after  the   completion   of
 in-vehicle testing (12-14 months).

     In  summary,   a  leadtime   period   of  two   years  appears
 reasonable to implement an onboard  requirement.   Of course,  the
model year  of  implementation  for an onboard  requirement  would
depend  on when a final rule was promulgated.
                               C-29

-------
                                                              -26-
                                                                                                                        CN
                                                                       '4-4
                                                                           oo
                                                                           §  i
                                                                                                                        oo
                                                                           oo
                                                                       M "%
0)
                                                                             00
                                                                          •H  I
                                                                          .M-l VO
                                                                          ftO
                                                                                               oo
                                                                                                                          CO
                                                                                                                          OS
                                                                                                                        CM H
CO
                                                                                               0)
a
CO

                                                                                      ao
                                                                                 J-   H
                                                                                               60
                                                                                               pit • Q
                                                                                                                          "3
                                                             C-30

-------
                              -27-
 VIII.   Onboard Control Versus Time

     When considering  the implementation  of onboard  controls,
 it^is  of value to  determine  how much time would  be  required to
 gain control of  a  majority of  the  annual  LDV and LOT  gasoline
 consumption.   This, of course, depends on  the  vehicle  scrappage
 and replacement  rates,  the annual  vehicle  miles of travel  and
 the vehicle  fuel  economies.  Consequently, the portion  of total
 LDV and LOT  fuel usage  (and  accompanying refueling  emissions)
 which  would  be controlled  as  a  function  of time  beginning  with
 the model year of  implementation  is estimated below.   For  this
 analysis  it   is assumed that  implementation  begins  in  the  1988
 model  year.

     A.    Total  Fuel  Consumption

     To   determine   the  portion  of  total  LDV   and   LDT  fuel
 consumption  controlled as a function of time the  controlled  and
 total  LDV   and   LDT  fuel consumption   must   be estimated   by
 calendar  year.   A  total  gasoline  consumption  by  a   specific
 model  year's  vehicles  in a given  calendar  can be derived  using
 the  expression given below;
     GC =  (REG)(SR)(VMT)(VMTGR)
                (MPG)(ODOM)
     where:

     GC = gasoline consumption  (gallons)

    REG = new vehicle registrations  for  that model year
          (function of model year)

     SR = survival rate of new vehicles  in the calendar year  of
          interest (function of age)

    VMT = average annual mileage of  the  vehicles  (function of
          age)

   VMTGR = growth rate in average annual mileage of the vehicle
          (function of model year)

    MPG = new vehicle in use fuel economy  (function of model
          year)

   ODOM = Usage pattern factor to account for the different mix
          of urban/rural driving and average daily  mileage on
          average in-use fuel economy  (function of age)

     Data for the  input parameters  described  above  is provided
and referenced in Appendix B.  However,  a  few explanatory notes
                               C-31

-------
                              -28-
are  appropriate.   First,  the approach  used here  models total
LDV  and  LDT fuel  consumption using  twenty  model  years  of LDV
and  LDT  registrations  (e.g.,  1988   fuel  consumption  would be
modeled  using  registrations  from   1988  to  1969   inclusive).
While  it' is recognized  that  there are  a  small  number  of  LDVs
and  LDTs  older than  20  years still  in-use,  their contribution
to  total  fuel  consumption is  relatively  insignificant  due to
their  low  registrations  and  average  annual  VMT.   Second, Table
B-2  contains  average annual  VMT  data for  LDVs  and LDTs.   This
data is applicable for  pre-1982 model year  LDVs  and LDTs.  For
1982 and  later LDVs and LDTs, average annual  VMT was projected
to  increase at a  rate of 0.8 percent per  year  for LDVs  and 0.4
percent  per  year  for   LDTS.[16]   Last,  calculation  of  fuel
consumption  included a usage pattern factor  (ODOM)  to  account
for  the  fact  that the mix of urban/rural  driving and the daily
vehicle miles  of  travel  both change as an  LDV  ages,  and  this
affects  the   in-use  fuel  economy   in  any • given  year of   a
vehicle's life.  This applies to  LDVs only.[16]

     Given  this data, total  LDV and LDT fuel consumption  in any
given  calendar year  can  be calculated by simply  determining the
fuel consumption  of each model years LDVs and  LDTs in the  year
of  interest and summing the  consumption from  each model  year's
LDVs and  LDTs  to  derive  a total.  This method of calculation  is
shown  mathematically in  the expression given below:
     GC
20
5
y
                  (REG
                      v

                                     v
(VMT  GRV^X)
                          (MPG,r  v)(ODOM
                              V f X      *
                                       ,) (VMT  GR
                               (MPG    )
                                  T. ,X
     v a  LDVs,  t  -  LDTs,  x =  model  year,  y =  years,
     z =  vehicle  age

In  this  method of calculation y  =  1 would be the calendar  year
of  interest,  and all  data used would begin  with that year  and
then  going  back  20  years.   Total  fuel  consumption  would  be
determined  by  summing the  consumption   of the  most  recent  20
model years LDVs  and  LDTs in  the  calendar year  of interest.
      B.
Controlled  Fuel  Consumption
      Calculation of  the  controlled  fuel  consumption  requires
 only two additions  to the discussion  given above.   First,  since
 controlled consumption is not  assumed to begin until  1988,  the
 period  over  which  controlled  consumption  will  be  calculated
 varies from 1 model year  in  1988  to' 13 model years  in 2000  (or
 presumably  longer   were  more  data  available  with  which   to
 calculate controlled  consumption  after  2000) .   Second, as  was
                                C-32

-------
                              -29-
 discussed    previously,    tampering    with   the   fillpipe   or
 evaporative   emission   system   will   eliminate   the   control
 effectiveness  of  the onboard  vapor  recovery  system  of  those
 vehicles.  Thus,  the fuel consumption of  tampered LDVs  and- LDTs
 must  be  factored  out.  This can be accomplished using  the  NTAMP
 factor  described  previously.   NTAMP  is   a  function  of  mileage
 and  is  different  for LDVs  and LDTs.  For any given mileage  in
 the  life of an LDV  or  LOT, NTAMP  = 1-TAMP,  where  TAMP is  the
 percentage  tampering calculated  using the projected  composite
 tampering rate  equations  given in Section  III.B.3.  The  mileage
 used  in  the tampering rate  equation for  each model years  LDVs
 and  LDTs includes the growth  rate  decribed  above  for LDVs  and
 LDTs.

      Controlled  fuel consumption in  any   calendar  year  is  then
 the   sum  of   the  fuel   consumption  of  each  model  year's
 non-tampered  LDVs and LDTs  in the  calendar  year  of  interest.
 This  method  of  calculation  is  shown mathematically below.   The
 only  difference between this and  the previous expression is  the
 limits  on the  summation  and  the   inclusion  of  the  tampering
 factor.
Controlled =

    GC
( z)
                                     x> (°DOMv,
                                      (MPG   )
                                         u, x
     C.
           Discussion of Results
     The  portion  of  the  total  LDV and  LOT  fuel consumption
controlled  in any  calendar  year,  1988  or  later,  can  now be
calculated.  Figure 2 compares  LDV  and  LDT gasoline consumption
which  would   be  controlled  by   an   onboard  vapor  recovery
requirement to total LDV  and  LDT gasoline consumption, assuming
onboard  controls were  first  introduced  with  1988  model  year
LDVs and  LDTs.   Table  3  is  a tabular summary  of  the graphical
information  presented  in  Figure  2.   This  data  shows   that
control of 50 percent of  all  LDV and LDT fuel consumption would
be  achieved  5 to  6 years  after  introducing an  onboard vapor
recovery requirement and  control  of more  than 84 percent of all
LDV and  LDT gasoline consumption would be  achieved  by 2000  (13
years  after   control   is   implemented).    Ttfithout   tampering,
control  in  the year 2000 would  exceed  92 percent;  control of
approximately  8.5  percent   of   consumption   is  lost  due  to
tampering.

     In terms of the separate LDV and  LDT fleets,  control of 50
percent  of  LDV fuel consumption would be  achieved  in  about 5
years and by  2000  89  percent of  LDV  gasoline consumption would
                               C-33

-------
                               -30-
                            Table  3

         Gasoline Consumption of Non-Tampered  Vehicles
                 With Onboard Emission Control
           Compared to Total Vehicle  Fuel  Consumption

           LDV Gas Consumption  (billions of  gallons)
Year

1988
1989
1990
1991
1992
1993
1994
1995
1995
1997
1998
1999
2000
Year

1988
1989
1990
1991
1992
1993
1994
1995
1996
1997
1998
1999
2000
  Total LDV
Gas Consumption

      49.0
      48.0
      46.9
      45.9
      45.0
      44.0
      43.1
      42.3
      41.6
      41.0
      40.5
      40.2
      40.0
LDV Gas Consumption
Controlled Vehicles

         6.0
        11.3
        15.8
        19.8
               Percent Control
                      12.2
                      23.5
                      33,
                      43,
   ,7
   ,1
        23.2
           ,1
           ,4
           ,3
26,
28,
30,
31.9
33.1
34.2
34.9
35.5
51.6
59.3
65.9
71.6
76.7
80.7
84.4
86.8
88.8
           LPT Gas Consumption  (billions of gallons)
  Total LDT
Gas Consumption

      24.4
      24.0
      23.4
      23.0
      22.7
      22.4
      22.2
      22.1
      22.1
      22.0
      22.0
      22.0
      22.2
LDT Gas Consumption
Controlled Vehicles

         2.4
         ,4.5
         6.3
         8.0
         9.5
        10.9
        12.1
        13 ,.2
        14,2
        15.0
        15,.8
        16,.5
        17.0
               Percent Control

                       9.8
                      18.8
                      26.9
                      34.8
                      41.9
                      48
   7
54.5
59.7
64
68
                         3
                         2
                      71.8
                      75.0
                      76.6
                               C-34

-------
                               -31-

                         Table  3-Cont'd

         Gasoline Consumption of Non-Tampered Vehicles
                 With Onboard Emission Control
           Compared to Total Vehicle Fuel Consumption

       LDV and LPT Gas Consumption  (billions of gallons)
      Total LDV & LOT
Year  Gas Consumption

1988        73.4
1989        72.0
1990        70.4
1991        69.0
1992        67.7
1993        66.4
1994        65.3
1995  '      64.4
1996        63.6
1997        63.0
1998        62.5
1999        62.2
2000        62.2
       LDV &
LOT Gas Consumption
Controlled Vehicles

         8.4
        15.8
        22.1
        27.8
       . 32.7
        36.9
        40.4
        43.5
        46.1
        48.2
        49.9
        51.4
        52.5
Percent Control

       11.4
       21.9
       31.4
       40.3
       48.3
       55.6
       61.9
       67.5
       72.5
       76.6
       79.8
       82.6
       84.4
                               C-35

-------
   80 -
   70 .
   60 .
   50 .
to

o
•H
H
H
•H
(0
c
o
a
e>
   40
   30
   20
   10
             -32-


          Figure 2


Controlled vs. Total Gasoline

Consumption for LDVs and LDTs


         1988 - 2000
Total Consumption


Controlled Consumption


Controlled Consumption


Controlled Consumption
                      LDV  & LDT


                      LDV


                      LDT
                                                            .-+—'
                #
               /
      1988  1989  1990  1991  1992  1993  1994  1995 1996 1997 1998 1999 2000
                                   C-36

-------
                              -33-
be  controlled.  For  LDTs,  50  percent  control  would  require  6
years and 77 percent  control would be achieved by  2000.

     This method  of determining the  time  for achieving control
of  refueling  emissions  differs  slightly  from  that   used  by
Lindgren,[17]  which  estimated  the  fraction   of   the  vehicle
population  which  would  be equipped with  onboard gasoline vapor
controls  over  time.   The   fraction   of  dispensed   gasoline
controlled  is  more  appropriate than  the  fraction  of  vehicles
controlled,  since refueling  loss  emissions  are  a  function of
the amount  of  gasoline dispensed  and  not simply  a  function of
the number  of  vehicles  in the  fleet.

IX.  Conclusions

     The  data  from  the  API  demonstration  program   and   the
manufacturers'   previous  comments   both  indicate  that  onboard
control  of  refueling  emissions from  LDVs,  LDTsr  and  lighter
weight  HDGVs   should  be  technologically   feasible   using   a
fillpipe  seal  and  an  integrated  onboard/evaporative   emission
control  system.   Onboard  control  should  also  be  feasible   for
the heavier HDGVs,  but  the systems used on  heavier  HDGVs would
be somewhat more complex  and  costly.   The  implementation issues
for the  control of  heavier. HDGV  refueling  emissions  could be
worked out  in  a  manner  similar  to  the  approach  used  in   the
recent HDGV evaporative  emissions final  rule,  so control  of
virtually  all  of  the  gasoline-fueled  motor  vehicles  may  be
possible.   Implementation of an  onboard requirement should have
a negligible impact on the vehicle's exhaust emission levels.

     An  in-use control  efficiency  of  98  percent  is  expected,
with  negligible  deterioration for a   well-maintained   vehicle.
Using  the  tampering  rates  expected  in  the  late  1980's   and
beyond,  owner  tampering with the filler  neck  restrictor  and  the
charcoal canister could reduce the average  lifetime efficiency
to 91.8 percent for  the sales-weighted fleet of LDVs  and LDTs^
Using 1988  projected  fuel  economies   for LDVs  and LDTs,   the
fleet average  lifetime  reduction in  refueling  HC  emissions is
.0196 tons per  vehicle.

     An integrated  onboard/evaporative  emission control system
is expected to carry a fleet  average  cost  of  $15 to  $25   per
vehicle,    although    the  average .  should   be  nearer   $15.
Implementation   of  an  onboard  requirement  would  not   increase
lifetime  operating  or  maintenance  costs.    At   $15-$25   per
vehicle,   an  onboard  requirement   would  have  no  perceivable
impact on manufacturer or dealer sales.

     Using  the costs  and emission  reduction  benefits  mentioned
above,  the  sales-weighted   lifetime   cost   effectiveness   for
                              C-37

-------
                              -34-
onboard  control  is  $766  to  $1,271  per  ton  of  HC controlled.
The annual cost effectiveness  is  $1,194  to  $1,990 per ton of HC
controlled.

     The  larger  charcoal  canister  of  an  integrated  onboard/
evaporative emissions  control system  could  potentially control
excess  in-use  evaporative emissions.   If the  preliminary  best
estimate  of  these  benefits  is   added   to  those  achieved  by
onboard control, the lifetime  cost-effectiveness values fall to
$435  to  $725  per  vehicle  and  the  annual cost effectiveness
becomes $678 to $1130 per vehicle.

     An  onboard  requirement  could  be   implemented  two  years
after  promulgation  of  a  final   rule.   Control  of   refueling
emissions  from 50 percent  of the total annual  nationwide  LDV
and LDT gasoline  consumption  could  be achieved  in  five years.
Control of refueling vapors  from more than 84  percent  of total
annual  nationwide  LDV  and  LDT  gasoline  consumption  could  be
achieved by 2000.
                               C-38

-------
                              -35-
                           References

     1.    "On-Board  Control  of  Vehicle  Refueling  Emissions
Demonstration  of   Feasibility,"   API  Publication   No.   4306,
October 1978.

     2.    "Recommendation    On    Feasibility    For    On-Board
Refueling Loss Control," U.S. EPA, OMSAPC, February 1980.

     3.    See reference 1, p. 25.

     4.    Letter,  F.  L. Voelz,  ARCO to E.  P.  Crockett,  API,
January 14,  1982,  and follow-up  telephone  conversation between
M. Reineman, U.S. EPA and F. L. Voelz, ARCO, August 18, 1983.

     5.    "Staff  Report,  Issue   Analysis  -  Final  Heavy-Duty
Engine HC  and  CO Standards,"  U.S.  EPA,  OANR, QMS,  ECTD,  SDSB,
March 1983.

     6.    "Transportation   Energy    Data    Handbook,"   Sixth
Edition, ORNL-5883, Oak Ridge National Laboratory, 1982.

     7.    "Motor Vehicle  Tampering  Survey  - 1982,"  U.S.  EPA,
National   Enforcement   Investigations   Center,   Larry   Walz,
EPA-330/1-83-001, April 1983.

     8.    "Draft  Mobile  3  Documentation,"  data  provided  by
Lois Platte, U.S. EPA, QMS, February 14, 1984.

     9.    "A  Report  to  the  Legislature  on  Gasoline  Vapor
Recovery  Systems For  Vehicle Refueling  at  Service  Stations,"
California Air Resources Board, March 1983.

     10.   "Compilation  of  Air   Pollutant  Emission  Factors,
AP-42, Supplement 9," U.S. EPA, OAQPS, July 1979.

     11.   "Trends  in  Motor  Gasolines:   1942-1981",  E. Shelton,
et al, U.S. Department of Energy, DOE/BETC/Rl-82/4, June 1982.

     12.   "Decision:    Vapor   Recovery   Control   Strategy,"
General Motors Corporation briefing to EPA, February 3, 1984.

     13.   "Cost  Comparison For  Stage  II and  On-Board Control
of   Refueling   Emissions",   American   Petroleum   Institute,
January 1984.

     14.   See for  example the Regulatory  Analysis  and Summary
and Analysis of  Comments prepared in  support  of the light-duty
diesel  particulate  regulations  for   1982  and later  model  year
light-duty  diesel  vehicles.   Both  are  available  in  Public
Docket No. OMSAPC 78-3.
                               C-39

-------
                              -36-
                      References (cont'd)

     15.   Trap  Oxidizer  Feasibility  Study",  U.S.  EPA,  OANR,
OMSAPC, ECTD, SDSB, March 1982.

     16.   "The   Highway   Fuel   Consumption   Model  -   Ninth
Quarterly   Report,  prepared   by   Energy   and   Environmental
Analysis, Inc., for U.S. Department of Energy, February 1983.

     17.   "Manufacturing  Costs  and  Automotive  Retail  Price
Equivalent    Of   On-Board    Vapor    Recovery   System    For
Gasoline-Filling Vapors,"  Leroy H.  Lindgren,  Consultant,  Draft
Report, June, 1983.
                               C-40

-------
            APPENDIX A
Recommendation on Feasibility for
 Onboard Refueling  Loss  Control
               C-41

-------
                                                          Appendix A
                          February  1980

                  Recommendation on Feasibility
                for Onboard Refueling  Loss  Control
                              NOTICE

Technical Reports do not necessarily represent final EPA decisions
or positions.  They  are  intended to present technical analysis of
issues using  data  which  are currently  available.   The purpose in
the  release  of  such reports is  to  facilitate  the  exchange of
technical information and to inform the public of technical devel-
opments which may form the basis  for a  final EPA decision, position
or regulatory action.

             Standards  Development  and  Support Branch
               Emission Control Technology  Division
           Office of Mobile Source  Air  Pollution Control
                Office  of Air,  Noise and Radiation
               U.S. Environmental Protection Agency
                                     C-42

-------
I.   Introduction

     Gasoline refueling  losses  estimated  to be in the range of 4-5
g/gallon,  can be  controlled by  use of  control equipment  at  the
service  station  (Stage II control)  or by use  &f control equipment
in  the  vehicle  (onboard control).   As  required  by the 1977 amend-
ments to the Clean Air Act,  the  Emission Control Technology Divi-
sion (ECTD)  of EPA has reviewed and analyzed available data on the
feasibility  and  desirability  of  onboard  refueling  loss  control
which will be discussed in this report.

II.  Summary of Conclusions  and Recommendations

     Several hardware  demonstrations and  paper studies,  Ref.  1,  2,
have been conducted to determine the technical feasibility and cost
effectiveness on onboard  refueling  loss control.  Much of  the
current  information  is from the American  Petroleum Institute (API)
onboard  demonstration  program,  Ref. 3.   Other current information
was obtained from motor vehicle manufacturers in response to a June
27, 1978  Federal  Register  (43FR 27892)  request for relevant infor-
mation.   • These demonstrations and  analyses deal  with .the state-of-
the-art  emission control technology.

     Analysis of  this  information  supports the  following  conclu--
sions:

     1.    Onboard  refueling  loss  control  is  feasible for  light-
duty vehicles.

     2.    The most probable control system uses hydrocarbon adsorp-
tion on  charcoal  (the same  strategy that is, used  for evaporative
emission control).

     3.    Control  effectiveness  can be  as high  at 97%, but  this
depends  especially upon the vehicle fillpipe/service station nozzle
interface and upon control technology design.

     4.    An analysis  of  data from  three fillpipe/nozzle concepts
(fillpipe  seals,   nozzle   seals,  and  combination  fillpipe/nozzle
seals)  shows  that the  effectiveness of  all three  concepts  is
approximately equal.   Durability  effects  have  not been extensively
evaluated, especially for the nozzle seal  concept.

     5.    A vapor/liquid  pressure  relief  valve  is  required  to
protect  the integrity of the vehicle fuel  tank during the refueling
process.  The pressure relief valve  can be designed to function on
the fuel nozzle,  or it may  be incorporated as  part of the fillpipe
seal mechanism, which  would  be sealed-off  by  the  fuel cap  during
vehicle  operation.  Durability  effects have  not  been evaluated  for
either   the  fillpipe or  nozzle pressure  relief.    ECTD  recommends
that the fillpipe/nozzle seal and pressure relief be located on  the
vehicle  if onboard controls are required.
                                     C-43

-------
                                 -2-
     6.   The consumer  cost  for light-duty vehicle  refueling  loss
control systems are estimated to be  $17/vehicle.   The  $17  estimate
does not  include  costs  for a seal or  pressure  relief.   Cost  for  a
seal and  pressure  relief,  if used on  the vehicle,  is  estimated  to
be about  $2.70.   The cost  of a seal  on  the nozzle should be  the
same as  the cost  for  a Stage  II nozzle.   Except for the as  yet
undefined durability of the interface seal no maintenance costs  are
expected.

     7.   The feasibility  of controlling refueling  loss  emissions
from gasoline fueled trucks and diesel fueled vehicles  has  not  been
evaluated to date.  Technical feasibility and cost effectiveness  of
controlling these sources should be determined.

     8.   Minor increases  in CO exhaust emissions  can  probably  be
controlled  by  minor changes  to  either the refueling  loss  control
system or  to  the  exhaust emission control system.   The ability  to
certify a vehicle  to a  3.4 g/mi  CO standard  to  50,000  miles should
not be seriously impaired.

     9.   The use  of a  bladder  in the fuel  tank appears to be  a
viable alternative control  strategy,  but  some  problems exist  and
technical feasibility is yet to be demonstrated.
           *
    10.   Considering the  lead  time  needed  for  regulation  develop-
ment  and review  within EPA and  the lead  time  required by  the
industry  for  development and application of  technology,  implemen-
tation of onboard controls  cannot occur before 1984.

     ECTD  recommends  that  the  choice between  onboard  control  and
Stage  II control  of  refueling  loss  emissions  be based upon  air
quality  considerations  and the relative cost effectiveness of  the
two strategies for the same overall level of control.

     It is recommended that methods of reducing the cost of onboard
refueling  control  systems  be  examined  by  considering  tradeoffs
between  control  system capacity and  cost.   It may  be  possible  to
sacrifice  some capacity that  is  only  required   under '. infrequent
conditions  and  achieve proportionately more  significant cost
savings.

     The  feasibility  and desirability  of  control of refueling
losses  from light  and  heavy-.duty  gasoline  fueled  trucks  and  from
diesel  fueled  vehicles  should  be  considered.   EPA  should  support
the development of the  bladder  tank  alternative for. refueling loss
control  strategy.   If  regulations are to be  developed  for  onboard
refueling  loss control,  a  certification  test  prpcedure  must  be
developed.
                                     C-44

-------
                                 -3-
 III.  Review  of Available  Information

      The  data and information summarized in this section are based
 on  material   submitted  to EPA by  the  American Petroleum Institute
 and  information  received in  response  to a request  for information
 (43FR 27892)  published  on June 27,  1978.  The API material, Ref. 3,
 is  the result of their most recent study to assess onboard techni-
 cal  feasibility and compare the  cost  effectiveness of  onboard
 refueling controls and  Stage  II controls.  This study was initiated
 at  the urging of EPA.  Respondents  to the Federal  Register notice
 included^ General Motors,  Ford,  and  AMC.    The  API,  GM,  and  Ford
 information  contain  data  from tests with onboard control hardware.
 All  respondents,  with the  exception  of  AMC, submitted information
 on  the cost and the desirability of  onboard control systems.
      1.
API Onboard Study
     The  API Onboard Control Study was  structured  to address
questions  regarding onboard  feasibility  which  were  posed  to  API in
a December 1977 meeting with EPA.  The API study consisted of three
tasks:   a vehicle  concept  demonstration, a fillpipe/nozzle concept
demonstration,  and a  cost/benefit  analysis.   Exxon Research  and
Engineering  Company  and  Mobil Research and  Development  Corpora-
tion were the API contractors  for  the  vehicle  concept  demonstra-
tion.   Atlantic  Richfield Company was  the API  contractor for  the
fillpipe/ nozzle concept demonstration.  Exxon R &  E completed  the
cost/benefit analysis for API.

     The vehicle concept modification task had  the following  design
objectives:

     1)   Minimum 90% overall refueling vapor recovery.

     2)   No significant effect on exhaust  emissions.

     3)   No significant effect on evaporative  emissions.

     4)   Design should be durable,  practical,  and safe.

     The  fillpipe/nozzle  demonstration  had the following  objec-
tives:

     1)   90% overall vapor control.

     2)   Compatible  with  existing vehicle  population.

     3)   Compatible  with  existing Stage  II nozzles.

     4)   Design should be durable,  practical,  and safe.

     Test procedure guidelines for the API  work were  discussed at  a
meeting with API  on  March 15,  1978.   Important  procedural  guide-
                                     C-45

-------
                                 -4-
lines which  resulted from" that meeting  are  summarized  as follows:

     Fuel specification:   Indolene  unleaded  test  fuel  was used for
all exhaust, evaporative, and refueling loss measurements.

     Dispensed fuel quantity:  Test  vehicles  were refueled to 100%
of capacity from a condition of 10% tank capacity.

     Fuel tank temperature/Dispensed fuel temperature;   The dispen-
sed  fuel  temperature was  selected  to be representative  of  summer
refueling conditions in  Los  Angeles  during  the month of August,  or
about 85*F.  The fuel  temperature  in the tank was also selected  to
be 85*F.  Thus., the refueling was isothermal.*

     Purge Cycle;   For  the  purposes  of  the  API study,   the  only
driving cycle  which  was used  for  purging the  refueling  loss  can-
ister is the LA-4 cycle.

     Individually,  these test procedure  guidelines  are considered
to represent  real world  situations in  a  high  oxidant forming
location, e.g.,  Los  Angeles during  the  month of  August.   Collec-
tively, these  guidelines imply that  the  API  vehicles  demonstrated
the  feasibility of onboard  control  systems  in an approximate worst
case  condition.    This  reasoning  is  consistent  with   earlier  EPA
recommenda't ions  that  API  err  on the  conservative  side  during
their study.   For  example,  Exxon used the  following test sequence
to quantify the exhaust emissions interaction between the  refueling
control system and the exhaust emission control system:

     1)   Load ECS (Evaporative  Control  System)  canister  to  break-
through .
     2)

     3)

     4)
through.
Condition the vehicle by driving 2 LA-4's.

Soak vehicle overnight.

Load  RCS  (Refueling Control  System)  canister to  break-
     5)   Condition  the  vehicle by  driving  5 to 6  simulated  city
driving days  (4.7  LA-4's  with one hour hot  soaks  in between  and  a
diurnal at  the  end of the day)  to  consume 90% of the  fuel in the
tank.
*This represents  a conservative situation as survey data,  Ref.  4,
 show  that  nationwide  dispensed  fuel  temperatures are  typically
 lower than  tank  fuel  temperatures,  thereby representing  a  vapor
 shrinkage situation during the refueling process.
                                     C-46

-------
                                 -5-


     6)   Drain the fuel tank.

     7)   Block RCS canister line.

     8)   Fill tank to 40%, unblock RSC cani-ste* lines.

     9)   Conduct diurnal evaporative test in SHED.

     10)  Drain tank to 10%.
                                            t
     11)  Bring fuel  tank  liquid and vapor to  equilibrium at 85°F
(shake the vehicle to accelerate the equilibrium process).

     12)   Refuel  the vehicle  to  100%  in  SHED with  85°F  fuel.

     13)  FTP

     14)  Hot soak evaporative test  in SHED.

     Obviously, these  test  procedures  do not lend themselves to a
routine laboratory certification test procedure.  They do, however/
permit  an approximation of how  an onboard  control  system would
function in a severe "real-world" situation.

     A review of the three  API contractor's activities is presented
below.

Exxon

     Exxon  assumed  the responsibilty for modifying four test
vehicles.   Their vehicles  included the following:

    • 1978  Chevrolet Caprice

     1978  Ford Pinto

     1978  Plymouth Volare*

     1978  Chevrolet Chevette

     All  vehicles  were  designed to  comply with  1978  California
exhaust  and  evaporative  emission standards  (.41  HC,  9.0  CO,
1.5 NOx, 6.0 Evap).
* Vehicle  subsequently  dropped from test  program  because  of high
baseline NOx levels.
                                     C-47

-------
     The  Caprice  is  a  conventional  oxidation  catalyst  vehicle,
while  the Pinto  is a three-way catalyst vehicle  with feedback
carburetor  control.   Vehicle  descriptions and  complete  refueling
loss control  system  descriptions  are presented  in  Table A-l  and
Figure A-l  of the Appendix.   The  refueling loss canisters  in  the
Caprice, Pinto and Chevette are described as follows:
              RCS
Vehicle Carbon Volume
Caprice
Pinto
Chevette
5.0P_
3.0JP
3. OP
                          Carbon Mass

                             1800 g

                             1100 g

                             1100 g
Carbon Type*

  BLP-F3

  BLP-F3

  BLP-F3
Location

Underhood

Underhood

  Trunk
* Same carbon currently used for controlling evaporative emissions.
     The Exxon  exhaust  and evporative emission  test  results  which
compare  baseline  and modified  versions  of the  Caprice,  Pinto  and
Chevette are summarized  in  Tables  1,  2 and 3.   Engine-out data are
summarized  in  Tables 4  and  5.   Refueling  loss  effectiveness  test
results  are summarized  in Table  6.   All  Exxon  refueling emission"
tests assum&d a no-leak  seal  at the  fillpipe/nozzle  interface.   In
laboratory  practice  this was  achieved with leak  free  connections
from the fuel nozzle to the fillpipe.

     Benzene  emissions  from  the  Caprice  and  Pinto  were  measured
during several of the refueling loss SHED tests.   These  results are
summarized  in  Table 7.   The  Exxon data indicate  that benzene
control  is  directly  proportional  to  refueling  loss control effect-
iveness, although  current benzene  levels  in  the  SHED  are at  the
detectable  limit of  the instrumentation.

     Table  8 presents  Exxon's manufacturer cost  estimates  for
onboard  control systems  for  the  1978 Caprice  and .Pinto.   These
estimates do not include the costs for fillpipe sealing  devices and
pressure  reliefs.    This  hardware represents  an additional  cost
of  approximately $1.50  (manufacturer's  cost)  per vehicle.   Exxon's
cost  estimates  assume an estimated $.50 credit  for  downsizing the
ECS canister,  which in  the two  canister system,  controls  only
carburetor  losses.   Exxon  estimates  the  incremental cost  of  two-
canister refueling  control systems to range from  $8.25  to $10.53.
This estimate includes the above mentioned $.50 credit but does not
include  the $1.50  cost  for the fillpipe seal  and  pressure relief.
The corresponding cost  range for single canister refueling control
systems  is  $6.75 to  $9.00.   For light-duty trucks, Exxon estimates
a cost range of $12  (large single  canister) to $20 (two
separate  refueling  loss  canisters  or multistage purge  systems).
                                     C-48

-------
-7-








 a
Ed
*-*
CO
C
3
Q






X
O
z

^
•4
E
so o
v^ 0

cn
3
eg
.C
X
(d U




















. ONOO -,-* r*
CM o en o

i— i « • » > • JZ
C < CO C < W O
c
cy
u
u
ceo;
o o CM
•*4 ,^4
iJ U
to eg
)-c u
3 3
eo oo
U-l U-l
e c
o o
o u
• 11 >• i
C < CO C < CO CJ
4^
e

-------
-8-


























en

o

o
eo
H
















































-
XJ
0)
01
O
1
01
e
o
en

'

•o
c -
CO

CO
3
CO
j:
•*• •
H

Cu

fa .















.


•

eg
O
H
60 CO
^ O
CO

O* 4^
CO O
>MM
iA
Id
 in
\O r*** eN "™"'
• • • n
«S 0 — 0



vO «l
co en en o
•* C5 O O O
II
e





0N -^ C^ *^
00 — 1 +

-^ o -< .




eN «) en
r-» en vo e>i i
• • • *
en o en o





P** CS \O W\ ^f
cs o e>i o 1
do' do



V
60
. . . . c
en « Q en eu C> co
n > • H > • jz
c < w e < ui u
. c
CU
u
e e n
o o - cu
•»J -«H
^J 1 if
CO co
h U
3 3
BO 60
••* — ^
1M U-4
C C
0 0
C3 CJ
CU tJ
C eu
13 -tl
CO *^
CO O
« s



















































en
1
Ij
*j
u
co
xj
03

i i
£

en
+
CU
H
fa
*
C-50

-------
                                           -9-
1)






















«
•- H
u
Q.
CO
1
CO
o
"
CO
CO
"I
3
o
a
c
60
c
Cd

























X

00

"M*
^
CM

p"*

b
o
z




o
o






u
=







0



I
oo
*
1,



o


0
























eo r* ff — i
«3- r» O <• CM
• • • • ^
vO O CM CM


CO in ON O> -*
CM O CM O +
— C3 — C3


CU
00
• • * * C
m cu Q m « O co
n > • u > • r;
c < co c < en u

c
cu
u
e c 
m \o +
-
CO vO 1^. 0\
CO O r* O CO
- 0 '
.

• • •
sr §1 a. ^r 5 Q.
II < CO II < 03
c , . c «
00
c
10
6
C C j-t
0 0 C
-*4 . -M* (j;

-------
                                                -10-
                          01
                          a*
                          9)

                          a
                         a
                         to
                                            o\
                                            o\
                                                                   co
                                           CO
                                           o\
\o

 0)
 CB
H
       w
       jj
       e
       
        •  •   • I  •

       — o\ m 
-------
                                            -11-
 to
H
       at

       o
       • ~4
       0}
       
-------
                                -12-
                               Table 8

                  COST ESTIMATES FOR ONBOARD SYSTEMS
                           (1)
Charcoal
        (2)
                   (3)
Canister and Valves

                  (4)
Tank Modifications

Hoses and Tubing

Assembling and Installing
      Q $20.00/hr.
(6)
Credit for Downsized
Evaporative Control System
Caprice
$4.96
2.50
0.30
1.57
1.50
$11.03
$0.50
$10.53
Pinto
$3.03
2.00
0.50
1.72
1.50
$8.75
$0.50
$8.25
 (1)  Estimates  are made  for manufacturer's  large  volume  production.

 (2)   1800  g  for the  Caprice canister,  1100  g  for  the  Pinto  canister
      at  $1.25/lbm (Calgon BPL-F3 carbon).

 (3)   Plastic container and valves.

 (4)   Larger  size float/roll-over valve.

 (5)   3/4" vapor  line  from fuel tank  to canister,  3/8" .purge  line.
      EPDM tubing for vacuum  control lines.

 (6)   Additional 4.5  minutes  labor at $20/hour.

 (7)   Reduced size evaporative control canister.
                                      C-54

-------
                                 -13-
     Exxon estimates  the average cost for onboard  control  systems
to  be  $9/vehicle.   This  is  based on  the  following  assumptions:

     1)   Onboard  systems  are  designed  to control  refueling  emis-
sions 'from light duty vehicles with an average fuel  tank size  of  17
gallons  refueled  to  100%  capacity from  a  condition  of 10%  tank
capacity.  The  onboard  systems  are  designed  to control hydrocarbon
emissions at a level of 6 g/gal.

     2)   70%  of   light-duty  vehicles and  single  tank  light-duty
trucks are  assumed  to use single canister (evap + refueling)
systems.

     3)   30%  of   light-duty  vehicles and  single  tank  light-duty
trucks are assumed to use two canister systems.

     4)   Light duty  trucks  with dual or large  fuel tanks  consti-
tute approximately 10%  of  the  light-duty  vehicle  and  light-duty
truck  population.

     In  summary,  Exxon finds  that  onboard  refueling 'controls for
light-duty vehicles are a technically  feasible, practical, and  cost
effective alternative to Stage  II vapor recovery.   They  are of the
opinion that the  same may  also be  said  for light-duty trucks.

Mobil

     Mobil R&D has modified  a  1978 Pontiac Sunbird for  control  of
refueling losses.   This vehicle  has  a  three-way catalyst  with a
feedback carburetor control system,  and  is  certified for  complaince
with  California   exhaust  and evaporative  emission  standards.
This modified  vehicle uses a  single  canister which contains  1550
grams  of Calgon BLP-F3  carbon.  The complete  vehicle and refueling
loss control  system  descriptions  are presented  in the  Appendix.
Table  9  presents  comparisons of exhaust and  evaporative emissions
from the  Sunbird  for  the  baseline  and  modified  configurations; a
summary of the  refueling emission data  is  presented  in Table 10.

     Similar   to Exxon's  findings,  Mobil  states that  their  test
results have demonstrated that  onboard controls  are a  feasible and
desirable method  of  controlling  refueling  losses  from  light-duty
vehicles and  light-duty trucks.

Atlantic Richfield Company

     One  of  the  requirements   for  the  operation of  an  effective
refueling loss  control system  is  a no-leak  seal at  the fillpipe
nozzle interface.   Atlantic  Richfield (ARCO) has designed and
tested  three  types of sealing systems.  They  included:

     1)   Modification  of  the  vehicle fillpipe  to  achieve  a  seal
when used with conventional lead-free  nozzles.
                                     C-55

-------
-14-



















,



T3
"Ja
e
a
CO
i
1 i
•
0)
O
CO
•••4

(0
O.
S
6
O
ON I

CO
O W

~« ti
H
0)

U
o
Q.

>
Ed

•O
C

JJ
0]
eo
i
cu













.—4
Q
1 1
0
H


-N .*
00 eo
~> O
en
Q. 4J
3 5

\ """






1—4
eo
h

• ••4
O







*

^
"^

^—^
^
i

"M c
•n
CO

Ed CJ















.





^t
O si- — < vo
O en — i u"> vo
• • • * ^
(NO CS O





eN en i*. r»-
—, _ CN en
-i d -< d



•





r«. tN en
co O r»- CN
do o o




eN
U -*
c n



CO ^^ ^^ ^^
CT\ O 0*> O »^
O O O O




_ — , \r\  . t
».p ^3 *-Q ^3

ON en o en
en o -TO en
• • * • ^*
o o. o o




ON CU O NO CU O CO
U > • H > • -00
. c < co c < co c
- eo
€
.u
c
e c cu
O O CJ
•H -^ 14
u u CD
co eo • O-i
l-i )i
3 3
00 60
(U Uj
C C
O O
CJ CJ
eu -a
c o
• •-4 -*4
~* U-l
CU **^
CO T3
eo O






CU
(•^
jQ
eo
0
>
eo

3


CU
^
eo

CD

^^
3
co
U

•^
CO
O
CO

4J
o -u
J- U
i c
CO 3
CA
f«4
i
3 to
"•0 0 C
— co
— e
eo CU CU
3 "-1 *•"
T3 ja 3
— i co en
> H RJ
! ^
•f-l CO
ca
cu o
U hJ
M
J 60
3 C
•^
QJ --
60 «
eo 3
<
!-> O
u e
i cu
O -H
O 0
U-l
[d




'^
CO
00
en ~»^
C 60
O ^
**
CO CO
CA JJ
-< C
= (U
CU
00 W
C 3
-i CO
i— * CO
CU CU
3 •£
U-l
U Q
OS W
33
CO





X
Ol

±t
cu
J_
tl.
c.
CO
CJ
C

a
4v
u
cu
t-^
c
CJ

C_J


a
6C
v^'

(
ca
C
cu
a
a
C.

i™"*
<
Z
b

P^ ON -T in
ON O> ON ON




















CO CN •£!" NO
— i O -N C
u
• ** M
O 'I— I
a.  — (
C Bi CO
3
CO •
e ja a;
O *"* ^J
U eo
U-l ON U
•a •— 60
CU ea C
oo c —
U '-4 r->
3 S «J
a. o 3
C U-4
Id CU
CU uj i_i

• • . • co S
\Q \f\ \Q f"N^ «^^ ^^ Q« •
_ — i — , — i c cu oo
sa 3'
CJ to CO
*


















































*























C-56 .

-------
                                  -15-
      2)   Modifications to  both  the  fillpipe and lead-free nozzle.

      3)   Modification  of  a Stage II  vapor  recovery  nozzle.

      A  description  of each type of seal ,and., a summary of the
 durability  data  collected  with  each system are presented  below:

      Fillpipe seals;    Two  types  of   fillpipe  seals  have been ex-
 amined.   They are a  rotary  grease  seal  (similar to grease  seals
 used on rotating machinery shafts), and  a doughnut shaped  seal.
 The material types for these two  seals are a compounded  nitrile and
 thermosetting urethane, respectively.   More complete descriptions
 of  these  seals,   including   durability  data, are  found in  Figure
 A-5 and Tables A-2 and A-3  of the Appendix.  Appproximately  thirty
 days of durability tests with both  types  of seals have demonstrated
 that the  rotary  seal is more effective,  basically  due  to the
 absence of  expansion  problems when exposed to gasoline  liquid and
 vapor atmospheres.  The seal effectiveness of the prototype  fill-
 pipe and nozzle hardware  are determined  by a bench  test apparatus
 which pressurizes a  particular  system  and measures  the resulting
 leak ^rates.    Seal  effectiveness  calculations   are  determined  by
 dividing the leak  rate by  a  nominal fueling rate  (assumed to be 7.5
.galIons/min.).   Durability   tests conducted  with the  rotary seal
 have demonstrated  that the rotary seal is  effective  after 700-1000
 nozzle   insertions, which  correspond   to  the number  of  fuel  fills
 expected during the life of  the vehicle.

      Combination fillpipe/nozzle seals;   These systems  consist  of
 connecting parts on both the fillpipe  and nozzle.  Figure A-6  is an
 example  of  a prototype design evaluated by ARCO.  Durability test
 results  with these systems are similar to results obtained with the
 rotary  seal..

      Nozzle  Modification;    Working   prototypes  of vapor  recovery
 nozzles, modified  for  refueling loss  control,  have  been developed
 by  OPW and Emco Wheaton and  evaluated by ARCO for effectiveness and
 durability.   These nozzles   are designed  to seal  on standardized
 fillpipes.   The modified vapor recovery nozzles  incorporate  a
 pressure relief valve, which  is located at  the vapor return exit  or
 cast  into  the nozzle  body  and designed  to  open  at  approximately
 14-17 in. water pressure*, thereby  permitting the nozzle to refuel
 onboard  control  vehicles  and in-use  vehicles.    Nozzle  durability
 data  are very limited  but  one nozzle  has  been inserted  and latched
 7500 ^times,  representatitve of one year of service at  a high volume
 station, and  showed  a seal effectiveness of greater  than  99%.

     ARCO concludes that  the  preferred seal  techniques  are  either
 the  fillpipe seal  method  or  the combination  fillpipe/nozzle  seal.
* Refueling  loss  control systems designed  by Exxon and Mobil  are
  designed to operate at fill pressures of less than 4 in.  of water
  pressure.
                                      C-57

-------
                                 -16-
     2.   Vehicle Manufacturer' Information

General Motors

     General Motors has  several  reservations concerning the appli-
cability of  onboard  controls,  citing such  things  as:   the uncer-
tainty  of  the  effectiveness  of  f illpipe/nozzle  seals,  potential
cost  increases  associated  with  exhaust  emission  control systems
which must be designed  to  control  increased CO  emissions, negative
fuel  penalties  which  are   the  result of  this  increased emission
control, and the long  lead  time which  is required  to  obtain a
substantial  reduction  in atmospheric hydrocarbon  and benzene
loading.  GM has  stated  that  refueling  losses can  be  controlled on
the vehicle  (feasibility for  trucks has not been  demonstrated) ^ or
at  the  service  station.   GM's disagreements with  controlling
refueling emissions  with  onboard  controls  are  primarily based on
the issue of cost/effectiveness.

     More recently, (Ref.  13  & 14),  GM  has  expressed  concern  about
possible inadequacies  in the  design of  the  API demonstration
vehicles.   Specifically, GM states that  a large diameter  refueling
vent and^ a  small  restricted vent are necessary  for protection from
overfTTTing  the  fuel  tank and   preventing excessive  hydrocarbon
loading  of   the  storage canister  during  high  temperature vehicle
operation.

     GM's March,  1978 submission to EPA presents a summary of  their
work  on the- control  of diurnal evaporative emissions  and  refueling
losses  using  fuel tank bladders.

      It  is EPA's  opinion that the  theoretical control  effectiveness
of  evaporative  and   refueling  loss  emissions  using   bladder  tank
technology  is  high   and  that problems  related to vapor storage,
pressure relief valves, and  bladder  materials  can be  solved.   It
is  recommended that bladder tank  feasibility be  researched  by
funding a. bladder tank hardware  demonstration contract.

      The March,  1978 submission presents calculations showing that
 the additional weight  of the  components of  an  onboard control
 system will  cancel  out any potential  energy saving  which  results
 from the combustion  of the  refueling  vapors.     ECTD agrees with
 this analysis.

      The June,  1978  submission  is  basically a cost  effectiveness
 analysis comparing onboard controls  with  Stage  II controls (balance
 displacement and  vacuum assist systems).  GM estimates that onboard
 control systems,  effective with  the 1982 model  year, will  range
 from §16 to $24.  These figures are  about  $5 to $9 higher than the
 March,  1978 estimates due  to higher  estimates  for larger canisters
 and a new vapor/ liquid  separator.  GM assumes,  that the seal at the
 fillpipe/nozzle  interface will  be  obtained using modified  vapor
                                      C-58

-------
                                  -17-
 covery.   Rather,  GM  emphasizes  certain technical  concerns  which
 they say  are  not  fully addressed by the API  study.   According to
 GM, these  include API's  unsubstantiated  support  for  the  onboard
 fillpipe  seal  and  pressure  relief  (lack  of  adequate  durability
 results),  an unknown CO penalty for light-duty vehicles (no sensi-
 tivity data  relating  CO  to  test procedure differences),  and  un-
 proven feasibility for trucks.

      GM is  of  the opinion that  accelerated  laboratory durability
 tests   are  not sufficient  to  prove  that  proposed  elastomer  type
 seals   will  be effective  in  the  extreme  usage and  environmental
 conditions of the real world,  particularly when considering  a  ten
 year average lifetime  for  a light-duty vehicle.

 Ford

    _ Ford  has  submitted  test  results  from four  1978 model year
 vehicles  (three  non-feedback systems  and one feedback control
 system) modified   for  refueling loss  control.   These  vehicles  are
 described  in  detail  in Table A-4 in  the  Appendix  and in their
 submission to EPA,  Ref.  7.    The purge control systems  for  these
 vehicles are shown in  Figures A-7 and A-8 in the Appendix.

     Ford  estimates the cost to the consumer of onboard controls  to
 range  from $15-$20.   They note that the $15-$20 estimate does not
 include additional expense for such items  as:    packaging costs,
 incremental  labor costs, or  the costs for  additional exhaust
 emission control,  such as  feedback  control  over a wider air/fuel
 ratio range.

     Recent Ford  material,  Ref.  12,  suggest  that -the  cost  of
 onboard syst.ems may range from $30 to $253.  The  $30 estimate
 includes costs  over  the original $15-20 estimate, including costs
 for  such items as vehicle modifications  to  package  onboard systems,
 incremental  assembly, and material substitution.  The  $253 estimate
 includes the cost  for  a feedback  fuel  system  and electronic con-
 trols for  vehicles which are not  planned to be  equipped  with these
 control devices.

     On the  basis  of their in-house test results,  Ford has  conclu-
 ded  that onboard controls  are not technically  feasible for light-
 duty vehicles.

 American Motors

     AMC has submitted a letter to EPA, Ref. 9, which  states their
 concerns with  the  possible use  of onboard  controls.   They state
 that packaging concerns,  reduced quantities  of  purge  air  from
downsized  engines,  and compliance  with  stringent  evaporative
 emission standards  are  unresolved  technical  issues which have not
been addressed by  the API work to  date.
                                        C-59

-------
                                 -18-


     AMC does not find that API has demonstrated light-duty  vehicle
technical feasibility.

IV.  Analysis of Available Information

     1.   API Work  .

Exxon

     Exxon  R&E  appears  to have  done  a credible job  in  character-
ising  the  components  of a hydrocarbon  adsorption  system.   An
examination  of  the  results from baseline tests  and  tests with  the
modified Pinto  (3-way + feedback carburetor  system)  show small  but
finite  increases in  engine-out  (14%)  and  tailpipe  (8%) CO  emis-
sions.   HC, CO, and  NOx  emissions  are still well  below statuatory
emission levels for low mileage vehicles.   Engine-out CO emissions
from  the Caprice are approximately 20% higher  than  baseline test
results;  tailpipe CO emissions  are approximately 10%  higher than
the  baseline results.    No  increase   in  tailpipe  CO was  observed
during  tests with the Chevette.  Exxon suggests  that  differences in
CO emissions for the Caprice  and  the  Pinto  can be further  reduced
by minor modifications  to the refueling loss control system or the
exhaust  emission control system, although this has not been demon-
strated.
             *
     Figure A-2 shows  canister  purge  as  a  function of  time.
Although  the data  are  bench  test results,  the  results are also
representative  of actual control system purge  data.  It  is  signifi-
cant to note that the refueling  loss canister  is essentially  purged
to its  working  capacity  after three LA-4  driving days.  _ This
 implies that the refueling control/exhaust  emission interaction is
 likely to  be less in a  typical  driving day  than Exxon has measured
 using   conservative   test  methods,  which  required running  a cold
 start  FTP  immediately after: a 90% refueling.
                           | j
      ECTD   expects  that  refueling loss control  systems  will  result
 in slightly higher CO  feedgas  levels.  Exxon estimates  that  the
 average increases in CO  feedgas between refuelings will be approx-
 imately 5%  for non-feedback control   systems  and  less  than  3%  for
 feedback control systems.  ECTD has no other data  concerning  either
 the magnitude  of the average CO  feedgas  penalty  or the resulting
 effect on catalyst  durability.  It is ECTD's opinion that the Exxon
 estimates  are  reasonable  and that these  -additional  CO penalties
 will make  it  more  difficult  for vehicle manufacturer's to certify
 some engine/families to the  3.4  g/mi CO standard.   The higher CO
 levels  somewhat reduce  the  margin available to  allow  for exhaust
 system deterioration over 50,000 miles.

      ECTD   finds that   light-duty  vehicles  equipped  with  onboard
 systems are  capable  of meeting a   2 gram evaporative  emission
 standard.
                                       C-60

-------
                                 -19-
     An analysis  of  the  control  effectiveness"of benzene emissions
during refueling,  Table  6,  indicates  that icharcoal. canisters can
control  in  excess of  99% of  the uncontrolled  benzene emissions.
Exxon  conducted  additional tests with the Caprice  and Pinto using
indolene test fuel with a high benzene content <4.2%X.  The results
from these five tests suggest that benzene emissions are controlled
in excess of 992 during refueling.

     Packaging  refueling  loss  control systems,  is :a difficult
problem,  but  definitely not  an  insurmontable one.   The refueling
loss canister  is located  behind  the  rear seat  and above  the rear
axle in  the  Caprice, and  in the engine  compartment  of .the Pinto.
It  is  Exxon's opinion,  and  ECTD agrees,- ;that  it  is. possible for
manufacturers  to  locate   a  refuelling  loss  canister  on downsized
vehicles without  major  engine compartment tor, sheetmetal modifica-
tions .                 .'.-,-  -.  , ..;;„ i":;,   ..'•-,:'• •:.

     The  feasibility of   refueling  lossr controls   for light-duty
trucks  has not been evaluated by  Exxon, but they are of  the opinion
that refueling  loss  control  is   f easib'le* ;f or, light-duty trucks by
using  larger  control systems  and more  sophisticated  -purging con-
trols  (refueling  loss  control  canisters for each  tank and/or two
stage  purging systems).   It-. is ECTD's opinion  that the control of
refueling losses  from  light-duty trucks  needs  to be demonstrated,
especially the ability to  comply with a  2  g evap  standard, before
onboard controls  are judged  to be effective  for these vehicles at
the costs Exxon has estimated.

     Table 8  shows Exxon's  detailed manufacturer's cost estimates
for refueling control systems which have  two canisters.  ECTD  finds
these  cost estimates to  be reasonable  for onboard  systems designed
to  control. 100%  of  refueling emissions  from  90%  fill  conditions.
Exxon  estimates  the  average  manufacturer's cost for the light-duty
truck  and light-duty vehicle population to  be about $9.   That
number is derived  as follows:
One-cansiter vehicles*
Two-canister vehicles*

6,000 to  8,500  Ibs.  trucks**
            Weighted average
  Assumed
Average Cost

  $7.88
  $9.38

 $16.00
  $9.00
   % of
Population

    70
    20

    10
*   Includes  light-duty vehicles and  light-duty trucks under  6000
   GVW - average  fuel  tank  size  =  17 gal.
** Average  fuel tank size = 35 gal.

The charcoal  cost per  gallon of tank volume  is  assumed  to  be  about
$0.20.
                                      C-61

-------
                                 -20-
     The  $9  incremental manufacturer  cost  may be translated  to  a
consumer  cost  estimate  of  $16.20 by multiplying the manufacturer's
cost estimate  by  a factor of  1.8  (Ref.  10,  EPA Report  "Cost Esti-
mations  for  Emission Control  Related Components/Systems  and  Cost
Methodology Description" by Rath  and  Strong-,  March  1978).   The 1.8
factor  is in  general agreement with  previous  EPA  studies,  such  as
the  EPA Report, Ref.  11,  "Investigation and  Assessment  of Light-
Duty Vehicle  Evaporative Emission Sources and Control," June 1976,
which  used a  manufacturer to  consumer  cost  factor  of 2.0.   The
$16.20  estimate is in good agreement  with  consumer  cost estimates
submitted  by  GM ($16-$24) and Ford ($15-$20).  It  is  possible  to
further  reduce the cost of  an onboard system by  trading  off  some
degree  of refueling loss control effectiveness.

     Exxon has designed  refueling loss  control  systems  based  on
conservative  criteria,  and thus a different set of design criteria
will  afford   reductions  in  the cost  of  onboard  control  systems.
Texaco  has submitted  data  (Figure A-ll)  Ref.   12,  which relates the
number  of light-duty  vehicle  refuelings  and the percent  of  tank
fill.   A  reasonable design criterion  is to size  the refueling
canisters to  control 90%  of  nationwide  refueling  emissions.
Calculations  (Figure A-12) show that  90%  control can be achieved by
designing systems  to control  100%  of "refueling emissions from fills
to  63% of fuel tank volume.   If onboard control  systems  are de-
signed  to control  emissions  from refueling to 63% of tank capacity
rather  than   90%  of tank  capacity,  the  Exxon estimate of  $9 per
vehicle can  be reduced by $1.60  as  the  result of reduced charcoal
quantity.  This cost  reduction is proportional to the reduction in
carbon  bed volume.  The net effect of this design change is a cost
reduction to the consumer of approximately $2.88.  Changes  in
design  specifications  such as  the 90% fill  requirement may afford
additional'cost reductions for  other control system components  as
well  as  a general reduction  in  the   problem  of  packaging onboard
control systems.
        *
     ECTD estimates  the consumer cost  of  light-duty vehicle onboard
control systems designed  for maximum control  effectiveness  to  be
about  $17.  This estimate  does  not include an  estimate  for the cost
of  the fillpipe seal or pressure relief valve.  The $17 estimate is
based  on Exxon estimates,  which when  translated to consumer costs,
are in agreement  with  consumer cost   estimates provided  by GM and
Ford.

     Exxon estimates the  manufacturer's  cost  for  a fillpipe  seal
and onboard  pressure relief valve to be approximately $1.50.  ECTD
estimates the consumer cost of  an onboard  fillpipe  seal and pres-
sure relief  to be  $2.70.

Mobil

      Comparisons   of  baseline  and modified  vehicle  test  results
 indicate that Mobil  R&D  is  able  to  add  refueling controls to the
                                      C-62

-------
                                 -21-
1978 Pontiac  Sunbird  (3-way + feedback carburetor system) without
adversely affecting exhaust  or evaporative emissions.  No changes
in engine-out  or  tailpipe CO emissions are observed.   Evaporative
emissions are also  unchanged,  with both  baseline  and  modified
vehicle  test  results  near the-  2  g evaportive emission  level.

     It must  be emphasized,  however, that  Mobil and Exxon use
different  test procedures  for measuring  the  refueling control/
exhaust emission  interaction.   Mobil's test procedure  consists of
the following sequence of events:

     1)   Load  canister  to  approximately one-half  of  working
capacity.

     2)   Condition vehicle  by driving  two simulated city driving
days (4.7 LA-4's with  one hour hot soaks in between  and a diurnal
at the end of the day).
     3)-
Drain fuel tank to 10% of volume.
     4)   Refuel to 90% of volume in SHED.             •

     5)   Conduct hot start emission test.

     6)   So'ak vehicle for 11 hrs.

     7)   Conduct diurnal evaporative test  in SHED.

     8)   FTP                                '

     9)   Hot soak evaporative test  in SHED.

     Steps 1,  2,  and 5  are  the  important  differences between the
test procedures used  by  Exxon and  Mobil.  Mobil  starts their  test
sequence with  a canister  loaded to one-half of  working  capacity,
versus a saturated condition for  the Exxon  procedure.  Mobil purges
the refueling loss canister with two LA-4  driving days, versus the
Exxon method  of purging by running  a  series of  LA-4 driving  days
until  the  fuel tank  reaches  10% of  capacity.    Mobil  runs  a hot
start emission test prior to the  FTP;  no such additional condition-
ing is used  in.the Exxon test  sequence.  It  is  ECTD's opinion  that
the the  Mobil  test  sequence,  particularly  the  addition  of  a hot
start exhaust emission test,  will result in a less severe  refueling
control/exhaust emission  interaction.   This  is  due  to the smaller
quantity of  hydrocarbon  which  is purged during  the  cold  start FTP
when using  the Mobil  test  sequence.    The actual emission sensi-
tivity  to  various  test   procedure  arrangements  has  not  yet   been
determined.

Atlantic Richfield Company

     ARCO  states  that the  fillpipe  modification approach and. the
                                      C-63

-------
                                -22-
combination  fillpipe/nozzle  seal concept  are  the preferred ^ tech-
niques for  achieving  a no-leak  seal.   This recommendation is not
supported from an analysis of leak rate and durability  data because
the  test results  show that  seal effectiveness among  all  three
concepts are equal.  Cost estimates  for  the three designs have not
been  submitted.    ARCO is continuing  to collect field  durability
data on their prototypes,  but the lack of a more  extensive durabil-
ity  demonstration  under simulated conditions  of real world usage
makes it questionable  to  assume that their seals will function as
well in the field as they have in the laboratory.

     In  particular, ARCO has  not adequately addressed  the  issue of
onboard pressure relief valves versus liquid pressure relief valves
located  on  the  fill .nozzle.   Pressure relief  valves  are necessary
to  prevent  over-pressurization  of the fuel tank in  the  event of a
failure of  the  automatic shutoff  on the fill nozzle.  For the
purpose  of  fuel tank  integrity  in  the  event  of a  vehicle crash,
NHTSA  recommends that  the pressure  relief not  be  located on the
fuel  tank.   However,  a relief  valve might be  incorporated safely
with  a fillpipe seal  mechanism,  which would be sealed-off by the
fuel  cap during  vehicle operation.

      The achievement  of a  safe and  durable seal  at the nozzle
fillpipe  interface is  critical  to  the  performance  of  an  onboard"
refueling  loss  control system.   ARCO  has demonstrated that the
effectiveness  of  fillpipe  seals,  combination  seals and nozzle
seals  are  equal; but  the  design, location, and durability of the
pressure relief  valve  have not  been  adequately addressed.

      Conceptually,  a  pressure  relief  may  be  designed to  function
properly when  located  on  the  vehicle or  on  the nozzle. However,  if
refueling  losses are  controlled  on  the  vehicle,  it  is recommended
that  the fillpipe/nozzle  seal   and  pressure  relief  valve also  be
located  on  the vehicle.'  Locating all  parts of an onboard system  on
the vehicle will prevent  the  potentially serious problem of refuel-
ing a controlled vehicle  without protection from overpressurization
 (no relief valve).  Administrative  and  certification concerns  also
suggest  that onboard  controls  are  practical only if  the  seal  and
 pressure relief  are located  onboard.

      An alternative technique of achieving a seal at the fillpipe/
nozzle interface is the  liquid trap or  submerged fill.   This  seal
 concept has not been" adequately  investigated.   Submerged  fill
 offers the  potential  for  significant advantages  in terms of simpli-
 city of operation and durability  (mechanical,  magnetic,  or  elas-
 tomer type seals  are avoided).  It is  ECTD's opinion that  the
 submerged  fill concept should be investigated. • Submerged fill (and
 seal  techniques  investigated  by ARCO)  must  be evaluated in  the
 context of  a  complete refueling and  evaporative emission control
 system.  This  includes incorporating features to provide^ adequate
 thermal expansion  capability  and rollover protection while  still
 permitting normal  safe refueling.
                                      C-64

-------
                                 -23-
     2.   Vehicle Manufacturer Information

General Motors

     General Motors has several reservations concerning the appli-
cability of  onboard  controls, citing such things  as:   the uncer-
tainty  of  the  effectiveness  of  fillpipe/nozzle  seals,  potential
cost  increases  associated  with  exhaust emission  control  systems
which must be designed to control increased CO emissions, negative
fuel  penalties  which  are   the  result  of  this  increased emission
control,  and the  long lead  time which is required to obtain a
substantial reduction in  atmospheric hydrocarbon and  benzene
loading.  GM has stated that  refueling  losses can be controlled on
the vehicle  (feasibility  for  trucks  has not  been demonstrated) or
at  the service station.   GM's  disagreements  with controlling
refueling emissions  with  onboard  controls are  primarily based on
the issue of cost/effectiveness.

     More recently, (Ref.  13 & 14), GM has expressed concern about
possible  inadequacies in  the design  of the  API demonstration
vehicles.   Specifically, GM states that a  large diameter refueling
vent and a small restricted vent  are'necessary for protection  from
overfilling  the fuel  tank  and   preventing  excessive  hydrocarbon
loading of 'the storage  canister  during high  temperature  vehicle
operation.

     GM's  March, 1978 submission to EPA  presents a  summary of their
work on the control of diurnal evaporative emissions and refueling
losses using fuel tank bladders.

     It is EPA's opinion that  the  theoretical control effectiveness
of  evaporative  and  refueling loss  emissions  using  bladder   tank
technology  is  high  and  that  problems  related to  vapor storage,
pressure relief valves, and bladder  materials  can be  solved.   It
is  recommended that  bladder  tank  feasibility be researched by
funding a bladder tank hardware demonstration contract.

     The March, 1978  submission presents calculations showing  that
the additional weight of  the  components of  an onboard  control
system will  cancel out any potential energy  saving  which results
from the  combustion  of the  refueling  vapors.     ECTD  agrees   with
this analysis.

     The June,  1978  submission is basically  a cost effectiveness
analysis comparing onboard  controls with Stage  II controls (balance
displacement and vacuum assist systems).  GM estimates that onboard
control systems,  effective with  the 1982  model  year,  will  range
from $16 to $24.   These figures are  about  $5 to $9 higher than the.
March, 1978 estimates due to  higher  estimates  for  larger canisters
and a new vapor/liquid separator.   GM assumes  that the seal at the
fillpipe/nozzle  interface  will be  obtained  using  modified  vapor
                                     C-65

-------
                                 -24-
recovery nozzles.  GM does not include seal costs in its estimate.
They assume  these  costs will  be  the same for either  Stage  II or
onboard controls and hence,  leave  these  costs  out  of their analy-
sis of  both  options.   General Motor's  onboard  cost estimates are
costs  to  the consumer.   These  estimates  kre based on  costs for
hydrocarbon adsorption  systems which control  evaporative and
refueling emissions  with one  canister  and systems  which  use two
separate  canisters for  containing  evaporative  (diurnal  and hot
soak)  and  refueling emissions.   The GM cost estimates  are con-
sistent with Exxon's  manufacturers cost estimates  for  onboard
controls.  As discussed  earlier,  it is  possible to  design cheaper
refueling  loss  control  systems  by not  providing  100% control _of
refueling emissions  under worst  case conditions.    If the  design
criterion of  100%  control  for a  90%  refueling   is changed to
100%  control for  a  63%  refueling,  it   is possible  to reduce the
required working  capacity of  the charcoal canister, thus  reducing
the average  system cost to the consumer  by about  $3.00.

     GM did  not comment  on  the  feasiblity  of refueling loss  con-
trols  for light-duty  trucks  and heavy-duty  gasoline  powered ve-
hicles.

Ford

     Ford  emphasizes  that  the   refueling  loss/exhaust   emissions
interaction is a function of  the test procedure and  that the
differences  between emissions interactions measured by  Exxon and
Mobil  are due  to  test procedure  differences.   This  statement  is
correct, although the  actual emission sensitivity  to  the test
procedure is unknown.

      Ford  attributes  the high CO effects which  they have  observed
with both  conventional oxidation  catalyst  systems and  three-way
plus feedback carburetor systems to the presence of refueling loss
controls.    However,  high CO  emissions  are  likely  the  result  of
using a manifold vacuum purge control system.  This   system results
 in  cold-start  hydrocarbon  loadings  that  are  two   to three  times
higher than results  obtained with venturi vacuum  control systems
 (Exxon system).  This  is  the  reason the Ford results  are  so high,
 particularly engine-out CO emissions.  Ford maintains  that refuel-
 ing loss control  systems  produce  peak  enrichment  effects  equal  to
 two  air/fuel  ratios, which  is beyond the  capability of  their
 current  feedback carburetor control  system.  Exxon has  demon-
 strated, however,  that venturi vacuum maintains the air/fuel ratio
 within the control limits of the feedback control system.   Problems
 with the existing Ford feedback  control  system are likely to be the
 result of  response  time problems,  not control  range problems.

      Some of Ford's  concerns with  onboard refueling  control sys-
 tems,  such  as packaging, weight of onboard systems, and  the design
 of vapor/liquid separators have been examined during the API study
                                      C-66

-------
                                -25-
and shown not to be significant  problem  areas.  Other concerns with
onboard controls,  including  system durability,  onboard feasiblity
for  light  and heavy  duty trucks, and  high  altitude feasibility,
have  not been  adequately  addressed  in any of  the information
submitted to  ECTD.   It remains ECTD's  judgment  that  these issues
need  further  examination, particularly before onboard controls are
determined to be feasible for  light  and heavy-duty trucks.   Al-
though onboard  durability data  are not  available,  ECTD finds that
onboard control  systems should be  as durable as current evaporative
emissions control  systems, which last for  the  lifetime  of the
vehicle.

     Ford estimates  the  consumer cost of onboard controls for
light-duty vehicles to range from  $30 to $253.  EPA estimates that
the  consumer  cost  of  onboard  control  systems  will be  about $20
(includes $2.70 for  the cost  of  an  onboard  seal and  pressure
relief).

American Motors
     AMC's concerns with the use of onboard controls are addressed
to  the  issues of  exhaust  and  evaporative emissions interactions,
feasiblity of  vehicles  using small engines,  costs, and light-duty
truck  feasibility.   With the  exception  of feasibility for light-
duty trucks', AMC's concerns  have  been  examined  in detail by the API
study.  EPA's analysis of that  data is that refueling loss controls
are feasible for light-duty vehicles  at a consumer  cost of approxi-
mately $17.
V.
Conclusions
Feasibility

     An  Analysis  of the  available  information  has shown that
onboard refueling  loss controls are  feasible  for light-duty
vehicles  designed  to  meet  low  exhaust  and  evaporative emission
standards  (0.41  EC,  3.4 CO,  1.0 NOx and  2.0 Evap.).  However, the
feasibility  for  light-duty trucks,  particularly  the  assurance that
onboard  control  systems are compatible  with  a 2 gram evaporative
emission standard, has not been established.   Feasibility for
heavy-duty gasoline vehicles  has not been established.

     An  analysis  of  information  and  test data  presented  to EPA
regarding  the control  of light-duty  vehicle  refueling  emissions
offers the following conclusions:

     1.   Onboard  control  systems  in  laboratory  use  situations can
control  in excess  of 97% of  the uncontrolled hydrocarbon refueling
losses.
                                     C-67

-------
                                -26-
     2.   The same systems in laboratory use situations can control
in excess  of  972 of  the uncontrolled  benzene  refueling  losses.

     3.   Test results  from  two light-duty.- vehicles  equipped with
three-way catalysts, feedback  carburetors,  and prototype refueling
loss systems  show that  tailpipe CO emissions  range  from a 0 to 82
increase.

     Test results from the  same  vehicles show  that  engine-out CO
emissions range from a 0 to 14% increase.

     4.   Emission  data from  two  conventional  oxidation catalyst
equipped light-duty vehicles  show that  tailpipe CO  emissions
range  from a 0 to  10% increase.

     Data from one of  the conventional oxidation catalyst vehicles
show that engine-out CO emissions increase by  10  to  20%.

     5.   The  addition of  refueling  loss  controls  to  light-duty
vehicles does not  significantly affect evaporative emission losses.

     6.   Minor  increases in  CO exhaust  emissions   seen  for  some
vehicles  can probably  be controlled by  minor  change to either the-
refueling  loss  control  system or to the  exhaust emission control
system.   However, the  addition of  refueling  loss  controls  will
likely make  it  more difficult  to certify some vehicles to the 3.4
g/mi standard at  50,000 miles.

     7.   Onboard controls  do  not  affect  vehicle  fuel economy.

     8.   Onboard controls  do  not  affect  vehicle  driveability.

     9.   Refueling  loss control  systems  for light-duty vehicles
are  estimated to add  $17 to  the vehicle  sticker  price.   ^The^$17
estimate does not include the costs  associated  with  the  fillpipe/
nozzle seal  or  pressure relief valve.   The consumer  cost  of  a seal
and  pressure relief  in  the fillpipe  is  estimated to  be about  $2.70.
The  cost of  a seal on the nozzle should be roughly  the same  as the
cost  for a  Stage II nozzle.   However, ' it is recommended that all
components of an onboard control system be located  on the vehicle.

Lead time

      Onboard refueling  loss control  can be  implemented for  1984
model   year  light-duty  vehicles,  provided  that potential  problem
                                      C-68

-------
                                  -27-
  areas  such  as  the design  and development  of  effective fillpipe/
  nozzle  seals and pressure relief  valves  do not require additional
  hardware  demonstration programs.    It  is anticipated that  the
  fillpipe/nozzle  seal  and the  control feasibility  for  light  and
  heavy-duty trucks  are  issues  which can be resolved during the NPRM
  process.

      ECTD   estimates  that a minimum of two years  leadtime  will be
  required by manufacturers for development  (purge  system optimiza-
  tion,  design  and  verification  of fillpipe  seal  mechanisms)  and
  production  tooling changes (tooling associated with fabrication and
  relocation  of new evaporative control components).   These estimates
  are  based  in part  upon data provided  by  manufacturers  relating to
  carburetor  tooling  changes,  and in part  upon  data supplied  by GM
  relating  tOf retooling changes   for  body  panel modifications.
  Additional  time will be required for EPA to develop a  certification
  type tes't  procedure and  issue  regulations;  however, the certifica-
  tion procedure development can overlap the  production  tooling lead
  tLme,'QOnTheref°re' the  Pr°Jection  ^  that  an NPRM  can  be published
  in  1980  with final  rules promulgated  by  1981  with  the earliest
 possible implementation  date  being  1984.   (See  lead   time  chart
 Figure  1).

 Compliance  Costs

      ECTD.estimates  that  certifying light-duty vehicles  for compli-
 ance with  a refueling loss standard will  require an additonal
 nfe7nn ^nerSTT.ar at the EPA-MVEL'   T^3  « based on  an estimate
 of  100-150  refueling loss tests per year.    Facility modifications/
 equipment procurements  will cost from $30K to $80K.

      A potentially  significant impact on  refueling loss  compliance
 costs  is  Inspection/Maintenance  testing  of  light-duty vehicles.
 EPA  has  not  developed,  and  is not  aware  of, a  valid  I/M test  for
 determining the performance  of evaporative  emission  control sys-
 tems.  Monitoring  the  performance  of in-use refueling  loss control
 systems will be difficult and  cumbersome.   At this  time, it  may be
 assumed  that the onboard compliance  costs  associated with  an I/M
 test will be  equal to the cost of Stage II enforcement.

VI.  Recommendations for Future Work

     1.    ECTD  recommends  that  additional hardware  testing  be
conducted to determine the optimal fillpipe-nozzle'seal.   Addition-
  ,?'*,? °Peratlon and durability of a fillpipe  or  nozzle pressure
relief (including overfill protection for liquid  expansion) must  be
demonstrated.  The  use of an  onboard  liquid trap  seal  (submerged
till; as  an alternative to elastomer  type  seals  should  be invest--
igated.
                                     C-69

-------
                                      -28-
                                  Figure 1

                                 Lead Time
             Quarter:

Develop Certification
Test Procedure

Continued Study of
Fillpipe/Nozzle Seal
Concepts

Decision on Seal
Concepts

EIS,  EIA, NPRM
Preparation

Publish NPRM

Final Rule

Manufacturers
Lead Time
U,
                                   1980
                                   234
   Calendar Year
1981      1982      1983      1984
234I1234I1234I1234I
    -(Decision to publish service station nozzle
      requirements or put seal on vehicle)
                                           "1984 MY
                                             C-70

-------
                                -29-
     2.   ECTD  recommends  that  additional  hardware  testing be
conducted to  assess the  feasibility  of controlling  refueling
losses  on  light-duty  trucks and  heavy-duty  gasoline  powered
trucks.

     3.   ECTD recommends that the need  for  controlling refueling
losses  from  diesel  powered vehicles  be  investigated  since these
vehicles are  predicted to represent a substantial fraction of the
entire motor vehicle  population in the 1980's.

     4.   ECTD recommends that the  bladder  fuel  tank be investi-
gated as  an  alternative  to carbon adsorption technology.    It is
ECTD's  opinion  that  the theoretical  control  of  ecvaporative  and
refueling  loss  emissions with  bladder  tanks  is  high and  that
technical problems  can be solved.  It is  recommended that  bladder
tanks feasibility be  researched by funding a hardware demonstration
contract.

     5.    Finally,  ECTD  recommends  that  methods  of  reducing  the
cost_of onboard refueling control  systems  be  examined.   Such
studies  should  be  directed  toward  tradeoffs between level of
control  effectiveness  and cost.   It may  be  possible  to sacrifice
control  capacity that is  required under only  infrequent conditions
to  achieve  a proportionally more  significant cost  savings. '
                                    C-71

-------
                               -30-
                           References

1.  "Control of Refueling Emissions," Statement by General Motors
    Corporation,  June 11,  1973.

2   "Control  of Refueling Emissions  with  an Activated Carbon
    Canister on  the  Vehicle  - Performance and Cost Effectiveness
    Analysis,"  Interim Report  Project  EF-14,  prepared   for  the
    American Petroleum Institute, Washington, D.C., October 1973.

3   "On-Board Control  of Vehicle Refueling Emissions - Demonstra-
    tion of Feasibility," API Publication No. 4306, October 1978.

4.  "Summary and Analysis of Data from Gasoline Temperature Survey
    Conducted  at Service Stations," Radian  Corporation,  Austin,
    Texas.   Prepared for the American Petroleum Institute, Wash-
    ington, D.C., November 1976.

5.  "General  Motors  Commentary to  the  Environmental Protection
    Agency Relative to On-Board Control  of Vehicle Refueling
    Emissions,"  March  1978.

6.  "Suppplement to  General Motors Commentary to the Environmental
    Protection  Agency  Relative to" On-Board  Control  of  Vehicle
    Refueling Emissions," June  1978.

 7.  "Ford  Motor  Company Response to EPA Concerning Feasibility  and
    Desirability of  a Vehicle  On-Board  Gasoline Vapor  Recovery
10,
                                                              and
    System."

8.  "Ford  Motor Company Position Concerning Feasibility
    Desirability of Vehicle On-board Refueling  Vapor Control
    Systems," November  6,  1978.

9.  AMC letter to Paul  Stolpman, August 3, 1978.

    "Cost Estimations for Emission  Control Related Components/Sys-
    cems  and Cost  Methodology Descriptions,"  Rath and  Strong,
    Inc.,  Lexington,  Massachusetts.    Prepared for  the  Environ-
    mental  Protection  Agency, Ann Arbor,  Michigan, March  1978.
11.  "Investigation and  Assessment
     tive  Emission  Sources  and
                                   of Light-Duty Vehicle Evapora-
                                  Control,"  Exxon  Research  and
 12.
    Engineering  Company,  Linden,  New  Jersey.   Prepared  for  the
    Environmental Protection Agency,  June 1976.

    Texaco  statement  submitted to Paul  Stolpman,  July 18, 1978.
                                     C-72

-------
                               APPENDIX
         ^ Appendix contains detailed descriptions and data from the
 test vehicles and fillpipe/nozzle  seals  which were used in the most
 recent testing  and  evaluation of  refueling  loss control systems.

 Exxon

      Table  A-l  presents a description of  all  Exxon test  vehicles.
 Figure A-l  is a schematic  of the basic control system designed for
 the  Chevrolet Caprice and  the Ford Pinto.   The refueling emissions
 (RCS)  canister  controls  both  refueling emissions  and diurnal
 evaporative   emissions;  the   evaporative  emissions  (ECS)  canister
 controls  carburetor  hot soak losses.    Exxon  investigated  several
 different  purge  mechanisms,  including  combinations  of  manifold
 vacuum and  venturi vacuum, and  two  stage purge  control valves
 controlled  by ^ fuel  volume,   but venturi  vacuum, which is  propor-
 tional  to  engine air  flow,  is the most effective  purging  method.
 Exxon's control  system is  designed to  maintain  the total  purge air
 volume  (RSC   +  ECS)  equal  to  the  purge  air volume  of.  the  unmodi-
 fied vehicle's evaporative control system.

     The air  bleed control valve, shown in Figure A-l,  is  necessary
 because the'RCS  canitser is  purged more  efficiently (higher hydro-
 carbon purge per unit  volume  of air) than  the  unmodified  ESC
 system-, thereby  resulting  in richer A/F mixtures.   This  air  bleed
may  not be  necessary  for  other  vehicles wich  feedback carburetor
controls.

     figure A-2  is a plot of the RCS canister purging as a function
of time.   These data  are  based  on consecutive  LA-4 driving  days.
As noted,  the RCS system  is  purged at  a  rate  of about  4  litres/
min., which  corresponds  to  a total canister purge volume of  about
40 litres during an LA-4 driving  cycle.

Mobil

     Specifications for the vehicle Mobil has modified  for refuel-
ing loss control are  summarized as  follows:

     Vehicle:  1978 California Pontiac  Sunfaird

     Engine Size:  151 cu.  in. L-4

     Interia Weight:   3000  Ibs.

     Emission Control  System:
          Exhaust:      3-way  catalyst  with  feedback  carburetor, •
                        EGR
          Evaporative:   Carbon canister
                                     C-73

-------
                                 A-2
      Fuel  Tank Capacity:   18.5  gallons

      The production vehicle  is  modified for controlling refueling
 emissions  by enlarging the existing carbon canister, (one canister
 controls  refueling,  diurnal,  and  hot  soak loss),-  enlarging  the
 vapor line between  fuel  tank and canister,  redesigning the vapor/
 liquid  separator,  and  installing a  purge  control orifice between
 the canister  and intake manifold.   A  schematic  of  the Sunbird's
 control system  is shown  in Figure  A-3.   Various flow  control
 orifices were inserted in the canister  purge line but best results
 are obtained  with  an orifice  of 0.100  in. diameter.   Mobil uses
 1550 grams of Calgon BPL-F3  carbon for  their control system, which
 assumes a 20% safety factor.   This  quantity  ofl  carbon is based on a
 90Z fill of the 18.5 gallon tank, and assumes ia hydrocarbon loading
 of six grams per gallon of dispensed  fuel.   The working capacity of
 the canister  is  approximately  240  grams.  The  basic components of
 the canister  control  system are shown  in  Figure A-4.   The  ported
 vacuum  purge control valve is from a  1978 Chevrolet  Impala  evapora-
 tive  canister,  while  the two  fuel tank  vapor valves  (two are
 used  to reduce  the pressure drop  during the  refueling operation)
 are carburetor bowl valves from a 1978 Impala.   Using two  fuel tank
 vapor valves  results in  fillpipe pressures  as  low as two  inches of
 water  pressure  during  refueling.  The  fuel tank vapor valves are
 also  controlled by manifold  vacuum such  that  the vapor valves
 are  closed  when manifold  vacuum is   present at the control  port.

 Atlantic  Richfield Company

       Figure  A-5- shows  the  fillpipe  seal which  ARCO has  developed
 and  tested  for durability.  Tables A-2 and A-3 are  typical  of the
 durability  results obtained with  this seal.   Figure A-6 is an
 example of-a prototype  combination  fillpipe/nozzle  seal  which has
 been  developed and evaluated by ARCO.
                                               •i'
 Ford

       The  vehicles which  Ford  has  used  for refueling  loss  testing
 are  shown in Table A-4.   A single 4.35 I  canister  is used  in the
 Mustang,  while a duel, canister  system,  829 ml  and  3.4 1,  are used
 for   controlling  carburetor vapors  and diurnal/refueling  losses,
 respectively,  in the  Pinto.   The purge systems for  the Mustang and
' the  Pinto are shown in  figures A-7 and  A-8.

       Figures A-9 and A-10 are  plots  of  canister  loading versus test
 procedure sequence.   These plots  indicate that Ford's  refueling
 loss control system  is quite  sensitive  to  the particular  test
 procedure which is used to quantify the refueling   control/exhaust
 emission interaction.
                                      C-74

-------
                                      A-3
                                    Table A-l

                              Vehicle Descriptions
   Make
          Engine Displacement/
Model         Configuration
Control Systems
Fuel Tank
Capacity
(gallons)
Chevrolet   Caprice    5.0 litre (305 CID)/V-8

Ford        Pinto      2.3 litre (140 CID)/L-4


Plymouth    Volare     3.7 litre (225 CID)/L-6

Chevrolet   Chevette   1.6 litre (98  CID)/L-4
                                     Ox. Cat., AIR, EGR

                                     3-Way, Ox. Cat.,
                                        AIR,   EGR

                                     Ox. Cat., AIR, EGR

                                     Ox. Cat., AIR, EGR
                         21.0

                         13.0


                         18.0

                         12.5
                                          C-75

-------
                            Table  A-2
                       FILLPIPE MODIFICATION

                        ROTARY SEAL-CR 7538
                      LEAK RATE AS AFFECTED BY

                     FILLNECK PRESSURE AND WEAR
NO. OF SPOUT
INSERTIONS
0
100
100
100
100
100
100
100
100
100
100
TYPE
SPOUT

Smooth
Rough
Smooth
Rough
Smooth
Rough
Smooth
Rough
Smooth
Rough
CUMULATIVE
INSERTIONS
0
100
200
300
400
500
600
700**
800
900
1000
FT3/MIN
@ 5" W.C.
0
0
0
0
0
0
0
0
0
0
0
LEAK *
(§'15" W.C.
0.001
0.001
0.001
0.001
0.001
0.001
0.001
0.001 '
0.002
0.00.2
0.002
*  Leak rate average of  six nozzle insertions.

** Expected number of insertions during vehicle  life.
RGJ:ip
7/13/78
                              C-76

-------
 HOURS  OF
 LIQUID
 SOAK

     0

    16

    35
                           Table  A-3


                     FILLPIPE MODIFICATION

                      ROTARY SEAL-CR. 7538
EFFECT OF LIQUID
SOAK ON SEAL ID-
TOTAL WEEKS
OF VAPOR
SOAK
0
0
0
2
3
4
5
6
7
8
AND VAPOR
AND' LEAK

SEAL
ID, IN.
.712
.712
.711
.705
.699
.701
.703
.698
.693
.691
GASOLINE
RATE*

FT3/MIN LEAK**
w J yv*~ (9 J.3 we
o o
o o
0 o
0 o
0 .001
0 .001
•001 ..001
0 o
o .001'
•001 .002
 *   Vapor and liquid soak at 72°F.

 **  Leak  rate average of nine nozzle insertions.
RGJrip
7/13/78
                             C-77

-------
                                      A-3
                                    Table A-l

                              Vehicle Descriptions
   Make
Model
Engine Displacement/
    Configuration
                                                   Control Systems
                                                            Fuel 'Tank
                                                            Capacity
                                                            (gallons)
Chevrolet   Caprice    5.0  litre  (305 CID)/V-8

Ford        Pinto      2.3  litre  (140 CID)/L-4


Plymouth    Volare     3.7  litre  (225 CID)/L-6

Chevrolet   Chevette   1.6  litre  (9.8^ CID)/L-4
                                     Ox.  Cat., AIR, EGR

                                     3-Way,  Ox.  Cat.,
                                        AIR,    EGR

                                     Ox.  Cat., AIR, EGR

                                     Ox.  Cat., AIR, EGR
                                                     21.0

                                                     13.0


                                                     18.0

                                                     12.5
                                            C-78

-------
I
    CO
    s
    UJ

    03

    03
    J
    O
    cc

    2
    o
    o
g
O3
03

2
LL!

o
3=

LU

LL
LU
CC

Q
    LU



    <
    CC
    o
    CL
    <

    LU
                                       C-79

-------
*>
 a
O
 s

-------
       CO


       O


       to



       UJ

       O
       •2.

       __i
       UJ
<
u
3
00
      o
      o:
O
O
      C/l
     O
     CO


     O.

-------
                                                                             3»  M » *• •
                                                                             Q- t/N ro
                           _l
                                                                              S .2>E
     (O
                                                                                            oo *"
                                                                                            "
                                                                                             I   ™
00
      o>
      
-------
  FILL PIPE MODIFICATIONS
  ROTARY SEAL     ,   ..
                    ROTARY SEAL
  TRAP DOOR
                                         SPOUT
           LEAD RESTRICTOR
FILL PIPE  MODIFICATIONS
ROTARY SEAL
                  ROTAPY SEAL
TRAP DOOR
                                           SPOUT
         LEAD RESTRICTOR
                 C-83

-------
           Figure  A-6
NOZZLE / F1LLPIPE MODIFICATION
CONE SEAL
                                           SPOUT
 LEAD RESTRICTOR
 TRAP DOOR
                                    •LATCH COLLAR
        CONICAL SEAL
NOZZLE / F1LLP1PE MODIFICATION

CONE SEAL

 LEAD RESTRICTOR
 TRAP DOOR
SPOUT
                                     LATCH COLLAR
         CONICAL SEAL
                C-84

-------
s
C=3
H

19
CO
     CO
     CO
         a
I
 O
     CO
      c co
      ffl
      ^j
      en
      3
      S
                                                                                                           Figure  A-7
                  a &a
                  EH O
                  H M
                  63 05 3
                  > O W
                  i-3    CU
                  < 3 O

                  >Ms
                  63 O O
                  O CO =3
                  tf rH Q
                  5   • •*!
                  CM O >
         63
            Ce5

So


5^
S a
=3 >

^5=
> >
to cu

> cH


> 2

EH S
a o

3 O
r-l EH
                                                    i
CO =>
3 CO Crf
I-H CO
<*<*;&
EH CU ^
2H ( ^
8X EH
C3
CS
a • o
> 3 S*
i-J 1— 1
«a! 63












> ^° «
-S- 1-1 CO
CO O Sai &3 J
~ * z)
S < 3



CO 3
° s
-j pH
P3
^
0









s:
o
i— <
EH
O
a
3
3
O
O
EH
*£t
&a
>
i
                                                                                                   .
                                                                                              C!  >
                                                              C-85

-------
p.
 S5S
 J3 ^ _
 """s.£ 5
     a
  S12
  z;
  o
                                                                Figure  A-8
                                       C-86

-------
                              Procecurs £2 Set 2
                         •   Mustang 5.CL (8218) 197$ 1*9  States
                                Date 7-17-78 Test 29
                 - 1*350 nl Canister, Tank & Garb. Eowl w/.lcG" Purge Orifice
                  |-      I    GrarLS of Vapor Purged (-)j Absorbed (+)
                                                               Figure  A.-9
             +55  I -98  I +27  1-50  I   +35 I  -38 I +39  I  +8   I +95  '  -7^ 1   +18 !
si
2-4
a
•p
03
•H
c

o
                                                              Gas.
                                     "Gss/ifi.  CO

                      ON BOARD SIS.  Bag 1,22:3 «J
                      	Bag 1 CO C-=3.
                                      Bag i CO- Gcs.

                      BASZLIIS SIS.  Sag 1,2*3 CO Grs.
                                      Gas/Mi  CO
                             15-9
                              205
                                                                       118
               r-4 O
                     r-l n
 CS
•H
-1-3
      . o   i
       '->   '-> O
       CU   HO
                                  CM '
                              r-l   S^
                            - d
                            32   ^
                            o -   o
                            — —   s
                              — !   U
                            r-i ^S   6-1 "
 ** S
'2. "
"~"
         ej   !-i
         a   =:
S   f-l JJ
     i  a
    3i-l  O
    iH C/3
                    n
                    en
                                                               o
                                W S
  ja

s- 3

§"
                                     Test  Seausnce
                                      C-87

-------
                                                                 to
                                                                 
03
O
     X f->
     O O
     CO SI
     •4^   T3
    I CO ^ O
    - o n to
d   co c a
T3 J  I  c5 p
O O C—O,
O  •      &-I
O lf\ O r-j O
p   -Pa


       JLPk 3
       CO t|
   Cj   ^" O
               9-
              s
ir\
^r
                 r-
                 RT
                 o
                                                               vo
                                                                                °aSES '-Z
                                                                                        H  T
                                                               CO
                                                               CO
                                                                                                 o

                                                                                                 o
                                                                                   TITJ
                                                                                    TCTc  <0^r
                      001 I
                               ^o trt -^ -3^37^0 -i=  Q^

-------
                                                                               o
                                                                               o
00
    co
    a
    co
    tr
    Cd
    2
    o
    CO
fc.
o
    M
    t^*

    r^
    co
                *
                «
                *
              o
              o
                                                                          *
                                                                          »
                                                                          •»
                                                                             h-

                                                                             o
                                          o

                                          ^



                                       0 I
                                       1C H>

                                          u.
                                          O

                                          H-
                                          z
                                       O LU
                                       T O
                                          c
                                          lit
                                          0.
                      o
                      
-------

-------
       APPENDIX B
  LDV and LDT Operation
and Usage Characteristics
           C-91

-------
                           Table  B-l

                        Survival Rates
Age
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
Survival
LDV
0.999
0.990
0.980
0.953
0.922
0.882
0.833
0.765
0.685
0*589
0.507
0.408
0.324
0^259
0.20-8
0..168
O.,142
0..134
0.126
0.118
Rates
LOT
1.000
0.999
0.996
0.972
0.940
0.917
0. 886
0.852
0.818
0.780
0.735
0.688
0.632
0.575
0 . 512
0.450
0.390
0.325
0.270
0.210
Source:    Letter,   David   Lax,   Energy   and   Environmental
           Analysis,   Inc.   to   Robert   Johnson,   U.S.   EPA,
           November 10, 1983.  Survival  rates  used  in  the Ninth
           Quarterly  Report  of   the  Highway  Fuel  Consumption
           Model.
                               C-92

-------
                           Table B-2
Annual
Year
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
Vehicle Miles

LDV
13,925
13,125
12,372
11,661
10,992
10,361
9,766
9,205
8,677
8,179
7,716
7,274
6,857
6,464
6,094
5,745
5,415
5,105
' 4,813
4,537
Travelled
VMT
LDT
17,739
16,425
15,208
14,082
13,039
12,074
11,180
10,352
9,585
8,875
8,218
7,610
7,047
6,525
6,042
5,595
5,180
4r797
4,443
4,113
Source:  Draft  MOBILES  Documentation,   data  provided  by  Lois
         Platte, U.S. EPA, QMS, February 14, 1984.
                              C-93

-------
                           Table B-3
         Registrations and Vehicle In-Use Fuel Economy
            Total Registration  [11
Fuel Economy [2]
Year

1969
1970
1971
1972
1973
1974
1975
1976
1977
1978
1979
1980
1981
1982
1983
1984
1985
1986
1987
1988
1989
1990
1991
1992
1993
1994
 1995
 1996
 1997
 1998
 1999
 2000
LDV
9.530
8.460
9.960
10.610
11.480
8.790
8.200
9.750
10.820
10.820
10.450
8.831
8.051
7.661
8.165
9.200
9.744
10.070
10.454
10.582
10.521
10.457
10.683
10.818
10.818
10.728
10.818
10.818
10.818
10.818
10.818
10.818
LDT
1.492
1.423
1.627
1.973
2.484
2.183
2.020
2.799
2.790
3.229
2.741
1.943
1.664
1.814
1.995
2.328
2.563
2.661
2.854
2.768
2.779
2.668
2.792
2.865
2.866
2.871
2.939
2.939
2.939
2.939
2.939
2.939
LDV
14.91
15.26
15.0"
15.17
14.71
15.53
13.46
14.87
15.59
16.95
17.25
20.05
21.44
22.23
22.20
22.80
23.50
24.30
25.20
26.30
27.50
28.80
30.30
31.60
33.00
33.63
34.22
34.22
34.22
34.22
34.22
34.22
LDT
10.88
10.98
10.99
11.02
11.02
10.91
10.49
11.16
10.91
10.92
11.73
14.08
15.68
15.96
16.37
16-. 82
17.31
17.98
18.61
19.28
19.93
20.60
20.66
20.73
20.79
20.85
20.91
20.91
20.91
20.91
20.91
20.91
 ?1?  "The   Highway  Fuel   Consumption   Model-Ninth  Quarterly
   1  Report, Energy  and Environmental Analysis,  Inc.,  for U.S.
      Department of Energy, February  1983   (millions  of vehicles)
 [2]  Historical and projected  in-use fuel  economies  (mpg).
                                C-94

-------
                           Table B-,4

             LDV Age-Odometer Usage Pattern Factor
Age
1
2
3
4
5
6
7
8
9
10
11
12
13 . "...
14
15
16
17
18
19
20
Odometer LDV
1.013
1.014
1.005
0.997
0.994
0.992
0.967
0.97.6
0.954
0.939
0.935
0.918
0.917
0.910
0.898
0.896
0.862
0.829
0.805
0.800
Source:    "The Highway  Fuel Consumption  Model-Ninth Quarterly
           Report, Energy and Environmental Analysis, Inc., for
           U.S. Department of Energy, February 1983.
                              C-95

-------
   Rate of Misfueling By Means of
     Riel FiUer Inlet Tampering
            Versus Mileage
                                     Legend
                                    o J/MLDV
                                     I/M Ragraaaion
                                    • I/M LOT

                                    • NorH/WLDT
01234   5   6789   10

Vehicle Mileage In 10,000 Mile Increments
                  C-96

-------
  Rate of Evaporative Canister Disablement

                   Versus Mileage
   25-
g.
C=3

co  15-

-------
                 APPENDIX C
                 Tables  from

 "Manufacturing Costs and Automotive Retail
     Price Equivalent of On-Board Vapor
Recovery System for Gasoline-Filling Vapors"
                     C-98

-------
                                   TaM*  1
                                  COMPARATIVE COSTS
EVAPORATIVE CONTROLS
BASELINE VEHICLE SYSTEM
Division
Component Vendor
Assembly Costs
Canister .- 2.698
Purge Control .740
Vacuun Signal
Theraal Switch
Liauid-Vapor .20
Separator
Fill Pipe ' .300
Seal Assembly
Hose .700
Tubing 1.200
Cost to
Customer
RPE
6.293
1.727
EVAPORATIVE CONTROLS ON-BOARD
VEHICLE FILLING VAPOR CONTROLS
Division
Vendor
Costs
5.396
1.480
2.631
I .441
.460 ! .910
.700 ' 1.894 ''
1.600 1.400
2.76 2.400.
Cost to
Cus toner
RPE
12.386 .
3.454
6.138
.1.028
2.123
4.410
3.200
5.42
Costs
Delta
Vendor
Division
2.698
.740
2.631
.441
.710
1 . 594
.700
RPE
6.293
1.727
6.138
1.028
1.953
1.194
. 1.600
1.200 ' 2.760
Hardvare           .40
  C Modifications
Vehicle  Assembly   .98
rOTAL
 .92
1.104
                                           .80
                                          2.200
-.720
                         1.84
                         5.06
1.656
                                                                   .80
                                                                  2.200
                                                                   .240
                                                                 12.954
                                                1.84-0
                                                5.060
                                                 .552
                                               29.855
                                      C-99

-------
                                      TYUe 2
                       RETAIL PRICE EQUIVALENT AT  THZ VEHICLE LEVEL
Part
Plant
  or
Vendor
Selling
Price    R&D
Invest
Tools &
Equip
                                                Corp
                                              Allocation
                                                .20 VC
Corp
Profit
.20 VC
Dealer
Markup
.40 SP
Canister           2.698

Purge Control        .740

Vacuun Solenoid    2.631

Thermal Switch       .441

Vapor-Liquid         .910
Valve

Fill Pipe  Seal     1,.894
                             .539

                             .148

                             .526

                             .088

                             .182


                             ,378
                        .539

                        .148

                        .526

                        .088

                        .182


                        .378
          2.517

           .691

          2.455

           .411

           .849


          1.76
Vehicle
Retail
Price
Equiva-
lent
          6.293

          1.727

          6.138

          1.028

          2.123


          Jt.41.
                                        C-100

-------













*n
_*>
_o
&
1-

































>
eg
o
C
2
O
O
CO
>
. Q
H
H
C
>••
ce
£3

b
<
u.

21
<
j
<
o
fr-
















ee
vt e
Q • ^4 4)
*3 Q.i-4 O
SS S- «-* —
Si O 9J I*
S» O so a,
^ C^
. -i 3C
i- "o o
O fa «M
U Bo •

• BS 3C
C* J^
u ca C
O C «N
U U •

C
00 p"*
e
o
o •
J- e.
K
Ed

i_i jg
C • *•>
« eo a
— lio C
a. S U




w
Q
01
u J= C
C U «T
Q U '



O
.0
93

•
a
Z





•



b
S
a.


CO O 1-4 -i
^^ • *? ^^ *(P
vo r» ss «r
«S CM

*^ ^D ^3 C^
£ ^^
ft O en o

— C O C


m  C cr>
CO «» CC - O 4»
0 3 «J JZ
O 0. > H


0
<_l


C
CN

O
••4

O


0

o
sO









2



o
o

f>



l^
o
a.
a
>
I

-4 V
3 >
CT i-*
«4 a
u >


1
—

PS)

g

O
'

in
O

vO
(S
*







«N
sC
-T
*


0
t~>
ri

^^
«
-------
                                         BILL OF MATERIAL
MANU7ACTi.fi.iNG COSTS
Component
Description
Canister Body

Grid

Int. Filter

Ext. Filter

Activated Carbon

Canister Cap
Canister Connectors

Assembly
TOTAL
Purge Control Valve
Valve Body
Diaphragm
El. Connector
TOTAL
Vacuua Solenoid
Housing
Cap
Coil
Armature
Spring
Vacuum Springs
Diaphragm
Vacuum Hose
Air Nozzle
Solenoid Connector
Material
DB437

67

AY332

KZ1-4

5448

DB437
DB437



Assam.
DB437
ED38
1742

As sea
Alum.
Alum.
1752 Cu-Fe
Fe-HPM
820
820
ED38
Aluminum
Fe-HPM
Cu-1752
Weight
Founds
.30

.10

.10

.20

.50

40
. .05

—

—
.20
.01
.0.2

—
.20
.10
.05
.03
.01
.02
.01
.03
.02
.01
•Material
Costs
.24

.04

.10

.20

.50

.08
.04

M
1.200
—
.16 "
.01
.02
• 19
_
.300
.150
.075
.045
.010
.016
.010
.045
.020
.010
Labor
Overhead
Hrs. 1.40 '
.10
.01
.01
.001
.01
.001
.01
.001
.01
.001
. .01
.001
.01
.001
.10

• 95
.05
.01
.01

.100-
. 100
.100
.150
.020
.010
.010
.010
.050
.010
.005
.14

.014

.014

.014

.014

• 014 •
.014

.14

.07
.07
.014
.014

.140
.140
.140
.070
.028
.014
.014
.014
.070
.014
.007
Mfg.
Costs
.480

.054

.124

.224

.524

.104
.064

.240
1.820
.120
.280
.034
.044
.478
.240
.540
.390
.195
.093
.034
.044
.034
.165.
.044
.022
Reference
Current Siz

See Sketch









•»•


See Sketc




See Sketc








TOTAL
                                                    .681
1.801
                                         C-102

-------
S -3.0 ( e S
BILL OF MATEXIAL
MANUFACTDSING
'oaponent
Ascription
"hermal Switch
•S - Body
pring
Tsytal
onnector
OTAL
apor-Liquid
eparator
ousing
tg. Flange
oao
-Ring
ronnet
sal
loac
Material
Assent*
Aluminum
820
Br-Si
Cu 1752

As sen.
1010 Steel
1010 Steel
Absorb
ED38
ED38
ED38
Steel
Weight
Founds
-
.05
.01
.01
.01

-
.20
.10
.05
.01
.02
.01
.05
Material
Costs
-
.075
.010
.030
.010
.125

.160
.080
.050
.005
.020
.005
.040
COSTS

Labor
Overhead
.03
.02
.005
.010
.005
.070
.01
.020
.010
.030
.005
.010
.005
.010
.042
.028
.007
.014
.007
.098
.014
.028
.014
.042
.007 .
.014
.007
.014

Mfg.
Costs Reference
.072 See Sketch
.123
.022
.054
.022
.293
.024 See Sketel
.208
.104
.122
.017
.044
.017
.064
OTAL
-.360
                                                       .10
.140
.600
                                         C-103

-------
                                         BILL OF MATERIAL
MANUFACTURING COSTS
Component
Description
Fill Pipe Rotary Seal
Lead Restrictor
Guide
Rotary Seal
Material
« Assent.
Steel
plastic
ED38
Weight
Pound s

.30
.20
.10
Material
Costs

.15
.20
.10
Labor
Overhead
.120
.220
.020,
.010
140
280
028
014
Mfg.
Costs
.240
.630
.248
.128
Refer en'c
Fig. 5A



TOTAL
                                                    .45
.330   .462  '  1.246
                                          C-104

-------
                APPENDIX D
DETERMINATION OF IN-USE EMISSION REDUCTION
      BENEFITS OF STAGE II PROGRAMS

-------

-------
                             I..   METHODOLOGY

      Chapter  5.0  of  this  document  sets  forth the emission  reduction
 benefits which would be attained by each Stage  II and onboard control
 program option if the control systems performed optimally  -i.e., at the
 efficiencies  achieved during the California Air Resources  Board (ARE)
 certification tests,  in the case of Stage  II systems, and  in the American
 Petroleum  Institute  (API) demonstration tests, in the case of onboard
 control systems.   The emission reductions which would actually be
 achieved in the field by  each program option are set forth as a function
 of the level of effort employed in enforcing each option — the level of
 effort being expressed in terms of frequency of in-use inspections, as
 well  as in terms  of  person-years of resources necessary to perform the
 inspections and the  associated .legal follow-up work.  Appendix C deals
 with  in-use (with  tampering) estimates for onboard controls.  This
 appendix deals strictly with Stage II controls.  The input data for
 this  Stage II analysis were derived as follows:
      1.  Potential failure modes were identified for each type of vapor
 recovery system (balance;  hybrid; and vacuum assist).  These failure
modes were grouped under the three general  headings of misinstallation,
 improper maintenance and tampering.  See Table D-l for a list of failure
modes and Part II  of this  Appendix (p. D-17)  for,an explanation of the
 system defects covered by  each failure mode.
     2.  The average percentage  reductions  in control system
efficiencies attributable  to the various failure modes were estimated.
See Table D-2 for  figures, and Part II for  assumptions.1
1
 The numbers appearing in Table D-2, as well  as those appearing in
 Table D-3 (see §3,  immediately following in  text)  represent the best
 estimates available at the present time.  They are based on the data
 presently available respecting Stage II technologies, or are estimates
 or extrapolations based on existing knowledge.  These numbers are
 subject to future revision as  refinements occur in the state-of-the-art
 control  systems and as further information becomes available regarding
 the in-use performance of the  various control  systems.
                                   D-3

-------
                                             TABLE D-l
                                      Potential Failure Modes*
                  Mi si retaliation
                  Improper Maintenance
                      Nozzles
                      Hoses
                      Processing Unit
                      Aspirator Units
                      Canister System
                      Fill pipe Seal
                  Tampering
                      Nozzles
                      (user-convehience,
                      economics or
                      fuel-switching
                      motivated)
                      Nozzles, hoses
                      (alleviate
                      misinstallatibn
                      problems)
                      Processing unit
                      Canister System
                      Fill pipe Seal
         Stage II Systems
Balance   Hybrid       Vacuum-Assist
XXX
X
X
X
X
X
X
X
                             X
*Explanation of the specific defects covered by each failure  mode  appears  in  Part
 II of this Appendix at pp. D-17 to D-44.
                                           D-4

-------
                               TABLE D-2
                 Average Percentage Reduction In System
                      Efficiency Per Failure Mode
Misinstallation
Improper Maintenance
    Nozzles
    Hoses
    Processing Unit
    Aspirator Unit
    Canister System
    Fill pipe Seal
Tampering
    Nozzles
    (User-convenience,
    economics or
    fuel-switching
    moti vated)
    Nozzles, hoses
    (Alleviate
    misinstallation
    problems)
    Processing Unit
    Canister System
    Fill pipe Seal
             Stage II Systems
 Balance   Hybri d          Vacuum-Assist
  5%
 22%
 10%
  10%

  15%
  10%

   7%
100%
 100%
100%
100%
  1%

  5%
 50%
 75% (Weighted
      Average)
100%
100%
                              100%
                              D-5

-------
     3.  The percentages of in-the-field systems2 which would exhibit

the various failure modes, assuming there was minimal in-use enforcement
(i.e., assuming compliance was essentially voluntary)3 were estimated.

See Table D-3 for estimates, Part II for assumptions.


     4.  Using the failure mode rates from Table D-3, and the recovery

efficiency decrements frpm Table D-2, the average in-use recovery

efficiency for nozzle-fill pipe losses for the various types of systems

(minimal enforcement case) were calculated.  See Table D-4 for results.


The calculations were made according to the following formula:
Average
In-Use
System
Efficiency
  n
 E
1 = i
                            [100
  N
 E
 j  = i

  n
 E
i = 1
(A,)
where
                                        N
                                       E
                                      j = i
(T.E.
     n = total number of permutations and combinations of failure modes
         occurring with each control system type

     A-} = percentage occurrence of each combination of failure modes

     Ei = in-use system efficiency associated with each combination of
          failure modes

     N = total number of failure modes for each recovery system type

     Bj = that portion of the percentage occurrence of each failure
          mode not occurring in combination with any other failure mode
 A "system." as used in this context, refers to the combination of
nozzle, return hose, and return line components peculiar to each
individual nozzle plus, in the case of manifolded Stage II systems, the
common components at the installation site.  The number of "systems" is
thus equal to the number of nozzles.
3
 "Minimal Enforcement," or "Voluntary Compliance" means, in the case of
Stage II programs, the situation resulting if virtually no resources,
State or federal, were allocated to program enforcement.
                                  D-6

-------
                              TABLE 0-3

                   PERCENTAGES OF IN-FIELD SYSTEMS
                   EXHIBITING VARIOUS FAILURE MODES
                      (Minimal Enforcement Case)
Misinstallation

Improper Maintenance

    Nozzles

    Hoses

    Processing Unit

    Aspirator Unit

    Canister System

    Fill pipe Seal

Tampering

    Nozzles
    (user-
    convenience,
    economics or
    fuel-switch ing
    motivated)

    Nozzles,
    Hoses
    (alleviate
    misinstallation
    problems)

    Processing Unit

    Canister System

    Fillpipe Seal
                Stage II Systems


      Balance     Hybrid

    ? TO-1 TO       1  ^%-7 fW
    L* • +J tQ j, *J fO       .L • fJfO I • -J fO
   45%

    5%
15-20%
.85-5%
   11%

    5%



   30%
10-15%
.4-2.5%
Vacuum-Assist

    0-2%



      7%

      2%

      2%
  10-15%
  negl.
                                         15%
                               D-7

-------
                                        TABLE  D-4

       Average  In-Use  System  Recovery  Efficiencies  for Nozzle-Fill pipe  Interface
       Losses (Minimal  Enforcement Case,  Steady-State)
  Balance

    68%
Stage II Systems

 Hybrid           Vacuum Assist

  78%                  69%
                                  TABLE  D-5           .   .  .

               Calculation of Stage II Weighted Average  System
                Recovery Efficiency for  Nozzle Fillpipe  Losses
            (Minimal  Enforcement Case/Non-Installation Rate  = Zero)
             WEIGHTED AVERAGE
             SYSTEM RECOVERY
             EFFICIENCY
                 68% x 80%
                 78% x 15%
                 69% x  5%
54.4
11.7
 3.5
7t5%~~
                                  TABLE P-6

                   Rates of Total  Noncornpliance (Failure to
                  Install) for Stage II Programs as Function
                      of Time (Minimal  Enforcement Case)
  Time Elapsed from Final
Compliance Deadline (Years)
          3  and thereafter
                 Rate of
            Total Noncompliance

                   40%

                   30%

                   20%
       Overall Stage II
      Program Efficiency

             42%

             49%

             56%
                                         D-8

-------
     Ej   = in-use system efficiency associated with each failure mode
     I.E. = optimum system efficiency (i.e., the efficiency achieved in
            the ARB or API tests).
     The quoted formula gives the weighted average of the efficiencies
     of all in-use systems of a given type.  The first part of the.
     formula represents the weighted average efficiency of those systems
     containing a combination of two or more failure modes.  The second
     represents the weighted average efficiency of systems with single
     failure modes; the third, of systems with no failure modes.  In
     the case of combined failure modes, the net in-use efficiency is
     determined by multiplying the efficiencies associated with the
     single failure modes making up the combination.  This procedure
     involves the dual assumption that the effect of any failure mode
     is independent of the effect of other failure modes and that the
     effect of simultaneous failure modes is multiplicative.  The
     distribution of failure modes among systems was assumed to be
     random, except in those instances where the types of failure mode
     were mutually exclusive.
     Sample Calculation - Assume that a group of balance systems has a 40%
     rate of improperly maintained nozzles, a 20% rate of improperly
     maintained hoses, and a 30% rate of tampered-with nozzles.  The
     average in-use system efficiency would be determined as follows:
          Average
          In-Use     = 8% (.95 x .78 x .9)  +   32% (.95 x .78)  + 6% (.95 x .90)
          System       [Nozzle,  Hose Defects]  [Nozzle Defects]  [Hose Defects]
          Efficiency
                            + 30% x 0       + 24% x .96
                         [Nozzle Tampering]   [No Defects]
          Average
          In-Use
          System
          Efficiency
= 57%
    ,5.   The aggregate in-use recovery efficiency (% of fill pipe losses
actually recovered)  of Stage II  systems,  assuming there was no absolute
noncompliance (outright failure  to install  systems)  was calculated,
                                  D-9

-------
using the percentage efficiency for each type of system from Table  D-4
and the following distribution of throughput coverage by system type:4
          Balance       -  80%
          Hybrid        -  15%
          Vacuum        -   5%
The results appear in Table D-5.
     6.  The percentage of gasoline-dispensing facilities absolutely
failing to comply with Stage II regulations (by not installing any
system whatsoever) in a minimal-enforcement situation, was estimated as
a function of time.  The results appear in Table D-6.  Bases for
projections are set forth in Part II of this Appendix (p. D-37).

     7.  Estimates were then made of the impact on the failure mode
rates set forth in Table D-3, and on the absolute noncompliance rates
set forth in Table D-6 of various levels of enforcement effort" ~
expressed in terms of frequency of in-use inspections of regulated
outlets.  Three scenarios are depicted:  bi-annual, annual, and quarterly
inspections.  The results appear in Tables D-7, D-8, and D-9, respectively.
The impact of the various levels of enforcement effort on misinstallations
and on absolute noncompliance (total non-installation) was determined,
in each case, by assuming that the level of enforcement resources
utilized in the installation phase of the program would be the same as
that utilized in later years to perform in-use inspections and that,
in the installation phase, these resources would concentrate upon
assuring proper installation of systems.  The results with regard to
absolute noncompliance rates appear in Table D-12.

     8.  Using the data set out in Tables D-7, D-8, and D-9, and the
computational method described in Paragraph 4 above, the average in-use
system recovery efficiencies for nozzle-fill pipe losses for Stage II
systems were calculated.  The results are set out in Table D-ll.  Also
set out in that table is the weighted average system recovery efficiency,
calculated according to the method set out in Paragraph 5 above.
 The throughput distribution set forth is based on estimates of the
 distribution of systems currently being installed in the San Diego and
 South Coast areas.
                                  D-10

-------
                                         TABLE D-7

                              Percentages of In-Field Systems
                              Exhibiting Various Failure Modes
                        (Steady-state,; Bi-annual  In-Use  Inspections)
Mis installation
Improper Maintenance
Nozzles
Hoses
Processing Unit
Stage II Systems
Balance Hybrid
Dir.* N.O.V. Dir. N.O.V.
1.5% to 9% .8% to 4.5%
27% 29% 5% 6%
3% 3% 2% . 3%
Vac.
Dir.
0
3%
1%
2%
Asst.
N.O.V.
to 1 .2%
. ' 4%
1%
. .2%.
    Aspirator Unit

Tamperi ng

    Nozzles
    (user-
    convenience,
    economics or
    fuel-switch ing
    motivated)

    Nozzles,
    Hoses
    (alleviate
    system
    problems)

    Processing
    Unit
           13%
                          30%
                6%
                           30%
            10%
negl
negl.
negl.
negl.
          6%
        10%
negl.   negl.
                                                        10%
*"Dir." signifies direct enforcement;  "N.O.V."  signifies  notice-of-violation
 enforcement.  For explanation of these terms,  see Part II  of  this Appendix.
                                           D-n

-------
                                        TABLE D-8

                             Percentages of In-Field Systems
                             Exhibiting Various Failure Modes
                         (Steady-state; Annual In-Use Inspections)
Stage
Balance
Dir.* N.O.V.*
Misinstallation ' .5% to 3%
Improper Maintenance
Nozzles 15% 21%
Hoses 2% 2%
Processing Unit
Aspirator Unit
Tampering
II Systems
Hybrid Vac.
Dir. N.O.V. Dir.
.3% to 1.5% negl.
3% 4% 2%
1% ' 2% 1%
2%-
30% 30%

Asst.
N.O.V.
negl
3%
1%
2% .


    Nozzles           2%
    (user-convenience,
    economics or
    fuel-switching
    motivated)

    Nozzles, Hoses    negl.
    (alleviate sys-
    tem problems)

    Processing Unit
5%
negl.
2%
         2%
4%
negl
negl.     negl.     negl
                                   1%
                               5%
*See footnote, Table D-7.
                                         D-12

-------
                                          TABLE 0-9
                               Percentages of In-Field Systems
                               Exhibiting Various Failure Modes
                         (Steady-state;  Quarterly In-Use Inspections)
Stage
Balance
Dir.* N.O.V.*
Misinstallation negl . neql .
Improper Maintenance
. Nozzles 4% 8%
• Hoses .5% 1% •
Processing Unit
II Systems
Hybrid Vac.
Dir. N.O.V. Dir.
negl. negl. negl.
1% . 2% negl.
negl. .5% negl.
2%
Asst.
N.O.V.
negl.
1%
negl .
2%
    Aspirator Unit

Tampering

    Nozzles
    (user-
    convenience,
    economics or
    fuel-switch ing
    motivated)

    Nozzles,
    Hoses
    (a! leviate system
    problems)

    Processing Unit
negl
negl
  5%
                            30%
negl .
                            30%
negl.
                                    negl.     negl
negl.
negl.
negl.
negl.     negl
                                               negl.      2%
*See footnote, Table D-7.
                                          D-13

-------
                                         TABLE D-10

                          Average In-Use System Recovery  Efficiencies
                          (Nozzle-Fillpipe Losses) as  a Function  of
                          Method of Enforcement and  Frequency  of
                          In-Use Inspections

Freq. of
Inspections
Quarterly
Annual ly
Bi -Annually
Balance Hybrid
N.O.V.* Dir. N.6.V. ' Dir.
95% 95%
86% 90% 88% 91%
77% 84% 82% 87%
Vac.
N.O.V.
-
84%
76%
Assist
Dir.
95%
90%
85%
*See footnote, Table D-7.
                                           D-14

-------
























i-H
,_^
1
CD

LU
I
CO
•a:







































(^ ^—
O) O
0 XJ
e c
O)  +J
LU O) O
S CU
>> Q-
i- 4*^ (/)
 • ~""*
0 1
§ sE »-H
•M LU
Wl M-
>)<4- O
00 O
O) C O
woe
=> -r- O
1 +J 3
e= o o-
i— i c eu
3 S-
 <
•^














l-H C
I — 1 1 I 1
• ?o co
o-o co r.
OO U_
LO

o
.£
0
•r"
<+_
LU

s_
0
o
O)
1
«^^
10
oo
0)

^
o> • cn
> <*-
eC C
_ LU
X? ^Q ^5
d> O) CO CM
!c la oo oo
CD -i->
•i- oo
 CO
IO >^
< • LO
> CT>

O O ^-  en
>> O 00 C\J
=c • oo oo




^
i_ &5 »5
•r- O «*
, |Z
o£ H ^ §
0 OJ r- C
• O) •*-> (8 C
CT Q. S_ = ef

-------
                               TABLE D-12
              Absolute Non-Compliance as Function of Time
  Time Elapsed from Final
Compliance Deadline (Years)
           2
           3
           4 and thereafter
Rate of Total  Non-Compliance
             30%
             15%
              5%
              0%
                                   D-16

-------
             II.   EXPERIMENTAL  DATA  AND  ANALYTICAL  ASSUMPTIONS
                         Tables D-l,  D-2,  and D-3
 MISINSTALLATION/TAMPERING:   NOZZLES/HOSES
 Stage II  --  The  nature  of this type  of  defect is improperly laid piping,
 resulting in lack  of  proper  drainback of  condensed vapors and, thereby,
 excessive back pressure  in the vapor return system.  Estimates received
 by  EPA of the proportion of  balance  systems which were misinstalled
 during the first round  of Stage II vapor  recovery in California range
 from  5 percent to  30  percent (Latter estimate is contained in ARB
 Staff Report Accompanying Proposed Revisions to-ARB Suggested Vapor
 Recovery  Rules (October  22,  1977, at p. 10)).  As a result of this
 uncertainty,  it was decided  to express the proportion of balance systems
 which  would  be misinstalled  in any future Stage II program as a range,
 using  the midpoint for calculation purposes, as expressing the most
 probable  estimate.                                            -
      It was  assumed that even  in a minimal-enforcement scenario, the
 major  oil companies who have been involved with Stage II in California
 would  learn  from experience  and be able to cut the number of misinstal-
 lations in half.    (It was not believed that a lower figure could be
 achieved  due  to difficulties in supervising the large number of
 contractors involved in an en masse installation  of Stage II  systems).
 It was assumed that independents and smaller concerns would experience
 the same installation problems  as were experienced in California.   As
 61 percent of stations fly the  brands of majors,  a  range of 3.3  percent
 to 20 percent1 represents the weighted average  of expectable  misinstalla-
tions for balance systems.
     The effect on system efficiency of a misinstallation varies with
 the extent to which the piping  has  been mislaid and the  type  of  system
 involved.  By requirement,  the  balance system  nozzles certified  by  the
ARB contain a device for shutting  off the flow  of fuel when the  back
 pressure in the vapor recovery  return system exceeds  a certain  limit.
Based on information from air pollution  control officials in  California
 hereafter, "California officials",  the point at which  balance  system
1
 E.g.,  3.3% = .61 x 2.5% + .39  x 5%.
                                  D-17

-------
back-pressure devices at the State and local  levels are shutting off
appears to be in the range of 6 inches of water pressure.   Based on
available information, EPA estimates that no more than 25  percent of
all misinstallations have the potential for producing this level  of
back pressure.  Accordingly, for the minimal-enforcement scenario,  the
range of estimates for severe misinstallations at balance  system sites
was determined to be .85 percent to 5 percent, and the range for non-
severe misinstallation was determined to be 2.5 percent to 15 percent.
     To the extent that the liquid trap in a mi sinstailed  system was so
large that it from time-to-time created a back pressure exceeding the
cutoff point, it is believed that a service station operator would, at
least in a minimal-enforcement scenario, solve the problem by either
replacing the vapor recovery nozzle with a conventional nozzle or
disconnecting the vapor return system (e.g.,  disconnecting the-vapor
return hose).  This would result in a 100 percent loss of  recovery
efficiency.  Because of the deliberate nature of this presumed "fix"
for this problem, severe misinstallations are listed in Tables D-l,
D-2, D-3, and analogous tables under the heading of "tampering."
     The effect on system efficiency caused by a misinstallation
depends on the amount of liquid present in the liquid trap at a given
time.  The impact of a non-severe misinstallation on the efficiency of
a balance system was estimated as follows:  First, it was  assumed that
the non-severe misinstallations -- i.e., those where the maximum liquid
column would be the equivalent of at most 6 inches of water pressure --
were distributed in such a way that the average maximum liquid trap
would be 3 inches.  Based on discussions with California technical
personnel, it was estimated that the loss in balance system efficiency
which could be expected from this size liquid trap would thus be in the
range of 10 percent.  The efficiency loss resulting from a balance
system with this average 3 inch liquid trap would thus range from zero
percent to 10 percent, based on the amount of liquid in the trap at any
given time.  No estimates of the frequency of a given size build-up of
liquid, nor of corresponding dispersion rates, were available; accordingly,
for purposes of estimation, the average value of 5 percent efficiency
loss was used.
                                  D-18

-------
               Some  hybrid  system manufacturers have adopted the practice of
          certifying installation contractors, with installation training a
          prerequisite  to certification.  Accordingly, it was assumed that the
          misinstallation rate  for the  hybrid system could be reduced by 50
          percent  over  the  balance system rate.  Hybrid systems, like balance
          systems, are  required to have backpressure shut-off devices.  Hence, a
          range of 1.25 percent to 7.5  percent for moderate-effect hybrid mi sin-
          stall ations and .4 percent to 2.5 percent for severe-effect hybrid
          mi sinstallations  was  estimated.  As in the case of balance systems, a
          deliberate "fix"  resulting in 100 percent loss in system efficiency was
          assumed  for severe mi sinstallations in the minimal-enforcement scenario.
          Likewise,  the average maximum liquid trap was estimated to be in the 3
          inch  water pressure range.  As hybrid systems continuously recirculate
          fuel  through  the  vapor return piping, it was estimated that the effect
          on  hybrid  efficiency  of this  size liquid trap would continually approxi-
          mate  the outer limit  of balance system losses — i.e.^ would approximate
          10  percent.
               EPA understands  that the Gulf Science and Technology Company --
          producer of the Gulf  Hasselman vacuum-assist system, has, like some
          manufacturers of  the  hybrid systems, adopted the practice of screening
          installation  contractors.  Accordingly, it is believed that the rate of
          misinstalled  systems  would be about the same for vacuum-assist systems
          as  for hybrid systems.  However, the Gulf Hasselman system vacuum pump
          has  a dead-head vacuum in excess of 20 inches of water column.2
          Accordingly,  the  percentage of vacuum-assisted systems whose recovery
          efficiencies  would be actually affected by the liquid traps associated
          with  mi sinstallations  should be small.  It was estimated that the
          percentage of systems  moderately affected woul d not exceed 1 percent
          and  that the  corresponding reduction in recovery efficiency would be
          nominal  -- also no more than  1 percent.  It was estimated that the
          percentage of systems  which would be severely affected by a misinstalla-
          tion  would be negligible, especially as vacuum-assist systems are not
          required to incorporate back-pressure shut-off devices.
_
          'Certification Test Report, Gulf-Hasselman VCP-2 Vapor Recovery System
           (ARB), at p.3.~	
                                           D-19

-------
IMPROPER MAINTENANCE
A.  Nozzles
Stage II - The nature of this defect is torn bellows or faceplate,
interfering with the nozzle's in-use recovery efficiency.   In the steady-
state condition, the probability that any individual nozzle will have such
a defect and, thus, the percentage of nozzles which can be expected to
be defective, can be expressed mathematically as follows:
                      b
                                      [Formula D-l]
P * 1 -
          1-a+b
where
     "P" is the probability that any nozzle will be defective at any
point in time.
     "b" is the probability that, in a given time period, a defective
nozzle will be repaired or replaced with a non-defective nozzle.
     "1-a" is the probability that a non-defective nozzle will become
defective within the given time period — i.e., 1-a represents the rate
of deterioration.
     "b," the probability that a defective nozzle will be repaired or
replaced can be re-expressed as:
          b = bl + b2, where
          bl » the probability that a nozzle will be repaired or
          replaced as the result of program enforcement, and
          b2 = the probability that a nozzle will be repaired or
          replaced for some other reason.
One "other reason" why a defective Stage II vapor recovery nozzle would
be replaced is the failure of some, aspect of the nozzle's fuel-dispensing
apparatus.  The fuel-dispensing portion of vapor recovery nozzles is no
different than that of conventional nozzles, the fuel-dispensing
apparatus of which eventually wears out or malfunctions.  Based on
information obtained from nozzle manufacturers, EPA has concluded
that on  average, a nozzle "wears out" or malfunctions at an age of
about  2  years.  Thus, in the steady-state, b2 is equal to .0416.  This
rate is  assumed independent of the level of enforcement effort.
                                  D-20

-------
      The value of bl, of course, depends upon the level of in-use
 enforcement effort.   In a no-enforcement scenario, bl is equal to zero.
 If in-use inspections were performed biannually, bl would be equal to .0416,
      "1-a," the remaining factor in the equation, should, like "b2,"
 be independent of the level of enforcement effort.  A representative
 value of this variable was determined using observational  data obtained
 in the South Coast area in October of 1979 and in the District of Columbia
 area in October of 1978.  In the South Coast, 150 certified balance
 system nozzles were examined.   These nozzles had been installed for
 varying lengths of time, ranging from 1 to 8 months.   The  percentages
 of nozzles with faceplate or bellows defects as  a function of length of
 time in-use were plotted as set forth in Figure  D-l  following.   It was
 assumed that a linear fit was  appropriate for this curve on the hypothesis
 that development of faceplate  and bellows defects are,  at  least during
 the periods  of time involved,  a random function  of nozzle  usage.   The
 slope of the  best  linear fit for the three  data  points yields a 4.2  percent
 probability  that any  one nozzle will  become defective  in a  given  month.
      In  early  October 1978,  EPA personnel  inspected 106  service
 stations  in  the  District of  Columbia.   The  rates  and  sizes  of nozzle
 defects  were recorded.   The  length  of  time  the vapor  recovery nozzles
 had been  in service at each  station  was  not recorded.  However, from
 information obtained  from District  of Columbia Department of  Environmental
 Services officials, EPA was able  to  estimate that the systems had
 been  in-use between 20 and 25 months, and that, on average, the systems
 had been in-use for 23 months.
      Inspections were made on 198 first-generation, Emco-Wheaton Model 3003
 nozzles.  So far as bellows and faceplates construction  is concerned,
 these nozzles are similar to the Emco-Wheaton nozzle presently certified
 by ARB for use with balance systems, the exception being that the ARB-
 certified nozzle has a 20 percent greater bellows area on its unleaded
 version.  Of the 198 nozzles examined, four had catastrophic defects,
 totally eliminating the vapor recovery function.   Seventy-three had
 defects .which,  while not catastrophic, were deemed sufficiently substan-
 tial  to affect vapor recovery efficiency.  Twenty-two  had defective
bellows,  forty seven  had faceplate  defects, and  four  had defects in
                                  D-2.1

-------
                                FIGURE D-l
o
CD
O

-------
 both areas.  Thus, at first glance, the approximate figure for the rate
 of defective nozzles would appear to be 33 percent (73 of 218).2a
      This/figure is lower than the true rate, however.  First, it does
 not take  into account the greater bellows area of the unleaded version
 of the ARB-certifled nozzle, which is the nozzle that will actually be
 used in Stage II programs.  In addition, the District currently allows
 the use of one conventional nozzle per product per service mode
 {attendant/self-serve) at each station, and the 33 percent figure does
 not take into account the fact that the vapor recovery nozzles inspected
 were presumably underutilized due to the availability of the conventional
 nozzles at the District's service stations.  These two factors were
 compensated for as follows:  First, to account for defects which would
 have been found had the conventional nozzles been Stage II nozzles, the
 number of nozzles with bellows defects was increased by 20 percent —
 from 22 to 27.   Secondly, it was estimated that at least one vehicle in
 fiye had deliberately chosen to use a conventional  nozzle in preference
 toii;vapor recovery nozzle^  This results in the need to multiply the
factual^throughput per nozzle; by a factor of 1.25 to equal  the throughput
 per'.nozzle vyhich woiild have p^         the absence ;pf the legal
: conyentional  nozzTes.  |y the same token, multiplying the total  number
 ^ de^ec^ive noz2les by this same factor gives the number of nozzles
 which  would have been found defective if al 1  nozzles had been used
 equally.   The combined effect of these two compensations was to  increase
 the number of defective nozzles from, 73 to 98 [(73 +5) x 1.25].
 Thus,  the corrected rate of defective nozzles was determined to  be 45
 percent (98/218).
     The rate of nozzle deterioration evidenced in  the District  was
 determined from this one data point (45 percent of  nozzTes defective,
 at the 23-month point)  by first developing a  model  for describing  the '",.'-':
 rate of nozzle  defects  as a function of time  in a Stage II program.    y
 The model  developed is  depicted in Figure D-2 above.   As can be  seen,
 the model  has the  number of defective nozzles rising  more  or less
 linearly  from zero  to a certain maximum point,  then falling to an
 2a         --       .      •••..-.-  •     . '  .  •••  ..  ••   -.   ..•  .   ,  .   :  ••••
  The total  of 218 includes  20 conventional  nozzles being  used illegally.
   (See discussion under Nozzles:   Tampering,  below at  p. D-32).
                                   D-2 3

-------
               n 9 J 6flii'x
                                 	Srt,S,,,,t:S,2Gl	LE	£53	U	1,6	;	*	§M[	:	:	?.SO!f6	1^	
                                  sd :oJ teaqqs b fuow  z
 intermediate value*tnetbd steady38*atestt SaM^aM-Snsrsbowgtfoarrflns  is
 the"way the!cufye wtiuld;  intact? njettavesio-'fa,  eaiSe-iwbepe&tbentiozz^e ion
 deteriorationffateTwas ab*ou.tr:eqdal ctoritRewsteadysstatetreplaGSfflfintn^  "io

 rate to be:initheif4apeGqent»rangeysandisEI?.AIs6*nformafta'on3sbows ibater!*
 nozz3e;rep1acementi'Rat§ iSJaftss app^^^Jraatieiy  ^Bp^Kcentv^fssNi'nfibnsd'i'fi)
'•\s. p&Qlfigure;D^2  stiowscthe nQit)bersdf defiecti«§  ndzzilesopeaching a;ipeakon
.:-at about the ;20-month:pdjntf.and5Sbe£^steaaylsstatejyatoeybeJflga;Keaehedla*
 about 32 months1.  :This?i'sTthe?bena»torftwfitGhc'Oc6ars't,if rtheriwotn-oatsion
 points of the  fuel^'dispemsn-ngDapparatus^ol  a;groupc'of£new5aokz3.esnat?eo3
 distributed norma]lySaboot theiaverageonbzzTecli^
      The; rate  of nozzleadeteriorationvdemoitstfated?6y Stagsosl I'M
 in the District  wasi'determfhedsfromathesfenown  45recent5"23-m©nifi %
 point "in a two-step pr6e§§IC3vFTrst,®theosteadyrf§
 defective^noizles wa§ determinecl^by•-fitting ascurvi patterned"*'$foi
 model'set but  fn:Figare?S-2 %Q thi§ik^ownvdatasp51
 steady-state^level bf^tntt flew ci5rv§iv6Tfiiqsteady^s*a'tefi^ye1&thusr;
 determined Was:40 perellntl ^^Usiftl-tfiifj^valSei  aftd
 formula; Set out;at-pag^D2f giWdvev-slh^ietsHora^
 tn h(=f-?'7R'- -s<: follbfti5'^. f fs "t! svf.tssleb bnuol' nssd avert b Fuow  riarriw
 UU DG  t«/Oj do lUI lUWa-.•-"'- *  "
                           if«OD ow? sesrlt to  J:>9it1'9 bsnrdmoa 9rtT'  .xrffiijps
                                                             Io  isdroun  sriJ
         P
         .4 = 1-
     3t o3- t\ ,!
  asw asfsson svrJosl-sb  to s^s*1* bs^DsttoD sH*
  .0416
l-a+.0416                          -!81S\8e}
                       to
                                                       to 9te*t aril
                                   Jnfoq'sJsb  sno arr!,t. mo -ft
s'4? ATI sstimate
  andJ):istricfco
  of the- South Coast ?
  (4.2SrX
                      cte da-ta ^iCliheR
                                                          feerl3f^a»|)ewtfent
                                                6 o* oiss
              rtrso as is son . fBnorJfl9l/fift»...QA	:«;a0!.jf?,(H	,SIS .^p JaJpJ  a
                                                                      sS

-------
        The  South Coast  and District of Columbia nozzle surveys were taken
   at  times  when little  or no in-use inspections of stations were occurring 3
   From  the  average deterioration figure, then, the typical, steady-state
   proportion of nozzle  deficiencies in a minimal  enforcement scenario can
   be calculated using the probability formula.  Specifically,
            P = 1  -     -0416       =  45
                    >.U33b + .0416

  Accordingly,  in  a minimal  enforcement  scenario,  EPA  estimates that
  the proportion of defective  balance  system  nozzles would  be approximately  .
  45 percent.                                                             J
      The defects  observed  in  the A-3003 nozzles, .as  already  noted  were
  of three types - catastrophic defects, faceplate rips, and bellows rios
  Four of  the 78 (73 + the 5 added to compensate for underutilization)  '  '
  nozzles  exhibited defects so major that a 100 percent loss of recovery
  efficiency was assumed.  In the remaining 74 defective nozzles  there
  were 51  instances of faceplate defects deemed minor enough to have only
  a marginal effect (assumed to average in the range of 10 percent)  on
  vapor recovery efficiency.
      There were also 27 (22 + 5 compensating) instances  of bellows
 rips.  The average bellows  defect was  determined  to be a rip of  1.3  inches
 m length.   The average reduction in  system  efficiency resulting from
 such  a  defect  was  determined  by calculating  the ratio of emission flow
 through such a rip  to. the uncontrolled emissions  generated  during a
 refueling.  This calculation was made as follows:
EFFICIENCY
REDUCTION
(as a fraction)
                                 n    r
                                 ijo~  [Formula °-
where:  Q  = volumetric flow of vapor out of rip - (ft3/ sec)
        Qo - total  uncontrojled emissions.available at 6 gal/min
                  l .UU4 ft-Vsec)  dispensing rate.
It was assumed that Q/Qo would be  roughly the sa«  at othe, dispensing  rates
                                                                 «••
                                 D-25

-------
Restating,
       Er =       A x V	
              .0134 ft3/sec
where:
A s area of rip, and
V = velocity of emissions passing through the rip (ft/sec)
The standard rip was judged to be of rectangular shape with width equal
to 1/32 x L, where L is the length of the rip.  (The proportion of width
to length is based on the engineering judgment that a 2-inch rip would
have a 1/16-inch cross section).
The velocity of flow out of the rip is expressed by the formula for
flow from a sharp-edged orifice at low pressure drops, namely:
     V =   cd
where:
Cd  - coefficient of discharge of a sharp-edged orifice
    = density of gasoline vapor
 p  = pressure drop across the orifice (i.e., rip)
Thus, restating:
    Er =   .0313 x L2 x C^     / 2 Ap
           .0134 ft3/sec
 L, as noted, was measured to be 1.3 inches.
 CH =  .60   Source:  Marks, Mechanical Engineers Handbook
      (5th  ed.), p. 239.
   *  .0982 Ib  /ft3.  The vapors were assumed to be a mixture of 40 percent
      air,  and  60 percent propane.  Sources of densities:  41 Federal Register
      48053; Marks, op. cit., p. 1909.
  p =  '.1 inch of water.  This is believed to represent a conservative
      estimate  of AD-  According to data in ARB Exec. Order No. G-70-17,
      certified balance systems operate roughly at  .3"Ap at 6 gallons per
      minute flow rates.  It is known that a rip would reduce  A_p_ and it is
                                   D-26

-------
     believed that a reduction to .1 inch water for a 1.3-inch rips if
     anything, overstates the case.
     Accordingly,
     ER =  (-0313) (1.3")2 (1. ft)2 (.60)  /   2 (.1" HoQ) (5.20 lbf/ft2)
                                          /	(1" H20)
             (.0134 ft3/sec) (12 1n)2     /    	:—^	
                                              (.0982 1bm) (1bf sec2)

                                                 (ft3)    (32.2 lbm ft)
             0.3
     Accordingly, the weighted average effect, on recovery efficiency
caused by defective nozzles was determined to be:
Weighted Average^
Efficiency Reduction =  (51 x 10%) + (27  x 30%)' + (4 x 100%) = 22%
Per Defective Nozzle                      ^8          ~~
     The expectable rate of defects in hybrid and vacuum-assist
nozzles (in a minimal enforcement scenario) was determined by examining
233 OPW-7V-A nozzles at District service stations.  The bellows design
of the 7V-A nozzle is similar to that of Red Jacket and Hasselman system
nozzles.  The differences are that the bellows on a Red Jacket nozzle
has a roughly 50 percent greater area than that of the OPW-7V-A and
the Red Jacket nozzle does not employ a faceplate; the Hasselman nozzle,
on the other hand, employs a different type of faceplate (concave metal
instead of flat rubber) than the 7V-A nozzle.
     The number of 7V-A nozzles with defective bellows was determined
to be 15.  As in the case of the Emco-Wheaton nozzles, this figure was
increased by 25 percent — to 19 — to compensate for the underutilization
attributable to the legal conventional nozzles in-use at D.C. service
stations.  To estimate the number of defective Red Jacket nozzles which
would have been observed at District stations, the number was increased
by 50 percent again -- to 29 — to compensate for the greater bellows
areas of the Red Jacket nozzle.  Accordingly, so far as the Red Jacket
system is concerned, the "measured" rate of nozzle defects at the
4
 Does not sum to 74 since 4 nozzles exhibited both faceplate and bellows
 defects.
                                  D-27

-------
23-month point was 12 percent .(29/244).4a  As was just done in the  case
of the balance system, the steady-state number of deficient nozzles can
be determined .using the model curve set forth in Figure D-2 and the
12 percent/23-month data point.  The extrapolated steady-state value
is 11 percent.  The corresponding deterioration rate (monthly basis),
calculated using Formula D-l, is .5 percent.
     In a similar manner, the proportion of Gulf-Hasselman system nozzles
which could be expected to be defective in the steady-state, in a no-
enforcement scenario, was determined to be 7 percent.  The corresponding
Gulf-Hasselman deterioration rate (monthly basis) was determined to be
.31 percent.
     The average size bellows rips in the OPW-7V-A nozzles was 1.6  inches
in length.  Using Formula D-2, the effect of this size defect on the >
recovery efficiency of an OPW-7V-A used as a balance system nozzle  would
calculate to 45 percent.  A hybrid nozzle operates at a slightly negative
pressure, however, and a vacuum-assist nozzle at a substantially negative
pressure.  Based on discussions with technical personnel in the San
Diego Air Pollution Control District (APCD) and the relative pressures
involved, it is EPA's judgment that the effect of a 1.6-inch rip on a
hybrid and a vacuum-assist nozzle would be in the range of 15 percent
and 5 percent, respectively.
B.  Hoses
Stage II - The nature of this defect is kinking or flattening of the
vapor return hose with a resultant increase of back pressure in the
vapor return line.  EPA's October 1978 survey at District of Columbia
service stations found the rate of such defects to be 18 percent
(77 hoses with defects, among 430 examined).  This compares with a
29 percent defect rate measured by the California Air Resources Board
in a 1977 survey of service stations in the San Francisco Bay area.5
4a
  The 244 figure includes 11 conventional nozzles being used illegally.
  See "Nozzles:  Tampering" discussion, below at p. D-32.
^Report:  Harmon Wong-Woo to William Lewis, Subject:  Field Survey of
 Bay Area Air Pollution Control District's Phase II Vapor Recovery
 Program; March 10, 1977.
                                     D-28

-------
      The causes of  these hose defects appear to be twofold.  First,
  vehicles can run over that portion of a hose overhanging a service
  station island.  Second, flattening could occur at the hose/return pipe
  interface when a hose was stretched to its full length at an angle to
  the return pipe riser (portion of return pipe aboveground).  At the
  time of each survey, the stations in both the District of Columbia and
  the Bay Area were using first generation Stage II technology.  Among
  other features of this technology is the use of the same length hoses
  (12 feet)  as are used with conventional  nozzles.  State-of-the-art
 balance and hybrid systems, however, are required to use either 8-foot-
 long vapor return hoses or standard length hoses attached to the dispenser
 in an overhead retractor arrangement.   Another feature which could be
 required is the use of swivel  connectors for attaching the vapor return
 hose to the riser as well  as  for attaching the return hose to the
 nozzle.  Requiring 8-foot-long return  hoses or an overhead retractor
 arrangement,  plus  requiring the  use of swivels,  should have  a significant
 impact  on  the rate of hose  defects.  EPA estimates that requiring
 such features would result  in  a  5 percent maximum rate of hose  defects
 in balance  and  hybrid systems  in  a no-enforcement  scenario.
      The impact of  a hose defect  on  system recovery efficiency  depends
 on the  degree of flattening or kinking.   Defects  in balance  and hybrid
 system  hoses  will  run the gamut from ones  causing  only a  slight impact
 on efficiency up to  defects of a  size  sufficient  to trigger  the balance
 and hybrid back pressure shut-off  mechanisms.   It  is  recognized that
 defects exceeding the upper bound  of this  range also  will occur,  but  it
 is assumed that hoses damaged to this extent will  be  repaired towing  to
 the effect of the damage on nozzle fuel-dispensing capability.   The 5 per-
 cent defect rate estimates refer only to defects within the specified range
     It is assumed that the average defect will be midway in the
 specified range — i.e., equivalent to the average misinstallation
 defect.   The impact of a given misinstallation defect, as already noted,
 however, depends upon the amount of fuel  caught in the liquid trap.  A
crushed  hose -- being a continual  phenomenon -- is analogous to a
misinstallation where the trap is full.  Accordingly,  the effect of a
crushed  hose is analogous to the  effect of misinstalled piping in a hybrid
                                  D-29

-------
system.  This is true for both balance and hybrid systems.  Accordingly,
the impact on balance and hybrid system efficiency of a hose defect is
estimated at 10 percent.
     Vacuum-assisted system nozzles do not employ back pressure shut-off
mechanisms.  Accordingly, there is no upper limit to the impact on
recovery efficiency which a hose defect might have.  Impacts will  range
from 100 percent, in the case of a totally collapsed hose, down to
negligible, in the case of a slight kink.  It is assumed that the
average effect would occur at the midpoint of the range -- i.e., will
equal 50 percent.
     Vacuum-assist systems use the same length hoses as conventional
nozzles.  Accordingly, the gross number of crushed hoses would be
greater with such systems than with balance and hybrid systems.  However,
because of the vacuum-assist system's negative operating pressure, it
is estimated that the proportion of hose defects capable of adversely
affecting vacuum-assist system performance will be low -2 percent at
most.
C.  Processing Unit
     The Hasselman vacuum-assist system processing unit consists of an
electrically powered blower, an electronically ignited incinerator, and
a control apparatus consisting of a number of solenoids and valves.
While the system is believed generally reliable, the electronic and
mechanical parts are, obviously, subject to wear and malfunction.
Accordingly, it was assumed that there would have to be some average
downtime associated with these units.  A nominal 2 percent rate was
assumed.   It was assumed that 50 percent of the downtime would be
downtime of the blower, with resulting 100 percent recovery efficiency
loss, and 50 percent would be downtime of the incinerator, with a
roughly 50 percent loss in recovery efficiency.  The weighted average
recovery loss is thus 75 percent.
D.   Hybrid Unit
     The nature of this defect is miscalibration (by human error or
natural drift of equipment settings).  In a study performed by the
South Coast Air Quality Management District, slightly over 30 percent
                                  D-30

-------
 of the units inspected were judged to be sufficiently out of calibration
 to have an impact on system recovery efficiency.  Fifty percent of the
 units were miscalibrated so as to increase the vacuum at the nozzle-
 fill pipe interface, and 50 percent miscalibrated so as to decrease the
 vacuum.  The engineer who conducted the South Coast test estimated that
 a hybrid unit with a severe miscalibration of the second type would
 collect about 75 percent of the vapors at the interface whereas a  unit
 with  a severe miscalibration of the first type would suffer some
 unspecifiable,  but relatively small,  incremental  vent losses.  On  the
 basis of this information, EPA estimates that the average efficiency
 loss  of a miscalibrated system of the second type would be 12 percent
 (the  midpoint between no loss and maximum loss)  and the average loss
 for a miscalibrated system of the first type would be perhaps 2.5  percent
 (midpoint between  zero  loss and assumed nominal  maximum loss  of
 5  percent).   The weighted average efficiency loss of miscalibrated
 systems in a  minimal  enforcement scenario calculates,  therefore, to  be
 about 7 percent  (.5  x 12%)  + (.5 x  2.5%).
                                TAMPERING
 A-  Nozzles (Economics,  Difficulties-of-Use,  or Fuel  Switching-Motivated)
      This  "deficiency"  consists  of  the  substitution  of  conventional
 nozzles  for required  vapor  recovery nozzles.   In  two  important ways,
 gasoline  dealers will find  vapor recovery nozzles  substantially less
 desirable than conventional  nozzles.  First,  vapor recovery nozzles are
 much more expensive than conventional nozzles. . A  new conventional  nozzle
 runs about $45 to $50;  a new balance system nozzle runs about $160.
 A rebuilt conventional  nozzle runs about $25  (net with  trade-in).  A
 rebuilt balance system  nozzle runs about $100  (net with trade-in).6
     Secondly, in a number of different ways, vapor recovery nozzles
 are less desirable to use than conventional nozzles.  All vapor recovery
 nozzles,for example, are to some extent heavier, bulkier and thus more
awkward to use with any vehicle than a conventional nozzle.  Balance
 Figures are all  1978 figures.  EPA estimates that a nozzle must be
 rebuilt (traded  in)  every 2 years on average.
                                  D-31

-------
systems suffer from the added handicap of requiring exertion of pressure
(estimate of 5 pounds required)  against the nozzle  bellows spring tension
in order to activate and maintain the flow of fuel  through the nozzle.
     In addition to the general  inconvenience problem,  certain features
of balance system nozzles make refueling either vehicles in general  or
some vehicles in particular more difficult.  On some vehicles (e.g.,
Econoline vans), it is difficult to depress the fill pipe lead restrictor
no matter how much pressure is exerted against the  bellows spring
tension.  For those systems which have been installed with large liquid
traps, the back pressure shut-off features of balance system nozzles
will render refuel ings difficult or impossible whenever liquid has
built up in the trap.7  On vehicles with certain fill pipe/fill tank
configurations, the flow of fuel can be frequently interrupted (much to
the user's annoyance) because a system pressure build-up (created in
part by the tight-seal feature) either activates the nozzle back
pressure shut-off mechanism or causes fuel to activate the automatic
shut-off mechanism.  Lastly, on vehicles with certain fillpipe
configurations  (particularly side-fill vehicles), the weight of the
balance  system  nozzle precludes latching the nozzle securely in the
fillpipe and  using  it in the so-called "hands-off" mode of operation.
      Given  the  sizable cost differences involved, and the substantial
differences in  ease of use, it  is not surprising that dealers would be
tempted  to  use  conventional nozzles whenever they could.  To the extent
enforcement was lax,  therefore, one would  expect to find a certain
number  of the dealers using conventional nozzles on one or more of
their dispensers.
      To  estimate the rate  at which this form of tampering might occur
in a  minimal  enforcement scenario, EPA personnel recorded occurrences
of the  phenomenon during EPA's  October 1978  survey  of District of
Columbia service stations.  At  the stations which  utilized  Emco-Wheaton
Type  A-3003 nozzles, there should  have been  a  total  of  218  vapor  recovery
nozzles in  place and functioning.  The  survey  showed, however,  that
three of the  required nozzles  had  no  bellows whatsoever,  and that,  in
  The same phenomenon will  occur to a lesser extent  with  hybrid  systems.
                                   D-32

-------
20 instances, a conventional  nozzle had been substituted for the required
vapor recovery nozzle.8
     At the stations utilizing OPW-7V-A nozzles, there should have been
a total of 244 vapor recovery nozzles in place and functioning,   the
survey showed, however, that 12 nozzles had no bellows and that  11
conventional nozzles were being used illegally.  Combining the figures
for the Emco-Wheaton A-3003 and OPW-7V-A stations, the overall observed
tampering rate was determined to be:                                  '
        Overall
        Tampering
        Rate
      (D.C. Survey)
 23 + 23   =  10%
218 + 244
     The number at interest is of course the steady-state tampering
rate.  To derive an estimate of this number from the 10 percent data-
point, it is necessary to make certain assumptions about what proportions
of that 10 percent are attributable to various motivations and how
much these proportions will change in the steady-state.  In general,  it
is assumed that most nozzle substitutions which would be attributable
to the inconvenience of the nozzles would occur soon after the nozzles
were put in place and that most of the nozzle substitution attributable
to economics would not occur until a nozzle had become defective due  to
some failure of its fuel-dispensing apparatus.  EPA estimates that
about three-fourths of the observed nozzle substitution is attributable
to inconvenience of use,  and the balance to economics.
     In the steady-state, given literally no enforcement whatsoever,  it
is likely that tampering  due to both rationales would increase very
substantially.  For the minimal enforcement case,  it was assumed that
the rate of nozzle substitution attributable to inconvenience of use
would not change materially in the steady-state.
8
 The District currently allows the use of one conventional  nozzle per
 service mode provided there is at least one vapor recovery nozzle per
 product per service mode (The two service modes referred to are attendant-
 serve and self-serve).  Any such "legal" conventional  nozzles were
 excluded from the count.
                                  D-33

-------
     So far as nozzle substitution attributable to economics is concerned,
a relevant consideration for the extrapola ion process is the length  of
use of the nozzles in the District prior to the EPA tampering observations.
As noted above in the section discussing "Improper Maintenance:  Nozzles,"
the nozzles at D.C. stations had been installed for only 23 months on
average and, further, were estimated to be 20 percent underutilized owing
to the presence of legal conventional nozzles.  Under these circumstances
it is believed conservative to estimate a fourfold increase in the rate
of nozzle substitution attributable to economics in the steady-state,
minimal enforcement scenario.  EPA thus estimates that the rate of
nozzle-switching in a minimal enforcement scenario would be 15 to 20 percent
at a minimum, with about 7 to 8 percent attributable to inconvenience
of use, and 8 to 12 percent based on economics. .
     This rate of nozzle-switching pertains to balance systems.  Hybrid
and vacuum-assist nozzles are much easier to use than balance-system
nozzles.  It is believed that the differences in ease of use would be
sufficient to materially reduce the incentive for nozzle substitution
attributable to convenience considerations with these types of systems
— to perhaps one-third the rate prevailing in the case of balance
systems.  On the other hand, it is believed that the incentives for
nozzle substitution attributable to economics would be about the same
for assisted systems as for balance systems.  Accordingly, for assisted
systems,  EPA estimates a steady-state nozzle substitution rate of
10 to  15  percent, with 2 to 3 percent attributable to inconvenience of
use and 8 to 12 percent attributable to economics.
B.  Processing Unit
     This "failure mode" consists of a user's deliberately turning off
the vacuum-assist  secondary unit with resultant 100 percent loss of
vapor  recovery efficiency.  The incentive for shutting off a vacuum-assist
 secondary unit would appear to be substantial.  To begin with, there is the
 fact that the unit consumes about $50 worth of electricity  (annually) at a
 typical station where  such a  system would be installed.9
  Because of the relatively  high  capital cost  involved,  it is assumed
  that vacuum-assist systems will  be  installed only at higher throughput
  stations.   The $50 figure  is  EPA's  estimate  for a nine-dispenser,
  60,000 gal. per month  station.
                                   D-34

-------
More significantly, typical annual maintenance costs of the unit are
estimated at $330 per year.10  Particularly at outlets where maintenance
costs run higher than average, this level of expense will  tend to induce
cost-saving system shutdowns.  Processing units were found turned off
at 10 percent of the vacuum-assist equipped stations surveyed in the
South Coast California area.  In a truly no-enforcement circumstance,
it is believed that in the long run the number of vacuum-assist processing
units which would typically be turned off would be quite high.  Given
the rate observed in the South Coast, and the economic incentives, it
is believed that an estimate of a 15 percent minimum rate  for this form
of tampering is reasonable for the minimal-enforcement case.
                                  Table D-5
     In order to determine the in-use efficiency of Stage  II programs,
it is necessary to estimate the proportion of regulated throughout
which would be covered by each type of Stage II system.  Officials in
the South Coast and San Diego areas — where systems are currently
being installed — appear to agree that balance systems are capturing
about 80 percent of the market (throughput basis), hybrid  systems
are capturing about 15 percent, and vacuum-assisted systems about
5 percent.  The balance systems, which require the lowest  capital
outlays, appear to be favored by the major oil companies for self-serve,
as well  as attendant-serve, applications.  The assisted systems appear
to be favored by some independents, particularly at high-throughput
self-service operations.  Because of their lower price, hybrid systems
appear to be capturing the lion's share of the assisted system market.
                               Table D-6
     The estimates of total noncompliance rates were obtained by using
rates of noncompliance experienced with Stage I vapor recovery regulations
10
  Source:   EPA Stage II  cost figures.
                   .   .            D-35

-------
as a baseline.  In the summer of 1978, sources in EPA's Region II office

estimated that the total  noncompliance rate at service stations was

12 percent, despite the fact that the Region had been active in enforcing

Stage I, and despite the fact that the regulation had then been in
effect for 2-1/2 years.  The EPA Division of Stationary Source Enforcement

indicated at the same time that the then rate of total noncompliance

with Stage I regulations at small bulk plants was quite large — probably

well in excess of 40 percent.  Given these two benchmarks, the 40 percent,

30 percent, and 20 percent estimates of total noncompliance appear

conservative, if anything:

     1.  Stage II regulations are substantially more onerous, both
         economically and as a burden on service station operations,
         than Stage I regulations (Stage I costs only about $900 per
         station and requires Little, if any, operating and maintenance
         costs; Stage II costs $7,000 at a typical station for.the
         cheapest system, and costs several hundred dollars a year
         to maintain).  The economic impact of Stage  I at bulk plants
         (10-to 11-thousand-dollar investment) and, accordingly, the
         rates of Stage II noncompliance, should be more akin to those
         for  Stage I at bulk plants.

     2.  The  Stage I compliance  figures reflect a situation where at least
         one  in ten stations per year were being checked for compliance,
         whereas Table D-6 assumes a voluntary compliance scenario.

     3.  At the time Stage I noncompliance estimates  were made,  there
         were not known to be any equipment-availability problems
         associated with  Stage  I implementation.  It  must be assumed
         that there will  be  some delays due  to equipment shortages  if
         Stage  II is implemented on  a large  scale.'n


                        Tables  D-7.  D-8, and  D-9

     Tables  D-7,  D-8,  and D-9,  set  forth the  rates of system defects which

 could  be expected to prevail in  each of  three scenarios where  enforcement
 is performed according to the  strategy  set  forth  in  Section  9.0  of  this
 11
   In 1976, for example, the Arthur Little Company estimated that an
   18 to 24-month lead time would be needed to produce  the requisite
   number of nozzles if Stage II were required in  only  11 AQCR's.
   Arthur D.  Little, Inc., Economic Impact of Stage II Vapor Recovery
   Regulations, November 1976, at p. 204.~
                                   D-36

-------
 document, and where in-use enforcement inspections occur on a biannual,
 annual, and quarterly basis, respectively.  The defect rates reflect
 EPA estimates of the impact on compliance of the enforcement strategy
 pursued at the three different levels of intensity.  A discussion of
 the assumptions and judgments underlying these estimates follows.

                        Misinstallation
Stage  II - The San Diego APCD has developed a strategy for insuring

that Stage II vapor recovery systems are installed properly.  The

mechanism generally consists of making installation of a certified and

properly functioning system a prerequisite for a permit to operate a

station where installation of a vapor recovery system is required.  The

vapor  recovery aspect of the permitting process consists of the following
steps:

     1.  The service station owner applies for a permit to construct a
         vapor recovery system at his location.  The application must
         set forth, among other things, an exact engineering specification
         of the system to be installed.

     2.  The application is reviewed and a determination made whether
         the proposed system is adequate.  Upon such determination being
         made, the permit to construct is issued.

     3.  Upon completion of construction, and prior to repaying the
         service station, the APCD is notified.  An on-site inspection
         to determine the adequacy of the installation is then made.
         In some cases, this inspection includes certain parameter
         tests — particularly pressure/flow and liquid-blockage tests.
         Alternatively, the installation contractor is required to
         certify that these tests have been performed.  If the instal-
         lation meets requirements,  the authority is given to repave the
         station and to resume business operations.

     4.  After a certain operational  period (somewhere between 30 and
         90 days), the station is reinspected to ensure that the.
         installed system is functioning properly.   Upon satisfactory
         completion of_this second inspection, a full  permit to operate
         is issued.
     The misinstallation deficiency rates set forth in Tables D-7,  D-8,
and D-9, were determined by assuming that a system modeled on the

San Diego mechanism would be used to monitor installation in the three

enforcement scenarios.  It was further assumed that the control  system

at any station subjected to such a procedure would be properly installed
                                  D-37

-------
and totally free of defects.  The only question to resolve, therefore,

was what proportion of stations could be subjected to the procedure in

each enforcement scenario.
     To estimate this latter variable, it was assumed that the resources

available (measured in person-years) to monitor installation would be

the same as the resources available in the steady-state to conduct in-

use inspections of currently installed systems.12  The amount of time per
station available for installation monitoring was assumed to be  .6 hours,

1.2 hours, and 4.8 hours for the biannual, annual, and quarterly

enforcement scenarios, respectively.
     To assess the time available per station to conduct installation-
monitoring, it was necessary to estimate the amount  of time required to

process a station through each stage of the procedure.  The various

times were estimated as follows:13
     1.  Review application for permit to construct         3.0  hours
         and prepare permit to construct

     2.  Conduct post-construction, pre-operational         1 hour,
         inspection                                           50 minutes^

     3.  Conduct follow-up  inspection after initial         1 hour,
         operational period                                   10 minutes^

     4.  Prepare permit to  operate                          1 hour,  30  minutes
         Total                                              7.5  hours
   In actual  practice,  in order to avoid the situation  where a dealer must
   re-install  vapor recovery upon  a misintallation's  being discovered after a
   station has been repaved, it may prove desirable  to  temporarily utilize
   additional  resources,  where necessary, to ensure  that all  installations can
   be monitored fully in  the first instance.

   All figures include  allotments  of time for discussion of the program
    • .I     t  _.__.	•	  _i_i__. _A._J.^«...*I«*UB MiiAA-ic^AM**  /% J«  "ivtsi^ifi/iitoi  c T* a T T ^ n
                                                                     in the
   San Diego area.

   Consists of 30 minutes to examine visually the installed equipment,
   30 minutes average per station (three nozzles randomly selected at every
   other station at 20 minutes per nozzle) to conduct liquid blockage tests,
   20 minutes to consult with the service station dealer and to fill  out an
   inspection form, and 30 minutes travel (Travel will have to be done on an
   individual station basis owing to the need to conduct the inspection
   before the station is repaved).
 15
   Consists of standard in-use inspection.
                                     D-38

-------
     It was assumed, in order to obtain optimum utilization of available
resources, that Stage II would be phased in equally over a 5-year period.16
Accordingly, the amount of time required per station for installation-
monitoring in each year of the 5 year-period would be 1.5 hours.
Comparing this 1.5-hour-per-station figure to the .6, 1.2, and 4.8 hours-
per-station time-available figures, the percentages of systems sure to
be properly installed (because fully installation-monitored) is estimated
to be 40 percent, 80 percent, and 100 percent, in the biannual, annual,
and quarterly enforcement scenarios, respectively.  Accordingly,  the
misinstallation figures in Tables D-7, D-8, and D-9, represent 40 percent,
80 percent and 100 percent reductions over the figures in Table D-3.]j
                            IMPROPER MAINTENANCE
Nozzles — Stage II - As set fort in the discussion of Tables D-l, D-2,
and D-3, under the subheading "IMPROPER MAINTENANCE:  Nozzles, Stage-II,"
the steady-state percentage of defective nozzles can be expressed as:
                 P = 1 -     bl + b2
                           1-a + bl + b2
where
     "P" is the probability that any nozzle will be defective at
     any point in time.
     "1-a" is the probability that a non-defective nozzle will
     become defective within the given time period — i.e.,
     1-a represents the rate of deterioration.
     "bl" = the probability that a nozzle will be repaired or
     replaced as the result of program enforcement, and
     "b2" = the probability that a nozzle will be repaired or
     replaced for some other reason.
As discussed earlier, b2 turns out to be the rate at which nozzles are
refurbished or replaced as the result of failures of the fuel-dispensing
16
  To the extent such a regime was not followed, supplementary installation-
  monitoring resources would need to be utilized in the years where
  installations would exceed a 5-year equal distribution.
  Should an area provide adequate personnel to fully monitor installations,
  the number of misinstallations should be negligible, regardless of the
  frequency of in-use inspections.
                                    D-39

-------
apparatus and is equal  to .0416.   On the assumption that every  nozzle
found to be defective during an enforcement inspection will  be  quickly
repaired or replaced, bl can be simply stated as .0416, .0832,  and
.33 for bi-annual, annual, and quarterly enforcement,  respectively.
     The only remaining factor needing to be re-examined in  an  in-use
enforcement context is 1-a -- the probability that a non-defective
nozzle will become defective in a given month.  To what extent  will  the
rate prevailing in a no-enforcement context be modified in an in-use
inspection context?  Given the financial disincentives to maintaining
nozzles properly, it is believed that little, if any,  deterrence would
occur with an M.O.V. type of enforcement mechanism.  With a  "direct"
enforcement type of mechanism, on the other hand, the possibility of
obtaining a certain amount of deterrence presents itself. Discussions
with a number of local  air pollution control officials, however,
support the conclusion that local officials may be reluctant to impose
a fine directly for each and every nozzle violation attributable solely
to a failure of maintenance.  Rather, it appears that fines  would be
directly imposed, mostly in cases where the service station  operator is
perceived to be a willful "repeat offender."  Since the capability to
detect willful offenders would be directly proportional to the  frequency
of inuse inspections, one could expect  the degree of deterrence to
vary with the frequency of inspections.  It is estimated that deterrence
rates of 10 percent, 33 percent, and 50 percent, could be achieved with
biannual, annual, and quarterly inspections, respectively.  Accordingly,
the factor 1-a is reduced by the appropriate proportion for these
scenarios compared to the value prevailing in the no-enforcement scenario.
Sample Calculation
     Hybrid System/Direct Enforcement/Annual  Inspections
             =    _bl + bZ
                      ^a + bl + b2
            P  =  1  -
P = 1 -
                            .0416 +  .0832
                      .0051  1.6/J +  .0416 + .0832
                             = 1 -  .97 =  .03 = 3%
                        .
                      .1282
                                  D-40

-------
 Hoses  --  It  seems  reasonable  to  assume that the impact of in-use
 enforcement  inspections upon  maintenance of hoses will be proportional
 to  the impact  of  such  inspections upon nozzle maintenance.  Accordingly,
 the "hose" figures  in  Tables  D-7, D-8, and D-9, are in the same ratio
 to  the corresponding figures  in  Table D-3 as the nozzle figures in
 Tables in D-7, D-8, and D-9,  are to the nozzle figures in Table D-3
 (The figures are  rounded to the  nearest half-percentage),
 Processing Unit - As this figure refers only to system "downtime," it
 would  be unaffected by in-use enforcement.
 Hybrid Unit  -- According to the  study performed by the engineering
 staff  of the South Coast Air Quality Management District, hybrid systems
 appear susceptible to  drifting substantially out of calibration even
 when recently  adjusted.  See Phase II Vapor Recovery Evaluation Program
 (SCAQMD), at pp. 28-29, 45-47.   Indeed, the magnitude of the drift is
 described as being comparable to that originally measured in the
 District's study.  _I_d_. ,-at 28.  There does not appear to be any reason
 to  believe that the problems causing the drift are not correctable.
 On  the  other hand, given the presently available data, it was decided
 to  treat this category of defect in terms of the worst-case analysis by
 assuming that the impact of in-use enforcement inspections upon the
 quality of hybrid calibrations would be relatively small  and nonquanti-
 fiable.  Accordingly, the "hybrid unit" figures in Tables D-7,  D-8, and
 D-9, are the same as the figure presented in Table D-3.
                               Tampering
 Nozzles, Stage II (User-convenience,  economics or fuel-switching
motivated):
 Stage  II -- As in the case of improperly maintained nozzles,  the dynamic
 rate of nozzle-substitution in the  steady-state context is expressed by
 the  formula  {see p. D-19  supra):

            P = 1  -     b
                     1-a + b

Where nozzle-substitutuion  is  concerned,  "b,"  the  probability  (in  a
given month)  that a substituted  conventional  nozzle will  be  replaced by
                                  D-41

-------
a vapor recovery nozzle is equal  to the probability in a given month
that a nozzle will  be inspected in-use — i.e.,  is  represented by the
inspection rate expressed on a monthly basis.
  The factor "1-a," the probability in a given month that a vapor
recovery nozzle will be replaced with a conventional nozzle,  is developed
as follows:  First, it is assumed that, upon the inception of any in-
use enforcement program, nozzle-substitution,  if it occurred  at all,
would occur when the fuel-dispensing portion of the nozzle apparatus
wore out — i.e., when the nozzle was otherwise being replaced.  With
that assumption, "1-a" can be expressed as .0416 x  S, where "S" is  the
percentage of service station operators inclined to indulge in the
substitution practice.  It is assumed that the steady-state substitution
rates estimated in the subsection "Improper Maintenance Nozzles" (see
discussion of Table D-3 supra) is the same as  the proportion  of dealers
having a proclivity to substitute nozzles in a minimal-enforcement
scenario.  EPA has made estimates of the impact which in-use  inspections
would have upon the percentage of operators inclined to engage in nozzle-
substitution in both "N.O.V." and "direct" type enforcement scenarios
and these are set forth as follows:

                               Table D-13
                  S, Percentages of Dealers Will ing to
                  Engage in Nozzle Substitutionas  a
                  Function of Enforcement Frequency
                  and type of Enforcement Mechanism
                                            Balance
Inspection
Frequency
Bi-annual

Annual

Quarterly
Motivation for
 Substitution
Inconvenience
Economics
Inconvenience
Economics
Inconvenience
Economics
N.O.V.    Dir.
  Assisted
N.O.V.    Dir.
4-6
8-12
2-4
5-10
0-2
1-2
1-3
3-5
0-2
2-3
negl .
negl .
2-3
8-12
1-3
5-10
0-1
1-2
0-2
3-5
0-2
2-3
negl
negl
                                  D-42

-------
 From these figures  for "S,"  the  factor  "1-a" was calculated.  From the
 determined values  for  "1-a"  and  "b," the various values of "P" reflected
 in Tables  D-7,  D-8,  and D-9,  were calculated.
 Sample Calculation  -
   Balance  System/Annual  Enforcement/N.O.V.-Type Enforcement Mechanism
   P  =  1  -
            1-a  +  b
= 1 _  	.0832
       (.0416)  x (.105)  + .0832
                     =  1  -
                                .0832
                            .00436 +  .0832

                    =  1  - '.95  =  .05  =  5%
                       =  1  -  -0832
                               .0876
 Nozzles,  Hoses  (Alleviate Misinstallation Problems):
      This form  of  tampering  is.assumed to become negligible in the
 steady-state  condition  under  any frequency of in-use inspections.  Once
 the deficiency  (misinstallation) is  remedied, the incentive for the
 nozzle-substitution would disappear.  Reinspection of stations where
 nozzles  had initially been substituted would assure correction of the
 deficiency.
 Processing Unit -- Tables D-7, D-8,  and D-9, set forth OMSNE's estimates
 of  the rates  of noncompliance to which the 15 percent figure set forth
 in  Tables D-3 would be  reduced as the result of in-use enforcement.
 Both  the  N.O.V.  and, particularly, the direct mechanism figures, show
 substantial reductions.  The  reductions should be especially significant
 with  a direct type of enforcement mechanism, as switching off the
 processing unit would undoubtedly be treated as a serious violation.
                      Tables D-10 and D-ll
      The  figures in Tables D-10 and D-ll are analogous to the figures in
 Tables D-4 and  D-5 and  have been calculated in the same manner.  In
 Table D-ll and  subsequent tables, the Stage II weighted average
 efficiencies  are broken down into Federal and State enforcement
 categories, rather than into N.O.V. and direct enforcement categories.
•Federal enforcement is  100 percent of the N.O.V. type, while state
 enforcement presupposes a 50 percent N.O.V. and 50 percent direct mix.
                                  D-43

-------
                               Table D-12
     Table D-6 sets forth estimates  of absolute  noncompliance  (failure
to install a system)  in a no-enforcement scenario,  as  a  function  of
time.  The noncompliance rates were  assumed to result  from  delays
attributable to lack  of station owner diligence  and/or to equipment  or
installation supply problems, as well  as from out-and-out lack of
cooperation.  The steady-state noncompliance rate in  the no-enforcement
scenario, and therefore the rate attributable to lack  of cooperation,
was estimated at 20 percent.
     As has been noted earlier, the first phase  of  enforcement would
consist of an information gathering  visit to all facilities, one  result
of which would be knowledge of when each facility could  be  expected  to
meet the various regulatory compliance dates.  It would  add little time
to the installation-monitoring effort described  above  ("Misinstallation")
to revisit, for enforcement purposes, the uncooperative  20  percent of
the subgoup expected to install Stage II during  each  year of  a five-
year phase-in program.  As a result, with any reasonable enforcement
effort, it is believed that the portion of total noncompliance attribu-
table to lack of cooperation could be eliminated.  That  would  leave  the
portion attributable to lack of due diligence and/or logistics problems
which, by implication, is estimated to be 20 percent,  10 percent, and
zero percent  in the first, second, and third years, respectively, following
the final compliance deadline.  These would be the respective yearly
rates --  if the lack of cooperation could be eliminated  immediately.
An assumption that a certain amount of foot-dragging, including legal
foot-dragging, would continue  to occur, even in the presence of an
enforcement effort is  reflected in the noncompliance rates set forth in
Table D-12.
                                   D-44

-------
              APPENDIX E
CUMULATIVE VALUES OF EMISSION REDUCTIONS
              (1986-2020)
                  E-l

-------

-------
                               Table E-i.  Cumulative  VOC  Emission  Reductions  (Mg/yr)
    Year
         1986
         1987
         1988
         1989
         1990
         1991
         1992
         1993
         1994
         1995
         13%
         1997
         1998
         1999
         sum
         2001
         £082
         2883
         2004
         2936
         2087
         2008
         2009
         2013
         2011
         2012
         2013
         2014
         2015
         201S
         2017
         2018
         2019
         2820

 Cum.  Total

NPV of Total
Bulk
Terminals
VOC
(Mg/yr)
0
17,222
50,272
65,088
63,906
62,471
61,627
60,867
59, 178
58,334
57,574
56,730
55,886
55,042
54,282
54,282
54,282
54,282
54,282
54,282
54,282
54,282
54,282
54,282
54,282
54,282
54,282
54,282
54,282
54,282
54,282
54,282
54,282
54,282
54,282
1,864,120
518,730
Bulk Plants
no ex.
VOC
(«g/yr)
0
26,754
78,097
101,114
99,278
97,048
95,737
94,556
91,933
90,622
89,442
88, 130
86,819
85,507
84,327
84,327
84,327
84,327
84,327
84,327
84,327
84,327
84,327
84,327
84,327
84,327
84,327
84,327
84,327
84,327
84,327
84,327
84,327
84,327
84,327
2,395,300
805,345
Bulk Plants
ex.
VOC
(Mg/yr)
0
23,998
70,052
90,597
89,050
87,051
85,874
84,816
82,463
81,286
80,228
79,051
77,875
75,699
75,640
75,540
75,640
75,648
75,648
. 75,640
75,640
75,640
75,640
75,640
75,640
75,640
75,540
75,640
75,640
75,640
75,640
75,640
75,540
75,640
75,640
2,597,579
722,831
Storage
Tanks
VOC
(Mg/yr)
0
6,381
19, 143
25,524
25,524
25,524
25,524
25,524
25,524
25,524
25,524
25,524
25,524
25,524
25,524
25,524
25,524
25,524
25,524
25,524
25,524
25,524
25,524
25,524
25,524
25,524
25,524
25,524
25,524
25,524
25,524
25,524
25,524
25,524
25,524
842,292
222,573
St 1-Nation
no ex.
VQC
(Mg/yr)
0
33,025
96,405
124,816
122,550
119,798.
118,179
116,722
113,484
111,865
110,408
108,789
107, 171
105,552
104,095
104,095
104,095
104,095
104,095
'104,095
104,095
104,095
104,095
104,095
104,095
104,095
104,095
104,095
104,095
104,095
104,095
104,095
104.095
104,095
104,095
3,574,752
994,749
St I-Nation
ex.
vac
(Mg/yr)
0
21,120
61,650
79,819
78,370
76,610
75,575
74,643
72,572
71,537
70,685
69,570
68.535
67,500
66,568
66,568
66,568
66,568
66,563
66,568
66.568
66,568
66,568
DO; 5ofl
66,568
66,568
66,568
66,568
66,568
66,568
66,568
66,568
66,568
66,568
66,568
2,286,031
636,136
                                                      E-3

-------
Table E-l.  Cuauiative VOC Emission Reductions (Mg/yr)


Year

1985
1987
1983
1989
1990
1991
1992
1993
1994
1995
19%
1997
1993
1999
2000
2001
2002
2003
2804
2005
2088
2007
2008
2009
2018
2011
2012
2013
2014
2015
2016
2017
2018
2019
2020
Cua. Total
PV of Total
St I-flll Nfl
no ex.
VOC
(Mg/yr)
2,819
5,491
5.343
5,188
5.894
4J979
4,912
4,852
4,717
4,550
4,589
4,522
4,455
4,387
4,327
4,327
4,327
4,327
4,327
4,327
4,327
4,327
4,327
4,327
4,327
4,327
4,327
4,327
4,327
4,327
4,327
4,327
4,327
4,327
4,327
156,860
49,814
St I-flll Nfl St
ex
VOC
(Hg/yr)
2,419
4,712
4.585
4,452
4,371
4,273
4,215
4,163
4,048
3,990
3,938
3,880
3,822
3,765
3,713
3,713
3,713
3,713
3,713
3,713
3,713
3,713
3,713
3,713
3,713
3,713
3,713
3,713
3,713
3,713
3,713
3,713
3.713
3,713
3,713
134,598
42,058
II-Nation St
no ex.
VOC
(Mg/yr)
0
72,613
211,370
281,673
284,232
277,849
274,095
270,715
263,206
259,451
256,072
252,317
248,562
244,808
241,428
241,428
241,428
241,428
241,428
. 241,428
241,428
241,428
241,428
241,428
241,428
241,428
241,428
241,428
241,428
241,428
241,428
241,428
241,428
241,428
241,428
8,266,963
2,287,545
II-Nation St
ex
VOC
(Mg/yr)
0
49,801
142,631
191,555
195, 100
190,719
188, 141
185,822
180,667
178,090
175,770
173, 193
170,616
168,039
165,719
165,719
165,719
165,719
165,719
165,719
165,719
165,719
165,719
165,719
165,719
165,719
165,719
165,719
165,719
165,719
165,719
165,719
165.719
165,719
165,719
5,669,446
1,566,057
II-flll Nfl St
no ex.
VOC
(Mg/yr)
45,391
90,975
93,825
93,636
91,936
89,872
88,657
87,564
85,135
' 83,921
82,828
81,613
80,399
79, 184
78,091
78,091
78,091
78,091
78,091
78,091
78,091
78.091
78,091
78,091
78,091
78,091
78,091
78,091
78,091
78,091
78,091
78,091
78,091
78,091
78,891
, 2,814,849
869,598
II-flll Nft
ex.
VOC
(Mg/yr)
31.029
62,895
66,065
66,547
65,338
63,871
63,008
62,231
60,505
59,642
58,865
58,902
57,139
56,275
55,499
55,499
55,499
55,499
55,499
55,499
55,499
55,499
55,499
55,499
55.499
55,499
55,499
55,499
55.499
55,499
55,499
55,499
55,499
55,499
55,499
1,9%, 880
614,676
                         E-4

-------
Table £-1.  Cumulative  VQC Emission Reductions (Mg/yr)


Year

1986
1987
1988
1989
1993
1991
1992
1993
1994
1995
1996
1997
1998
1999
2008
2081
2882
2003
2004
2005
• ' 2036
2087
2088
2009
2010
201 1-
2012
; 2813
2814
' ' 2015
281S
2017
2018
2019
2828
Cum. Total
NPV of Total
St II-Sel Nfl
no ex.
VOC
(Mg/yr)
15,046
32,151
33, 158
33,101
32,500
31,778
31,341
38,955
38,895
29,657
29,288
28,851
28,422
27,992
27,686
27,606
27,685
27,505
27,686
27,S06
27,586
27,686
27,586
27,686
27,686
27,686
27,686
27,685
27.686
27,606'
27,506
27,686
27,686
27,686
27,686
995, 875
387,411
St II-Sel Nfl
ex.
VGC
(Mg/yr)
11,225
22,752
23,899
24,873
23,636
23,105
22,793
22,512
21,887
21,575
21,294
213,982
20,678
28,357
28,876
28,076
28,876
20,876
20,876
28,876
28,876
20,876
28,876
20,076
20,876
20,076
28,875
20,875
28,875
28,076
20,876
28,875
20,876
20,376
28,075
722,363
222,356
St II-Nation
no ex.
Combo VOC
(Mg/yr)
0
72,513
188, 120
221,676
197,541
168,377
143,626
123,446
138,281
83,543
67,859
54,753
43,881
32,884
24,867
19,797
13,846
5,019
574
13
0
8
8
0
0
0
0
0
8
8
0
8
8 "
0
0 :
1,561,758
938,893
St I I -tot ion
ex
CoBibo VOC
(Mg/yr)
8
49,801
126,942
158,754
135,595
115,576
98,586
84,735
68,834
57,345
45,579
37,583
29,517
22,517
17,869
13,589
9,553
3,566
471
18
0
0
8
' • . 8
0
0
8
0
0
0
8
0
0
8
8
1,867,828
548,338
St II-flll Nfl
no ex.
Coabo VOC
(Mg/yr)
45,391
98,^975
83,585
73,692
63,895
54,462
46,456
39,929
32,436
27,022
21,949
17,718
13,909
10,611
8,843
3,695
613
155
5
0
0
0
0
8
0
0
0
0
0
8
8
0 '
0'
8
8
634,454
428.588
St II-flll NA
ex.
Combo VOC
(Hn/yr)
31,829
62,895
58,797
52,372
45,418
38,786
33,016
28,377
23,852
19,285
15,599
12,586
9,885
7,541
5,716
2,594
534
143
6
0
8
8
0
0
0
8
8
8
0
8
0
0
fl
0
8
447,555
381.349
                      E-5

-------
Table E-l.  Cuaulative VOC Emission Reductions  (Mg/yr)


Year

1965
1987
1388
1989
1338
1991
1992
1993
1934
1935
19%
1997
1998
1993
2000
2891
2022
2003
2004
2035
2006
2007
2808
2893
2810
2011
2812
2013
2814
2815
2816
2817
2818
2619
2820
Cum Total
NPV of Total
St II-Sel Nfl
no ex.
Co«bo VOC
(Mg/yr)
16,046
32,161
25,528
26,051
22,588
19,253
16,423
14,115
11,467
9,553
7,759
6,251
4,917
3,751
2,843
1,386
217
55
2
8
8
8
8
8
8
8
8
0
8
8
8
0
8
8
0
224,285
151,587
St II-Sel Nfl
ex.
Combo VOC
(Mg/yr)
11,225
22,752
21,270
18,345
16,427
14,002
11,343
18,255
8,333
6,347
5,543
4,553
3,576
2,728
2,068
375
133
52
2
0
0
0
0
0
8
0
0
0
8
0
0
8
8
0
0
161,385
103,012
                   Refueling   Refueling     Evao.        Evap.     Ref.+Evap   Ref.+Evap
                    Era.  Red."    Em.  Red.    EM.  Red.     Era. Red.    Em. Red.    Em. Red.
                    Cura.  VOC    Cua.  VOC    Cum.  VOC    Cua. VOC    CUB. VOC    CUR. VOC
                   (No Tamo.)   (W.  Tarap.)   (No Tamp.)   (W. Tamp.)  (No Tamo.)  (M. Taap.)
                    (Xg/yr)      (Mg/yr)      (Mg/yr)     (Mg/yr)     (Mg/yr)     (Mg/yr)
                       3
                       9
                     34495
                     S5385
                     92149
                     116538
                     138219
                     157854
                     173488
                     187533
                     199523
                     209940
                     £18353
                     £25555
                     £31174
                     235581
                     23838G
                     240332
                     242392
                     243134
                     243995
                     244395
                     244395
                     244395
                     £44395
                     £44395
                     £44395
                     244395
                     244395
                     £44395
                     244395
                     244335
                     244395
                     244395
                     244395
  8
33694
63303
88544
111380
131812
147839
161862
1742B1
184698
193112
199923
205933
210340
215148
218753
221959
223562
223952
2£4363
£24353
224363
224353
224363
224363
£24363
224353
224363
224353
224363
224353
224363
224363
224353
  0
  0
26918
51878
75018
96388
117788
136630
153538
168878
182918
193838
284230
211900
219180
219180
219180
213188
213180
219188
219188
219180
219188
219188
213188
219188
219180
219188
219180
219188
213188
219188
219180
219180
219180
  8
£6398
58440
72540
93210
112320
129610
144690
158340
178560
179538
188110
195268
200720
288720
200728
£08720
280728
208720
200728
£00720
200728
208728
288720
280728
200720
2887£0
£007£0
200728
200720
200728
200720
£00728
208720
  8
  8
61405
117175
167159
213568
£55999
293684
327818
356373
382433
403778  '
422583
437465
458354
454761
457556
459572
451572
452374
453175
463575
453575
463575
463575.
463575
463575
463575
463575
463575
453575
463575
463575
463575
463575
68384
113743
161084
284590
243332
£77449
386552
332621
355258
372642
388833
481193
411850
41586B
419473
422679
424282
424682
425883
425883
425883
425883
425883
425083
425083
425083
425083
425033
425083
425083
425883
425883
425883
                    5914319     6374748     5223238     5736120     13138843     12118868

                    1421343     1320882     1265531     1173718     2637844      2500528
                    E-6

-------
                             Table E-2.  Cuaulative Benzene Reductions (Ma/vr)
Year
     1988
     1987
     1988
     1989
     1993
     1991
     1992
     1993
     1994
     1995
     19%
     1997
     1998
     1999
     2683
     2804
     2885
     2087
     2009
     2018
     2011
     2012
     2013
     2014
     2015
     2016
     2017
     2018
     2019
     2020
Bulk
Terminals
Bz
<«g/yr)
8
103
332
391
383
375
370
365
355
358
345
340
335
330
326
326
326
326
326
326
326
326
326
326
326
326
326
326
326
326
326
326
325
325
326
11,185
3,112
Bulk Plants
no ex.
Bz
(Mg/yr)
8
161
459
607
596
582
574
567
552
544
537
529
521
513
506
506
5%
586
506
505
506
536
506
506
5S5
506
506
506
506
506
506
506
506
506
506
17,375
4,835
Bulk Plants
ex.
Bz
(Hg/yr)
0
144
420
544
534
522
515
589
495
488
481
474
467
46®
454
454
454
454
454
454
454
454
454
454
454
454
454
454
454
454
454
454
454
454
454
15,585
4,337
    38
   115
   153
   153
   153
   153
   153
   153
   153
   153
   153
   153
   153
   153-
   153
   153
   153
   153
   153
   153
   153
   153
   153
   153
   153
   153
   153
   153
   153
   153
   153
   153
   153
   153

5,354

1,335
St !-Nation
no ex.
Bz
(Mg/yr)
0
198
578
749
735
719
709
700
681
671
662
653
643
633
625
625
625.
625
525
625
625
525
625
625
625
625
625
625
625 ,
625
625
625
525
625
625
21,449
5,958
St I-Nation
ex.
Bz
(Mg/yr)
a
127
378
479
470
450
453
448
435
429
424
417
411
485
399
399
• 399
399
399
399
399
399
399
399
399
399
399
399
399
399
399
399
399
399
399
13,716
3,817
                                                  E-7

-------
                            Table E-2. Cumulative Benzene Reductions (Mg/yr)
Year




ises
19B7
1SS8
1989
1990
1991
1992
1993
1994
1995
19%
1997
1998
1999
2888
2901
2882
2003
2084
2005
2006
2097
2003
2009
2010
2011
2012
2013
2014
2015
2016
2017
2018
2019
2020


St I-flll Nfl
no ex.
Bz
(Mg/yr)
17
33
32
31
31
30
29
29
2B
28
28
27
27
26
26
26
26
26
26
26
26
26
26
26
26
26
26
26
26
26
26
26
26
26
26
941
294
St I-flll Nfl St II-Nation St II-Nation St II-flll Nfl St II-flll Nfl
ex
Bz
(Mi/yr)
15
28
28
27
26
26
25
25
24
24
24
23
23
23
22
22
22
22
22
22
22
22
22
22
22
22
22
22
22
22
22
22
22
22
22
808
252
no ex.
Bz
(Mg/yr)
0
477
1,388
1,850
1,867
1,825
1,800
1,778
1,728
1,704
1,682
1,657
1,632
1,608
1,585
1,585
1,585
1,585
1,585
. 1,585
1,585
1,585
1,585
1,585
1,585
1,585
1,585
1,585
1,585
1,585
1,585
1,585
1,585
1,585
1,585
54,288
15,022
ex
Bz
(Mg/yr)
0
322
937
1,258
1,281
1,252
1,235
1,220
1,186
1,169
1,154
1,137
1,120
1,103
1,088
1,088
1,088
1,068
1,088
1,088
1,088
1,088
1,088
1,088
1,088
1,088
1,088
1,388
1,088
1,088
1,088
1,088
1,088
1,088
1,088
37,230
10,284
no ex.
Bz
(Mg/yr)
£98
597
616
615
604
590
582
575
559
551
544
536
528
520
513
513
513
513
513
513
513
513
513
513
513
513
513
513
513
513
513
513
513
513
513
18,485
5,711
ex.
Bz
(Mg/yr)
204
413
434
437
429
419
414
409
397
392
387
381
375
370
364
364
364
364
364
364
364
364
364
364
364
364
364
364
364
354
364
364
3£4
364
364
13,113
4,036
                                                     E-3

-------
Table E-2.  Cumulative Benzene  Reductions (Mg/yr)


Year

1986
1987
1988
1989
1990
1991
1992
1993
1994
1995
1996
1997
1998
1999
2000
2001
2002
2003
2094
2005
2006
2007
2008
2009
2910
£011
2012
2013
2014
2015
2016
2017
2018
2019
2020


St Il-Sel NR
no ex.
Bz
(Hg/yr)
105
211
218
217
213
209
206
203
198
195
192
189
187
184
181
181
181
181
181
181
- 181
181
181
181
181
181
181
181
181
181
181
181
181
181
181
6,535
2,019
St Il-Sel Nfl
ex.
Bz
(Mg/yr)
74
149
157
158
155
152
150
148
144
142
140
. 138
136
134
132
132
132
132
132
132
132
132
132
132
132
132
132
. 132
132
132
132
132
132
132
132
4,744
1,460
St Il-ifation
no ex.
Combo Bz
(Mg/yr)
0
477
1,235
1,456
1,297
1,106
943
811
659
549
446
360
282
215
163
130
91
33
4
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
10,256
6,160
St II-Nation
ex
'Coabo Bz
(Mg/yr)
0
322
834
990
890
759
647
555
452
377
306
. 247
194
148
112
89
63
23
3
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
7,012
4,205
St 11-911 Nfl
no ex.
Combo Bz
(Mg/yr)
298
597
548
484
420
358
305
262
213
177
144
116
91
70
53
24
4
1
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
4,166
2,814
St II-flll Nfl
ex.
Combo Bz
(Mg/yr)
204
413
386
344
298
254
217
186
151
126

83
65
50
38
18
4
1
8
0
0
0

0
0
0
0
0
0
0
0
0
0
0
0
2,939
1,979
                   E-9

-------
                             Table E-2.  Cuaulative Benzene Reductions (Mg/yr)
Year




1986
1987
1988
1989
1999
1991
1992
1993
1994
1995
1995
1997
1998
1999
2389
2001
2002
2003
2004
2005
2086
2007
2803
2809
2818
2811
2012
2813
2814
2815
2816
2817
2818
2819
8828


St II-Sel Nfl
no ex.
Cofflbo Bz
(«g/yr)
105
211
194
171
148
126
108
93
75
63
51
41
32
25
19
9
1
0
0
0
8
0
0
0
9
0
0
0
8
0
8
0
0
0
9
1,473
995
St II-Sel Nfl
ex.
Conbo Bz
Wg/yr)
74
149
140
124
108
92
78
67
55
46
37
38
23
18
14
6
1
0
0
8
0
0
0
0
0
0
0
0
0
0
8
0
0
0
8
1,053
716
Refueling
Bz Em Red
   Cua.
(No Tamp.)
 (Mg/yr)

    0
    0
   228
   431
   608
   769
   912
   1937
   1145
   1238
   1317
   1388
   1441
   1489
   152S
   1555
   1573
   1587
   1608
   1685
   1619
   1613
   1613
   1613
   1613
   1613
   1613
   1613
   1613
   1613
   1513
   1513
   1613
   1613
   1513

   45638

   9381
Refueling
Bz En Red
   Cua.
(U. Tamp.)
 (Mg/yr)

    Hi
    %
   222
   418
   5(54
   735
   865
   976
   11968
   1150
   1219
   1275
   1319
   1359
   1388
   1420
   1444
   1465
   1475
   1478
   1481
   1481
   1481
   1481
   1481
   1481
   1481
   1481
   1481
   1481
   1481
   1481
-  1481
   1431
   1481

   42873

   8717
  Evao.        tvap.
Bz EM Red   Bz En Red
   Cua.         CUB.
(No TaraD.)  (W. Tamp.)
 (Mg/yr)      (Mg/yr)
                                                                           0
                                                                          178
                                                                          342
                                                                          495
                                                                        .  549
                                                                          777
                                                                          902
                                                                          1013
                                                                          1115
                                                                          1207
                                                                          1279
                                                                          1348
                                                                          1399
                                                                          1447
                                                                          1447
                                                                          1447
                                                                          1447
                                                                          1447
                                                                          1447
                                                                          1447
                                                                          1447
                                                                          1447
                                                                          1447
                                                                          1447
                                                                          1447
                                                                          1447
                                                                          1447
                                                                          1447
                                                                          1447
                                                                          1447
                                                                          1447
                                                                          1447
                                                                          1447
                                                                          1447

                                                                         41873

                                                                          8359
                9
                0
                174
                333
                479
                515
                741
                855
                955
                1045
                1126
                1185
                1242
                1289
                1325
                1325
                1325
                1325
                1325
                1325
                1325
                1325
                1325
                1325
                1325
                1325
                1325
                1325
                1325
                1325
                1325
                1325
                1325
                1325
                1325

               37358

                7785
 Onboard
Ref.+Evao.
Bz EM Red
(No Tatap.)
 (Mg/yr)
    0
   405
   773
   1103
   1410
   1630
   1938
   2158
   2352
   2524
   2655
   2789
   2887
   2972
   3001
   3020
   3033
   3046
   3052
   3057
   3050
   3050
   3050
   3050
   3060
   3050
   3060
   3060
   3060
   3060
 Onboard
Ref. +Evap.
Bz Ea Red
(W. Tamp.)
 (Mg/yr)

    0
    0
   397
   751
   1052
   1350
   1686
   1831
   2023
   2195
   2345
   2459
   2561
   2548
   2713
   2745
   2769
   2790
   2803
   2803
   2S8&
   2806
   2806
   2805
   2886
   2806
   28%
   2806
    3050
    3060
    3060

   86711

   17740
   2806
   2886
   2306
   28%
   2806
   2386

  79932

  16584
                                                E-10

-------
                             Table E-3. Cuaulative EDB Emission  Reductions  (Mn/vr)
Year




1986
1987
1988
1989
1990
1991
1992
1993
1994
1995
19%
1997
1998
1999
2800
2001
2082
2003
2004
2005
2006
2007
2808
2009
2010
2011
2012
2013
2014
2015
2016
2017
2818
2019
2020


St I-Nation
no ex.
EDB
(Mg/yr)
8.880
8.401
1.868
1.234
1.865
8.942
0.799
0.555
3.543
8.437
0.361
0.298
8.217
0.146
0.072
8.072
0.072
0.872
0.87S
8.072
0.072
0.872
0.072
0.872
8.872
0.872
8.872
0.872
0.072
8.872
8.872
0.872
0.072
0.072
0.072
9.659
5.239
St I-Nation
ex.
EDB
(Mg/yr)
0.380
0.256
3. 678
8. 789
8.681
8.683
0.511
0.419
8.347
8.288
0.231
8.186
0.139
0.093
0.046
0.846
0.046
0.846
0.046
0.846
0.346
8.846
0.846
8.046
3.846
0.846
0.046
8.046
8.846
8.846
8.846
8.846
0.846
8.046
8.846
6.177
3.350
St I-flll Nfl
no ex.
EDB
(Mg/yr)
0.838
8.867
0.859
8.851
0.844
8.039
8.833
0.827
3.023
0.818
9.H15
0.812
3.009
8.806
8.003
0.803
8.803
0.883
0.803
0.803
8.803
8.803
0.883
0.803
0.803
8.003
8.883
0.083
0.803
0.883
0.003
0.803
0.383
8.803
3.383
8.584
8.313
St I-flll Nfl
ex
EDB
(Mg/yr)
8.033
8.857
3.350
8.844
8.336
0.834
3.329
8.023
0.819
8.016
0.813
8.018
0.808
0.885
8.003
8.883
0.803
0.033
0.803
8.803
0.803
0.803
0.803
0.003
0.003
0.803
0.803
0.803
0.003
8.803
0.303
8.803
8.803
8.883
3.383
8.433
8.269
St II-Nation
no ex.
EDB
(Mg/yr)
0.003
8.998
2.610
3.129
2.776
2.455
2.082
1.706
1.414
1.139
0.939
0.756
8.565
8.379
3.187
0.187
8.187
0.187
8.187
0.187
0.187
0. 187
8.187
8.187
8.187
0.187
8.187
8.187
8.187
0.187
3.187
8.187
8.187
8.187
3.187
24.872
13.488
St II-Nation
ex
EDB
(Wg/yr)
8.088
8.668
1.761
2.128
1.905
1.685
1.429
1.171
8.971
8.782
0.645
0.519
8.388
0.260
0.128
8.128
8.128
0.128
8.128
8.128
3.128
0.128
0.128
0.128
8.128
8.128
3.128
8.128
8.128
0.128
8.128
0.128
8.128
0.128
8.128
17.011
9.153
St II-flll Nfl
no ex.
EDB
(Mg/yr)
8.690
1.241
1.159
1.048
0.898
8.794
3.673
8.552
0.457
8.369
0.304
8.245
0.183
8.123
8.061
0.061
0.061
8.061
0.061
8.061
8.061
8.061
0.061
0.861
8.061
8.061
8.061
8.861
0.061
0.861
8.061
8.861
0.861
0.061
0.061
9.997
6.144

-------
                            Table E-3. Cuaulative EDB Emission Reductions  (Mg/yr)
Year
St II-flll Nfl St II-Sel Nfl St II-Sel Nfl St II-Nation St II-Nation St II-flll Nfl



1986
1987
1988
1989
1990
1991
1992
1993
1994
1995
199S
1997
1998
1999
2809
mi
2102
20S3
2004
£865
£006
£807
£008
2009
8010
£011
2012
£013
2014
£015
£016
2017
2018
£019
2020
ex.
EDB
(Mg/yr)
0.471
8.658
0.816
ft.739
0.638
0.564
0.479
0.392
0.325
0.262
0.216
0.174
0.130
0.087
0.043
8.043
8.043
8.043
0.043
0.043
8.043
0.043
0.043
0.043
8.043
0.043
0.043
8.043
8.043
8.043
0.043
0.843
0.043
0.043
0.043
no ex.
EDB
(Mg/yr)
0.244
0.439
0.410
0.368
0.317
0.281
0.238
0.195
0.162
0.130
8.107
0.886
0.865
0.043
0.821
0.821
0.021
8.821
0.821
8.821
8.021
0.821
0.021
8.021
8.021
0.821
8.821
3.321
0.821
0.021
0.021
0.821
0.021
0.821
0.821
ex.
EDB
(Mg/yr)
0.171
0.318
8.295
8.257
8.231
0.284
8.173
0.142
8.118
0.895
0.878
3.363
8.847
0.332
8.815
0.816
8.816
0.016
0.816
. 0.816
0.016
0.816
0.016
0.316
0.816
0.816
0.016
8.316
0.016
8.315
0.816
8.016
0.816
8.815
0.816
no ex.
Cosibo EDB
(Mg/yr)
8.888
8.990
2.323
2.463
1.929
1.488
1.891
8.778
8.539
8.357
8.249
8.164
8.898
3.351
8.019
3.815
8.011
8.884
.088
.888
0.808
8.088
0.888
8.388
8.888
8.388
3.888
8. '388
0.888
3.383
8.888
8.388
3.800
8.888
3.888
ex
Combo EDB
(Mg/yr)
3.888
8.653
1.558
1.675
1.324
1.821
3.749
8.534
8.370
' 3.252
8.171
0.113
8.367
3.335
3.313
0.811
8.387
3.333
.388
.388
8.338
8.880
8.838
3.388
3.083
3.388
8.823
3.383
0.038
3.388
3.833
3.333
8.388
3.333
8.333
no ex.
Combo EDB
(Mg/yr)
8.698
1.241
1.031
0.319
0.624
3.481
8.353
3.252
0.174
3.119
8.881
3.853
0.832
0.816
3.385
3,383
.383
.380
.838
3.388
3.338
3.833
3.838
3.388
3.333
3.030 •
£.338
3. 338
3.3S3
2. -333
?.v38
3.338
' 3.380
3. 333
8. 333
St II-flll Nft
ex.
Combo EDB
(Mg/yr)
8.471
8.858
8.725
8.532
3.443
3.342
3.251
0.179
3.124
3.884
3.857
3.838
8.32E
3.312
•3. 884
3.382
.m
.338
.388
3.388
0.383
3.308
2.380
3.388
J.388
3.388
3.32-3
3, 383
3.32-3
3.333
3.238
3. 338
'J.M0'
3. 333
3, eft
                   7.855




                   4.320
3.534




2.172
2.552




1.563
12.58




 8.55
8.58




5.S2
5,97




4.54
3.18
                                                 E-12

-------
                             Table E-3. Cumulative  EDB  Snission Reductions (Hg/yr)
           St II-Sel Nfl   St II-Sel Nfl
              no ex.           ex.
Year         Cosbo EDB      Combo EDB
              (Hg/yr)        (Mg/yr)
1986
1987
1988
1989
1990
1991
1992
1993
1994
1995
19%
1997
1998
1999
2933
2001
0.244
0. 439
0.365
0.289
0.221
0.170
0.125
0.089
0.062
0.042
0.028
0.919
0.011
0.006
0.082
0.001
0.171
8.31S
(3. £63
0.218
8.168
0.124
0.891
0.355
0.345
0.031
0.021
0.014
0.008
0.284
0.302
0.001
    2085
    2006
    2007
    2008
    2010
    2011
    2012
    2013
    2014
    2015
    2016
    2017
    2018
    20-19
    2020
0.000
0.000

8,00?!

0.900
8. £08
0.000
                                  1.52

                                  1.15
                                                   E-13

-------
                             Table E-4. Cuauiative EBC Emission Reductions (."ij/yr)
Year




1986
1987
1988
1989
1990
1991
1992
1993
1994
1995
19%
1997
1998
1999
2808
2001
2002
2003

2085
2066
2097
2008
2009
2010
2611
2012
2013
2014
2015
2316
2017
2018
2019
2020


Bulk
Terainals
EDC
(Mg/yr)
0.00
2.14
5.65
6.58
5.68
5.82
4.26
3.49
2.89
2.33
1.92
1.55
1.16
0.78
0.38
0.38
0.38
0.38
0.38
0.38
0.38
0.38
0.38
0.38
0.38
0.38
0.38
0.38
0.38
0.38
0.38
0.38
0.38
0.38
0.38
51.47
27.91
Bulk Plants
no ex.
EDC
(Mg/yr)
0.00
3.32
8.77
10.22
8.82
7.80
6.61
5.42
4.49
3.62
2.98
2.40
1.80
1.21
0.59
0.59
0.59
0.59
0.59
0.59
0.59
0.59
0.59
0.59
0.59
0.59
0.59
0.59
0.59
0.59
0.59
0.59
0.59
0.59
0.59
79.95
43.36
Bulk Plants
ex.
EDC
(Mg/yr)
0.00
2.38
7.87
9.16
7.91
7.00
5.93
4.86
4.03
3.25
2.68
2.15
1.61
1.08
0.53
0.53
0.53
0.53
0.53
0.53
0.53
0.53
0.53
0.53
0.53
0.53
0.53
0.53
0.53
0.53
0.53
0.53
0.53
0.53
0.53
71.72
38.90
Storace
Tanks
EDC
(Mg/yr)
3.20
8.79
£.15
2.58
2.27
2.05
1.76
1.46
1.25
1.82
0.85
0.70
0.53
0.36
0.18
0.18
0.18
0.18
0.18
0.18
0.18
0.18
0.18
0.18
0.18
0.18
0.18
0.18
0.18
0.18
0.18
0.18
0.18
0.1.8
0.18
21.55
11.31
St I -Mat ion
no ex.
EBC
•Mg/yr)
3.30
4.10
10.83
12.61
13.89
3.63
8.16
6.69
5.55
' 4.47
3.68
2.97
2.22
1.49
0.73
0.73
0.73
0.73
0.73
0.73
0.73
0.73
0.73
0.73
0.73
0.73
0.73
0.73
0.73
0.73
0.73
0.73
3.73
0.73
8.73
98.69
53.53
St I-Nation
ex.
EBC
(Mg/yr)
3.38
2.62
E.33
3.07
5.36
6.16
5.22
4.28
3.55
2.66
2.36
1.30
1.42
0.95
0.47
0.47
0.47
0.47
0.47
0.47
0.47
3.47
0.47
8.47
0.47
0.47
0.47
0.47
3.47
8.47
0.47
0.47
0.47
0.47
0.47
63.11
34.23
St I-fill Mfl
no ex.
EDC
(Mg/yr)
8.39
0,53
3.63
3.52
3.45
3. 40
3.34
2.28
3.23
0.19
a. is
0.12
0,89
0.06
8.33
0.33
OS
0.03
0.03
0.03
8.03
0.03
0.03
0.03
0.93
0.03
8.83
0.93
0.93
0.03
0.93
S8.33
3.33
0.93
2.03
5.15
3.28
                                                 E-14

-------
                             Table E-4.  Cuaulative EDC Emission Reductions (Mg/yr)
Year
     1986
     1987
     1588
     1989
     1998
     1991
     1992
     1993
     1994
     1995'
     19%
     1997
     1998
     1999
     2882
     2883
     21904
     2085
    ,2886
     2007
     2888
     2889
     2810
     2811
     2012
     2813
     2014
     2815
     2816
     2817
     2818
     2019
     2020
St I-flll Nfl
ex
EDC
(Sg/yr)
8.33
8.58
0.51
8.45
8.39
8.34
8.29
8.24
0.20
0.16
8.13
8.11
8.08
0.05
0.03
8.83
8.83
8.83
0.03
0.03
0.83
0.03
0.03
0.83
0.03
0.83
8.83
8.03
0.83
0.03
0.03
0.03
0.03
0.83
0.83
St II-Nation
no ex.
EDC
(«g/yr>
0.08
10.10
26.61
31.98
28.29
25.32
21.22
17.39
14.42
11.62
9.58
7.71
5.76
3.87
1.91
1.91
1.91
1.91
1.91
1.91
1.91
1.91
1.91
1.91
1.91
1.91
1.91
1.91
1.91
1.91
1.91
1.91
1.91
1.91
1.91
St II-Nation
ex
EDC
(Wg/yr)
8.88
6.81
17.95
21.59
19.42
17.18
14.57
11.94
9.98
7.97
6.57
5.29
3.95
2.66
1.31
1.31
1.31
1.31
1.31
1.31
1.31
1.31
1.31
1.31
1.31
1.31
1.31.
1.31
1.31
1.31
1.31
1.31
1.31
1.31
1.31
St II-flll Nfl
no ex.
EDC
(Mg/yr)
.7.03
12.65
11.81
18.68
9.15
8.89
6.86
5.63
4.66
3.76
3.10
2.49
1.86
1.25
0.62
0.62
0.62
0.62
9.62
8.62
0.62
0.62
0.62
0.62
0.62
0.62
0.62
8.62
8.62
0.62
8.62
0.62
8.62
0.62
0.62
St II-flll Nfi
ex.
EDC
(Hn/yr)
4.80
8.75
8.32
7.54
6.58
5.75
4.88
4.88
3.31
' 2.57
2.20
1.77
1.32
8.89
0.44
8.44
0.44
0.44
0.44
8.44
0.44'
0.44
0.44
0.44
8.44
0.44
8.44
0.44
0.44
8.44
3.44
8.44
8.44
3.44
3.44
St II-Sel Nfl
no ex.
EDC
!!>!g/yr)
2.48
4.47
4.18
3.75
3.24
2.86
2.43
1.99
1.65
1.33 .
1.09
8.88
8.66
3.44
8.22
8.22
8.22
0.22
8. £2
8.22
8.22
0.22
8.22
0.22
0.22
8.22
8.22
0.22
3.22
8.22
3.22
0.22
8.22
0.22
3.22
St II-Sei Nfl
ex.
EDC
(Mg/yr)
1.74
3.16
3.81
2.73
2.35
2. '28
1.76
1.45
1.20
3.97
a.ae
8.64
8.48
0,32
3.16
0.16
3.16
3.15
3. IS
3.15
3. 16
8.15
3.16
9.15
0.16
8.15
3.15
8.16
3.15
0.16
3.16
0.15
3.15
2.15
(2.15
                   4.42

                   2.75
253.53

136.68
173.40

 93.38
181.91

 52.63
71.91

44.83
36.83

22.14
26.01

15. S3
                                                E-15

-------
                            Table E-4.  Cuaulative EDC Emission Reductions (Mg/yr)
Year
St II-Nation St II-Nation St II-flll NR St Il-flll Nfl St II-Sel Nfl St II-Sel NR



1986
1987
19B8
1989
1990
1991
1932
1993
1994
1995
19%
1997
199B
1999
em
2881
2882
2883
2004
2885
2886
mi
2888
2009
2818
2811
2912
2013
2814
2815
2816
2817
2818
2819
2028


no ex.
Coabo EDC
(Mg/yr)
8.88
18.10
23.68
25.18
19.66
15.16
11.12
7.93
5.49
3.74
2.54
1.67
1.88
8.52
8.28
8.16
8.11
0.84
.88
.08
8.88
0.88
8.88
0.00
8.88
8.88
8.08
0.08
8.00
0.80
0.08
8.88
8.80
0.88
8.08
128.22
87.11
ex
Coabo EDC
(Mg/yr)
8.08
6.81
15.98
17.07
13.50
18.41
7.63
5.44
3.77
2.57
1.74
1.15
8.68
0.36
0.13
0.11
0.88
0.83
.88
.08
8.88
8.88
0.88
8.08
8.00
0.08
0.08
0.00
0.80
0.00
0.08
0.08
0.80
8.00
0.88
87.46
59.34
no ex.
Combo EDC
(Mg/yr)
7.83
12.65
10.51
8.34
6.36
4.91
3.69
2.57
1.78
1.21
8.82
8.54
0.32
0.17
0.86
0.83
.80
.80
.00
0.90
0.00
0.08
8.38
8.08
8.38
0.08
0.00
8.08
8.80
0.80
0.00
0.88
8.88
8.88
8.08
60.98
46.25
ex.
Combo EDC
(Mg/yr)
4.88
8.75
7.48
5.93
4.52
3.49
2.56
1.82
1.25
8.86
0.58
8.38
8.23
0.12
0.05
8.82
.88
.88
.08
8.88
. 8.08
' 0.88
0.00
0.88
8.88
0.88
8.80
8.%
8.08
0.88
8.00
0.88
8.00
8.88
0.00
42.78
32.40
no ex.
Combo EDC
(Mg/yr)
2.48
4.47
3.72
2.95
2.25
1.73
1.27
8.91
8.63
0.43
8.29
8.19
0.11
0.86
8.82
0.01
.08-
.08
.00
8.00
8.80
0.®
0.00
8.89
0.88
8.00
0.08
0.00
0.00
0.00
0.98
0.88
8.88
0.00
0.88
21.53
16.35
ex.
Combo EDC
(Mg/yr)
1.74
3.16
2.58
2.15
1.64
1.26
8.92
3.66
2.46
8.31
2.21
0.14
a. 0s
0.84
8.82
0. 01
.m
.00
.88
8.08
8.28
%.W
8.80
0.1213
*,m
8.*B
8.N
0.80
8.88
0.83
9. 83
0.80
3.89
8.08
8.120
15.47
11.72
                                                   E-16

-------
               Table E-5.  Total  Baseline  Emissions Given No Additional Controls
Irm — — T-r-Fij 	
REFUELING
<«g/yr)
1982 407,308
1983 388,952
1984 365,893
1985 353,683
1986 341,866
1987 332,112
1988 323,158
1989 313,797
1998 388,899
1991 381,188
1992 297,116
1993 293,447
1994 285,387
1995 281,237
19% 277,574
1997 273,584
1998 269,434
1999 265,364
2888 261,781
2001 261,781
2882 261,781
2083 261,781
2804 251,781
2805 261,701
2886 261,781
2887 261.781
2088 261,701
2809 261.701
2818 261,781
2011 261,781
2812 261,781
2013 261,701
2014 261,781
2815 261,781
2016 261,701
2017 261,781
2018 251,701
2019 251.701
2828 261,781
VOC -
EMISSIONS
STflGE I
(Mg/yr)
621,088
581.256
558,279
539,649
520,398
506,736
493,074
478,791
478,897
459,540
453,338
447,741
435,321
429,111
423,522
417,312
411,182
484,892
399,383
399,383
399,383
399,383
399,383
399,303
399,383
399,383
399,383
399.303
399,383
399.383
399,383
399,303
399,383
399.383
399,383
399,383
399,303
399,383
399,383

TOTflL
(Mg/yr!
1,823,888
962,288
924, 172
893,332
361,464
838,848
816,232
792,588
778, 195
768,728
758,448
741,188
720,528
710,348
781,896
690,316
688,536
678,256
661,884
661,084
661,884
661.004
661,884
661,884
661,884
661.884
661 j 804
661,884
661,884
661,884
661,884
561.884
661,884
661,004
661,804
661,884
661,884
651,884
661,884
———"-—
REFUELING
(Mg/yr)
2,686
2,514
2,415
2.334
2,251
2,192
2,133
2,071
2,833
1.988
1,961
1,937
1,883
1,856
1,832
1,885
1,778
1,751
1,727
1,727
1,727
1,727
1,727
1,727
1,727
1,727
1,727
1.727
1,727
1,727
1,727
1,727
1,727
1,727
1,727
1,727
1,727
1,727
1,727
BENZENE -
EMISSIONS
STflGE I
(Mg/yr)
3,726
3,488
3,358
3,233
3,122
3,840
2,958
2,873
2,821
2,757
2,720
2,686
2,512
2,575
2,541
2.584
2,467
2,429
2,396
2,3%
2,396
2,3%
2,396
2,3%
2,395
2,3%
2,396
2.3%
2,396
2,3%
2,396
2,3%
2,396
2.3%
2,396
2,3%
2,396
2.396
2,396

TOTflL
(Hg/yr)
6,412
6.882
5,765
5,572
5,373
5,232
5,391
4,944
4,854
4,745
4,681
4,623
4,495
4,431
4,373
4,389
4,245
4,181
4,123
4,123
4,123
4,123
4, 123
4.123
4,123
4.123
4,123
4.123
4,123
4.123
4,123
4,123
4,123
4,123
4,123
4.123
4,123
4,123
4, i£3
	
REFUELING
(Mg/yrO
18
9
7
6
5
5
4
4
3
3
2
3
£
1
1
1
1
3
8
8
8
8
8
3
3
0
3
0
3
3
8
3
0
3
0
•8
8
••*
V
8
EDB
EMISSIONS
STflGE I
(Mg/yr)
14
12
18
a
• • . 7
5
5
5
4
4
3
3
2
2
1
1
1
i
t.
0
0
8
8
.0
0
0
0
0
0
8
0
0
0
8
3
3
8
fl
a
8
	
TOTflL
(Mg/yy)
£4
23
17
14
12
11
9
a
^
i
6
5
4
4
3
2
2
1
i
8
e
3
3
8
8
8
3
8
$
0
3
0
3
0
8
3
3
0
US
0
11155638   17335514    28282152
73693
182219
                                                           175912
                                      59
                                       94
                                                                                                 153
                                       E-17

-------
       Taole £-5.  Total Baseline Emissions  Giver; No additional Controls
inn


1982
1983
1984
1985
1966
1987
1988
1989
1990
1991
1992
1993
1994
1995
19%
1997
1998"
1999
2000
2001
2082
2003
2084
2005
2006
2007
2008
2009
2010
2011
2012
2013
2814
2015
2016
2017
2018
2019
2020

REFUELING
(Hg/yr)
104
88
74
62
53
46
41
36
31
27
23
19
16
13
10
8
6
4
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
euu
EMISSIONS
STflSE I
(Mg/yr)
140
119
101
84
72
63
55
48
42
37
31
26
21
17
14
11
9
6
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3

TOTflL
(Mg/yr)
244
266
175
147
125
189
%
84
73
64
54
45
37
30
25
20
15
18
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
706
955
1651
                               E-18

-------
  MINLMflL
    VOC
   NESHflP
    EX.
  (Mg/yr)

  257,728

     9
   28,827
   84.047
  112,957
  115,806
  112,424
  110,904
  109,537
  106,499
  104,979
  103,612
  102,393
  188.574
  99,354
  97,687
  97,687
  97,687
  97,587
  97.687
  97,587
  97,687
  97,587
  97,687
  97,587
  97,687
  97,687
  97,687
  97,537
  97,687
  97,537
  97,687
  97,687
  97,687
  97,687
  97.687

3,341,952

 923.110
               FflBLE    E-6.
  MINIMflL
    VOC
flLL Nfl flREflS
   NO-EX.
  (Mg/yr)

  121,448

   26,730
   53,628
   55,286
   55,196
   54,194
   52,977
   52,251
   51,617
  •50,185
   49,469
   48,825
   48,109
   47,393
   46,677
  46,033
  46,633
  45,033
  46,033
  45,833
  45,033
  46,033
  46,033
  46,033
  46,033
  46,033
  46,033
  45,033
  45,033
  45,033
  46,033
  45,033
  45,033
  46,033
  45,033
  46,033

1,559,238

 512,560
:RVICE STATION IN--USE
MINIMflL
VOC
flLL Nfl flREflS
EX.
(«g/yr)
86,312
18,303
37,895
38,961
39,227
38,515
37,650
37, 141
36,684
35,666
35,157
34,699
34,199
33,682
33, 173
32,715
32,715
32,715
32,715
32,715
32,715
32,715
32,715
32,715
32,715
32,715
32,715
32,715
32,715
32,715
32,715
32,715
32,715
32,715
32,715
32,715
1, 177, 159
362,381
MINIMftL
VOC
SELECTED Nfl flREflS
EX.

-------
Bl-ftNN.
VOC
SLIM AREAS
NO-EX.
(Hg/yr)
39,140
78,526
88,954
38,323
79,355
77,573
76,525
75,582
73,485
72,437
71,493
78,445
69,397
68,348
67,485
67,405
67,495
67,485
67,485
67,485
67,485
67,485
67,485
67,485
67,485
67,485
67,485
67,485
67,485
67,485
67,485
67,485
67,485
67,485
67,485
2,429,587
750,535
BI-fflN.
VOC
fiLL Nfl fiREflS
EX.
(«g/yr)
26,881
54,318
57,858
57,448
55,397
55,131
54,386
53,715
52,225
51,488
58,818
58,865
49,320
48,575
47,984
47,984
47,984
47,904
47,984
47,904
47,904
47,904
47,984
47,904
47,984
47,904
47,984
47,984
47,904
47,984
47,984
47,904
47,984
47,904
47,904
1,723,697
530,629
BI-fWN.
VOC
SELECTED Nfl flREflS
EX.
(Mg/yr)
9,695
19,649
20,637
20,779
28,401
19,943
19,674
19,431
18,892
18,623
18,388
18,111
17,841
17,572
17,329
• 17,329
17,329
17,329
17,329
17,329
17,329
17,329
17,329
17,329
17,329
17,329
17,329
17,329
17,329
17,329
17,329
17,329
17,329
17,329
17,329
523,540
191,953
ANNUftL
VOC
NESHflP
NO-EX.
(Mg/yr)
9
65618
191346
254886'
257305
251527
248128
245869
238271
234872
231812
228413
225814
221615
218556
218556
218556
218556
218556
218556
2185%
218556
218556
218556
218555
218556
218555
218556
218556
218556
218555
218556
218555
218556
218555
.7483551
2078641
fiNNUfll
VOC
NESHflP
EX.
(Mg/yr)
0
44270
129073
' 173485
176617
172651
178318
168218
163551
161218
159119
156785
154452
152119
158019
158819
158819
150819
150019
158819
158819
158819
150019
150019
150019
158819
158819
158019
158819
158019
150019
150819
150019
158019
150019
5132283
1417633
ftNNUfiL
VOC
fill Nfl flREflS
NO-EX.
(Mg/yr)
42004
84272
36878
' 86737
85162
83249
82124
81112
78852
77737
76724
75599
74474
73349
72337
72337
72337
72337
72337
72337
72337
72337
72337
72337
72337
72337
72337
72337
72337
72337
72337
72337
72337
72337
72337
2607362
805452
E-20

-------
ANNUAL
voc
LL NA AREAS
EX.
(Mg/yr)
28762
58293
61224
61643
68524
59165
58365
57646
56846
55S47
54527
53728
J2928
£129
(1489
J1409
51409
51489
51489
51489
51409
51409
51409
51489
51409
51409
51409 '
51409
51409
51409
51409
51409
51409
51409
51409
1349821
569455
ANNUAL
VOC
SELECTED NA AREAS
EX.
(Mg/yr)
10405
21087
22147
22299
21894
21403
21113
20853
28275
19985
19725
19436
19147
18857
18597
18597
18597
18597
18597
18597
18597
18597
18597
18597
18597
18597
18597
18597
18597
18597
18597
18597
18597
18597
18597
669165
205998
QUflRTERLY
VOC
NESHflP
NQ-EX.
(Mg/yr)
0
68661
200246
266741
269273
263226
259669
256467
249353
245796
242594
239037
235480
231923
228722
228722
228722
228722
228722
£28733
228722
228722
228722
228722
228722
228722
228722
228722
228722
228722
228722
228722
228722
228722
228722
7831623
2166950
QUARTERLY
VOC
NESHAP
EX.
(Mg/yr)
0
46329
135076
181554
184832
180681
178239
176842
171159
168717
166519
164078
161636
159195
155997
156997
156997
156997
155997
156997
156997
156997
156997
156997
156997
155997
156997
155997
156997
156997
155997
155997
155997
156997
156997
5370994
1483570
QUARTERLY
VOC
ALL NA AREAS
NO-EX.
(Mg/yr)
43913
38102
90827
90579
89033
87033
85857
84799
82447
81278
80212
79036
77850
76584
75525
75625
75525
75525
75625
75625
75525
75625
75625
75625
75525
75625
75525
75625
75625
75625
75525
75525
75625
75525
75525
2725878
842863
                            QUARTERLY
                               VOC
                           ALL NA AREAS
                               EX.
                             (Mg/yr)
                              30059
                              60942
                              64007
                              64445
                              63275
                              61854
                              61018
                              60266
                              58594
                              57758
                              57805
                              56170
                              55334
                              54498
                              53746
                              53746
                              53746
                              53745
                              53746
                              53746
                              53746
                              53745
                              53745
                              53746
                              53745
                              53746
                              53746
                              53746
                              53746
                              53746
                              53746
                              53746
                              53746
                              53746
                              53745

                             1933904

                              595339
  QUARTERLY
     VOC
SEL. NA AREAS
     EX.
   (Mg/yr)
    10877
    22046
    23154
    23313
    22889
    22375
    22073
    21801
    211%
    20894
    20522
    20319
    20017
    19715
    19442
    19442
    19442
    19442
    19442
    19442
    19442
    19442
    19442
    19442
    19442
    19442
    19442
    19442
    19442
    19442
    19442
    19442
    19442
    19442
    19442

    699581

    215362
E-21

-------
KINIMftL
  3z.
 NESHftP
 NO-EX.
(Mg/yr)

375,472

   0
  281
  818
 1,1993
 1,1%
 1,076
 1,061
 1,048
 1,819
 1,284
  991
  977
  958
  948
  935
  935
  935
  935
  935
  935
  935
  935
  935
  935
  935
  935
  935
  935
  935
  935*
  935
  935
  935
  935
  935

 32, Ml

 8.855
MINIHflL
  Bz.
 NESHflP
  EX.
(Mg/yr)

257,728
  189
  552
  742
  755
  738
  728
  719
  699
  689
  680
  670
  660
  650
  642
  642
  642
  642
  642
  642
  642
  642
  642
  642
  642
  642
  642
  542
  642
  642
  642
  642
  642
  642
  642

 21,947

 6,062
  MINIMflL             MINIMflL             MINIMflL             BI-flNN.
    Bz.                  Bz.                  Bz.                  Bz.
flLL Nfl flREflS        flLL Nfl flREflS     SELECTED Nfl flREP.S         NESHflP
   NO-EX.               EX.                  EX.                 MO-EX.
  (Mg/yr)             (Mg/yr)             (Mg/yr)             (Mg/yr)

  121,448              86,312              31,223

    175                 120                  43                  0
    352                 244                  88                 391
    363                 256                  93                1,148
    362                 258                  93                1,518
    356                 253                  91                1,533
    348                 247                  89                1,498
    343                 244                  88                1,478
    339                 241                  87                1,460
    330                 234                  85                1,419
    325                 231                  84                1,399
    321                 228                  82                1,381
    316                 225                  81                1,360
    311                 221                  80                1,340
    307                 218                  79                1,320
    302                 215                  78                1,302
    302                 215                  78                1,302
    302                 215                  78                1,302
    302                 215                  78                1,302
    302                 215                  78                1,302 '
    302                 215                  78                1,302
    302                 215                  78                1,302
    302                 215                  78                1,302
    302                 215                  78                1,302
    302                 215                  78                1,302
    302                 215                  78                1,302
    302                 215                  78                1,302
    302                 215                  78                1,302
    302                 215                  78                1,302
    302                 215                  78                1,302
    302                 215                  78                1,302
    302                 215                  78                1,302
    302                 215                  78                1,302
    302                 215                  78                1,302
    302                 215                  78                1,302
    302                 215            .      78                1,302

   10,896              7,730               2,796               44,573

   3,366               2,380                861                12.333
                                                          E-22

-------
Bi-fim,
Bz.
NESHfiP
EX.
(Mg/yr)
8
264
769
1,033
1,852
1,823
1,814
1,002
974
953
348
934
920
985
894
894
894
894
394
894
894
894
894
894
894
894
894
894
894
894
894
894
894
894
894
38,568
8,444
BI-flNN.
Bz.
flLL N8 flREflS
NO-EX.
(Hg/yr)
257
516
532
531
521
589
583
4%
483
476
469
463
456
449
443
443
443
443
443
443
443
443
443
443
443
443
443
443
443
443
443
443
443
443
443
15,955
4,929
BI-flW.
Bz.
fill Nfl flREflS
EX.
(Mg/yr)
176
357
375
377
370
362
357
353
343
338
334
329
324
319
315
315
315
315
315
315
315
315
315
315
315
315
315
315
315
315
315
315
315
315
315
11,320
3,485
     64
    129
    136
    136,
    134
    131
    129
    128
    124
    122
    121
    119
    117
    115
    114
    114
    114
    114
    114
    114
    114
    114
    114
    114
    114
    114
    114
    114
    114
    114
    114
    114
    114
    114
    114

   4,095

   1,261
  431
  1257
  1674
  1690
  1652
  1629
  1689
  1565
  1542
  1522
  1500
  1478
  1455
  1435
  1435
  1435
  1435
  1435
  1435
  1435
  1435
  1435
  1435
  1435
  1435
  1435
  1435
  1435
  1435
  1435
  1435
  1435
  1435
  1435

49144

13598
   0
  291
  348
  1139
  1160
  1134
  1118
  1105
  1074
  1059
  1045
  1039
  1014
  999
  985
  985
  985
  985
  985
  985
  985
  985
  985
  985
  985
  985
  985
  985
  985
  985
  985
  985
  985
 985
 985

33704

 9310
E-23

-------
imxL
Bz.
ALLNRflRQS
HQ-EX.
(Mg/yr)
276
553
571
578
559
547
539
533
518
518
584
495
489
482
475
475
475
475
475
475
475
475
475
475
475
475
475
475
475
475
475
475
475
475
475
fi}«UflL
Bz.'
flLL Nft flREAS
EX.
(Mg/yr)
189
383
482
405 '
397
389
383
379
368
363
358
353
348
342
338
338
338
338
338
338
338
338
338
338
338
338
338
338
338
338
338
338
338
338
338
mm.
Bz.
SELECTED Nfi RREflS
EX.
(Hg/yr)
68
138
145
146
144
141
139
137
133
131
138
128
126
124
122
122
122
122
122
122
122
122
122
122
122
122
122
122
122
122
122
122
122
122
122
QUftRTERLY
Bz.
NESKflP
NO-EX.
(Mg/yr)
0
451
1315
' 1752
1768
1729
1785
1684
1638
1614
1593
1570
1546
1523
1.502
1502
H502
1582
1502
1502
1502
1502
1502
1502
1502
1502
1502
1502
1502
1502
1502
1502
1502
1502
1502 .
QUfiRTERLY
Bz.
NESHflP
EX.
(Hg/yr)
9
304
887
1192 '
1214
1187
1170
1156
1124
1108
1094
1077
1061
1045
1031
1031
1031
1031
1031
1031
1031
1031
1031
1031
1031
1031
1031
1031
1031
1031
1031
1031
1031
1031
1031
8USRTERLY
Bz.
flLL Nfl PRESS
NO-EX.
(Mg/yr)
288
579
596
595
585
572
564
557
541
534
527
519
511
504
497
497
497
497
497
497
497
497
497
497
497
497
497
497
497
497
497
497
497
497
497
QUfiRTERLY
Bz.
fill Nfl flREflS
EX.
(Mg/yr)
197
400
420
• 423
416
406
401
' 396
385
379
374
369
363
358
353
353
353
353
353
353
353
353
353
353
353
353
. 353
353
353
353
353
353
353
353
353
17123




 5290
12148




 3740
4394




1353
51430




14230
35271




9743
17931




5530
12700




 3910
                                                        E-24

-------
  QUflRTERLY
     Bz.
SEL  Nfl flREflS
     EX.
   (Mg/yr)
      71
     145
     IK
     153
     150
     147
     145
     243
     139
     137
     135
     133
     131
     129
     128
     128
     128
     128
     128
     128
     128
     128
     128
     128
     128
     128
     128
     128
     128
     128
     128
     128
     128
     128
     128

     4594

     1414
                                                           E-25

-------

-------
       APPENDIX F

EXPOSURE AND HEALTH RISK
        ANALYSIS
             F-l

-------

-------
                          Table  F-l.   CONVERSION  FACTOR  CALCULATIONS
Pollutant

Gasoline Vapors
  - Plausible  Upper Limit
      rat studies  (kidney)
      mice studies (liver)
                           Seventy year unit risk   Molecular
                             factor (per ppro)
                                3.53 x ID"3
                                2.14 x 10-3
                                                                        Ratio to YOC emissions
                                 From storage tanks

                                    1
                    From  refueling

                       1
  - Maximum Likelihood Estimate
      rat studies  (kidney)       2.01 x 10-3
      mice studies (liver)       1.44 x 10-3

                                2.2  x ID'2

                                4.2  x 10-1

                                2.8  x 10-2
Benzene

EDB

EDO
                       78.11

                      187.87

                       98.97
0.0060

0.000046

0.00047
0.0066

0.000052

0.00053
                    For Sulk Plants, Bulk Terminals  and  Service Station Inloading
      Risk  from gasoline vapors
         Risk from Benzene
                                = 3.53 x ig~3 gasoline  vapors  incidence/ppm gasoline vapor x
                                        2.2 x 10-2 benzene  incidence/ ppm benzene

                                x 78.11 g/g-mole Bz  x 1 g  gasoline vapor/gYOC = 31.6 (Plausible Upper Limit,  rat)
                                  66 g/g-mole gasoline    0.0060 g benzene/gVOC
                                     vapor
                                =  2.14 x 10-3 x  (78.11) x  (1)
                                   272X 10-2 (66)  X  (0.0060)
                                                                    19.2 (Plausible Upper Limit,  mice)
                                •  2.01 x 10-3  x  (78.11) x (1)   =  18.0 (Maximum Liklihood Estimate,  rat)
                                   2.2 X 1Q-2 x (66)  X  (0.0060)
                                =  1.44 x 10-7  x  (78.111 x  (1)
                                   2.2 x 10-2 (66) X  (0.0060)
             Risk from EDC
             Risk from EDB

             Risk from EDB
             Risk from benzene

             Risk from EDC
                                                                =   12.9  (Maximum Likelihood Est


                                                                     =  1.29


                                =  4.2 x 10-1 x (78.11)  x  (0.000046)   =  6.09 x 10-2
=  2.8 x 10-2 x  (187.87) x  (0.00047)
   4.2 X 10-1 x  (98.97) X (0.000046)
                                   Z7Z X 10-^ X (187.87) X  (0.0060)
                       	  =  2.8 x 10-2 x (78.11)  x  (0-00047)
              Risk from benzene"     2.2 x 10-2 x (98.97)  x  (0.0060)
                                      =  7.87 x 10-2
                              For Service Station  Refueling (Outloading)
      Risk  from gasoline vapors
         Risk from benzene
              Risk from EDC
              Risk from EDB

              Risk from EDB
              Risk from Benzene

              Risk from EDC
              Risk from benzene
=  3.53 x 10-3 x  (78.11) x  (1)
   2.2 X 10-2 x (66)  X  (0.0066)

=  2.14 x 10-3 x  (78.11) x  (1)
   2.2 X ID'2 X (66)  X  (0.0066)

=  2.01 x 10-3 x  (78.11) x  (1)
   2.2 x 10-2 x (66)  x  (0.0066)

=  2.gi x IP"3 x  (78.11) x  (1)
   2.2 X 10-2 x (66)  X  (0.0066)

=  2.8 x 10-2 x (187.87) x  (0.00053)  =
   4.2 x 10-1 x (98.97)X  (0.000052)

=  4.2 x 10-1 x (78.11) x (0.0000521  =
   2.2 x 10-^ x (187.87) x  (0.0066)

=  2.8 x 10-2 x (78.il) x (g.gggss)
   'i.'i X 10-2 x (98.9/i X (0.0066)
                                                                        28.8 (Plausible Upper Limit,  rat)


                                                                        17.4 (Plausible Upper Limit,  nice)


                                                                        16.4 (Maximum Likelihood Estimate,  rat)


                                                                        11.7 (Maximum Likelihood Estimate,  mice)


                                                                        1.29


                                                                        6.25 x 10-2


                                                                        8.07 x 10-2
                                                       F-3

-------
                                Table F-2.  Bulk Terminal Incidence  - Theoretical
Ytar
  1985
  1937
  1238
  1583
  1538
  1391
  1S9£
  1993
  1934
  1S95
  1395
  1997
  1998
  1999
  2001
  £§02
  £603
  2804
  2S05
  SMI
  3)38
  2011
  i«i£
  £013
  £614
  £015
  £.315
  £017
  £918
  £319
  £923
Baseline
Bulk
Terainal
Incidence
Due To
Benzene

8.3883
0.0788
0.3773
0.0753
3.3749
3.0738
0.3731
0.0725
0.0712
0.07%
0.3699
0.0693
0.8686
0.0680
0.3574
0.0674
0.3674
0.0674
0.2674
0.0674
8.0674
0.0674
0.8674
0.0674
0.0574
0.0674
0.2574
0.2574
•3.3674
0.0674
3.8674
0.C674
3.3674
0.0674
8.2674
Controlled
Bulk
Terminal
Incidence
Due To
Benzene

0.0803
0.0723
0.0582
0.3508
0.0501
0.0493
•3.0483
0.0484
0.0474
0.0459
0.0465
0.0460
3.0455
0.0451
0.0446
0.0446
0.0446
. 0.0446
0.0445
0.0445
0.0445
0.0446
0.0446
0.0446
0.3445
0.3445
3.0445
0.3445
3.0445
, 0.9445
0.0445
0.0446
0.0445
0.0446
3.0445
Baseline
Bulk
Terainal
Incidence
Due To
Sas Vauors
(PUL,rat)
2.540
2.493
2.447
2.408
2.359
2.335
2.315
2.294
2.254
2.233
2.213
2.193
2.172
2.152
2.132
2.132
£.132
2.132
2.132
2.132
2.132
2.132
2.132
2.132
2.132
2.132
2.132
2.132
2.132
.2.132.
2.132
2.132
2. 132
2.132
2.132
Controlled
Bulk
Terminal
Incidence
Due To
Sas Vaoors
(PlL,rat)
£.540
2.289
1.843
1.608
' 1.585
1.560
1.545
1.530
1.501
1.486
1.471
1.456
1.441
1.426
1.412
1.412
1.412
1.412
1.412
1.412
1.412
1.412
1.412
1.412
1.412
1.412
1.412
1.412
1.412
. 1.412
1.412
1.412
1.412
1.412
1.412
Baseline
Bulk
Terminal
Incidence
Due To
Sas Vapors
(PULaice)
1.540
1.512
1.483
1.455
1.436
1.415
1.403
1.391
1.366
1.354
1.342
1.329
1.317
1.305
1.292
1.292
1.292
1.292
1.292
1.292
1.292
1.292
1.292
1.292
1.292
1.292
1.292
1.292
1.292
. 1.292
1.292
1.292
1.292
1.292
1.292
Controlled
Bulk
Terminal
Incidence
Due To
Sas Vauors
(PULBiice)
1.540
1.388
1.117
3.975
0.961
3.946
3.937
0.928
0.913
0.901
0.892
0.883
0.874
3.855
3.856
8.856
0.856
3.856
3.356
3.855
3.856
0.855
3.855
3.856
3.856
0.855
0.855
3.855
3.856
0.855
3.856
9.855
3.856
3.856
0.855
Baseline
Bulk
Terminal
Incidence
Due To
Gas Vapors
(MLLrat)
1.446
1.420
1.393
1.357
1.349
1.338
1.318
1.305
1.283
1.272
1.263
1.249
1.237
1.225
1.214
1.214
1.214
1.214
1.214
1.214
1.214
1.214
1.214
1.214
1.214
1.214
1.214
1.214
1.214
. 1.214
1.214
1.214
1.214
1.214
1.214
                 2.44

                0.773
 1.67

3.565
77.18

24.45
52.92

17.88
45.79

14.33
32.38

13.84
43.95

13.93
                                                   F-4

-------
Table F-2.  Bulk Terainal  Incidence - Theoretical
fear






1986
1387
1388
1389
1390
1331
1992
1933
1394
1935
19%
1937
1393
1339
2330
£881
£002
2883
£384
2005
2306
2007
2888
2909
2818
2311
2012
2013
2014
2015
2916
2017
2018
2319
2320
SUM =
NPV =
Controlled
Bulk
Terminal
Incidence
Due To
Gas Vaoors
OLE, rat)
1.446
1.333
1.349
0.915
3,903
0.388
• 8.389
8.871
3.354
8.846
3.338
3.329
8.821
0.812
8.384
3.884
9.884
3.384
8.384
0.304
3.884
8.834
8.384
8. 304
0.804
0.804
•3.884
8.834
3.384,
3. 384
3.884
0.884
0.884
3.304
8.804
38.13
10.18
Baseline
Bulk
Tersinal
Incidence
Due To
Gas Vapors
(ME, Mice)
1.336
1..017
8.338
3.379
8.367
0.352
3.344
3.336
3.919
0.311
8.983
3.894
8.886
8.878
8.878
8.878
8.878
8.873
8.878
3.873
0.878
8.878
8.878
8.378
8.878
3.873
3.870
0.870
3.878
3.878
8.870
3.870
0.870
3.878
8.873
31.43
9.38
Control led '
Bulk
Terainal
Incidence
Due To
Gas Vapors
(ME, Bice)
1.835
8.334
8.752
3.655
8.647
8.636
0.638
8.624
8.512
3.636
8.688
8.534
8.588
8.582
0.576
8.576
8.576
3.576
8.576
8.576
8.576
8.576
8.576
3.576
3.576
3.576
8.575
0.575
8.576
8.576
8.576
8.576
8.575
8.576
8.575
21.53
7.23
Baseline
Bulk
Terminal
Incidence
Due To
EDB

8.88144
8.88131
8.88121
8.88183
3.30094
3.83385
8.38872
8.88364
3.88858
8.30344
8.88841
3.80837
3.83833
8.83329
8. 08825
8.08825
3.00825
8.03825
8.63325
3.38825
8.38825
0.88325
8.08825
3.83825
3.88025
8.08825
8.08825
8.38825
8.88325
3.80825
8.08825
8.88325
0.83825
8.83825
8.S8825
3.8157
8.8878
Control led
Bulk
Terminal
Incidence
Due To
EDB

3.88144
8.88128
3.38898
8.38063
8.33862
3.08856
3.88847
3.88342
3.33833
8.38823
8.88826
3.08823
3.83823
8.38817
8.88814
8.38814
3.88814
3.88814
8.08814
3.88314
3.88814
8.38814
8.33814
0.38814
8.68814
8.33314
3.83814
3.33314
3.38814
8.88814
8.38814
8.08814
8.83814
0.38814
3.38814
8.8188
3.08597
Baseline
Bulk
Terrainai
Incidence
Due To
EDC

3.38186
0.88178
3.38157
8.88133
8.88122
3.88189
8.38833
8.83883
8. 83865
0.38857
3.38852
8.38347
3.38842
8.88937
8.38832
8.88832
8.38832
3.88832
0.33832
3.83332
8.88832
8.33832
3.88832
8.83832
8.08332
8.83832
8.08332
3.33332
8.88832
0.83332
3.S3832
3.88832
8.38832
3.00032
3.88832 .
0.8283
0.8101
Controlled
Bulk
Terminal
Incidence
Due To
EDC

3.38186
3.88156
8.03117
3.88389
8.86383
0.88872
8.33361
8.83854
3.88043
8.08837
8.38833
8.83838
3.88826
3.88822
8.33819
8.33819
0.33019
3.08819
8.88819
0.88813
0.88819
8.88819
8.88319
3.83819
3.88819
0.38819
8.80813
3.38319
8.33319
3.83319
0.88819
8.88819
3.S8819
8.33813
3.33319
8.8139
8.0877
                   F-5

-------
                Table F-3.  Bulk  Plant  Incidence - Theoretical
ar






1986
1937
1988
1989
1990
1991
1992
1993
1994
1995
1996
1997
1998
1999
sees
£00i
2002
m$
sew
2B05
sm
£097
20ea
£899
2813
1311
2012
£913
2814
£015
2016
£617
2918
£019
2028
Baseline

Bulk Plant
Incidence
Due To
Benzene

3. 8439
0.0427
0.0415
0.0484
0.93%
0.0387
0. 8382
3.0377
0.0367
0.0362
0.0357
3.0352
8.0346
0.8341
0.8336
8.0335
0.0336
0.8336
0.8336
8.8336
0.0336
0.0336
0.8336
0.8336
0.8336
3.3335
8.0336
8.8336
0.8336
8.3336
0.0336
0.0336
8.3336
St. 8335
0.8336
Controlled
(EX)
Bulk Plant
Incidence
Due To
Benzene

0.0439
0.8365
0.0238
0.0174
8.0178
8.8167
0.0164
0.8162
0.0158
0.0156
0.8153
0.0151
0.0149
0.0147
0.0145
0.8145
8.0145
0.0145
0.8145
0.0145
0.0145
0.0145
0.0145
0.0145
0.0145
0.0145
8.8145
3.0145
0.8145
0.8145
0.8145
0.8145
0.0145
8.8145
0.3145
Controlled
(NO EX)
Bulk Plant
Incidence
Due To
Benzene

8.3439
8.8358
8.8215
0.0144
0.0141
8.8138
8.8136
0.0134
0.0131
0.8129
0.0127
8.8125
0.0124
0.0122
0.8120
0.0120
0.0120
8.0120
0.0128
0.0120
0.8120
0.0120
0.0120
0.8120
8.8128
8.8128
8.8128
8.8120
8.8128
8.0128
0.8120
0.0128
3.8120
0.8120
0.8128
Baseline

Bulk Plant
Incidence
Due To
Bas Vapors
(Pll,rat)
1.389
1.351
1.314
1.277
1.253
1.226
1.218
1.193
1.161
1.145
1.129
1.113
1.097
1.881
1.864
1.064
1.864
1.064
1.864
1.864
1.064
i.064
1.864
1.964
1.864
1.064
1.864
1.864
1.864
1.864
1.864
1.854
1.064
1.854
1.864
Controlled
(EX)
Bulk Plant
Incidence
Due To
Gas Vapors
(PUL,rat)
1.389
1.159
8.752
8.549
0.539
8.527
8.520
0.513
0.499
0.492
8.485
0.479
3.472
8.465
0.458
8.458
8.458
0.458
8.458
0.458
0.458
0.458
0.458
8.453
0.458
8.458
0.458
0.458
8.458
3.458
3.458
8.458
0.458
8.458
0.458
Controlled
(NO EX)
Bulk Plant
Incidence
Due To
Gas Vapors
(PUL,rat)
1.389
1.134
8.6B8
8.455
8.447
0.437
8.431
8.426
0.414
8.488
8.403
8.397
0.391
8.385
8.388
0.388
0.380
8.388
0.388
8.388
8.338
8.380
8.380
0.388
8.388
8.388
8.380
0.380
8.380
. 8.388
8.388
8.388
8.388
8.388
0.388
Baseline

Bulk Plant
Incidence
Due To
Gas Vapors
(Pit, mice)
8.842
8.319
0.797
8.774
0.760
8.743
0.733
8.723
0.704
0.694
8.684
0.675
8.665
8.655
8.645
8.645
8.645
8.645
0.645
3.545
0.645
8.645
8.645
8.645
0.645
8.645
8.645
3.645
0.645
8.645
0.645
8.645
3.645
8.545
0.645
 1.24




0.403
0.583




8.220
0.498




8.196
39.29




12.75
18.45




 S.96
15.77




 6.21
£3.82




 7.73
                                    F-6

-------
Table F-3.  Bulk Plant  Incidence - Theoretical
Year






1985
1987
i'388
1989
1958
1991
1992
1993
1994
1995
19%
1997
1998
1999
£088
20(91
£00£
2003
£884
£835
2036
£007
£088
£339
£010
£911
2012
2813
£014
£015
£216
£017
£018
£019
20£0
SUM =
NPV =
Controlled
(EX)
Bulk Plant
Incidence
Due To
Gas Vaoors
(PUL, rnics)
3.842
0.703
0.455
0.333
3.327
0.328
0.315
3.311
0.383
3.298
0.294
0.290
0.286
0.282
0.277
3.277
9.277
0.377
8.277
3.277
8.277
3.277
9.277
8.277
9.277
3.277
8.277
0.277
9.277
3.277
0.277
3.277
0.277
0.277
0.277
11.19
4.22
Controlled
(NO EX)
Bulk Plant
Incidence
Due To
Sas Vaoors
(PUL,fflice)
0.842
0.686
0.412
0.276
3.271
0.265
0.261
0.258
0.251
0.248
0.244
0.241
8.237
0.234
0.230
0.230
d.230
0.230
0.233
0.230
0.220
0.230
0.230
8.238
0.230
0.230
0.238
0.230
0.230
0.238
0.230
0.230
0.233
0.230
0.£33
9.56
3.76
Baseline

Bulk Plant
Incidence
Due To
Sas Vapors
(BLE.rat)
3.791
0.778
0.748
8.727
0.713
0.698
0.689
0.688
0.661
0.652
8.643
0.634
0.624
0.615
0.686
8.606
„. ,0.685
0.606
8.686
0.606
0.606
0.686
0.686
0.606
0.636
0.686
0.606
0.606
0.606
8.686
0.606
0.686
8.606
0.686
0.606
22.37
7. £6
Controlled
(EX)
Bulk Plant
Incidence
Due To
Sas Vapors

-------
Table F-3.  Bulk Plant  Incidence - Theoretical
ar






1986
1S87
1958
1989
1930
1991
1932
1993
1934
1995
19%
1997
1958
1999

2001
2022
2e03
2C34
2885

2807
2828
2639
2913
2011
2012
2013
2014
2015
2016
2017
2818
2919
202®
8UN-
»V«
Controlled
(NO EX)
Bulk Plant
Incidence
Due To
Gas Vaoors
(ME, aice)
0.566
0.463
0.277
9.186
0.182
3.178
0.176
0.174
0.169
0.157
0.164
0.162
0.160
0.157
0.155
0.155
0.155
0.155
0.155
0.155
0.155
0.155
0.155
0.155
0.155
0.155
0.155
0.155
0.155
0.155
0.155
0.155
0.155
0.155
0.155
6.43
2.53
Baseline

Bulk Plant
Incidence
Due To
EDB

0.S00784
0.000685
0.000605
0.000527
0.000456
0.000403
0.000241
0.000279
0.000233
0.000186
0.000155
0.000124
0.000893
0.000062
0.000031
0.000031
0.000031
0.000031
0.000031
0.000031
0.000031
3.000031
0.000031
0.000031
0.600031
0.000031
0.000031
0.000031
0.008031
0.300031
0.800031
3.000031
0.000031
0.000031
0.000031
0.00558
0.00362
Controlled
(EX)
Bulk Plant
Incidence
Due To
EDB

0. (500784
0.000588
0.000346
0.800227
0.000196
0.800173
0.300147
0.000120
0.000130
0.030080
0.000867
0.000053
0.003040
0.000027
0.000013
0.000013
0.000013
0.000013
0.800013
0.300013
0.000013
0.000013
0.006013
0.030013
0.800013
0.i30013
0.000013
0.000013
0.000013
0.000013
0.003813
0.030013
0.000013
0.238313
0.000013
0.80323
3.00234
Controlled
(NO EX)
Bulk Plant
Incidence
Due To
EDB

3.300784
0.003575
0.003313
0.000188
0.'800163
0.000144
8.300122
0.000100
' 0.000083
0.303066
0.S00055
0. '380044
0.000033
0.003022
3.303011
0.030011
0.000011
0.000011
3.033011
0.380011
0.008811
0.030011
0.000011
0.000011
0.300011
8.800311
9.323011
0.303011
0. 000611
0.333011
8.000011
3.000011
3.003911
0.000011
3.390011
0.89292
3.550218
Baseline

Bulk Plant
Incidence
Due To
EDC

0.001314
0.000886
0.803782
3.300682
3.300583
0.300521
3.003441
0.000361
0.000301
0.000241
0.303200
0.030160
0.303120
0.030080
0.000040
0.030040
0.000040
0.003040
0.303040
0.030040
3.800340
0.000040
3.303840
0.300040
9. 338940
0.000040
0.333040
3.030040
0.308040
0.033040
0.308040
3.330040
3.003040
3.030040
3.303040
8.00722
0.30468
Controlled
(EX)
Bulk Plant
Incidence'
Due To
EDC

3.301014
0.300760
0.000448
3.000293
0.000253
0.003224
0.380190
0.300155
0.390129
0.000103
0.300086
3.000069
0.000052
0.000034
0.300017
3.300017
0.330017
0.303017
0. 000017
0.300017
0.038017
0.000017
0.000017
3.300017
3.030017
3.300017
0.303017
3.300017
0.300017
0.000317
' 0.033017
3.303017
3.300017
3.300017
3.303017
0.00417
0.30383
Controlled
(NO. EX)
Bulk Plant
Incidence
Due To
EDC

3.001014
0.300744
0.300405
0.033243
3.300210
3.300186
3.300157
3.000129
0.300107
' 0.1300086
0.000071
0.000057
0.000043
0.300029
3.000014
0.300014
3.303014
3.030014
3.303014
0.333014
3.303014
0.303014
0.303014
3. 303014
0.308014
3.300014
3.030314
3.303014
3.030314
3.333014
' 3.800014
0.300014
8. 338814
8.330014
8. 803014
3.30378
3.33281
                      F-8

-------
Table F-4.  Service Station  Incidence Due To Benzene - Theoretical
year




1985
1987
1388
1939
1993
1591
1992
1993
1994
1995
1996
1997
1998
1999
ma®
2881
2882
2383
£8**
£005
SM&
2887
£3ea
•2239
£312
£8ii
23:2
£013
£814
2315
i316
2017
£818
£819
£828
SUN =
•iPV =
Baseline

Total
Incidence Due
To Benzene
9. ££91
3. £233
3.2159
3.2138
3.2368
S.£322
3.19%
.3. 1959
3.1916
«. 1889
3.1863
3.1835
3. 1818
3. 1783
3.1756
3.1756
3.1756
3. 1756
3. 1756
3. 1756
3.1756
3.1756
8. 1755
3.1756
8. 1755
3. 1755
8. 1756
3.1756
3. 1755
3.1756
3. 1756
3. 1756
3. 1755
3. 1756
3. 1755
6.43
£.13
St.II-Nfl*
(EX)
Total
Incidence Due
To Benzene
3. £117
3.1877
3. 1799
3. 1734
3. 1782
3. 1564
•3. 1642
3. 1621
3.1577
3.1555
3. 1533
3.1511
8. 1489
3. 1467
8. 1445
3. 1445
3. 1445
3.1445
•3. 1445
3. 1445
3. 1445
3.1445
0. 1445
3. 1445
3. 1445
3. 14*5
3. 1445
3. 1445
3. 1445
3. 1445
3. 1445
3. 1445
3. 1445
3. 1445
•2. 1445
5.35
1.76
St. II-Nfl*
(NO EX) .
Total
Incidence Due
To Benzene
3. 1938
•3. 1623
3. 1543
3.1483
3.1455
3.1423
3. 1434
8.1385
•3. 1348
3. 1329
3. 1311
0.1292
8. 1273
0.1255
8.1236
3. 1235
0. 1236
3. 1236
8. 1236
3. 1235
3. 1236
3. 1236
3. 1236
3. 1235
8. 1235
3. 1235
3. 1235
3. 1235
3. 1236
3. 1235
3. 1235
3. 1236
3. 1235
3.1235
8. 1235
4. Si
1.53
St.II-NA
(EX)
Total
Incidence Due
To Benzene
3. 1933
8. 1584
3, 1436
3.1339
8. 1313
8.1285
8. 1258
3. 1251
3. 1217
3. 1288
•3. 1183
8.1156
8. 1149
8.1133
8. 1115
8.1116
3.1116
3. iliS
•2.1116
8.1115
3.1 US
8.1116
3. 1116
8.1116
3.1115
3.1116
3.1115
8. 1115
3.1116
8.1115
3.1116
3.1116
'2.1115
3.1115
3.1115
4.18
1.48
.St. II-Nfl
!NO EX)
Total
Incidence Due
To Senzene
3. 1712
8. 1378
3. -3974
3.8914
3.3897
8.3377
3. 3365
8.3854
8.^331
3.3819
•3. 3888
3.87%
3. 2785
8.8773
3.3762
3.8762
§. 8762
8.8762
3.3752
3.8762
,8.3762
3.8762
3.8762
3.3762
3.3762
8.3752
3.3752
3.8762
3.0752
8.3752
3.8762
8.3762
3.3762
8.3752
3.3752
2.98
1.08
St.I-Naticn
(EX)
Total
Incidence Due
To Benzene
3.2291
3.2174
3.2084
3. 1894
•3. 1858
3. 1817
3. 1794
3. 1778
3.1722
3. 1598
3. 1574
3. 1650
3.1625
3. 1682
•3, 1578
3. 1578
8. 1578
8. 1578
•3. 1578
3.1578
3. 1578
8. 1578
8.1578
3. 1578
3. 1578
8. 1578
•3. 1573
8. 1578
3. 1578
3. 1578
3. 1578
3.1573
3. 1578
3. 1578
3. 1578
5.87 '
1.93
St. I-Nation
(NO EX)
Total
Incidence Due
To Benzene
3.2291
3.2136
3. 1895
3. 1754
3.1723
3. 1683
3. 1651
3. 1639
8. i594
8. 1572
3. 1558
3. 1528
3. 1586
3.1484
3. 1461
8.1451
3. 1451
8. 1461
8. 1461
8.1461
3. 1461
8. 1461
3. 1451
3. 1461
8. 1461
3. 1461
8. 1451
3.1461
8. 1461
3. 1461
3. 1461
8.1461
3.1461
3. 1451
8. 1461
5.47
1.82
                              F-9

-------
                       Table F-4.  Service Station Incidence Due To Benzene - Theoretical
Year
St.II-Nation
    '.EX)
St.II-N'ation
  API (EX)
St.II-Xation  St. I ill-Nat ion St.Iill-Nation Onboard-Nation
   (NO EX)         (EX)          (NO EX)
                                                                                            St.II-Nfl*
                                                                                              (EX)
                                                                                            & Onboard
    •"oral          Total          Total          Total          Total          Total          Total
Incidence Due  Incidence Due  Incidence Due  Incidence Due  Incidence Due  Incidence Due  Incidence Due
 "o Benzene     To Benzene     To Benzene     To Benzene     To Benzene     To Benzene     To Benzene
1936
1987
1988
1989
1990
1991
1952
1993
1394
19S5
19%
1997
1938
1999
283d
ae0i
23K
c«03
•a»
2385
2835
cw?
zm
£289
•914
a3h
e'312
£313
1-314
£$15
2916
2017
318
S919
£?££
£12 -
vpy =
3.2291
9. 1991
0.1473
0. 1172
9.1116
3. 1892
8. 1377
0.1@63
8. 1834
0.1820
0. 1005
0.0991
0.3977
0.0962
0.0948
8.8948
8.3948
0.0948
0.0948
0.8948
0.3948
8.8948
8. 8948
0.0948
0.8948
0.0948
3.0948
3.3948
3.3948
0.0948
8.8948
0.0948
8.3S43
3.3948
0.2948
3.72
1.34
0.2291
0.2210
0. 1959
3.1555
3. 1259
0. 1134
3.1141
3.1381
3.1034
0. 1020
0. 1005
0.0991
0.0977
0.3962
0.0948
0.0948
0.8948
0.0948
0.0948
0.0948
0.0948
0.8948
3.0948
• 0.0948
3.3948
8.0948
'3.3948
8.0948
3.8948
0.0948
3.8948
0.8948
•3.3943
0.0S48
0.3948
3. 85
1.45
0.2291
0.1S63
3. 1102
3.2685
3.S633
0.0619
3.3611
0.0603
3.0537 '
0.3579
0.0570
0.0562
0.0554
0.8546
0.0538
0.0538
•3.3538
S.0538
•3.3538
3.8538
3.3538
0.0538
0.8538
•3.3538
0.0538
8.8538
3.8538
3.2538
•3.8538
0.0538
3.3538
3.2538
0.8338
8.8538
0.8538
2.31
3.35
8.2291
3. 1334
3.1308
8.3959
3.8986
8.0887
0.5575
8.8863
3.8843
3.0828
0.8817
8.8885
3.8793
3.8782
8.8770
0.8778
0.8770
3.0778
8.3773
•3.9778
8.1773
3.0773
0.3773
0.3778
•3.8773
8.3773
8.8773
3.3773
3.8773
0.3778
8.3770
8.3778
3.8778
8.3778
8.3773
3..11
1.17
8.2291
3. 1785
3.13374
0.3391
3.8347
8.3339
3.3334
8.8333
3.3321
3.8317
3.3312
3.8388
3.8333
8.8299
3.8294
8.8294
8.3294
8.8294
•3.8294
0.8294
8.8294
3.8294
3.3234
3.8294
3.3294
3.3294
3.8234
3.8294
3.8294
3.0294
•3.3294
8.8294
8.0294
8.8294
3.8294
1.47
3.72
2.2291
8.2233
3.2014
3. 1815
3.1655
3. 1588
•3. 1377
3.1266
3. 1139
3. 1849 "
3.8969
3.38%
3.3831
3.8772
3.0721
8.3721
3.3721
3.3721
3.3721
3.8721
8. 3721
3.3721
3.3721
3.3721
3.3721
8.3721
3.3721
3.8721
3.3721
0.8721
8.3721
3.8721
8.8721
3.0721
8.8721
3.49
1.48
3.2117
3. 1877
<3. 1583
3. 1515
3. 1393
3. 1273
3. 1179
8. 1894
8.8995
0.3926
8.8864
3.8887
3. 3757
3.8711
3.8670
0.2691
3.3714
3.8719
3.8721
3.3721
8.8721
3.8721
3.3721
3.8721
3.8721
3.8721
3.3721
3.3721
3.3721
3.3721
8.3721
8.8721
. 3.3721
3.8721
3.3721
3.22
1.33
                                                   F-10

-------
Table F-4.  Service Station Incidence DUB To Benzers -Theoretical
'aar





1586
1587
1938
1589
1998
1931
1952
1533
1994
1395
1996
1557
1998
1999
3580
£381
2822
£303
£084
£635
3826
£237
£328
£039
2018
£811
£212
£813
£814
£315
£316
£817
£018
£219
£028
m =
*V =
St.lI-Nfl*
(NO EX)
i Cr.-board
"otai
Inci de?scs Due
To Benzene
3. 1988
8. 1623
3.1448
3.1332
2.1199
8.1299
•2. 1321
8.0953
3. -2367
8.363-3
3.0757
3.3739
3.2667
8.3623
8. 3594
8.0646
3. 3737
8.8717
0.3721
3.3721
3.3721
0.8721
3.3721
3. S721
•2.3721
3.5721
3, 3721
3.3721
8. 07£i
3.3721
8. 3721
3.3721
8.3721
0..3721
3.0721
3.83
-..15
3t.II-.Nfl
(EX)
4 Onboard
Total
Incidence Due
To Benzene
8.1933
8. 1584
0. 1327
3.1189
0.1181
8.1817
•2.0958
3.8898
8.8318
8.3769
8.3724
8.8633
8.8547
8.8614
8.0584
8.8640
a. 8703
8.8715
8. 3722
8.3721
2.8721
8.8721
8.8721
8.8721
8.8721
3.8721
8.8721
8.8721
8.0721
8.8721
6.8721
3.8721
8.3721
3.8721
«. 3721
2.51
1.10
St.II-Nfl
(NO EX)
4 Onboard
Total
Incidence Due
To Benzene
" 0. 1712
3.1870
8.3929
8.8829
8. -2775
8.8725
0.0685
8.8649
8.8684
8.8574
0.8547
0.8522
0.2508
3.0479
8,0460
3.3570
8.3S93
0.8713
8.0720
3.8721
3.3721
8.8721
• 3.3721
0.0721
3.3721
8.3721
3.3721
3.3721
0.8721
8.8721
3.3721
3.3721
3. 3721
8.0721
3. 3721
2.53
3.37
St. I (EX)
& Onboard-
Nation
Total
Incidence Due
To Benzene
3.2291
3.2174
3. 1849
8. 1681
0. 1445
8.1255
8.1174
8.1866
8.0945
0.8858
3.3780
,8.0710
0.0648
8.3592
0.0543
3.3543
3. 3543
8.3543
0.0543
3.3543
0.8543
0.3543
3.8543
3.3543
3. 8543
3.3543
3.2543
3. 3543
3. 3543
3. 3543
8.3543
3. 3543
3.3543
0.3543
3. 0543
£.83
1.31
St. I (NO EX)
J Onboard-
Nation
Total
Incidence Due
To Benzene
8.2291
8.2136
8. 1741
8. 1461
0.1388
3.1151
3.1041
8.8935
3.8817
0.3732
8.3655
8.3587
3.8527
0.8473
8.8426
0.0426
3. 3426
8.0426
8.0425
8.8425
3.3426
0.3426
3.3425
3.3426
8.8425
3.0426
3.8426
3.3426
3.8425
3.0426
3.8425
3. 3425
•3. 34£5
8.3425
3.8425
£.48
i.£9
                                                             & St. I (EX)   i St.I  (NO EX)
                                                            it Gnbd-Nation  & Onbd-Nation
;al
rice Due
izene
•3. 1933
3.1464
3. 1218
8. 1037
3.0953
8.8871
8.2807
3.0748
3.0631
0.8533
3.3591
0.0552
3.0517
0.3486
8.0453
8.0493
3. J532
8.8539
3.3543
3. 3543
8.8543
3.3543
0.0543
3. 3543
3.3543
3.3543
3. 3543
3. 3543
3.8543
3.3543
-3.3543
3.8543
3. 3543
3.3543
3. 0543
Total
Incidence Due
To Benzene
. 0. 1712
8.1828
3.3782
3.8639
3.8590
3.8543
0.8505
3.3472
3.8432
3. 3484
3.0388
3.8357
3.8337
3.8318
3.3332
8.0354
8. 8413
3.8422
3.8425
3.8425
3.8425
3.0425
3. 0425
3.3425
3.3425
3. 3426
3.8426
3. 34£6
8. 3425
3. 0426
3.3426
3. 34£b
3.3425
3.8425
3.3425
                                                                    £.37

                                                                    3.97
1.72

8.69
                           F-ll

-------
                     "able F-4.  Service  Station
                                          Due To Benzene - Theoretical
ar





1336
13S7
.338
1*83
.353
1=31
*:3£
13S3
1H&
1595
13%
1997
.393
1S99
22TO
£231
£232
2233
2234
£225
£?5?5
2237
£w38
£.235
I!lO
2S11
•II 12
£?13
£2114
£315
£316
2217
2! 13
£215
£323
5t.II JEX)
i Qnboard-
\a*ion
Total
incidence Due
~o Senzane
£ flJW
3. .991
3. 1422
3. 1375
2. -2979
S.391S
2. 2372
3.0829
0.3775
0.8741
3.3739
8.8679
8.3652
8.0627
3.3634
8.3684
3.3538
0.8683
3.3715
3.8721
3.3721
0.3721
3.3721
3.3721
•3.9721
0.3721
3.3721
3.2721
•3.8721
3.3721
3.3721
3.3721
3.8721
8.8721
3.3721
St. II (NO EX)
i Ondoarc-
Nasion
Total
Incidence Due
~; Banzsne
3.2291
8. 1363
3. 1397
3.8674
3.0618
3.3681
8.8589
8.8578
3.0559
8.8549
3.8539
0.3529
3.0519
0.3510
3.0501
0.8501
3.8550
3.0651
3.8712
0.3721
8.8721
0.8721
3.3721
0.0721
•3.8721
3.3721
3.0721
3.3721
3.3721
3.8721
3.0721
3.3721
3.8721
3.3721
•3.8721
St. I 411 (EX)
& Onboard-
Mat ion
"otai
Incidence Due
To Senzene
. 3.2291
3. 1934
3.1257
8.3862
8.8769
0.8713
3.8578
8.0638
8.8582
3.8558
0.8520
•3,8493
3.8469
8.0446
0.8425
8.0426
0.0452
0.0505
3.8537
8.0542
3.3543
8.8543
3.8543
8.0543
8.8543
8.8543
3.8543
3.8543
3.8543
8.3543
3.3543
8.3543
3.8543
8.8543
3.8543
St.Iill {MO EX)
i Onboard-
Nation
Total
Incidence Due
To Benzene
3.2291
3.1785
8.8868
8.3381
3.3332
3.0328
0.8312
0.3385
0.8293
•3. 3287
8.8280
8. 8274
8.8268
0.8263.
8.0257
8.8257
8.8295
0.8373
3.8419
8.8426
8.8426
3.8425
8.8426
3.8426
3.8425
3.3426
8.8425
0.8426
3.8425
3.8426
3.8426
3.8426
•3.3425
3.3425
•3.8426
Slil =
2.93
2.53




8.97
2.32




•3.98
1.67




3.72
                                  F-12

-------
Table F-5.  Service Station  Incidence Due To Gasoline Vapors (Plausible Upper  Liait) - Theoretical
iar




1986
1987
1988
1989
1990
1991
1992
1993
1994
1995
19%
1997
1998
1999
2888
2001
2882
2083
2084
2085
2036
2007
2088
2089
2010
2011
2012
2313
2014
2015
2016
2017
2018
2019
2020
m =
>v =
Basel ire

Total Incidence
Due To Gasoline
Vapors (PUL, rat)
5.759
S.588
6.488
6.227
6.189
5.975
' 5.897
5.818
5.661
5.582
5.583
5.425
5.346
5.268
5.189
5.189
5.189
5.189
5. 189
5.189
5.189
5.189
5.189
5.189
5.189
5.189
5.189
5.189
5. 189
5.189
5.189
5.189
5.189
5.189
5.189
191.54
62.19
St.II-Nfl*
(EX)
Total Incidence
Due To Gasoline
Vapors (PUL, rat)
6.265
5.567
5.335
5.145
5.348
4.937
4.872
4.807
4.678
4.613
4.548
4.483
4.418
4.353
4.288
4.288
4.288
4.288
4.288
4.288
4.288
4.288
4.288
4.288
4.288
4.288
4.288
4.288
4.288
4.288
4.288
4.283
4.288
4.288
4.288
159. 11
52.20
St.II-HW*
(NO EX)
Total Incidence
Due To Gasoline
Vapors (PUL, rat)
5.884
4.815
4.579
4.481
4.317
4.223
4.167
4.112
4.081
3.945
3.889
3.334
3.778
3.723
3.657
3.667
3.667
3.667
3.667
3.667
3.667
3.667
3.667
3.667
3.667
3.667
3.657
3.667
3.667
3.667
3.667
3.667
3.567
3.667
3.667
135.58
45.24
St.II-Nfl
(EX)
Total Incidence
Due To Gasoline
Vaoors(PUL,rat)
5.724
4.472
4.185
3.985
3.911
3.325
3.775
3.725
3.624
3.574
3.523
3.473
3.423
3.372
3.322
3.322
3.322
3.322
3.322
3.322
3.322
3.322
3.322
3.322
3.322
3.322
3.322
3.322
3.322
3.322
3.322
3.322
3.322
3.322
3.322
124.35
41.58
St.II-Nfl
(NO EX)
Total Incidence
Due To Gasoline
Vapors (PUL, rat)
5.874
3.192
2.386
2.738
2.578
2.619
2.585
2.552
2.481
2.447
2.413
2.373
2.344
2.309
2.275
2.275
2.S75
2.275
2.275
2.275
2.275
2.275
2.275
2.275
2.275
2.275
2.275
2.275
2.275
2.275
2.275
2.275
2.275
2.275
2. 275
86.47
29.72
St.I-Nation
(EX)
Total Incidence
Due To Gasoline
Vapors (PUL, rat)
6.769
5.410
5.386
5.551
5.445
5.326
5.255
5.185
5.846
4.975
4.985
4.335
4.765
4.596
4.626
4.626
4.626
4.625
4.S26
4.626
4.625
4.625
4.525
4.525
4.626
4.526
4.5£6
4.625
4.6££
4.625
4.S2S
4.626
4.S2S
4.526
4.5£S
172. 19
56.38
                                          F-13

-------
Table F-5.  Service Station Incidence Due To Gasoline Vapors (Plausible Upper Liait) - Theoretical
ar




198S
1987
1988
1989
1998
1991
1933
1993
1994
1995
19%
1997
1998
1999
2000
2801
2082
2083
2084
2083
28@6
2037
28€8
2089
2910
2811
2312
2013
2014
2015
2016
2017
2018
2019
2020
UN*
PV =
St.I-Nation
(fffl EX)
Total Incidence
Due To Gasoline
Vapors«PUL,rat)
6.769
6.232
5.543
5.106
5.310
4.900
4.336
4.771
4.642
4.578
4.513
4.449
4.384
4.320
4.255
4.255
4.255
4.255
4.255
4.255
4.255
4.255
4.255
4.255
4.255
4.255
4.255
4.255
4.255
4.255
4.255
4.255
4.255
4.255
4.255
159.48
53. 2S
St.II-Nation
(EX)
Total Incidence
Due To Gasoline
Vapors (PUL, rat)
5.769
5.897
• 4.395
3.522
3.357
3.283
3.240
3.197
3.110
3.067
3.024
2.981
2.938
2.894
2.851
2.851
2.851
2.851
2.851
2.851
2.851
2.851
2.851
2.851
2.851
2.851
2.851
2.851
2.851
2.851
2.851
2.851
2.851
2.851
2.851
111.55
40.09
St.II-Nation
ftPI (EX)
Total Incidence
Due To Gasoline
Vapors (PUL, rat)
6.769
6.529
5.801
4.628
3.769
3.579
3.425
3.249
3.110
3.1357
3.024
2.981
2.938
2.S94
2.851
2.851
2. 851
2.851
2.851
2.851
2.851
2.851
2.851
2.851
2.851
2.851
2.851
2.851
2.851
2.851
2.851
2.851
2.851
2.851
2.851
115.64
43.25
St.II-Nation
(NO EX)
Total Incidence
Due To Gasoline
Vapors (PUL. rat)
6.769
5.524
3.318
2.896
1.344
1.932
1.877
1.852
1.801
1.776
1.751
1.726
1.781
1.676
1.651
1.651
1.651
1.651
1.651
1.551
1.651
1.651
1.651
1.651
1.651
1.651
1.651
1.651
1.651
1.651
1.551
1.651
1.651
1.651
1.651
78.39
28.66
St. nil-Nation
(EX)
Total Incidence
Due To Gasoline
Vapor s (PIL, rat)
6.759
5.718
3.874
2.846
2.593
2.634
2.600
2.555
2.4%
2.461
2.425
2.392
2.357
2.322
2.288
2.288
2.288
2.288
2.288
2.288
2.288
2.288
2.288
2.288
2.288
2.288
2.288
2.288
2.288
2.288
2.288
2.288
2.288
2.288
2.288
92.28
34.70
St. Hill-Nation
(NC EX)
Total Incidence
Due To Gasoline
Vapors (PUL, rat!
S.763
S.27&
2.588
1.166
1.337
1.314
1.881
3.987
8.951
3.947
0.934
3.921
8.907
3.894
0.881
8.881
0.881
3.381
8.881
3.881
3.881
3.881
0.881
3.381
2.881
3.881
0.381
8.881
0.881
9.381
0.881
3.881
9. 881
a. 381
0.881
43.89
21.28
                                            F-14

-------
        Table F-5. Service Station Incidence Due To Gasoline Vanors (Plausible Uocer Liait)  - Theoretical
jar





198&
1987
1988
1989
1990
1991
1992
1993
1994
1995
19%
1997
1998
1999
2080
2801
£082
2083
2084
200S
2006
2007
2008
2009
2010
2011
2012
2012
2014
2815
2315
2017
2018
£019
2020
Onboard-Nation


Total Incidence
Due To Baseline
Vapors (PUL. rat)
6.769
6.588
5.963
5.385
4.321
4.472
4.115
3.794
3.425
3.165
2.932
2.719
2.532
£.360
£.209
£.289
2.209
2.209
2.209
£.£09
2.209
2.209
2.289
2.209
2.209
2.209
2.289
2.209
2.209
2.289
£.£89
2.209
2.289
2.209
£.209
St.IHW*
(EX)
J Onboard
Total Incidence
Due To Gasoline
Vapors (PUL, rat)
5.265
5.567
5.002
4.515
4.159
3.813
3.539
3.293
3.004
2.804
2.523
2.458
2.31£
2.177
£.358
2.120
2. 198
2.203
2.209
2.289
2.289
2.289
2.209
2.209
2.209
2.209
2.289
2.209
2.209
2.209
2.209
2.209
2.239
2.209
2.289
St.IHift*
(NO EX)
J Onboard
Total Incidence
Due To Gasoline
Vapors (PUL, rat)
5.884
4.815
4.334
3.879
3.581
3.292
3.053
2.857
2.615.
2.447
2.296
2.157
2.1234
1.921
1.821
1.985
2.168
2.198
2.209
£.£09
2.209
2.209
2.209
2.209
2.239
2.209
2.239
2.209
2.289
2.209
2.289
2.209
2.209
2.209
2, £'39
St.II-Nfl
(EX)
1 Onboard
Total Incidence
Due To Sasoline
Vapors (PUL, rat)
5.724
4.472
3.357
3.554
3.301
3.054
2.860
2.585
2.475
2.333
2.203
2.384
1.978
1.880
1.792
1.962
2. 155
2.193
2.208
2.209
2. £39
2.209
2. £89
2.209
2. £39
2.209
2.209
2.209
£.£39
2.209
£.£39
2.209
2. £09
£.£09
£.239
St.IHffl
(NO EX)
& Onboard
Total Incidence
Due To Sasoline
Vapors (PUL, rat)
5.374
3.192
2.776
2.434
£.332
2. 131
2.256
1.951
1.833
1.743
1.663
1.538
1.524
1.462
1.406
1.746
£. 124
2.185
2.208
£.£09
2.209
2.239
£.£39
2.239
2.239
2. £&3
2. £39
£.239
£.£39
£.£99
2.239
2. £09
2. £39
2. £05
£.£39
St. I (EX)
t Onboard-
Mat ion
Total Incidence
Due To Basoline
Vapors (PUL, rat)
6.759
6.418
5.441
4.739
4.258
3.3£4
3.475
2. 162
£.313
£.559
£.334
2.133
1.951
1.788
1.546
1.546
1.546
1.S46
1.646
1,546
1,646
1,846
1.546
1.S46
1.S46
i.G4S
I. £45
i.546
1.346
L.34&
i.546
1.346
1. 546
i.S4b
t •' ; C
it wTQ
SUM =




NPV =
135.53




 44.19
97.66




39.65
93.87




34,89
88.22




33.12
-26,14
85.13




38. Z3
                                                  F-15

-------
Table F-5.  Service Station Incidence  Due To Gasoline Vapors (Plausible Upper Limit)  - Tnsoretical
ar





1986
1987
1968
1989
1993
1991
1993
1993
1994
1995
19%
1997
1996
1999
2888
2081
2882
2803
2684
2805
28%
2887
2608
2839
2818
2811
2812
2313
2814
2815
2816
£817
2018
2819
2828
IH*
pys
St. I (NO EX)
i Onboard-
Nation
Total Incidence
Due To Gasoline
VaoorstPUL.rat)
6.769
6.292
5. 898
4.265
3.822
3.397
3.854
2.747
2.406
2.161
1.942
1.743
1.578
1.412
1.276
1.276
1.276
1.276
1.276
1.276
1.276
1.276
1.276
1.276
1.276
1.276
1.276
1.276
1.276
1.276
1.276
1.276
1.276
1.276
1.275
73.47
35.26
St.II-Nfl (EX)
& St. I (EX)
& Onbd-Nation
Total Incidence
Due To Gasoline
Vapors(PULrat)
5.724
4.345
3.588
3.876
2.832
2.595
2.487
2.238
2.841
1.984
1.788
1.667
1.557
1.475
1.393
1.495
1.613
1.636
1.645
1.646
1.646
1.646
1.646
!• D*Tu
1.646
1.646
1.646
1.646
1.646
1.646
1.646
1.646
1.646
1.646
1.646
71.35
28.97
St.II-Nfl (NO EX)
£ St. I (NO EX)
4 Onbd-Nation
Total Incidence
Due To Gasoline
Vapors (PUL, rat)
5.374
3.333
2.313
1.884
1.743
1.685
1.497
1.488
1.284
1.285
1.132
1.866
1.888
8.954
8.986
1.862
1.236
1.265
1.275
1.276
1.276
1.276
1.276
1.276
1.276
1.276
1.276
1.276
1.276
1.276
1.276
1.276
1.276
1.275
1.276
51.35
20.46
St. 1 1 (EX)
& Onboard-
Nation
Total Incidence
Due To Gasoline
Vapors (PUL, rat)
5.769
5.897
4.248
3.243
2.953
2.785
2.649
2.525
2.369
2.266
2.171
2.883
2.204
1.938
1.863
1.863
1.939
2.8%
2.192
2.289
2.209
2.289
2.289
2.289
2.289
2.209
2.289
2.289
2.209
'2.209
2.209
2.289
2.209
2.289
2.289
89.20
34.73
St. II (NO EX)
i Onboard-
Nation
Total Incidence
Due To Sasolir,e
Vapors (PUL. rat)
5.759
5.524
3.294
£.£56
1.982
1.848
1.813
1.779
1.721
1.698
1.659
1.630
1.681
1.572
1.545
1.545
1.694
2.000
2.182
2.209
2.209
2.289
2.209
2.289
2.289
2.209
2.289
2.289
2.209
2.239
2.209
2.209
2.209
2.209
2.209
79.18
29.19
St.IJII (EX)
& Onboard-
Nation
Total Incidence
Due To Gasoline
Vapors (PUL, rat)
5.759
5.718
3.726
2.557
2.259
£.136
2.339
1.594
1.754
1.650
1.574
1.494
1.424
1.358
1.308
1.300
1.375
1.533
1.629
1.645
1.546
1.646
1.645
1.546
1.546
1.546
1.646
1.646
1.646
1.S46
1.546
1.546
1.645
1.546
1.646
69.35
£9.34
                                            F-16

-------
Table F-5.  Service Station Incidence Due  To Baseline Vapors  (Plausible Upper Limit)  - Theoretical
Year


1986
1987
1988
1989
1990
1991
1992
1993
1994
1995
1996
1997
1998
1999
2000
2001
2002
2003
2004
2005
2006
2007
2008
Oaon
COCJ
2010
2011
2012
£013
2014
2015
2016
£017
2018
2019
2020
JM =
>V =
St.Itll (NO EX)
& Onboard-
Nation
Total Incidence
Due To Gasoline
Vapors (PUL, rat)
6.769
5.276
2.572
1.136
0.994
0.950
• 0.937
0.915
0.881
0.861
0.842
0.824
0.806
0.790
0.774
0.774
0.887
1.118
1.255
1.275
1.276
1.276
1.276
1.276
1.276
1.276
1.276
1.275
1.276
1.276
1.275
1.275
1.276
1.275
1.276
49.78
21.52
Year


1986
1987
1988
1989
1990
1991
1992
1993
1994
1995
19%
1997
1998
1999
2000
2001
2002
2003
£004
2005
2006
2007
2008
2009
2010
2011
2012
2013
2014
2015
2016
2017
2018
2019
2020
sua =
NPV =
Baseline

Total Incidence
Due To Gasoline
Vapors (PUL, lice)
4.10
3.99
3.88
3.77
3.70
3.52
3.57
3.53
3.43
3.38
3.34
3.29
3.24
3.19
3.15
3.15
3.15
3.15
3.15
3.15
3.15
3.15
3.15
3.15
3.15
3.15
3.15
3.15
3.15
3.15
3.15
3.15
3.15
3.15
. 3.15
116. 12
37.7-3
' St.II-Nfl*
(EX)

Total Incidence
Due To Gasoline
Vapors (PUL, aice)
3.80
3.37
3.23
3.12
3.06
2.99
2.95
2.91
2.84
2.80
2.76
2.72
2.68
2.54
2.60
2.50
2.60
2.50
2.60
2.50
2.60
2.50
2.60
2.50
2.60
2.60
2.60
2.50
2.50
2.60
2.50
2.50
2.60
2.50
2.50
95.46
31 . fiR
, St.II-Nfl*
(NO EX)

Total Incidence
Due To Gasoline
Vapors (PUL, aice)
3.57
2.92
2.78
2.57
2.62
2.56
2.53
2.49
2.43
2.39
2.36
2.32
2.29
2.26
2.22
£.22
2.22
2.22
2.22
2.22
2.22
2.22
2.22
2.22
2.22
2.22
2.22
2.22
2.22
2.22
2.22'
2.22
2.22
2.22
2.22
32.85
•37 A-5
St. II-Nfl
(EX)

Total Incidence
Due To Gasoline
Vapors (PUL, aice)
3.47
2.71
2.54
2.42
2.37
2.32
•p oa
2.2S
-O <7
Ut i l
2.14
2.11
2.87
o 84
i~* VT
2.81
2&1
» Vl
2.81
2.31
2.31
2.31
2.91
2.01
2.31
2.31
? 31
u. -01
2. 01
2.81
2.01
2.81
2.01
2.31
3 31
L.« -Si.
2.91
. 2.01
75.39
S*^ J f
                                                                                        25.16
                                         F-17

-------
      Table F-5.  Service Station Incidence Due To  Baseline Vapors  (Plausible Upper Limit) - Theoretical
ar




19BS
1987
1388
1369
1990
1991
1992
1993
1994
1995
1995
1997
1993
1999
2000
2061
2062
2803
2804
£005
2006
2007
2008
2009
2010
2011
2012
2013
2014
2015
2016
2017
2018
£819
2020
St.IHW
(NO EX)
Total Incidence
Due To Gasoline
Vapors (PUL, dice)
3.08
1.93
1.76
1.65
1.62
1.59
1.57
1.55
1.59
1.48
1.46
1.44
1.42
1.40
1.38
1.38
1.38
1.38
1.38
1.38
1.38
1.38
1.38
1.38
1.38
1.38
1.38
1.38
1.38
1.38
1.38
1.38
1.38
1.38
1.38
St.I-Nation
(EX)
Total Incidence
Due To Baseline
Vapors (PUL, aice)
4.18
3.89
3.57
3.36
3.30
3.23
3.19
3.14
3.06
3.02
2.97
2.93
2.89
2.85
2.80
2.80
2.80
2.80
2.80
2.80
2.80
2.80
2.80
2.80
2.80
2.80
2.80
2.80
2.80
2.80
2.80
2.80
2.80
2.80
2.80
St.I-Nation
(NO EX)
Total Incidence
Due To Baseline
Vapors (PUL, sice)
4.10
3.81
3.36
3.10
3.94
2.97
2.93
2.89
•2.81
2.78
2.74
2.70
2.66
2.62
2.58
2.58
2.58
2.58
2.58
2.58
2.58
2.58
2.58
2.58
2.58
2.58
2.58
2.58
2.58
2.58
2.58
2.58
2.58
2.58
2.58
St. II-Nation
(EX)
Total Incidence
Due To Baseline
Vapors (PUL, aice)
4,10
3.57
2.55
2.14
2.33
1.99
1.95
1.94
1.89
1.86
1.83
1.81
1.78
1.75
1.73
1.73
1.73
1.73
1.73
1.73
1.73
1.73
1.73
. 1.73
1.73
1.73
1.73
1.73
1.73
1.73
1.73
1.73
1.73
1.73
1.73
St. II-Nation
flPI (EX)-
Total Incidence
Dus To Gasoline
Vapors (PUL, aice)
4.10
3.96
3.52
£.81
2. 25
2.17
2.88
1.97
1.89
1.86
1.83
1.81
1.78
1.75
1.73
1.73
1.73
1.73
1.73
1.73
1.73
1.73
1.73
1.73
1.73
1.73
1.73
1.73
1.73
1.73
1.73
1.73
1.73
1.73
1.73
St. II-Nation
(NO EX)
Total Incidence
Due To Sasoline
Vapors (PUL, mice)
4.10
3.35
2.31
1.27
1.18
1.15
1.14
1. 1£
1.89
1. 88
1.86
1.05
1.03
1.82
1.88
1.88
1.80
1.88
1.38
:.8iB
1.00
1.38
1.38
1.03
1.80
1.80
1.06
1.08
1.38
1.88
1.00
1.39
1.63
1.83
1.00
9UM»
52.42




18.02
104.39




 34.44
36.68




32.29
67.63




24.30
78.11




26.22
42.67




17.37
                                                      F-18

-------
         Table F-5. Service Station Incidence Due To  Gasoline Vapors  (Plausible Upper Limit) - Theoretical
fear





19%
1987
1988
1989
1998
1991
1992
1993
1994
1995
19%
1997
1998
1999
2880
2081
2082
2883
2884
2885
2886
2887
2888
2889
2818
2811
2012
2813
2814
2815
2816
2017
2013
2019
2828
St.IHI-Nation
(EX)

Total Incidence
Due To Gasoline
Vapors 
-------
       Table F-5. Service Station Incidence Due To Gasoline Vapors (Plausible Upper Limit)  - Theoretical
ar *




1985
1987
1988
1989
1990
1991
1992
1993
1994
1995
19%
1997
1998
1999
2880
2001
2302
2003
2004
2005
2005
2007
2088
2009
2010
2011
2012
2313
2914
2015
2016
2017
2018
2019
2029
St.II-NA
(NO EX)
& Onboard
Total Incidence
Due To Gasoline
Vapors (PUL,*iee)
3.08
1.93
1.68
1.51
1.41
1.32
1.25
1.19
1.11
1.86
1.01
0.95
0.92
0.89
0.85
1.06
1.29
1.32
1.34
1.34
1.34
1.34
1.34
1.34
1.34
1.34
1.34
1.34
1.34
1.34
1.34
1.34
1.34
1.34
1.34
St. I (EX)
J Onboard-
Nation
Total Incidence
Due To Gasoline
Vapors (PUL.sice)
4.10
3.89
3.30
2.85
2.58
2.32
2.11
1.92
1.70
1.55
1.42
1.29
1.18
1.08
1.00
1.00
1.00
1.83
1.00
1.00
1.00
1.00
1.00
1.00
1.00
1.00
1.00
1.00
1.00
1.00
1.00
1.00
1.38
1.00
1.00
St. I (NO EX)
& Onboard-
Nation
Total Incidence
Due To Sasoline
Vapors (PUL,i raice)
4.19
3.81
3.39
2.59
2.32
2.85
1.85
1.67
•1.45
1.31
1.18
1.06
0.95
0.86
0.77
0.77
0.77
0.77
0.77
0.77
0.77
0.77
0.77
0.77
8.77
0.77
0.77
8.77
0.77
0.77
0.77
8.77
0.77
8.77
3.77
St.II-Nfl (EX)
J St. I (EX)
J Onbd-Nation
Total Incidence
Due To Gasoline
Vapors (PUL, aice)
3.47
2.63
2.17
1.86
1.72
1.57
1.46
1.36
1.24
1.15
1.08
1.01
3.95
0.89
0.34
0.91
0.98
0.99
1.00
1.00
1.00
1.30
1.38)
1.00
1.00
1.80
1.190
1.00
. 1.89
1.00
1.38
1.00
1.80
1.00
1.83
St.II-Nfl (NO EX)
i St. I (NO EX)
4 Onbd-Nation
Total Incidence
Due To Sasolins
Vapors (PUL,;sice)
3.38
1.84
1.40
L14
1.26
0.97
0.91
0.35
0.78
0.73
0.69
8.55
0.61
0.58
0.55
0.64
8.75
0.77
0.77
0.77
0.77
0.77
0.77
0.77
0.77
8.77
0.77
0.77
0.77
0.77
0.77
0.77
8.77
0.77
0.77
St. II (EX)
J Onboard-
Nation
Total Incidence
Due To Gasoline
Vapors (PUL,raice)
4.13
3.57
2.53
1.97
1.39
1.69
1.51
1.53
1.44
1.37
1.32
1.26
1.21
1.17
1.13
1.13
1.18
1.27
1.33
1.34
1.34
1.34
1.34
1.34
1.34
1.34
1.34
1.34
1.34
1.34
1.34
1.34
1.34
1.34
1.34
M>V*
46.62




15.85
52.25




23.52
                                                      44.54




                                                      21.38
43.26




17.56
31.13




12.41
54.08




21.05
                                                   F-20

-------
        Table F-5. Service Station Incidence Due To Gasoline Vapors  (Plausible Uoper  Liait) - Theoretical
Year     St.II (NO EX)
           J Onboard-
             Nation
        Total Incidence
        Due To Sasoiine
  St.IJII  (EX)
   S Qnboard-
     Nation
Total Incidence
Due To Gasoline
St.ISII '(NO EX)
   4 Onboard-
     Nation
Total Incidence
Due To Baseline
        Vapors (Pll, mice)   Vapors (PUL, nice)  Vapors (PIL, mice)
1986
1987
1988
1989
1990
1991
1992
1993
1994
1995
19%
1997
1998
1999
2888
2801
£00£
2803
2004
2005
2906
2007
2008
2009
2910
2011
2012
2013
.2014
2015
2016
2017
2018
2019
2020
SUM =
M>V =
4.18
3.35
2.00
1.25
1.15
1.12
1.10
1.08
1.04
1.02
1.01
0.99
0.97
0.95
0.94
0.94
1.03
1.21
1.32
1.34
1.34
1.34
1.34
1.34
1.34
1.34
1.34
1.34
1.34
1.34
1.34
1.34
1.34
, 1.34
1.34
48.@0
17.69
4.10
3.47
2.25
1.56
1.39
1.29
1.22
1.15
1.06
1.01
0.95
0.91
0.86
0.82
0.79
0.79
0.83
0.93
0.99
1.00
1.00
1.00
1.00
1.00
1.00
1.00
1.00
1.00
1.00
1.00
1.00
1.00
1.00
1.00
1.00
42.35
17.79
4.13
3.23
1.5S
0.69
0.60
8.58
0.57
3.55
•0.33
9.52
0.51
3.50
0.49
0.48
0.47
8.47
0.54
'0.68
0.75
0.77
0.77
3.77
0.77
3.77
3.77
3.77
0.77
3.77
0.77
3.77
3.77
3.77
3.77
3.77
8.77
33.18
13.04
                                              F-21

-------
Table F-6. Service Station Incidence  Due To Gasoline Vapors (Maxiaua Likelihood Estimate)  - Theoretical
iar




1986
1987
1968
1989
1999
1991
1992
1993
1994
1995
19%
1997
1999
1999
2000
2881
2082
2383
2004
2@65
2006
2007
2008

-------
Table F-6.  Service Station  Incidence  Due  To Sasoline Vapors  (Maxiaum Likelihood Estiaate)  - Theoretical
Year




1986
1987
1988
1989
1990
1991
1992
1993
1994
1995
19%
1997
1998
1999
ma
2081
2002
2003
£004
2005
20%
2007
2008
2009
2010
2011
2012
2013
£014
2015
2016
2017
2018
2019
2020
SUS =
NPV =
St.I-Nation
(NO EX)
Total Incidence
Due To Baseline
Vapors (MLE, rat)
3.354
3.583
3.156
2.908
2.853
2.790
2.753
2.717
2.643
2.607
2.570
2.533
2.496
2.460
2.423
2.423
2.423
2.423
2.423
2.423
2.423
2.423
2.423
2.423
2.423
2.423
2.423
2.423
2.423
2.423
2.423
2.423
2.423
2.423
2.423'
90.81
33.33
St.II-Nation
(EX)
Total Incidence
Due To Sasoline
Vapors (MLE, rat)
3.854
3.358
2.503
2.006
1.911
1.859
1.845
1.820
1.771
1.746
1.722
1.697
1.673
1.648
1.623
1.623
1.623
1.623
1.623
1.623
1.623
1.623
1.623
1.623
1.623
1.623
1.623
1.623
1.623
1.623
1.523
1.623
1.623
1.623
1.623
63.52
£2.83
St.II-Nation
fiPI (EX)
Total Incidence
Due To Gasoline
Vapors (MLE, rat)
3.854
3.718
3.303
2.635
2.146
2.038
1.951
1.850
1.771
1.746
1.722
1.697
1.673
1.643
1.623
1.623
1.523
1.623
1.623
1.623
1.623
1.623
1.623
1.623
1.623
1.623
1.623
1.623
1.623
1.623
1.523
1.623
1.523
1.623-
1.523
55.35
24.53
St.II-Nation
(NO EX)
Total Incidence
Due To Sasoline
Vapors (MLE, rat)
3.854
3.145
1.885
1.194
1.107
1.083
1.869
1.054
1.326
1.012
8.997
8.983
0.959
0.955
3.940
0.940
0.940
0.940
0.940
0.940
0.940
0.940
0.940
0.940
9.948
0.940
3.940
0.940
0.940
0.940
8.940
0.940
8.940
3.940
3.940
40.08
15.32
St.Iill-Nation
(EX)
Total Incidence
Due To Sasoline
Vapors (MLE, rat)
3.354
3.256
£.236
1.621
1.534
1.500
1.480
1.461
1.421
1.401
1.382
1.362
1.342
1.322
1.303
1.303
1.303
1.303
1.303
1.303
1.303
1.303
1.303
1.303
1.303
1.303
1.303
1.303
1.303
1.303
1.303
1.303
1.303
1.303
1.383
52.53
19.76
St. IJII-Nation
(NO EX)
Total Incidence
Due To Sasoline
Vapors (SLE, rat)
3. 854
3.304
1.474
3.564
0.590
0.577
0.570
3.562
•3.547
3.539
0.532
3.524
3.517
3.509
3.501
0.501
3.501
3.581
3.581
3.501
8.501
3.581
0.501.
0. 581
3. 531
e.501
0.501
3.581
3. 501
3.531
3.501
3.581
3. 531
3.581
0.501-
24.99
12.11
                                             F-23

-------
    Table F-6. Service Station Incidence Due To Gasoline Vapor?; (Maxiaua Likelihood Estimate)  - Theoretical
ar Onboard-Nation





1986
1987
1988
1989
1990
1991
1992
1993
1994
1995
19%
1997
1998
1999
2000
2091
2002
2£03
2004
2005
2006
2897
sm
2039
2010
2011
2312
2013
2014
2615
2315
2317
2018
2019
2020


Total Incidence
Dug To Gasoline
Vaoors(HE,rat)
3.854
3.751
3.395
3.066
2.682
2.547
2.343
2.150
1.950
1.802
1.669
1.548
1.442
1.344
1.258
1.258
1.258
1.258
1.258
1.258
1.258
1.258
1.258
1.258
1.258
1.258
1.258
1.258
1.258
1.258
1.258
1.253
1.258
1.258
1.258
St.II-Nfl*
(EX)
& Onboard
Total Incidence
Due To Gasoline
Vapors (MLE. rat)
3.567
3.170
2.848
2.571
2.358
2.171
2.015
1.875
1.711
1.597
1.494
1.399
1.316
1.240
1.172
1.207
1.247
1.255
1.258
1.258
1.258
1.258
1.258
1.258
1.258
1.258
1.258
1.258
1.258
1.258
1.258
1.258
1.258
1.258
1.258
St.II-Nfl*
(NO EX)
4 Onboard
Total Incidence
Due To Gasoline
Vapors (ICE, rat)
3.350
2.741
2.451
2.209
2.039'
1.874
1.744
1.627
1.489
1.394
1.307
1.228
1.158
1.894
1.037
1.130
1.235
1.251
1.258
1.258
1.258
1.258
1.258
1.258
1.258
1.258
1.258
1.258
1.258
1.258
1.258
1.258
1.258
1.258
1.258
St.II-Nfl
(EX)
4 Onboard
Total Incidence
Due To Gasoline
Vapors (MLE, rat)
3.259
2.545
2.253
2.024
1.880
1.739
1.629
1.529
1.410
1.328
1.254
1.185
1.126
1.070
1.020
1.117
1.227
1.249
1.257
1.258
1.258
1.258
1.258
1.258
1.258
1.258
1.258
1.258
1.258
1.258
1.258
1.258
1.258
1.258
1.258
St.II-Nfl
(NO EX)
4 Onboard
Total Incidence
Due To Gasoline
Vapors (MLE, rat)
2.889
1.817
1.581
1.415
1.328
1.242
1. 176
1.115
1.842
0.992
0.947
0.905
0.867
0.832
0.801
0.994
1.209
1.244
1.257
1.258
1.258
1.258
1.258
1.258
1.258
1.258
1.258
1.258
1.258
1.258
1.258
1.258
1.258
1.258
1.258
St. I (EX)
4 Onboard-
Nation
Total Incidence
Due To Gasoline
Vapors (MLE, rat)
3.854
3. 850
3.298
2.681
2.424
2.177
1.978
1.803
1.690
1.457
1.329
1.213
1.111
1.018
0.937
0.937
0.937
3.937
0.937
0.937
0.937
0.937
0.937
0.937
0.937
8.337
0.937
0.937
0.937
0.937
0.937
3.937
0.337
3.937
0.937
NPV*
60.09




25.16
55.51




22.23
51.74




19.87
50.23




18.85
                                                                                          43.78




                                                                                          14.89
49.87




22.03
                                                    F-24

-------
Table F-6.  Service Station Incidence Due To Gasoline Vapors  (Maxiraua Likelihood Estimate)  - Theoretical
Year St. I (NO EX) St.lI-Nfl (EX) St.II-Ntt (NO EX) St. II (EX) St. II (NO EX) St.IJII (EX)
1 Onboard- 4 St. I (EX) | St. I (NO EX) i Onboard- & Onboard- J Onboard-
Nation ( Onbd-Nation ft Onbd-Nation Nation Nation Nation
Total Incidence Total Incidence Total Incidence Total Incidence Total Incidence Total Incidence
« i?f I! SUB tSaS°lira °Ue T° SaS°Hne DueT° Sa5oiine D» T° Gasoline Sue To Gasoline
Vaoors(8LE,rat) Vapors (!t£, rat) Vapors (MLE, rat) Vapors (MLE, rat) Vapors (ME, rat) Vapors OLE, rat)
1986
1987
1988
1989
1990
1991
1992
1993
1994
1995
1996
1997
1998
1 fwi
1999
2080
mi
2002
2203
2084
2005
2006
2007
2008
2009
2010
2011
2012
2013
2014
2015
2016
2017
2018
2019
2020
5m =
!*V =
3.854
3.583
2. 583
2.428
2.176
1.934
• 1.739
1.564
1.378
1.230
1.106
0.992
0.894
0.804
0.726
0.725
0.726
0.726
0.726
0.726
0.726
0.726
0.726
0.725
0.726
0.726
0.726
0.726
0.725
0.726
0.726
0.726
0.725
0.726
0.726
41.83
20.08
3.259
2.474
2.043
1.751
1.612
1.477
1.371
1.275
1.162
1.084
1.014
0.949
0.892
0.840
0.793
0.852
0.918
0.932
0.937
0.937
0.937
0.937
0.937
0.937
0.937
8.937
0.937
0.937
0.937
0.937
0.937
0.937
0.937
3.937
0.937
40.53
15.50
2.889
1.727
1.317
1.073
0.992
0.914
0.852
0.797
8.731
0.686
0.545
0.607
0.574
0.543
0.516
0.505
0.784
0.720
0.726
0.725
0.725
0.725
0.726
0.726
0.726
0.726
0.726
0.725
0.725
0.726
3.726
0.726
3.725
0.726
0.726
29.24
11.65
3.854
3.358
2.419
' 1.847
1.587
1.586
1.508 •
1.438
1.349
1.290
1.236
1.186
1.141
1.099
1.061
1.861
1.134
1.194
1.248
1.258
1.258
1.258
1.258
1.258
1.258
1.258
1.258
1.258
1.258
1.258
1.258
1.258
1.258
1.258
1.258
50.79
19.77
3.354
3.145
1.876
1.176
1.083
1.8S?
1.332
1.013
0.980
0.962 -
0.945
0.928
0.911
0.895
0.880
0.880
0.965
1.139
1.242
1.258
1.258
1.258
1.258
1.258
1.258
1.258
1.258
1.258
1.258
1.258
1.258
1.258
1.258
1.258
1.258
45.09
15.62
3.354
7 1*C
ujt i_UD
2 l?o
*-• l&to
1.452
1.309
1.216
1.144
1.878
0.999
8.945
0.395
0.851
0.811
0.773
3.740
37&ffl
* i *rtf
0.784
3.873
0.927
3.937
a 537
v« Jw (
3.937
0.937
3.937
Q Q77
u. j&t
3.937
0.937
3.937
a 077
V. 3wl
a 
-------
Table F-6. Service Station Incidence Due To Gasoline Vacors  (Maxiaura Likelihood Estimate) - Theoretical
ar





1386
13S7
19BB
1989
1939
1991
1333
1993
1934
1995
19%
1997
1998
1399
203®
2001
2882
2883
2394
2035
28%
2887
2803
2809
£010
2811
2912
2013
2814
2315
2016
2817
2818
2813
2828
St.UII (NO EX)
4 Onboard-
Nation
Total Incidence
Due To Gasoline
Vaoors(i4£,rat)
3.854
3.884
1.465
8.647
8.566
8.547
8.533
8.521
8.581
8.438
8.479
8.463
8.459
8.458
8.441
8.441
8.585
8.636
8.715
8.726
8.726
8.726
8.725
8.726
8.725
8.726
8.726
8.726
8.726
8.726
8.725
8.726
8.725
8.726
8.726
Year





1986
1387
1388
1383
1998
1991
1992
1993
1934
1995
1936
1337
1998
1993
2888
2881
2882
2883
2884
2085
2886
2887
2888
2883
£818
2811
2812
2813
2814
2815
2816
2817
2818
2819
2828
                                  Baseline
         St.II-Nfi*
            (EX)
         St.II-Nfl*
          (NO EX)
          St.II-Nfl
            (EX)
                              Total Incidence   Total  Incidence   Total  Incidence   Total Incidence
                              Due To Gasoline   Due To Gasoline   Due  To Gasoline   Due To Gasoline
                              Vapors (lfl-E, sice)   Vapors (MLE, sice)   Vapors (!t£, nice)  Vapors («LE,aice)
               28.34   SUM =

               12.25   NPV =
2.75
2.59
2.61
2.54
2.43
2.44
2.41
2.37
2.31
2.28
2.25
2.21
2.18
2.15
2.12
2.12
2.12
2.12
2.12
2.12
2.12
2.12
2.12
2.12
2.12
2.12
2.12
2.12
2.12
 2.12
 2.12
 2.12
 2.12
 2.12
 2.12

78.14

25.37
2.58
2.27
2.18
2.18
2.86
2.01
1.93
1.%
1.31
1.88
1.86
1.83
1.88
1.78
1.75
1.75
1.75
1.75
1.75
1.75
1.75
1.75
1.75
1.75
1.75
1.75
1.75
 1.75
 1.75
 1.75
 1.75
 1.75
 1.75
 1.75
 1.75

64.91

21.39
2.48
1.%
1.87
1.88
1.76
1.72
1.78
1.68
1.63
1.61
1.59
1.56
1.54
1.52
1.58
1.53
1.58
1.58
1.58
1.58
1.50
1.53
1.50
1.58
1.50
1.58
1.58
1.58
 1.58
 1.53
 1.58
 1.53
 1.50
 1.50
 1.58

55.75

18.45
 2.34
 1.32
 1.71
 1.63
 1.60
 1.55
 1.54
 1.52
 1.48
-1.46
 1.44
 1.42
 1.48
 1.38
 1.36
 1.36
 1.3S
 1.36
 1.35
 1.36
 1.36
 1.36
 1.36
 1.36
 1.36
 1.36
 1.36
 1.36
 1.36
 1.36
 1.36
  1.3S
  1.36
  1.36
  1.36

 50.73

 16.93
                                                   F-26

-------
        Table F-6. Service Station Incidence Due To Gasoline Vapors (Maximum Likelihood Estimate)  -  Theoretical
Year
St. II-Nfl St
(NO EX)
:.I-Nation St.I-Nation St.II-Nation St.IHtotion St.II-Nation
ltJU (m Ex> (EX) flPi (EX) (NO EX)
Total Incidence Total Incidence Total Incidence Total Incidence Total Incidence Total
Due To Sasohne Due To Gasoline Due To Basoline Due To Gasoline Due To Saso'iw Due To
Vapors OLE, nee) Vapors (HLE, rice) Vapors OLE,riw) Vapors OLE, rice). Vapors «€£, rice) Vapors
1986
1387
1988
1989
1990
1991
1992
1993
1994
1995
19%
1997
1998
1999
•'MVVk
2000
2001
2802
2003
2004
3l>iV?
CB05
2006
2007
•3AAn
2008
2009
2010
2011
2012
2013
2014
2015
2016
2017
2018
2019
2020
2.87
1.30
1.19
1.11
1.09
1.07
1.35
1.04
1.01
1.00
0.98
0.97
0.%
0.94
0.93
0.93
0.93
0.93
0.93
0.93
0.93
0.93
0.93
0.93
0.93
0.93
0.93
0.93
0.93
0.93
0.93
0.93
0.93
0.93
0.93
2.76
2.61
2.40
2.26
2.22
2.17
2.14
2.12
2.0S
2.03
2.00
1.97
1.94
1.32
1.89
1.89
1.89
1.89
1.89
1.89
1.89
1.89
1.89
1.89
1.89
1.89
1.89
,1.89
1.89
1.89
1.89
1.89
1.89
1.89
1.89
2.76
2.57
2.25
2.08
2.84
2.00
1.97
1.95
'1.39
1.87
1.84
1.81
1.79
1.76
1.74
1.74
1.74
1.74
1.74
1.74
1.74
1.74
1.74
1.74
1.74
1.74
1.74
1.74
1.74
1.74
1.74
1.74
1.74
1.74
1.74
2.75
2.41
1.79
1.44
1.37
1.34
1.32 •
1.30
1.27
1.25
1.23
1.22
1.20
1.18
1.16
1.16
1.16
1.16
1.16
1.16
1.16
1.16
1.16
1.16
1.16
1.16
1.15
1.16
1.15
1.16
1.16
1.16
1.15
1.16
1.16
2.76
2.66
2.37
1.89
1.54
1.45
1.40
1.33
1.27
1.25 '
1.23
1.22
1.20
1.18
1.16
1.16
1.16
1.16
1.16
1.16
1.16
1.16
1.16
1.16
1 
-------
     Table F-6.  Service Station Incidence. Due To Gasoline Vapors (Maxiau* Likelihood Estimate)  -  Theoretical
Year
St.ISlI-NaUon
(EX)
Total Incidence
Due To Gasoline
Vapors (HE, aice)
St. Ifcll-Nation
(NO EX)
Total Incidence
Due To Gasoline
Vapors (MLE, mice)
Onboard-Nation
Total Incidence
Due To Gasoline i
Vaoors(MLE,niice) '
St.II-Nfl*
(EX)
& Onboard
fatal Incidence
St.II-Nfl*
(NO EX)
& Onboard
' Total Incidence
i Due To Gasoline
>) Vapors (MLE.ciice)
St.II-Nfl
(EX)
& Onboard
Total Incidence
Due To Sasoline
Vapors (MLEjfflics)
19B6
1987
1988
1989
1999
1991
1992
1993
1994
1995
19%
1997
1998
1999
2031
2302
2803
£034
2805
2036
2087
2889
2310
2311
2312
2313
2314
2815
2816
2817
2818
2019
2329
SUH*
WV =
2.76
2.33
1.58
1.16
1.18
1.37
1.36
1.35
1.32
1.00
3.99
8.98
8.%
3.95
8.93
3.93
8.93
8.93
• 8.93
8.93
8.93
8.93
3.93
8.93
8.93
8.93
8.93
0.93
8.93
0.93
8.93
3.93
3.93
3.93
0.93
37.61
14.16
2.76
2.15
1.36
8.48
8.42
8.41
8.41
8.43
8.39
8.39
3.38
8.38
8.37
8.36
8.36
8.36
8.36
8.36
8.36
8.36
8.36
3.36
8.36
8.36
3.36
8.36
8.36
8.36
8.36
8.36
3.36
8.36
8.35
8.36
8.36
17.98
8.68
2.76
2.69
2.43
2.28
2.81
1.82
1.63
1.55
•1.48
1.29
1.28
1.11
1.83
3.%
0.98
3.93
8.98
0.98
3.98
8.98
8.98
8.98
8.98
8.93
3.98
0.98
8.98
8.93
8.38
0.98
0.90
8.90
0.98
8.30
3.30
43.05
18.33
2.56
2.27
2.84
1.84
1.78
1.56
1.44 •
1.34
1.23
1.14
1.37
1.08
8.94
3.89
8.84
8.86
8.89
8.90
0.90
3.93
0.98
8.93
3.38
3.90
8.90
3.98
8.93
8.90
8.98
8.30
8.33
0.33
3.93
8.93
39. 84
15.33
2.40
1.96
1.76
1.58
1.46
1.34
1.25
1.17
1.37
1.00 "
8.94
8.88
0.83
8.78
3.74
0ni
.ol
8.88
3.30
0.30
3.30
8.30
3.30
3.30
3.90
3.90
8.38
a oa
V* J*f
3.98
0.38
8.90
3.38
8.93
0.98
3.93
3.98
37.37
14.23
2.24
1.82
1.61
1.45
1.35
1.25
1.17
1.10
1.31
3.35
8.98
0.85
8.81
8.77
3.73
a aa
v> UV
3.38
9.89
8.38
8.30
3.90
3. 30
0.30
3.30
3.90
3.90
3.98
3.90
8.98
' 3.30
3.38
a. 30
8.33
0.30
8.90
35.39
13.51
                                                     F-2S

-------
Table F-6.  Service Station Incidence Due To Gasoline Vaoors (Maxim™ Likelihood Estimate)  - Theoretical
                                                                           St.II-Nfl(NOEX)     St. I!  (EX)
Year        St.II-NA         St. I (EX)         St. I (NO EX)      St.II-Nfl (EX)

           4 Onboard           Nation            Nation        $ Onbd-Nation      8° Qnbd-Nation'         Nation
        ,otal Incadence   Total Incidence   Total  Incidence   Total  Incidence    Total  Incxdence    Total  I,
        Due To Gasoline   Due To  Sasohne   Due To Gasoline   Due To Gasoline    Due To Sasoli^    Due To Gasolin-
                 Mice)   VaporsfKLMice)   Vapors    Manors (MLE, .ice)   Vapors (HLE, lice)   Vapors (MLE,««)
1986
1987
1988
1989
1990
1991
1992
1993
1994
1995
19%
1997
1998
1999
<*WlflU&
2000
2001
2062
2003
2004
2005
2006
2007
2008
2889
2018
2011
2012
2313
2014
2015
2015
2017
2018
2019
2020
SUM =
KPV-
2.37
1.30
1.13
1.01
0.95
0.89
3.84
0.80
3.75
0.71
0.68
0.65
0.62
0.60
0.57
0.71
8.87
0.89
0.90
0.90
0.90
0.90
0.90
0.90
3.90
0.93
3.93
0.90
0.90
0.98
3.98
0.90
3.90
/ 8.98
0.90
31.37
10.67
2.76
2.61
2.22
1.92
1.74
1.56
1.42
1.29
1.15
1.04
3.95
0.87
3.30
0.73
0.57
0.67
8.67
0.67
3.57
0.67
0.57
0.67
3.57
0.67
3.57
0.57
0.57
0.57
0.57
0.57
3.67
0.67
3.67
3.67
3.67
35.16
15.33
2.75
2.57
2.38
1.74
1.55
1.39
1.25
1.12
•0.98
0.38
3.79
0.71
8.64
0.58
8.52
8.52
0.52
3.52
8.52
0.52
0.52
8.52
0.52
0.52
0.52
0.52
8.52
8.52
0.52
0.52
3.52
3.52
3.52
0.52
0.52
29.97
14.39
2.34
1.77
1.45
1.25
1.16
1.86
3.98 •
8.91
3.33
0.78
0.73
3.68
3.54
8.60
3.57
8.61
8.66
0.67
3.67
8.67
3.67
8.67
3.67
8.57
3.67
0.67
8.67
3.67
3.57
0.67
3.67
3.67
8.67
8.57
3.67
29.11
11.82
2.37
1.24
3.94
8.77
3.71
3.65
3.61
3.57
8.52
3.49 -
0.46
8.44
3.41
0.39
3.37
0.43
0.58
0.52
3.52
0.52
0.52
0.52
8.52
3.52
8.52
0.52
8.52 .
9.52
8.52
0.52
8.52
8.52
8.52
3.52 .
28.95
8.35
2.75
2.41
1.73
1..32
1.21
1.14
( fflQ
i. BO
1.03
3.97
3,92
0.89
8.85
8.82
a 70
VI 1 J
8.76
0.76
0.79
2.86
3.89
3.90
8.98
8.90
3QA
• yo
3.90
8.98
3.30
3.90
3.90
8.98
8.30
3.98
a. 90
3.93
3.30
3.98
35.39
14.17
                                            F-29

-------
     Table F-6.  Service Station Incidence Due To Baseline Vapors (Maxin.ua  Likelihood  Estimate) - Theoretical
Year
St. II (NO EX)
1 Onboard-
Nation
Total Incidence
Due To Gasoline
Vapors 
-------
                       Table F-7.  Service  Station  Incidence  Due  To  EDB  find  EDC - Theoretical
Year


1985
;3S7
1388
1389
1393
1331
1532
1933
1334
1955
i err
1535
; QC"?
i957
1333
193?
2888
2831
2282
2834
28:35
2235
2887
2288
28S9
2310
2311
2312
2313
2314
2315
--•il •< r
CtiO
£317
2318
2313
2222
Baseline

Total
Incidence Due
To EDS
3.38418
8.30365
3.30322
3.88281
3.33243
0.08215
3.C81S2
8. 80143
3.80124
8.88939
3. 30083
3.23858
3.80333
3; 80817
8.88817
3.S3317
3. 88317
8.88317
3. 30317
3. 88817
8. 08317
8.38817
S. 30317
3.30317
3. 83817
3.30817
3.S8817
3.88317
3. 28017
3. 33017
8. 02317
3.88317
3. 33017
3.80817
Si.II-Nfl*
/r*v\
(EX)
Total
Incidence Due
To EBB
3.38386
0.33337
3.80257
3.30231
8.33280
8.88177
3.33149
8.30122
3.88102
3.30881
3.88868
3.30854
8.38041
8.38827
3. 30814
3.88314
a. 38814
3.33814
g. 88814
3,33314
3.38814
3.80014
3.22814
3.88814
8.33814
•3.38814
3.83314
'3.38014
3.38314
3.83814
3.32814
3.33014
8. 38314
3.80814
8. 38314
St. U-Nfl*
(NO EX)
Total
Incidence Due
To EDB
3.88353
3.88265
3.80229
0.80197
8.88171
3. 80151
3.30128
8.80104
3.88087
0. 88873
3.32-858
8.00846
8.88835
3.88823
8.88812
0.30812
3.30312
3.88812
8. '30312
0.30812
8. 38812
8.33812
8.38812 •
8.38812
8. 88812
3.30812
8. S0812
8.38812
3.30012
2.30012
0.30812
•3.38812
3.38312
0.38812
St.II-Nfl
(EX)
Total
Incidence Due
To EDB
8.08352
8.30245
5.80288
8.80178
8.88154
3.88136
3.38115
0.80094
3.38873
3.88063
3.30352
3.89842
3.30831
3.80821
3.80810
3.88818
8.00818
3.88818
3.38810
3.00013
0.30818
3.88813
3.SS813
' -3.83013
3. 38813
3.88818
3.38813
3.88810
8.28818
3. 8331 C
3.83910
3.30810
3.38818
3. £091 3
3.38818
St.II-Nfl
(NO EX)
Total
Incidence Due
To EDB
3.80312
0. 00175
3.30144
3. 80121
8.08185
3.28893
. 3.38875
3. 88864
8.80854
3.80843
0.80836
3.38829
3.88821
3.88814
3.88807
3.38887
8. 03807
3. 38987
3. 88887
8.20387
8.88087
8.30087
3.83387
3. 88897
3.80387
3. 88387
3.88337
8.88887
3.88887
3.30887
3.38837
3. 38387
3.38037
3. 88887
3.38387
St. I -Nat ion
(EX)
Total
Incidence Due
To EDB
3.30418
3.30355
a. 38238
3.38253
3.30213
3. 33133
3.38164
3.38134
3.30112
8.33889
3.38874
3.88868
3.38345
3.30038
8.88015
3.30015
8.38015
3.88815
3.38015
3.38815
8.88815
3.38015
3.38315
3.30015
a. 33815
8.88815
3.28015
3. 88815
3.88815
3.38815
3.38815
3.28815
3.38815
3.88815
3.88015
St.I-Nation
(NO. EX)
Total
Incidence Due
To EDB
3.80418
0.30358
8.38232
3.30235
3.88283
3.80179
0.88152
3.30124
3.S8134
8.38883
8.83369
8.88855
3.08041
3.28028
3.38814
8. §3014
3.38814
3.38814
3.88014
3.38814
3.38014
3.38814
3.83014
3. S3314
3. 38814
8.88814
•3.38014
3. 33814
5.38014
3.33814
3.38014
3.33314
3.83014
3. 30814
3.88814
CI:« -
iDu:'; ™
              3.3238




              3.3193
3.3252




3.5153
2.3217




•3.8143
3.3159




8.3133
3.3144




3.3893
•2.3275
                                                                                         3.3181
3.32S1




3.3173
                                                   F-31

-------
                     T53i2 r-7. -Service Station Incidence Due To EBB  And  EDC  -  Theoretical
tear
St. II-N'ation
    fEX)
St.II-Nation
  flP! (EX)
St.II-Nation  St.IJII-Nation St.UII-Nation Onboard-Nation
   
-------
                      Tasle F-7. Service Station Incidence Due To EDS find EDC - Theoretical
YH" ™ S';<^ *^* ' «.1^» B>
Totai Totai Total Total Total
Incidence Due Incidence Due Incidence Due Incidence Due Incidence Due Incidence Due Incidence Due
'° EDB '° m To ™ T° EDB To EDB To EDB To EDB
13SS
1987
13S8
1333
1332
1331
.332
1333
1334
1355
15%
1337
1938
1339
2328
2331
2333
2034 .
-•*,/» fj. 17
c«95
2337
£328
2889
2813
"21 '
^Vl A
£312
2813
2314
2315
£315
£317
231S
£319
2023
8. '28353
3.88255
3.38229
3.83197
0.33151
3.88123
8.38104
3.S8387
0.33373
8.38853
8.33346
3.28335
3.30323
3. 88312
3.30814
0. 08315
3.38316
8. 33317
3. 30817
3.33317
3.83817
3.38817
3.33817
8.80317
•> »*rtfiH7
Vm Vi&GlI
3.83317
3.33317
8.33817
3.83817
3.88817
3.38817
8.38017
8.20817
3. 08817
3.28352
8.33246
3.88288
8.88178
3.30154
0.33135
8.88115
8.33394
3. 38879
3.30353
3.38852
3.83842
3.38831
3.38821
8. 33818
3.33313
d. ®SiS
3.38816
2.88817
.3.33317
0.88817
3.33817
8.38017
3.33817
8.88017
3{%fin t -7
.33817
3.J8317
8.80817
8. 38017
8.80017
3.30817.
3.88017
8.08817
8.80017
3. 83017
8.38312
0.38175
3.08144
8.88121
3.88185
3.88893
3.28879
3.80354
8.38854 -
3.88343
8.83836
3.88829
3.08821
3.88814
8.88887
3.88811
3.33(315
0.83815
3.33817
3.38317
3.80317
0.88817
8.23017
8.88317
0.88817
8. 80317
3.38817
3.83817
0.00017
0.33817
3.30017
3.38317
3.33017
3.88817
-3.33317
0.08418
3.88355
3.88298
3.83253
8.38213
0.88193
0.00164
3.33134
0.00112
8.88889
8.38374
3.38358
3.03845
8.08838
8.80815
0.88315
0.88815
3.83815
8.83815
0.88815
8.88815
3.88815
0.88815
8.88315
8.88315
8.38315
0.30815
8.88015
0.38815
8.80815
3.33315
8.38815
8.88815
9.30815
3. £381 5
3.28418
8.30350
3.38282
3.38235
3.33203
8.38179
3.88152
3.88124
3.88855
8.38841
0.30028
3. 88814

3.83814
3.83014
8.38814
3.38314
3.33014
3.38314
3.88814
3.38814
3.83814
0.38814
3.88014
8.30814
8. -£8814
3.33814
3. £3314
3.80814
<3.£0014
3.83814
3.33352
3.83239
3.08191
3.38158
8. 30137
3.33121
3.88182
8.33384
3.38878
3.38856 *
8.38847
3.38837
3.08028
3.88019
3 iMfliW
v* Wvvj
3.80812
3.30014
3.38815
3. 33815
8.80815
3.38815
8.38815
3.88815
8.38815
3.08815
0.88015
3.88815
3.38815
0.38815
8.38815
3.-S3015
3.38815
8.88815
03fflffli^
. vvviu
8.83815 .
3.38312
3.08167
8.38123
3.33897
3.130884
3.88074
-3.38863
3.23851
3.38343
0.88834
8.83828
3.38823
8.S8817
8.38811
3aat-ȣ
o tiwvb
a^iifloq
• tf'QV'Qj
8.88013
3.38014
Sj 30014
V« WVA"
8.38814
3.38314
3.38814
3.38814
8.38814
8.38314
3.80814
3. 38814
3.08014
3.08814
3.33814
3.38814
8.88814
8.03014
SUM =
3.8227




8.3144
3.0211




3.3134
3.3152




3.3138
a. 3276




3.*
3.8261




0.3173
                                                                                        3.3194




                                                                                        0.8125
3.3148




3.30S8
                                                 F-33

-------
   'Table F-7. Service Station Incidence Due To EDB find EDC - Theoretical
ar



St. II (EX) St.
i Onboard- i
Nation
Total
II (NO EX) St.ISI! (EX) a.IHI(NOEX)
Onboard-
Mat ion
Total
Incidence Due Incidence Sue

IS8S
19S7
1983
1989
199-3
1991
1992
1993
19%
1S95
19%
1997
1958
1999
283$
2831
£232
£?32
2234
2fc«5
&&
£837
£388
sm
5318
2*11
2*212
£013
2814
8215
seis
2817
2®18
S319
£2£3
To EDB
3.80418
9. 20326
0.30218
0.83156
8.90138
0.33115
9.33898
0.03083
0.33067
0.30853
3.88844
8.88035
0.e3027
8.22313
0.33809
8.80339
8.&811
8.23814
8.83016
0.WB17
0.03317
8.28017
3.83017
8.80317
0.82(317
8.83317
0.28917
0.83817
8.33317
8. 38317
0.28917
8.38017
8.^817
8.30317
3.22317
To EDB
3.38418
0.83385
8.30163
3.83098
8.30374
0.03365
3.08855
0.83345
9.30038
0.83038
•3.08025
3.38020
8.00015
3.03310
0.30905
3.88835
8.83003
0.33813
3.38016
0.30317
3.83017
3.33817
3.38017
0.88017
3.08017
0.88317
8.83817
8.33317
3.80017
0.83017
3.00817
8.88017
8.80917
0.88317
8.38817
i Onboard-
Nation
Total
Incidence Due
To EDB
8.38418
0.38317
3.30194
3.83128
8.30106
3.88094
8.880B0
8.83865
8.83354
8.38843
8.88826
0.30329
8.33822
0.30814
0.80387
0.38807
8.30889
8.88812
0.88815
8.33815
8.08015
3.88315
8.80015
8.08015
8.88015
0.88015
0.38015
3.38815
8.38015
8.88015
8.38015
3.80015
3.88015
3.38815
3.38315
J Onboard-
Nation
Total
Incidence Due
To EDB
3.33418
3.80292
3.33130
8.38352
3.80841
8.88036
3.38338
0.812025
3.38321
8.33017
8.80814
3.08811
8.88808.
0.08086
0.88803
8.08883
0.88805
8.80010
0.88813
3.812314
0.39814
8.33014
3.33814
3.33814
3.33814
3.30814
9.S3014
3.83014
0.30014
3.33814
3.30814
8.83814
3. 2001 4
8.88814
0.^3014
Baseline

Total
Incidence Due
To EDC
8.30540
0.08471
3. 3341 6
3.80363
3.00313
8.03277
• 3.30235
0.00192
0.80160
0.28128
3.80107
8.00085
0.30064
3.38843
3.80021
8.00021
8.00021
0.00021
3.80021
8.00021
3.88021
8.00021
8.88321
3.08021
8.88021
8.38821
3.38321
0.83021
3. 83821
3.00821
3.38821
3.88821
3.33821
3.38821
3.83821
St.II-Nft*
(EX)
Total
Incidence Due
To EDC
3.38498
3.883S5
3. £3344
3.38238
3.38256
3.33228
3. '58193
3.33158
3.33131
r. 88185
8.33088
0.88370
3.38353
3.80035
3.33818
8.88818
3.30018
3.33018
0.3801'8
8.88818
3.38818
0.38018
8.38318
8.88818
3.33013
3.S8018
3.38813
3.83818
3.30818
3.33018
8.38818
3.8081B
8.83818
3.33813
3.83318
3.3211
3.3166




3.3113
3.3139




3.3128
3.31.35




8.3397
3.8384




3.3249
3.3322




3.3211
                                  F-34

-------
"able F-7. Service Station  Ircidence Due To EDS find EDC - Theoretical
Year

1985
1937
1988
1989
1593
1991
1592
1SS3
13S4
1555
13SS
1397
1998
1999
2383
2881
£382
£283
£334
£385
2306
£-337
338
£313
2311
£312
£813
2314
£315
£31S
£317
£313
£215
£323
m =
IPV = .
St.IJ-Nfi*
(NO EX)
Total
Incidence Due
To EBC
3.83468
3. 33343
8.38295
3.83255
3.802E8
3. 30495
3.30155
3.20135
3.88112
8.33398
8.33875
8. £0368
8.30045
3. 23838
3.00815
0.80315
3. 28015
8.82315
3. 33015
8.3S015
-3. £2815
3.30815
3. 3S015
3.33315
3. 33315
3.33315
8.30815
3. 33815
0. 23815
8.^015
2.33315
:. i'280
3.3135
St.II-Nfl
(EX)
Total
Incidence Due
To EDC
0.08455
3.30317
0.38269
3.30238
8.80199
3. 30175
0.30149
3.38122
8.38181
3.38881
3.88858
8.30054
8.38041
0.30027
0.00014
0. 30814
8.33814
8. 33814
8.30814
3.30814
8.38314
8.88814
8.88814
3.08814
8.88014
3. 80814
3.03314
0.30014
3.38814
3iTAAi ft
.!ffio814
3.3S014
3.33014
3.33314
3.30814
8.38214
.3.3257
3.3171
St. II-Nfl
(NO EX)
Total
Incidence Due
To EDC
8.S3483
3.08225
3.38185
3.38157
0.83136
3.38120
0.38181
3.00883
8.38069
3.33855
8.83046
0.38837
8.888£8
3.08818
8.30339
0.30809
0. 30639
3.30389
3.83809
0.30309
3.82339
8.30099
3. 33389
3.08809
3.8S339
3.38089
3.83339
0. 38389
8. 022129
3. 30809
3.32339
3.38309
3.33339
0.30039
3.33389
3.3166
3.3127
St.I-Nation
(EX)
Total
Incidence Due
To EDC
3.38548
3.08468
3.80385
0.30325
8.38282
0.30253
0.38211
0. S3 173
0.S0144
0.30115
3.88096
0.33377
3.0885S
0.38333
8.03819
8.83319
0.00019
3.33819
8.33019
3.33319
8.33819
8.33813
3.88019
. 0.30813
0.88019
3.33*19
3,33319
3.33319
3.80819
0.38019
8.33819
3. '2031 9
0.3801-9
0,33319
0.30019
3. 3356
3.3233
St.I-Naiion
(NO EX)
Total
Incidence Due
To EDC
2.03548
3.88452
8.30364
3. 33383
3.83262
3.38231
3.38195
3.03163
3.30134
3. 08137
3.80839
3.33071
8.80053
8.08836
8.83818
3.33813
8.83818
3.38818
3.33318
3.30018
6.33818
8. 38818
3.33818
3.33318
8.30018
3.33013
3.33318
3.38318.
2.38318
0.33818
3.38818
3.23318
3.80818
3.00818
3.33818
3.3337
3. 3223
St.Il-Nation
(EX)
Total
Incidence Due
To EDC
3.33540
3.33423
0.38282
3.08231
3.88168
3.08149
8. 33125
3.03183
8.33886
3. '38059
8.30357
3. 38846
8.88034
8.08323
3.88811
3.33011
0.03811
3.03311
8.33811
3.00011
3.33011
8.33811
3.88811
8.30011
3.38311
0.00811
0.33811
0. 3001 1
3.33011
3.33811
3.38811
3.33311
8.30011
3. 03811
3.38311
3.3254
_ - , .,_
St.II-hiation
flPI .(EX)
Total
Incidence Due
To EDC
0.33540
3.33467
3.30375
0.30267
•3.03198
3.38163
3.03134
8.20185
3.38085
3.03069
3.38057
8.33346
8.08034
8.33823
3.38011
3.33011
8.88011
3.33011
3.03811
3.38311
3.68811
8.30011
8.00011
3.33311
3.83011
3. 3001 1
3.38011
3.30311
3.88011
0.30311
3.381311
0.20011
3.83811
3.33811
3.33011
3. 3233
a *3
-------
                     Taale "-?. Service Station Incidence Due To cuB find EDC - Theoretical
Yiar    .St. !I-Natiort  St. ISII-Naticn St. 1411-Nation Onboard-Nation
           JNC EX)         (EX)           (NO EX)
            Total
Total
Total
                                                         Total
St. II-Nfl*
06
C*. 33806
8.3188
3.8143
3.S0540
8.S0489
3.88251
8.80165
3.38137
8.80121
•3.80183
3.33084
3.88878
3.80356
3.08047
8.80037
9.08028
8.88819
3.88009
3.90889
3.30009
9.08889
3.08009
9.33889
0.88009
9.00089
8.08009
3.33889
3.08009
3.3S309
3.00009
8.38089
0.88809
8.30009
3.38809
8.3-3809
8.88809
3.38089
0.88809
2.2225
8.3154
9.88543
0.38377
8.88167
3.38067
9.88052
8.30345
8.08839
3.22032
8.88827 '
9.38821
0.80018
8.30014
0.08911
3.28307
0.80804
3.98384
3.08904
3.88084
0.88004
8.30384
0.08004
3.0S084
9.08804
3.32384
9.08804
3.3-3084
0.00084
8.28084
0.80004
3. £8304
3.03004
8.88884
0.00004
•3.30004
9.S8004
2.3149
0.3122
0.^543
3.08471
^0. 88416
3.558363
9.28313
8.03277
3.38235
8.08192
3.03163
8.00128
0.00187
8.00085
0.IS364
0.2(2043
0.128021
0.80921
0.33821
0.8C021
3.S3321
8.88021
•3. 63821
3.30821
3.80321
3.80321
0.S021
0.08021
0.30321
8.08321
0.83821
3.33321
•3. -331321
8.00821
3.33021
0.0S821
@.§0021
8. -3384
•3.0249
8.28498
8.303%
0.38344
3.30298
0.08258
3.30228
3.30193
3.93158
9.88131
3.00185
8.38088
8.08070
8.88853
3.00035
8.08018
0.30819
9.38021
8.30321
3.39021
3.38021
0.00021
0.150021
0. 20921
•3.20821
0.08021
8.83021
3.80821
0.08921
3.08021
8.08821
3. 00021
3.88821
3. 130021
3.08021
3.88021
3.8338
3.0212
3.30463
3.38343
3.802%
3.88255
0.38220
3.30195
8.38165
3.08135
3.88112
0.30898 "
8.08875
3.08868
3.03845
8.88830
3.00015
8.00818
3. -38021
3.88021
0.08021
8.88021
0.00821
0.80021
9.30021
3.88021
3.08821
3.80021
3.23021
3.00021
0.20021
3.30821
3.33«21
•3.32021
0.08021
9.30021
0. 120021
0.3293
8.0136
8.00455
3.00317
0.33269
0.30230
0.03199
3.88176
3.00149
0.83122
8.38181
0.83881
0.88063
3.88054
8. '20341
3.88027
8.88014
3.00017
0. '38920
3.88021
8.80021
3.08821
0.08321
8.08021
3.08821
3.38021
0.30021
3.30021
3.&J021
0.00021
0.S0021
3.88021
3.30321
0.30021
3.38921
3.38021
0. 80021
3.2272
•3.3173
                                                    F-36

-------
Table F-7.  Service Station Incidence Due To EDB find  EDC -  Theoretical
Year





1386
1987
1388
1989
1998
1391
1992
1393
1994
1395
1996
1397
1998
1993
mm
2881
£302
2383
£"234
2885
2886
2307
2008
2089
•313
2811
2812
2813
314
2815
2216
2817
2318
2013
£323
SUM =
NPV =
St.IIHW
(NO EX)
S Critoard
Total
Incidence Due
To -EDC
3.38433
3.38225
0. 03186
0.00157
a. eei3&
3.30120
3.88181
3.38083
3.2-0859
3.82855
3.38846
3.38337
3.33028
a. 08818
3.38889
8. 08014
3. 30020
0.08821
0:08821
3.08821
0. 08021
0.88021
0.80621
2.03021
3. 30021
0.00021
0.38821
•2.08821
3.50021
3. 38021
3.28021
0. 3882!
3. 08021
3.e8321
0.30021
3.0209
0.0132
St. I (EX)
J Onboard-
Nation
Total
Incidence Due
To E33C
0.30549
0.88468
0.00385
0.30325
8.09282
0.30258
3.88211
0.88173
8.08144
0.38115
8.00095
8.08077
0.08858
8.08838
8.88819
0.88813
0. 08013
3.88019
0.88819
3.88019
0. 00819
0.08019
0. 30013
3.88813
8.88813
3.88019
3.30013
8.08319
0,30313
8.08019
8.00019
3.38319
8.03819
3.38019
3.00819
8.5356
3.0223
St. I (NO EX)
I Onboard-
Nation
Total
Incidence Sue
To .EDC
0.83548
8.38452
0.88364
0.38383
3.88262
3.38231
8.881%
3.00163
8.08134
3.88137
8.38889
8.88871
0.88853
3.38836
3.88813
8.90018
0.88013
8.38018
0. 28013
3.-38818
0.88818
8.88818
8. 33813
8.88818
0.80818
0.28318
0. 00813
8.38818
3.38318
3,38318
'3. 30013
3.88818
3.38813
3.38318
3.30318
3.3337
3.3223
                                 J  St. I  (EX)
                                4 Onbd-Nation
                                   Total
                                Incidence Due
                                   To EDC
                                     3.08455
                                     8.88389
                                     0.33247
                                     0.88284
                                     8.88177
                                     0.88156
                                     8.38132
                                     0.33898
                                     3.38872
                                     8.83848
                                     8.03836
                                     0.88824
                                     0.88812
                                     0.88015
                                     •3.03818
                                     0.38319
                                     •3.03819
                                    3.08819
                                    3.38819
                                    3.38819
                                    8.38019
                                    3.33019
                                    3.88819
                                    •3.03013
                                    3.33319
                                    3.33019
                                    3.88019
                                    3.33819
                                    3.83819
                                    0.88819
                                    3.38819
                                    0.00819

                                    3.3251

                                    a. 3162
[-NA (NO EX
,1 (MO EX)
'.fad-Nation
Total
:dence Due
o EDC
8.23433
0.38215
3.83159
8.88125
8.33188
3.03395
3.88381
3.88866
0. 08-355
3.38844
3.03837
8.38829
3.88822
8.08815
8.38887
3. 38012
0.88017
8.38317
0.30818
8.88318
3.38818
0.88318
3.88818
8.88818
3.88813
8.38318
8.08813
0.38013
3.00818
3.38818
3.03013
3.33318
3.83013
3.83318
8.08818
3.8181
0.8115
St. 1 1 (EX)
& Onboard-
Nation
Total
Incidence Due
To EDC
8.88540
3. 38428
3.83282
3.80281
3.83168
3.S8149
3.28125
3.30183
3.38885
3.88869
0.33857
3.38846
0.38834
0.38023
3.08811
3.38811
8.88814
3.38018
3.03821
3.88821
3.30821
0.30821
8.38021
3.88821
8.80821
3.30821
3.38021
8.30321
3.38821
3.38321
0.80021
3.38821
3.38821
3.83821
3.08821 •
3.3272
0. 0131
St. II (NO EX)
4 Onboard-
Nation
Total
Incidence Due
To EDC
0.38548
3.80393
8.03210
0.38117
3.08395
3.38884
3.39871
3.38858
8.38848
8.38039
0.38032
3.80826
3.33819
3.38013
3.38886
0.38885
8.08818
3.08817
3.00321
0.80821
8.38821
8.08821
0.88321
3.83321
3.03821
0.83821
3.38821
8.03021
3.03321
0.33321
8.00021
8.33821
8.83821
3.30021
8.0902!
0.3215
0.8146
                             F-37

-------
                   • Table F-7.  Service Station Incidence Due To EDB find EC - Theoretical
Year    Si. III!  (EX)  St. IHI (NO EX)





1366
1387
1388
1333
1233
1991
1392
1993
1994
1993
1926
1997
1998
1999
2«08
2831
2002
2283
sew
£895
2836
2837
2e0a
i239
2818
1311
2812
£313
2814
2015
£315
2817
ftift
3®19
£828
&M*
NPV =
S Onboard-
Nation
Total
Incidence Due
ToEDC
3.23543
3.C8A03
3.^51
§.30165
3.00137
3.^121
9.^103
•3. 38884
9.20070
•3.300S&
3.8^47
3. 00837
a.eeesa
3.?^319
a.e-eera
0.03^9
3.8^12
ia.e-2816
3.80019
9.^319
8.89919
8.^819
@.0eei9
8.88819
3.88819
3.83019
8.88819
0.03319
8.33319
3.33819
8.28819
8.*819
8.38819
8.83819
3.28319
0.3244
3.81S5
4 Onboard-
Nation
Total
Incidence Due
To EDC
3.38548
8.83377
3.80167
8.80867
0.88052
0.20846
'3.03039
8.80832
8.08827
8.20821
0.88018
0.80314
8.08011
8.08807
8.08884
8.08834
0.03087
0.00313
0.00817
0.08818
0.08018
0.00018
8.03018
0.00018
0.00018
8.08318
0.08018
0.00018
3.00318
8.00818
0.88018
3.00318
8.88018
0.00318
0.88018
8.3175
3.3125
                                                     F-38

-------
                Table F-8.  Self-Ssrvice Incidence Due To Benzene - Theoretical (Selfsine.wks)
  Year
        1986
        1987
        1988
        1989
        1990
        1991
        1992
        1993
        1994
        1995
        19%
        1997
        1998
        1999
       20191
       2802
       2004
       2005
       2006
       2007
       2808
       2009
       2010
       2011
       2012
       2013
       2014
       2015
       2016
       2017
       2013
       2019
      2020
Baseline

Incidence
Due To
Benzene
(Theoretical)
4.42
4.24
4.08
3.92
3.80
3.68
3.59
3.50
3.38
3.30
3.23
3.16
3.10
3.03
2.%
2.96
2.%
2.%
2.%
2.9S
2.96
2.95
2.%
2.96
2.%
2.96
2.96
2.96
2.96
2.36
2.96
2.96
2.%
2.36
2.%
St.II-Nfl*
(EX)
Incidence
Due To
Benzene
(Theoretical)
4.25
3.32
3.74
3.59
3.48
3.37
3.29
3.21
3.89
3.02
2.96
2.98
2.83
2.77
2.71
2.71
2.71
2.71
2.71
2.71
2.71
2.71
2.71
2.71
2.71
2.71
2.71
2.71
2.71
2.71
2.71
2.71
2.71
2.71
2.71
St.II-Nfl*
(NO EX)
Incidence
Due To
Benzene
(Theoretical)
4.18
3.78
3.61
3.45
3.35
3.25
3.17
3.09
2.98
2.91
2.85
2.79
2.73
2.57
2.61
2.51
2.51
2.61
2.51
2.61
2.61
2.61
2.61
2.61
2.51
2.61
2.61
2.61
2.51
2.61
2.61
2.61
2.61
2.51
2.61
St. II-Nfl
(EX)
Incidence
Due To
Benzene
(Theoretical)
3.98
3.37
3.17
3.01
2.92
2.83
2.76
2.69
2.60
2.54
2.49
2.43
2.38
2.33
2.28
2.28
2.28
2.28
2.28
2.28
2.28
2.28
2.28
2.28
2.28
2.23
2.28
2.28
2.28
2.28
2.28
2.28
2.28
2.28
2.28
St.II-Nfl
(NO EX)
Incidence
Due To
Benzene
(Theoretical)
3.81
3.134
2.85
2.71
2.62
2.54
2.48
2.42
2.33
2.2B
2.23
2.19
2.14
2.89
2.05
2.85
2.35
2.05
2.35
2.35
2.S5
2.35
2/35
£.85
2.85
2.05
2.35
2.05
2.05
2.65
2.35
2.35
2,85
2.35
2.95
SUM =
NPV =
112.61

 37.98
133.35

 34.95
99.66

33.76
87.33

29.85
73.58

27.36
                                              F-39

-------
              Table F-8.  Self-Service  Incidence Due To Benzene - Theoretical (Selfsine.wks)
Year
 sua =


\




198S
1987
1933
1989
1999
1991
1992
1993
1994
1995
19%
1997
1998
1999
2000
2001
2332
2003
2004
2005
2006
£007
2208
2009
2018
2011
£012
2013
2014
£015
2016
2817
£918
•019
5320


St.II-Nation
(EX)

Incidence
Due To
Benzene
(Theoretical)
4.42
3.57
2.16
1.38
1.23
1.19
1.15
1.13
1.09
1.07
1.35
1.03
1.93
0.98
0.96
0.%
8.96
0.96
0.%
0.%
0.96
0.%
0.96
0.%
0.95
0.9S
0.S5
0.96
0.26
0.95
0.96
0.95
0.95
0.%
0.95
42.63
18.32
St.II-Nation
flPI (EX)

Incidence
Due To
Benzene
(Theoretical)
4.42
4.18
3.58
2.41
1.62
1.47
1.33
1.18
1.09
1.07
1.05
1.03
1.88
0.98
0.%
0.96
0.%
0.95
0.96
0.96
0.96
0.96
0.95
0.95
0.95
0.%
3.95
0.95
0.95
2.55
0.96
0.95
9.95
3.95
0.%
46.47
21.03
St. II-Nation
(NO EX)

Incidence
Due To
Benzene
(Theoretical)
4.42
3.28
1.33
0.29
0.19
0.18
0.18 .
0.17
0.17
0.16
0.16
0.16
0.15
0.15
0.15
0.15
3.15
0.15
0.15
3.15
0.15
0.15
0.15
0.15
3.15
0.15
3.15
0.15
0.15
0.15
0.15
0.15
0.15
3. 15
S.15
14.25
9.37
Onboard


Incidence
Due To
Benzene
(Theoretical)
4.42
4.24
3.59
3.18
2.75
2.35
2.83
1.72
1.41
1.17
0.97
0.79
0.62
0.47
0.34
8.34
3.34
0.34
0.34
•3.34
0.34
0.34
8.34
3.34
3.34
0.34
9.34
0.34
3.34
0.34
0.34
3.34
0.34
3.34
3.34
37.00
22.08
St.II-Nfl*
(EX)
w/ Onboard
Incidence
Due To
Benzene
(Theoretical)
4.25
3.92
3.39
2.91
2.52
2.16
i.86
1.58
1.29
1.83
0.89
0.72
8.57
0.44
0.32
0.33
0.34
0.34
3.34
0.34
3.34
3.34
3.34
0.34
0.34
3.34
3.34
0.34
3.34
3.34
3.34
3.34
0.34
3.34
3.34
34.71
20.53
                                              F-40

-------
                Table F-8,  Self-Service Incidence Due To Benzene - Theoretical (Self sine, wks)
  Year
        I9B5
        1337
        19-38
        1983
        1590
        1331
        1932
        1333
        19-34
        1335
        13%
       1337
       1398
       1939
       2800
       2001
       •2802
       2203
       2034
       2005
       283S
       2607
       2208
       2303
       2810
       2011
       2012
       2813
       2314
       2015
       2316
       2017
       2318
       2013
SUM =
St. II-Nfi*
(NO EX)
w/ Onboard
Incidence
Due To
Benzene
(Theoretical}
4. IS
3.78
3.26
2.80
2.43
2.28
1.79
1.52
,1.25
1.34
8.86
8.70
0.55
0.42
0.31
0.32
3.34
0.34
0.34
0.34
0.34
0.34
0.34
0.34
0.34
0.34
0.34
0.34
0.34
0.34
3.34
3.34
0.34
3.34
0.34
33.73
13.90
St.II-Nfl
(EX)
w/ Onboard
Incidence
Due To
Benzene
(Theoretical)
3.98
3.37
2.87
2.45
2.12
1.82
1.56
1.33
1.09
0.31
0.75
0.61
0.4S
0.37
0.27
0.38
0.33
0.34
0.34
0.34
0.34
0.34
0.34
0.34
0.34
0.34
3.34
0.34
0.34
0.34
0.34
0.34
0.34
0.34
3.34
30.75
17.87
' St. II-Nfl
(NO EX)
w/ Onboard
Incidence
Due To
Benzene
(Theoretical)
3.81
3.04
2.58
2.20
1.91
1.64
1.48
1.20
0.98
0.82
0.68
0.55
0.44
0.34
8.25
0.29
0.33
0.34
0.34
0.34
0.34
0.34
0.34
0.34
0.34
0.34
3.34
0.34
0.34
0.34
0.34
0.34
0.34
0.34
0.34
28.58
16.38
St.II-Nation
(EX)
w/ Onboard
Incidence
Due To
Benzene
(Theoretical)
4.42
3.57
1.35
1.11
0.30
0.78
0.67
0.57
8.47
0.40
0.33
0.27
0.22
0.17
0.13
0.13
0.18
0.27
0.33
0.34
0.34
0.34
8.34
0.34
0.34
0.34
0.34
0.34
0.34
0.34
0.34
•3.34
0.34
0.34
0.34
22.34
13.31
St. II-Nation
(NO EX)
w/ Onboard
Incidence
Due To
Benzene
(Theoretical)
4.42
3.28
1.20
3.25
0.15
0.13
0.12
0.10
0,09
0.08
0.07
0.06
0.06
0.05
0.04
0.04
0.11
0.25
0.33
0.34
0.34
0.34
0.34
0.34
0.34
0.34
0.34
0.34
0.34
0.34
0.34
0.34
8.34
3.34
0,34
16.31
9.67
                                                 F-41

-------
       Tablg F-9.  Self-Service Incidence Due To Gasoline Vaqors (Plausible linear Limit)  - Theoretical
Year





1985
1987
1988
!989
i2sa
1591
1592
1993
1994
1995
19%
1997
1998
1999
2M3
£031
302
£003
2004
2805
2006
£007
m&
2309
2010
2011
2812
£013
2014
2015
£016
2317
£318
2319
5028
Baseline
Incidence
Due To
Sas Vaoors
{PUL,ratJ
40. £6
38. S3
37. 66
36.40
35.51
34.60
33.56
33.32
32.£9
31.70
31.16
30.62
20.38
£9.54
£9.01
29.91
29.81
29.01
29.31
£9.01
29.01
29.01
29.01
29.01
29.01
29.01
29.01
£9.01
29.01
29.01
29.01
29.01
29. '31
29.01
£9.01
St.II-Nfl*
(EX)
Incidence
Due To
Gas Vaoors
(PUL,rat)
38.80
35.99
34.59
33.32
32.51
31.67
31.38
30.50
£9.56
£9.02
£8.5£
£8.03
27.54
27.04
£6.55
25.55
£5.55
£6.55
£5.55
£6.55
25.55
26.55
25.55
26.55
26.55
£5.55
26.55
£6.55
£6.55
25.55
£6.55
£5.55
26.55
£5.55
26.55
St.II-Nfl*
(NO EX)
Incidence
Due To
Sas Vaoors
(PUL,rat)
38.14
34.71
33.33
32.10
31.32
30.51
£9.95
£9.38
28.48
27.95
£7.48
£7.01
25.53
£6.06
25.58
£5.58
£5.58
£5.58
25.53
£5.58
25.58
25.58
25.58
£5.58
£5.58
25.58
£5.58
£5.58
25.58
£5.58
£5.58
£5.58
25.58
£5.58
25.58
St.II-NA
(EX)
Incidence
Due To
Gas Vaoors
(PULrat)
35.27
30. S0
29. £8
£8.30
£7.31
£5.61
25.12
25.62
• £4.83
24.38
£3.97
£3.55
£3.14
22. 7£
22.31
22.31
22.31
££.31
££.31
£2.31
£2.31
22.31
22.31
22.31
22.31
22.31
£2.31
££.31
22.31
££.31
£2.31
22.31
22.31
£2.31
22.31
St.II-Nfl St.II-Nation St.II-Nation St.II-Nation
(NO EX) (EX) P.PI (EX) . (NO EX)
Incidence
Due To
Sas Vauors
(PULrat)
34.71
27.88
£5.33
£5.14
£4.53
23.90
23.45
23.01
22.30
21.90
21.52
21.15
£0.78
£0.41
£0.84
£0.04
20.04
£0.04
£0.04
£0.04
£0.134
20.04
£0.04
20.04
23.84
£0.04
20.04
20.04
23.84
20.04
20.34
23.04
20.84
£0.04
£0.04
Incidence
Due To
Gas Valors
(PUL,rat)
40. £6
32.88
19.92
12.58
11.51
11.21
• 11.00
10.79
10.45
10.27
10.10
9.92
9.75
9.57
9.40
9.40
9.40
9.40
9.40
9.40
9.40
9.40
9.40
9.40
9.40
9.40
9.4«
9.40
9.40
9.40
9.40
9.40
9.40
9.40
9.40
Incidence
Due To
Gas Vaoors
(PULrat)
40.26
38.40
32.31
2£.3B
15.11
13.78
12. SI
11.25
10.46
10.27
10.10
9.92
9.75
9.57
9.40
9.40
9.40
9.40
9.40
9.40
9.40
9.40
9.40
9.40
9.40
9.40
9.48
9.40
9.40
9.40
9.40
9.40
9.40
9.40
9.40
Incidence
Due To
Gas Vaoors
(PUL,rat)
40. £5
38.15
12.27
2.73
1.75
1.70
1.67
1.64
1.59
1.56
1.53
1.51
1.48
1.45
1.43
1.43
1.43
1.43
1.43
1.43
1.43
1.43
1.43
1.43
1.43
1.43
1.43
1.43
1.43
1.43
1.43
1.43
1.43
1.43
1.43
 SIM*
1085.14




 358.19
995.72




333.23
960.15




318.97
841.16




£81.92
757.75




255.47
407.60




158.66
443.53




196.48
131.30




 90.90
                                                    F-42

-------
Table F-9.  Self-service  Incidence Due To Gasoline Vasors (Plausible Unper Li.it)  - Theoretical
Year

1385
1387
1988
1389
1S33
1591
1392
1933
1934
1395
1396
1997
1398
1999
--. Tk|>/%
dew
£202
•2233
£834
2335
£306
;307
2388
369
£318
£311
2312
.£•313
2814
£315
£316
£317
2018
£019
2320
Su* =
"IrV =
Onboard
Incidence
Du= To
Sas Vapors
CPUL,rat) '
42.25
33.33
33.32
29.12
25.24
21. S3
13.55
15.81
12.95
13.38
8.92
7.22
5.75
4.^8
3.24
3.24
3.24
3.24
3.24
3.24
3.24
3.24
3.24
3.24
3.24
3,24
3.24
3.24
3.24
3.24
3.24
3.24
2.24
3.24
3.24
341.41
£32.53
St.II-Nft*
(EX)
w/ Onboard
Incidence
Due To
Bas Vapors
(POL, rat)
38.83
35.99
31.35
25.65
23.11
19.73
16.39
14.49
11.88
9.91
8.19
S.S3
5.28
4.35
2. '39
3.39
3.21
3.23
3.24
3.24
3.24
3.24
3.24
3.24
3.24
3.24
3.24
3.24
3.24
3.24
3.24
3.24
3.24
3.24
3.24
323.42
188.41
St.II-Nfl»
(NO EX)
w/ Onboard
Incidence
Due ~o
Sas Vaoors
(Pit, rat)
38.14
34.71
29.94
25.53
22.27
19.37
16.38
13.97
11.45
9.55
7.98
5.48
5.10
3. '32
2.89
3.84
3.28
3.23
3.24
3.24
3.24
3.24
3.24
3.24
3.24
3.24
3.24
3.24
3.24
3.24
3.24
3.24
3.24
3.24
3.24
311.92
182.61
St.II-Nfl
(EX)
w/ Onboard
Incidence
Due To
Sas Vapors
(?UL,rat)
3S.27
33.93
26.30
22.42
19.44
16.65
14.31
12.21
' 10.81
8.35
6.92
5.51
4.48
3.45
2.56
2.34
3.15
3.21
3.24
3.24
3.24
3.24
3.24
3.24
3.24
3.24
3.24
3.24
3.24
3.24
3.24
3.24
3.24
3.24
3.24
28^. 15
153.99
St. II-Nfl
(NO EX)
w/ Onboard
Incidence
Due To
Gas Vapors
(PUL,rat)
34.71
27.88
23.65
20.14
17.47
14.37
12.87
18.98
9.01
7.53
6.24
5.07
4.05
3.13
2.33
2.71
3.14
3.21
3.24
3.24
3.24
3.24
3.24
3.24
3.24
3.24
3.24
3.24
3.24
3.24
3.24
3.24
3.24
3.24
3.24
264.15
158.35
St.II-Nation
(EX)
w/ Onboard
Incidence
Due To
Sas Vapors
(PUL,rat)
40.26
32.80
17.98
18.28
8.26
7.18
6.13
5.25
4.34
3.66
3.06
2.52
2.34
1.51
1.24
1.24
1.68
2.53
3.14
3.24
3.24
3.24
3.24
3.24
3.24
3.24
3.24
3.24
3.24
3.24
3.24
3.24
3.24
3.24
3.24
236.84
122.11
St.II-Nation
(NO EX)
w/ Onboard
Incidence
Due To
Sas Vapors
(PUL,rat)
40.26
30.16
11.05
2.26
1.35
1.28
1.38
0.95
0.84
0.75
3.57
8.58
0.54
0.48
8.43
3.43
1 flfi
x. TO
2 35
W« Ww
3.12
3.24
3.24
3.24
3.24
3.24
3.24
3.24
3.24
3.24
3.24
3.24
3.24
3.24
3.24
3.24
151.43
88.83
                                     F-43

-------
       Taale F-9.  Self-Service Incidence Due To  Gasoline Vaoors  (Plausible Uoser Limit) - Theoretical
Year





1985
1337
13S8
1983
1993
1991
1932
1993
1994
1995
19%
1997
1998
1999
20SS
2201
seas
seas
384
2205
22%
2837
2008
2309
£31®
2311
£812
2313
£014
2915
2016
2317
2018
2819
2820
Baseline
Incidence
Due To
3as Vacors
{pUL.Hice)
24.25
23.55
22.79
22.22
21.49
20.93
20.54
20.16
19.54
19.18
18.85
18.53
18.20
17.88
17.55
17.55
17.55
17.55
17.55
17.55
17.55
17.55
17.55
17.55
17.55
17.55
17.55
17.55
17.55
17.55
17.55
17.55
17.55
17.55
17.55
St.II-Nfl*
(EX)
Incidence
Due To
Gas Vanors
(PUL. Eice)
23.45
21.73
23.88
20.12
13.63
19.12
18.77
18.41
17.34
17.52
17.22
15.92
16.53
15.33
15.03
16.03
16.03
16.03
16.03
16.03
15.33
16.83
16.03
16.03
16.03
16.03
16.03
15.03
16.03
16.03
15.03
16.03
16.03
16.03
15.03
St.II-Nftt
(NO EX)
Incidence
Due To
Sas Vaoors
(PUL, nice)
23.35
20.34
20.11
19.35
18.89
18.40
18.06
17.72
17.18
15.86
15.58
16.29
16.00
15.72
15.43
15.43
15.43
15.43
15.43
15.43
15.43
15.43
15.43
15.43
15.43
15.43
15.43
15.43
15.43
15.43
15.43
15.43
15.43
15.43
15.43
St.II-Nfl
(EX)
Incidence
Due To
Sas Vaoors
(PUL. Mice)
£1.33
18.59
17.60
16.82
15.41
15.39
15.63
15.40
' 14.92
14.65
14.40
14.15
13.90
13.55
13.40
13.40
13.40
13.40
13.48
13.40
13.40
13.40
13.40
13.40
13.40
13.40
13.40
13.40
13.40
13.40
13.40
13.40
13.40
13.40
13.40
St. II-NA St. II-Nation St. II-Nation St. II-Nation
(NO EX) (EX) flPI (EX) . (NO EX)
Incidence
Due To
Sas Vaoors
(PUL. sice)
23.32
16.71
15.77
15.05
14.53
14.31
14.05
13.78
13.36
13.11
12.83
12.57
12.44
12.22
12.00
12.30
12.00
12.30
12.tt
12.00
12.80
12.00
12.00
12.00
12.30
12.80
12.20
12.30
12.^0
12.00
12.08
12.30
12. S3
12.80
12.30
Incidence
Sue To
Gas Vapors
(PUL, mice)
24.35
13.76
11.80
7.34
5.63
6.46
6.34
6.22
5.33
5.32
5.32
5.72
5.52
5.51
5.41
5.41
5.41
5.41
5.41
5.41
5.41
5.41
5.41
5.41
5.41
5.41
5.41
5.41
5.41
5.41
5.41
5.41
5.41
5.41
5.41
Incidence
Due To
Eas Vapors
(PUL,!aice)
24.36
23.23
13.48
13.34
8.36
3.05
7.33
6.50
6.33
5.92
5.82
5.72
5.62
5.51
5.41
5.41
5.41
5.41
5.41
5.41
5.41
5.41
5.41
5.41
5.41
5.41
5.41
5.41
5.41
5.41
5.41
5.41
5.41
5.41
5.41
Incidence
Due To
Sas Vaoors
(PUL, mice)
24.36'
18.12
7.37
1.18
8.59
3.57
0.55
3.35
3.54
3.53
0.52
8.51
0.53
3.49
3.48
3.48
3.48
3.48
3.48
0.48
3.43
8.48
0.48
3.48
3.43
3.48
3.48
3.48
3.48
0.48
0.48
3.48
3.48
8.48
3.43
 SUM
6w&«55




216.72
601.21




199.42
579.20




192.45
535.55




169.51
453.92




153.14
237.20




 99.42
259.44




115.63
66.19




51.29
                                                    F-44

-------
-able F-9.  Self-Service Incidence Due To Baseline Vapors (Plausible Upper Liait)  - Theoretical
Year



198S
1937
1988
1939
1952
1991
1992
1393
1994
1995
19%
1997
1998
1999
c'380
£031
£002-
£033
£884
£085
2086
•2007
2S08
£039
£018
2011
2012
2013
2014
2015
2016
£017
2018
2019
2020
SUM =
NPV =
Onboard

Incidence
Bus To
Gas Vapors
(PlL,sice)
24.35
23.55
20,45
17.62
15.27
13.07
' 11.22
9.57
7.34
6.54
5.40
4.37
3.48
2.66
1.96
1,95
1.96
1.X
1.96
1.X
1.X
1.95
1.96
l.X
1.96
1.96
1.96
l.X
1.X
1.96
i.X
1.96
l.X
1.95
1.96
266.57
122.54
St.II-Nfl*
(EX)
w/ Onboard
Incidence
"us To
•Sas Vaoors
(PUL,3iice)
23.45
21.73
18.75
15.10
13.96
11.95
10.26
8.75
7.17
5.98
4.94
4.01
3.19
2.45
1.31
1.87
1.94
1.95
1.95
1.95
1.96
1.%
1.96
1.95
1.X
1.96
1.96
1.96
l.X
1.96
1.96
1.96
1.96
1.95
l.X
193.59
113.80
St. II-Nfl*
(NO EX)
w/ Onboard
Incidence
Due To
Sas Vapors
(POL, mice)
23.05
20.34
18.05
15.50
13.44
11.50
9.88
8.43
6.91
5.76
4.76
3.86
3.08
2.36
1.74
1.34
1.94
1.95
1.X
1.95
l.X
l.X
1.95
1.36
1.96
1.96
1.96
1.96
1.96
1.95
1.95
1.95
1.95
1.X
1.96
188.32
110.21
'St.II-Nfl
(EX)
w/ Onboard
Incidence
Due To
Sas Vapors
(PH., nice)
21.89
18.59
15.81
13.47
11.68
10.00
8.50
7.33
' 6.02
5.02
4.16
3.38
2.59
2.07
1.54
1.71
1.91
1.94
1.95
1.96
1.95
i.X
1.96
1.95
1.X
1.96
1.95
1.X
1.95
1.96
l.X
1.96
1.96
1.36
1.96
171. 14
98.69
St.II-Nfl
(NO EX)
w/ Onboard
Incidence
Due To
Gas Vapors

-------
   Table F-10.  Self-Service  Incidence Due To Sasoline Vaoors (Haxiaua Likelihood Estisate)  - Theoretical
Year
 NPV






1986
1387
1988
1983
1933
1991
1992
1993
1994
1925
19%
1997
1998
1999
sm
2031
2002
5003
2884
£835
2866
2007
2038
2009
2010
2011
2012
£013
2014
2015
2016
£017
2018
£019
2029
=
a
Baseline

Incidence
Due To
Gas Vapors
(ME, rat)
22.92
22.17
21.44
20.73
20.22
19.70
13.33
18.97
18.39
18.95
17.74
17.44
17.13
16.82
16.52
16.52
16.52
16.52
16.52
16.52
16.52
16.52
16.52
15.52
16.52
15.52
16.52
16.52
16.52
16.52
16.52
16.52
16.52
15.52
16.52
617.88
203.95
St.II-Nfl*
(EX)
Incidence
Due To
Gas Vapors
(ME, rat)
22.89
20.49
19.69
18.97
18.51
18.03
17.70
17.36
16.83
15.52
16.24
15.%
15.68
15.40
15.12
15.12
15.12
15.12
15.12
15.12
15.12
15.12
15.12
15.12
15.12
15.12
15.12
15.12
15.12
15.12
15.12
15.12
15.12
15.12
15.12
555.97
188.04
St.II-Nfl*
(NO EX)
Incidence
Due To
Gas Vapors
(ME, rat)
21.72
19.76
18.98
18.28
17.83
17.37
17.05
16.73
16.21
15.92
15.65
15.38
15.11
14.84
14.57
14.57
14.57
14.57
14.57
14.57
14.57
14.57
14.57
14.57
14.57
14.57
14.57
14.57
14.57
14.57
14.57
14.57
14.57
14.57
14.57
546.72
181.62
St.II-Nfl
(EX)
Incidence
Due To
Gas Vapors
(ME, rat)
20.65
17.60
16.57
15.94
15.55
15.15
14.87
14.59
• 14. 14
13.88
13.65
13.41
13.17
12.94
12.70
12.70
12.70
12.70
12.70
12.70
12.70
12.70
12.70
12.70
12.70
12.70
12.70
12.70
12.70
12.70
12.70
12.70
12.70
12.70
12.70
478.96
160.53
St. II-Nfl St. II-Nation St. II-Nation St. II-Nation
(NO EX)
Incidence
Due To
Gas Vapors
(ME, rat)
19.76
15.37
14.99
14.32
13.97
13.51
13.36
13.10
12.70
12.47
12.25
12.04
11.83
11.52
11.41
11.41
11.41
11.41
11.41
11.41
11.41
11.41
11.41
11.41
11.41
11.41
11.41
11.41
11.41
11.41
11.41
11.41
11.41
11.41
11.41
431.47
145.45
(EX)
Incidence
Due To
Gas Vapors
(ME, rat)
22.32
18.68
11.34
7.22
6.55
5.38
• 6.26
6.15
5.96
5.85
5.75
5.55
5,55
5.45
5.35
5.35
5.35
5.35
5.35
5.35
5.35
5.35
5.35
5.35
5.35
5.35
5.35
5.35
5.35
5.35
5.35
5.35
5.35
5.35
5.35
232.09
96.24
flPI (EX)
Incidence
Due To
Gas Vapors
(ME, rat)
22.32
21.87
. 18.40
12.74
8.60
7.85
7.18
5.40
5.95
5.85
5.75
5.65
5.55
5.45
5.35
5.35
5.35
5.35
5.35
5.35
5.35
5.35
5.35
5.35
5.35
5.35
5.35
5.35
5.35
5.35
5.35
5.35
5.35
5.35
5.35
252.55
111.88
(NO EX)
Incidence
Due To
Sas Vapors
(ME, rat)
22.32
17.17
6.39
1.55
1.00
0.97
8.95
0.93
3.91
0.89
0.87
0.86
0.84
0.83
0.81
0.81
0.81
0.81
8.81
0.81
0.81
0.81
0.81
0.81
0.81
0.81
0.81
0.81
0.81
0.81
0.81
0.81
0.81
0.81
0.81
74.75
51.76
                                                   F-4b

-------
Table F-10.  Self-Service  Incidence Due To Basoiine Vauors (Maximum Likelihood Estiaate)  - Theoretical
Year



1S86
1987
1988
1989
1998
1991
1992
1993
1994
1995
19%
1997
1998
1999
2089
2081
2082
2883
2024
2085
2086
2087
2088
2889
2818
2911
2312
2013
2014
£015
2016
2017
2318
2019
2820
SUM =
hPV =
Onboard

Incidence
Due To
Gas Vaoors
(ME, rat)
22.32
22.17
19.25
16.58
.14.37
12.38
10.56
9.00
7.38
6.15
5.08
4.11
3.27
2.51
1.34
1.84
1.84
1.34
1.84
1.84
1.34
1.84
1.84
1.84
1.34
1.84
1.34
1.84
1.34
1.84
1.34
1.84
1.34
1.84
1.34
194.40
115.32
St. II-N'ft*
(EX)
w/ Onboard
Incidence
Due To
Sas Vacors
(ME, rat)
22.39
20.49
17.69
15.18
13.16
11.27
9.68
8.25
6.76
5.54
4.66
3.78
3.81
2.31
1.70
1.76
1.33
1.84
1.84
1.84
1.84
1.84
1.34
1.84
1.34
1.84
1.34
1.84
1.34
1.84
1.34
1.84
1.34
1.84
1.34
132.45
137.28
St.II-Nfl*
(NO EX)
H/ Onboard
Incidence
Due To
Sas Vapors
, (ME, rat)
21.72
19.76
17.05,
14.63
12.68
10.86
9.33
7.95
5.52
5.44
4.50
3.64
2.93
2.23
1.65
1.73
1.82
1.84
1.84
1.84
1.84
1.84
1.84
1.34
1.34
1.84
1.34
1.84
1.84
1.84
1.84
1.84
1.84
1.84
1.34
177.51
103.98
St.II-Nfl
(EX)
w/ Onboard
Incidence
Due To
Gas Vaoors
(ME, rat)
20. 55
17.63
14.97
12.76
11.07
9.48
3.15
6.95
' 5.78
4.76
3.94
3.20
2.55
1.%
- 1.46
1.61
1.79
1.33
1.84
1.84
1.84
1.84
1.84
1.84
1.34
1.84
1.84
1.84
1.34
1.84
1.84
1.84
1.84
1.84
1.34
161.30
93.33
St.II-Nfl
(NO EX)
w/ Onboard
Incidence
Due To
Sas Vapors
(ME, rat)
19.76
15.87
13.47
11.47
9.95
8.52
7.33
5.25
5.13
4.29
3.55
2.89
2.31
1.78
1.32
1.54
1.79
1.83
1.34
1.84
1.84
1.84
1.84
1.34
1.34
1.84
1.84
1.84
1.84
1.84
1.84
1.34
1.84
1.84
1.84
150.41
55.51
St.II-Nation
(EX)
M/ Onboard
Incidence
Due To
Sas Vapors
(ME, rat)
22.92
18.68
10.19
5.81
4.70
4.04
• 3.49
2.99
2.47
2.88
1.74
1.43,
1.16
0.92
3.71
0.71
3.96
1.47
1.79
1.84
1.84
1.84
1.84
1.84
1.34
1.34
1.34
1.34
1.84
1.84
1.84
1.84
1.84
1.84
1.84
117.78
59.53
St.II-Nation
(NO EX)
H/ Onboard
Incidence
Due To
Sas Vapors
(ME, rat)
22.92
17 17
* * • 4 /
S 29
U* ^J
1.2fl
4 • UW
0.77
0.68
0.61
3.55
0.48
0.43
w» ^w
3.38
8.34
0.31
0.27
0.25
3.25
0.60
1.34
1.78
1.84
1.84
1.84
1.84
1.84
1 34
*« l/T
1.84
1.84
1.84
1 S4
i« G^
1 84
A • \fT
1.34
1.34
1.34
.1.84
1.34
36.22
50.56
                                         F-47

-------
   Table F-I0.  Self-Servics Incidence  Due  To Gasoline Vaoors  (Maximum Likelihood Estisats)  - Theoretical
Year





1986
1987
1988
1989
1990
1991
1992
1993
1994
1995
1995
1997
1998
1999
2003
2081
2002
2803
2884
2085
2806
2807
2008
2889
2318
2811
2012
2013
2814
2015
2016
£017
2018
2019
2020
-
s
Baseline
Incidence
Due To
Sas Vapors
(M.E,aice)
16.42
15.88
15.36
14.85
14.49
14.11
13.85
13.59
13.17
12.93
12.71
12.49
12.27
12.05
11.83
11.83
11.83
11.83
11.83
11.83
11.83
11.83
11.83
11.83
11.83
11.83
11.83
11.83
11.83
11.83
11.83
11.83
11.83
11.83
11.83
442.66
146. 12
St. II-Mfl*
(EX)
Incidence
Due To
6as Vapors
(MLE,aice)
15.83
14.68
14.11
13.59
13.26
12.32
12.68
12.44
12.06
11.84
11.64
11.43
11.23
11.03
10.83
18.83
10.83
10.83
10.33
10.83
18.83
10.83
10.83
10.33
10.83
10.83
10.83
10.83
10.83
10.83
10.83
10.83
10.83
10.83
10.83
485.19
134.71
St. II-Nfl*
(NO EX)
Incidence
Due To
Gas Vaoors
(MLE,aice)
15.56
14.15
13.68
13.18
12.78
12.45
12.22
11.39
11.62
11.40
11.21
11.02
10.82
10.63
10.44
18.44
10.44
10.44
18.44
10.44
10.44
18.44
10.44
10.44
10.44
10.44
13.44
10.44
10.44
10.44
18.44
13.44
10.44
10.44
18.44
391.68
138.12
St. II-Nfl
(EX)
Incidence
Due To
Sas Vapors
(MLE,iaice)
14.83
12.61
11.94
11.42
11.14
13.85
18.65
10.45
' 18. 13
9.95
3. 78
9.61
3.44'
9.27
9.10
9.18
9.10
9. 18
9.10
9.1.0
9.10
9.10
9.1.8
3.18
9.10
3.18
9.10
9.10
9.13
9.18
9.18
9.10
9.18
9.10
9.10
343.14
115.88
St. II-Nfl St.II-Nation St.II-Nation St.II-Nation
(NO EX) (EX) AP! (EX) . (NO EX)
Incidence
Due To
6as Vapors
(MLE,3ice)
14.16
11.37
18.74
18.26
18.81
9.75
3.57
9.39
9.13
3.93
8.78
8.63
8.48
8.32
8.17
8.17
8.17
8.17
8.17
3.17
8.17
8.17
3.17
3.17
8.17
8.17
3.17
3.17
8.17
8.17
3.17
3.17
8.17
8.17
8.17
389.11
104.21.
Incidence
Due To
3as Vaoors
(ME,aice)
16.42
13.38
3.12
5.17
4.53
4.57
4.43
4.40
4.27
4.19
4.12
4.85
3.38
3.31
3.83
3.83
3.83
3.83
3.83
3.83
3.83
3.83
3.83
3.83
3.83
3.33
3.83
3.83
3.83
3.33
3.83
3.83
3.83
3.83
3.83
155.27
68.80
Incidence
Due To
Sas Vaoors
(MLE,wice)
16.42
15.57
13. 18
3.13
6.16
5.62
5.14
4.53
4.27
4.13
4.12
4.85
3.58
3.31
3.83
3.83
3.83
3.83
3.83
3.83
3.83
3.83
3.33
3.83
3.83
3.33
3.63
3.83
3.33
3.33
3.33
3.33
3.33
3.33
3.83
188.93
58.15
Ineidarice
Due To
Gas VsDors
!MLE,;aic5)
IS. 42
12.33
C ••£«
1.11
3.71
0.S3
3. £3
8.57
8.55
0.54
3.63
8.61
3.60
8.59
8.53
3.58
3.58
'3.58
3.58
0.58
3.5B
3.53
8. 53'
2.58
2.53
8.58
2.53
3.58
3.58
8.58
g.53
8.58
3.53
8.53
8.58
53.56
37.i28
                                                F-48

-------
Table F-10.  Self-Service  Incidence Due-To Gasoline Vapors (Maxisw  Likelihood Estiaate) - Theoretical
Year





1985
1987
1988
1989
1990
1991
1992
1993
1994
1995
19%
1997
1998
1999
2000
2001
2002
2003
2004
2005
2006
2007
2008
2009
2010
2011
2012
2013
2014
2015
2016
2017
2018
2019
2020
SUM =
NPV =
Onboard

Incidence
Due To
Sas Vaoors
(HE, Bice)
16.42
15.88
13.79
11.88
10.30
8.81
• 7.57
6.45
5.28
4.41
3.64
2.95
2.34
1.80
1.32
1.32
1.32
1.32
1.32
1.32
1.32
1.32
1.32
1.32
1.32
1.32
1.32
1.32
1.32
1.32
1.32
1.32
1.32
1.32
1.32
139.27
82.62
St.II-Nft*
(EX)
H/ Onboard
Incidence
Due To
Sas Vapors
(MLE, sice)
15.83
14.68
12.67
10.88
9.43
8.07
6.93
5.91
4.34
4.04
3.34
2.71
2.15
1.65
1.22
1.25
1.31
1.32
1.32
1.32
1.32
1.32
1.32
1.32
1.32
1.32
1.32
1.32
1.32
1.32
1.32
1.32
1.32
1.32
1.32
130.71
76.85
St.II-Nft*
(NO EX)
»/ Onboard
Incidence
Due To
Sas Vaoors
(«LE,aice)
15.55
14.16
12.21
10.48
9.09
7.78
5.68
5.70
4.67
3.90
3.22
2.61
2.08
1.60
1.18
1.24
1.31
1.32
1.32
1.32
1.32
1.32
1.32
1.32
1.32
1.32
1.32
1.32
1.32
1.32
1.32
1.32
1.32
1.32
1.32
127.24
74.49
Sf.II-Nfl
(EX)
«/ Onboard
Incidence
Due To
Sas Vapors
(MLE. slice!
14.80
12.51
10.73
9.14
7.93
6.79
5.84
4.98
' 4.08
3.41
2.82
2.29
1.83
1.41
1.04
1.16
1.29
1.31
1.32
1.32
1.32
1.32
1.32
1.32
1.32
1.32
1.32
1.32
1.32
1.32
1.32
1.32
1.32
1.32
1.32
115.91
56.90
St. II-Nft
(W EX)
«/ Onboard
Incidence
Due To
Sas Vapors
fflLE.aice)
14. 16
11.37
9.65
8.22
7.13
5.10
5.25
4.48
3.68
3.07
2.55
2.07
1.65
1.28
0.95
1.11
1.28
1.31
1.32
1.32
1.32
1.32
1.32
1.32
1.32
1.32
1.32
1.32
1.32
1.32
1.32
1.32
1.32
1.32
1.32
107.76
51.33
St. II-Nation
- (EX)
K/' Onboard
Incidence
Due To
Sas Vapors
(MLE. sice)
16. 42
12.38
7.33
4.15
3.37
2.89
2.50
2.14
1.77
1.49
1.25
1.83
0.83
0.56
0.51
0.51
0.69
1.06
1.28
1.32
1.32
1.32
1.32
1.32
1.32
1.32
1.32
1.32
1.32
1.32
1.32
1.32
1.32
1.32
1.32
84.38
49. 81
St. II-Nation
(NO EX)
w/ Onboard
incidence
Due To
Sas Vapors
iME.snce)
15.42
:5. 38
4.51
3.3£
3.55
8.49
3.44
3.39
2.34
3.31
0.27
0.25
0.22
3.20
0.18
9.18
0.43
2.96
1.27
1.32
1.32
1.32-
1.32
1.32
1.32.
1.32
1.32
1.32
1.32
1.32
L32
1.32
1.32
i.32
1.32
61.77
36.22
                                          F-49

-------
Table F-ll. Self-Service Incidence Due To EDB And EDC - Thaoreticai
Year



Baseline St.II-NA*
(EX)
Incidence
Due To
ma
Incidence
Due To
EEB
(Theoretical) (Theoretical)
1986
1987
1988
1989
1990
1991
1992
1993
1994
1995
1996
1997
1998
1999
2900
2001
2982
2993
2994
2895
2906
2097

2999
2919
£911
£912
2913
2914
2915
2916
£917
2918
2919
2829
SUM =
NPV =
9.9296
9.8253
9.0228
0.0199
0.8172
9.9152
0.0128
8.0185
0.0988
0.0070
0.0058
0.0947
9.0935
9.0923
9.9912
9.0912
0.0912
0.8012
0.0012
9.0012
9.0812
0.0012
0.0912
9.8912
0.9912
0.9912
0.0912
0.9912
0.8312
9.3912
9.8012
9.8912
9.3812
0.0912
0.8812
8.218
8.136
3.3284
9.0238
3.0209
0.0181
0.0157
0.0139
9.0117
0.9996
0.8880
9.0864
9.0053
0.9843
0.8832
0.9921
9.0011
0.8911
9.9811
9.0011
0.8911
9.0911
0.8011
8.0911
0.0911
0.0911
9.8911
0.0911
9.0311
9.8911
0.8911
9.8811
9.8311
0.8011
9.8911
0.3911
0.8311
8.194
3.125
St.IHW*
(NO EX)
Incidence
Due To
EDB
(Theoretical)
3.8279
0.0223
9.0281
0.9174
8.9151
3.9133
8.0113
3.8992
9.8977
8.8952
0.8051
0.8041
0.8031
8.8921
8.8910
9.0910
9.8919
9.8919
8.8918
8.0819
9.8810
8.8810
0.8919
9.8910
9.8919
0. '3819
8.8818
3.8810
0.3818
8.0919
9.8810
3.8819
8.8818
9.8919
8.9318
9.187
8.1S2
St.II-Nfl
(EX)
Incidence
Due To
EDB
(Theoretical)
3.8265
9.8293
3.3176
8.9151
8.8131
3.3116
0.8998
3.8888
• 8.8367
3.8353
0.8045
3.8836
0.0827
8.8818
8.8889
3.3039
8.3039
3. £009
8. (3889
3.3889
0.0089
8.8039
0.8803
8.8389
8. (500-3
8.3S99
8.3389
3.8889
3.3889
3.3889
•3.8389
8.8809
9. 8889
3, 0889
0.88(59
3. 165
8.189
St.II-Nfi ot.II-Nation :
!KO EX) (EX)
Incidence
Due To
ESB
(Theoretical)
8.3E54
•3.8183
8.0157
3.3135
8.3117
3.3133
3.8338.
3.3372
9.3868
3.8948
9.3040
3.2332
8.8924
3.0316
8.3988
8.8888
'8.8998.
2.3388
8.0088
3. -2838
3.2369
3. 3368
0. 8883
•8. 3338
3.8388
•3. 8388
8.8828
3.3888
3.3868
8. £388
8.8888
3. mm
9.2088
3. S088
3.0338
8. 149
8.399
Incidence
D-iS "o
E"3
(Theoretical)
•3. 3235
3.8216
8.3117
0.3365
9.8852
3.8846
8. £233
3.3832
3.88£7
3. 2821
0.M13
8.2314
0.8811
3.3837
. 8.9884
3. 3834
8.0884
3.2234
8.3804
8. 2K4
9.0F34
9. 3834
9.2834
3.3804
9.8804
3. C034
3.3834
3.C034
3.3884
8.8884
3.3884
8.23S4
3.8884
3. 2824
8.3884
3. i'33
9.575
St.II-Nsiiori
Ir.cicsr-.c9
Due ":•
EJB
(Tnsorsvlcal)
:.329&
3. C2£5
3.C134
8.3113
3.387®
8. 2858
8. 30i5
3.S323
2. 3827
- 3.3021
3. 3316
8. 8814
3.0811
9. 3387
3.8884
3.2.824
2. -2804
8. 3834
9. 8934
8. 3334
3. 3034
3. 3084
3. 3884
3.3824
3. C924
3. -3884
3. 3804
3. B334
•8. -3334
3. 3084
3. 3834
3. 8834'
8. 3934
8. £384
8. 8884
8.124
8.1335
                              F-50

-------
Taois F-li.  Seif-Ssrvice  Incidence Due To EDB find EDC - Theoretical
Year






1986
1987
1988
1989
1998
1991
1992
1993
1994
1995
1996
1997
1998
1999
2888
2081
£802
2003
2004
2005
2006
2087
2888
2889
2818
2011
2812
2013
2814
2315
2016
2017
2818
2319
2028
SUM =
NPV =
St.II-Nation
(NO EX)

Incidence
Due To
EDB
(Theoretical)
3.82%
0.3198
8.8869
0.3809
0.8083
3.8083
0.8883
0.0802
3.8082
8.0801
0.8001
0.8881
0.8881
.8088
.8888
.0088
.8888
.0008
.0088
.8808
.8888
.8080
.0088
.8000
.8888
.3338
.3888
.3888
.8880
.8388
.0088
.8888
.8883
.8888
.8888
0.859
8.855
Onboard


Incidence
Due To
EDB
(Theoretical)
8.8296
0.3258
8.8228
8.8199
8.0172
0.3152
3.8128
0.3185
8.8888
0.8870
0.8858
8.8047
8.8835
0.8823
8.8812
0.8812
8. 8812
0.0012
8.8812
8.8012
8.8812
0.8812
3.8812
0.8812
8.8812
8.3312
0.8812
3.8812
8.8812
8.3312
8.8012
8.8012
8.8812
0.3312
0.3812
3.218
3. 135
St.II-Nfl*
(EX)
w/ Onboard
Incidence
Due To
EDB
(Theoretical)
8.8234
3.8238
3.8289
3.3181
3.8157
3.0139
3.3117
3.38%
3.3880
8. £064
3.8853
8.8843
3.8332
8.3821
3.8811
8.8811
8.8812
8.8812
3.8812
8.8312
3.8812
3.8812
3.8812
8.3312
3.8812
8.0312
3.3812
3.3012
3.3812
3.8312
8, 3812
3.8012
-3,3812
3. 3312
3.3012
8.195
3. 125
St. II-Nfl*
(NO EX)
w/ Onboard
Incidence
Due To
EDB
(Theoretical)
8.3279
3.8229
8.8281
8.3174
0.3151
3.0133
3.3113
8.8892
' 3.3877
8.0062
8.8851
3.3841
3.3831
0.8821
3.3810
3.3311
8. S812
0.8012
3.3312
0.0812
3.8312
3.0812
. " 3. 3812
0.8812
3.2312
3.8012
3.5312
3.8012
8.3012
0.8812
8.8312
8.8812
3.J312
3. 8812
3.8012
0. 198
3.122
St.II-Nfl
(EX)
w/ Onboard
Incidence
Due To
EDB
(Theoretical)
3.S255
3. 8283
3.3176
3.3151
3. 3131
8.3116
3.3393-
0. 8888
3. «67
8. 3853
3.8945
3.3835
3.3827
0.3813
0.3889
0. 8818
3. Mil
8.3012
3.3812
0.8012
3. 3812
3.3812
3.8812
3.3012
3.3312
3. 8812
3.3312
•2. 3812
3. 3812
3.3812
3.8312
3.8812
3.3812
3. 3312
3.3912
3.17!
3.139
St.II-Nfl
(NO EX)
w/ jncoara
Incidence
Due To
EDS
(Theoretical i
0.3254
3.3133
3.3157
3.3135
3.3117
0.3133
8. 2833
0.JS7E
0. 3858
0.5048
3. 8348
3. 3332
3. 3824
3.3S215
3.3888
8.3813
8.2811
8.3812
3.3312
3.881E
.3. S3 12
3.S012
3.312
3. £812
3.C312
2. C812
3.8012
3.3812
3.S312
3. £312
3,8812
3.2812
3.3312
2, £3:2
8.2812
3.157
.*.m
St.II-Nation
!EX!
w/ Onboard
Incidence
Due "To
3B
(Theoretical)
*.&%

3.3117
3. 5855
3.335E

8. 2333
3. 2832
0.3027
' 3.3821
3.3818
3.301*
8.3311
3.3887
3.3384
3. 3834
3.3885
3. 3889
3.3311
3. 3812
3. 2012

5.3312
3.3012
3.C312

3. t31?
2. "31?
3. C312
C» ••Jdl.L
3.5-312
T< *ii i J-:
<-« v'oiC
•i *•*«•••
V
-------
     Table F-ll. Self-Service Incidence Due To EBB Qnd EDC - ineorsticai
ar






198&
1987
1988
1389
1930
1991
1992
1393
1994
1995
19%
1397
1998
1999
2903
2081
2$02
2803
2084
2995
2006
2007
2828
2809
2919
2911
2012
2313
2914
2915
2016
2917
2918
2919
2029
St.II-Nation
(NO EX)
H/ Onboard
Incidence
Due To
EDB
(Theoretical)
0. '32955
0.91982
0.00595
9.90092
0.80334
0.00039
9. 30925
8.99921
0.90817
9.00814
9.09012
0.99909
0.90307
0.00905
0.99902
0.00002
0.90928
0.00081
0.00112
0.09117
9.90117
8.90117
0.90117
9.00117
9.30117
0.00117
0.30117
0.00117
0.99117
8.80117
8.00117
0.09117
3.30117
0.00117
8.30117
Baseline

Incidence
Due To
EDC
(Theoretical)
8.3381
0.0333
3.3294
0.0256
0.3221
0.9196
3.3166
8.0136
3.3113
0.3890
8.3375
9.3060
8.3345
9.0033
8.3015
0.3015
3.3315
3.3015
3.3815
9.3915
8.3015
8.0315
0.0015
8.0815
3.8815
3.3815
8.3015
8.0915
8.3815
8.3315
3.3815
9.0815
3. '3815
8.8815
9.6315
St.IHM*
(EX)
Incidence
Due To
EBC
(Theoretical)
3. 3367
8.0387
9.3269
8.8234
3.0232
3.3179
3.3151
3.8124
3.8133
8.3833
3.8869
8.3855
8.3841
9.302B
9.3014
8.3814
0.3014
3.0814
8.3314
0.3814
8.3314
9.0014
3.3014
3.8014
9.3814
3.3814
3.3014
3.8814
0.'8014
3.0314
9.S814
0.8914
0. 3014
3. 6014
9.8314
St. II-Nfl*
(NO EX)
Incidence
Due To
EDC
(Theoretical)
3.3351
3.3296
0. 3259
3.8225
8.3195
8.3172
3. 8146
3.3119
3.3899
0.3879
3. 3066
3.5053
8.3840
8.3326
3.3313
9. £013
8.3813
3.3813
3.3813
3.3913
3.3313
3.3813
3.3013
3.3313
3.3313
8.2313
8.8313
0.0813
3.8013
3.13813
3. '2013
9.3813
3. 3013
3.0313
3.3813
St. II-N'fi
(EX)
Incidence
Due To
EDC
(Theoretical)
3.0342
3.3262
0.822S
3.31S5
3.3169
8.3149
8. 8125
0.3103
3. 2066
9.8369
8. 3357
3.3046
3. £334
3. §323
9.3311
8.3811
3. 2311
0.12811
0.3311
0.2811
3.3311
9.C311
8.3311
3.331;
3.3311
3.0311
3.3311
0,3811
8.3-311
3.3311
3.5311
8.S011
8.0311
9.3311
3.8311
                                                                                  3.S327
                                                                                  3.J23S
                                                                                  •3.0293
                                                                                  3.3174
                                                                                  3.3151
                                                                                  •3.3133
                                                                                  3.8113
                                                                                  3.2092
                                                                                  3.2-377
                                                                                  3.3362
                                                                                  0.8351
                                                                                  3.S041
                                                                                  3.3031
                                                                                  2.3021
                                                                                  3.8310
                                                                                  3. '2318
                                                                                  3.8313
                                                                                  3.3813

                                                                                  3. m 13
                                                                                  0.3013
                                                                                  3.3818
                                                                                  3.3310
                                                                                  3.3313

                                                                                  8.6310
                                                                                  3.3312
                                                                                  3.0910
                                                                                  8.3012!
                                                                                  8. S318

                                                                                  s.;
0.980

3.957
8.271

8.176
8. £58

3.163
3.241

3.157
0.213

3.143
8.8310

 9.193

 3.127
                                   F-52

-------
Table F-ii.  Self-Service Incidence Due To EDE And EDC  -  Theoretics!
Year






1985
1987
1988
1989
1998
1991
1992
1993
1994
1995
1996
1997
1998
1999
2800
2001
2802
£003
2004
2805
2006
2007
2008
2009
2010
£811
2012
2013
£014
2015
2016
2017
2018
2019
2020
SUM =
NPV =
St.II-Nation
(EX)

Incidence
Due To
EDC
(Theoretical)
0.0381
9.0279
0.0151
0.0084
0.0057
0.3059
•0.0050
0.0041
0.0034
0.8027
0.0023
0.0818
0.0014
0.0009
0.0005
0.0005
0.0005
0.0085
0.0005
0.0005
0.0005
0.8885
' 0.0005
0.0005
0.0005
0.0005
0.0005
0.0005
0.0005
0.0005
0.0005
0.8085
0.0005
3.S005
0.0005
0.133
0.102
St.II-Nation
API (EX)

Incidence
Due To
EDC
(Theoretical)
0.0381
0.0328
0.0251
0.8154
0.0099
3.®74
9.0058
0.3843
0.0034
0.3027
0.0023
0.0818
0.0014
0.0009
0.0005
0.0005
0.12085
0.0005
0.0005
0.0005
0.0005
0.0005
0.0085
0.0005
0.0005
0.0005
0.0005
0.0005
0.0005
0.0305
0.8805
0.0005
0.3005
3.0005
0.0305
0.1£0
0.123
St.II-Nation'
(NO EX)

Incidence
Due To
EDC
(Theoretical)
3.03812
0.82555
3.30895
0.80118
3.30044
0.33339
3.03033
0.38827
0.33322
0.00018
0.38815
3. 03012
0.03089
8.30006
3.83333
8.30803
0.00093
3.00003
0.00083
3.00003
8.38833
0. 83883
0.80803
8.08003
0.08083
3.00003
8.88833
8.30003
0.38833
8.08803
8.03003
8.03803
8.00803
3.89883
0.08003
3.377
0.871
Onboard


Incidence
Due To
EDC
(Theoretical)
0.3381
8.8333
3.0294
3. '3255
0.3221
3.3135
3.8155
8.3136
0.0113
8.0890
0.8375
0.3860
8.3845
0.3038
8.0015
. 3.3315
9.3015
8.3815
0.0815
8.8815
8.8815
9.8015
8.0015
3.8015
3.3015
3.0315
0.3815
3.8815
3.0315
0.@815
3.3015
3.8815
8.8815
3.S815
0.3015
3.271
3. 176
St.II-Nfl*
(EX)
w/ Onboard
Incidence
Sue To
EDC
(Theoretical)
3.3367
3.3337
0.8269
0.3234
3.3202
0.3173
3.0151.
3.3124
3.8103
3.3383
8.8059
8.8855
8.3841
8.2328
3. 3814
3.2314
9. 8015
0.2015
3.3815
3.esi5
0.3815
0.3315
3.3015
3. -3815
3.3015
«. 88 15
3.3315
8.3815
0.2315
8.8315
3.0015
3.2315
0.8815
8.8815
3.0015
3.253
8.163
St.II-Nfl*
(NO EX)
w/ Onboard
Incidence
DUB To
EDC
(Theoretical)
2.3351
3.3296
0.0253
3.3225
8.3155
•3.3172
0.8146
8.3119
3.0099
8. 8879
3.0865
8.3053
0.0343
8.2826
8.6313
0.2314
8.M15
3.3315
• 3.2015
3.3315
3.3015
8.2815
0.3315
3.3315
3.3815
3.8315
3.0815
3.2315
8.8815
3.0315
3.8015
8.3315
0.8815
3.3815
3.8315 .
8.245
3. 158
St.II-NS
(EX)
w/ Onboard
fncic'pncs
Due To
EDC
(Theoretical)
3. 3342
3.3262
3.322S.
3.3195
3.3159
3. 3149
8. 3126
3.3103
3. 2986
8. 8863
3.8057
3.3045
3. 3034
3.3323
0.3911
0.8313
8.3315
3.3315
8. 3015
8.S315
3. 8015
8.8315
3,8315
3. £315
6. 2815
3. -3315
3. 3315
0.3315
3. S3 15
8. 8815
3.3815
3.3315
3. 3015
8.3015
3. 301E
0.223
& 141
                        F-53

-------
                      Table F-il.  Self-service  Incidence Due To EDB ftrid EDC - Theoretical
Year
St.II-Nfl     St.II-Nation   St.II-Nation






1SBS
1987
1988
1989
1950
1991
1992
1993
1994
1995
19%
1997
1998
1999
2888
2881
2802
2802
2834
2885
280S
2887
2883
2839
2818
2811
2812
2813
2814
2815
2816
2817
£818
2919
£828
SUM-
m *
(NO EX)
H/ Onboard
Incidence
Due To
EDC
(Theoretical)
8.3327
8.8235
0.8283
8.8174
0.0151
8.0133
0.3113
8.0892
0.6877
0.0862
0.0051
0.0841
0.0831
0.0021
0.0810
0.8012
0.0015
0.0015
8.0015
0.8815
8.8015
0.0015
8.0015
0.0015
0.0015
0.0815
8.3015
0.0815
0.0815
0.0015
0.0015
0.0815
0.0815
0.8915
8.8815
8.282
3.128
(EX)
w/ Onboard
Incidence
Due To
EDC
(Theoretical)
8.0381
0.8279
0.0151
0.8884
3.8867
0.8859
0.8058
8.8041
0.0034
3.8027
8.0023
0.8018
0.0314
0.8089
0.8005
0.0805
0.8007
0.0812
0.0815
0.0315
0.0315
0.8015
0.0315
0.0815
0.8815
0.8315
0.0315
8.0315
3.8815
3.3315
0.8015
0.3315
3.3315
8.8815
8.3815
8.152
3. 133
(NO EX)
«/ Orboard
Incidence
Due To
EDC
(Theoretical)
3.33312
3.02555
9.38896
8.38113
3.03844
0.80039
3.33033
3.8C327
8.28822
0.38018
8.33815
0.83312
3.03339
3.83385
3.39303
3.03333
8.38336
0.38184
3.33145
0.33151
3.30151
0.88151
3.33151
3.03151
9.W151
0.03151
8.23151
3.83151
3.33151
0.09151
8.30151
3.S3151
3.130151
3.08151
3.33151
0.133
3.874
                                                      F-54

-------
Table F-12.  Service Station  Incidence Due To Benzene find 3as Vaoors - in-usa,  finnuai  Insoections
Year




1986
1987
1988
1989
1990
1991
1992
1993
1994
1995
1996
1997
1998
1999
2900
2001
2002
2003
2004
2005
2006
2007
2008
2009
2010
2011
2012
2013
2014
2015
2016
2017
2013
2019
2020
SU« =
i*V =
Baseline

(Ann. Inso. )
Total
Incidence Due
To Benzene
3.2316
0.2255
0.2193
0.2131
0.2090
0.2045
3.2018
0. 1991
0. 1937
0. 1910
0.1883
0.1856
0.1829
0.1803
0. 1776
0. 1776
0. 1776
0.1776
0. 1776
0.1776
0. 1776
0.1776
0. 1776
0.1775
0. 1776
0.1775
0. 1775
3.1776
3. 1776
0.1776
0. 1775
3. 1776
0. 1775
0.1776
0. 1776
6.55
2.13
St. I-Nation
(EX)
(finn. Inso. )
Total '
Incidence Due
To Benzene
0.2316
0.2198
3.2028
0. 1917
0. 1881
0.1840
3. 1815
0. 1791
3. 1743
0. 1719
0.1594
0. 1670
0.1646
0.1622
0.1598
0.1598
3.1598
0. 1598
0.1598
0. 1598
0.1598
0.1598
0. 1598
0. 1598
3. 1598
0. 1598
3. 1598
0. 1598
3. 1598
3. 1598
3.1598
8. 1598
3. 1598
0. 1598
3. 1598
5.94
1.36
St.II-NA*
(EX)
(Ann. Inso. )
Total
Incidence Due
To Benzene
3.2153
3. 1924
3. 1845
0. 1781
0. 1747
0.1709
0. 1586
0.1664
0.1619
8. 1596
0.1574
0. 1551
8*1529
0.1506
0. 1484
0.1484
0. 1484
0. 1484
0.1484
3. 1484
0. 1484
0.1484
0.1484
0.1484
0.1484
3.1484
3. 1434
3.1484
0.1484
0.1484
0. 1434
3.1484
3. 1484
3.1484
0. 1484
5.53
1.33
St.II-Nfl
(EX)
(Ann. InsD. )
Total
Incidence Due
To Benzare
3. 1983
3. 1579
3.1483
3. 1416
0. 1389
0. 1359
3. 1341
3. 1323
3. 1287
3.1259
8. 1251
3. 1234
0.1216
0. 1193
0.1180
0.1180
8. 1130
3. 1180
3. 1130
8.1188
3. 1180
3.1180
0.1180
3.1183
0. 1180
3. 1180
3.1130
3. 11S3
0. 1133
3.1180
3. 1180
3.1188 -
3.1183
3. 1183
3. 1130
4.41
1.47
St.II-Nation
(EX)
(Ann. Inso. )
Total
Incidence Due
To Benzene
3.2315
3.2037
3. 1553
0. 1288
3.1225
3. 1198
0.1182
0.1167
8. 1135
3.1119
8.1134
0.1838
3. 1372
0. 1356
3.1840
8.1340
3. 1340
3.1340
3. 1340
0. 1043
3. 1040
8.1340
0. 1048
3. 1340
3. 1343
3. 1343
3. 1340
3. 1343
3.1040
3. 1348
3.1343
0. 1343
3. 1340
0.1343
-3. 1948
4.24
1.43
St.II-Nation
(NO EX)
(firm. InsD. )
Total
Incidence Due
3.231S
0. 1323
3. 121S
3.8831
3. 3780
3.3763
3.3753
3. 8743
3.3723
0.3713
3. 3733
3.3693
0. 3583
0.3673
3.3563
8.3653
3. 3663
3.0653
3.2663
2.3563
3.3663
%. 3563
3.0653
3.3663
3.3563
3. 2653
3. (3663
8.3663
iV3653
3.3563
3. 3663
3. £663
3. '3553
-0.0563
3.S653
£.74
1.37
Onboard-Nation
(w/ Tasncer. )
(Ann. Inso. )

Incidence Due
To Bsnzsr.a
6. 2316
3.2339
0. 1S4£
2. 1537
2. 1537
8 '317
VI iVj! :
3. 1288
3. 11 IS
3.3977
v* &Jt. 3
0.0655
3. & 13
8. 2813
8.3818
8.2810
3.2318
3. 2318
3. £815
3.2313
2. 2313
3. -:-3ia
3.3813
3.2813
8 2315
3.3318
3.0818
3.3313
3.2313
•3.0818
8.3813
3. 3818
3.77
1.54
                                        F-55

-------
       Table F-12.  Service Station Ir.cider.ee Due To Benzene find 3as Vanors - In-use.  Rnnuai Insoections
Yiar
  1586
  1987
  1588
  1989
  1930
  1991
  1592
  1993
  1994
  1995
  19%
  1997
  1998
  1999
  2000
  2001
   2003
   2084
   2085
   2006
   2007
   2008
   2009
   £010
   2011
   2012
   2013
   2014
   2015
   2015
   2017
   2018
   2019
   2020

  SUH*

  NPV =
Baseline

(Ann. Inso. )
Total
Incidence Due
To Gas Vapors
(PH., rat)
5.842
6.659
5.476
6.293
6.174
6.^39
5.950
5.880
5.721
5.642
5.552
5.483
5.403
5.324
5.244
5.244
5.244
5.244
5.244
5.244
5.244
5.244
5.244
5.244
5.244
5.244
5.244
5.244
5.244
5.244
5.244
5.244
5.244
5.244
5.244
St. I-Nation
(EX)
(flnn. Inso. )
Total
Incidence Due
To Sas Vapors
(PUL,rat)
6.842
6.480
5.954
5.617
5.511
5.390
5.319
5.248
5.107
5.036
4.365
4.894
4.823
4.752
4.681
4.681
4.681
4.681
4.681
4.681
4.681
4.681
4.681
4.681
4.681
4.681
4.681
4.681
4.681
4.681
4.581
4.681
4.681
4.681
4.581
St.II-Nfi«
(EX)
(flnn. Insn. )
Total
Incidence Due
To 3as Vapors
(PUL,rat)
6.358
5.780
5.469
5.278
5.178
5.365
4.998
4.332
'4.7S9
4.732
4.665
4.599
4.532
4.455
4.399
4.339
4.399
4.399
4.393
4.399
4.393
4.399
4.333
4.399
4.333
4.339
4.333
4.399
4.393
4.339
4.333
4.339
4.393
4.399
4.399
St.il-Nfl
(EX)
(flnn. Insp. )
Total
Incidence Due
To Gas Vapors
(PUL,rat>
5.859
4.589
4.408
4.208
4.129
4.838
3.985
3.932
3.825
3.773
3.719
3.666
3.613
3.553
3.507
3.537
, 3.507
3.587
3.507
3.537
3.507
3.507
3.507
3.507
3.507
3.5S7
3.507
3.507
3.507
3.537
3.507
3.537
3.507
3.507
3.507
St.II-Nation
(EX)
(flnn. Insp. )
Total
Ircidence Due
To Sas Vapors
OIL, rat)
6. 842
5.333
4.645
3.833
3.573
3.590
3.543
3.495
3.401
3.354
3.306
3.259
3.212
3.155
3.117
3.117
3.117
3.117
3.117
3.117
3.117
3.117
3.117
3.117
3.117
3.117
3.117
3.117
3.117
3.117
3.117
3.117
'3.117
3.117
3.117
St.II-Nation
(NO EX!
(Ann. Insp. )
Total '
Incidence Due
To Gas Vancrs
(PUL,rat)
£.342
5.535
3.S44
2.517
2.266
2.315
£.284
2.254
2.193
"2.162
£.13£
2.101
2.071
2.341
2.818
2.013
2.013
2.313
2.81-3
2.013
2.013
2.313
2.313
2.013
£.310
£.818
2.313
2.910
£.310
2.313
£.313
2.018
2. 313
2.218
£.013
193.59

 62.65
174.24

 57.47
153.15

 53.48
131.35

 43.-60
120.81

 42.75
82.82

32.20
                                                      F-56

-------
Table F-12.  Service  Station  Incidence Due To Benzene find Sas Vapors - In-use, Annual Inspections
Year



1985
1987
1988
1969
1998
1991
1992
1993
1994
1995
1996
1997
1998
1999
£000
2802
£003
2004
d805
2086
2887
2888
2809
2810
£311
£812
2813
2814
2815
2816
2017
2018
2019
£828
SUH =
M3V =
Onboard-Nation
(w/ Tamper. )
(flnn. Insp. )
Total
Incidence Due
To Sas Vaoors
(PH., rat)
6.842
6.659
6.834
5.464
5.814
4.579
' 4.243
3.943
3.600
3.358
3.142
2.952
2.783
2.625
2.488
2.488
2.488
£.488
2.488
2.488
2.488
2.488
2.488
2.488
2.488
2.488
2.488
2.488
2.488
2.488
2.488
2.488
2.488
2.488
2.488
113.48
45.98
Baseline
(flnn. Insp. )
Total
Incidence Due
To Gas Vapors
(PUMice)
4.15
4.84
3.93
3.82
3.74
3.66
3.61
3.56
3.47
3.42
3.37
3.32
3.28
3.23
3.18
3.18
3.18
3.18
3.18
3.18
3. 18
3.18
3.18
3.18
3.18
3.18
3.18
3.18
3.18
3.18
3.18
3.18
3.18
3.18
3.18
117.36
38.19
St.l-Nation
- (EX)
(flnn. Insp. )
Total
Incidence Due
To Sas Vapors
(PUL.raice)
4.15
3.93
3.51
3.41
3.34
3.27
3.££
3.18
' 3.10
3.05
3.31
2.97
2.92
2.88
2.84
2.84
2.84
2*84
2.84
2.84
2.84
2.84
2.84
2.84
2.84
2.84
2.84
2.84
2.84
2.84
2.84
2.84
2.84
2.84
2.84
185.63
34.34
St. II-Nfl*
(EX)
(flnn. Insp. )
Total
Incidence Due
To Eas Vapors
(PUL.Hice)
3.66
3.46'
3.32
3.28
3.14
3.07
3.33
2.99
2.91
2.87
£.83
2.79
2.75
2.71
2.67
2.67
2.67
2.67
2.67
2.67
2.67
2.67
2.67
2.57
2.57
2.67
2.67
2.67
2.67
2.57
2.57
2.67
2.S7
2.67
2.67
98.91
32.4?
St. II-Nfl
SEX)
(flnn. In=c. )
Total
Incidence Due
To Sas Vapors
(PL'L, mice)
3.55
2.S4
2.67
2.55
2.53
2.45
2.42
2.38
2.32
2.29
2.25
2.22
2.39
2.16
2.13
2.13
2.13
2.13
2.13
2.13
2.13
£.13
2.13
2.13
2.13
2.13
2.13
2.13
2.13
2.13
2.13
2.13
2.13
2.13
2.13
79.45
?A i?
St. II-Nation
(EX)
(finr>. Insp. )

Incidence Due
To Gas Vapors
iPULjMice)
4.15
- ff
u. uu
2. 2°
2.22
2 IS
I~« Aw
2.15
2.1£
2.86
" 2 SR
l»4 OO
£.88
i 9S
1* JO
1 95
*» Jw
1.92
1.89
i.89
1 flq
*. OJ
1.39

1m QJ
1.89
1.89
i 89
A« \JJ
i an
A« U-/
i 39
*• \JJ
< aq
*• Q J
1 flQ
*» 33
i.83
1 39
A* w J
i aq
A* U.7
i.SS
. 1.89
73.24
oe; n<
                                                                                         25.91
                                           F-57

-------
Table F-12.  Service Station Incidence Due To Benzene find Gas Vapors - In-use.  flnnual  Insoections
Year






1986
1987
1988
1989
1999
1991
1992
1993
1994
1995
19%
1997
1998
1999
20®
2091
2902
2083
2884
2835
20%
2807
2808
2809
2818
2811
£812
2913
2814
2815
2816
2817
2818
2819
2828
SUMs
NPV =
St.II-Nation Onboard-Nation
(NO EX)
(Ann. Inso. )
Total
Incidence Due
To Gas Vapors
(PUL,aice)
4.15
3.45
2.21
1.53
1.43
1.48
1.38
1.37
1.33
1.31
1.29
1.27
1.26
1.24
1.22
1.22
1.22
1.22
1.22
1.22
1.22
1.22
1.2E
1.22
1.22
1.22
1.22
1.22
1.22
1.22
1.22
1.22
1.22
1.22
1.22
58.21
19.52
(H/ Tastier.)
(firm. Inso. )
Total
Incidence Due
To Gas Vapors
(PUL,aice)
4.15
4.84
3.66
3.31
3.84
2.78
2.57
2.39
2.18
2.64
1.98
1.79
1.69
1.59
1.51
1.51
1.51
1.51
1.51
1.51
1.51
1.51
1.51
1.51 "
1.51
1.51
1.51
1.51
1.51
1.51
1.51
1.51
1.51
1.51
1.51
58.79
27.83
Baseline

(flnn. Insp. >
Total
Incidence Due
To Sas Vaoors
(?LE,rat>
3.895
3.792
3.687
3.583
3.516
3.439
3.393
3.348
'3.258
3.212
3. 167
3.122
3.877
3.831
2.986
2.986
2.986
2.986
2.986
2.986
2.986
2.986
2.385
2.986
2.985
£.986
2.986
2.586
2.985
2.986
2.986
2.986
2.985
2.986
2.986
113.23
35.79
St. I-Nation
(EX)
fftnn. Inso. )
Total
Incidence Due
To Gas Vapors
(ME, rat)
3.896
3.598
3.398
3.198
3.138
3.069
3.029
2.988
2.908
2.367
2.827
2.787
2.745
2.786
2.665
2.655
2.665
2.665
2.665
2.655
2.555
2.665
2.665
£.665
2.665
2.665
2.665
2.665
2.565
2.665
2.565
2.665
2.665
2.665
2.565
99.21
32.72
St. I MB*
(EX)
(flnn. Inso. )
Total'
Incidence Due
To Gas Vapors
OLE, rat)
3.525
3.246
3.114
3.3K
2.949
2.884
2.346
2.808
2.732
2.594
2.655
2.618
2.581
2.543
2.585
2.535
2.505
2.585
2.505
2.535
2.583
2.585
2.585
2.585
2.595
2.585
2.505
2.585
2.585
2.585
2. 585
2.585
£.585
2.535
£.585
92.93
38.45
St.II-Nft
(EX)
(flnn. Insp. )
Total
Incidence Due
To Sas Vaoorc
(MLE,rat)
3.342
2.578
2.513
2.396
2.351
2.299
2.269
2.239
2. 178
'2.148
2. US
2.088
2.857
2.027
1.397
1.997
1.997
1.997
1.997
1.997
1.937
1.997
1.937
1.997
i.937
i.397
i.937
1.397
1.997
1.997
1.997
1.997
1.397
1.397
1.957
74.53
24.33
                                                F-58

-------
Table F-i£.  Service Station  Incidence Due To Benzene find Sas Vaoors - In-use,  Annual  Insosctions
Year






1935
1987
1988
1989
1990
1991
1992
1993
1994
1995
1996
1997
1998
1999
2088
2001
2802
2083
2004
2005
20%
2007
2008
2009
2818
2011
2012
2813
2014
2015
2016
2017
2318
S019
2020
SUM =
!*V =
St. Il-Nation
(EX)
(flnn. InsD. )
Total
Incidence Due
To Sas Vapors
(ME, rat)
3.396
3.433
2.S45
2.182
2.090
2.044
2.817
1.990
1.935
1.910
1.383
1.855
1.829
1.802
1.775
1.775
1.775
1.775
1.775
1.775
1.775
1.775
1.775
1.775
1.775
1.775
1.775
1.775
1.775
1.775
1.775
1.775
1.775
1.775
1.775
68.79
24.34
St.II-Nation
(NO EX)
(flnn. Inso. )
Total
Incidence Due
To Sas vapors
(ME, rat)
3.395
3.237
2.875
1.433
1.347
1.318
1.301
1.283
1.249
1.231
1.214
1.197
1.179
1.162
1.145
1.145
1.145
1.145
1.145
1.145
1.145
1.145
1.145
1.145
1.145
1.145
1.145
1.145
1.145
1.145
1.145
1.145
1.145
1.145
1.145
47.15
18.33
Onboard-Nation
(w/ Taaper. >
(flnn, Inso. )
Total
Incidence Due
To Sas Vapors
(ME, rat 5
3.395
3.792
3.435
3.111
2.855
2.507
2.415
2.245
2.858
1.912
1.789
1.581
1.585
1.495
1.417
1.417
1.417
1.417
1.417
1.417
1.417
1.417
1.417
1.417
1.417
1.417
1.417
1.417
1.417
1.417
1.417
1.417
1.417
1.417
1.417
54.62
26. 14
Baseline

(flnn. Inso. )
Total
Incidence Due
To Bas Vacors
(ME, nice)
2.73
2.72
2.64
2.57
2.52
2.45
2.43
2.49
2.33
2.30
2.27
2.24
2.20
2.17
2.14
2.14
2. 14
2.14
2.14
2.14
2.14
2.14
£.14
2.14
£.14
2.14
2.14
2.14
2.14
2.14
2.14 .
2.14
2.14
2.14
2. 14
78.97
£5.64
St.I-Nation
(EX)
(flnn. Ir.so. }
Total
Incidence Due
To Gas Vapors
(ME. fliics}
2.79
2.64
£.43
2.29'
2.25
2.29
2.17
2.14
2.128
2.35
2.83
2.30
1.97
1.94
1.91
1.91
1.91
1.91
1.91
1.91
1.91
1.91
1.91
1.91
1.91
LSI
1.91
1.31
1.91
1.91
1.91
- 1.91
1.91
1.91
1.91
71.88
£3.44
St. II-Nfl*
(EX)
(flnn. InsD. )
Total
Incidence Due
To Sas Vapors
(ME. a ice)
2.68
£.33
£.23
£.15
2.11
£.37
2.04
£.31
•1.95
" 1.93
1.90
1.S8
1.85
1.82
1.79
1.79
1.79
1.79
1.79
1.79
1.79
1. 79
1.79
1.79
1.79
1.79
1.79
1.73
1.79
1.79
1.79
1.73
1.79
1.73
1.73
66. 5&
£1.82
                                        F-59

-------
     Table F-12.  Service Station Incidence Due To Benzene find  Gas  Vaoors -  Inruse. fir.nual  inspections
ar






1986
1987
1988
1989
1923
1991
1992
1993
1994
1995
1996
1997
1998
1999
2000
mi
2302
£833
2384
2005
2006
2007
2008
2039
2010
2011
2812
2013
2014
2015
2016
2017
2018
2019
2020
St.II-Nfl
(EX)
(flnn. Iriso. )
Total
Incidence Due
To 5as Vaoors
(!
-------
Table F-13.  Service Station  Incidence Cue To Benzene find Sas Vaoors - in-use,  No  Inscections
/ear





1986
1987
1988
1989
1990
1991
1992
1993
1994
1995
19%
1997
1998
1999
2200
2301
2002
2003
2004
2005
2336
£307
2338
am
2313
2011
2212
2S13
2314
2315
2216
3317
2S18
2319
£S23
Baseline

(flnn. Inso. )
Total
Incidence Due
To Benzene
3.2315
0.2255
0.2193
0.2131
0.2096
3.2045
•0.2018
0. 1991
8. 1937
0. 1910
0.1883
0.1856
0.1329
0.1803
0. 1776
0. 1776
0. 1776
0.1775
0.1776
0. 1775
8. 1776
3. 1776
0. 1776
0. 1775
3. 1775
3. 1775
3.1775
0. 1775
0.1776
3. 1775
3, 1775
3. 1775
3. 1775
3. 1776
e.1775
St.II-NA*
EX)
(No Inso.)
Total
Incidence Due
To Benzene
3.2247
0.2113
0.2044
0. 1981
0.1944
0. 1901
0. 1876
0. 1851
0.1801
0. 1776
0.1751
0. 1726
0.1701
0. 1675
0.1651
0. 1651
3. 1651
0. 1651
3. 1651
3.1651
3.1651
3. 1551
3. 1651
8. 1651
3.1551
0. 1551
3. 1651
3. 1551
3. 1651
3. 1651
3. 1651
3. 1651
0. 1651
3. 1S51
3. 1551
St.II-NA '
(EX)
(No Insp.)
Total
Incidence Due
To Benzene
8.2123
3.1363
0.1782
0. 1717
0.1684
0. 1647
0.1626
8.1604
0. 1561
0.1539
0.1517
0.1496
0.1474
0.1452
0.1431
0. 1431
0.1431
0.1431
8. 1431
3. 1431
0.1431
0. 1431
0.1431
3. 1431
0. 1431
3. 1431
3. 1431
3. 1431
3. 1431
3. 1431
3. 1431
3. 1431
3. 1431
3. 1431
8. 1431
St. II-Nation
(EX)
(No Insp.)
Total
Incidence Due
To Benzene
0.2316
3.2128
0. 1825
0.1638
3.1589
3. 1554
. 3. 1534
0. 1513
0.1472
0.1452
0.1431
8. 1411
0.1390
0. 1370
0.1350
8. 1350
8. 1350
3.1350
3.1350
8.1353
3. 1350
8. 1353
3. 1350
0. 1350
3. 1353
0. 1358
3.1353
3. 1353
0. 1353
3,1350
8. 1350
3. 1358
0. 1353
3. 1350
3. 1358
St.II-Nation
(NO EX)
(No Insp. )
Total
Incidence Due
To Benzene
0.2315
3.2345
3. 1584
0. 1319
8. 1272
3.1244
3. 1228
3. 1211
0. 1179
0. 1152
0.1145
0.1129
0.1113
0.1397
0.1088
3.1088
0. 1880
3. 1083
8. 1880
3. 13S8
3.1388
8. 1880
8. 1083
3.1330
3. 13£3
3.1380
3. 1080
3. 1380
0. 1380
3. 1383
3. 1388
3. 1088

3. 1383
3. 1383
Onboard-Nation
(w/ Tasoer. )
(flnn. Inso. )
Total
Incidence Due
To Benzene
3.2316
3.2255
3.2039
3. 1842
3. 1587
3. 1537
3.1421
3. 1317
0.1200
3.1116
8.1342
0.0977
3.9919
8.3655
3.3818
8.0818
0.0818
3.3818
3.3818
0. 2818
3. 3318
3.0813
3.3818
3.3818
8.3813
0.3318
8. 0818
0.3813
3.3818
3.3818
3.3818
8. 3818
3. 3813
3. -3818
3.3818.
5.55




2.13
5.11




1.99
                                   5.31
5.10




1.73
                                                                 4.17




                                                                 1.47
3.77




1.54
                                       F-61

-------
        "acls F-S3; -Ssrvice Stat::-n Ircidsncs Sue To Benasne find Gas  Vaaors -  In-usa, Mo InsDections







1S23
* "27
--W/
.538
1339
.5%
1931
1?:2
1S33
122*
«9SS
!9S=
19S7
less
1939
£823
£?*i
£332
cees
ia?4
2M5
28-36
£237
•sea
2^
281S
sen
22!£
£213
sei4
£315
=216
£817
£318
£319
£223
Baseline

ffir.n. Insc. 5
Total
Incidence Due
"o Sas Vapors

-------
Table F-13.  Service Station iraidsnce Dus To Benzene find Sas Vapors - In-use.  No Inspections
Year






133S
1337
1SS3
195?
1932
19?1
13-32
1533
1934
1935
1996
1997
1993
1393
mm
£081
2832
2803
2M4
2805
2205
2827
£038
2839
2310
2311
2312
2313
2814
£215
£815
2317
2818
2013
2323
Baseline

(Ann. Insp. )
Total'
Incidence Due
To Gas Vapors
«'Pll,iiica)
4.13
4.34
3.93
3.32
3.74
3.S6
3.51
3.55
3.47
3.42
3.37
3.32
3.28
3.23
3.18
3.13
3.13
3.13
3.18
3.13
3.18
3.13
3.18
3.13
3.18
3.18
3.13
3.:8
3.13
3.18
3.18
3.13
3.13
3.18
3.13
St.II-NA*
(EX)
(No InsD.)
Total
Irsciderice Due
To Sas Vapors
(PUL,sice}
4.22
3.73
3.65
3.55
3.48
3.41
3.36
3.32
3.23
3.18
3.14
3.09
3.35
3.30
2.96
2.96
£.36
2.95
2.95
2.96
2.36
2.36
2.36
2.96
2.96
2.36
2.96
2.95
2.95
2.96
2.96
2.95
2.36
2.95
2.96
St.II-Mfl
(EX)
(No Inso.)
Total
Incidence Due
To Sas Vapors
(P'JL,mice)
3. as
3.34
3.19
3.38
3.02
2.95
2.91
2.87
2.S0
2.76
2.72
2.68
2.54
2.60
2.56
2.56
2.55
2.56
2.56
2.56
2.55
2.55
2.56
2.55
2.55
2.56
2.56
2.56
2.56
2.56 .
2.55
2.56
2.55
2.55
2,55
St.II-Nation
(EX)
(Mo Inso.)
Total
Incidence Due
To Sas vapors
(PUL, mice}
4.15
3.81
3.23
2.95
2.86
2.80
2.76
2.72
2.55
2.61
2.53
2.54
2.50
2.47
2.43
2.43
2.43
2.43
2.43
2.43
2,43
2.43
2.43
2.43
2.43
2.43
2.43
2.43
2.43
2.43
2.43
2.43
2.43
2.43
2.43
St.II-Nation
(NO EX)
(No -Inso. )
Total
Incidence Due
To Sas Vapors
(PUL, slice)
4.15
3.67
2.85
2.33
2.29
2.24
2.21
2.18
2.12
2.139
2. '37
2.34
2.31
1.98
1.95
1.95
•1.95
1.95
1.95
1.95
1.95
1.95
1.95
1.95
1.35
1.35
1.35
1.35
1.95
1.95
1.95
1.95
1.95
1.55
1.95
Onboard-Nation
(w/ Tamoer, )
(flnn. Inso. )
Total
Incidence Due
To Sas Vanors
(PIL.aice)
4.15
4.04
3.56
3.31
3.134
2.73
2.57
2.39
2.13
2.34
1.30
1.79
1.69
1.53
.1.51
1.51
1.51
1.51
1.51
1.51
1.51
1.51
1.51
1.51
1.51
1.51
1.51
1.51
1.51
1.51
1.51
1.51
1.51
1.51
1.51
   117.35




    35.12
139.48




 35,55
95.19




31,23
91.59




31.31
75.16




26.43
53.79




27.33
                                    F-63

-------
Tafaie ~-:3.  Service Station Incidence Due To Berizene  find  Gas Vaoors - In-use, No Insoections
Yaar






• C3C
« .~C1
.;c<
:ssa
1989
1993
1«S!
i«X
1953
19:4
1S95
1595
1997
:938
1399
s?a>
2231
£222
2233
r?34
2205
226
2M7
•3S8
;eag
2318
2811
1812
2213
£314
£815
3?16
£817
$18
£819
•?28
Sift*
•43V =
Baseline

(flnn. !n=n. 5
Total
Incidence Dae
To Sas Vaacrs
f ME, rat)
3.S96
3.792
3.687
MI wQu
3.516
3.439
3.393
3.343
3.258
3.212
3.167
3.122
3.377
3.831
2.986
2.986
£.986
2.986
2.986
2.986
2.986
2.985
2.986
2.985
2.986
2.985
2.986
2.985
2.986
2.985
£.986
2.985
2. 586
2.985
2. '336
113.23
35.79
St. II-NA*
EX)
(No Inso.)
"otal
Incidence 3ue
To Sas Vaaors
(MLE.rat)
3.779
3.556
3.440
3.334
3.271
3.199
3.157
3.115
3.331
2.989
2.947
2.985
2.853
2.820
2.778
2.778
2.778
2.778
2.778
2.778
£.778
2.778
2.778
2.778
2.778
2.778
2.778
2.778
2.778
2.778
£.778
2.778
2.778
2.778
2.778
102.75
33.49
St. II-Nfl
(EX)
(No Inso. )
Total
Ir-cicenca Due
To Sas Vaoors
(MLE,;«at)
3.572
3.135
2.999
2.889
2.834
2.772
2.736
2.699
2.626
2.590
2.553
2.517
2.489
2.444
2.497
2.487
2.487
2.407
£.407
2.437
2.407
2.407
2.487
2.487
2.407
2.487
2.407
2.487
2.407
2.487
2.407
2.487
2.407
2.487
2.437
89.40
29.38
St. I I -Nat ion
(EX!
(No Inso.)
Total
Ircidencs Due
To Bas Vaoors
CMLE,rat)
3.895
3.583
3.881
2.758
2.686
2.627
2.593
2.558
2.489
2.454
2.420
2.385
2.351
2.316
2.232
2.282
2.282
2.282
2.282
£.282
2.282
2.282
2.282
2.282
2.282
£.282
2.282
2.282
2.282
2.282
2.232
2.282
2.282
2.282
2.282
86.12
29.13
St.II-Nation Onboard-Nation
(NO EX)
(No Inso. )
Total
Incidence Due
To Sas Vaoors
(MLE.rat)
3.896
3.443
2.674
2.232
2.153
2.186
£.878
2.351
1.995
1.367
1.948
1.912
1.884
1.857
1.829
1.329
1.829
1.329
1.329
1.829
1.829
1.829
1.829
1.829
1.329
1.329
1.329
1.829
1.829
1.829
1.329
1.329
1.329
1.829
1.829
78.68
24.82
(w/ Taffioer.)
(flnn. Inso. )
Total ,
Incidence Due
To Sas Vaoors
(ME, rat)
3.396
3.792
3.436
3.111
2.355
£.687
2.416
2.245
2.853
1.912
1.789
1.681
1.585
1.495
1.417
1.417
1.417
1.417
1.417
1.417
1.417
1.417
1.417
1.417
1.417
1.417
1.417
1.417
1.417
1.417
1.417
1.417
1.417
1.417
1.417
54.62
26.14
                                       F-64

-------
Table F-13.  Service Station  Incinence Due To Benzene find Gas Vapors - In-use,  No  Inspections
Year






1"85
1*87
1588
-.389
1C32
- ~CJ.
1992
1553
139'
;9'r5
19%
1997
1998
1999
2808
2801
2082
2833
2384
£335
£886
£887
2888
2889
£818
£211
2012
313
2814
2315
2816
2817
2318
£319
2823
SUM =
NPV =
Baseline

(flrei. InsD. )
Total
iTiciusree Due
To 035 Vapors
«LE,aice)
£.79
2.72
2.54
2.57
2.52
2.46
. 2.43
2.48
2.33
£.38
2.27
2.24
2.20
2.17
2.14
2.14
2.14
2.14
2.14
2.14
2.14
2.14
£.14
£.14
2.14
2.14
2.14
£.14
2.14
£.14
2.14
2.14
2.14
2.14
£.14
78.97
£5.54
St.II-Nfl*
(EX)
(No Insp.)
Total
Incidence Due
To Sas Vapors
(MLE, fitice)
£.71
£.55,
2.46
2.39
£.34
2. £9
£.26
2.23
2.17
£.14
2.11
2.88
£.05
2.82
1.99
1.99
1.99
1.99
1.99
1.99
1.99
1.99
1.99
1.99
1.99
1.99
1.99
1.99
1.99
1.99
1.99
1.99
1.99
1.99
1.99
73.61
23.99
St.II-Nfl •
(EX)
(No Insp.)
Total
Incidence Due
To Sas Vaoors
(MLE, sice)
2.55
£.25
2.15
2.37
2.83
1.99
1.95
1.93
1.88
1.86
1.33
1.38
1.73
1.75
1.72
1.72
1.72
1.72
1.72
1.72
- 1.72
1.72
1.72
1.72
1.72
1.72
1.72
1.72
1.72
1.72
1.72
1.72
1.72
1.72
1.72
64.85
21.35
St.II-Nation
(EX!
(No Inso.)
Total
Incidence Due
To Gas Vaoors
MLE, Bice)
2.79
2.57
2.21
1.98
1.92
1.88
1.85
1.83
1.78
1.76
1.73
1.71
1.68
1.56
1.63
1.53
1.63
1.63
. 1.63
1.53
1.63
1.53
1.63
1.63
1.63
1.63
1.63
1.63
1.53
1.63
1.63
1.53
1.63
1.53
1.53
51.78
20.87
St.II-Nation
(NO EX)
(No Inso.)
Total .
Incidence Due
To Gas Vaoors
(!4LE,3ice)
2.79
£.47
1.92
1.50
1.54
1.51
1.49
1.47
1.43
1.41
1.39
1.37
1.35
1.33
1.31
1.31
1.31
1.31
1.31
1.31
1.31
1.31
1.31
1.31
1.31
1.31
1.31
1.31
1.31
1.31
1.31
1.31
i.3i
1.21
1.31
58.58
17.78
Onboard-Nation
(w/ Tanner. )
iflnn. IKSD. )
Total
Incidence Due
To Sas Vaoors
(«LE,raice)
2.79
2.72
2.46
£.£3
2.35
1.87
1.73
i.51
1.47
1.37
1.28
1.28
1.14
1.07
1.81
1.01
1.31
1.31
1.01
1.31
1.81
1.81
1.01
1.31
1.01
1.31
1. 81
1.31
1.31
1.01
1.81
1.81
LSI
1.81
1.81
45. £9
18.72
                                     F-65

-------
Table F-14.  Self-Service Incidence Due To Benzene find Gas Vaoors - In-Use
ar





1986
1987
1988
1989
1990
1991
1992
1993
1994
1995
19%
1997
1998
1999
2336
2901
2002
2303
2204
2S85
2006
2287
2008
2239
2810
2011
2912
2013
2914
2815
2016
2017
2818
2019
2820
Ifil s
PV 2
Baseline

Incidence
Due To
Benzene
(In-use, ftnn. 5
4.39
4.22
4.06
3.91
3.79
3.68
3.59
3.51
3.39
3.31
3.24
3.18
3.11
3.05
2.98
2.98
2.98
2.98
2.98
2.98
2.98
2.98
2.98
2.98
2.98
2.98
2.98
2.98
2.98
2.98
2.98
2.98
2.98
2.98
2.98
113.06
37.91
St.II-Nfl*
(EX)
Incidence
Due To
Benzene
(In-use, ftnn. )
4.24
3.93
3.76
3.51
3.50
3.40
3.32
3.24
3.13
3.06
3.00
2.93
2.87
2.31
2.75
2.75
2.75
2.75
2.75
2.75
2.75
2.75
2.75
2.75
2.75
2.75
2.75
2.75
2.75
2.75
2.75
2.75
2.75
2.75
2.75
104.60
35.23
St.II-Nfl
(EX)
Incidence
Due To
Benzene
(In-use, flnn.)
3.99
3.43
3.24
3.09
3.00
2.91
2.84
2.77
2.68
2.62
2.56
2.51
2.46
2.41
2.35
2.36
2.36
2.35
2.36
2.36
2.36
2.36
2.36
2.36
2.36
2.36
2.36
2.36
2.36
2.36
2.36
2.35
2.36
2.36
2.35
89.98
30.59
St.II-Nation
(EX)
Incidtsnce
Due To
Benzene
(In-use, flnn. )
4.39
3.63
2.35
1.62
1.49
1.45
1.41
1.38
' 1.33
1.30
1.28
1.25
1.23
1.20
1.17
1.17
1.17
1.17
1.17
1.17
1.17
1.17
1.17
1.17
1.17
1.17
1.17
1.17
1.17
1.17
1.17
1.17
1.17
1.17
1.17
49.95
20.05
St.II-Nation
(NO EX)
Incidence
Due To
Benzene
(In-use, ftnn. )
4.39
3.37
1.60
0.66
0.56
0.54
0.53
8.51
0.50
0.49
0.48
0.47
0.46
0.45
0.44
0.44
0.44
0.44
0.44
0.44
0.44
0.44
0.44
0.44
0.44
0.44
0.44
0.44
0.44
0.44
0.44
0.44
0.44
0.44
0.44
24.17
12.71
St. II-Nft*
(EX)
Incidence
Due To
Benzene
(In-use, No)
4.43
4.16
3.99
3.83
3.72
3.61
3.52
3.44
3.32
3.25
3.18
3.12
3.05
2.99
2.92
. 2.92
2.92
2.92
2.92
2.92
2.92
2.92
2.92
2.92
2.92
2.92
2.92
2.92
2.92
2.92
2.92
2.92
2.92
2.92
2.92
111.02
37.33
St.II-Nfl
(EX)
Incidence
Due To
Benzene
(In-use, No)
4.27
3.84
3.65
3.50
3.39
3.29
. 3.22
3.14
3.03
-2.96
2.90
2.84
2.79
2.73
2.67
2.67
2.67
2.67
2.67
2.57
2.67
2.67
2.67
2.67
2.67
2.57
2.67
2.67
2.67
2.67
2.67
2.67
2.67
2.67
2.57
101.58
34.33
St.II-Nationl
(EX) 1
Incidence 1
Due To 1
Benzene 1
(In-use, No) 1
4.52J
3.951
3.861
2.531
2.401
2.33J
2.28J
2.221
2.15J
2.101
2.861
2.0il
1.971
1.931
1.89J
1.89J
1.891
1.391
1.39
1.89
1.89
1.89
1.39
1.39
1.39
1.89
1.89
1.89
1.89
1.39
1.89
1.89
1.89
1.89
1.89
TS.igl
27.37
                                F-66

-------
Table F-14.  Self-Service Incidence Due To Benzene find Sas Vapors - In-Use
Year





1986
1987
1988
1989
1990
1991
1992
1993
1994
1995
19%
1997
1998
1999
2000
2081
2802
2003
2024
2005
£906
2007
2008
2389
2310
2011
2012
2813
2014
2015
2016
2017
2018
2019
2020
iM =
'V =
St. II-Nation
(NO EX)
Incidence
Due To
Benzene
(In-use.No)
4.52
3.79
2.58
1.90
1.79
1.74
1.70
1.66
1.60
1.56
1.53
1.50
1.47
1.44
1.41
1.41
1.41
1.41
1.41
1.41
1.41
1.41
1.41
1.41
1.41
1.41
1.41
1.41
1.41
1.41
1.41
1.41
1.41
1.41
1.41
58.35
22.57
Onboard
W/ Taaoer.
Incidence
Due To
Benzene
(In-use.flnn.)
4.39
4.22
3.69
3.21
2.83
2.48
2.20
1.95
1.70
1.51
1.35
1.21
1.39
0.97
0.87
0.87
0.87
0.87
0.87
0.87
0.87
0.87
0.87
0.87
0.87
0.87
3.37
0.87
3.37
0.87
0.87
3.87
0.87
3.87
0.87
51.16
24.63
Baseline

Incidence
Due To
Gas Vapors
(PUL,rat>
(In-use,flnn.)
40.58
39.24
37.%
36.69
35.79
34.87
34.22
33.58
32; 54
31.95
31.41
30.36
30.32
29.78
29.23
29.23
29.23
29.23
29.23
29.23
29.23
29.23
29.23
29.23
29.23
29.23
29.23
29.23
29.23
29.23
29.23
29.23
29.23
29.23
29.23
St. II-Nfl*
(EX)
Incidence
Due To
Gas Vapors
(PUL,rat)
(In-use,flnn. )
39.24
36.54
35.14
33.87
33.04
32.19
31.59
. 31.00
30.04
29.49
28.99
28.49
27.99
27.49
26.98
25.98
26.98
25.98
25.98
25.98
25.98
26.98
25.98
25.98
25.98
26.98
26.98
26.93
26.98
26.98
26.98
26.96
26.33
26.98
26. SB
St. II-Nfl
(EX)
Incidence
Due To
Sas Vapors
(Pit, rat)
(In-use,flnn. )
35.92
31.88
30.28
28.99
28.28
27.55
•27.04
26.53
25.71
25.24
24.81
24.38
23.%
23.53
23.10
23. 10
23.10
23.10
23.10
23.10
23.10
23.10
23.10
23.10
23. 10
23.10
23.10
23.10
23.10
23.13
23. 10
23. 10 .
23.13
23. 10
23.13
St. II-Nation
(EX)
Incidence
Due To
Gas Vapors
(PUL,rat)
(In-use.flnn.)
40.62
33.72
21.92
15.24
14.08
13.72
13.47
13.21
12.81
12.57
12.36
12.14
11.93
11.72
11.50
11.50
11.50
11.50
11.50
11.50
11.50
11.50
11.50
11.50
11.50
11.50
11.50
11.50
.11.50
11.50
11.53
,11.53
11.50
11.50
11.53
St. II-Nation
(NO EX)
Incidence
Due To
Gas Vapors
(PUL,rat)
(In-use.flnn.)
40.62
31.33
15.00
6.22
5.24
5.11
5.02
4.92
4.77
4.68
4.60
4.52
4.44
4.36
4.28
4. 28
4.28
4.28
4.28
4.28
4.28
4.28
4.23
4.28
4.23
4.23
4.28
4.28
4.23
4.28
A. 23
4.23
4.23
4, £3
4.23
                           1093.69




                           351.01
1011.75




 335.40
870.15




231.13
481.27




1SS.73
                                                                                   115.25
                           F-67

-------
                   Table F-:4. Self-Service Incidence Due To Benzsrs Snc- Sas Vapors - In-Use
ar






1986
1987
1588
1989
1950
1991
1992
1993
1994
1995
1995
1997
1998
1999
em
£901
3202
2883
2334
mm
£$86
2887
2808
£889
2813
2811
2812
2813
2014
2315
2816
2817
2818
2919
£828
St.II-Nfl*
(EX)
Incidence
Due To
Gas Vaoors
CPUL,rat)
(In-use,No)
40.94
38.58
37.29
35.98
35. IB
34.19
33.56
32.93
31.91
31.33
38.88
30.25
29.73
29.28
28.67
28.67
28.67
28.67
28.67
28.57
£8.67
28.57
28.67
28.67
28.67
28.67
£8.67
28.67
£8.67
£8.67
£8.67
28.67
£6.67
£8.67
£8.67
St.IHB
(EX)
Incidence
Due To
Gas Vapors
iPUL,rat)
(In-use, No)
39.45
35.68
34.15
32.83
32.02
31.20
33.62
30.04
29.12
£8.59
28.10
27.61
27.13
26.64
26.16
26.15
£6.16
26.16
25.16
26.16
25.15
26.16
£5. 16
25.16
£5. 16
25.16
25.16
25.16
26.15
26.16
25.16
25.16
25.16
25.16
26.16
St.II-Nation
(EX)
Incidence
Due To
Sas Vapors
(?'JL,rat)
(In-use, No)
41.31
36.38
28.61
23.77
22.67
£2.09
21.58
21.27
£0.6£
£0.£4
19.98
19.55
19.21
18.86
18.52
18.52
18.52
IB. 52.
18.52
18.52
18.52
18.52
18.52
18.52
18.52
18.52
18.52
18.52
18.52
18.52
18.52
18.52
18.52
18.52
18.52
St.II-Nation
(NO EX)
Incidence
Due To
Gas Vapors
{?UL,rat)
(In-use,'fo)
41.81
35.24
24.89
17.88
16. '98
15.46
16. 15
15.86
- 15.37
15.89
14.83
14.57
14.32
14.06
13.89
13.8&
13.88
13.88
13.88
13.80
13.80
13.88
13.80
13.80
13.88
13.23
13.80
' 13.33
13.80
13.80
13. S0
13.68
13.80
13.88
13.30
Onboard
W/ Tamper.
Incidence
Due To
Gas Vapors
(PUL,rat)
(In-use, Arm. i
40. 58
3?. £4
34.23
£3.34
26. 33
23.34
23.44
18. 16
15.38
14.35
12.57
11.29
10. 17
9.12
8..22
8.22
8.22
8.22
3.22
3.22
8.22
8.22
3.2£
3.22
3.22
8.22
3.2E
8.22
8.22
8.22
8.22
8.22
8.22
8.22
3.22
Baseline

Incidence
Due "o
3as Vapors
'(PUL. slice)
24.63
£3.79
£3.81
£2.24
£1.73
21.1*
28.75
28.35
. 19.73
19.37-
19.24
18.71
18.35
13.35
17.72
17.72
17.72
17.72
•7.72
17.72
17.72
17.72
17.72
17.72
17.72
17.72
17.72
17.72
17.72
17.72
17.72
17.72
17. 7£
17.72
17.72
St.II-Nfi*
(EX)
Ittcitiencs
Due To
Gas Vapors
(PULraicsi
£3.79
22.15
21.33
23.53
2J.03
19. 51
IS. 15
13.79
15.51
17.33
17.57
17.27
16.57
16.66
16.35
16.35
16.35
16.35
16. 35
15.35
iS.35
15.35
15.35
IS. 35
1S.3S
15.36
1S.3S
16.35
15,35
18.35
16. ZS
15.35
15.35
15.35
16.35
SUH «



NPV*
1073.91



 355.48
982.47



325.32
725.%



259,81
552.49



£13.81
477.55



223.74
663.33



£18.33
 61 -7 ~.r
 -3. wO




263.33
                                                   F-68

-------
Table F-14.  Self-Service Inciderce Due To Benzene Srid Gas Vasors  - In-use
Year






1986
1987
1588
1989
1990
1991
1992
1993
1994
1995
19%
1997
1998
1999
2000
2001
2002
2003
2004
2005
2006
2007
2008
2009
2010
2011
2012
2013
2014
2015
2015
2017
2018
2019
2020
SUM =
NPV =
St.II-Nfl
(EX)
Incidence
Due To
Sas Vapors
(PUL,sice)
(In-use,flnn.)
22.38
19.33
18.35
17.57
17.14
16.70
•16.39
15.03
15.59
15.30
15.04
14.78
14.52
14.26
14.00
14.00
14.00
14.00
14.00
14.00
14.00
14.00
14.00
14.00
14.20
14.00
14.00
14.30
14.00
14.00
14.08
14.20
14.00
14.80
14.00
527.51
176. 49
St. II-Nation
(EX)
Incidence
Due To
Sas Vapors
(POL,r8ice)
(In-use,flnn.)
24.52
20.44
13.29
9.24
8.54
8.32
8.16
3.01
7.75
7.62
7.49
7.36
7.23
7.10
5.97
6.97
5.97
5.97
6.97
6.97
6.97
. 6.97
6.97
6.97
5.97
6.97
5.97
6.97
6.97
6.97
5.97
5.97
6.97
6.97
6.97
291.54
115.00
St. II-Nation
(NO EX)
Incidence
Due To
Sas Vapors
(PUL,rsice)
(In-use.Ann. )
24.52
18.99
9.09
3.77
3.18
3.10
3.04
2.98
2.39
2.84
2.79
2.74
2.59
2.65
2.68
2.59
2.60
2.50
2.60
2.60
2.60
2.60
2.60
2.50
2.60
2.50
2.60
2.50
2.50
2.60
2.50
2.60
2.50
2.60
2.50
139.92
72.30
St. II-Nft*
(EX)
Incidence
Due To
Sas Vapors
(PIL, mica)
(In-usBjNo)
24.32
23.45
22.61
21.81
21.28
20.73
20.35
19.55
• 19.35
18.59
18.67
18.35
18.02
17.70
17.38
17.38
17.38
17.33
17.38
17.38
17.38
17.38
17.38
17.33
17.38
17.38
17.38
17.33
17.38
17.33
17.38
17.38
17.38
17.38
17.38
651.04
215.46
St. II-Mfi
(EX)
Incidencs
Due To
Sas Vacors
(PUL,:;,ice)
(In-use,Nc)
23.51
21.53
20.78
19.98
19.41
13.91
18.55
13.21
17.65
17.33
17.04
16.74
16.45
16.15
15.85
15.86
15.36
15.36
15.36
15.86
15.35
15.86
15.36
15.86
15.36
15.86
15.85
15.86
15.86
15.86
15. 3S
15.86
15. 86
15.86
15.35
595.50
153. 13
St.II-Naticn
(EX)
Incidents
. 3u= To
3as Va;-crs
(?UL,5iics)
( la-use, No)
£3.34
22.31
17.34
14.41
13.75
13.39
13.14
12.90
12.53
12.27
12.25
11.85
11.54
11.44
11.23
11.23
11.23
11.23
11.23
11.23
11.23
11.23
11.23
11.23
11.23
11.23
11.23
11.23
11.23
11.23
11.23
11.23
11.23
11.23
11.23
440. 10
157.51
St. II-Naticn
(NO EX)
Incidence
Due ""o
Sas Vascrs
(PUL,:r.ica)

25.34
21.35
14.53
IS. 34
13.25
9. S3
9. S3
9.51
9.32
"9.15
3.59
8.83
3. S3
8.52
3.37
3.37
8.37
3.37
8.37
3.37
8.37
3.37
8.37
3.37
3.37
8.37
3.37
8.37
3.37
8.37
3.37
8.37
8.37
8.37
8.37
341.3S
129.52
Onboard
«7 Tanioer.
Incidence
Due TD
Sas Vacc-rs

( In-uss. fif.fi. :
Oi -»
'"2 7^
• 23.75
IS. 33
15.94
13.37
12.35
11.31
5.53
8.52
7.52
5.35
6.17
5.53
4.98
4.38
4.93
4; 93
4.98
4. SB
4.98
4.98
4.98
4.98
4.98
4.98
4.98
4.9S
4.98
4.53
4.98
4.58
4.98
4.98
4.93
289.58
133.57
                               F-69

-------
Table F-14.  Self-service  Incidence Due To Benzene find  Sas Vanors - In-use
Year





Baseline

Incidence
Due To
Gas Vaoors
(HUE. rat)
St.II-Nfl*
(EX)
Incidence
Due To
Sas Vaoors
«MLE.rat)
St.!I-Nfl St.II-Nation St.II-Nation
(EX)
Incidence
Due To
Sas Vapors
(ME. rat)
tin-use, fern. ) (In-use, flnn. ) (In-use, flnn. )
1986
1987
1388
1989
1990
1991
1932
1993
1994
1995
19%
1997
1998
1999
em
2801
mz
2883
2384
gees
2005
2807
sm
2009
2910
2911
2312
2013
2014
2015
2016
2917
2018
2019
2020
SUM*
NPV =
23.11
22.34
21.61
20.89
20.38
19.85
19.49
19.12
18.53
18.19
17.88
17.57
17.26
16.%
16.65
16.65
16.65
16.65
16.65
16.65
16.65
16.65
16.65
16.65
16.65
16.65
16.65
16.65
16.65
16.65
16.65
16.65
16.65
16.65
16.65
522.75
205.55
22.34
20.81
20.01
19.28
18.81
18.33
17.99
17.65
17.10
16.79
16.51
16.22
15.34
15.65
15.37
15.37
15.37
15.37
15.37
15.37
15.37
15.37
15.37
15.37
15.37
15.37
15.37
15.37
15.37
15.37
15.37
15.37
15.37
15.37
15.37
576.10
158.98
21.82
18.15
17.24
16.51
16.10
15.69
15.40
15.11
14.54
14.37
14.13
13.88
13.54
13.40
13.15
13.15
13.15
13.15
13.15
13.15
13.15
13.15
13.15
13.15
13.15
13.15
13.15
13.15
13.15
13.15
13.15
13.15
13.15
13.15
13.15
455.47
(EX)
Incidence
Due To
Bas Vapors
(MLE,rat)
( In-use, flnn. )
S3. 13
19.20
12.48
8.58
8.02
7.81
7.67
7.52
7.29
7.16
7.04
5.91
6.79
6.67
6.55
6.55
6.55
6.55
6.55
6.55
6.55
6.55
6.55
6.55
6.55
6.55
6.55
6.55
6.55
5.55
6.55
6.55
5.55
6.55
6.55
273.92
165.77 108.31
(MI EX)
Incidence
Due To
Sas Vapors
(ME, rat)
( In-use, Ann. )
23.13
17.34
8.54
3.54
2.99
2.31
2.86
2.30
2.72
2.67
2.62
2.58
2.53
2.48
2.44
2.44
2.44
2.44
2.44
2.44
2.44
2.44
2.44
2.44
2.44
2.44
2.44
2.44
2.44
2.44
2.44
2.44
2.44
2.44
2.44
. 131.42
67.91
St.II-Nfl*
(EX)
Incidence
Due To
Sas Vaoors
(MLE.rat)
(In-use, No)
£3.31
•22.33
21.23
28.49
19.98
13.47
19.11
18.75
lfi.17
17.84
17.54
17.23
15.33
16.53
16.32
16.32
16.32
15.32
15.32
16.32
16.32
16.32
15.32
16.32
16.32
16.32
15.32
16.32
15.32
16.32
15.32
16.32
15.32
16.32
16.32
611.49
282.37
St.II-Nfl St.II-Nation
(EX)
Incidence
Due To
Sas Vaoors
(MLE,rat)
(In-use, No)
22.45
23.31
13.44
18.59
18.23
17.75
17.44
17.11
16.58
15.28
16.00
15.72
15.45
15.17
14.39
14.39
14.85
14.83
14.39
14.89
14.39
14.89
14,89
14.89
14.89
14.89
14.93
14.89
14. S9
14.83
14.83
14.39
14.89
14.89
14.83
559.42
186.39
(EX)
Incidence
Due To
Sas Vaoors
!XLE,f£t5
(In-use, No)
£3. S3 '
22. S5
IS. 2?
13.53
"} 0«
*C. J x
12.53
12. 34
12.11
11.74
11.52
11.33
11.13
18.94
10.74
13.54
13.54
13.54
10.54
13. 54
18.54
13.54
13.54
18. 54
10.54
13. 54
10.54
13.54
13.54
13.54
13.54
12.54
18.54
IS. 54
10.34
13.5^
413.37
147.94
                                   F-70

-------
                   Table F-14. Self-Service Incidence Due To Servers find Sas Vaoors - In-uss
Year   St.II-Nation    Onboard






1986
1987
1988
1989
1998
1991
1992
1993
1994
1935
19%
1997
1998
1999
2000
2001
2002
2203
2004
2005
2006
2807
2098
2009
2010
2011
2012
2013
2014
2015
2015
2017
2018
2013
2020
11 =
'V =
(NO EX)
Incidence
Due To
Sas Vaoors
(ICE, rat)
(In-use,No)
23.80
20.06
13.72
10.18
9.62
9.37
9.20
9.03
8.75
8.59
8.44
8.39
8.15
8.01
7.86
7.86
7. Sfi
7.86
7.85
7.86
7.86
7.86
7.86
7.86
7.86
7.S6
7.86
7.85
7.85
7.86
7.86
7.86
7.86
7.86
7.86
320.29
121.74
NX Taraoer.
Incidence
Due To
Sas Vaoors
(ME, rat)
(In-use,flnn.)
23.11
22.34
19.52
16.99
14.98
13.12
11.64
13.34
9.08
8.01
7.16
6.43
5.79
5.19
4.68
4.68
4.68
4.68
4.68
4.68
4.68
4.58
4.68
4.68
4.68
4.58
4.68
4.68
4.68
4.58
4.58
4.58
4.58
4.68 '
4.68
271.92
139.24
Baseline     St.II-*ifi*    St.II-Nfl   St.II-Nation St.II-Naiion

Incidence
Due To
Sas Vaoors
(In-use,flnn. )
15.55
15.01
15.48
14.37
14.60
14.22
13.96
13.70
13.28"
13.33
12.31
12.59
12.37
12.15
11.33
11.33
11.93
11.93
11.93
11.33
11.33
11.33
11.93
11.93
11.33
11.33
11.33
11.33
11.93
11.93
11.93
11.33
11.93
11.93
11.93
446. 15
147.27
(EX)
Incidence
Due To
Sas Vacors
(Ml£,aice)
(In-use,flnn. )
15.31
14.31
14.34
13.81
13.48
13.13
12.89
12.54
12.25
12.33
11.83
11.62
11.42
11.21
11.01
11.01
11.01
11.01
11.01
11.01
11.01
11.01
11.01
11.01
11.01
11.01
11.01
11.31
11.01
11.31
11.01
11.31
11.01
11.01
11.01
412.73
135.82
(EX)
Incidence
Due To
Sas Vapors
(*LE,Bice)
(In-use,firm. )
15.05
13.01
12.35
11.82
11.54
11.24
11.03
10.32
10.43
10.33
18.12
3.35
9.77
3,60
9.42
9.42
9.42
3.42
9.42
9.42
3.42
9.42
9.42
9.42
3.42
9.42
9.42
9.42
9.42
9.42
9.42
9.42
9. «
9.42
9.42
354.95
113.75
(EX)
Incidence
Dua To
Gas Vapors
(ME, -.nee)
15.57
13.75
3.34
S.c2
5.75
5.50
5.49
5.39
5.22
5.13
5.84
4.95
4.37
4.78
4.63
4.63
4.59
4.69
4.59
4.53
4.69
4.63
4.53
4.55
4.59
4.63
4. S3
4.53
4.59
4.63
4.53
4.69
4.53
4.69
4. S3
135.24
77.33
(NO EX).
Incidence
Due To
Sas Vaoors
(MLE, 12 ice)
(In-use.Pnn. )
1 U« •-! .
12.75
5.12
£.54
£.14
2. S3
2.35
2.31
1.35
*. -•.
1.88
1.84
1.81
1.78
1.75
1.75
1.75
1.75
1.75
1.75
1.75
1.75
1.75
1.75
1.75
1.75
1.75
1.75
1.75
1.75
1.75
i.75
1.75
1.75
1.75
94.15
43.55
                                                   F-71

-------
Table F-14.  Self-service  Incidence Due To Benzene find  Gas Vapors - In-use
ar






1986
1987
1988
1989
1990
1991
1992
1993
1994
1995
1996
1997
1998
1999
2000
mi
2002
2003
2004
2005
2006
2807
2008
2009
2010
2011
2012
2013
2014
£015
2016
2017
2018
2019
2020
m =
3y =
St.II-Nft*
(EX)
Incidence
DUE To
Gas Vapors
(«LE,aice)
(In-use, No)
16.70
15.78
15.21
14.58
14.32
13.95
13.69
13.43
13.02
12.78
12.55
12.35
12.13
11.91
11.69
11.59
11.69
11.69
11.69
11.59
11.69
11.69
11.69
11.69
11.69
11.59
11.69
11.69
11.69
11.69
11.59
11.69
11.69
11.69
11.69
438.38
144.98
St.II-NA St.II-Nation
(EX)
Incidence
Due To
Gas Vanors
(«LE,nice)
(In-use, No)
16.09
14.55
13.93
13.39
13.06
12.73
12.49
12.25
11.88
11.55
11.46
11.25
11.07
10.87
10.67
10.67
10.67
10.67
10.67
10.67
10.67
10.67
18.67
10.67
10.57
10.67
10.67
10.67
10.67
10.67
10.67
10.67
10.67
10.67
10.67
480.78
133.32
(EX)
Incidence
Due To
Sas Vaoors
(MLE,aice)
(In-use, No)
17.35
15.31
11.67
9.59
9.25
9.31
8.84
8.58
8.41
8. £5
8.12
7.38
7.fl4
7.69
7.55
7.55
7.55
7.55
7.55
7.55
7.55
7.55
7.55
7.55
7.55
7.55
7.55
7.55
7.55
7.55
7.55
7.55
7.55
7.55
7.55
296. J.4-
105.99
St.II-Nation
(NO EX)
Incidence
Due To
Sas Vaoors
(MLE. aice)
tin-use, No)
17.35
14.37
3.83
7.29
6.89
5.72
6.59
6.47
6.27
6.15
6.05
5.94
5.84
5.74
5.63
5.53
5.63
5.63
5.63
5.53
5.63
5. -S3
5.53
5.63
5.63
5.63
5.53
5.53
5.63
5.53
5.63
5.63
5,63
5.63
5.63
223.46
87.22
Onboard
W/ Tamper.
Incidence
Due To
Sas Vaoors
(ME, mice)
(In-use, Ann. !
16.55
16.31
13.38
12.17
13.73
9.40
8.24
7.41
5.45
5.74
5.13
4.51
4.15
3.72
3.35
3.35
3.35
3.35
3.35
3.35
3.35
3.35
3.35
3.35
3.35
3.35
3.35
3.35
3.35
3.35
3.35
3.35
3.35
3.35
3.35
134.31
93.31
                               F-72

-------
                    TABLE  F-15.   VEHICLE OPERATIONS  INCIDENCE  DUE  TO  BENZENE
Without  additional Evaporative Esissions-
   flnnual       flnnual        flnnual
 Incidence    Incidence    Incidence
 No Bz Red   62.4* Bz Red  81.3* Bz Red
                                      flnnual
                                      Incidence
                                      No Bz Red  62.4* Bz Red
-With Additional Evaporative EtnisBions-
  finnual       flnnual       flnnual
Incidence     Incidence    Incidence
            81.3* Bz Red    Onboard
                         U/o Tamper.
  flnnual
Incidence
 Onboard
y/ Tamper.
12.13
11.53
19.99
10.52
10.13
9.76
9.59
9.45
9.36
9.34
9.37
9.37
9.39
9.42
9.43
9.43
9.43
9.43
9.43
9.43
9.43
9.43
9.43
9.43
9.43
9.43
9.43
9.43
9.43
9.43
9.43
9.43
9.43
9.43
9.43
11.72
11.18
10.63
13,26
9.91
9.57
'9.41
9.28
9.19
9.1fl
9.21
9.22
9.24
9.27
9.28
9.28
9.28
9.28
9.28
9.28
9.28
9.28
9.28
9.28
9.28
9.28
9.28
9.28
9.28
9.28
9.28
9.28
9.28
9.28
9.28
11.60
11.07
ia.se
18.18
9.84
9.50
9.35
9.22
9.14
9.13
9.17
9.17
9.28
9.23
9.24
9.24
9.24
9.24
9.24
9.24
9.24
9.24
9.24
9.24
9.24
9.24
9.24
9.24
9.24
9.24
9.24
9.24
9.24
9.24
9.24
 338.44
 107.52
332.29
105.16
                          330.39
                          184.44
12.35
11.74
11.20
10.73
10.34
9.97
9.80
9.55
9.57
9.56
9.59
9.59
9.61
9.64
9.66
9.66
9.66
9.66
9.66
9.66
9.66
9.66
9.66
9.66
9.66
9.66
9.66
9.66
9.66
9.66
9.66
9.66
9.66
9.66
9.66
346.16
109.81
11.86
11.32
10.82
10.39
10.04
9.69
9.53
9.40
9.32
9.31
9.34
9.35
9.38
9.40
9.42
9.42
9.42
9.42
9.42
9.42
9.42
9.42
9.42
9.42
9.42
9.42
9.42
9.42
9.42
9.42
9.42
9.42
9.42
9.42
9.42
336.91
106.57
11.72
11.19
10.71
10.29
' 9.95
9.61
9.45
9.33
9.25
9.23
9.27
9.28
9.30 .
9.33
9.34
9.34
9.34
9.34
9.34
9.34
9.34
9.34
9.34
9.34
9.34
9.34
9.34
9.34
9.34
9.34
9.34
9.34
9.34
9.34
9.34
334.98
105.58 •
12.23
11.63
11.%
10.58
10.17
9.79
9.60
9.45
9.35
9.32
9.34
9.33
9.35
9.37
9.38
9.38
9.38
9.38
9.38
9.38
9.38
9.38
9.38
9.38
9.38
9.38
9.38
9.38
9.38
9.38
9.38
9.38
9.38
9.38
9.38
337. S2
107.67
12.23
11.63
11.96
10.58
18. 17
9.79
9.60
9.45
9.35
9.33
9.35
9.34
9.36
9.38
9.39
9.39
9.39
9.39
9.39
9.39
9.39
9.39
3.39
9.39
3.39
9.39
9.39
9.39
3.39
9.39
3.39
9.39
3.33
9.39
9.39
337.91
107.73
                                                F-73

-------
             Table F-16.  PREDICTED MAXIMUM BENZENE CONCENTRATION (jug/m?)
Facility
1986 1990 1995 2000
Terminals
Controlled
Uncontrolled

Bulk Plants

Controlled
Exempt
Uncontrolled

Service Station Inloading

Uncontrolled (300, -30)a
Uncontrolled (600, -30)*
 3.756
22.91
 0.3124
 0.3151
 1.191
 0.1548
 0.2322
Service Station Refueling (outloading)
Uncontrolled (300, -30)a
Uncontrolled (600, -30)a
 0.2787
 0.2039
 3.406
20.74
 0.2822
 0.2846
 1.076
 0.1398
 0.2097
 0.2518
 0.1842
 3.120
18.98
 0.2575
 0.2598
 0.9819
 0.1276
 0.1914
 0.2298
 0.1681
 2.913
17.69
 0.2395
 0.2416
 0.9137
 (LI 188
 0.1781
 0.2139
 0.1564
*The two numbers in parantheses are two grid points.  The sum of service station
 inloading and refueling (outloading) result in a maximum concentration that can
 be at either point, depending on the control alternative.
                                        F-74

-------
  Table F-17.    RISK FROM  HIGH  EXPOSURE TO  BENZENE   (x  10~7)
 Facility
                                 1986
                                            1990
                                                        1995
                                                                    2000
                                                                                  Total
                                                                             70-yr.  Lifetime
 Terminals

 Uncontrolled                     22.2
 Controlled                        3.65

 Bulk Plants

 Uncontrolled                      1.156
 Controlled with exemption         0.3059
 Controlled                        0.3033

 Service Station Inloading

 Uncontrolled                      0.229
 Stage I                           0.0195
 Onboard                           0.229
 Stage II                          0.218
 Stage I & II                      0.0122
 Stage I & Onboard                 0.0195
 Stage II 4 Onboard                0.218
 Stage I 5 II S Onooard            0.0122

 Service Station Refueling (Outloadi'ng)
Uncontrolled
Stage I
Onboard
Stage II
Stage I &
Stage I 4
          II
          Onboard
Stage II  & Onboard
Stage I 4 II  &  Onboard
Service Station Total

Uncontrolled
Stage I
Onboard
Stage II
Stage I 4 II
Stage I Si Onboard
Stage IIS Onboard
Stage I 4 II S Onboard

Self Service
0.201
0.275
0.201
0.0238
0.0326
0.275
0.0238
0.0326
                                 0.430
                                 0.294
                                 0.430
                                 0.242
                                 0.0448
                                 0.294
                                 0.242
                                 0.0448
                                            20.1
                                             3.31
                                             1.045
                                             0.2763
                                             0.2740
                                             0.207
                                             0.0176
                                             0.207
                                             0.197
                                             0.0110
                                             0.0176
                                             0.197
                                             0.0110
0.182
0.248
0.131
0.0215
0.0294
0.179
0.0200
0.0273
            0.388
            0.266
            0.338
            0.218
            0.0405
            0.197
            0.217
            0.0384
                       18.4
                        3.03
                        0.9534
                        0.2523
                        0.2500
                        0.189
                        0.0161
                        0.189
                        0.180
                        0.0101
                        0.0161
                        0.180
                        0.0101
0.166
0.227
0.0633
0.0197
0.0269
0.0865
0.0165
0.0226
            0.354
            0.243
            0.252
            0.199
            0.0369
            0.103
            0.196
            0.0327
                       17.2
                        2.83
                        0.8872
                        0.2346
                        0.2325
                        0.176
                        0.0150
                        0.176
                        0.167
                        0.00937
                        0.0150
                        0.167
                        Ov00937
0.154 •
0.211
0.0280
0.0183
0.0250
0.0383
0.0145
0.0198
            0.330
            0.226
            0.204
            0.18S
            0.0344
            0.0533
            0.182
            0.0291
                         l,232a
                           203a
                            63.68a
                            16.84a
                            16.693
                            _b
                            _b
                            _b
                            _b
                            _b
                            _b
                            _b
                            _b
_b
_ij
_b
_b
_b
_b
_b
_b
                23.7a
                16.2a
                15.5a
                13.3a
                2.473
                5.10a
                14.2C
                3.38C
Uncontrolled"1
Stage Id
Onboard"
Stage IId
Stage I S IId
Stage I 4 Onboard4*
Stage II & Onboardd
Stage I 4 II 4 Onboardd
113
113
0.759
5.09
5.09
0.759
0.759
0.759
aTotal  70-year lifetime  risk = 4    [(1986 Risk + 1990 Risk)/2]  + 5  [(1990 Risk + 1995 Risk)/2]

                            + 5    [(1995 Risk + 2000 Risk}/2]  +  56 (2000 Risk)

'Total  is  not  calculated here.  Service station inloading  and refueling = Service
 Station Total.

cTotal  70-year lifetime  risk for combinations of Stage II  and Onboard

                            ={4    [(1986 Risk +• 1990 Risk)/2]  +• 5 [(1990 Risk +'1995 Risk)/2]

                            + 5    [(1995 Risk + 2000 Risk)/2]  + 2 (2000 Risk}} all  with Stage  II

                            + 4    [(2000 Risk with Stage II + 2000 Risk without Stage ID/2]  •

                            + 50   [(2000 Risk without Stage II}]

dRisk fron self-service was not calculated for individual  years, rather a 70-year lifetime
 risk was  calculated  directly.
                                       F-75

-------
                     Table  F-18.    RISK  FROM  HIGH  EXPOSURE  TO  GASOLINE  VAPORS
                        (PLAUSIBLE  UPPER  LIMIT  UNIT  RISK  FACTOR)  (  x  1C)-')
Facility
Terminals
Uncontrolled*
Controlled*
1986
437 or 704
70.0 or 115
Total
1990 1995 2000 70-year Lifetime
386 or 637 354 or 583 330 or 544 23,640 or 39,000
63.4 or 105 58.1 or 95.9 54.3 or 89.5 3,890 or 6,417
Bulk Plants

Uncontrolled*
Controlled with exemption*
Controlled*

Service Station Inloading

Uncontrolled*
Stage I*
Onboard*
Stage II*
Stage I 4 II*
Stage I S Onboard*
Stage II & Onboard*
Stage I 4 II 4 Onboard*
  22.2  or 36.6
  5.87  or 9.68
  5.82  or 9.60
  4.39 or 7.24
0.375 or 0.618
  4.39 or 7.24
  4.18 or 6.89
0.234 or 0.386
0.375 or 0.618
  4.18 or 6.89
0.234 or 0.386
Service Station Refueling (Outloading)

Uncontrolled0
Stage Ij
Onboard0
Stage lie
Stage I 4 IIC
Stage I 4 Onboard0
Stage II 4 Onboard0
Stage I 4 II 4 Onboard6

Service Station Total
  3.51 or 5.78
  4.79 or 7.90
  3.51 or 5.78
0.416 or 0.686
0.568 or 0.937
  4.79 or 7.90
0.416 or 0.686
0.568 or 0.937
 Uncontrolled
 Stage  I
 Onboard
 Stage  II
 Stage  I 4  II
 Stage  I 4  Onboard
 Stage  II 4 Onboard
 Stage  I 4  II 4 Onboard

 Self-Service

 Uncontrgll6df
 Stage  I*
 Onboard*
 Stage  IIf    .
 Stage  I 4  II*
 Stage  I 4  Onboard^
 Stage  II 4 Onboardf
 Stage  I S  II 4 Onboardr
  7.89 or 13.0
 5.17 or 8.52
  7.89 or 13.0
  4.59 or 7.58
 0.802 or 1.32
  5.17 or 8.52
  4.59 or 7.58
 0.802 or 1.32
  20.1  or 33.1
  5.30  or 8.75
  5.26  or 8.67
  3.97 or 6.54
0.339 or 0.558
  3.97 or 6.54
  3.77 or 6.23
0.212 or 0.349
0.339 or 0.558
  3.77 or 6.23
0.212 or 0.349
  3.17 or 5.22
  4.33 or 7.14
  2.29 or 3.77
0.376 or 0.619
0.513 or 0.847
  3.13 or 5.16
0.349 or 0.575
0.477 or 0.787
  7.13 or 11.8
  4.67 or 7.70
  6.25 or 10.3
  4.15 or 6.85
 0.725 or 1.20
  3.47 or 5.72
  4.12 or 6.80
 0.688 or 1.14
  18.3  or 30.2
  4.84  or 7.98
  4.80  or 7.91
  3.62  or  5.97
 0.309  or  0.510
  3.62  or  5.97
  3.45  or  5.68
 0.193  or  0.319
 0.309  or  0.510
  3.45  or  5.68
 0.193  or  0.319
  2.89 or 4.77
  3.95 or 6.52
  1.10 or 1.82
0.343 or 0.565
0.469 or 0.773
  1.51 or 2.49
0.289 or 0.476
0.394 or 0.651
  6.51 or 10.7
  4.26 or 7.03
  4.72 or 7.79
  3.79 or 6.25
 0.662 or 1.09
  1.82 or 3.00
  3.73 or 6.16
0.588 or 0.969
  17.0  or 28.1
  4.50  or 7.42
  4.46  or 7.36
   3.37 or 5.56
 0.288 or 0.474
   3.37 or 5.56
   3.21 or 5.29
 0.180 or 0.296
 0.288 or 0.474
   3.21 or 5.29
 0.180 or 0.296
  2.69 or 4.44
  3.68 or 6.06
0.488 or 0.805
0.319 or 0.526
0.436 or 0.719
 0.668 or 1.10
0.252 or 0.416
0.345 or 0.569
  6.06 or 9.99
  3.96 or 6.54
  3.86 or 6.36
  3.53 or 5.82
 0.616 or 1.02
 0.955 or 1.58
  3.46 or 5.70
 0.524 or .865
1,222 or 2,015
323 or 533
320 or 528
     N.A.b
     N.A.b
     N.A.b
     N.A.b
     N.A.b
     N.A.b
     N.A.b
     M.A.b
    N.A.b
    N.A.b
   N.A.b
    N.A.b
    N.A.b
    M.A.b
    H.A.b
    N.A.b
435 or 717d
285 or 469d
293 or 484d
253 or 417d J
44.2 or 72.9d
90.9 or 150d
269 or 444e
60.7 or 100e
                                                                          547 or 903C
                                                                          547 or 903C
                                                                          4.39 or 7.24C
                                                                          24.6 or 40.6C
                                                                          24.6 or 40.6C
                                                                          4.39 or 7.24C
                                                                          4.39 or 7.24°
                                                                          4.39 or 7.24C
 aRisk from high  exposure to gasoline vapors =  risk from high exposure to benzene x 1.71 or 9.42.

 *>Not Applicable.  Total is not calculated here.  Service Station inloading +  refueling = Service Station Total.

 °Risk fro* high  exposure to gasoline vapors =  risk from high exposure to benzene x 1.56 or 8.56.

 (fatal 70-year lifetime risk * 4 [(1986 Risk + 1990 Risk)/2] + 5 [(1990 Risk  + 1995 Risk)/2]

                              + 5 [(1995 Risk  +  2000 Risk)/2] + 56 (2000 Risk).

 *Total 70-year lifetime risk for options combining Stage II and Onboard

                             =  { 4 [(1986 Risk + 1990  R1sk)/2] + 5 [(1990 Risk + 1995 R1sk)/2]

                             +   5 [(1995 Risk + 2000  Risk)/2] + 2 (2000 Risk)} all with Stage II

                             +   4 [(2000 Risk with Stage  II + 2000 Risk without Stage  ID/2]

                             +  50 [(2000 Risk without Stage II)].

 'Risk fro* self-service was  not calculated for individual  years, rather a 70-year  lifetime  risk was calculated  directly.
                                                      F-76

-------
               Table F-19.   RISK  FROM HIGH EXPOSURE  TO  GASOLINE  VAPORS
              (MAXIMUM LIKELIHOOD   ESTIMATE  UNIT  RISK FACTOR)  (  x  lO'7)
Facility
Terminals
Uncontrolled3
Controlled3
Bulk Plants
Uncontrolled3
Controlled with exemption3
Controlled3
Service Station Inloading
Uncontrolled3
Stage I3
Onboard3
Stage II3
Stage I S II3
Stage I & Onboard3
Stage II S Onboard3
Stage I & II S Onboard3
Service Station Refueling
Uncontrolled0
Stage 1°
Onboard0
Stage 11°
Stage I & 11°
Stage I & Onboard0
Stage II $ Onboard0
Stage I 4 II 4 Onboard0
Service Station Total
Uncontrolled
Stage I
Onboard
Stage II
Stage I * II
Stage I 4 Onboard
Stage II & Onboard
Stage I S II 5 Onboard
Self-Service
Uncontrolled*1
Stage I*1
Onboard*1
Stage IIf
Stage I & IIf
Stage I & Onboard*1
Stage II 4 Onboard*1
Stage I S II 4' Onboard*1
1986

287 or 401
47.1 or 65.7

14.9 or 20.8
3.95 or 5.51
3.92 or 5.47

2.95 or 4.12
0.252 or 0.352
2.95 or 4.12
2.81 or 3.92
0.158 or 0.220
0.252 or 0.352
2.81 or 3.92
0.158 or 0.220
(Outloading)
2.36 or 3.29
3.22 or 4.50
2.36 or 3.29
0.280 or 0.390
0.382 or 0.534
3.22 or 4.50
0.280 or 0.390
0.382 or 0.534

5.31 or 7.41
3.48 or 4.85
5.31 or 7.41
3.09 or 4.32
0.540 or 0.754
3.48 or 4.85
3.09 or 4.32
0.540. or 0.754

_
-
1990

260 or 363
42.7 or 59.6

13.5 or 18.8
3.57 or 4.98
3.54 or 4.94

2.67 or 3.72
0.228 or 0.318
2.67 or 3.72
2.54 or 3.55
0.142 or 0.199
0.228 or 0.318
2.54 or 3.55
0.142 or 0.199

2.13 or 2.97
2.91 or 4.07
1.54 or 2.15
0.253 or 0.353
0.345 or 0.482
2.10 or 2.94
0.235 or 0.328
0.321 or 0.448

4.80 or 6.70
3.14 or 4.38
4.21 or 5.87
2.79 or 3.90
0.488 or 0.681
2.33 or 3.26
2.77 or 3.87
0.463 or 0.647

—
-
1995

238 or 332
39.1 or 54.6

12.3 or 17.2
3.26 or 4.55
3.23 or 4.51

2.44 or 3.40
0.208 or 0.290
2.44 or 3.40
2.32 or 3.24
0.130 or 0.181
0.208 or 0.290
2.32 or 3.24
0.130 or 0.181

1.94 or 2.71
2.66 or 3.71
0.743 or 1.04
0.231 or 0.322
0.315 or 0.440
1.02 or 1.42
0.194 or 0.271
0.265 or 0.370

4.38 or 6.11
2.87 or 4.00
3.18 or 4.44
2.55 or 3.56
0.445 or 0.622
1.22 or 1.71
2.51 or 3.51
0.395 or 0.552


-
2000

222 or 310
36.5 or 51.0

11.5 or 16.0
3.03 or 4.23
3.00 or 4.19

2.27 or 3.16
0.194 or 0.270
2.27 or 3.16
2.16 or 3.01
0.121 or 0.169
0.194 or 0.270
2.16 or 3.01
0.121 or 0.169

1.81 or 2.53
2.47 or 3.45
0.329 or 0.459
0.215 or 0.300
0.293 or 0.410
0.449 or 0.627
0.170 or 0.237
0.232 or 0.324

4.08 or 5.69
2.57 or 3.72
2.59 or 3.62
2.37 or 3.31
0.414 or 0.578
0.643 or 0.897
2.33 or 3.25
0.353 or 0.493


-
Total
70-year Lifetime

15,910 or 22,200
2,618 or 3,654

822 or 1,148
217 or 304
216 or 301

N.A.b
iv .n.
N.A.b
N.A.b
" "• .
N.A.D
'••"••
N.A.D

N.A.b
" "*.
N.A.b

N.A.b
k
N.A.b
l,.0«
N.A.b
U
N.A.b
" •"*«
N.A.b
N.A.b
NiA*.b

293 or 408d
191 or 267 d
197 or 27 5 d
170 or 238d
29.7 or 41 .5d .
61.2 or 85. 4d
181 or 253e
40.9 or 57e

368 or 514°
368 or 514°
2.95 or 4.12°
16.6 or 23.1°
16.6 or 23.1°
2.95 or 4.12°
2.95 or 4.12°
2.95 or 4.12°
aRisk from high exposure  to gasoline vapors =  risk from high exposure to benzene x 0.821 or 6.28.
bHot Applicable.  Total is not calculated here.  Service Station inloading + refueling = Service Station Total.
°Risk from high exposure  to gasoline vapors =  risk from high exposure to benzene x 0.747 or 5.72.
dTotal 70-year lifetime risk = 4 [(1986 Risk + 1990 R1sk)/2] + 5 [(1990 Risk + 1995 R1sk)/2]
                           + 5 [(1995 Risk  + 2000 Risk)/2] + 56  (2000 Risk).
STotal 70-year lifetime risk for Stage II in combination with Onboard
                          = | 4 [(1986 Risk + 1990 Risk)/2] + 5  [(1990 Risk + 1995 Risk)/2]
                          +   5 [(1995 Risk + 2000 Risk)/2]}a11 with Stage II
                          +   4 [(2000 Risk with Stage II + 2000 Risk without Stage ID/2]
                          +  50 [(2000 Risk without Stage II)].
 Risk from self-service was not calculated for individual years,  rather a 70-year lifetime risk was calculated directly.
                                                  F-77

-------
     Table  F-20.  RISK  FROM  HIGH  EXPOSURE  TO  EDB  (x  10~9)
Facility
Teralnals
Uncontrolled
Control 1 ed
Sulk Plants
Uncontrolled0
Controlled with exemption0
Controlled0
Service Station Inloadlng
Uncontrolled0
Stage 1°
Onboard0
Stage 11° .
Stage I 4 11°
Stage I 4 Onboard"
Stage II 4 Onboard0
Stage I 4 II 4 Onboard0
Service Station Refueling
Uncontrolled"1
Stage Id
Onboard^*6
Stage IId .
Stage 1.4 11° .
Stage I 4 Onboard'1.6
Stage II 4 Onboard1"-8
Stage I 4 II 4 Onboard"1.6
Service Station Total
Uncontrolled
Stage I
Onboard8
Stage II
Stage I 4 II
Stage I 4 Onboard6
Stage II 4 Onboard6
Stage I 4 II 4 Onboard6
Self-Service
Uncontrolled?
Stage IS
Onboard6-?
Stage 119
Stage I 4 119
Stage I 4 Onboard6. 3
Stage II 4 Onboard5 -9
Stage I 4 II 4 Onboard6. 9
1986

39.7
6.61

2.07
0.547
0.543

0.409
0.0349
0.409
0.389
0.0218
0.0349
0.389
0.0218
(Outloadlng)
0.369
0.504
0.369
0.0437
0.0598
0.504
0.0437
0.0598

0.778
0.539
0.778
0.433
0.0816
0.539
0.433
0.0816

.
-
-
-
-
1990

23.4
3.95

1.20
0.318
0.315

0.238
0.0203
0.238
0.226
0.0127
0.0203
0.226
0.0127

0.214
0.293
0.214
0.0254 '
0.0347
0:293
0.0254
0.0347

0.452
0.313
0.452
0.252
0.0474
0.313
0.252
0.0474

.
-
-
-
-
1995

9.59
1.74

0.493
0.131
0.129

0.0964
0.008?.
0.0964
0.0913
0.0052
0.0082
0.0092
0.0052

0.0869
0.119
0.0869
0.0103
0.0141
0.119
0.0103
0.0141

0.183
0.127
0.183
0.102
0.0192
0.127
0.102
0.0192

~
-

-
-
2000

1.95
0.498

0.081
0.021
0.021

0.0166
0.0014
0.0166
0.0158
0.0009
0.0014
0.0016
0.0009

0.0150
0.0205
0.0150
0.0018
0.0024
0.0205
0.0018
0.0024

0.0316
0.0219
0.0316
0.0176
0.0033
0.0219
0.0176
0.0033

-
—
-
-
-
Total
70-yr. Lifetime

347
68.8

16.8
4.43
4.39

.c
.c
.c
-c
.c
.c
.c

' _c
_c
_c
_c
_c
_c
_c
_c

6.36a
4.40a
6.36a
3.54«
0.667a
4.40a
4.27?
1.63?

196"
i QKn
iy o
1^6
3*33
8 23^
. 196"
8.83"
8.83"
Total  70-year lifetime risk - 4 [(1986 Risk * 1990 Rislc)/2] * 5  CU990 R1s* * 199S
                           * 5 CI1995 Risk + 2000 R1sk)/2] + 56  (2000 Risk)
Nlsk fro* high exposure to EDB - (risk fron high exposure to benzene) x (7.56 x 10"2) x (leaded gasoline
 throughput  1n year/total  gasoline throughput 1n year).
«Total  Is not calculated here.  Service Stations Inloadlng * refueling =• Service Station Total.
d!l1sk fron high exposure to EDB • (Risk from high exposure to benzene) x (7.76 x lO"2) x (leaded gasoline
 throughput  In year/total  gasoline throughput In year).
*S1sk froa high exposure to EDB not affected by Onboard controls, because Onboard controls will be  Installed only on
 cars using  unleaded gasoline  only and because unleaded gasoline does not contain EDB.
 Total 70-year Hfetlne risk  for combinations of Stage II and Onboard
                            - 14 [(1986 Risk * 1990 R1sk)/2] + 5 [(1990 Risk * 1995 R1sk)/2]
                            +  5 [(1995 Risk + 2000 R1sk)/2] * 2 (2000 Risk)} all with Stage  II
    ».                       +4 [(2000 Risk with Stage II  +.2000 Risk without Stage ID/2]
                            * 50 (2000 Risk without  Stage II)
 3R1SK  from self  service was not calculated for  individual years,  rather a 70-year  lifetime risk was calculated directly.
 "-otal risk fro. nigh  exposure to  EDB  = total  risk  fron high exposure to benzene  x 7.7,6  x ID'2, since assumed Individual
  using only all  leaaed gasoline.
                                                 F-78

-------
Table F-21.      RISK  FROM  HIGH  EXPOSURE  TO  EDC  (x  10~9)
Facility
Terminals
Uncontrolled3
Controlled3
Bulk Plants
Uncontrolled3
Controlled with exemption3
Controlled3
Service Station Inloading
Uncontrolled3
Stage I3
Onboard3
Stage II3
Stage I 4 II3
Stage I 4 Onboard3
Stage I! 4 Onboard3
Stage I 4 II 4 Onboard3
Service Station Refueling
Uncontrolled1-
Stage Ic
Onboard0*1*
Stage IIC
Stage I 4 lie.
Stage I 4 Onboardc-d
Stage II 4 OnhoardC.d
Stage I 4 II 4 Onboardc'd
Service Station Total
Uncontrolled
Stage I
Onboard"
Stage II
Stage I 4 II
Stage I 4 Onboard1*
Stage II 4 Onboardd
Stage I 4 II 4 Onboardd
Self-Service
UncontrolledS
Stage 19
Onboarddi9
Stage 119
Stage I 4 119
Stage I 4 Onboardd-9
Stage II 4 Onboard11 -3
Stage I 4 II 4 Onboardd-9
1986

51.3
8.55

2.68
0.703
0.702

0.522
0.0446
0.522
0.497
0.0279
0.0446
0.497
0.0279
(Outloading)
0.476
0.651
0.476
O.OS65
0.0772
0.651
0.0565
0.0772

0.999
0.695
0.999
0.554
0.105
0.695
0.554
0.105


-
_
.
_
-
-
1990

30.2
5.11

1.55
0.411
0.408

0.304
0.0259
0.304
0.289
0.0162
0.0259
0.289
0.0162

0.277
0.378 '
• 0.277
0.0328
0.0449
0.378
0.0328
0.0449

0.580
0.404
0.580
0.322
0.0611
0.404
0.322
0.0611


.
_
_
_
-
-
1995

12.4
2.24

0.638
0.169
0.167

0.123
0.0105
0.123
0.117
0.0066
0.0105
0.117
0.0066

0.112
0.153
0.112
0.0133
0.0182
0.153
0.0133
0.0182

0.235
0.164
0.235
0.131
0.0248
0.164
0.131
0.0248


_
_
_
_
-
-
2000

2.53
0.644

0.105
0.028
0.028

0.0212
0.0018
0.0212
0.0202
0.0011
0.0018
0.0202
0.0011

0.0194
0.0265
0.0194
0.0023
0.0031
0.0265
0.0023
0.0031

0.0406
0.0283
. 0.0406
0.0225
0.0043
0.0283
0.0225
0.0043


_
_
_
„
•
-
Total
70-yr. Lifetime

448
89.0

21.7
5.73
5.68

_b
_b
_b
_b
_b
_b
_b
_b

_b
_b
_b
-b
_b
.b
.b
.0

8.16e
5.68e
8.166
4.536
0.8596
5.68«
5.47f
2.11f

253C
253<=
253C
11. 4=
11. 4C
253<=
11. 4C
11. 4C
      Risk  from high exposure to EDC  »  risk from high exposure to  EDB x 0.992.
     b
      Total  is not calculated here.  Service station inloading and refueling =  Service
      Station Total.
      Risk  from high exposure to EDC  =  risk from nigh exposure to  EDB x 0.989.
     d
      Risk  from high exposure to EDC  not affected by Onboard controls, because  Onboard
      controls will be installed only on cars using unleaded gasoline only and  because unleaded
      gasoline does not contain EDC.
     STotal  70-year lifetime risk =  4  [(1986 Risk + 1990 Risk)/2] + 5 [(1990 Risk * 1995 Risk!/2]
                                   5  [(1995 Risk * 2000 Risk)/2] +  56 (2000  Risk)
     'Total  70-year lifetime risk for combinations of Stage II and Onboard
                                = |4  [(1986 Risk + 1990 Risk)/2] » 5 [(1990 Risk +• 1995 Risk 1/2]
                                +  5  [(1995 Risk + 2000 Risk)/2] + 2 (2000 Risk)|all with Stage II
                                <•  4  [(2000 Risk with Stage :i + 2000 Risk without Stage ID/2]
                                i- 50  (2000 Risk without Stage II!
      Risk  from self-service was not  calculated for individual years, rather a  70-year lifetime
      risk  was calculated directly.
                                        F-79

-------

-------
                  APPENDIX G
CUMULATIVE VALUES OF CAPITAL AND ANNUALIZED COSTS
                  (1986-2020)
                     G-l

-------

-------
                                      Table S-i.  Baseline Marketing Ootion Costs
                Tersiinal
        Tensinal
    Year
       1982

       1906
       !937
       1983
       1989
       1990
       1991
       1992
       1993
       1994
       1995
       1996
       1997
       1998
       1999
       2002
       2083
       2004
       2885
       2006
       2007
       2008
       2009
       2010
       2011
       2012
       2013
       2014
       2015
       2016
       2017
       2018
       2019
       2320

Total Costs

NPV of Costs
Annual ized
Costs
($ Million)
0.0
5.8
17.9
24.6
25.0
25.6
25.9
26.2
26.8
27.1
27.4
22.4
22.7
23.0
23.3
23.3
23.3
23.3
23.3
23.3
23.3
23.3
23.3
23.3
23.3
23.3
23.3
23.3
23.3
23.3
23.3
23.3
23.3
23.3
23.3
Capital
Costs
($ Million!
0.0
85.4
85.4
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
71.3
71.3
0.9
0.0
0.0
0.0
0.0
0.0
0.0
0.0
71.3
71.3
0.0
0.0
9.0
0.0
0.0
0.0
0.0
0.0
71.3
71.3
0.0
3.0
      Storage Tank     Storage Tank     Bulk Plant
                        Annualized       Capital
                          Costs           Costs
                       t$ Million)     ($ Million)
                                           Bulk Plant
789

214
598

221
     $0
  ($1.9)
  ($5.6)
  ($7.5)
  ($7.5)
  ($7.5)
  ($7.5)
  ($7.5)
  ($7.5)
  ($7.5)
  ($7.5)
  ($7.5)
  ($7.5)
  ($7.5)
  ($7.5)
  ($7.5)
  ($7.5)
  ($7.5)
  ($7.5)
  ($7.5)
  ($7.5)
  ($7.5)
  ($7.5)
  ($7.5)
  ($7.5)
  ($7.5)
  ($7.5)
  ($7.5)
  ($7.5)
  ($7.5)
  ($7.5)
  ($7.5)
  ($7.5)
  ($7.5)
  ($7.5)

($246.6)

 ($65.1)
                                                  $0
                                                $8.2
                                                $8.2
                                                  $0
                                                  $0
                                                  $0
                                                  $0
                                                  $0
                                                  $0
                                                  $0
                                                  $0
                                                  $0
                                                  $0
                                                  $0
                                                  $0
                                                  $0
                                                  $0
                                                  $0
                                                  $0
                                                  $0
                                                  $0
                                                $8.2
                                                $8.2
   $0
   $0
   $0
   $0
   $0
   $0
   $0
   $0
   $0
   $0


$32.7


$16.3
Annual i zed
Costs
($ Million)
(No ex.)
0.0
9.6
28.9
38.7
38.9
39.0
39.1
39.2
39.4
39.5
39.6
39.5
39.7
39.8
39.9
39.9
37.1
37.1
37.1
37.1
37.1
37.1
37.1
37.1
37,1
37.1
37.1
37.1
37.1
37.1
37.1
37.1
37. 1
37.1
37.1
1255
337
Capital
Costs
($ Million)
(No ex.)
0.0
113.6
113.6
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
106.5
106.5
0.8
0.0
0.0
0.0
0.0
0.0
0.8
0.0
0.9
0.0
0.0
0.0
0.0
136.5
186.5
0.0
0.0
653
, 252
                                                      G-3

-------
                                    Table G-l.  Gasoline Marketina Ootion Costs
r



1982
1986
1987
1988
1989
1990
1991
1992
1993
1994
1995
19%
1997
1998
1999
28W
2001
2882
2003
2004
20d5
2806
2007
2888
2389
2810
2811
2812
2813
2814
2315
2816
2817
2818
2019
2828
flnnualized
Costs
(S Million)
(Ex.)

0.0
5.7
17.4
23.5
23.6
23.8
23.9
24.0
24.3
24.4
24. S
24.6
24.7
24.8
24.9
24.9
23.3
23.3
23.3
23.3
23.3
23.3
23.3
23.3
23.3
23.3
23.3
23.3
23.3
23.3
23.3
23.3
23.3
23.3
23.3
Capital
Costs
($ Million)
(Ex.)

8.8
79.7
79.7
0.0
0.0
8.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.8
0.0
75.6
75.6
0.0
0.0
0.8
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
75.6
75.6
0.0
0.0
Total Costs

NPV of Costs
781

208
462

177
 0.0
 2.8
 8.3
11.1
11.1
11.1
11.1
11.1
11.1
11.1
11.1
11.1
11.1
11.1
11.1
11.1
10.8
10.8
10.8
10.8
10.8
.10.6
10.8
10.8
10.8
10.8
10.8
10.8
10.8
10.8
10.8
10.8
10.8
10.8
10.8

 361

   96
 0.0
18.3
10.3
 0.8
 0.0
 0.0
 8.0
 0.0
 0.0
 0.0
 0.0
 0.0
 0.0
 0.0
 0.0
 0.0
 9.2
 9.2
 0.8
 8.0
 0.0
 0.0
 8.0
 8.8
 0.0
 0.8
 0.8
 0.8
 0.0
 0.0
 0.0
 9.2
 9.2
 8.0
 0.8

  57

  23
Bulk Plant
Trucks
Annual i2ed
Costs
($ Million)
(No Ex.)
31.1
8. a
7.8
23.4
31.1
31.1
31.1
31.1
31.1
31.1
31.1
31.1
31.1
31.1
31.1
31.1
31.1
30.4
39.4
30.4
39.4
30.4
30.4
30.4
30.4
30.4
30.4
30.4
30.4
30.4
30.4
30.4
30.4
30.4
38.4
30.4
Bulk Plant
Trucks
Capital
Costs
<* Million)
(No Ex.)

0.0
28. 5
28. 5
0.0
0.0
0.0
0.0
- 0.0
0.8
0.0
0.0
0.0
0.0
0.0
0.0
0.0
25.9
25.9
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
25.9
25.9
0.0
0.0
1014

 278
161

 &3
                                                        6-4

-------
                                 Table 6-1. Gasoline Marketing (lotion Costs
Year
   1982
1985
1987
1988
1989
1998
1991
1992
1993
1994
1995
1996
1997
1998
1999
2880
2001
£002
2083
2004
2305
2008
£087
2008
£389
2810
2011
2012
2813
2014
2015
2016
2017
2018
2019
2020
Total Costs
NPV of Costs
8.0
4.5
13.5
18.0
18.0
18.0
18.0
18.0
18.0
18.0
18.0
18.0
18.0
18.0
18.0
18.0
17.6
17.6
17.6
17.6
17.5
17.6
17.6
17.6
17.6
17.6
17.5
17.5
17.6
17.5
17.6
17.6
17.6
17.6
17.6
585
156
0.0
16.5
16.5
0.0
8.0
0.0
8.0
8.0
0.0
8.0
8.0
0.0
0.0
8.0
0.0
0.8
14.9
14.9
0.0
8.0
0.0
8.0
0.0
0.0
0.0
8.0
0.8
0.8
0.8
0.8
0.0
14.9
14.9
8.8
0.0
93
36
0.8
16.9
50.8
67.7
67.7
67.7
67.7
67.7
67.7
67.7
67.7
67.7
67.7
67.7
67.7
67.7
67.7
67.7
57.7
67.7
57.7
67.7
67.7
67.7
67.7
67.7
57.7
67.7
67.7
57.7
67.7
67.7
67.7
67.7
67.7
2234
598
8.0
168.8
168.0
0.8
0.0
8.8
0.0
8.0
8.0
0.0
0.0
0.0
0.0
8.0
8.0
8.0
168.8
168.8
0.0
0.0
0.0
0.0
0.0
8.0
0.0
8.8
0.0
8.8
0.8
0.8
0.0
168:8
158.8
8.8
8.8
1088
378
0.0
7. 1
21.3
28.4
28.4
28.4
28.4
28.4
28.4
28.4
28.4
28.4
28.4
28.4
28.4
28.4
28.4
28.4
28.4
28.4
28.4
28.4
28.4
28.4
28.4
28.4
28.4
28.4
28.4
28.4
28.4
28.4
28.4
28.4
28.4
938
248
0.0
78.6
70.6
8.0
8.8
0.8
0.0
8.8
0.0
8.0
0.9
0.8
0.0
8.8
0.0
0.0
78.6
70.6
0.0
8.8
0.0
0.0
0.0
8.8
0.0
0.0
8.0
0.0
0.8
0.0
8.0
. 78.5
78.6
8.0
0.0
423
159
                                                 G-5

-------
                                    Table G-i.  Gasoline Marketing Ootion Costs


tr



1982
1985
1987
1988
1989
1990
1991
1992
1993
1994
1995
19%
1997
1998
1999
2000
2001
2082
2003
2904
2005
20%
2007
2008
2009
2010
2011
2012
2013
2014
2015
2016
2017
2018
2019
2020
Ser. Sta.
Stage MB
ftnnualized
Costs
($ Million)
(No Ex.)
2.4
1.2,
2.4
2.4
2.4
2.4
2.4
2.4
2.4
2.4
2.4
2.4
2.4
2.4
2.4
2.4
2.4
2.4
2.4
2.4
2.4
2.4
2.4
2.4
2.4
2.4
2.4
2.4
2.4
2.4
2.4
2.4
2.4
2.4
2.4
2.4
wSi • t3 6 a*
Stage I-Nfl
Capital
Costs
W Billion)
(No Ex.)

5.9
5.9
0.0
0.0
0.0
0.8
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
5.9
5.9
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
5.9
5.9
0.0
0.0
0.0
Ser. Sta.
Stage I-Nft
ftnnualized
Costs
($ Million)
(Ex.)
1.0
0.5
1.0
1.0
1.0
1.0
1.0
1.0
1.0
1.0
1.0
1.0
1.0
1.0
1.0
1.0
1.0
1.0
'l.0
1.0
1.0
1.0
1.0
1.0
1.0
1.0
1.0
1.0
1.0
1.0
1.0
1.0
1.0
1.0
1.0
1.0
Ser. Sta.
Stage I-Nft
Capital
Costs
<$ Million)
(Ex.)

2.5
2.5
0.8
0.0
0.8
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
2.5
2.5
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
2.5
2.5
0.0
0.0
0.8
Total Costs



NPV of Costs
81




24
35




14
34




10
15



 &
                                                    G-6

-------
Table S-2. Stage II and Onboard Ootion Costs
YEAR





1986
1937
1988
1989
1998
1991
1992
1993
1994
.1595
1996
1997
1998
1999
2008
2081
sm
2803
2884
2885
2886
2887
2888
2889
2810
2811
2812
2813
2814
2815
2816
2817
2818
2019
2020
Total Costs
NPV of Costs
St. II
Nation.
(No Ex. )
Net firm.
Cost
($ Si 11 ion)
8
129
388
537
554
557
558
568
563
565
566
568
569
571
573
573
573
573
573
573
573
573
573
573
573
573
573
573
573
573
573
573
573
573
573
$18,718
$4,869
St. II
Nation.
(No Ex. )
Cap. Cost
Total
($ Million)
0
1,156
1,155
134
8
8
8
8
8
113
118
14
0
0
0
8
1,039
1,156
238
14
0
0
0
0
0
118
118
14
8
0
0
1,039
1,039
238
118
$7,823
$2.847
St. II
Nation.
(Ex.)
Net Ann.
Cost
($ Million)
8
42
128
175
182
184
185
186
188
189
190
192
193
194
195
195
195
195
195
195
195
195
195
195
195
195
195
195
195
195
195
195
195
195
195
$6,323
$1,628
St. II
Nation.
(Ex.)
Can. Cost
Total
($ Million)
0
444
444
47
0
0
8
0
0
41
41
4
0
0
0
0
403
444
85
4
0
0
0
0
0
41
41
4
0
8
0
403
403
85
41
$2,977
$1.086
St. II
fill m
(No Ex.)
Net flnn.
Cost
($ Million)
72
150
168
155
166
157
167
168
169
169
170
178
171
171
172
172
172
172
172
172
172
172
172
172
172
172
172
172
172
172
172
172
172
172
172
$5,841
$1,672
St. II
All m
(No Ex.)
Cap. Cost
Total
($ Million)
653
48
48
8
8
0
0
0
66
4
4
8
0
0
0
587
102
48
4
8
0
0
0
8
66
4
4
0
0
0
587
35
102
4
4
$2,350
$977
St. II
All Nfl
(Ex.)
Net Ann.
Cost
($ Million)
24
49
52
54
54
55
55
56
55
57
57
58
58
- 58
59
59
59
59
59
59
59
59
59
59
59
59
59
59
59
59
59
59
59
59
59
$1,373
$558
St. II
All Nfl
(Ex. )
Cao. Cost
Total
($ Million)
253
14
14
0
0
0
0
8
24
1
1
0
0
0
0
229
37
14
i
0
0
8
0
0
24
i
1
8
0
0
229
13
37
i
1
$895
$373
               G-7

-------
Table G-2.  Stage II  and  Onboard Ootion Costs
YEflR




($
1986
1987
1988
1989
1990
1991
1992
1993
1994
1995
1996
1997
19%
1999
2888

2882
£083
2884
2085
2886
2007
2883
2009
2010
2011
2012
2013
2014
2015
2016
2817
2018
2019
2028
Total Costs
HPV of Costs
St. II
Sel. NA
(No Ex.)
Net ftnn.
Cost
Million)
27
57
61
63
63
63
64
64
64
64
65
65
65
65
65
65
65
65
65
65
65
65
65
65
65
65
65
65
65
65
65
65
65
65
65
$2,220
$636
St. II
Sel. Nfl
(No Ex.)
Cao. Cost
Total
($ Million)
248
15
15
0
0
8
0
8
25
2
2
0
8
0
0
223
39
15
2
0
0
0
0
8
25
2
2
8
8
8
223
14
39
2
2
$893
$371
St. II
Sel. Nfl
(Ex.)
Net Ann.
Cost
($ Million)
9
19
20
21
21
21
21
21
21
22
22
22
22
22
22
22
22
22
22
22
22
22
22
22
22
22
22
22
22
22
22
22
22
22
22
$758
$212
St. II ONBOflRD ONBOARD
Sel. Nfl CAPITAL
(Ex.)
Cap. Cost ($
Total
($ Million)
%.8
5.4
!i.4
0.0
0.0
0.0
0.0
8.0
9.8
8.4
0.4
0
0.0
19.0
8.0
8S. 9
14.8
5.4
8.4
8.0
0.0
8.0
0.8
0.0
9.8
8.4
0.4
0.0
0.8
8.0
86.9
5.8
14.0
8.4
0.4
$348
$142
COSTS
Million) ($


0
8
281
281
198
283
287
287
286
288
288
288
288
288
288
288
288
288
208
288
268
288
288
288
288
288
288
288
288
288
288
288
288
288
288
$6,836
$1.787
ANNUAL
COSTS
Million)


0
0
32
64
%
129
162
195
228
261
294
328
338
339
341
342
342
342
343
343
343
343
343
343
343
343
343
343
343
343
343
343
343
343
343
$9,666
$1.921
St. II Coai.
Nation.
(No Ex. )
Net Ann.
Cost
($ Million)
8
129
399
563
592
604
615
624
634
641
648
654
659
663
667
669
519
211
28
1
0
8
0
0
0
0
0
0
0
0
0
0
0
0
0
$9,521
$4,261
St. II Coa.
Nation.
(No Ex.)
Cap. Cost
Total
($ Million)
8
1,156
1,156
134
8
8
ei
0
8
118
118
14
8
8
0
O
&
0
0
0
8
8
8
0
0
8
8
0
0
0
8
8
0
8
8
$2,695
$2,287
                   6-3

-------
Table G-2.  Stage  II and Onboard Ootion Costs
3R





1986
1987
1988
1989
1990
1991
1992
1993
1994
1995
19%
1997
1998
1999
2000
2001
2002
2003
2004
2005
2006
2007
2008
2009
2010
2011
2012
2013
2014
2015
2016
2017
2018
2019
2020
Costs
Costs
St. II COH.
Nation.
(Ex.)
Net flnn.
Cost
($ Billion)
0
42
135
195
209
218
226
. 232
239
245
250
254
257
261
263
265
206
83
10
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
$3,589
$1,575
St. II COM.
Nation.
(Ex.)
Can.' Cost
Total
($ Million)
0
444
444
47
0
0
0
0
0
41
41
4
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
$1,022
$842
St. II Cora.
fill Nfl
(No Ex. 5
Net flnn.
Cost
($ Million)
72
150
164
173
177
181
184
187
190
192
194
1%
198
199
200
116
21
6
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
$2,803
$1,429
St. II Cos.
fill Nfl
(No Ex. )
Cap. Cost
Total
($ Million)
653
40
40
0
0
0
0
0
66
4
4
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
$808
$757
St. II Com.
911 Nfl
(Ex.)
Net flnn.
Cost
($ Million)
24
49
55
60
63
65
68
70
72
73
75
76
77
78
79
45
8
2
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
$1,040
$518
St. II Coo.
fill Nfl
(Ex.)
Cao. Cost
Total
($ Million)
253
14
14
0
0
0
0
0
24
1
1
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
$307
$289
St. II COIB.
Sel. Nfl
(No Ex. )
Net flnn.
Cost
($ Million)
27
57
62
66
67
69
70
71
72
73
74
75
75
76
76
44
8
2
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
$1,065
$543
St. II Cos.
Sel. Nfl
(No Ex.)
Can. Cost
Total
($ Million)
248
15
15
0
0
0
0
0
25
2
2
0
0
8
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
$387
$288
                 G-9

-------
                                   Table 6-2.  Stage II and Onboard (lotion Costs
   YEAR
St. II Cos.
  Sal. m
   (Ex.)
  Net ftnn.
    Cost
($ Million)
St. II Cca.
  Sel. Nfl
   (Ex.)
 Cap. Cost
   Total
($ Million)
1986
1987
1988
1989
1998
1991
1992
1993
1994
1995
19%
1997
19%
1999
2808
2801
2802
2883
2884
2885
2086
2887
2888
2818
2811
2812
2813
2814
2815
2816
2817
2818
2819
2828
9
19
21
23
24
25
26
26
27
28
28
29
29
38
38
17
3
1
8
8
8
8
0
8
0
0
0
0
0
0
8
8
8
8
%.0
5.4
5.4
8.0
8.0
0.8
0.0
0.0
9.0
8.4
0.4
8.0
0.0
0.0
0.0
0
0
0
0
0
0
0
0
0
8
0
8
0
0
0
0
0
0
0
 Total Costs

(PV of Costs
        $395
        $197
        $117
        $110
                                                     G-10

-------
                            TfiBLE
                                      IiM-U5E   STAGE  II
  YEflR
 TOTflL NET flNNUftL
    COSTS($MM)
NOTION (BI-flNNUflL)
      (EX.)
 TOTflL NET ftNNUfiL
    COSTS($MK)
NflTIQN !BI-flNNUflL)
     (NO EX!
   YEAR
TOTflL NET flNNUftL
   COSTS($MM)
NOTION (MINIMflL)
     (EX.)
TOTftL NET fiNNUftL
   COSTS($M!"i)
NATION (MINIMflL)
    (NO EX)
  198£
1966
1987
1988
1939
1998
1991
1992
1993
1994
1995
19%
1997
1998
1999
mm
2001
2882
2083
2884
2085
2886
2087
2888
2889
2810
2811
2012
2013
2014
2815
2816
2817
2818
2819
2828
—
46
139
192
198
199
288
281
283
284
285
206
207
208
289
209
289
209
289
209
289
209
209
289
289
289
289
289
289
289
289
289
289
289
289
                                        135
                                        485
                                        558
                                        576
                                        578
                                        588
                                        581
                                        584
                                        585
                                        586
                                        ^flfl
                                        WWW
                                        589
                                        590
                                        591
                                        591
                                        591
                                        591
                                        591
                                        591
                                        591
                                        591
                                        591
                                        591
                                        591
                                        591
                                        591
                                        591
                                        591
                                        591
                                        591
                                        591
                                        591
                                        591
                                        591
                                                       198£

                                                       1986
                                                       1987
                                                       1988
                                                       1989
                                                       1998
                                                       1991
                                                       1992
                                                       1993
                                                       1994
                                                       1995
                                                       1996
                                                       1997
                                                       1998
                                                       1999
                                                       2888
                                                       2081
                                                       2082
                                                       2803
                                                       2884
                                                       2885
                                                       2886
                                                       2087
                                                       2888
                                                       2089
                                                       2810
                                                       2811
                                                       2812
                                                       2013
                                                       2014
                                                       2815
                                                       2016
                                                       2017
                                                       2018
                                                       2819
                                                       2820
                                                      29
                                                      92
                                                     139
                                                     158
                                                     165
                                                     166
                                                     167
                                                     168
                                                     169
                                                     169
                                                     170
                                                     171
                                                     171
                                                     172
                                                     172
                                                     172
                                                     172
                                                     172
                                                     172
                                                     172
                                                     172
                                                     172
                                                     172
                                                     172
                                                     172
                                                     172
                                                     172
                                                     172
                                                     172
                                                     172
                                                     172
                                                     172
                                                     172
                                                     172
                                          82
                                         261
                                         396
                                         451
                                         469
                                         472
                                         473
                                         475
                                         476
                                         477
                                         477
                                         478
                                         479
                                         480
                                         480
                                         488
                                         480
                                         488
                                         480
                                         488
                                         480
                                         480
                                         488
                                         488
                                         488
                                         438
                                         488
                                         488
                                         488
                                         488
                                         488
                                         488
                                         488
                                         488
TOTflL COSTS
 Wi
       6.790
       1,757
       19357
                                         5848
TOTflL COSTS
                                     (*»))
      5,546
                    1,486
                                                                                                          15553
                                                                                             3966
                                                         G-ll

-------
   YEflR
 TOTAL NET ANNUflL
    COSTSHMi)
NflTION (QUflRTERLY)
      (EX.)
 TOTflL NET flNNUflL
    COSTS($MM)
NOTION (QUflRTERLY)
     (NO EX)
   YEAS
   ma

   1986
   1987
   1S88
   1983
   1990
   1991
   1992
   1993
   1994
   1995
   1996
   1997
   1998
   1999
   em
   2891
   2882
   2093
   2934
   2935
   2986
   2037
   2098
   2099
   2019
   2911
   2912
   2913
   2014
   2015
   2016
   2017
   2018
   2919
   2029
        43
       131
       181
       186
       IBS
       189
       199
       193
       194
       195
       1%
       197
       198
       199
       199
       199
       199
       199
       199
       199
       199
       199
       199
       199
       199
       199
       199
       199
       199
       199
       199
       199
       199
       199
       131
       393
       543
       561
       563
       565
       566
       569
       571
       572
       574
       575
       577
       578
       578
       578
       578
       578
       578
       578
       578
       578
       578
       578
       578
       578
       578
       578
       578
       578
       578
       578
       578
       578
   198£

   1986
   19S7
   1988
   1989
   1990
   1991
   1992
   1993
   1994
   1995
   1996
   1997
   1998
   1999
   2001
   2902
   2093
   2004
   2005
   2006
   2007
   2009
   2010
   2011
   2012
   2013
   2014
   2015
   2016
   2017
   2018
   2019
   2020
TOTAL NET flNNUflL
COSTS (*!*!)
NflTION (flNNUflL)
(EX.)
TOTAL NET ANNUAL
COSTS («"!«)
NATION (ANNUAL)
(NO EX)

44
134
185
190
192
193
194
196
197
198
199
200
201
202
202
202
202
202
202
202
202
202
202
202
202
202
202
202
202
202
202
202
202
202
132
397
548
566
568
570
571
574
576
577
578
580
581
583
583
583
583
533
583
583
583
583
583
583
583
583
583
583
583
583
583
583
583
583
TOTftL COSTS
   t«w<)
 NPV
      6,462
      1,667
      18902
       4922
TOTflL COSTS
   («MM)

 NPVCSMH)
6,571
1.697
19053
 4964
                                                        G-12

-------
   YEflR
TDTflL NET flNMJfiL
   COSTS(*MM)
  flLL Nfl (MIN)
     (EX.)
   1982

   1956
   1987
   1988
   1989
   1998
   1991
   1992
   1993
   1994
   1995
   1996
   1997
   1998
   1999
   2(981
   20192
   2804
   2885
   2007
   2008
   2010
   2011
   2012
   2313
   2014
   2015
   2016
   2017
   2018
   2019
   2820
       22
       45
       47
       49
       49
       50
       50
       50
       50
       51
       51
       51
       51
       51
       52
       52
       52
       52
       52
       52
       52
       52
       52
       52
       52
       52
       52
       52
       52
       52
       52
       52
       52
       52
       52
TQTfiL NET fiNNUflL
   COSTS («ffl)
  ftLL Nfi (MIN)
    (NO EX)
  YEflR
       61
      128
      136
      140
      141
      141
      141
      142
      142
      143
      143
      143
      143
      144
      144
      144
      144
      144
      144
      144
      144
      144
      144
      144
      144
      144
      144
      144
      144
      144
      144
      144
      144
      144
      144
 TOTfiL NET flNMJflL
    COSTS ($WI)
flLL Nft (BI-flNNUflL)
      (EX.)
  1982

  1986
  1987
  1988
  1989
  1990
  1991
  1992
  1993
  1994
  1995
  1996
  1997
  1998
  1999
                                                                       2081
   2003
   2004
   2005
   2006
   2087
   2008
   2009
   2310
   2011
   2012
   2013
   2814
   2015
   2016
   2017
   2018
   2019
        26
        53
        56
        58
        58
        59
        59
        59
        60
        60
        60
        61
        61
        61
        62
        62
        52
        62
        62
        62
        62
        62
        62
        62
        62
        62
        62
        62
        62
        62
        62
        62
        62
        62
        62
 TOTflL NET flNNLML
    COSTS<$MM>
ALL m (BI-flNNUflL)
     (NO EX)
        75
       155
       165
       171
       171
       172
       172
       173
       174
       174
       174
       175
       175
       176
       176
       176
       176
       176
       176
       176
       176
       176
       176
       176
       176
       176
       176
       175
       176
       176
       176
       176
       176
       176
       176
TOTflL COSTS
   ($M«)
     1,730
      4850
TOTflL COSTS
       2,068
        5923
 NPV (JW)
      500
       1411
     ($»»
        593
        1721
                                                          6-13

-------
  YEflR
TQTflL NET ftNNUftL
   COSTS <«*«
 fiLL m (WML)
     (EX.)
TQTftL NET ftNNUflL
   COSTS($HM)
 ai N
    (NO EX)
                                                                 YEflR
  1982

  1986
  1987
  1963
  1989
  1993
  1991
  1993
  1993
  1994
  1995
  199S
  1997
  1998
  1999
  2008
   2802
   2§83
   2884
   2805
   2006
   2908
   2811
   2912
   £813
   2014
   2015
   2916
   2817
   2918
   2019
   2929
       25
       51
       54
       56
       56
       57
       57
       58
       58
       59
       59
       59
       60
       60
       60
       60
       60
       60
       60
       60
       69
       60
       60
       60
       60
       60
       60
       60
       60
       60
       60
       60
       60
       60
       60
       74
      153
      163
      168
      169
      173
      178
      170
      171
      172
      172
      173
      173
      174
      174
      174
      174
      174
      174
      174
      174
      174
      174
      174
      174
      174
      174
      174
      174
      174
      174
      174
      174
      174
      174
               TOTfiL NET flNNUAL
                  COSTS($MM)
               flLL Nfl  (QUART.)
                     (EX.)
              TOTfiL NET flNNUAL
                 COSTS($MM)
              fill Nfl  (QUfiRT.)
                  (NO EX)
  198E

  1986
  1987
  1988
  1989
  1990
  1991
  1992
  1993
  1994
  1995
  19%
  1997
  1998
  1999
   2001
   2002
   2003
   2005
   2006
   2007
   2008
   2009
   2010
   2011
   2012
   2013
   2014
   2015
   2016
   2017
   2018
   2019
   2020
 24
 50
 53
 55
 55
 56
 56
 57
 57
 58
 58
 58
 59
 59
 59
 59
 59
 59
 59
 59
 59
 59
 59
 59
 59
 59
 59
 59
 59
 59
 59
 59
 59
 59
 59
 73
151
161
167
167
168
169
169
178
178
171
171
172
172
173
173
173
173
173
173
173
173
173
173
173
173
173
173
173
173
173
173
173
173
173
TOTftL COSTS
 NPV ($SM)
      2,009
       577
       5852
       1699
TOTftL COSTS
1,975
                      566
5805
                     1684
                                                      G-14

-------
                                   TECHNICAL REPORT DATA
                            (Please read Instructions on the reverse before completing)
 1. REPORT NO.
  EPA-450/3-84-012a
                                                           3. RECIPIENT'S ACCESSION NO.
4. TITLE AND SUBTITLE
  Evaluation  of Air Pollution Regulatory Strategies
  for the  Gasoline Marketing Industry
                                                         5. REPORT DATE

                                                           July 1984
                                                         6. PERFORMING ORGANIZATION CODE
7. AUTHOR(S)
                                                           8. PERFORMING ORGANIZATION REPORT NO
9. PERFORMING ORGANIZATION NAME AND ADDRESS                    "
  Director, Office  of Air Quality Planning and  Standards,
  Director, Office  of Mobile Sources
  U.S. Environmental  Protection Agency
  Washington,  D.C.   20460
                                                         10. PROGRAM ELEMENT NO.
                                                         11. CONTRACT/GRANT NO.
                                                           68-02-3060
 12. SPONSORING AGENCY NAME AND ADDRESS
  Assistant Administrator for Air and Radiation
  U.S. Environmental  Protection Agency
  Washington,  D.C.   20460
                                                           13. TYPE OF REPORT AND PERIOD COVERED
                                                           . SPONSORING AGENCY CODE

                                                           EPA/200/04
15. SUPPLEMENTARY NOTES
   I.6S1
       The gasoline  marketing industry (bulk terminals,  bulk plants, service station
  storage tanks,  and service station vehicle refueling operations) emit to the atmos-
  phere several organic compounds of concern.  These  include:   volatile organic
  compounds  (VOC)s>which contribute to ozone formation;  benzene, which has been listed
  as a hazardous  air pollutant based on human evidence of carcinogencity; and ethylene
  dichloride  (EDC),  ethylene dibromide (EDB), and gasoline vapors, for which there  is
  animal evidence of carcinogencity.  This report contains an  analysis of the health,
  emission, cost,  and economic impacts of several regulatory strategies for addressing
  organic compound emissions from gasoline marketing  sources.   The regulatory strategic:
  considered  are:  (1)  service station controls  (Stage II) for vehicle refueling
  emissions only  in  areas requiring additional VOC control to  attain the national ozone
  ambient standard;  (2)  service station controls (Stage  II)  for vehicle refueling
  emissions on a  nationwide basis; (3) Onboard vehicle controls for vehicle refueling
  emissions on a  nationwide basis; (4) bulk terminal, bulk plant, and service station
  storage tank controls on a nationwide basis; and (5) various permutations and
  combinations of these alternatives.
 7.
                               KEY WORDS AND DOCUMENT ANALYSIS
                  DESCRIPTORS
                                              b.lOENTIFIERS/OPSN ENDED TERMS
                                                                        c.  COS AT I Field/Group
Gasoline
Air Pollution
Pollution Control
Stationary Sources
Mobile Sources
Volatile Organic  Compounds Emissions
Benzene Emissions
                                              Air Pollution  Control
 8. DISTRIBUTION STATEMENT
  Unlimited
                                            19. SECURITY CLASS I This Report)
21. NO. OF PAGES
                                              Unclassified
    644
                                              20. SECURITY CLASS (This page)
                                                Unclassified
                                                                      22. PRICE
EPA Form 2220-1 (Rav. 4-77)   PREVIOUS EDITION is OBSOLETE

-------

-------

                          EPA-450/3-84-012a
 Evaluation of Air Pollution
  Regulatory Strategies for
Gasoline  Marketing Industry
    OFFICE OF AIR QUALITY PLANNING AND STANDARDS
                 AND
          OFFICE OF MOBILE SOURCES
      U.S. ENVIRONMENTAL PROTECTION AGENCY
           Office of Air and Radiation
            Washington, DC 20460
               July 1984

-------
This report has been reviewed by the Office of Air Quality Planning and Standards and the Office of Mobile
Sources, EPA, and approved for publication. Mention of trade names or commercial products is not in-
tended to constitute endorsement or recommendation for use. Copies of this report are available through
the Library Services Office (MD-35), U.S. Environmental Protection Agency, Research Triangle Park,
N.C. 27711, or from the National Technical Information Services, 5285 Port Royal Road Sprinafield
Virginia 22161.

-------
                           TABLE OF CONTENTS
Title
1.0  EXECUTIVE SUMMARY  	 .  	
     1.1  Operations, Emissions and Control  Technology. .  .
          1.1.1  Bulk Terminals 	
          1.1.2  Bulk Plants. . .  .	
          1.1.3  Tank Trucks. . .	
          1.1.4  Service Stations  ....... 	
     1.2  Analyses of Regulatory Strategies 	
          1.2.1  Regulatory Strategies, Model  Plants, and
                 Projections. ........  	
          1.2.2  Air Pollution Emissions,  Health-Risk, and
                 Control Cost Analyses	
     1.3  Results of Regulatory Strategy Analyses 	
                                              i     .    •
          1.3.1  Nonattainment Area Strategy Results  . .  .
          1.3.2  Nationwide Strategy Results  	
          1.3.3  Cost Per Incidence Reduction  .......
2.0  GENERAL DESCRIPTION AND PROFILE. .............
     2.1  General Industry Description	
     2.2  Gasoline Marketing Operations and Their Emissions
          2.2.1  Gasoline Composition 	
          2.2.2  Bulk Terminals	
          2.2.3  Storage Tanks at Terminals	  .
          2.2.4  Bulk Plants. ...............
          2.2.5  Tank Trucks	 .:  .
          2.2.6  Service Stations  . 	 	
     2.3  Baseline Emissions. .	
     2.4  References	
3.0  CONTROL TECHNOLOGY	
     3.1  Introduction	
     3.2  Control Technology for Bulk Gasoline Terminals.  .
     3.3  Control Technology for Storage Tanks. . . ....  .
          3.3.1  Fixed-Roof Tanks	
          3.3.2  Internal  Floating-Roof Tanks  	
          3.3.3  External  Floating-Roof Tanks  	
Page
 1-1
 1-2
 1-2
 1-4
 1-5
 1-5
 1-7

 1-7

 1-12
 1-17
 1-20
 1-22
 1-29
 2-1
 2-1
 2-3
 2-3
 2-8
 2-9
 2-13
 2-13
 2-15
 2-16
 2-20
 3-1
 3-1
 3-1
 3-4
 3-4
 3-5
 3-5
                                    m

-------
                       TABLE OF CONTENTS (Continued)
Title
     3.4  Control Technology for Bulk Gasoline Plants  .  .
          3.4.1  Submerged Fill 	
          3.4.2  Vapor Balance System 	
          3.4.3  Efficiency of Control Technologies .  .  .
     3.5  Control Technology for Tank Trucks	
          3.5.1  Description of Control  Technologies.  .  .
          3.5.2  Effectiveness of Technologies	
     3.6  Control Technology for Transfers into Service
          Station Underground Storage Tanks (Stage I)  .  .
          3.6.1  Description of Technology	
          3.6.2  Effectiveness of Technology	
     3.7  Vehicle Refueling 	
          3.7.1  Stage II Vapor Control  Systems 	
          3.7.2  Onboard Vapor Control Systems	
          3.7.3  Effectiveness of Technologies	
          3.7.4  In-Use Effectiveness of Control
                 Technologies 	
     3.8  References	
4.0  MODEL PLANTS AND REGULATORY STRATEGIES 	
     4.1  Model Plants	
          4.1.1  Bulk Terminal  Model  Plants 	
          4.1.2  Storage Tank Model  Plant 	
          4.1.3  Bulk Plant Model  Plants	
          4.1.4  For-Hire Tank Truck  Population 	 ,
          4.1.5  Service Station Model Plants  	 ,
     4.2  Gasoline,  Facility, and Vehicle Projections . . ,
          4.2.1  Gasoline Consumption ....  	 ,
          4.2.2  Gasoline Marketing  Facilities	,
          4.2.3  Light Duty Vehicles  and Light Duty Trucks,
     4.3  Regulatory Strategies 	
Page
 3-6
 •3-6
 3-8
 3-8
 3-10
 3-10
 3-11

 3-11
 3-11
 3-13
 3-15
 3-15
 3-19
 3-22

 3-23
 3-27
 4-1
 4-1
 4-1
 4-3
 4-3
 4-4
 4-7
 4-13
 4-13
 4-17
 4-20
 4-23

-------
                       TABLE OF CONTENTS  (Continued)
Title
     4.4  Source Category Control Options  	
          4.4.1  Bulk Terminals  	
          4.4.2  Bulk Plants	  .  .  .
          4.4.3  Service Stations	.  .  .  .
     4.5  References	
5.0  ENVIRONMENTAL AND ENERGY IMPACTS	  .  .  .
     5.1  Air Pollution Emission Impacts	
          5.1.1  Phase-In-Schedules for Control Options
          5.1.2  Discounting of Emission Reductions  .  .
          5.1.3  Emission Reduction Methodology .  .  .  .
     5.2  Other Environmental  Impacts 	
          5.2.1  Water Pollution Impacts.  .......
          5.2.2  Solid Waste Impacts. .  	
          5.2.3  Other Environmental Impacts	
     5.3  Energy Impacts. ......... 	
     5.4  References	
6.0  EXPOSURE/HEALTH-RISK ANALYSIS	
     6.1  Unit Risk Factors	
          6.1.1  Credibility of Risk Estimates	
     6.2  Exposure  and Risk  Methodology  and Assumptions
          6.2.1  General  Assumptions	
          6.2.2  Incidence Analysis  	
          6.2.3  Lifetime Risk  Analysis  	
     6.3  Presentation of Risk  Estimates for Regulatory
          Strategy	. .  . .	
     6.4  References	
Page
 4-29
 4-29
 4-32
 4-33
 4-41
 5-1
 5-1
 5-1
 5-6
 5-7
 5-19
 5-19
 5-24
 5-24
 5-24
 5-28
 6-1
 6-1
 6-3
 6-4
 6-5
 6-7
 6-18

 6-27
 6-40

-------
                       TABLE OF CONTENTS (Continued)
Title
7.0  COST IMPACTS	
     7.1  Introduction	
     7.2  Individual Facility Costs ... 	
          7.2.1  Bulk Terminals . . . . . 	
          7.2.2  Storage Tanks	
          7.2.3  Bulk Plants	
          7.2.4  For-Hire Tank Trucks 	
          7.2.5  Service Stations 	
     7.3  Nationwide Costs of Control Options 	
     7.4  Nationwide Costs of Regulatory Strategies 	
     7.5  Cost Per  Incidence Reduction.  .	
     7.6  References	
8.0  ECONOMIC IMPACT OF THE REGULATORY STRATEGIES  	  .
     8.1  Scope and Method	
     8.2  Summary of Cost Comparisons	•
          8.2.1  Total Cost by Regulatory Strategy  	
          8.2.2  Total Cost by Sector	
          8.2.3  Sources of Variation in Cost	
          8.2.4  Unit Cost and Quantity  Impacts  	
     8.3  Distributive Impacts	«
          8.3.1  Petroleum Refineries  	
          8.3.2  Bulk Terminals and Bulk Plants  	
          8.3.3  Service Stations  	
     8.4  References	
9.0  ENFORCEMENT STRATEGIES AND COST CONSIDERATIONS  ..'...
     9.1  Enforcement Strategies	
          9.1.1  Stage II  Programs	
          9.1.2  Onboard Control  Programs  	
     9.2  Resources Required  to  Pursue Enforcement Strategies
          at Various  Levels of Effort  	
          9.2.1   Installation Monitoring Resources	
Page
 7-1
 7-1
 7-2
 7-2
 7-8
 7-9
 7-15
 7-20
 7-34
 7-37
 7-44
 7-50
 8-1
 8-1
 8-3
 8-3
 8-5
 8-10
 8-18
 8-26
 8-29
 8-31
 8-34
 8-37
 9-1
 9-1
 9-1
 9-5

 9-5
 9-5
                                  vi

-------
                       TABLE OF CONTENTS  (Concluded)
Tttl e
          9.2.2   In-Use  Inspections Resources  	  .  .  .
          9.2.3   Test Observation Resources  	
          9.2.4   Legal-Clerical Resources	  .  .  .
          9.2.5   Onboard Control Inspection Resources  .  .  .  .
     9.3  Enforcement Costs	
     9.4  Enforcement Cost-Effectiveness Analysis 	
     9.5  References		
APPENDIX A  HISTORY/BACKGROUND.	  .  .
APPENDIX B  BASELINE EMISSIONS ANALYSIS 	
APPENDIX C, ASSESSMENT OF ONBOARD CONTROLS,  ... 	
APPENDIX D  DETERMINATION OF IN-USE EMISSION REDUCTION
            BENEFITS OF STAGE II PROGRAMS	
APPENDIX E  CUMULATIVE VALUES OF EMISSION REDUCTIONS
            (1986-2020)	
APPENDIX F  EXPOSURE AND HEALTH-RISK ANALYSIS ........
APPENDIX G  CUMULATIVE VALUES OF CAPITAL AND ANNUALIZED COSTS
            (1986-2020)  	
                                                                    Page
9-5
9-15
9-15
9-17
9-17
9-17
9-28
A-l
B-l
C-l

D-l

E-l
F-l

G-l
                                    vn

-------
                             LIST OF TABLES
Title
1-1  Gasoline Marketing Regulatory Strategies ...........    1-8
1-2  Major Analytical Considerations ...............    1-H
1-3  Unit Risk Factor Summary ...................    i"14
l-4a Estimated Risks from Gasoline Marketing Source Categories
     (Using Plausible Upper Limit Unit Risk Factor for
     Gasoline Vapors) .......................    1-18
l-4b Estimated Risks from Gasoline Marketing Source Categories
     (Using Maximum Likelihood Estimate Unit Risk Factor for
     Gasoline Vapors) .......................    1-19
1-5  Vehicle Refueling Controls in Nonattainment Areas  ......    1-21
1-6  Control of Benzene from Gasoline Marketing Sources ......    1-23
1-7  Summary of Theoretical  Impacts  for Selected Regulatory
     Strategies             .....  ...............    !-25
1-8  Impacts Based  on  "In-Use" Effectiveness for Stage  II and
     Onboard Controls  (1986-2020) .................    1-26
1-9  Economic  Impact of Regulatory  Strategies  Based on  Theoretical
     Efficiencies .........................    i'30
1-10 Economic  Considerations  .....  .... ..........    1-31
1-11 Estimated Regulatory Costs  and VOC Benefits  .........    1-33
 1-12 Benzene  Regulatory Costs  and Incidence Reduced ........    1-34
 1-13 Benzene  Regulatory Costs  Per Cancer  Incidence Avoided
      (Assuming VOC Benefits)  ...................    !-35
 1-14 Benzene  and Gasoline Vapors Costs Per Cancer  Incidence
     Avoided  (Using Rat Data Unit Risk Factor  for  Gas Vapors).  .  .    1-36
 2-1  Uncontrolled Emissions from Gasoline Tank Truck
      Loading Operations at a Typical Bulk Gasoline Terminal.  .  .  .    2-10
 2-2  Emissions from Gasoline Storage Tanks Located at a
      Typical  Terminal .......................     2~12
 2-3  Uncontrolled Emissions from a Typical Bulk Plant .......     2-14
 2-4  Uncontrolled Emissions from a Typical Service Station ....     2-17
                                      vm

-------
                          LIST OF TABLES  (Continued)
Title
                                                                      Page
2-5  Summary of Baseline Emissions for Gasoline Marketing
     Facilities for Base Year 1982	   2-19
3-1  In-Use Efficiencies of Stage II and Onboard Technologies.  .  .    3-25
4-1  Bulk Gasoline Terminal Model Plant Parameters  	    4-2
4-2  Bulk Plant Model Plant Parameters	    4-5
4-3  Method of Calculating the Number of Uncontrolled
     Terminal-Owned Trucks	 .  .  .  *.  . -   4-6
4-4  Estimates of 1982 Service Station Population	.  .    4-9
4-5  Estimated 1982 Service Station Size Distribution.  ......    4-11
4-6  Alternate Gasoline Consumption Projections
     (billion liters [gallons])	.  .    4-is
4-7  Estimated Number of Facilities in the Base Year 1982.  .  . ".  .    4-19
4-8  Onboard Consumption Projections	    4-21
4-9  Gasoline Marketing Regulatory Strategies Standard Numbers and
     Titles.	    4-24
4-10 Composition of Regulatory Strategies by Source Category  .  .  .    4-25
4-11 Gasoline Marketing Facility Model  Plants	 .  .  .  .  .    4-30
4-12 Number of Facilities Affected by Gasoline Marketing Control
     Options	    4-31
4-13 Ozone Nonattainment Areas (NA)  Assumed Affected by a CTG for
     Gasoline Marketing Evaluation ...  	 ......    4-35
4-14 Relative Size of Nonattainment Areas Affected by Additional
     Stage II Controls	    4-37
4-15 Nonattainment Areas Committed to or  Scheduling Stage II
     Vehicle Refueling Controls (NA)  	  ...    4-38
5-1  Effective Dates and Phase-In Schedules for Gasoline
     Marketing Regulatory Strategies  	 	    5-5
5-2  Summary of Control  Requirements  and  Affected Emission Factors
     for Gasoline Marketing Control  Options	      5-9
5-3  Gasoline Vapor Emission  Reductions in 1982 Associated with
     Control Options .	      5-11

-------
                          LIST OF TABLES (Continued)
Title
5-4  Summary of Projected Total Onboard Emission Impacts in
     Each Year of the Study (1988-2020) ..............    5-13
5-5  Projected Total Gasoline Consumption Changes
     from the Base Year ......................    5-15
5-6  Projected Leaded Gasoline Consumption  ............    5-16
5-7  Nationwide Emission Reductions from Gasoline Marketing
     Control Options .......................    5-18
5-8  Emission Reductions from Stage II  Options When Combined with
     Onboard ...........................    5-20
5-9  Nationwide Emission Reductions from Gasoline' Marketing
     Regulatory Strategies  (1986-2020)  ..............    5-21
5-10 Emission Reduction Comparison Between  Stage II and  Onboard
     Considering  In-Use Efficiencies  .  .  .............    5-23
5-11 Gasoline Recovery Ratios ......  .............    5-25
5-12 Energy Savings  Associated with Gasoline  Marketing Regulatory
     Strategies ............  •  .............    5-27
6-1  Unit  Risk  Factor Summary ...................    6-2
6-2  Emission Sources Considered  in Risk Analysis .........    6-6
6-3  Bulk  Terminal  and Bulk Plant Annual  Incidence  Analysis.  ...    6-9
6-4  Service Station Annual  Incidence Analysis  ..........    6-12
6-5  Self-Service Incidence Analysis  ...............    6-15
6-6  Bulk  Terminal  and Bulk Plant Lifetime  Risk  Analysis .....    6-20
6-7  Service Station Lifetime Risk Analysis ............    6-25
6-8  Self-Service Lifetime Risk  Analysis  .....  :  .......    6-28
6-9  Estimated  Cumulative  Incidence from  Benzene and  Gasoline
     Vapors (Plausible  Upper Limit Unit Risk  Factor)  from Gasoline
     Marketing  Source  Categories under Each Regulatory  Strategy.  .    6-29
 6-10 Estimated  Cumulative  Incidence from  Benzene and  Gasoline  Vapors
      (Maximum Likelihood Unit Risk Factor)  from Gasoline Marketing
      Source Categories  under Each Regulatory Strategy .......    6-30
 6-11 Estimated  Cumulative  Incidence from  EDB and EDC  from Gasoline
      Marketing  Source Categories under Each Regulatory  Strategy.  .    6-31

-------
                          LIST OF TABLES (Continued)
Title
                                                                     Page
6-12 Estimated Cumulative Incidence from Benzene and Gasoline
     Vapors from Vehicle Operations under Selected Regulatory
     Strategies	    6-32
6-13 Estimated Lifetime Risk from Benzene and Gasoline Vapors
     (Using Plausible Upper Limit Unit Risk Factor)  from
     Gasoline Marketing Source Categories under Each Regulatory
     Strategy	    5.34
6-14 Estimated Lifetime Risk from Benzene and Gasoline Vapors
     (Using Maximum Likelihood Estimate Unit Risk Factor)  from
     Gasoline Marketing Source Categories under Each Regulatory
     Strategy	    6-35
6-15 Estimated Lifetime Risk from EDB and EDC from Gasoline
     Marketing Source Categories under Each Regulatory Strategy.  .    6-36
6-16 Estimated In-Use Cumulative Incidence from Benzene and  Gasoline
     Vapors (Plausible Upper Limit Unit Risk Factor)  from  Gasoline
     Marketing Source Categories under Selected Regulatory Strategies
     and Enforcement Levels	 .    6-37
6-17 Estimated In-Use Cumulative Incidence from Benzene and  Gasoline
     Vapors (Maximum Likelihood Estimate Unit Risk Factor) from
     Gasoline Marketing Source Categories under Selected Regulatory
     Strategies and Enforcement Levels ...'...	    6-38
7-1  Bulk Terminal  Bottom Load Control  Costs (Thousands of
     4th Quarter 1982 Dollars)	  .    7-5
7-2  Bulk Terminal  Top Load Control  Costs (Thousands of
     4th Quarter 1982 Dollars)	    7-4
     Footnotes for Tables 7-1 and 7-2	    7-6
7-3  Average Bulk Terminal  Control  Costs
     (Thousands of 4th Quarter 1982 Dollars)	    7-7
7-4  Bulk Terminal  Average Weighted Costs (Thousands of 4th  Quarter
     1982 Dollars)	   7-3
7-5  Cost of Installing a Bolted Internal  Floating Roof on an
     Existing Fixed-Roof Tank (4th Quarter 1982 Dollars)   ...  .  .  .   7-10

-------
                          LIST OF TABLES (Continued)

Title                                                                 Page
7-6  Estimated Control Costs for Bulk Plants (No Exemptions)
     (4th Quarter 1982 Dollars)	     7-12
7-7  Estimated Control Costs for Bulk Plants (Exempt < 4,000  gal/day)
     (4th Quarter 1982 Dollars)	     7-14
7-8  Cost for the For-Hire Tank Trucks at Terminals
     (4th Quarter 1982 Dollars)	     7-17
7-9  Cost for the For-Hire Tank Trucks at Bulk Plants
     (4th Quarter 1982 Dollars)	     7-19
7-10 Service Station Stage I Capital and Net Annualized
     Cost Estimates (4th Quarter 1982 Dollars)	     7-21
7-11 Stage I Control Costs (Millions of 4th Quarter
     1982 Dollars)	". .     7-22
7-12 Average Stage  II Costs per System (4th Quarter
     1982 Dollars)	     7-23
7-13 Stage II Recovery Credit Calculations	     7-25
7-14 Weighted Average Stage II Costs (4th Quarter
     1982 Dollars)	     7-26
7-15 California Air Resources Board  (ARB) Stage II
     Control Costs  	     7-28
7-16 Service Station Stage II Weighted Average Costs
     (4th Quarter 1982 Dollars)	     7-29
7-17 Onboard Vapor  Control Hardware  Costs	
     (1983 Dollars)	     7-32
7-18 Cost Comparison between Stage  II and Onboard Controls
     Considering  In-Use Efficiencies (4th Quarter  1982 Dollars)
     (1986-2020)	     7-35
7-19 Number of Facilities Requiring  Controls  for Gasoline
     Marketing Options  ....  	     7-36
7-20 Nationwide Costs of Gasoline Marketing Control  Options.  .  . .    7-38
7-21 Cost Effectiveness of Gasoline  Marketing Control Options.  . .     7-39
7-22 Costs for Stage  II Options when Combined with Onboard ....    7-41
7-23 Nationwide Costs of Gasoline Marketing Regulatory Strategies.    7-42
                                      xn

-------
                          LIST OF TABLES (Continued)
Ti tl e
Page
7-24 Cost Effectiveness of Regulatory Strategies for
     Gasoline Marketing. . .	     7-43
7-25 Estimated Regulatory Costs and VOC Benefits	      7-45
7-26 Benzene Regulatory Costs and Incidence Reduced 	      7-46
7-27 Benzene Regulatory Costs Per Cancer Incidence Avoided
     (Assuming VOC Benefits)	      7-47
7-28 Benzene and Gasoline Vapor Costs Per Cancer Incidence Avoided
     (Using Rat Data Unit Risk Factor for Gas Vapors)                  7-48
8-1  1986 NPV of the Costs of the Regulatory Strategies
     (109 1982 dollars). .	     8-4
8-2  Estimated Reduction in Benzene Content of Gasoline Resulting
     From Regulatory Strategy XIV .	..."..    8-6
8-3  1986 NPV of the Costs of Benzene Reduction (109 1982 Dollars).    8-6
8-4  1986 NPV of Stage I Control  and Enforcement Costs for Bulk
     Terminals, Bulk Plants,  and For-Hire Trucks
     (106 1982 dollars)	 .	     8-7
8-5  NPV of Stage I and II Control and Enforcement Costs for Service
     Stations (10^ 1982 dollars)	    8-9
8-6  1986 NPV of the Total Costs of the Regulatory Strategies Under
     Alternative Assumptions  (lO9 1982 dollars in 1986) 	    8-11
8-7  Average Unit Cost and Quantity Effects for Nationwide
     Regulatory Strategies under Base Case Assumptions. ......    8-19
8-8  Average Unit Cost Increases for Gasoline under Alternative
     Cost Assumptions (1982  /liter)	    8-20
8-9  Reductions in Gasoline Consumption Attributed to Average Unit
     Cost Increases under Alternative Cost Assumptions
     (106 liters/year)	    8-23
8-10 Gasoline Quantity Impacts:  Average National Reductions in
     Consumption under Constant Gasoline Consumption With Various
     Elasticity Assumptions (10^ liters/year)  	    8-24
                                       xm

-------
                        LIST OF TABLES (Concluded)

Title                                                                 Page
8-11 Reductions in Vehicle Consumption Attributable to Unit Cost
     Increases under Alternative Cost Assumptions 	   8-25
8-12 Reductions in Vehicle Consumption Attributable to a Unit Cost
     Increase of $13/Vehicle Tank With Various Price Elasticities .   8-25
8-13 Difference in Average Unit Cost for Small and Large Facilities
     under Declining Gasoline Consumption 	   8-28
8-14 Increased Cost to Petroleum Refineries Due to Benzene Reduction
     in Gasoline for a 10,000-Barrel/Stream-Day Refinery vs. the
     U.S. Average	   8-30
8-15 Increased Unit Cost to Petroleum Refineries Due to Benzene
     Reduction of Gasoline by PADD District ( ?/liter, t/galIon
     in parentheses)	   8-30
8-16 1986 NPV of Control Cost per Station by Model Plant
     (1982 Dollars)	   8-35
9-1  Inspection and Re-Inspection Time Assumed for
     Enforcement Cost Analysis	   9-7
9-2  Probability That Individual Control Unit will have at Least
     One Defect, as a Function of Enforcement Effort	   9-13
9-3  Percentages of Facilities (Steady-State Average) Which
     Would be in Violation as a Function of Frequency of
     In-Use Inspections 	   9-14
9-4  Cumulative Enforcement Costs of Control Options (1986-2020). .   9-19
9-5  Cumulative Enforcement Costs of Regulatory Strategies
     (1986-2020)	   9-22
9-6  Theoretical and In-use Effectiveness Comparison among Stage II
     Control Options	   9-27
                                       xiv

-------
                                 LIST OF FIGURES
Title
                                                                      Page
1-1  Gasoline Marketing in the U.S. (1982 Baseline
     VOC Emissions)	     1_3
1-2  Effect of Onboard and Stage II Controls on Benzene Incidence
     (Based on Theoretical Efficiencies)	     1-28
2-1  Gasoline Distribution in the U.S	     2-2
2-2  Emission Equations 	     2-5
2-3  Mass Emission Rates ^Saturation	     2-6
2-4  Mass Emission Ratio of BZ, EDC, & EDB to VOC in Vapor. ...     2-7
3-1  Example of Gasoline Loading at Bulk Terminals	     3-2
3-2  Gasoline Tank Truck Loading Methods	     3-7
3-3  Vapor Balance Systems at Bulk Gasoline Plants	     3-9
3-4  Tank Truck Vapor Collection Equipment for
     Bottom Loading Operations	.	     3-12
3-5  Vapor Balance System at a Service Station	     3-14
3-6  Stage II Vapor Recovery Balance System 	     3-16
3-7  Stage II Vapor Recovery Vacuum Assist System 	     3-18
3-8  Stage II Vapor Recovery Hybrid System	     3-20
3-9  Onboard Controls for Vehicle Refueling Emissions 	     3-21
4-1  Estimated Flow of Gasoline Through the U.S. Gasoline
     Distribution System	     4_14
4-1  Footnotes for Figure 4-1	     4_15
4-2  Gasoline Consumption Projections 	     4-16
4-3  Total  Gasoline Consumption vs. Gasoline Consumed by Onboard
     Controlled Vehicles (1988-2020)	     4-22
5-1  Linear Phase-In-Schedules	     5-3
6-1  Map of Bulk Terminal  Complex	     6-21
6-2  Map of Bulk Plant Complex	     6-23
6-3  Map of Service Station Complex	 .     6-26
9-1  Probability of an Individual Defective Unit	     9-11
                                        xv

-------

-------
                          1.0  EXECUTIVE  SUMMARY

     The  purpose  of  the  study  was  to  evaluate  the  air  pollution
 regulatory  strategies available to reduce emissions from  gasoline
 marketing operations of benzene (Bz), ethylene  dibromide  (EDB), ethylene
 dichloride  (EDO, and gasoline vapors (GV).   Gasoline vapors or
 volatile organic compound (VOC) emissions contribute  to ambient ozone
 concentrations and, thus,  in  some areas contribute to a failure to
 attain the  national ambient air quality standard  for  ozone.  Benzene is a
 known carcinogen, which  has been  listed as a  hazardous air pollutant
 under Section 112 of the  Clean Air Act and. is present in  varying amounts
 in gasoTine.  In addition, EDB, EDC  and gasoline  vapors each have been
 shown to cause cancers  in  laboratory animals.   EDB and EDC are-generally
 added to leaded gasoline,  but are not present in  unleaded gasoline.
 The  following segments of  the gasoline marketing  industry were considered:
 bulk terminals (including  storage tanks and tank  trucks), bulk plants
 (including  storage tanks and tank trucks) and service stations (both
 inloading of underground  storage tanks and refueling  of vehicles).   The
 regulatory  strategies examined controls on all  segments of the industry,
 both with and without selected size cutoffs for small  facilities,  as
 well  as controls onboard the vehicle to reduce  refueling emissions.
    As noted, there are still  areas of the country which have"not yet
 attained the national  ambient air quality standard (NAAQS) for ozone.
 The Clean Air Act requires that all  areas achieve the NAAQS by
 December 31, 1987.   Some States,  as part of their State implementation
 plans to meet the statutory requirement, are considering control  of
 gasoline marketing sources, especially the refueling  of motor vehicles.
Thus, an analysis of gasoline marketing regulatory strategies must
 address the need to attain the ozone NAAQS in selected areas.  However,
 the emissions from gasoline marketing sources may induce public health
 risks which require control on a  national  basis.  The  analysis  evaluated
 regulatory  strategies which address both the more limited nonattainment
 issue in part of the country and  the broader question  of the  need  for a
 national  control  program to limit potential  hazardous  exposure.
                                  1-1

-------
1.1  OPERATIONS, EMISSIONS AND CONTROL TECHNOLOGY
    This section briefly outlines the operations and emissions of each
source category and major associated type of facility in the gasoline
marketing industry, as well  as the commonly used control techniques.
The segments of the gasoline marketing industry analyzed in this study
include all elements and facilities that move gasoline starting from
the bulk terminal to its end consumption.  Gasoline produced by refiner-
ies is distributed by a complex system comprised of wholesale and
retail outlets.  Figure 1-1 depicts the main elements in the marketing
network.  The flow of gasoline through the marketing system is shown
from the refinery, through bulk terminals, and sometimes bulk plants,
to retail service stations or commercial or rural dispensing facilities,
primarily via pipeline and tank truck.  The wholesale operations storing
and transporting gasoline including delivery and storage in a service
station underground tank are commonly called Stage I operations-.
Retail-level vehicle refueling operations are commonly termed Stage II.
    The baseline nationwide VOC emission estimates are also given for
the various source categories on Figure 1-1.  VOC emission factors for
the individual point source operations at each source category were
estimated.  Emissions at baseline were calculated based on the gasoline
throughput and current regulations for each source category in each
county in the nation.  Emission estimates for the other pollutants (Bz,
EDB, EDO were calculated using a ratio of the vapor pressures and
thus, vapor emission rates.
1.1.1  Bulk Terminals
    Bulk gasoline terminals serve as the major distribution point for
the gasoline produced at refineries.  Gasoline is most commonly delivered
to terminal storage tanks by pipeline with no emissions.  Gasoline is
stored in large aboveground tanks and later pumped through metered
loading areas, called loading racks, and into delivery tank trucks,
which service various wholesale and retail accounts in the marketing
network.
    Most tanks in gasoline service at terminals have an external or,
less commonly, an internal floating roof to prevent the loss of product
through evaporation and working losses.  Floating roofs rise and fall
with the liquid level preventing formation of a large vapor space
                                  1-2

-------
              Imported
              Gasoline
                                        Barge
                                      Pipeline
                                       Tanker
                 '200,000  MG/YR
        222,000  !%/YR
Servlca
Station
407,000 MG/YR
                                Commercial,
                                   Rural
             Consumer
                               Imcortad
                                  or
                               Domestic
                                Crude
                        Wholesale
                        Distribution
                        Level
208,000 MS/YR
                                    	J ~ Storage
                                                                    =  iransaort
                  Figure 1-1.  Gasoline Marketing in the U.S.
                         (1982 Baseline VQC Emissions)
                                     1-3

-------
and resulting emissions.  Fixed-roof tanks, which are still  used for
gasoline in some areas, use pressure-vacuum (P-V) vents to control  the
smaller breathing losses and may use processing equipment to control
the much greater working (filling and emptying) losses.  Breathing
losses result from volume variation due to daily changes in temperature
and barometric pressure.  Emptying losses occur when air drawn into the
tank during liquid removal saturates with hydrocarbon vapor and expands
beyond the vapor space of a fixed-roof tank.  Filling losses occur  when
the vapors in the fixed-roof tank are displaced by the incoming liquid
and forced to the atmosphere.  The largest potential source of losses
from external floating-roof tanks is an improper fit between the seal
and the tank shell.  Withdrawal loss from exposed-wet tank walls is
another source of emissions from floating-roof tanks.
    Emissions from the tank truck loading operations at terminals occur
when the product being loaded displaces the vapors in the delivery
truck tank and forces the vapors to the atmosphere.  In order to con-
trol these loading emissions, the displaced vapors can be ducted to a
vapor processor such as a carbon adsorber, thermal oxidizer, or refrig-
erated condenser for recovery or destruction.  The quantity of emissions
generated during loading a tank truck are dependent on the type of
loading.  Splash loading from the top of the truck creates considerable
turbulence during loading and can create a vapor mist resulting in
higher emissions.  Top submerged loading, which uses an extended fill
pipe, or bottom loading admit gasoline below the liquid level in the
tank and can be used to reduce turbulence and emissions (about a 60
percent reduction).  The recently promulgated bulk terminal new source
performance  standards  (NSPS), as well as a large number of State regula-
tions, currently require the use of vapor processors and submerged
loading at bulk terminals.  Most State regulations limit truck loading
emissions to 80 mg/liter transferred  (equivalent to about 90 percent
reduction) and require the tank truck to be vapor tight.  The NSPS is
more stringent and requires a lower emission limit of 35 mg/liter.
1.1.2  Bulk  Plants
    Bulk gasoline plants are secondary distribution facilities that
typically receive gasoline from bulk  terminals via truck transports,
                                   1-4

-------
 store  it  in aboveground storage tanks, and subsequently dispense it
 via  smaller account trucks to local firms, businesses and service
 stations.  As discussed in the previous section, vapors can escape from
 fixed-roof storage tanks at bulk plants due to breathing losses
 even when there is no transfer activity.  The majority of bulk plants
 already use top or bottom submerged loading, largely in response to
 State  regulations.  Vapor balancing is required by many State regulations,
 primarily for incoming loads, but also for outgoing loads in some
 instances.  Vapor balancing enables the vapors from the tank being
 filled to be transferred via piping to the tank being emptied.  Thus, the
 vapors are not forced to the atmosphere, and as a result, working losses
 are greatly reduced (by 90 percent or greater).
 1.1.3  Tank Trucks
    Gasoline tank trucks are normally divided into compartments with a
 hatchway at the top of each compartment.  Loading can be accomplished by
 top splash or submerged fill  through the hatch, or by bottom filling.
The majority of trucks have dual  capability.   Either top or bottom
 loading can be adapted for vapor collection.   However,  the trend is
toward bottom loading because of State vapor recovery regulations and
operating and safety advantages.   The vapor collection equipment is
basically composed of vapor domes enclosing each top hatch along with
various connectors and pipes (some removable)  that enable the vapors
 from the tank being filled to be transferred to the tank being emptied
of liquid.  Tank trucks with vapor collection equipment can become a
 separate source of emissions when leakage occurs (estimated to average
about 30 percent of potentially captured emissions).  Many States
require gasoline tank trucks equipped for vapor collection to pass an
annual  test of tank vapor tightness and pressure limits for the tanks
and vapor collection equipment (reducing average leakage to about
10 percent).
1.1.4  Service Stations
    Gasoline handling operations,  emissions,  and controls at service
stations are basically divided into two steps:   the filling (or inloading)
of the underground storage tank,  commonly called Stage I, and vehicle
 refueling, commonly called Stage II.  The filling of underground tanks
                                  1-5

-------
at service stations ends the wholesale gasoline marketing chain.  The
automobile refueling operation at service stations is the part of the
marketing chain that interacts directly with the general public.
    Emissions from underground tank filling operations at service
stations can be reduced significantly (by about 95 percent) by the use
of a vapor balance system.  Instead of being vented to the atmosphere,
the vapors are transferred into the tank truck unloading at the service
station and, ultimately, to the terminal vapor processor for recovery
or destruction.  Such controls have been incorporated into many State
regulations.
    Vehicle refueling emissions are another major source of emissions,
attributable to spillage and to vapor displaced from the automobile
tank by dispensed gasoline.  The two basic vehicle refueling regulatory
strategies are:  (1) control systems on service station equipment
(termed Stage II controls), and (2) control systems on vehicles and
trucks (termed onboard controls).  Stage II controls consist of either
vapor balance systems or assisted systems.  Assisted systems use a
variety of means to generate a more favorable (negative or zero) pressure
differential at the nozzle-vehicle interface so that a tight seal  is
not necessary between the vehicle and the nozzle "boot" (a flexible
covering over the nozzle which captures the vapor for return to the
underground tank via a vapor hose).  Stage II controls are currently
being used in 26 counties in California and the District of Columbia and
are being considered for other ozone nonattainment areas.   Onboard vapor
controls consist of a fillpipe seal and a carbon canister that adsorbs
the vapors displaced from the vehicle fuel  tank by the incoming gasoline.
The onboard system has undergone only limited testing to date.   It is
unclear what design problems could be encountered if onboard were
required for the entire vehicle fleet;  however, the technology is an
extension of a system already installed on light-duty cars and trucks.
Since 1971, new cars have been equipped with similar carbon canister
systems for collecting evaporative emissions (breathing losses caused
by temperature changes in the vehicle tank and carburetor).
     Both Stage II and onboard controls can be highly effective (as
high as 95 and 98 percent, respectively).   However, these  high  theoretical
                                  1-6

-------
 efficiencies  are  likely to be somewhat reduced in-use  (to as low as
 56  percent  for Stage  II programs under a minimal enforcement scenario
 considering a 20  percent rate of noncompliance, and to about 92 percent
 for onboard controls  with the expected level of tampering).
      It  should be noted that vehicle refueling controls not only reduce
 ambient  concentrations of VOC and hazardous emissions dispersed from
 the service station,  but also reduce the much higher exposures to
 hazardous pollutants  during self-service refueling (as discussed later).
 In addition,  it has been found that the present canisters for controlling
 evaporative emissions on many models of new vehicles are undersized.
 The expansion of  the  onboard system to control refueling emissions
 could achieve additional evaporative emission reductions roughly equal
 to the emission reductions achieved through refueling control.   The
 estimate of excess evaporative emissions is preliminary because EPA
 testing  is not yet complete.
 1.2  ANALYSES OF  REGULATORY STRATEGIES
    The  regulatory strategies selected for this evaluation were assessed
 with regard to their air pollution emissions, health-risk and cost
 impacts.  Impacts analyses were conducted using a model plant approach
 for most industry segments and source categories, along with certain
key assumptions.   Economic impacts were also assessed, as were the
effects of various enforcement levels on in-use effectiveness of the
vehicle refueling control  systems.   The following sections summarize
the regulatory strategies and analytical  methods and assumptions used.
1.2.1   Regulatory Strategies, Model  Plants, and Projections
    A total  of 14 industry-wide regulatory strategies were selected for
evaluation.   These strategies,  which are presented in
Table 1-1, are composed of a mixture of control  options for the individual
 source categories.  For the strategies calling for additional  controls
assessment was made of the relative  emissions, risks, costs, and cost
effectiveness of:
     (1) nationwide control  of Stage I  sources,
                              1-7

-------
              TABLE 1-1.  GASOLINE MARKETING REGULATORY STRATEGIES3
No Additional Controls (Baseline)
Stage II - Selected Nonattainment Areas (NA*)b
Stage II - All Nonattainment Areas (NA)C
Stage I - Nationwide
Stage II - Nationwide
Stage I and Stage II - Nationwide
Onboard - Nationwide
Stage II - Selected Nonattainment Areas & Onboard - Nationwide
Stage II - All Nonattainment Areas & Onboard - Nationwide
Stage I & Onboard - Nationwide
Stage II - All Nonattainment Areas and Stage I & Onboard - Nationwide
Stage II & Onboard - Nationwide
Stage I & Stage  II & Onboard - Nationwide
Benzene Reduction in Gasoline^

Facility Size Cutoffs:
Stage I:
      (1) bulk plants with  throughputs <4000 gal/d from balance controls on
         outgoing loads; and
      (2) service stations  with throughputs <10,000 gal/mon.
Stage II:
      (1)  all service  stations with  throughputs <10,000 gal/mon; and
      (2)  all independent  service  stations with throughputs <50,000 gal/mon,
Ozone  nonattainment  areas  needing  vehicle refueling controls  to help meet
 their  ozone  attainment goals by  1987.
 Areas  predicted by State or EPA  to be  nonattainment for ozone in 1982.
d
 Benzene reduction:
     A.   removal of  94.5 percent of  Bz  from  reformate  fraction for total
          reduction  of  62.4 percent;
      B.   removal of  94.5 percent of  Bz  from  reformate  and  fluid catalytic
          cracker (FCC)  fractions for total reduction of 81.3  percent.
                                    1-8

-------
      (2) nationwide control  of vehicle refueling emissions
          (Stage II controls, onboard controls,  or both),
      (3) ozone nonattainment area control  of vehicle refueling
          emissions (selected or.all  ozone  nonattainment areas),  and
      (4) combinations of the above.
 The regulatory strategies consider these approaches  either singly or  in
 combination,  both for controlling all  facilities and for  including size
 cutoffs  for some  facilities.   The facility  size  cutoffs were assumed
 based on the  relatively  higher costs of control  for  small  facili-
 ties, existing size  cutoffs  under State and local  regulations, and
 statutory  requirements for small  and medium throughput  independent
 service  stations  under Section  325 of  the Clean  Air  Act (section titled
 "Vapor Recovery for  Small  Business Marketers of  Petroleum  Products").
 If  a  Section  112  standard is pursued requiring Stage  I or  Stage II
 controls,  the actual  size cutoffs could vary from  these assumptions
 based upon  a  more  thorough economic  analysis for  small businesses and
 an  assessment of whether  Section 325 applies to  Section 112 standards.
 For the  purposes of  the analysis, initial  installation of  Stage I and
 II  control  equipment was  assumed  to  occur in 1986  for the  nonattainment
 area  strategies, in  1987  for nationwide strategies, and on new vehicles
 beginning with the 1988 model year for onboard controls.   All of the
 strategies  were compared  with a baseline reflecting 1982 Federal, State,
 and local  regulations.  The base year of 1982 was  selected because this
 represented the final implementation year for many State regulations
 affecting gasoline marketing sources, and because  at the beginning of
 the analysis  the most recent complete data reflected 1982  totals.
    A number  of model plants were developed to represent the entire
 spectrum of facilities in the analyses.  Four model plants were used
 for both bulk terminals and bulk plants while five model plants were
used  for service stations.  The model plants for each source category
were  differentiated on the basis of size,  in terms of gasoline  throughput.
Estimates of  typical  costs, emissions, and  resultant health-risks could
 then  be generated for each model plant and,  thus, for the entire  population
spectrum of each facility type.
                                  1-9

-------
    Because onboard controls would be installed only on new vehicles
(this analysis did not consider a retrofit option), the onboard regulatory
strategies take a number of years after initial implementation to
control the entire vehicle fleet.  Therefore, in order to evaluate the
comparison of onboard with other controls during both phase-in and full
Implementation, the analyses examined the time period from 1986 (when
the first controls would begin to be implemented) through 2020.  Thus,
the projection of certain basic parameters was necessitated.
     Total and leaded gasoline consumption were extrapolated to the
year 2000, based on the projections by EPA through 1990 for the phasedown
of lead in gasoline (47 FR 49329).  Due to a lack of confidence in
extrapolating beyond 15 years, gasoline consumption was assumed to
remain constant from the year 2000 to the year 2020.  The number and
fuel consumption of onboard controlled vehicles in a given year, were
estimated based on projections of new vehicles, retirement rates, fuel
economy, and mileage accrual rates through the year 2000.  These parameters
were also assumed constant from 2000 through 2020.  Although the number
of bulk plants and service stations have been decreasing, no quantitative
data was readily available with which to project facility populations.
Therefore, the numbers and size distributions of facilities were assumed
to remain constant at the values estimated for the base year of 1982.
However, the throughputs per facility and corresponding recovery credits
were decreased in proportion to the projected decline in nationwide
gasoline consumption.  The economic effects  of alternative assumptions
(e.g., constant gasoline consumption, declining marketing facilities,
etc.)  are examined in Chapter 8.
    A  summary of major analytical considerations that should be noted
in assessing the results is given in Table 1-2.  Both costs and emissions
were summed to a cumulative value, and also  were discounted (at 10 percent)
to a net  present value in 1986 (by summing the equivalent worth in 1986
of each annual amount).  Discounting was necessary because impacts are
not  uniform over the time period analyzed due  to:  1) the -slower phase-in
period of onboard controls compared to Stage I and II equipment; 2) the
respective useful lives of service station and onboard control equipment;
and  3) the declining gasoline consumption which  directly influences the
                                   1-10

-------
                                   TABLE 1-2
                        MAJOR ANALYTICAL CONSIDERATIONS
    A standard exposure lifetime of 70 years was used for exposure  to  ambient
    concentrations (i.e., those away from the immediate vehicle  fueling  area)
    A period of 50 years was used for self-service fueling exposure because
    one would not be expected to operate a vehicle for a complete lifetime.

    Because of the different phase-in times for  Stage II and  onboard  and  to
    analyze the impact of the strategies after onboard was  fully implemented
    the analysis covered a period from 1986 to 2020.

    In this analysis,  it was assumed that a national  Stage  II program  would be
    in place in 1989,  and that 1988  model  year vehicles  would be the first to
    incorporate onboard controls.   It is expected  that  it will take  about 20
    years  to convert the entire vehicle fleet to onboard control.

    Both costs and emissions capture over the 35-year analysis period were
    discounted at 10 percent to calculate cost effectiveness.  This was done
    to address the difference in  phasing-in  Stage  II and onboard controls.
   —  Model Plants
   --  Typical Locations of Plants
   —  Number of Facilities Within Source Categories
   --  Distribution of all Sources

o  Sources were distributed by best estimate considering size and population
   densities.

o  Risk is assumed to be linearly related to dose (concentration x duration
   of exposure) and combinations of concentration and duration yieldinq the
   same dose are assumed to be equivalent for risk estimation purposes.

o  The impacts of exposure to other substances beyond benzene, EDC,  EDB  and
   gasoline vapor are not addressed.  Health Effects other than cancer are
   not explicitly addressed.

o  Total  gasoline and leaded gasoline consumption is based on EPA's  lead
   phase-down projection extrapolated to the year 2000.   The consumption
   estimate for the year 2000 is assumed for all  years from 2001 to  2020.

o  Fleet average cost estimate of onboard systems used in this analysis was
   approximately $15 per vehicle.  Recent studies by API  and Ford Motor
   Company estimate average onboard costs of $13  and $53  per vehicle
   respectively.                                                     '

o  The average  capital  costs (equipment and  installation  costs)  per  station
   for Stage  II  control  systems  used in this analysis are $5,700  $6 100
   $6,600   $9,800  and $14,800 for 5,  20,  35, 65,  and 185  thousand gallon
   per month  throughput stations, respectively.
                                     1-11

-------
recovery cost credits.  Cost effectiveness was calculated using the
discounted costs and emissions values.
1.2.2  Air Pollution Emissions, Health-Risk, and Control  Cost Analyses
    Several underlying methods and assumptions were made  in all  of the
emissions, health-risk, and control cost analyses,  in addition to the
projections noted in the previous section.  Generally, emissions and
health-risk impacts of the various regulatory strategies  (including
baseline, which reflects current controls) were estimated for a base
year (1982) and then extrapolated to the years 1986 through 2020 in
proportion to the total (for benzene and gasoline vapors) or leaded
(for EDB and EDO gasoline throughput for each source category.   In
addition, the phasing-in of control installations with time were
considered in accordance with statutory requirements.  All  affected
facilities were assumed to install controls (linearly with  time) within
one year for a CT6 and within 2 years for a NESHAP  except for independent
service stations, which may be allowed up to 3 years in accordance with
Section 325 of the Clean Air Act.  Capital costs were attributed to the
year of installation.
     Estimates of annualized costs, emissions, and subsequent health-risks
during phase-in periods were based on the number of facilities and
corresponding gasoline throughput controlled for an entire  year.  The
impacts of vehicles with onboard controls in each year were calculated
based on the vehicle fleet projections noted previously.  After nearly
the entire vehicle fleet was projected to be equipped with  onboard
controls (in about 2002-2003), Stage II controls were not replaced, but
instead gradually phased out after the completion of useful equipment
lives for those strategies combining Stage II and onboard.
    The health-risk analysis estimated both annual  cancer incidences
nationwide and lifetime risk from high exposure, assuming a linear
dose response relationship with no threshold.  The term "lifetime risk
from high exposure" is conceptually similar to the term "maximum lifetime
risk" which has been presented in other EPA documents, including those
on benzene sources regulated or considered for regulation under Section
112 of the Clean Air Act.  The term "lifetime risk  from high exposure"
rather than "maximum lifetime risk" is used in presenting the risk
                                  1-12

-------
 calculations for the gasoline  marketing  study  because  EPA  is  less
 certain  in  this  case that  the  assumptions  used result  in the  maximum
 exposure to any  single  person  or  group.  For example,  high  exposure
 from self-service was assumed  to  occur to  a person pumping  40 gallons
 per week.   To the extent that  some  people  may  pump more gas than that,
 risks may be underestimated.
     The  estimates of risk,  in  terms of individual lifetime  risk from
 high exposure and aggregate incidence, are applicable  to the  public in
 the vicinity of  gasoline marketing  sources and  those persons  who refuel
 their vehicles at self-service pumps.  This analysis did not  examine
 the risk to workers  from occupational exposure  (e.g.,  terminal operators
 and service station  attendants).  The lifetime  risk from high exposure
 for these workers  is probably  substantially higher than for the general
 public.  In addition, the estimates of aggregate incidence  would be
 higher if such worker populations were included in the analysis.  Of
 course, any controls to reduce gasoline marketing emissions would reduce
 exposure for workers as well as for the general public.
      The unit risk factors used in the analysis for the four pollutants,
 presented in Table 1-3,  were developed by the EPA's Carcinogen Assessment
 Group  based on available health studies.   Two values of unit risk are
 shown  for gas vapors - a "maximum likelihood estimate" and  a "plausible
 upper  limit."  Both  values are used in the analysis to provide a broader
 base  for evaluating  the impacts of exposure to gasoline vapors.   For a
 detailed description of the derivation of the gasoline vapors unit risk
 numbers, see  the EPA staff paper "Estimation of the Public Health Risk
 from Exposure to Gasoline Vapor Via the Gasoline Marketing System",
June 1984.   The risk factor for benzene is based on studies of humans
occupationally exposed to benzene.  The risk factors for gasoline
 vapors, EDC,  and EDB are based on animal  studies only.   Because  of the
 significance  of the gasoline vapor animal studies,  conducted for the
American Petroleum Institute,  they are examined in  detail  in the EPA
 staff  paper cited above.  The  staff paper was  submitted on  June  .22,
 1984,  to the EPA Science Advisory Board for review.
     There can be substantial  uncertainty in unit risk  factors.   Reasons
 for this uncertainty include extrapolations which must be  made from
                                  1-13

-------
                   TABLE 1-3.   UNIT RISK  FACTOR SUMMARY
  Pollutant
      Unit R1ska
(probability of cancer
given lifetime exposure
     to 1 ppm)
   Health Effects
      Summary
                                                                            Comments
 Gasoline Vapor

   Plausible Upper Limit:b
     Rat Studies
     Mice Studies
    3.S  x  ID-3
    2.1  x  10-3
  Maximum Likelihood Estimates:
    Rat Studies
    Mice Studies
Benzene
Ethylene
  Oibrcmide
Ethylene
  Bichloride
   2.0  x 10-3
   1.4  x 10-3

  2.2 x 10-2
   4.2 x 10-1
   2.3 x 10-2
                           Kidney tumors in
                           rats, liver tumors
                           in mica.
                                               Human evidence of
                                               leukernogenicity.
                                               Zymbal  gland
                                               tumors  in rats,
                                               lymphoid and  other
                                               cancers in mice.
Evidence of carci-
nogenicity in
animals by. inhalation
and gavage. Rats:
nasal tumors; Mice:
liver tumors.

Evidence of carci-
nogenicity in
animals.  Circulatory
system,  forestomach,
and glands;  Mice:
Liver,  lung,  glands,
and uterus.
                        Gasoline  test samples in
                        the animal  studies were
                        completely  volatilized,
                        therefore may not be
                        completely  representative
                        of  ambient  gasoline vapor
                        exposures.
EPA: listed as a hazardous
air pollutant, emission
standards proposed.
IARCC: sufficient evidence
to support a causal associ-
ation between exposure and
cancer.

 EPA:  suspect human carci-
 nogen;  recent restrictions
 on pesticidal  uses.
                                                  EPA:  Suspect human
                                                  card nogen.  Draft
                                                  health assessment
                                                  document released
                                                  for review March 1984.
          -i     *?  I"  te"?n °f  ttf Probability of a cancer incidence (occurrence)  in
          individual  for a 70-year lifetime of exposure to l ppm of pollutant.
                  x   1 1™1 *  is Ca1cu1ated as «» 95 percent upper-confidence  limit  of  the
  «               f to exposure to 1 ppm of gasoline vapor,  using the multistage dose-
 response model  for low-dose extrapolation.
 IARC:  International Agency for Research on Cancer.
                                              1-14

-------
workers or animals to the general population and from the higher concen-
trations found in studies to the lower concentrations found in the
ambient air.
     Estimates of risk due to exposures from bulk terminals, bulk
plants, service stations, and self-service vehicle refueling were
generated for each of the four pollutants.  In order to calculate
community exposure to emissions (and the- resultant risk) from bulk
terminals and plants, assumptions were made concerning their geographical
distribution.  The fundamental assumption was that facilities were
located in proportion to the gasoline throughput for an area—for
example, the largest model plants would be located in large urban areas
where throughput (and population density) were highest.  Further, each
model plant type in each source category (bulk terminals and bulk
plants) was distributed over a range of ten urban area sizes.  The
largest terminals, for instance, were assumed to be located in cities
ranging in size from New York City to Des Moines, Iowa; the smallest
terminals were assumed to be located in cities ranging in size from
Spokane to Effingham, Illinois.  Estimates were also made of the extent
of existing control  at these terminals.  Most of those in the large
cities (likely to be ozone nonattainment areas) were considered controlled,
based upon existing regulations, with proportionately fewer facilities
controlled in the smaller areas.
    Thus, for both terminals and bulk plants, there were 40 model plant
locations (four model plant sizes each distributed to 10 representative
areas) for which estimates of ambient concentrations, population exposure
and incidence were made.  Total national incidence was calculated by
multiplying the model plant incidence by the number of facilities
represented by each model plant.  In somewhat similar fashion, model
service stations were allocated to 35 localities (multi-county metropol-
itan areas or single counties) and grouped by seven population size
ranges.  The model  plants were selected to be representative of total
national  service station distribution.  The localities and seven popula-
tion size ranges were selected to be representative of the total national
population distribution.
                                  1-15

-------
    Ambient concentrations, exposure, and incidence for bulk terminals,
bulk plants and service stations were calculated using the SHEAR version
of the EPA Human Exposure Model (HEM).  The HEM is a model capable of
estimating ambient concentrations and population exposure due to
emissions from sources located at any specific point in the contiguous
United States.
     Annual incidence due to self-service vehicle refueling was
estimated based on benzene and VOC concentrations in the region of
the face of a person filling the tank, as measured in a study for the
American Petroleum Institute (API).*  API selected thirteen gas stations
in 6 cities in which samples were collected to characterize typical
exposures to total hydrocarbons, benzene and eight other compounds.
Samples were collected using MSA-type battery operated pumps operated at
a one liter per minute flow rate and analyzed using a gas chromatograph.
Results were expressed as mg/nP air and ppm (vol.).
     The lifetime risk analysis was designed to estimate high exposures
of the four pollutants.  The Industrial Source Complex (ISC) dispersion
model was used to calculate annual concentrations in selected years  at
a number of receptors in the vicinity of a bulk terminal complex, a
bulk plant complex, and a service station complex.  Meteorological  data
for several cities expected to produce high concentrations were used.
The highest concentration at any receptor under a given regulatory
strategy was used to estimate the risk over a 70-year lifetime.  The
lifetime risk due to self-service exposure was estimated based on the
API measurements and an assumed lifetime exposure pattern of an individual
using a relatively high amount of gasoline (i.e. traveling salesman):
the risk was based upon an assumed exposure to four 10-gallon self-service
refuel ings per week for a working lifetime (estimated as 50 years).
    The control cost estimates were developed using a method similar to
that for emissions.  Capital and operating cost data were obtained
(largely from previous EPA studies) and developed on a per facility
*C1 ayton E'nvironmental Consultants, Inc.  Gasoline Exposure Study for
 the American Petroleum Institute.  Job No. 18629-15.  Southfield,
 Michigan.  August 1983.
                             1-16

-------
 basis for each model plant size of each source category.  These per
 facility costs were then combined with data on the number of facilities
 requiring controls within each source category.  Capital costs over the
 35 years of the analysis were incorporated during the initial phase-in
 years and then repeated in the years in which the economic life of the
 equipment ended if replacement equipment was required.  Annualized
 costs reflected the capital  costs and also were adjusted each year, as
 appropriate, to reflect reduced recovery credit due to the assumed
 decreases in gasoline throughput.
 1.3  RESULTS OF REGULATORY STRATEGY ANALYSES .
      Although only results of strategies involving size cutoffs  are
 given in this summary,  all  strategies are discussed in detail  in Chapters  2
 through  9.   The estimated  risks from the various  source categories  are
 given in Table 1-4 for  baseline (no additional  controls)  and  when
 controlled.   Table l-4a contains estimated risks  using the  plausible
 upper limit  unit risk factor  for gasoline  vapors  and  Table  l-4b  contains
 the estimates  using the maximum likelihood estimate unit  risk factor
 for gasoline vapors.
      The lifetime  risk  from high exposure  estimates the probability
 that  exposure  by an individual  to a  relatively high ambient concentration
 throughout his  lifetime  would  result  in  a  cancer  incidence.  The  lifetime
 risk  from high  exposure  to bulk terminal emissions  is  higher than the
 lifetime risk  from  high  exposure to uncontrolled  emissions from any of
 the other source categories.
     The average annual   incidence for each  regulatory  strategy is the  sum
 of the estimated average annual incidences  from each industry segment
 expected to  result  from  exposures during a  given year.  (The estimated
 annual incidences decrease during the study period in proportion to a
 projected decrease  in gasoline consumption.) The average annual  inci-
 dences given are estimates of cancer incidence due to benzene and
 gasoline vapors; for the latter, results are shown based on both mice
 and rat health studies.   Subsequent incidence numbers  in this Executive
Summary will  be given in terms of rat data only (for both plausible   .
upper limit and maximum  likelihood  estimates) as the rat numbers  are
higher and the tables can be simplified.   The incidences due to  EDB  and

                                  1-17

-------
    TABLE l-4a.   ESTIMATED RISKS  FROM  GASOLINE  MARKETING
                         SOURCE CATEGORIES
(USING  PLAUSIBLE UPPER LIMIT-UNIT RISK FACTOR  FOR  GASOLINE  VAPORS)
                                   A.  BASELINE
Source Category
Bulk Terminals
Bulk Plants
Service Stations
Self-service

Source Category
Bulk Terminals
Bulk Plants
Service Stations0
Stage I controls
only
Stage II controls
only
Onboard controls
only
Lifetime Risk from
High Exposure
(probability of effect)
Bz(GVa)
1.2 x 10-4(2.4 or 3.9
6.4 x 10-6(1.2 or 2.0
2.4 x 10-6(4.4 or 7.2
1.1 x 10-5(5. 5 or 9.0
B. CONTROLLED WITH SIZE
x 10-3)
x 10-4)
x io-5)
x 10-5)
CUTOFFSb
Lifetime Risk from
High Exposure
(probability of effect)
Bz(GVa)
2.0 x 10-5(3.9 or 6.4
1.7 x 10-6(3.2 or 5.3
1.6 x 10'6(2.9 or 4.7
1.3 x 10-6(2.5 or 4.2
1.6 x 10-6(2.9 or 4.8
x 10-4)
x 10-5)
x 10-5)
x 10-5)
x 10-5)
Average Annual Incidence
Over 35 years (1986-2020)
Bz(GVa)
0.07(1.3 or 2.2)
0.04(0.68 or 1.1)
0.19(3.3 or 5.5)
3.2(19 or 31)
-
Average Annual Incidence
Over 35 years (1986-2020)
Bz(GVa)
0.05(0.92 or 1.5)
0.02(0.32 or 0.53)
0.17(3.0 or 4.9)
0.11(1.9 or 3.2)
0.10(1.8 or 3'.0)
       Self-service0
        Stage II controls    5.1 x 10~7(2.S  or 4.1 x  10~6)    1.2(6.8 or 12)
        only
        Onboard controls     7.6 x 10~3(4.4  or 7.2 x  10-?)    1.1(5.9 or 9.8)
        only

       a
       8z * benzene, GV = gasoline vapors (mice or rats studies).  Incidences and
       lifetime risk due to gasoline vapors  are presented to  reflect the two unit
       risk factors (for liver cancer in mice or for  kidney cancer in rats).
       b
       Based on theoretical control efficiencies.

       Indicates reduction of lifetime risk  from high exposure for controlled sources.
                                       1-18

-------
           TABLE  l-4b.   ESTIMATED RISKS  FROM GASOLINE  MARKETING
,llt.T                              SOURCE  CATEGORIES
(USING  MAXIMUM  LIKELIHOOD  ESTIMATE UNIT RISK  FACTOR  FOR GASOLINE  VAPORS)
                                         A.  BASELINE
Source Category
Bulk Terminals
Bulk Plants
Service Stations
Self-service
Lifetime Risk from Average Annual Incidence
(proSaSlitroTeVfect) . ^^ ** ^ (1986-2020>
Bz(Gva) BzfGvai
1.2 x 10-4(1.6 or 2.2 x 10-3)
6.4 x lO-^S.Z x ID'5 or 1.1 x 10"4)
2.4 x 10-6(2.9 or 4.1 x 1Q-5)
1.1 x 10-5(3.7 or 5.1 x 10-5)
0.07(0.90 or 1.3)
0.04(0.46 or 0.64)
0.19(2.2 or 3.1)
3.2(13 or 18)
B. CONTROLLED WITH SIZE CUTOFFSb
Source Category
Bulk Terminals0
Bulk Plants
Lifetime Risk from
High Exposure
( probability of effect)
Bz(GVa)
2.0 x 10-5(2.6 or 3.7 x 10~4)
1.7 x 10-6(2.2 or 3.0 x 10'5)
Average Annual Incidence
Over 35 years (1986-2020)
8z(GVa)
0.05(0.62 or 0.86)
0.02(0.22 or 0.30)
  Service Stations0
    Stage I  controls
    only
    Stage II controls
    only
    Onboard  controls
    only
1.6 x 10-6(1.9 or 2.7 x 1Q-5)
1.3 x 0-6(1.7 or 2.4 x 10-5)
1.6 x 10-6(2.0 or 2.8 x lO'5)
0.17(2.0 or 2.3)
0.11(1.3 or. 1.8)
0.10(1.2 or 1.7)
  Self-service0
    Stage  II controls
    only
    Onboard controls
    only
5.1  x 10-7(1.7 or 2.3 x 10"6)
7.6  x 10-3(3.0 or 4.1 x 10~7)
1.2(4.8 or 6.6)
1.1(4.0 or 5.6)
   Bz -  benzene, GV = gasoline vapors.  Incidences and  lifetime risk due to
   gasoline vapors are presented  to reflect the two unit risk factors (for
   liver cancer in mice or kidney cancer in rats.            racwrs i ror
  b
   Based on theoretical  control efficiencies.
   Indicates reduction of lifetime risk from high exposure for controlled sources.
                                       1-19

-------
EDC only increase the incidences due to benzene by less than 3 percent
in most cases and by 5 percent or less in all cases.  Because the
estimated incidences due to EDB and EDC are relatively small, they were
omitted from the summary tables.  The average annual incidence from
self-service refuel ings at service stations contributes about 80
percent of the total incidence from all source categories.  The annual
incidences due to service stations without any additional controls are
approximately 3 for benzene and from about 15 to 36 for gasoline vapors,
considering both community exposures to ambient concentrations and
individual exposures to self-service refueling concentrations.
1.3.1  Nonattainment Area Strategy Results
     The effects of vehicle refueling controls in nonattainment areas
are  shown in Table 1-5.  The  primary focus of the nonattainment area
strategies is to reduce VOC emissions in order to attain  the  national
ambient  air  quality standard  (NAAQS) for ozone; reduction of  risk due
to hazardous  pollutants is an added benefit.  If onboard  controls
nationwide  under Section 202(a)(6) of  the Clean Air Act  are  used  to
replace  or  supplement  nonattainment area regulatory strategies, in
addition to  the primary aim  of reducing VOC  emissions  in  some or  all
nonattainment areas, VOC and  hazardous emissions are also reduced
nationwide.   It  should be  noted that  the strategy of Stage  II controls
in all  nonattainment areas also includes the costs, emissions, and
risk reductions  for Stage  I  controls  at  service  stations  in  two  non-
attainment  areas  (Atlanta  and Phoenix) where they currently  are  not
 installed.
     The average  annual  baseline level  of VOC emissions (and zero  addi-
 tional  control  cost)  from  service stations  of  91,300 Mg/yr  is given  in
 parentheses in  Table  1-5.  The average annual  VOC  emission  reduction
 from the baseline, net present value  of  control  costs, and  discounted
 cost effectiveness due to  Stage II refueling controls  were  estimated  to
 be 20,600 Mg/yr,  $210 million, and $940/Mg  VOC,  respectively, for the
 "selected nonattainment areas" (NA*)  strategy  and 60,900 Mg/yr,  $570
 million, and $870/Mg VOC,  respectively,  for the "all  nonattainment
 areas" (NA) strategy.   The costs and cost  effectiveness values for
 strategies involving onboard controls are  much higher (costs about $2
                                   1-20

-------












oo
^^
LU
Q£
^f


	
"^
Ly
S



• i
*?
^^
"^

LU
	 l
•!•
«? 1
|_

1
1






C r-
o 10 .
-o li =Jaj ^

C CJ CJ O
o a> cn-o i _.
4J -^ S. "tj vo N
"3 > 3) C CO 03
SE ^ ^ i"^ en
a> < — i
oo •«-*







tn
tn
t3 a>
^^ CJ C3
C 4J > >
3  tn oo
z 4-> en
o ^™*-
§ =
=" 5« -~
>— CO
ro c - cj encM
3 O<0 = E O
S-r- S S C\J
c tn O •*- .
a> "s o -u S to
C»LU 3 ,
3
IO
1 i
»
> en co r- rt 5
^f ^P CO







O O O O o
C f-l !•« O O O
i CM LO en — i 5
-H CM CM


•^
§O O C3 O a
O C3 O O o
co vo en (*) en ^^
ca* c^ i«* r^ *


*.
^ =2
•*-* WJ «^» *W 4^ «
5 S til jS 2 afi^
*^- SO i> 3 S (/^ '
1 S C +4  Z h4 44 4-> !
00 Z 44 (U •*- 44 3 !

•-•a; c—o»<0 Z t? U ^ 2 Z N S

C < r— ^ -— ^—44 "S P— JO ^-= rs o*j^- o»i- O crjojc: a> &. c '
t/» ^ UJ C5 "3(Oi/J (*3tO .2 ^SO fO(OO J
CD OO Q£ V> (/} O <-O OO


i

i








I
i


,1
1








in
•2
u
c
CJ
u

'£
s
4J
0
en
c
1O
>>
o

en
1
Ik.
jj



U
u-

c
n reductio
0
(/}
t/»
ff
LU
/«
CJ
u
2
a>
tn
"3

en
c
*^
3
"°
tn
"ie
3
^
>
•*••

C

u_
o
xposure
CJ
§
^


c


0*
CJ
 o
c 
*•»



.££
42
£

c
3
a)
re
^
4J
a>

•o
o
£

P~
0)
^£
r~*
maximum
en
^
tn
3

tn
a>

IO
!
c •
CJ >,
tn I—
a> 
s. u
O)
c o.
> O)
tn s,"
CJ 4-1
.a u
3 <2

0 "«
4J S.
— * •*—
^ 3
C
0 4->
10 "=
4-» •*—


4-> CJ
10 a.
i- a.
~ 3
in c;
S- i—
o ja
CL-^
10 tn
> 3
IO
a> •—
c a.
i— -a
o c
ie
u; s-
a
;» 10
































s
«
c
Ol
IO
4->
«
C
•*-
c

1
1

^a)
y
>^
.c
CJ

L.
o

>

•—
; emission
c
*i
fO
_m















I^J
CO
en


_>>

o
en
4->
c
i
J:
10
'4->
4->
rO

j:
4J

er
S
CJ
en
4J
•5
c

S
«
t

^
^
Z

(O
OJ

44

§
i Nonattai
OJ
o
CD
'S
OO

w
0)

'2

c
o
N
o
ti_
c
CJ
1
IO
4J
4->

C

J
48
cu
.£? *
o "a.
4J =
IO
4J
S 4J

is
0.1
w >s
S ,3
4ta C
 •—
IQ IO
«s
•a c

'wli

*io i— i
4J
. 4->
• OO
"c o*
in °.
a> o
i

1 1
5"^
44
« -*•
O *" *
c *
cS
'"* *^^

o "^

o *w

a>
O C3
in r^
tn en
•a -a
10 10
c: "5
0 0


o*
i —
T.
-1

s-
IO
o
c
°
-(•
^
^

-------
billion and cost effectivenesses about $5,000/Mg VOC) when considering
only the emission reductions in the affected nonattainment areas and
nationwide costs for onboard controls.  The annual average incidences
presented are the cumulative nationwide cancer incidences expected to
result from service station community and self-service exposures during
the 35-year period under the given regulatory strategy averaged over the
35 years.  As can be seen from Table 1-5, the incidence reduction due to
Stage II controls in nonattainment areas are relatively small  (less than
one incidence per year due to benzene) compared with that associated with
onboard controls nationwide (more than two incidences per year due to ben-
zene reduced).  The results shown in Table 1-5 are based on theoretical
control efficiencies and do not account for reduced efficiency in-use.
1.3.2  Nationwide Strategy Results
     The issues involved in gasoline marketing operations do not" relate
to ozone attainment only -- there is concern for hazardous exposure also.
This section presents summaries of the results of analyses of alternative
nationwide regulatory strategies and combined nationwide/nonattainment
strategies.  Table 1-6 shows the estimated baseline annual average
incidence for benzene and gasoline vapors, and the residual incidence
and cost for each of the three primary nationwide strategies:   Stage I,
Stage II and onboard.  The baseline benzene incidence attributable to
vehicle operations emissions (tailpipe and evaporative) is also shown;
this incidence is more than twice as large as that from gasoline marketing
operations.  In addition, the baseline incidence from possible additional
evaporative emissions is shown separate from vehicle operations.  This
value was separated because of its uncertainty, being based on preliminary
test results.  Controls on gasoline marketing sources will not reduce
incidences associated with vehicle operations.
    For gasoline marketing sources, the average annual reduction in
benzene incidence is estimated to be 0.1 with Stage  I controls, 2.1
with Stage II controls and 2.2 with onboard.  For gasoline vapors, the
comparable numbers are 1.0 or 1.8, 12.3 or 21.6, and 13.5 or 23.7.
Only limited benzene incidence reduction is achieved by removing benzene
from gasoline (2.4-3.2 per year).  This reduction primarily results
from reduced exposure from gasoline marketing sources.  Benzene tailpipe
                                  1-22

-------
           TABLE  1-6.   CONTROL  OF  BENZENE AND  GASOLINE  VAPORS
                FROM  GASOLINE  MARKETING SOURCES
    Regulatory
    Strategy
    Average Annual
      Incidence
  Of Cancers Expected
   From Exposures
    Over 35 years
     (1986-2020)
       8z (GV«)
   Cost Impacts'
   of Strategies
     (SBillion)
 BASELINE

 Gasoline Marketing
 Vehicle Operations0
 Evap. Emissions Mot Captured6
  Total

 IMPACTS AFTER SELECTEDf
 NATIONWIDE STRATEGIES

 Stage I - Nationwide
 (with size cutoffs)

 Stage II - Nationwide
 (with size cutoffs)

 Onboard Controls
 Nationwide
   - w/o Evap
   - w/  Evap

 Benzene Reduction
 in Gasoline
  Gasoline Marketing
  Vehicle Operations0
  Evap.  Emissions  Not Captured
 3.5 (23 or 40)
 9.7      (NAd)
 0.2 (4.0 or 7.0)
13        (NA)

AVERAGE ANNUAL
INCIDENCE REDUCTION

 0.1 (1.0 or 1.8)
 2.1  (12  or 22)
 2.3  (14 or 24)
 2.5  (18 or 31)
 2.2-2.9  (0 or 0)
 0.18-0.23 (NA)
 0.09-0.12 (1.5 or 3.3)
                                                              1986 NPVb   Total  1982
                                                              of Costs    Dollars Spent
                                                            (1986 - 2020) (1986 - 2020)
0.9


1.6
1.9
1.9
                               7.4-22
3.2.


6.3
 9.7
 9.7
                                               30-90
 GV = gasoline vapors  (rat data only), two numbers  given  represent estimates using
 maximum likelihood  estimate unit risk factor and plausible upper limit unit risk
 factor, respectively.
b
 MPV = Net Present Value.

 Incidences due  to exhaust and evaporative benzene  emissions during vehicle operations
 £I!,?!*1T!!  "Sin9 !" area.source approach similar  to  that used for service stations.
 Jnanticipated evaporative emissions were not considered  here (see Footnote e).
d
 Not applicable.

 Based on  preliminary estimate of possible emissions not  captured by the existina
 Impacts  based on theoretical  control  efficiencies.
                                       1-23

-------
emissions are not affected substantially by benzene removal,  since
benzene is formed in the combustion process.*
     Costs of additional controls beyond baseline are presented both as
the net present value of cost (discounted at 10 percent to 1986}  and the
cumulative value of the estimated expenditures from 1986 through  2020
(all in 1982 dollars).  The costs of all available nationwide strategies
are greater than $800 million net present value of costs or $3 billion
cumulative costs.  The cost of benzene reduction in gasoline is a
factor of 2 to 10 greater than for the next most costly strategy.
     Table 1-7 summarizes the estimates of average annual  incidence,
emission reductions, cost, and cost effectiveness for the nationwide
control strategies (with size cutoffs) evaluated..  These impacts  also
are based on theoretical control efficiencies.  This table presents
more detailed results: average annual incidence under the strategy,
cumulative VOC and benzene emission reductions, net present value of
costs, and discounted cost effectiveness for both basic regulatory
strategies and combinations of strategies.  Although Stage I controls
result in large emission reductions at relatively low cost, they  result
in substantially less incidence reduction than Stage II or onboard.
Strategies with either Stage II or onboard refueling controls achieve
greater incidence reductions because of their effect on self-service
emissions.
     Table 1-8 presents costs, emission reductions, incidence reduction
and cost effectiveness with theoretical and in-use efficiencies for
Stage II and onboard, and gives two levels of enforcement for Stage II.
The in-use efficiency of Stage II programs is highly dependent on the
level of enforcement used, varying from 56 percent with no inspections
to 86 percent with annual inspections.  Enforcement costs are not
included in the cost-effectiveness figures given (their impact is addressed
in Chapter 8).  Although average annual enforcement costs for Stage II
nationwide with annual  inspections 'are about $7.7 million, including
* Black, P.M., I.E. High, and J.M. Lang.  Composition of Automotive
  Evaporative and Tailpipe Hydrocarbon Emissions.  Journal of the Air
  Pollution Control Association. 30:1216-1221.  November 1980.
                                  1-24

-------
 CO

 LjJ
 oo
 a:
 o
 UJ
 a
 UJ
vo

Q±
O

4-1 0) 0
§ •££=•
o o 4-> en
O CJ O S
in oj ^
5 i*-
|jj
=
4J r— O
= 14. d ^

m O E CM
ai 15
S- CD 1
a. 3 *
r— .Q IO
•w re in co
a) >• 4>* en
Z in --
CJ
re en—
3 s'o"
C C CM
C O "O
< •— = CM
in o
01 in "^ i
cn-r- 4J
re s o ic
i- UJ 3 CO
ai -a cn
> N cu X!
< eo ai —

_

=3 0 SO
C i- CM
C in «o
•C in c CM
•— O
 o-o en
5 * a-""





i





3 O
C 0) CM —
c u o re
*C C CM ^
CD en
at -a i — .
re u to M
1- C CO CO
aj *— » en



i
•




i
Regulatory
Strategies
with size cutoffs)

!

1
1
i
1

0 O
i i O ro
1* f
|
1
1



O O
0 O
i to en
1 * 0









-o

g1"1 o o
o o
"— '





r- 1
O O O
O O O
O » "
.H ro r*»
f^ to en
U^ 1—4 *— 1











O co -<
t- !. 0
O O
CM
ro o •
CM — en
uo ^- ro
m — i — ^







IV4
Basellnef No Addition;
Controls)
Stage II - Nationwide
Onboard -
Nationwide








o
CO
_4*






o
0
CM










O
o







i
o
s
CM











in

CO
CM
—i








CD
in -r-
re 3
CD C
s. o
* re
«c z
z
•o
1— 1 S-
— re
o
CD .a
en c
re o








S °
G"> — t
ro ^
— *






8 3
"*!. ^
CM










0 0
0 0
"1 ro







O o
0
O* f."
CM CM











— CO
ro

0 °
» S
— <
sn
re
CO
o
•o -c
re "3
— 0
re -z








o
en







o
co
CM










i
in
CM






0
2











s
o
1—4
co
CM
_i








Stage I and Onboard-
Nationwide








o
o
_J"






o
m
«/










o
0
^
CM






|
!?











CM
cn •
o
P^ |
in
CO |
o i

:


'



Stage I, Stage II,
Onboard - Nationwide



]


i
!
i
!

!

i







.;
^


e •»-
3 4->
S U
— a;
x a.
re in
fs qj
t
en
c *

in o
3 4-1
u
in re
01 14.
+•*
JJ 11
in
c 3
CD
in 4^
a) *v
s- s
Q.1—
a) p—
^

c oj-
CD Q.
> CL
•<- 3

CD
*• S
.a in
= 3
3 re
c: r—
O-
f-o
•*-» c
re
— t-
>. o
*= 0
o re

re

re in
•o —
gasoline vapors (rat
11 hood estimate unit r
u a>


O t—
re x





















in
O
u


tfc.
O
jr
^
§

4J
C
CD
in
a>
CL
O)
(4.
O
3
in
o>
-C


in


present value of costs
counted at 10 percent)
in

OJ "d
z —
3



















4>J
C
CD

CD
CL
O
C

CD
J=
4-1


-a
•c
01
•o
•5
W)

in
0
u
o
CD
3


> •
01 (J
ai CD
i- CX
CL
a) ""
C 4-1
ai

4J
in c
M— 3
O
in y
in in
a) i—
c -o
CD
> j=
— 4J
4J 0
<-> 43
£ .
K- in
CD C
« °
in in
O in
u "^
•o S
a)
4^ 14—
= O
3
o a>
U 3


a >
U 1










































in


-------
CJ
CO
CT»
CO
             c  u-
                                                                                                                E -r-
                                                                                                                3 4->
                                                                                                                S  U
                                                                                                                                  S CD
£
to
to
LU
SS
Ul

                                                                                                             !   "S
                                                                                                                                  0) 4->
UJ
             O A9r-
^      !
UJ
UU
CO
             T5 8)-—
                                                                                                                                  -i
                                                                                                                               !   = r-
a
LU
to
             (00   —
                                                                                                                                  3 -a
ca
to
                                            -<      o
co
UJ
              £"5
             ,-:
                             LU      O>
                                               •r-    t-   O 1-
                                                                                           18 T- JJ   — +J
                                                                                             "
                                                                                             i 7- 4->   — 4J   !
                                                                                           •—-os-   i- i.    — -a
                                                                                           .„    -
                                                                                           01 -a  10    iJ      -a •—
                                                                                                                       —  .    !   Q -P
                                                                                                                                  O 4
                                                                                                                                  a. >
                                                                                                                                  (O  =   —
                                                                                                                          S      — .
^-   — 4)
i—    U CTi
                                                                      1-26

-------
 them would cause only a slight increase in cost effectiveness.  The
 in-use efficiency of onboard controls is expected to be about 92 percent
 (based on current levels of tampering to use leaded gasoline, disregarding
 the phase-out of leaded gasoline).  The enforcement costs for onboard are
 lower than Stage II (average annual cost of $0.1 million) since they
 are for the incremental  cost above the current certification and in-use
 testing program, and the incremental  costs of inspecting onboard rather
 than evaporative control  systems on selected vehicles  at the assembly
 line.
      Figure 1-2  graphically depicts annual  incidences  due to benzene
 with either Stage II or  onboard regulatory  strategies  and with no
 additional  controls (baseline).   Estimated  incidence with Stage II  is
 shown  both  for the assumed statutory  phase-in requirements and for  an
 alternative phase-in schedule  suggested by  the American Petroleum
 Institute (API).   The  assumed  statutory phase-in  requirements  used  in
 this analysis  are installation within 2 years for non-independents  under
 a  NESHAP  (Section 112) and 3 years for independents  assuming Section  325
 of the  Clean Air Act applies to  CTG's and Section 112  standards.  The
 API  phase-in schedule  assumes  3 years for nonindependents  and  7 years
 for independents.   Installation  of Stage II  controls was  assumed  to
 begin in  1987  for both the  statutory  and API  phase-in  scenarios.
 Onboard controls  were  assumed  to  be installed on  new vehicles  beginning
 with the  1988  model  year.   Therefore,  all of  the  strategies  shown begin
 at baseline levels  of  about 4.8 incidences expected from 1986  benzene
 emissions from the  entire  gasoline  marketing  system.  The  baseline  (no
 further controls) levels of annual  incidence  also  decrease with time  in
 proportion to  the projected decrease  in  gasoline consumption.  The
 Stage II  strategies  reduce  incidence more rapidly  than  the onboard
 strategy.  The numbers in parentheses on the  graph indicate the differences
 in cumulative  incidence before or after  1995 when  the onboard strategy
 is projected to reduce incidence to below the level reached by the
Stage II strategies.  Thus, although Stage II can  achieve incidence
reduction sooner than onboard,  by 2020 the cumulative incidence reduction
with onboard controls has surpassed the cumulative reduction with
Stage II controls, since the steady-state levels of annual benzene
incidence are about 1.2 for Stage II versus 0.5 for onboard.
                                  1-27

-------
                  Figure  1-2.    Effect  of Onboard  and  Stage  II  Controls  on
                                          Benzene  Incidence
                              (Based on Theoretical  Efficiencies)
5.0
STAGE II  (WITH SIZE CUT-OFFS)   	
   STATUTORY PHASE-IN (2 YRS. NOH-INO., 3 YRS. IND.)

STAGE II  (WITH SIZE CUT-OFFS) 	
   API PHASE-IN  (3 YRS. NON-INO., 7 YRS. INC.)

ONBOARD  (BEGINS IN 1988) 	
                                       BASELINE INCIDENCE  	£r
                                                            A	A
                                                                                A
                                                                                          A	A
                                       (   )  DIFFERENCE IN  INCIDENCE DUE TO EFFECT OF CONTROL OPTION
                                                                                                   A
                                    (4.9 FROM 1986 THROUGH 1995 )
                                              (4.0 FROM 1986 THROUGH 1994 )
                                                           (14.7 FROM 1996 THROUGH 2020 )
     36    38     90
                                                                                              16    18     20
                                                   1-28

-------
     The  control  costs  associated with  each of  the  regulatory strategies
 are  assumed  in  this  study  to be passed  on by producers to consumers of
 gasoline  and vehicles in the form of higher prices.  The magnitude of
 these  price  increases for  components of nationwide  regulatory strategies
 with size cutoffs  are presented in Table 1-9.   Most show gasoline price
 increases of less  than  half a cent per  gallon of gasoline.  Price
 increases for benzene reduction strategies range from 1.5 to nearly
 5 cents per  gallon.  These are average  figures; in  practice they would
 vary both with  location and over time.
     Consumer resistance to these price increases can reduce the sale
 of vehicles  and gasoline.  Estimates of the reductions in the rate of
 consumption  are displayed  in percentage terms in.Table 1-9.  The
 estimated reduction  in  gasoline consumption ranges  from 38 million
 gallons a year  for Stage I to 128 million gallons a year for a-combination
 of Stage I,  Stage  II, and onboard controls, and to  over 1,200 million
 gallons a year for the  most costly benzene reduction option.  For
 regulations  involving onboard controls, annual  LDV  and LOT rate of sales
 are  estimated to decline by 17.7 and 5.3 thousand vehicles, respectively.
     Other economic  impacts of the regulatory strategies were also
 examined.   These included an analysis of the sensitivity of cost
calculations  to underlying assumptions and consideration of distributional
 impacts of the regulatory strategies by firm size.   Results are summarized
 in Table 1-10.
1.3.3  Cost Per Incidence Reduction
    An analysis was performed to determine the residual  costs expended
per cancer incidence avoided for selected nationwide and nonattainment
area regulatory strategies.  The residual  costs were determined by
obtaining the annualized costs of the controls  associated with the
regulatory strategy and subtracting a range of assumed benefit values
per megagram of YOC emissions reduced.   The assumed YOC  benefits are
those in addition to cancer prevention, such as reduction in non-cancer
health effects and agricultural  damage due to ozone.  The residual  cost
per  incidence was then calculated by dividing residual  costs by the
appropriate amount of cancer incidences avoided.
                                1-29

-------

















y?
o

o
LU
•* 2£
0^
H- 0
15 tJU
"•••% 1 1_
O LU
LU
06 a*


fcS
«C 0
a, uj

o
»~j

C^l
•y
o
o
LU


CTl


»— 4

LU
i
H*









?
i
• i
t

i i
1 "i
i
' £
i
" & i
J 11
« «-i
C£ O— •
' U JJ
, r-0. C
j «•£;. CJ
; o o u
! 3 4J a.

! S S
! CJ in
j SPg
: fe°
-p ^
s ^^
1
1
1
r
i
i


i

f
t


i
! ** o
' i»
flj U

S ""*
< 0
u

1
i
i
1
n
t
t
i
*
t
j



i
i





*

§

•
i
— .
CO
O
*-4
^— """*
fl^

tn •— •
•M l_ *^ O
JS:| f

o"~"
^J

ft*



•
i
•&
^~^
^^


*n — .
g ^
S S
o u
 ^



fel
8^-S
•^ fMV C7
pHB "^^ «^-
d W> W
S 4J 4J
re c c
c3 a> a>
~*









S


^Ij'lj
4> »*-
«3 O
*•* 3
U) O
>» w
4^ CO

•is

a —






co en

in in

CO CO
d d

r*» r^



to 10
^4 *•*
d d




r~ m CM r~
r~I — en i~l




in » ao in
So in in
^4 ^H O

odd d






CM CM
CM CM

 en
-H CM tn CM
odd "^
^^ *-• ~~ o

v in en ej
m to en <»
o o o  O
*O •— .fl
*» p c
^ *; o
^ c ^
e o c
re u re
_ _ _ -a —

cj cj cj re cj
en en en o en

crt co en o vi






fn

in

co
d

^
S


to
^M
d




in
§





in
























s. re
a. 1
c o
CJ u-
M CJ
1= S-
ej
CQ





.:•,: i:l:,,,::



















tA
0)


E

t«-
^^
o
i.
0)

e

e
re

tA
e
o
u
£

X
e
o
ex

tA
IA
re
iJ
•a
CJ

u
>^
u
CJ
o
0

U)
re
1 m
! re


!::' 	 "J:""


































tA
e
o

4J
a
3

V
S-
i
o
0
ii
IA
e
a.
e

0)
j=

o

a)


re
, 4M
3
J3
lil
4->
4^
a
V)

O
(J
3
T3
CJ
a:



:~s 	 • :'"



















4J
e
0)
u
 a
C 0)
0 >,
u — .
IA
0) i.
C CJ
"o i^
IA
reen

^ ""^
a x
3
re -i
IO
g,~
re ..
u s_
cj re
> CJ
e "IA
o e
o
•a i—
01 r—
u> re
re cn
cn
c
CJ X

CJ IO
a- ^^
-o
1-30

-------
                           TABLE 1-10
                     ECONOMIC CONSIDERATIONS
The costs of nationwide strategies without facility  exemptions  are  80
to 100 percent higher than for the comparable strategies with  exemp-
tions.  For strategies covering only nonattainment areas,  costs with-
out facility exemptions are 15 to .35 percent higher  than for comparable
strategies with exemptions.

Stage I and Stage II controls will cost 1/2 to 2  1/2 cents more per
gallon of throughouput at small gasoline marketing facilities  than  at
large ones.  Facility exemptions reduce this cost differential, but
do not eliminate it.

Facility exemptions improve the competitive position of  the smallest
facilities, but small facilities tend to be less  efficient than large
f aci 1 i ti es.

This analysis assumes the per vehicle cost for onboard controls is.
$13 per tank.  If, however, the cost really is $25,  the  net present
value of nationwide onboard control  costs would increase by more than
50 percent, and would then exceed those of nationwide Stage I  and
Stage II controls.

This analysis assumes gasoline consumption will decline  in the years
ahead.  If, however, consumption holds at current levels,  then  costs
for Stage I and Stage II controls would be less because  there  would be
more recovery credits.

This analysis assumes the number of gasoline marketing faci1ities re-
mains constant in the years ahead.  If, however,  the number declines,
then costs for Stage I and Stage II controls would be less because
there would be less control equipment needed.
                                     1-31

-------
     In Table 1-11, several of the regulatory strategies are presented
with their corresponding emission reductions.  The emission reductions
are presented as the net present value of all the annual emission
reductions over the study period and as'an annualized value representing
equal emission reductions for each year of the study period.  The
residual costs were determined assuming several  different dollar values
for the benefit of reducing each megagram of YOC emissions.  For example,
in Table 1-11 the annualized emission reduction associated with Stage I
is 0.218 million Mg.  Multiplying this emission reduction by each of
the assumed YOC benefit values yields the annualized VOC benefit in
dollars.
    Table 1-12 presents annualized cost (control equipment and
enforcement costs) and annualized incidence reduction due to benzene
exposure associated with several of the regulatory strategies. " The
cost per cancer incidence avoided, assuming no additional benefits, is
calculated by dividing the annualized costs by the annualized incidence
reduction.  Table 1-13 takes this one step further by incorporating the
annualized VOC benefits into the analysis.  The values presented represent
the residual cost, assuming varying benefits for reducing VOC emissions,
of reducing cancer incidences due to benzene exposure.
    Table 1-14 contains a similar analysis to that used in Table 1-13,
except  that Table 1-14 was developed using the sum of the incidences
due to  benzene and gasoline vapors.  It is assumed that the incidences
due  to  benzene exposure and the  incidences due to gasoline vapor exposure
are  additive since the respective exposure results in different types
of cancer incidences (leukemia  in the case of benzene exposure and
liver or kidney tumors in the case of gasoline vapor exposure).  Net
costs per annual  incidence avoided are given using both plausible upper
limit and maximum likelihood estimate risk factors for gasoline vapors.
                                  1-32

-------
  OO
 UJ
 ca
 Q
 «C
 00
 OO
 o
 UJ

 UJ
 <:
UJ
CO
1 t s
i ' §
o
i CM

1 1
 i cn
— +; , ^

CO *^ O
c o> i o
o> c o
CO QJ 1 .
Tl .
So '"
> 25 i
! -a *" i cn
co cn s
| M C 1 ^.

fO 3 in
3 I/> 1 4*
C in
5< •
| i cn
O
I Si

i
i
CO
S-
^3 C CO
o> c o sa.
N 0 •—
i—  U S
(a in 3
3 i- -o c
= S co o
i
w -g;
 o c
O QJ i^ O
™ 3 « —
a. »— « •—
fa *p- f^
O) tu £
z —
en ">i*3)
CO (J U_
4-J C u_

n 3 JT
3 | +J
cn = —
cc "" S

1
10 co in in _ «• cn
co •— * r*. cn cn r*^ ,— i
•a- <-< CM — i CM in
I
1
1

I
!
p^. CTI vo •- * n in cn
CM co in CM <}• o co
f> CM ^H CM f*0

j





eo cn r— r~- in r-» cn
— i in n  —i — < OJ '
o o o o o o o ;
i


^-t ^o *f rr cn mm;
CM O O -H O -* CM


^ tJ
co uo >  to ;
^^r:r;r:-o 5 •
i. o
QJ <1> OJ <3J 4) fQ "*^» *^^
cncnCTioicno S3

-------
          TABLE 1-12.  BENZENE REGULATORY COSTS AND INCIDENCE REDUCED
Regulatory Strategy
(with size cutoffs)
(In-use efficiency)
Stage I
Stage II-NA (87%)
Stage II-NA (56%)
Stage II-Nation (86%)
Stage II-Nation (56%)
Onboard (92%);
w/o evaporative
w/ evaporative
Annual i zed
Costs
($ Millions)3
91
62
52
183
146

199
199
Annuali zed
Benzene
Incidence
Reduction"
0.06
0.83
0.41
1.92
1.13

1.44
1.66
Costs
($ Millions
per Benzene Cancer
Incidence Avoided)
1,564
75
126
95
-128

138
120







Includes control equipment and annual enforcement costs.

Incidence reduction after controls.  Before-control  annualized incidence:
Stage I - 0.18, Vehicle Refueling = 4.09.
                                 1-34

-------














UJ
o
UJ
0
1— 4
0

I-H

o;
UJ
o

^C •*— ™*
<_5 00
1—
CU H-.
UJ U_
Q- UJ
z
OO UJ
K- 03
OO
0 0
0 0
^>
>•
o =i
h- >-i
< s:
_1 =>
=3 00



•
oo
i
i— i

UJ
_j
ca
•a:
1—

















^—,
•a
CO
•o
• fH.
0
>

CO
c i-'
CU 0
•o
•i- CO
(J 3
C p™
(-4 (B
s_ >
CO -P
O •!-
C M-

Q.
cn
to c
C -1-
0 &
•i- 3

t— 
to
0
o





•a
CO CO
M 0
•r- C
r— CO
3 -f
C u
C C
«t 1—4







^^
cn
+j
(^
s_
4^
OO

>^
SM.
O
4»
rO
i—
3
CD
CO
a;
cn
s:

o
0
o
CM


CD
21

O
O
tn
i— i
*«•


cn
s:

o
o
o
t— 1




cn
^F
^•^
o
o
uo
4V^


cn
2!
0
to
1 ^,J
1 ^^






C -^•^
O CO
•r— C
•P CO
3 £=
•o co
CU CO
o;— •







^~»
>>
c
CO
•r—
0

*4™
^4—
CO

CU
to
3
1
C
1—4
— •



O O O O O O O








O O O O CM O O









o ^r uo cri «* LD o









.a
^3 ^^ t-H i^* ^O CO f^O
CO CO LO CO O^ ^"






^^ ^^ ^^ ^^ r^^ ^7 co
CM lf> O f-> O CM CO
IO r- 4 ,— 1 '^4









*«o
co co ^^ CT^ *"•' ^f r*^
O O O ^H r— 1 r-l »-H









S* 3*
to vo
CO UO CO CO
&$ ^S *p™ *r-
r-» to c: c: -P -p
CO IjO ^D ^3 * * ^O ^3
— ~— -r- -i- — . i. S.
•P -*-> &S O O
^ ^ (O fQ CM CL CL
'z. -z. "zz. "z. cn  >
i— i t— i i— i i— i CO CO
1— 1 1— 1 )— 4 . )— 4 1 	 1 "O
1- O
CO CO CO CO CU (O *^> • —
cn cn cn cn cn o 5 !5
(T5 (O rt3 (O (O <^
•P 4*^ 4~* P P C
oo oo oo -oo oo o


^^
1—4
,— H
1
r— 1
CO

"re
1 —



s_

"~^
I/)
^(J
>rw
<4-
co

CO
ca •

o
o


-a
cu
N
•r—


to
J=
"*"*
CO
-p

o
•p-
•a
(^
•t—

t/1
o
i.
CO
M

1-35

-------













03
O

f*S

O

^g*


LU


jgg* i^i

>-« o
0 Q.
•Si <
»— I *rm
^^ ^^
O± CO
LU <£
0 OS
22
00

fHft

of O
I—

//S ^tf*

C? utv
^-<
ooi
o.




S
• •>» 1 Q-
* '8 ,
<0 | in •
VI *» | UJ
CJ ill ' vt | _J
O I •
k. . E


(Q 19 _J
O > 1
.Q *J 1 Cn 1 •
^ *** E Cw
O»- 1 •». 1
0) o
•f* C 1 CO 1 •
01 o LU
jMca i - i
CO -M — 1
^ ^^ ^*
of
ra i _j
§e i .
— 1 =5
•»•• g | O •
>- 9 s i a-
I— U> t ^
•»- at o I
E< t 0
in i uj
44- | V* •
^•^ 1 _J
1 •
4** 1
O 1
O •
1 _J

1 3
i E ol
. 0 '
in i •
i cv ui
i _J
( '

i



•a aj c
 2 o *^
r— -OotS — I- 4J _J

C C CJ CQ (O QB>^ •
< a. —






•a ffl c= 'o
 .
*^cj4J^ ^j=:iouj
r— T3 O O •<- — S •
<0— 3 Hh X^— -^ _J
3U13IM W U *» .
c c 01 m EJKV>E
«C «J






Si s;
, •£

: 2 «
* M Si
' -3 ?
en c
1 QJ »^«
r** •*—

o o o o o o o





o o o o o o o


o o o o o o o





O O O O O 'O O






O o **"> e»j ^* ^r o





O — « Lf) CO ^JO IO O

O C*) *-O IO CO GO PO

.

o ir> ^-* eo <*) ro in
*— * f— t ^H



O ^0

1—4


CO ^f — i ^O IO O
ro






01 «-l
•M o t" CNJ f r^» **^
^* OU ^ -HCJ







^4 —I —4 CO
~« to ro en ^ o^j







9* &* 4) 4J
 >
co u^ *^ ^

6-Q (V<9 (Q to
r- uo c c i- i-
co to o o o o

< «C « (0 CSJ > >
^ z z sz en o> — t »— 1 *— « O
S- 5 S
OJ OJ  «3 ^
(/> t/> CO tO t/> O
1-36





























«««
^4
f— t
1
	 «
i
H-


2
**•

(/I
4J
•^
 4->
•^ w)
— o
> °
IO IT
HI
CU N
e 1^
HI IO
•O 3
•f- C
I 5

s_
D
u
IO
(J
u
CJ

VI
o
ia























o
4^
u

u_

&.
V)
s-
4J

c
3
+J
"s
r—

HI -
a.
a.
3
0)

•^

t/>
3
"S.
•a
c
ia
^
o
*J
(J
IO
JH
VI

u
4->
e

0)
4->

13
VI
at

•o
o
o
j=
1 S
IO
> S

IO f-
cn x
10

^ "™
i— CU
a. ^
j->
HI



3
+. (J
N (O
.-.-




















































,
VI
4J
VI
o
u
HI
J=
4-* •
J=
O>
^



VI
4->
•t—
(fa.
OJ
e
01

s
>.

J=


4J
(O
-C
4->

CD

y
•o
c

VI
2
HI
rvi
•a


-------
                 2.0   INDUSTRY DESCRIPTION AND PROFILE

2.1  GENERAL INDUSTRY  DESCRIPTION
     The gasoline marketing network consists of all the storage and
transfer facilities that move gasoline from its production to its
end consumption.  The  network includes tanker ships and barges, pipelines,
tank trucks and rail cars, storage tanks, and service stations.  Crude
petroleum is shipped to refineries, which manufacture the wide range of
liquid petroleum products.  Finished gasoline is then distributed in a
complex system comprised of wholesale and retail  outlets.  Figure 2-1
depicts the main elements in the marketing network.
     Gasoline is delivered to bulk terminals from refineries by way of
pipeline, ship, or barge.  Large transport trucks (30,000 to 38,000
liters, or 8,000 to 10,000 gallon capacity) deliver the gasoline to
service stations or to intermediate bulk storage facilities known as
bulk plants.  Generally, a terminal is defined as any bulk wholesale
gasoline marketing outlet that receives product by pipeline, ship, or
barge, and delivers it in tank trucks to customers.  A bulk plant
typically receives product by truck from a terminal and has a smaller
storage capacity than a terminal.  In addition, daily product throughput
at a terminal  is much greater, averaging about 950,000 liters
(250,000 gallons), in contrast to about 19,000 liters (5,000 gallons)
for an average size bulk plant.
     Both bulk terminals and bulk plants deliver gasoline to private,
commercial, and retail  accounts.   Bulk plants, using 5,700 to 11,000
liter (1,500 to 2,900 gallon)  capacity delivery trucks, service primarily
agricultural accounts and service stations that are either long distances
from terminals or inaccessible to the large transports.  The trend in
recent years has been toward more terminal deliveries at the expense of
bulk plant deliveries.   Retail and commercial  level businesses include
the familiar service stations, as well  as commercial  accounts such as
fleet services (rental  car agencies,  private companies, governmental
                                  2-1

-------
  Imported
  Gasoline
Ref i nery
w i
p
                             Barge
                           Pi pel-
                            Tanker  /
A
me )
                             Bulk
                           Terminal
                'Imported
                   or
                Domestic
                  Crude
         Wholesale
         Distribution
         Level
                    Commercial,
                       Rural
Consumer
 Figure 2-1 .  Gasoline Distribution in the U.S.
                         =  Storage

                         =  Transport
                        2-2

-------
agencies), parking garages, and buses.   Another important consumer
category is about 2.7 million small  farms.
2.2  GASOLINE .MARKETING OPERATIONS AND  THEIR EMISSIONS
     The pollutants from each of the gasoline marketing facilities are
essentially the same; however, the operations which occur at each  of
the facilities differ, and the rates of emissions to the atmosphere
differ.  The emissions consist of a mixture of volatile organic  compound
vapors and air, and are discussed in Section 2.2.1.  Because of  the
complexity of the industry, Sections 2.2.2  through 2.2.6 will  present
separate discussions of the operations  at each industry sector and of
the associated emission rates of gasoline vapors, benzene, ethylene
dibromide (EDB), and ethylene dichloride (EDC) from typical facilities.
2.2.1  Gasoline Composition
     Motor gasoline is a complex mixture of varying amounts of paraffins,
naphthenes, olefins and aromatics (such as  benzene).  In addition,
small amounts of additives (such as tetraethy lead, 1,2-dichloroethane
[or Ethylene dichloride (EDC)] and 1,2-dibromoethane [or ethylene  dibromide
(EDB)] are used to achieve specific product qualities.  This section
discusses and provides estimates of the amount of benzene, EDB and EDC
vapor in gasoline vapor.
     It is well known that wide variations  in benzene content exist in
gasoline.  Analysis of 1977 gasoline pools  produced by various refin-
eries showed that benzene content in the liquid gasoline varied  from
0.15 to 4.26 volume percent with a national average of 1.3 volume
percent.1  In a review of more limited data over the period since  1977,
the benzene content variation was within the 1977 range while no
difference below or above the 1977 average was apparent.2  Only  one
published source was found reporting the concentration of EDB and  EDC
in liquid gasoline.3 This study estimated that the EDB and EDC
concentrations varied from 80 to 150 and 150 to 300 ppm, by volume, in
leaded gasoline, respectively.
     An earlier EPA study on benzene emissions from gasoline marketing
concluded that about 0.008 grams of benzene per gram of hydrocarbon
vapor at 26.7°C (SOT) was an appropriate figure to use for 1.3 volume
percent benzene in liquid gasoline.^  The figure of 0.008 was based on
laboratory and field tests conducted at temperatures varying  from
                                  2-3

-------
25 to 31°C.5»6>7  The study also warned that any attempt to adjust the data
to other temperatures would introduce an indeterminable degree of
error.  In addition, it is well understood that temperature has a major
influence on vapor-liquid equilibrium concentrations.  The remainder of
this section describes the approach used to estimate the relationship
of temperature on the gasoline vapor-liquid equilibrium.
     The equations shown in Figure 2-2 were used to calculate the
emission estimates for VOC, Benzene, EDC, and EDB.  Equation 1 is the metric
equivalent to the AP-42 emission formula for YOC loading and was derived
from the ideal gas law.  Instead of obtaining the true vapor pressure
from the ASTM distilation curve and the molecular weight from tables,
equations 2 and 3 were used to add the capability of varying temperature
and Reid vapor pressure (RVP).8»9  Equations 4, 5, and 6, which were
used for determining Benzene, EDC and EDB emissions, were developed
from the ideal gas law, assuming equal vapor to liquid volume ratios
and temperatures, an activity coefficient expression, and the Antoine
vapor pressure equation and coefficients.
     Uncontrolled and controlled VOC emission factors and data are
readily available from the many sources referenced in subsequent sections
of this report.  In order to estimate benzene, EDC and EDB losses,
uncontrolled emissions for the four pollutants were estimated by using
the equations presented on Figure 2-2.  Figure 2-3 graphically presents
the mass emission factors, assuming a saturated vapor space.  Benzene,
EDC and EDB to VOC emission factor ratios were calculated by simply
dividing each of the three pollutants emission factors by the VOC
emission factor.  Figure 2-4, graphically presents these emission
ratios.  Throughout the evaluation discussed in this report, these
ratios (for RVP 10) were used to estimate benzene, EDB and EDC emis-
sions in known amounts of hydrocarbons.  Different ratios, based on
product temperature, were used to calculate emissions from vehicle
refueling operations than were used for all other gasoline marketing
sources.  Ratios based on 70°F were used for vehicle refueling (BZ/VOC-
0.0066, EDC/VOC-0.00053, EDB/VOC-0.000052) while ratios based on 60°F
were  used for all other gasoline marketing operations (BZ/VOC-0.0060,
EDC/VOC-0.00047, EDB/VOC-0.000046).  These ratios were used for community
exposure from service stations, but were not used to calculate risks
                                  2-4

-------
                        FIGURE 2-2   EMISSION  EQUATIONS
      = 1492
          (   P M SF \
          \ T + 459.S/
                                                                   (1)
      (r         / 413.0   \-|                           /   1042
= 6XpJ  0.7553 - \T + 459.6jl  S0-5 log (RVP)  -   1.854  - \T  + 459.
      U                                       L

   [/   2416    \       -,            /   8742     \         )
+  |_\T + 459.6 /- 2.013J log (RVP) - \T +  459.6 )  +  15.64>
                                                                        0-5
 M = A + B(RVP)  + C(RVP)2  +. D(RVP)3

                    \ v  SF
 EBZ = 18.34
                             )           2788.51
                         exp U5.9008  -  K-52.36
 EEDC = 25
          (_LL§F\    \          I   2927;17   \
       .58\    K  ybxp \ 16.1764  -\K  -  50.22   /
                       \       .                i
EEDB = 44.95
 Where:
                 \ v  SF
                  K
                               '.568  -
                                       1903.6
                       10
(2)


(3)


(4)



(5)





(6)
   A = Constant for Gasoline   =    72.833
   B = Constant for Gasoline   =    -1.3183
   C = Constant for Gasoline   =    0.15079
   D = Constant for  Gasoline  =  -  0.0087302
     = Loading Emission Factor (mg  BZ/liter loaded)
     = Loading Emission Factor (mg  EDB/liter  loaded)
EEDC = Loading Emission Factor (mg  EDC/liter  loaded)
EVQC = Loading Emission Factor (mg  VOC/liter  loaded)
   K = Stock Temperature (°K)
   M = Molecular Weight of Vapors (Ib/lb-mole)
   P = True Vapor Pressure (psia)
 RVP = Reid Vapor Pressure (psi)
   S = Slope of ASTM distillation curve  at 10 percent  evaporated;
       for gasoline, S = 3
  SF = Saturation Factor
   T = Stock Temperature (°F)
   V = Volume Percent of Compound in Gasoline (Assume  EDB = 0.015% by Volume,
       EDC = 0.03% by Volume)3
     = Activity Coefficient (Bz = 1.17,  EDB and  EDC  =  2.1)

 exp  x [ =  ex
  Reference 3.
                                    2-5

-------
                     Figure 2-3. Mass  Emission Rates ^Saturation
     10000
       1000
-o
O)
 (A
 c
 ra

 •4-s
 52
 E
 
 rc>
 to
 •fj
 (13
 o:

 c
 o
       0.01
                  30     40     50    60     70    80    90    100   110    120


                                        Stock Temperature  (°F)
                                              2-6

-------
              Figure 2-4. Mass Emission  Ratio of BZ, EDC, &  EDB  to  VOC in Vapor
     0.01


- 1






	 j 	 _ 	
	 1 	 	

	

-=.•• 	 	 l
	




• 	 ^


~^= 	 -1


	 ^



	 ~- 	 .
1
r i n RVD ir^=-
 to
 a>
 l/l
 c:
 QJ


-o
 i.
 o
o
o
CO
a
o
Q
O

O
to
1/1
in
1/1
(D
    0.001
0.0001
    0.00001

                                         Benzene g-	=^=--5=
                  30    40     50    60    70     80     90    100   110   120
                                       Stock Temperature (°F)
                                               2-7

-------
from benzene and-gasoline vapor exposure during self-service refueling,
which were based on measured exposure data.
2.2.2  Bulk Terminals
     Bulk gasoline terminals serve as the major distribution point for
the gasoline produced at refineries.  Movement of gasoline at a bulk
terminal involves loading, unloading, and transfer of the liquid from
storage tanks into tank trucks.  Gasoline stored in large aboveground
tanks is pumped through metered loading areas, called loading racks,
and into delivery tank trucks, which service various wholesale and
retail accounts in the marketing network.  Loading racks contain the
equipment (such as pumps, meters, piping, grounding, etc.) necessary
to fill delivery tank trucks with liquid products.  Terminals generally
utilize two to four rack positions for gasoline, but there can be as
many as eight to ter\rack positions at large throughput terminals.
Each loading rack will typically have from one to four loading arms,
depending on the products available for loading at that rack position.
Each arm is dedicated to one product.
     Emissions from the tank truck loading operations at terminals
occur when the product being loaded displaces the vapors in the delivery
tank and forces the vapors to  the atmosphere.  Loading may be performed
using either splash, top submerged, or bottom loading methods.  Top
loading involves loading of products  into  the tank truck compartment
via  the hatchway which is located on  top of the truck tank.  Gasoline
can  be  loaded directly into the compartment through  a top loading fill
pipe (splash fill).  Attachment of  a  fixed or extensible  downspout  to
the  fill  pipe provides a means of introducing the product near  the
bottom  of  the tank  (submerged  fill).  Top  splash loading creates
considerable turbulence  during loading  and can create a vapor mist
resulting  in higher  emissions  from  the  truck  loading operation.   Submerged
loading greatly  reduces  the turbulence,  and therefore reduces  the
emissions.  Bottom  loading  refers simply  to the loading of  products
into the  cargo  tank  from  the  bottom.  This results  in the same  emission
reduction  as associated  with  top  submerged loading.  The  trend  in the
industry  is to  build new terminals  with bottom loading  racks  and  to
convert existing terminal  top  loading racks to bottom loading.  Some  of
                                   2-8

-------
the advantages cited for bottom loading include:   (1)  improved safety,
(2) faster loading, and (3) reduced labor costs.   Emission factors and
emissions from loading rack operations at a typical  950,000 liters/day
(250,000 gallons/day) terminal  are summarized in Table 2-1.
2.2.3  Storage Tanks at Terminals
     A typical terminal  has four or five aboveground storage tanks for
gasoline, each with a capacity ranging from 1,500 to 15,000 m^ (9,400
to 94,000 barrels).10 Most tanks in gasoline service have an external
floating roof to prevent the loss of product through evaporation and
working losses.  Fixed-roof tanks, still used in some areas to store
gasoline, use pressure-vacuum (P-V) vents to control breathing losses
and may use vapor balancing or processing equipment to control working
losses.  A typical  fixed-roof tank consists of a cylindrical steel
shell /with a cone-  or dome-shaped roof that is permanently affixed to
the tank shell.  A breather valve (pressure-vacuum valve), which is
commonly installed on many fixed-roof tanks, allows the tank to operate
at a slight internal pressure or vacuum.  Because this valve prevents
the release of vapors only during very small changes in temperature,
barometric pressure, or liquid level, the emissions from a fixed-roof
tank can be appreciable.
     The major types of emissions from fixed-roof tanks are breathing
and working losses.  Breathing loss is the expulsion of vapor from a
tank vapor space that has expanded or contracted because of daily
changes in temperature and barometric pressure.  The emissions occur in
the absence of any liquid level change in the tank.  Emptying losses
occur when the air that is drawn into the tank during liquid removal ~
saturates with hydrocarbon vapor and expands, thus exceeding the fixed
capacity of the vapor space and overflowing through the pressure vacuum
valve.  Combined filling and emptying losses are called "working losses."
     A typical external  floating-roof tank consists of a cylindrical
steel shell equipped with a deck or roof that floats on the surface of
the stored liquid, rising and falling with the liquid level.  The liquid
surface is completely covered by the floating roof except in the small
annular space between the roof and the shell.  A seal attached to the
                                  2-9

-------
   Table 2-1.   UNCONTROLLED EMISSIONS FROM  GASOLINE TANK  TRUCK
          LOADING OPERATIONS AT A TYPICAL BULK  GASOLINE TERMINALa
Emissions, Mg/yr
Loading Method
Submerged Loading^
Top Splash Loading
Balance Service
Gasoline
Vapor
Emission
Factor5
mg/ liter
600
1440
960
Gasoline
Vapors
200
500
300
Benzene EDBC EDCC
1 0.004 0.05
3 0.01 0.1
2 0.01 0.1
 Typical  terminal  gasoline throughput =950,000  liters/day
 (250,000 gallons/day).

3Ref.  11.
<%
'Occurs from leaded gasoline only.   Assumes  for  the  base year  that 48
 percent of total  throughput at this typical  facility  is leaded  gasoline
 (Ref.  18).

 Submerged loading could be either  top submerged loading or  bottom
 loading.
                               2-10

-------
roof touches the tank wall  (except for small  gaps in some cases)  and
covers the remaining area.   The seal  slides against the tank wall  as,,  —
the roof is raised or lowered.
     An internal floating-roof tank has both  a permanently affixed roof
and a roof that floats inside the tank on the liquid surface (contact
roof), or supported on pontoons several inches above the liquid surface
(noncontact roof).  The internal floating roof rises and falls with the
liquid level.
     Standing-storage losses, which result from causes other than a
change in the liquid level, constitute the major source of emissions
from external floating-roof tanks.  The largest potential source of
these losses is an improper fit between the seal and the tank shell
(seal losses).  As a result, some liquid surface is exposed to the
atmosphere.  Air flowing over the tank creates pressure differentials
around the  floating roof.  Air flows into the annular vapor space on
the leeward side and an air-vapor mixture flows out on the windward side.
     Withdrawal loss is another source of emissions from floating-roof
tanks.  When liquid is withdrawn from a tank, the floating roof is
lowered, and a wet portion of the tank wall is exposed.  Withdrawal
loss is the vaporization of liquid from the wet tank wall.
     As ambient wind flows over the exterior of an internal floating
roof tank,  air  flows into the enclosed space between the fixed and
floating roofs  through some of the shell vents and out of the enclosed
space through others.  Any vapors that have evaporated from exposed
liquid surface  and that have  not been contained by the floating deck
will be swept out of the enclosed space.  The withdrawal loss from an
internal floating-roof tank is similar to that discussed for external
floating roofs.  The other losses, seal losses, fitting  losses and deck
seam losses, occur not only during the working operations of the  tank
but  also during free standing periods.
     Table  2-2  illustrates the magnitude of fixed  roof and  floating
roof emissions  from  a 950,000 liter/day terminal with  four  storage
tanks  for  gasoline.
                                   2-11

-------
            Table  2-2.   EMISSIONS  FROM  GASOLINE  STORAGE  TANKS  LOCATED  AT
                                A  TYPICAL TERMINAL3
Emissions, Mg/yr
Storage Method
Fixed Roofd
Working Losses
Breathing Losses
Floating Roof6
Working Losses
Storage Losses
Gasoline
Vapor
Emission
Factor"5 Gasoline
Mg/yr Tank Vapors

34.2 140
8.8 40

f 0.1
9.6 40
Benzene EDBC EDCC

0.8 0.003 0.03
0..2 0.001 0.008

0.001 <0.001 <0.001
0.2 0.001 0.009

 Terminal  with 950,000 liters/day (250,000  gallons/day)  with  four  storage
 tanks  for gasoline.

 See Appendix B (Section B.2.2).
•*
'Pertains  to leaded gasoline only.   Assumes for the  base year that 48  percent
 of the gasoline throughput for this typical  facility  would be  leaded  gasoline
 (Ref.  18).

 Typical fixed-roof tank based upon capacity  of 2680 m3  (16,750 bbls.)
a
"Typical floating-roof tank based upon capacity of 5760  m3  (36,000 bbls.)

 Emission  Factor =  (0.46 x 10~7 Q)  Mg/yr, where Q is the throughput through  the
 tanks  in  barrels.
                                     2-12

-------
•2.2.4  Bulk  Plants
     Bulk  gasoline  plants  are  typically secondary distribution
facilities which receive gasoline from bulk terminals by truck transports,
store  it in  above ground storage tanks, and subsequently dispense it
via  smaller  account trucks to  local farms, businesses, and service
stations.  A typical bulk  plant has a throughput of about 19,000 liters
(5,000 gallons) of  gasoline per day with storage capacity of about
189,000 liters (50,000 gallons) of gasoline.12  EPA defines the bulk
plant as having a throughput of less than 76,000 liters (20,000 gallons)
of gasoline  per day averaged over the work days in one year.
     As discussed in the previous section, vapors can escape from
fixed-roof storage  tanks at bulk plants, even when there is no transfer
activity.  Temperature induced pressure differentials can expel vapor-
laden air or induce fresh  air  into the tank (breathing loss).- Liquid
transfers create draining and filling losses which combined are called
"working losses".
     The two basic  types of gasoline loading into tank trucks at bulk
plants are the same as those used at terminals.  The first is the
splash filling method, which usually results in high levels of vapor
generation and loss.  The second method is submerged filling with
either a submerged fill pipe or bottom filling, which significantly
reduces liquid turbulence and vapor-liquid contact,  resulting in much
lower emissions.   Table 2-3 indicates the uncontrolled emissions from a
typical bulk plant.   In a 1976 survey of bulk  plants,  75 percent used either
top-submerged filling or bottom filling and 25 percent used top splash
filling.13  These bulk plants which use top splash filling are typically
located in States with no control  regulations, or in attainment areas of
those States with regulations where no control is required.
2.2.5  Tank Trucks
     Gasoline tank trucks have been demonstrated to  be major sources of
vapor leakage.   Some vapors may leak uncontrolled to the atmosphere
from dome cover assemblies, pressure-vacuum (P-V)  vents,  and vapor
collection  piping and vents.   Other sources of vapor leakage on tank
trucks that occur less frequently  include  tank shell  flaws,  liquid and
vapor transfer  hoses,  improperly installed or  loosened overfill  protection
                                  2-13

-------
     Table 2-3.  UNCONTROLLED EMISSIONS FROM A TYPICAL BULK PLANT3
                                               Emissions,  Mg/yr
  Emission Source

Storage Tanks

  - Breathing Loss
  - Filling Loss
  - Draining Loss

Gasoline Loading
 Racks
Gasoline
 Vapor
Emission
Factorb
mg/1i ter
   600
  1150
   460
Gasoline
  Vapors
    3
    7
    3
Benzene
  0.02
  0.04
  0.02
 EDBC    EDCC
<0.001
<0.001
O.001
0.001
0.002
0.001
-
-
-

Splash loading
Submerged loading
Submerged loading
(Balance Service)
1440
600
960

8
3
5

0
0
0

.05
.02
.04

<0
<0
<0

.001
.001
.001

0
0
0

.002
.001
.001

a
 Typical bulk plant with a gasoline throughput of 19,000 liters/day
 (5,000 gallons/day).
 Reference 11.

'Pertains to leaded gasoline vapors only.
 percent of the throughput of this typical
                      Assumes for the base year that 48
                      facility is leaded gasoline (Ref.  18)
                                   2-14

-------
sensors, and vapor couplers.14  This leakage has been estimated to be
as high as 100 percent of the vapors which should have been captured and
to average 30 percent.15  Since terminal  controls usually coincide in
areas where trucks are required to collect vapors after delivery
of product to bulk plants or service stations (balance service), the
emission factor associated with uncontrolled truck leakage was assumed
to be 30 percent of the balance service truck loading factor (960
mg/liter x 0.30  =  288 mg/liter).
2.2.6  Service Stations
     The discussion on service station operations is divided into two
areas:  the filling of the underground storage tank and automobile
refueli ng.  Although termi nals and bul k pi ants also have two di sti net
operations (tank filling and truck loading), the filling of the
underground tank at the service station ends the wholesale gasoline
marketing chain.  The automobile refueling operations interact directly
with the public and control of these operations can be performed by
putting control equipment on either the service station or the
automobile.
     Normally, gasoline is delivered to service stations in large tank
trucks from bulk terminals or smaller account trucks from bulk plants.
Emissions are generated when hydrocarbon vapors in the underground
storage tank are displaced to the atmosphere by the gasoline being
loaded into the tank.  As with other loading losses, the quantity of
the service station tank loading loss depends on several variables,
including the quantity of liquid transferred, size and length of the
fill pipe, the method of filling, the tank configuration and the gasoline
temperature, vapor pressure, and composition.  A second source of
emissions from service station tankage is underground tank breathing.
Breathing losses occur daily and are attributed to temperature changes,
barometric pressure changes, and gasoline evaporation.
     In addition to service  station tank loading losses, vehicle
refueling operations are considered to be a major source of emissions.
Vehicle refueling emissions  are attributable to vapor displaced from
the automobile tank by dispensed gasoline and to spillage.  The major
                                  2-15

-------
factors effecting the quantity of emissions are gasoline temperature,
auto tank temperature, gasoline Reid vapor pressure (RVP), and dis-
pensing rates.  Table 2-4 illustrates the uncontrolled emissions from a
typical gasoline service station.  The emission factors presented in Table
2-4 are from EPA's AP-42 document.16  The California Resources Board (ARB)
has performed more recent testing and has estimated that refueling emissions
are about 1200 mg/liter.17  The AP-42 emission factors for vehicle refueling
have not yet been revised, and AP-42 factors were used to estimate emissions
from all other gasoline marketing sources.  Therefore, as a matter of
consistency, the AP-42 factors were used to estimate vehicle refueling
losses throughout this report.
      Emissions due to breathing losses from the vehicle fuel tanks are
controlled by existing carbon canister systems on the vehicle {evapora-
tive control systems).  However, preliminary data from EPA's emission
factors program indicates that in-use evaporative emissions appear to
significantly exceed the evaporative HC standard, primarily because in-
use fuels typically have higher volatility and produce larger amounts
of evaporative HC (when compared to the fuels used for certification
testing) which cannot be absorbed by the current charcoal canisters.
Preliminary estimates of the level of these excess evaporative emissions
are further discussed in Section 3.7.3 (Effectiveness of Technologies).
2.3  BASELINE EMISSIONS
     Baseline emissions are the emissions from gasoline marketing
sources in some selected "base" year.  The purpose of establishing an
emission baseline is to be able to estimate the impacts of reducing
emissions from this baseline through the implementation of additional
control measures.  The baseline emissions must take into account the
level  of control already in place in the base year to get an accurate
assessment of the impacts of the control  alternatives.  The base year
for the gasoline marketing source category was selected as 1982.   This
year was selected because this was the final  implementation date for
many State regulations concerning gasoline marketing sources and
                                  2-16

-------
 Table 2-4.  UNCONTROLLED  EMISSIONS FROM A TYPICAL SERVICE STATION3
                                          Emissions, Mg/yr
Gasoline
Vapor
Emission
Factors'3
Emission Source mg/liter
Gasoline
Vapors Benzene EDBC EDCC
Underground Storage
Tanks
-



Tank fil ling losses
e Submerged fill 880 -
• Splash fill 1380
- Breathing losses 120

2 0.01 <0.001 <0.001
3 0.2 <0.001 <0.001
0.3 0.002 <0.001 <0.001
Automobile Refueling
-
-
Displacement losses 1080
Spillage 84
2 0.01 <0.001 <0.001
0.2 0.001 <0.001 <0.001
a . .
 Typical  service station has a gasoline throughput of 190,000 liters/month
(50,000 gallons/month).
b
 Reference 11.
c
 Pertains to leaded gasoline only.   Assumes for the base year that 48
 percent  of the total  throughput is leaded gasoline (Ref. 18).
                                 2-17

-------
because the latest data available on facilities and gasoline consump-
tion at the time of the development of this document were representa-
tive of 1982.  Table 2-5 summarizes the baseline emissions calculated
for the gasoline marketing industry in base year 1982.
     The general approach for establishing the emission baseline was
basically the same for each sector of the industry.  Data was obtained
on the level of control already used by the States and anission factors
were selected to represent this level of control.  Uncontrolled areas
were defined and emission factors were selected to represent the type
of loading or type of operations in those areas.  Emissions were calcu-
lated by multiplying the emission factors by the.corresponding throughput
for the controlled and uncontrolled areas.  Appendix B contains a
detailed discussion of the procedures used to establish the emissions
baseline.
                                  2-18

-------
                           Table  2-5.   SUMMARY  OF  BASELINE  EMISSIONS

                               FOR  GASOLINE  MARKETING  FACILITIES

                                       FOR BASE YEAR 1982
                                            Emissions, Mg/yr
Faci
Bulk
-
-
Bulk
Gasoline
lity Vapors
Terminals
Truck Loading 140,000
Storage Tanks 56,000
Plants 208,000
Benzene EDBa EDCa

840 3 30
340 1 15
1,250 5 50
Service Stations
-
-

Underground tanks 222,000
Automobile refueling 407,000
Total 1,033,000
1,330 5 50
2,690 10 100
6,450 24 245
a
 Applies to leaded gasoline only.   Leaded gasoline consisted of approximately
 48 percent of total  consumption in 1982.&                         "*„„<»MSijr
                                            2-19

-------
2.4  References

1.  "Cost of Benzene Reduction in Gasoline to the Petroleum  Refining
    Industry," A.D. Little, Inc.  Report to EPA.   EPA Publication  No.
    450/3-78-021.  April 1978.

2.  "Motor Gasolines," Semi-Annual Reports U.S.  Department of Energy
    by Bartlesville Energy Technology Center. Reports covering 1977 to
    Winter 1982-83.
3.
4.
     McDermott, H.J., and Killiany, S.E. "Quest for a Gasoline TLV"
     American Industrial Hygiene Association Journal  (AIHA).  February  1978.

      Standard Support Environmental Impact Statement for Control  of
      Benzene from Gasoline Marketing Industry. U.S.  Environmental
      Protection Agency, Office of Air Quality Planning and Standards,
      Emission Standards and Engineering Division.  Research Triangle
      Park, N.C.  (Draft Report, never finalized). June 21, 1978.

 5.  Reference 3.

 6.  Runion, Howard E.  "Benzene in Gasoline."  American Industrial  Hygiene
     Association Journal.  May 1975.

 7.   Scott Environmental Technology.  Analysis of Vapor Samples from
      Gasoline Storage Tanks, Colonial Pipeline Company, Greensboro.
      florth Carolina.  Report for U.S. Environmental  Protection Agency.
      'Office of Air Quality Planning and Standards. November 1977.

 8.  American Petroleum Institute.  Evaporation Loss from Internal
     Floating-Roof Tanks.  American Petroleum  Institute, Washington,
     D.C.  API Publication 2519, Third Edition. June 1983. p. 18.

 9.  Beychok, M.R.  "Calculate Tank  Losses Easier." Hydrocarbon Processing.
     March 1983.  p.  72.

10.   Bulk Gasoline  Terminals -  Background Information  for Proposed
      Standards.   U.S.  Environmental Protection Agency.  Research
      Triangle Park,  N.C.   Publication No. EPA-450/3-80-038a.
      December 1980.   p 6-11.


 11.   Transportation and Marketing  of Petroleum Liquids.   In:   Compilation
      of Air Pollutant Emission  Factors.  U.S.  Environmental  Protection
      Agency.   Research Triangle Park, N.C.   July  1979.


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

  13.  Reference 12, p. 3-5.
                                   2-20

-------
14.  Reference 10, p. 3-15 and 3-17.
15.  Reference 10, p. 3-15.
16.  Reference 11.
17.  Memorandum from Norton,  R.L.,  Pacific  Environmental  Services,  Inc.
     to Shedd, S.A., U.S.  Environmental  Protection  Agency.  December 20,  1983.
     Trip Report to California Air  Resources  Board.
18.  National  Petroleum News.   1983 Factbook  Issue.  Mid-June  1983
     Volume 75,  No. 7A. p.  103.
                                2-21

-------

-------
                        3.0  CONTROL TECHNOLOGY
3.1  INTRODUCTION
     This chapter describes available control techniques which can be
used to reduce emissions from the gasoline marketing network.  A large
portion of the gasoline marketing industry employs vapor control technology
which has been demonstrated and has been installed and operated at
facilities for many years.  The control strategy for storage tanks at
terminals has been to reduce emissions (i.e., by use of submerged fill
and/or floating roofs).  The control strategy for truck loading and
unloading areas at bulk terminals, bulk plants, and service stations,
has been to collect and transfer vapors back to the bulk terminal vapor
processor for treatment.  Controls for storage tanks, bulk plants, bulk
terminals, and inloading at service stations is commonly referred to as
Stage I.  Controlling emissions as a result of vehicle refueling at
service stations has been demonstrated in California and the District
of Columbia by using control technology installed at the service station
(known as Stage II controls).  Another alternative for controlling
vehicle refueling emissions, but not practiced, is a control system on
the vehicle (onboard).  Since these two vehicle refueling control
systems are a key consideration in this study, the in-use effectiveness
of these two control systems will also be reviewed.

3.2  CONTROL TECHNOLOGY FOR BULK GASOLINE TERMINALS
     Emissions from bulk terminals occur during gasoline storage and
transfer operations as shown in Figure 3-1.  This section will  limit
the discussion to the emssions from transferring gasoline to outgoing
transport trucks.  Control of emissions from gasoline storage is discussed
in Section 3.3.  Vapor controls have been used at terminals for many
years and the baseline analysis (Appendix B) estimates that approximately
70 percent of the bulk terminals had some type of controls required in
the base year of 1982.
     Emissions resulting from outgoing transfer operations 'at terminals
are controlled by two main elements, a vapor processing system (or
                                  3-1

-------
                                                5-
                                                OJ
                                                3
                                               CO
                                                a
                                               -a
                                                ea
                                                o
                                                (O
                                                CD
                                                4-
                                                O
                                                a>

                                                "CL
                                                fO
                                                X
                                                 I
                                                oo
                                                01
\
\

\

\

\

\
\
4— i/)
O -^
O E
C£. (O
| —
CD
C O)
•r- OJ
•)-> (0
(0 i-
0 0
u_ oo
                      o
                      S-
                            
-------
 vapor  processor)  in  conjunction  with  a  vapor  collection  system.  The
 vapor  collection  system  consists  of all  the piping and components
 necessary  to  safely  transfer  the  air-vapor mixture from  the  loading rack
 and  tank truck  to  a  vapor processor.  A properly designed  vapor collection
 system at  the terminal should not result in excessive backpressure at
 the  tank truck  during loading and should have no vapor leakage during
 transfer.  Check  valves  are also  commonly installed in the vapor collection
 system at  the racks  so that vapors displaced from loading at one rack
 position will not  be simply emitted at  an adjacent unused rack position.
     There are  three major types  of new  generation vapor processors
 commonly used at bulk terminals:  carbon  adsorbers, thermal oxidizers
 and  refrigeration  systems.
     The carbon adsorption vapor  recovery system uses beds of activated
 carbon  to  remove gasoline vapors  from the air-vapor mixture.  These
 units  generally consist of two vertically positioned carbon be'ds and a
 carbon  regeneration system.   During gasoline tank truck loading activity,
 one  carbon bed  is  being used  for  absorption while the other bed is being
 regenerated,  usually by an air and vacuum purge.
     Thermal oxidation units are used to control  emissions from bulk
 terminals without  recovering any  gasoline.  The gasoline vapor-air
mixture generated  from transfer operations at the loading rack can be
 piped to either a  vapor holder or directly to the oxidizer unit.  The
 vapor holder  stores the air-vapor mixture from the loading rack so that
the  system can process gasoline vapors at a relatively constant concen-
tration and flow.  Once ignition has been initiated in the thermal
oxidizer, the air-vapor mixture serves as the fuel  and the combustion
process continues until  all  of the vapors have been burned.
     Refrigeration type recovery units recover gasoline vapors from the
loading operation in the form of a liquid product.   In the refrigeration
 system, the air-vapor mixture from the loading racks  is routed to  a
condensation chamber and passed over a series of cooling coils.  Tempera-
tures in the condensation section can  be as  low as  -84°C.  The
gasoline vapors  condense, with some water vapor in  the air, and are
separated in a gasoline/water separator.
     These  three vapor processing techniques  were evaluated recently
by EPA.132   The  test data considered in  evaluating  the three  control
                                  3-3

-------
technologies represent terminals ranging in gasoline throughput from
190,000 liters per day (50,000 gal/day) to 5,700,000 liters per day
(1,500,000 gal/day).  Sixty-one tests were evaluated, totaling over
100 days of testing.  These data are considered representative of the
conditions at a wide range of terminal sizes and indicate that these
three vapor processors can operate at or below 35 mg/liter.3  An emission
factor of 35 mg/liter was used to represent the installation of new vapor
processors in subsequent emission estimation analysis.
     Several other technologies exist and have been used for many years
at terminals.  These include compression-refrigeration-absorption (CRA),
compression-refrigeration-condensation (CRC), and lean oil absorption (LOA)
These technologies were considered adequate technology to meet the CTG
requirements for bulk terminals and have been shown to reduce emissions
to 80 mg/liter.4  Therefore, 80 mg/liter was used to represent controls
equivalent to those recommended in the bulk terminal CTG.

3.3  CONTROL TECHNOLOGY FOR STORAGE TANKS
     Storage tank emissions occur from breathing losses, and from
filling, and emptying losses  (working losses).  There are three major
types of storage vessels, fixed-roof  tanks, internal floating-roof
tanks and external  floating-roof tanks.  Each tank type has its own
associated emission rate.
3.3.1  Fixed-Roof Tanks
     A fixed-roof tank is the minimum acceptable equipment currently
employed for the storage of gasoline.  Working losses (filling and
emptying losses) and breathing  losses normally incurred from the storage
of  gasoline  in  fixed-roof tanks can be reduced in the following ways:
      (1)  by the installation of an internal floating roof with rim
          seals; or
      (2)  by the installation and use of a  vapor processing system  (e.g.,
          carbon adsorption,  incineration  or refrigerated condensation)
          or a  vapor balance  system
      Fixed-roof tank emissions  are most readily controlled by  the
 installation of internal  floating roofs.
                                   3-4

-------
3.3.2  Internal Floating-Roof Tanks
     Internal floating roofs can be used directly as a control device
for existing fixed-roof tanks.  An internal floating roof, regardless
of design, reduces the area of exposed liquid surface in the tank.
Reducing the area of exposed liquid surface, in turn, decreases the
evaporative losses.  The largest emission reduction is achieved by the
presence of the floating-roof vapor barrier that precludes direct
contact between a large portion of the liquid surface and the atmosphere.
All internal floating roofs share this design benefit.  The relative
effectiveness of one internal floating-roof design over another, is a
function of how well  the floating roof can be sealed.
     From an emissions standpoint, the most basic internal floating-
roof design is the noncontact, bolted, aluminum,, internal floating roof
with a single vapor-mounted wiper seal.  The four types of losses
from this roof design are rim or seal  losses; fitting losses;-deck
seam losses; and withdrawal losses.  Rim or seal losses and fitting
losses constitute the largest percentage contribution to the total loss
from an internal floating roof tank.
     Depending on the type of roof and seal system selected, installing an
internal  floating roof in a fixed-roof tank will reduce the total  emission
by 68.5 to 97.8 percent.5  The currently available emissions test data6
suggest that the location of the seal  (i.e., vapor- or liquid-mounted)
and the presence of a secondary seal are the primary factors affecting
the effectiveness of seal systems.  A liquid-mounted primary seal  has a
lower emissions rate and thus a higher control  efficiency than a vapor-
mounted seal.  A secondary seal, be it in conjunction with a liquid- or
a vapor-mounted primary seal, provides an additional level of control.
3.3.3  External Floating-Roof Tanks
     External floating-roof tanks do not experience the fitting losses
or deck seam losses that occur with most internal floating-roof tanks.
The external floating-roof tanks are constructed almost exclusively of
welded steel.  This accounts for the absence of the deck seam losses.
Further,  because of the roof design, few if any deck penetrations are
necessary to accommodate fittings.
     Rim seal losses and withdrawal losses do occur with external
floating-roof tanks.   The only difference between external floating-
                                  3-5

-------
roof tanks and internal floating roofs is that the external  floating-
roof seal, losses are believed to be dominated by wind induced mechanisms.'
Withdrawal losses in external floating-roof tanks, as with internal
floating-roof tanks, are entirely a function of the turnover rate and
inherent tank shell characteristics.  No control measures have been
identified that are applicable to withdrawal losses from floating-roof
tanks.
     Rim seal losses from external floating-roof tanks vary depending
on the type of seal system employed.  As with internal floating-roof
rim seal systems, the  location of the seal  (i.e., vapor- or liquid-
mounted) is the most important factor affecting the effectiveness of
resilient seals for external floating-roof  tanks.  The relative effec-
tiveness of the various types of seals can  be evaluated by analyzing
the seal factors  (Ks factor  and wind velocity as shown in emission equa-
tions in Appendix B).  These seal factors were  developed on the basis
of emission tests conducted  on a pilot scale tank.8  From such an
analysis it is clear that liquid-mounted seals  are more effective than
vapor-mounted seals at reducing rim seal losses.  Metallic shoe seals,
which commonly are employed  on only external floating-roof tanks, are
more effective than vapor-mounted resilient seals but less effective
than liquid-mounted resilient seals.

3.4  CONTROL TECHNOLOGY FOR  BULK  GASOLINE PLANTS
     Control of  gasoline working  and breathing  losses resulting from
storage  and handling of gasoline  at bulk plants can  be accomplished
through  submerged fill and a vapor balance  system.   The  following
sections  will describe these two  types of control technologies.
3.4.1   Submerged  Fill
     One method  of reducing  vapors  generated during  the  loading of
gasoline  into tank  trucks and storage tanks is  by using  submerged fill.
Submerged fill  is the  introduction  of liquid gasoline into the tank
being filled with the  transfer  line outlet  being  below the liquid
surface.  Submerged filling  minimizes droplet  entrainment, evaporation,
and  turbulance.   This  is compared to  splash loading  where the transfer
line outlet is  at the  top of the  tank  (Figure  3-2a).
                                   3-6

-------
                    Vapor Emissions
                                   V
          Vapors
                            Gasoline
                             Vaporsx
                                 x-
                               X
            Fill Pipe

                 Hatch Cover
                                     Gasoline
                  Product
                                                     Tank Truck Compartment
                 A.  Top Splash Loading Method
                    Vapor Emissions
                                   v
                Product
                                    -*•
                                       co
                                       O
I
             Fill Pipe

            	 Hatch Cover
                                                     Tank Truck Compartment
                  B.  Top Submerged Method
                 Vapor Vent to
                 Recovery Equipment
                 or to Atmosphere
                                       Vapors
               Product
            C. Bottom Loading Method
                                                    Tank Truck Compartment
               Gasoline
                                                "ft"-  Fill Pipe
FIGURE 3-2.  Gasoline Tank Truck  Loading Methods
                                3-7

-------
     Submerged filling of tank trucks at outgoing loading racks can be
either by a submerged fill pipe or bottom loading.   In the top submerged
fill pipe method, the fill pipe descends to within 15 centimeters of
the bottom of the tank truck (Figure 3-2b).  In the bottom filling
method, the fixed fill pipe enters the tank truck from the bottom
(Figure 3-2c).
3.4.2  Vapor Balance System
     The vapor balance system consists of a pipeline between the vapor
spaces of the truck and the storage tanks which essentially creates a
closed system allowing the vapor spaces of the storage tank and the
truck to balance with each other.  Figure 3-3 shows the balance system
at a bulk plant.  The net effect of the system is to transfer vapor
displaced by liquid in the storage tank into the transport truck during
transfer of gasoline into the storage tank.  This prevents the compression
and expansion of vapor spaces which would otherwise occur in a filling
operation.  If a system is leak tight, very little or no air is drawn
into the system, and venting, due to compression, is also substantially
reduced.  Also vapor balancing of storage tanks and outgoing account
trucks reduces account truck filling losses and virtually eliminates
emptying losses from storage tanks (i.e., displaced vapors are returned
to the storage tank in this closed balance system).
3.4.3  Efficiency of Control Technologies
     Submerged filling of tank trucks can reduce vapor loss by 58 percent
when compared to splash loading.  This reduction was based upon the use
of an emission factor of 1440 mg/liter to represent splash loading and an
emission factor of 600 mg/liter for submerged loading.^
     The balance system has proven to be effective in bulk plant
applications for both the delivery of gasoline by transport truck
to the bulk plant and for loading account trucks.  Based upon EPA test
data, controls on bulk plant storage tanks can reduce filling losses
by greater than 95 percent, and draining and tank truck loading losses
by greater than 90 percent.10  An emission factor of 57.5 mg/liter was
used to represent the balance system control technology for tank filling
losses based upon 95 percent control  of the uncontrolled emissions
                                  3-8

-------
                                                                    CD
                                                                                CO
                                                                                (O

                                                                               SI
                                                                                OJ
                                                                                c
                                                                                O
                                                                                
                                                                                ea
                                                                               CQ
O
a
03
 >
CO

 
-------
(1150 mg/liter).11  Emission factors for storage tank draining losses
and tank truck loading losses were assumed to be 46 mg/liter and
96 mg/liter respectively, based upon 90 percent control  of the
respective uncontrolled emission factors (draining losses - 460 mg/liter,
truck loading losses (balance service) - 960 mg/liter).12,13  Maintaining
the integrity of the storage tanks, tank trucks, and associated vapor
collection systems, and ensuring that proper connections are made,  are
necessary for obtaining high efficiencies.

3.5  CONTROL TECHNOLOGY FOR TANK TRUCKS
     Just as there are several loading methods and types of rack
equipment at terminals and bulk plants to fill tank trucks with gasoline,
there are several compatible truck loading systems.  Gasoline tank
trucks are normally divided into compartments with a hatchway at the
top of each compartment.  Top loading can be accomplished by opening
the hatch cover and dispensing product directly through the hatch by
splash or submerged fill.  A top loading vapor system,  compatible with
the hatch, permits loading through the hatch while vapors are collected.
A better vaportight seal is realized when bottom loading is used.  A
1979 surveyl4 covering approximately 1,900 tank vehicles, or about 2
percent of the gasoline tank truck population, indicated that 22.8 percent
of tank trucks have only top loading, while the remaining 77.2 percent
can be either top or bottom loaded.  The trend is toward more trucks
using bottom loading, due to State vapor recovery regulations and the
advantages cited earlier.
     Tank trucks become a separate source of emissions when leakage
occurs from the truck-mounted vapor collection systems and truck
compartment dome covers.  This leakage has been estimated to be as high
as 100 percent of the vapors which should have been captured and to
average 30 percent.15
3.5.1  Description of Control Technologies
     Vapor leakage can be minimized by the requirement of all tanks to
pass an annual leak-tight test.  To meet these annual requirements,
maintenance of the vapor containing equipment such as the hatch
cover seals and pressure-vacuum vents must be conducted, and repairs
                                  3-10

-------
performed.  Figure 3-4 Illustrates the tank truck vapor collection-
equipment.  The control techniques guidelines (CTG) for gasoline tank
trucks recommends pressure limits for the annual test on the tanks and
their vapor collection equipment.^  These pressure limits represent
the way in which the P-V vent valves operate on tank trucks.  The CTG
recommends that the tank trucks pass an annual certification test which
verifies the vapor tightness of the tank.  The monitoring provisions of
the regulations recommended by the CTG permit monitoring as needed
using a portable combustible gas detector.
3.5.2  Effectiveness of Technologies
     The effectiveness of vapor control  systems at bulk terminals and
bulk plants is dependent upon the absence of leaks in the vapor-containing
equipment on the tank truck.  In EPA-sponsored tests, the average
vapor loss due to tank truck leakage was determined to be 30 percent in
areas having no tank truck vapor tightness regulations.17  In June 1978
EPA conducted a series of vapor leak tests on 27 tank trucks that were
required to undergo an annual  leak tightness test.18  Tests were
conducted on the tank trucks before any maintenance was performed to
establish the truck leakage rate since the last certification.   Evalu-
ation of this data indicated that the average leak rate for those tanks
tested prior to maintenance was approximately 10 percent, meaning that,
on the average, approximately 10 percent of the air-vapor mixture
exhausted from a regulated gasoline tank truck during product loading
would leak to the atmosphere without reaching the vapor processor.19

3.6  CONTROL TECHNOLOGY FOR TRANSFERS INTO SERVICE STATION UNDERGROUND
     STORAGE TANKS (STAGE I)
     3.6.1.  Description of Technology.   Emissions from underground
tank filling operations at service stations can be reduced by the use
of a vapor balance system (Stage I control).   In the service station
balance system, vapors which would normally be vented to the atmosphere
are routed back to the delivery truck,  which  unloads gasoline,  through
a vapor collection system.  The truck transfers the vapors to the
terminal  or bulk plant for ultimate treatment by the vapor processors
at the terminal .
                                  3-11

-------
   OVERTURN
(VAPOR RETURN)
     RAIL
                                                           RUBBER BOOT
                                                               OR
                                                            ETAL COVER
                                                                VENT
                                                               VALVE
                                                      OVERFILL SENSOR
                                                       DOME LID SEAL
                                                   BASE RING GASKET
   TANK SHELL
       FIGURE 3-4. Tank Truck Vapor Collection  Equipment
                  For Bottom Loading Operations
                             3-12

-------
     Gasoline is loaded by gravity into the underground storage tanks
via a flexible hose.  Liquid gasoline displaces a nearly equal  volume
of partially saturated gasoline vapors.  The vapor is vented through a
pipe and flexible hose connected to a vapor collection system (i.e., a
manifolded pipe) on the transport truck.  Liquid transfer creates a
slight pressure in the storage tanks and a slight vacuum in the truck
compartment.  These pressure differences effectively cause the transfer
of displaced vapor to the truck.  Because of a phenomenon known as
vapor growth caused by liquid temperature differences, the truck volume
cannot always accommodate all of the vapors.  Any excess vapor is
released through the vapor vent line shown in Figure 3-5.  This technology
has been demonstrated and installed in service stations across the
country for over 10 years.  EPA has also provided design guidance for
Stage I controls as far back as 1975.20
     3.6.2    Effectiveness of Technology.  The effectiveness of the
Stage I vapor balance system is adversely affected by leaks.  Truck
hatches should be closed and hose connections should be tight during
loading.  Tests demonstrate balance systems to be greater than 95 percent
efficient for reducing underground storage tank filling losses.21
Note that breathing and emptying losses are not controlled by this
method.  These two losses account for 5 percent of total station losses.
     In order for the vapor balance system's performance to be maintained
at design efficiency levels, the following objectives must be met:
     (a)  Assure that the vapor return line will  be connected during
tank filling,
     (b)  Assure that there are no significant leaks in the system or
tank truck which reduce vacuum in the truck or otherwise inhibit vapor
transfer,
     (c)  Assure that the vapor return line and connectors are of
sufficient size and sufficiently free of restrictions to allow transfer
of vapor to the tank truck and achieve the desired recovery, and
     (d)  Assure that gasoline is discharged below the gasoline surface
in the storage tanks.
                                  3-13

-------
MANIFOLD FOR RETURNING VAPORS
                                                VAPOR VENT LINE
              TRUCKSTQRAG
              COMPARTMENTS
               INTERLOCKING VALVE
UNDERGROUND
STORAGE TANK
    /HI l\m 11 ltt rrrrr
                                        SUBMERGED FILL PIPE =
           FIGURE 3-5.  Vapor Balance System at a Service Station
                                3-14

-------
3.7  VEHICLE REFUELING
     In addition to service station tank loading losses, vehicle
refueling operations are considered to be a major source of
emissions {see Table 2-4.) Vehicle refueling emissions are attributable
to vapor displaced from the automobile tank by dispensed gasoline and
to spillage.  The quantity of displaced vapors is dependent on gasoline
temperature, auto tank size and temperature, fuel level, gasoline RVP,
and dispensing rates.
     The two basic refueling vapor control  alternatives are:  (1) control
systems on automobiles (onboard), and (2) control systems on service
station equipment (Stage II).  Onboard controls are basically limited
to two systems - the carbon canister and the collapsible fuel holder.
This study limits its analysis to the carbon canister system.  The
collapsible fuel  holder or bladder, which attempts to eliminate the vapor
emissions source by eliminating the vapor space in the fuel tank, is a
recently developed system and will  not be considered in this document.
3.7.1  Stage 11-Vapor Control  Systems
     Loading losses due to refueling motor vehicles can be significantly
reduced by Stage II systems.  There are currently three types of Stage  II
systems in limited use in the United States: the vapor balance, the
hybrid, and the vacuum assist systems.  These systems are currently
installed on many of the stations in California and the District of Columbia,
Stage II systems have been required in parts of California since the
early 1970s.  The performance of each of these three types of Stage II
systems is discussed below.
     3.7.1.1  The Vapor Balance System.  The simplest of the three
Stage II systems is the vapor balance system.  A generalized schematic
drawing of a balance Stage II system is presented in Figure 3-6.  As
gasoline vapor in the automobile fuel  tank is displaced by the incoming
liquid gasoline,  it is prevented from escaping to the atmosphere at the
fill neck/nozzle interface by a flexible rubber "boot." This boot is fitted
over the standard nozzle and is attached to a hose similar to the liquid
hose.  The hose is connected to piping which vents to the underground tank.
An exchange is made--vapor for liquid—as the liquid displaces vapor to
the underground storage tank.  The underground storage tank assists
                                  3-15

-------
                                                                   LL.
                                                                   UJ
                                                                   a:
                                                                    3
                                                                    j_i
                                                                    en
                                                                    OJ
                                                                    a
                                                                    C
                                                                    (U

                                                                    o
                                                                    a
                                                                    
-------
this transaction by drawing in a volume of vapor equal  to the  volume  of
liquid removed.
     The effectiveness of this system is dependent on a tight  seal
between the boot and vehicle fillneck.   If a tight seal  is not maintained
the collection efficiency of this system is severly impaired.   Since  a
slight pressure is generally created at the nozzle/fillpipe interface
with balance systems, effective operation requires that a seal  be  made
at the interface during vehicle refuel ings to minimize  vapor leakage  to
the atmosphere.  The balance system nozzle is equipped  with a  spring-
loaded bellows which must be compressed before dispensing can  occur.
To assure that the bellows is sufficiently compressed,  the spout has  a
latch band which when properly hooked onto a fillpipe lip, causes  an
interlock mechanism to deactivate to allow dispensing,  and to  open the
vapor passage valve.
     3.7.1.2  The Vacuum Assist System.  The vacuum assist system
differs from the balance system in that a "blower"—a vacuum pump—is
used to provide an extra pull at the nozzle/fill neck interface
(Figure 3-7).  Assist systems can recover vapors effectively without  a
tight seal at the nozzle/fillpipe interface because only a close fit  is
necessary.  A slight vacuum is maintained at the nozzle/fill neck interface
allowing air to be drawn into the system and not allowing the  vapors  to
escape.  Because of this assist, the interface "boot" need not be  as
tight fitting as with balance systems.   Further, the vast majority of
assist nozzles do not require interlock mechanisms.  Assist systems
generally have vapor passage valves located in the vapor passage somewhere
other than in the nozzles, resulting in a nozzle which  is less bulky
and cumbersome than nozzles employed by vapor balance systems.
     Because of the vacuum, the hydrocarbon/air mixture volume drawn
into the underground storage tank is more than the tank can accommodate.
Consequently, a vacuum assist system results in some venting of excess
vapors to the ambient air from the storage tank, requiring some form  of
secondary processing such as adsorption, incineration,  or condensation.
Vacuum assist systems typically in operation use incineration  for
secondary processing.
                                  3-17

-------
                                                                 '.O
                                                                 OJ
                                                                 0)
                                                                 -p
                                                                 00
                                                                 u
                                                                 (O
                                                                 O)
                                                                 u
                                                                 O)
                                                                 Q.
                                                                 03
                                                                 OJ
                                                                 cn
                                                                 ra
                                                                 en

                                                                 uj

3-1-8

-------
     3.7.1.3  The Hybrid System.  The hybrid system borrows from the
concepts of both the balance and vacuum assist systems.  It is designed
to enhance vapor recovery at the nozzle/fill neck interface by vacuum,
while keeping the vacuum low enough so that a minimum level of excess
vapor/air is returned to the underground storage tank (Figure 3-8).
     With the hybrid system a small amount of the liquid gasoline (less
than 10 percent) pumped from the storage tank is routed (before metering)
to a restricting nozzle called an aspirator.  As the gasoline goes
through this restricting nozzle, a small vacuum is generated.  This
vacuum is used to draw vapors into the rubber boot at the interface.
Because the vacuum is so small, very little excess air, if any, is
drawn into the boot, hose and underground storage tank, and thus there
is no need for a secondary processor, such as the vacuum assist's
incinerator.
3.7.2  Onboard Vapor Control Systems
     Onboard vapor control systems consist of carbon canisters installed
on the vehicle to control  refueling emissions.   The carbon canister
system adsorbs, on activated carbon, the vapors which are displaced
from the vehicle fuel tank by the incoming gasoline.  A schematic
diagram of a carbon canister system is presented in Figure 3-9.  Onboard
control  systems have been installed and evaluated on test vehicles.
Appendix C contains an assessment of the carbon canister-type onboard
control  system.
     Such a system first adsorbs the emissions  released during refueling
and subsequently purges these vapors from the carbon to the engine
carburetor when it is operating.  This system is essentially an expansion
of the present evaporative emissions control system used in all new
cars to minimize breathing losses from the fuel  tank and to control
carburetor evaporative emissions.  However,  unlike the present system,
a refueling vapor recovery system will require  a tight seal at the
nozzle/fill neck interface during refueling operations to ensure vapors
flow into the carbon canister and are not lost  to the atmosphere.
     As  shown in Figure 3-9, the vapors displaced during fuel  tank
refueling first flow from the fuel  tank to a vapor/liquid separator
where any entrained liquid is separated from the vapor and returned to
the fuel  tank.  The vapor then flows into a canister filled with
                                  3-19

-------
                                                           CM
                                                            0)
                                                           or
                                                            O)
                                                            •o
                                                            •I—
                                                            1-
                                                            
                                                            O
                                                            O
                                                            OJ
                                                            i.
                                                            O
                                                            Q.
                                                            fO
                                                            
                                                            oo
                                                            CO
                                                             I
                                                            co

                                                            LU
3-20

-------
               ONBOARD VAPOR CONTROL  SYSTEM
       CARBURETOR
                t
          CARBON
         CANISTER
 PURGE                 VAPOR/LIQUID
CONTROL                 SEPARATOR
                  FILL  PIPE MODIFICATIONS
 TRAP DOOR
                         SEAL
                        GUIDE


              LEAD NOZZLE RESTRICTOR   .
                                                   SPOUT
FIGURE  3-9 . Onboard  Controls  for Vehicle Refueling Emissions,
                          3-21

-------
activated carbon.  The canister size required is estimated to be about
two or three times the size of the present canisters used for evaporative
emissions.  The vapor line from the fuel  tank to the canister would
need to be larger than that presently used in order to accommodate the
higher vapor flow rate during refueling.22'23
     The carbon is regenerated during the driving cycle with air drawn
through the canister to desorb the gasoline vapors.  The air/vapor
mixture is sent to the carburetor and burned in the engine.   Onboard
control systems require purge control systems which will purge
hydrocarbons without producing significant increases in exhaust or
evaporative emissions.
     One  issue concerning onboard control systems is the best technique
to assure a tight seal at the nozzle-fill pipe interface.  Three concepts
were investigated, a fillpipe modification, a combination nozzle/fill-
pipe modification, and a nozzle modification.  Any nozzle modification
would  be  simpler in design than the nozzle used for service station
control systems  (Stage II) because there is no need for a double hose or
vapor  passage  in the nozzle.  In order to simplify emissions and cost
analyses  performed in this document, the fillpipe modification approach
was employed as  all components of the onboard control system would be
located on the vehicles.
3.7.3  Effectiveness of Technologies
     The  California Air Resources Board  (ARB) performs  certification
testing on all Stage II vapor recovery systems installed in the state.
Results of this  testing indicate that all of the Stage  II vapor recovery
systems discussed are capable of achieving an emission  reduction of
95 percent.24'25
      In addition to controlling automobile refueling  losses, a well
maintained-Stage II system eliminates underground  storage tank emptying
losses since gasoline vapors, not  fresh  air, are drawn  into  the tank
when  liquid gasoline  is pumped into  the  motor vehicle.   Because a
secondary processor is  employed, the vacuum  assist system is also
effective in reducing breathing losses and Stage I  inefficiencies.
      Onboard vapor control systems  are effective at reducing refueling
 vapor emissions, but  unlike  Stage  II systems, they  provide  no control
 of underground storage  tank  breathing, emptying  or Stage I  inefficiency
                                   3-22

-------
 losses.  An efficiency of 98 percent has been reported for control
 of automobile refueling losses using onboard control  systems (see
 Appendix C, Section II).
      The discussion on onboard controls to this point has only
 considered the emission reductions derived from eliminating refueling
 losses.  However,  preliminary data from EPA's emission factors program
 indicates that in-use fuels  typically have higher volatility than the
 fuel  specified for certification testing and, therefore,  produce
 larger  amounts of  evaporative HC which cannot be  adsorbed by the
 current charcoal canisters  (see Appendix C,  Section VI).   Preliminary
 estimates of  the level  of these excess evaporative, emissions can be
 made  using the data currently available from EPA's evaporative emission
                     • • •       .    .-           .-,-•,•. £.*   rr       ...
 factors testing program which is now in progress.  This program
 involves evaporative  emission testing  using  Indplene  {certification)
 and commercial  fuel  in  carbureted and  fuel-injected vehicles. - Based
 on preliminary  data from'this program,  it is estimated that light-duty
 vehicles (LDVs) have  evaporative emissions in  the  range of  0.23  to
 0.44  g/mi  using commercial fuel  and  0.16  to  0.24  g/mi  using  certifica-
 tion  fuel, yielding  an  excess in the range of  0.07 to 0.20  g/mi.  A
 best  estimate  at this time based on  this  preliminary  data is  evaporative
 emissions  of 0.33  g/mi  using  commercial  fuel  and 0.20  g/mi  using
 certification  fuel,  for  an excess  of 0.13 g/mi.  Although data  is not
 available  for  light-duty trucks  (LDTs) , one  would  expect  results  in
 the same  ranges since LDV and LOT  evaporative  control   systems  are
 very  similar.  Since refueling only occasionally coincides with the
 occurrence of evaporative emissions, the  larger charcoal   canister
 associated with an  intergrated onboard/evaporative emission  control
 system could also control these  excess  evaporative emissions at
 little or no extra cost.
3.7.4  In-Use Effectiveness of Control Technologies
      It would be unrealistic to assume that Stage II  and onboard control
 systems would achieve, in normal use, the level of recovery efficiencies
achieved during certification tests, which are performed under idealized
conditions.  The reliability of components, routine maintenance, and
public acceptance of the systems will all affect "downtime."  An assessment
of actual (i.e., in-the-field) effectiveness is based  upon engineering
                                  3-23

-------
considerations respecting the technologies involved, in-field experience
to date with Stage II systems (California and the District of Columbia),
and upon EPA's Mobile Source Enforcement Division's past experience
with the enforcement of mobile source and mobile-source-related programs
(e.g., Fuels, Stage I Vapor Recovery).
     Potential failure modes were identified for each type of vapor
recovery system (balance, hybrid, vacuum assist and onboard cases).
These failure modes were grouped under the three general headings of
misinstallation, improper maintenance and tampering (see Appendix D for
explanation of what system defects are covered by each mode).
     For Stage II and onboard technologies, steady-state in-use
efficiencies were determijied at enforcement frequencies of quarterly,
annual, bi-annual and minimal enforcement under both State and federal
enforcement scenarios (Table 3-1).  As expected, the greater the
enforcement frequency or effort, the closer 'the i'n-use efficiency
approaches the theoretical efficiency.  "Minimal enforcement" or
"Voluntary compliance" means, in the case of Stage  II programs, the
situation resulting if no resources, State or federal, were allocated
to program enforcement.  In the case of onboard programs, the enforcement
effort would consist solely of incorporation of the onboard vapor recovery
function into the ongoing certification and in-use  testing programs.
Appendix D discusses the analysis for estimating in-use effectiveness
in more detail.
     EPA experience with contractor personnel involved in pollution
control program enforcement at the Federal level indicates that an
inspector, on average, spends about 75 percent of a man-year in the
field doing inspection work.  Enforcement officials in California's
South Coast and San Diego areas indicate that their inspectors also
spend about 75 percent of their time in the field.  The slightly higher
in-use efficiency observed for Stage  II systems under the State enforcement
scenario (shown in Table 3-1) is a result of the assumptions made
regarding the method in which violations are handled.  Direct enforcement
is considered more effective in bringing violators  into compliance than
the  notice of  violation  (N.O.V.) type enforcement.  The analysis
                                   3-24

-------
                      Table 3-1.  IN-USE EFFICIENCIES OF STAGE II
                                AND ONBOARD TECHNOLOGIES
Program
Option
Theoretical
Efficiency
Frequency of
Enforcement
    Steady-State
 Average Efficiency
of Installed Systems
                                                       State
                                                     Federal
ONBOARD

Modified
Fill pipe
    99%
                                    92%
STAGE II
Balance
Systems


Hybrid
Systems


Vacuum
Assist
Systems

Weighted3
Average
of All
Systems
95% Quarterly
Annual
Bi-Annual
Minimal
95% Quarterly
Annual
Bi-Annual
Minimal
95% Quarterly
Annual
Bi-Annual
Minimal
95% Quarterly
Annual
Bi-Annual
Minimal
91
88
81

93
90
85

93
87
81

92
88
82

89
86
77
54%
93
88
82
62%
92
84
76
55%
90
86
78
56%b
aDetermined by using an average weighted according to the population of
 Stage II systems currently installed (i.e., 80 percent balance, 15 percent
 hybrid,  5 percent vacuum assist).
bProgram efficiency reduces to 56 percent when the rate of noncompliance
 percent)  is considered.  Actual  average control  efficiency of installed
 systems is estimated at 70 percent.
                                                          (20
                                                          recovery
                                      3-25

-------
assumed all N.O.V. type enforcement on the Federal  level  and half N.O.V
and half direct enforcement on the State level  (see Appendix D).   The
efficiency of onboard systems is estimated to decrease from 98 percent
to 92 percent with the expected level  of tampering  (Appendix C, Section
III).
                                 3-26

-------
                             3.8   REFERENCES

 1.   Bulk Gasoline Terminals  -  Background  Information  for  Proposed
     Standards.   U.S.  Environmental  Protection  Agency.   Office  of Air
     Quality Planning  and Standards.   Research  Triangle  Park, N.C.
     Publication  Number EPA-450/3-80-038a.  December  1980.

 2.   Bulk Gasoline Terminals  -  Background  Information  for  Promulgated
     Standards.   U.S.  Environmental  Protection  Agency.   Office  of
     Air Quality  Planning and Standards.   Research  Triangle  Park, N.C.
     Publication  Number EPA-450/3-8Q-038b.  August  1983.

 3.   U.S. Environmental  Protection Agency.  Federal Register, Vol 45,
     Number 244,  December 17, 1980.   p.  83126-83153.

 4.   Polglase, W., W.  Kelly,  and  J.  Pratapas.   Control of  Hydrocarbons
     from Tank Truck Loading  Terminals.  U.S. Environmental  Protection Agency,
     Research Triangle Park,  N.C.  Publication  Number  EPA-450/2-77-026.
     October 1977.

 5.   Control  of Volatile Organic  Compound  Emissions from Volatile Organic
     Liquid Storage in Floating and Fixed  Roof  Tanks  - Guideline Series.
     U.S. Environmental  Protection Agency.  Office  of  Air  Quality Planning
     and Standards. Research Triangle Park,  N.C.   Draft.  August 1983,
     pg. 3-4.

 6.   VOC Emissions from Volatile  Organic Liquid Storage  Tanks-Background
     Information  for Proposed Standards.   U.S.  Environmental  Protection
     Agency,  Research  Triangle  Park,  N.C.   Publication No. EPA/450/3-81-003a
     June 1983.   Appendix C.

 7.   The American Petroleum Institute (API) Draft Document,  Evaporation
     Loss from  Internal  Floating  Roof Tanks,  API Publication 2519.
     Third Edition. 1982.

 8.   Reference 4, p. 4-18 and Appendix C.

 9.   Transportation and Marketing of Petroleum  Liquids in  Compilation  of
     Air Pollutant Emission Factors.   Third Edition.   U.S. Environmental
     Protection Agency, Research  Triangle  Park, N.C.   April  1977.   p.  4.4-7.

10.   Pacific Environmental Services,  Inc.   Compliance  Analysis  of Small
     Bulk Plants.  Report to  U.S. Environmental Protection Agency,
     Region VIII.  Denver, Colorado.   Contract  68-01-3156, Task 17.
     December 1976.

11.   Control  of Volatile Organic  Emissions from Bulk  Gasoline Plants.
     U.S. Environmental Protection Agency.  Research  Triangle Park,  N.C.
     Publication  Number EPA-450/2-77-035.   December 1977.

12.   Reference  11.
                                   3-27

-------
13.  Reference 9.

14.  Hang, J.C. and R.R. Sakaida.  Survey of Gasoline Tank Trucks
     and Rail Cars.  U.S. Environmental  Protection Agency.
     Research Triangle Park, N.C.  Publication Number EPA-450/3-79-004.
     March 1979.  p. 3-15.

15.  Reference 1, p. 3-15.

16.  Shedd, S.A. and N.D. Mclaughlin.  Control of Volatile Organic
     Compound Leaks from Gasoline Tank Trucks and Vapor Collection
     Systems.  U.S. Environmental Protection Agency.  Research Triangle
     Park, N.C.  Publication Number EPA-450/2-78-051.  December 1978.

17.  Reference 1, p. 4-2.

18.  Scott Environmental Technology.  Leak Testing of Gasoline Tank
     Trucks.  U.S. Environmental  Protection Agency.
     Research Triangle Park, N.C.  Contract No. 68-02-2813.
     August 1978 (Draft).

19.  Norton, R.L.  Evaluation of Vapor Leaks and Development of Monitoring
     Procedures for Gasoline Tank Trucks and Vapor Piping.  U.S.
     Environmental Protection Agency.  Research Triangle Park, N.C.
     Publication Number EPA-450/3-79-018.  April  1979.   94 p.

20.  Design Criteria for Stage I  Vapor Control Systems, Gasoline Service
     Stations.  U.S. Environmental  Protection Agency.  Office of Air
     Quality Standards and Planning.  November 1975.

21.  Report to the Legislature on Gasoline Vapor Recovery Systems for
     Vehicle Fueling for Service  Stations (Sacramento,  California:
     California Air Resources Board, March 1983).  p. 9.

22.  Luken, Ralph A.  Cost and Cost-Effective Study of  Onboard and
     Stage II Vapor Recovery Systems.  U.S. Environmental  Protection
     Agency.  Research Triangle Park, N.C.  August 1978.  39 p.

23.  Reineman, Martin E.  Recommendation on Feasibility for Onboard
     Refueling Loss Control.  U.S.  Environmental  Protection Agency.
     Research Triangle Park, N.C.  December 1978.  29 p.

24.  Reference 14, p. 8.

25.  Memorandum from Norton, R.L.,  Pacific Environmental Services,  Inc.
     to Shedd, S.A., Environmental  Protection Agency.  December 20, 1983.
     Trip Report to California Air Resources Board.

26.  Report of the Stage II Gasoline Vapor Recovery Study Group on  the Impact
     of the Implementation D.C. Law 3-127, Section 2:2:707(d): Phase I.
     Bureau of Air and Water Quality, Department of Environmental  Services,
     District of Columbia Government.  October 20, 1982.
                                     3-28

-------
              4.0  MODEL PLANTS AND REGULATORY STRATEGIES

    This chapter presents the model plants used in the analysis to
represent the actual facilities in the estimation of the impacts,  the
projections used to estimate gasoline consumption, facilities,  and
vehicles, and the regulatory strategies examined.  Section 4.1  presents
the model plants for bulk terminals, bulk plants, service stations and
associated facilities.  Section 4.2 outlines the projections used  for
the analysis.  Section 4.3 discusses the regulatory strategies  and
their component control options for each source category.
4.1  MODEL PLANTS
 4.1.1  Bulk Terminal Model Plants
    This section defines the methodology used to select model  plant
parameters to represent the range of new and existing bulk gasoline
terminals and provide a basis for comparison of environmental  and
economic impacts of proposed regulatory strategies.  A bulk gasoline
terminal is typically any wholesale marketing facility that receives
gasoline from refineries by pipeline, ship, or barge, stores it in
aboveground tanks, and dispenses i^t into tank trucks for delivery  to
customers.1  The data base for determination of the model plant
parameters was derived primarily from operating data on 40 terminals of
various ages.  Data presented in reports of EPA-sponsored terminal
source tests, data from plant visits, and data from information requests
submitted under authority of Section 114 of the Clean Air Act were used
as further input for the selection of model plant parameters.2
    Since terminal gasoline throughputs are distributed over a  wide
range, several model plant sizes (given on Table 4-1) were considered
in order to best represent the industry.  A recent EPA-sponsored report2
discussed the distribution of gasoline terminals by throughput  within the
industry:
     1.  Almost half of the existing terminals (48 percent) are less
         than 757,000 liters per day so a model plant size of 380,000
         liters per day was selected to represent the subset.
                                  4-1

-------
i
i
i
i
ii
it
ik
II
•i:
&s
j 1 1 p
[— l:
DC|
Ou|
3 »
•M* II
If
	 } 1
*** i
§
S* P



**.
1 H
UJ*

^_m^
""""^ If
H -
s^ **
-J g
co 1
N
. M
1 1
*^* N
LU ''

•~ n
S? ^s
£E »
il
^
H
M
»
X
'•
f" o
li j.^
;! <->
« ij-
: o
I*
»»'
t^
lui
il
O
o
o
o
0
o
T-l

0
0
o
0
o
in




o
o
o
cT
LO
CM






0
o
c_^
o
o









-^
03
"O
f
cu
CL

c
o
t £1
S ^
i .
* J_s
Throughpul
«'

O *d-
o
0
o
o
CO
CO

O CO
0
o
o"
g
n




O CO
o
o
o
LO
o






O CM
o
<_^
o
oo
co









^
•ra to
c
t. o
(U •!""
CL 4-*
»f—
t/1 t/>
S- O
 g ^
jQ O
oo a.
o
~
o
T3 -P
CU -P
cn o
S- CQ
CT> 3

^•^


•a 4->
O> -P
O) O
^O t- CQ
CD
1°
to CL
O
u—
i^











in
j-

cn
•5 -a
o -g
CU
*[ ^^
o
i_ =
CU •!-
.a T3
£E 'w
3 O
:z — J

T3 CU
i— D.
CD
O
r- o
CM O
CM —
0.1=
r— O.
cn
o
r-- o
CM O
CM -— •




HH
r— Q.
O
P**" ^D
CM O
•» ^o
CM ^—





E <•—
O.E
i— CL
CT1
0
r^ o
CM O

CM •—










£=
S_

cn
•1—
T3
03
O
_J
CO
03
C£
cn
Q.
E
a.

'•±-
&. e:
cu o
*~ cn
0
0 0
CM O
"LO
CM «
CO CO
to to
cu o
•p r—
•B— »—
r~ cn
o
0 0
CM O
CM "
CO OO
S- C
cu o
•p r—

r^ "tO
• cn
o
0 0
CM O
"LO
CM "
CO CO
*•"•*
to to
S- C
OJ o

•r- r—
t— to
cn
O
0 0
CM O
OJ "
CO CO
— •














.JJ
o
to
Q.
o
o
3
I—
c
1—

cn
to
as
•P
3
•r*
E
O
CM
O)
•P
3

'E
o
CM

to
CU
•P
3
•i—
E
O
CM




to
CU
•p
3
'i
O
CM













CU
E
£
cn
c
•o
03
O

CJ
3
S-
1—
c:
03

^
E E
CL CL
r- cn
0 0
0 0
CO CM
n *\
CM —•
s"s
CL CL
i— cn
0 0
0 0
•3- «*
O LO


*f~*
E E
CL CL
r— CD
0 0
0 0
«\ «
O LO
CM —


«•»•••»
E E
Cu CL
i— cn
0 0
0 0
CO O
fO co
I— 1 •—»












to
3
o
cu
c.
03
•P
C
03
to cn
1 	 1 •!—
-o
E 03
3 O
X
03
s:

•r-
5_
03
CU
to
-o
O
CO
cu
to
•a
o
co
i-
(O
cu
>^
to
•a
o
^st°
CO

S-
03
CU
>^
*^^.
to
•o
0
^"
CO














cu
1—
3
-o
cu
3=.
CJ
CO
en
c:
•p
03
O)
CL
O

•i-2
r— °
C*O jD
O
O O
0 0^
"O
LO O
co _o
o
o o
r- o
CO **
co r~-
«* CM
P
S
co _a
E
O
o o
oo o
CO "
"O
CO LO
CM.t— H
^— ••


.^•^
r—
co .a
E J2

CD O
«* o
CO O
« n
O LO
r-< CO
^^^*











•P^
•p-
CJ
(O
a.
03
o
cu
cn
to
i.
o
oo
a>
c
"o
to
03
CJ3

^
U3







^









co









cu
c
•r-
*O
to
03
CIS

0
M-
to
-^
03
(—
4-
0
S-
cu
1
z

,_
o
CM

CT)
-




-









co






to

CJ

i-
H-
TD
O)
C
3
O
r~~
03
C
•i^
E
cu
1—
tt-
o
S-
cu
3
Z

E
"*

r— 1
CM





1 —
CM









CO










to
cu
•r~
-P
•r—
1^
o
03
U_
1o
O
1—
4-
Percent o

c
o
CO

LO
t— <
CO





LO
o









0
CM







r—
cu

o to
2H t—
s_ e:
a> T-
CL E
to o
CU -P
•f^«
r^ -P
•t- C3
CJ -P
03
U- 0
O
4- LO
O r-H
cu -P
.a £=
E o3
3 r—
Z 0-

o
LO
t— 1
*
o





CO
CM









CO
i— 1





{i 	 ;! ^ii 	 i '



r_
to
•P^
(E

4-
O
•P
d
CU
o
S-
cu
a.
O)
•P -P
Approxima
Throughpu

CL
1
1
|
1
1!
H
i
\\

1
1
i
I
1
|l
1
t
||
II
li
ii









il
1!
11
1




li
• H
n
ii
i



1!
i
1

i
ii
i
i
i


4-2

-------
      2.   Approximately  27  percent  of  the  gasoline  terminals  have  a
          throughput  between 757,000 and 1,514,000  liters/day.  A  model
          plant with  a throughput of 950,000 liters/day was selected to
          represent these plants.
      3.   An  additional  21  percent  of  the  terminals have a throughput
          between 1,514,000 and 2,270,000  liters/day.  A model plant size
          of  1,900,000 liters/day was  selected to represent these  facilities,
      4.   A model plant  size of 3,800,000  liters/day was selected  to
          represent the  4 percent of existing terminals greater than
          2,270,000 liters/day
4.1.2  Storage Tank Model Plant
    As disussed in Section 2.2.2,  a typical terminal has four or  five
above-ground storage tanks for gasoline,  each with a capacity ranging
from  1,500 to 15,000 m3 (9,400 to 94,000  barrels).  Most tanks in
gasoline  service have a floating roof to  prevent the loss of product
from  breathing and working losses.   The fixed-roof tank is the
least expensive to construct and is generally considered as the minimum
acceptable tank for the storage of petroleum products.  Emissions from
existing  fixed-roof tank are most readily controlled by the installation
of an internal  floating roof.   A set of model  plant parameters were
developed to describe the physical  characteristics of a typical  fixed-roof
tank at a bulk terminal.  This typical storage tank has a volume of
2,680 m3  (16,750 bbl), based on available EPA data on fixed-roof tanks
at terminals.  A diameter of 15.2 meters  (50 feet), and a height of
14.6 meters (48 feet), were assumed as typical  values for a tank of this
capacity.3 These parameters were then used in emission equations to
estimate  emission rates from fixed-roof tanks (see Appendix B)  and to
calculate capital  costs in Section  7.2.2.
4.1.3  Bulk Plant Model  Plants
    As described in Section 2.2.3,  bulk gasoline plants are secondary
distribution  facilities within the  gasoline marketing network.  The
typical  bulk  plant facilities  include tanks for storage of gasoline;
loading racks;  and incoming and outgoing tank  trucks.   Regardless of
throughput,  typical  bulk plants have the same  numbers of tanks,
loading racks,  and account trucks,  as  follows.4  The typical  bulk
                                  4-3

-------
plant  utilizes  two  relatively  small  above-ground storage tanks ranging
in capacity between 50,000  to  75,000  liters for gasoline storage.
Usually a  plant will  have one  loading  rack using top filling by either
a top-splash method or a top-submerged fill pipe.  Transport trucks
supply bulk plants with gasoline while account trucks deliver gasoline
to bulk plant customers.  Bulk  plants  typically average two account
trucks each.
    Bulk plant  operations are  based upon a large number of annual tank
turnovers, with the result  that most facilities tend to be small.  An
EPA-sponsored report  4>5 discusses the distribution of bulk gasoline
plants by  throughput  within the industry.  Over 90 percent of all bulk
plants have an  average daily gasoline throughput that is less than
30,280 liters per day (8,000 gal/day).  Of these almost half (42 percent)
are less than 15,140  liters per day  (4,000 gal/day) so a model plant
size of 11,350  liters per day  (3,000 gal/day) was selected to-represent
this subset.  Another 50 percent of the bulk plants have a throughput
between 15,140  and 30,280 liters per day (4,000 and 8,000 gal/day).  A
model plant with a throughput of 24,600 liters per day (6,500 gal/day)
was selected to represent these plants.  Only 7 percent of bulk plants
have a gasoline throughput  between 30,280 and 64,350 liters per day
(8,000 and 17,000 gal/day).  A model plant with a throughput of 47,300
liters per day  (12,500 gal/day) was selected to represent these plants.
An additional 1 percent of  plants have throughputs between 64,350 and
75,700 liters per day (17,000 and 20,000 gal/day).  A model plant size
of 64,350 liters per day (17,000 gal/day) was selected "to represent
these facilities.  A summary of the bulk plant model  plants is shown in
Table 4-2.
4.1.4  For-Hire Tank Truck  Population
    The trucking industry generally consists of two major groups, for-
hire and private.  Private  carriers are firms which transport their own
goods in their own trucks.  Examples of private carriers  are the  oil
companies which use their own tank trucks to deliver gasoline from
their terminals or bulk plants.  For-hire carriers transport freight
which belongs to others,  renting out the hauling  services of their
trucks.
                                  4-4

-------
                  41 4->
                  4J O 4J


                  IC1
                  x o  o>


                  ^1
                     41
                     a.
OO


UJ
C£

a.
•=C


Q-
o
CQ
CM
  I
 CQ
 
                     o
                     O  tn
                       *>
                     +•»  C
                     c  a
                     41i—
                     u o.

                     41
                     a.
                     IS

                     a.
                     41

                     1
                     •o
                      4)
41 0)
C in
T- 4)

'o o. >>
in 41 IB
 O>f—
                   C 3

                   IQ O
                               CO
                               CM
                   OO
                   in
              o
              o
              ro
                                     §
                                      O
                                      CO
                                      CVJ
                                           g   8
                                »T   O    ^
                                I-H   ro    10
                                  -    i     i
                                in   o    O
                                ^H   *f    oo
                                V   i-<    CM

                                     in    o
                         5-
                   41 3   _
                   C O. 01
                          in
                                                               IO    — 3
                                                              r-  4) "O O
                                                               O. C 41
                                             41 O 41
                                            *O t/1 g O)


                                             g O) 3 i-
                                                  «> t.
                                            4J -O    41
                                             C = V. >

                                             UJ 3 O -^
                                            r- O
                                            .^  41 fO "

                                            "3  o "o. <
                                            4J  41 O +J


                                             en r— O) C
                                             41    =3
                                             N O i- O
                                            •>- o 
-------
       It is estimated that 85,000 tank trucks are used for the delivery
  of gasoline.6 About 31 percent of the gasoline tank trucks or 26,300
  vehicles are used at bulk terminals.  This total includes only tank
  trucks of greater than 15,100 liter (4,000 gallon)  capacity in order to
  avoid the inclusion of small tank trucks operating from bulk plants.
  The remainder, or 58,700 vehicles, are smaller tank trucks used pri-
  marily to transport gasoline from bulk plants.
       For purposes of determining the number of uncontrolled tank trucks
  at terminals it was assumed that if a terminal is uncontrolled, the tank
  trucks loading at these terminals are also uncontrolled.  Since it was
  estimated that 33 percent of terminals are uncontrolled, then 33
  percent of 26,300 trucks results in 8,700 uncontrolled tank trucks.
  Given that 500 terminals would be regulated, then the number of
  uncontrolled, terminal-owned tank trucks was calculated using the
  method shown in Table 4-3.  Of the total 8,700 uncontrolled tank trucks
  at terminals, it was estimated that 2,900 tank trucks are terminal-owned
  and the remaining 5,800 tank trucks are owned by for-hire tank truck
  companies.
                Table 4-3.  METHOD OF CALCULATING THE NUMBER OF
                      UNCONTROLLED TERMINAL-OWNED TRUCKS
No. of Affected No. of Trucks owned
Terminals Percent of Terminals in ,by terminals in
Model Plant 1 thru 4a Model Plant 1 thru 4a
500 x 48
500 x 27
500 x 21
500 x 4
Total uncontrol 1 ed
x 3
x 6
x 9
x 20
terminal -owned tank trucks =
or =
Total
Trucks
720
810
945
400
2,875
2,900
aSee Table 4-1 (from Reference 2).
                                    4-6

-------
     At bulk plants It was estimated that 45 percent, or 26,400, of the
58,700 tank trucks will be controlled by a vapor balance system.  [The
45 percent was obtained by dividing the number of controlled bulk plants
(7,000) estimated at baseline (Table 4-7) by the total  number of bulk
plants (15,000)].  Thus, the number of tank trucks without provisions
for vapor balance was estimated as 32,300 vehicles.   About 8,000 bulk
plants were assumed to be uncontrolled (see Tables 4-7 and 4-10).  The
typical number of tank trucks owned by a bulk plant (determined from an
earlier study6) is two vehicles.  Thus about 16,000 tank trucks (the
product of 8,000 x 2 account trucks) are owned by bulk plants and the
remaining 16,300 tank trucks are owned by for-hire tank truck companies.
     All  cost estimates for the for-hire tank trucks in Chapter 7.0
were calculated based on these vehicle population figures rather than
on a model plant approach.  The costs for the private,  facility-owned
trucks are already included in the cost estimates for bulk plants and
termi nal s.
4.1.5  Service Station Model Plants
     Service stations, as defined in this document,  include all motor
vehicle refueling operations that receive revenue from sales of gasoline
(public retail  outlets) and that service governmental,  commercial, and
industrial fleet operations (private outlets).  For this report the
category  does not include agricultural  outlets.  Miscellaneous retail
outlets that were considered service stations for this  study include
convenience stores, mass merchandisers, marinas, parking garages and
others which obtain less than 50 percent of revenue from gasoline
sales.  In contrast, the U.S. Census Bureau counts as service stations
only those outlets that do 50 percent or more of their dollar business
in petroleum products.
    In the development of a  representative estimate of the total
service station population in the United States, Census Bureau data and
the Lundberg Survey served as primary references.  An accurate count is
difficult because convenience stores (C-stores), and others who sell
motor fuel do not necessarily fit the "service station" definition used
by the Census Bureau.
    The Lundberg Survey, based on refiner and marketer reports as
well as grass roots data, was discussed in the September 1983 issue of
National  Petroleum News7 (NPN) for 1977, 1980 and 1982.  The Lundberg
                                  4-7

-------
1977 total is 263,348 - 50 percent more than the Census Bureau figure
of 176,465.7  For 1982, the Lundberg total is 210,875, compared to the
Census Bureau's 144,690.10 Based on the Census Bureau definition of
service station and Lundberg's emphasis on recording all public outlets
which sell gasoline, an approximate ratio of the two figures, or 2/3 was
determined as the fraction of total public outlets thought to be defined
as marketers of gasoline that would be counted as service stations by
the Census Bureau.  This fraction was used in the percentage determination
of outlets estimated to be independent marketers of gasoline under
Section 324 of the Clean Air Act (see discussion on page 4-12).
    Estimates of 1982 service station population are presented in
Table 4-4.  The number of public outlets was based on the Lundberg
estimate.  Surveys confirm that the spread of C-store/gasoline combina-
tions will continue at the expense of the traditional service station
population9 and the Lundberg estimate more accurately reflects- the
number of C-stores, mass merchandisers and others which obtain less
than 50 percent of revenue from gasoline sales.7  In addition to
"public" outlets, there are a significant number of "private" facilities.
These outlets are maintained by governmental, commercial, and industrial
consumers for their own fleet operations.  Government agencies with
central  garages are typically regional  locations for the postal service,
Federal  government agencies, and state and county agencies.   Other
miscellaneous facilities include utility companies, taxi fleets,  rental
car fleets, school buses and corporate fleets.   The agricultural  sector
of private outlets, including farms, nurseries and landscaping firms,
was not considered.  In general, agricultural outlets would have
throughputs of less than 37,850 liters per month (10,000 gal/mo)  and,
collectively, a gasoline throughput representing approximately 3  percent
of the nationwide total.8
    Service station model  plant categories and the approximate distribution
of facilities within each model  plant category  were derived from  size
ranges used by the Bureau of Census, total  facilities reported for
19777 and 198210 and the total  consumption of gasoline (excluding
agricultural)  for each year.11  The total  population of retail  outlets
in the U.S.  continues to decline,  and as many  of these stations  disappear
they are replaced by units which pump more gasoline.   The number  of
                                  4-8

-------
     Table 4-4.   ESTIMATES OF 1982 SERVICE  STATION  POPULATION3
Public Outlets (i.e,  service stations  and
                 convenience stores)
"Private" Outlets0
     Government (Federal,  military,  State,  local)
     Miscellaneous (auto rental,  utilities, others)
     Trucking and Local  Service
     Taxi s
     School Buses

           Total
210,875b

 85,450
 94,530
 21,900
  5,380
  3,070

421,125
 Not including the about 2.5 million agricultural  outlets.
 Source:  Lundberg Estimates.  National  Petroleum News,
 September 1983  (Reference 7).
'Source:  Arthur D. Little, inc.  The Economic Impact of Vapor
 Recovery Regulations on the Service Station Industry^
 (Reference 8).
                                       4-9

-------
service station closures which occurred from 1977 to 1982 was assumed
to be distributed evenly across all model plant sizes.   Service station
openings were assumed to occur primarily within the two largest model
plant sizes.  Data on the number of facilities and gasoline consumption
reported for 1977 and 1982 established an average annual  throughput per
station for those years.  As expected, the average annual throughput
for all stations increased during this five-year period.   Therefore,
the Bureau of Census size distribution available for 1977 was adjusted
to reflect the increase in average annual throughput.
    The five model plants and the estimated size distribution of model
plants within each of the five categories are presented in Table 4-5.
The number of nozzles per station representative of each  category, as
well as corresponding average throughput values and throughput ranges,
were determined.  As a check, the total number of facilities per category
were multiplied by the average monthly throughput value determined as
representative of each category and then by 12 months per year.  The
total nationwide annual throughput calculated in this manner was within
3 percent of the throughput figure used as baseline.
    Model plants 1 and 2 were originally one category,  however, a fifth
model plant representing a throughput range of 0-37,850 1/mo (0-10,000
gal/mo.) was included (based on a graph of the size distribution for
the other four model plants) in order that the effect of  a 37,850 1/mo
(10,000 gal/mo.) exemption could be determined.  Public and private
facilities are distributed among the five model plant categories of
Table 4-5.  The distribution of "public" facilities across the first
three model plant sizes is approximately uniform (i.e., 25 to 30 per-
cent) declining to 14 percent and 3.5 percent of facilities for model
plants 4 and 5, respectively.  Based on information from  Arthur D.
Little, Inc., and the U.S. Census Bureau, it was estimated that
approximately 90 percent of "private" outlets have throughputs of less
than 37,850 1/mo (10,000 gal/mo.).12»13  The remaining  10 percent of
private facilities were distributed among model plants  2  through 5 in
proportions representative of the public service station  distribution.
Additional information within Table 4-5 includes percent  of total
                                  4-10

-------






o
1— H
i |
1 "'
03
2
I—
OO
3
LU
£!
00

•^
*-—
j
tA
19
0




LO


1
1

,j.



O i
1— H 1
1— 1
<
i— '
j
LU II
CJ 1
> 1
LU
CM
03
cn
Q
LU
g
•™^
1 —
OO
LU


•
LT>
|

"*
LU
— 1
CO

1—








CM



r- (













m
O
z
c
CO
a.

•o
o


i



O LO
o cc
^•*



LO LO
«»• LO
CM ^.*





CM LO
cn cn
^H*-»


LO O
r~ ""*
g?
CO
CC
•W
2
a.
f
cn
s.
.c
1—
1


1



VO
1
CM
•— 1


CM
i— 1
1
cn






o>
i



to
i



CM













CU
Is4
M
O
z
g°
cn o
o"2
f— 4
CM


«s- cn
*
^

§2
LO
Ol
CM




O LO
O •
cn -o
"CM
LO
LO


0 0
o cn
cn
en


O CM
CO

Id

o'2
«— t
CM


g^o
,— i o
t— 4

0 °*
O CM
**•





^D Lrt
^* CO
^


0 -H
o •
to «^
co"


0 0
^J Ol
A
f" cn
LO


CO
»— 1

1
1




c
0
•f—
CO
oo
CU
o

2 1
CU •>-> +J
CO CO C
tw
O

S.
CU
= "3 u


V)
c
o
13
(13

U
> c
S- 0
CU 1-
CO •*-* +J

= f— 0»
CU 3 0
CJ O. S- -J-> 0- e_
••- 0 CU
r- a. a.
-9 -9
I -^
IQ O CU
> 0. Q. ]
a.

g °
CM O
"O
t— f T-*
CM



0 CM
LO
CO

O CO
o
LO
cn
cn




O LO
cn
cn


§i —
»-*
cn
A
r-H


O CO
O LO
O
^f.
^p
CM







^
O
O
rH



r-4
CO
f^



LO






**:
if*
CM


.
•*
i

CM
2










! §

a.
1 U) i
C 1
"J i o
• c 1 ^
i -2 ^ ! -g
•?-  o
(J r— CJ
Ll_ Q. &_ '

2 °" °~
h-
o. i
C- Ij j
O) o 1
o »*
h-
4-n

-------
facilities (i.e., public and private) per model plant and the percent
of total gasoline throughput generated per model plant.  Several
proposed service station Stage II regulatory strategies include options
with and without site exemptions or cutoffs (the terms size exemption
and size cutoff are used interchangeably in this document).  Examined
size exemptions include any service stations generating less than
37,850 liters/month (10,000 gal/mo) throughput (based on existing State
and local regulations and expected relative economic impacts)  and
independent dealers below 189,300 liter/month (50,000 gal/mo.)  sales
volume (in accordance with Section 324 of the Clean Air Act and assuming
that Section 324 applies to Section 112 regulations).  These stations,
for which size exemptions are examined, represent the segment of the
retail industry that may be most vulnerable to changes in marketing
economics as well as external costs such as vapor recovery costs.
     Service stations were classified into several operational- groups^
and those stations which, by definition, qualified as independents were
grouped into a total percentage by model plant.  This percentage was
then reduced by one-third as only two-thirds of the total  "public"
outlets are estimated to fall within the definition of marketers of
gasoline, according to the Census Bureau and Clean Air Act definitions
(see discussion on page 4-8).  Independents as a percentage of total
"public" facilities were estimated at 18 percent, 31 percent, and 45
percent for model plants 1 through 3.  It was not necessary to calculate
the percentage of independents within model plants 4 and 5, as the
proposed size exemption applies only to facilities of less than 189,300
liters/month (50,000 gal/mo) throughput.  These percentages were used
together with the number of public facilities and the annual throughput
representative of each model plant to determine the throughput attribu-
table to independents and thus, excluded from the impacts analysis.
Independents with throughputs less than 189,300 liters/month (50,000
gal/mo) were estimated to pump approximately 15 percent of the total
gasoline throughput (10 percent in model plant 3, 5 percent in model
plant 2 and less than one percent in model plant 1).  If the size
exemption for all facilities less than 37,850 liters/month (10,000
gal/mo), representing 14.2 percent of total consumption (Table 4-5),
                                  4-12

-------
 were considered  together with the size exemption for independents,
 total throughput excluded from regulation would represent approximately
 29 percent of nationwide consumption.
 4.2 GASOLINE, FACILITY, AND VEHICLE PROJECTIONS
      Because the comparison of regulatory strategies, which will be
 discussed in Section 4.3, requires the evaluation of emissions,
 cost, and risk impacts far into the future, projections had to'be made
 for gasoline throughput, gasoline marketing facilities, and automobile
 fleet characteristics.  This analysis assumed that onboard controls
 would be installed only on new light-duty vehicles (passenger cars)  and
 light-duty trucks (<8500 Ibs.  i.e.,  vans, pickup trucks,  etc.).  Thus,
 the effectiveness of onboard controls increases with time.   For this
 reason,  the analysis was extended through the year 2020 so  that
 essentially the  entire light-duty vehicle fleet would be  controlled.
 In this  way,  onboard could be  compared with Stage  II in both the short
 and long term.   Therefore,  to  fully  analyze the regulatory  strategies,
 estimates for the base year  of 1982  and projections  to  the year 2020  '
 were  performed.
 4.2.1  Gasoline  Consumption
     Total  consumption of gasoline in  the base year  of  1982  was
 389 billion liters  (102.7 billion gallons).^   F1gure 4.! presents
 estimates of  how  this  consumption or throughput was  distributed throughout
 the gasoline marketing  chain.
     Several  sources were contacted to  determine the extent  of  data
 available on  gasoline consumption projections.  An EPA Federal  Register
 notice (47 FR 49329)16  dealing with phasing down the lead contemTi^
 gasoline contained projections of total gasoline consumption and the
 decrease in leaded gasoline usage through the year 1990.  The Department
 of Energy was contacted and both short-term (through 1985) and mid-term
 (through 1990) projections were obtained.17  However, no long-term
 projections were available. In Figure 4-2, the available consumption
 projections were plotted.  In addition, actual consumption for
1976-1982 was plotted from data available  in several  issues of the
National  Petroleum News Factbook.H   The original  analyses were to be
                                  4-13

-------
                      c
                      o
O
UJ
o
00
   00
   >-
   00
S2
         O)
O

CO
c
o
_J 00
u. *-«    c
   O   T-
O
UJ UJ   •—•
»— -z.   o-
»— oo
OO rf
UJ c3
CO

o
 * 00   i—
UJ
                                         4-14

-------
                    Footnotes  for  Gasoline  Distribution  System
                                    Figure  4-1
 1982  estimates.

 Obtained from  National  Emissions  Data  System  (NEDS),  1982  calendar year.
•»
'A number of  bulk  plants in  Southeastern  Alaska  and  some  islands  off
 Puget Sound  receive  gasoline  reportedly  by  barge  from marine  terminals.
 At this  time,  data have not been  received to  support  estimates of
 either facility population  or total  throughput  for  these bulk plants.

 Weighted average  based  on 1977 Bureau  of Census data  was determined  to
 be 25 percent  to  bulk plants, 75  percent to service stations.
a
"Based on U.S.  Department of Agriculture  data, Arthur  D. Little estimates
 that  3 percent of the total U.S.  gasoline volume  is consumed  by  the
 agricultural sector  (Reference 8).
                                       4-15

-------
                                                                                    'f—s
             u

             5
                                             p


                                             O)
                    !J
                    C
          o  — J
                                                                                           o

                                                                                           a.
                                                                                                      +J
                                                                                                      Q.
                                                                                           O
                                                                                           o

                                                                                           OJ
                                                                                           c:
                                                                                                      o
                                                                                                      CO
                                                                                                      fd
                                                                                                     CD
                                                                                                     C\J

                                                                                                     <3-
                                                                                                      O)
                                                                                                      CD
                                                                                       -s
en
 o
                              O
                              CO
                                                           O
                                                           OJ
           I     I
                                                                i     I
                                                I     f     I
CM  O
 o  in
o
CM
•s-
o   o
en   -3
to   n
o

CM
o
<3-
CVJ
                             uoLq.duinsuoo
                                         4-16

-------
based  on  projections  to  the year 2000,  so a graphical estimation
based  on  EPA  projections was made.  The EPA projections were used since
they were in  the mid-range of DOE projections.   It  is recognized that
there  is  a degree of  uncertainty within the assumptions and estimations
employed  to project gasoline consumption far into the future.  For this
reason and because no additional information was available,  projected
consumption of gasoline was assumed to be constant  from the year 2000
to the year 2020.  The graphical extrapolation that was used as a best
estimate  is depicted  in Figure 4-2 as a dashed line.
     A similar approach was used to project leaded gasoline consumption.
This was  necessary to estimate projected EDB and EDC emissions.  The
rapid  decrease in leaded gasoline corresponds,to. the phase out of
leaded gasoline production.  Again, projections were assumed constant
from the year 2000 to 2020.  The estimated leaded gasoline con-sumption
is also shown on Figure 4-2 as a dashed line.
     After these projections were developed and used as the basis for
the analysis, three other projections were obtained and evaluated,
one by the Department of Energy (DOE), one by the American Petroleum
Institute (API), and one in the Oil & Gas Journal.  Table 4-6 compares
the three projections at selected years.  Because of the uncertainties
inherent  in all such projections, the original  assumptions and
extrapolations were retained.
4.2.2  Gasoline Marketing Facilities
     Data on the number of gasoline marketing facilities are necessary
since nationwide cost information is generated from costs per facility.
Several information sources were used to estimate the number of facilities
in the base year of 1982.  Table 4-7 summarizes the number of facilities
in each industry sector.   To determine the number of facilities in
controlled areas at baseline, the throughputs for all  the controlled
areas were summed from the tables in Appendix B and the percentage of
total  nationwide consumption was determined.   This percentage was then
applied to the total  number of facilities.   For example, 67 percent of
the throughput for terminals was determined to be in areas that were
already controlled.   Applying this percentage to the total  estimated
number of terminals  (1,500)  resulted in 1,000 controlled terminals.
                                  4-17

-------
TABLE 4-6. ALTERNATE GASOLINE
CONSUMPTION PROJECTIONS (billion liters [gallons])
:
>
:
I
I
«
e
R
!
i
i
C
•
fc
r
«
Ik
%
»
t>
f
t
I
*
•
p
(
i
i
*
,;,
V
f
f'
I"
»'
V
»>
»
J
y
*
»
f
*>
#
r
Tnt»i (Unleaded and Leaded) Gasoline . ,. _. 	 , 	 	 ,__ 	 Leaded Gasoline 	
11
>> c
S-2
.= **
si
VI
JO
11
o. «
eu
*t
•oto
IA!C
vt •*-•
•C
|
St.
CD */* ^™
«^
a^
IS
01 CJ
C± VI
o  co r^ to m


t


§^MCT\r— mroeivi«-**-*oooooo ,
i 	 	
to o" I •
• • I
Ch O i
co co
2 ., - 1 , .
R S i ^
Is
I^.H.^.it. U- >*• U— **- l^tfc-»*-U- ^H
1—1 *~1 r^ *"""' f~1 *~1 ^ " ^ ^ ^! " ^ ^ ^ 'o^
r^T i o" icuimt^-i^Ti«— '^ w °? Ci^ « .^^ . ^— • — • — • «-» • ••— • • ^ • ^— • — • • S ^
OJ Cft
t a: c^
J trt OJ
*-4 r*» cn «-H ^-« in O
^ U3  r* o ***
co co co r»* i^. £
.is-sss si
j « n rt tn CM >, -
1 	 	 	 	 : 	 , 	 ", 	 : 	 : S B
j r-i 1 -
?"eou>~i-.e3C3e9oe30CSC3Cse; "^S1
:!«"«o^»^o"^^-^^o s1!
JIPJ^^SSSSSSSSSS^CM • 3 -g
I , 	 -D U_
3

CFrom Personal Communication with Debra Paxton, Department of Energy (DOE), September 30, 1983 (Reference 17).
dFrom "Motor-gasoline forecasting: an embryo science," OJJ «_Gas_ Jour_nal_, November 14, 1983 (Reference 18).
6From Department of Energy (DOE/PE/70045-1), Appendix B, February 1983 (Reference 19).
                                                  !£   S
                                               *«—   O
                                               s-
                                               •§
4-18

-------
                    Table 4-7.   ESTIMATED NUMBER OF  FACILITIES
                                  IN THE BASE  YEAR 1982
Facility Type
Bulk Terminals
Bulk Plants
Service Stations
Total
l,500a
15,000a
421,100b
Controlled At
Baseline
1,000
7,000

Uncontrolled
Baseline
500
8,000

At



    - Underground                              223,200                197,900
       storage tanks
    - Automobile                                37,900                383,200
       refueli ng
Storage Tanks                  5S470C             4,840d                  630e
For-Hire Tank Trucks
    - Terminals                                                       5,800^
    - Bulk plants                                                     16,300f
 Based on 1983 NPN Factbook estimates  (Reference 15).
b
 See Table 4-4.
c
 References 22 and 23.
 Controlled by floating roof with primary or secondary  seals  (see Appendix  B).
e
 Fixed-roof tanks.
 Independent trucks only.   Other uncontrolled trucks at bulk  terminals (2,870)
 and bulk plants (16,000)  included in  bulk terminal  and bulk  plant costs.
                                       4-19

-------
The remainder (500 terminals) were assumed to be uncontrolled (i.e. no
vapor processors).
     Attempts were made to project the number of gasoline marketing
facilities into the future.  Several references have documented the
fact that the bulk plant and service station populations are declining.11
However, no quantifiable data were available to project facility popul-
ation into the future.  For this reason, no projections for facilities
were attempted.  Instead, for costing purposes, the number of facilities
was kept constant through the year 2020 while the throughputs and cor-
responding recovery credits were decreased proportional to the decrease
in gasoline consumption.  This probably biased the cost estimates for
service stations and bulk plants to the high side, but the degree of
the bias is unknown.
4.2.3  Light Duty Vehicles and Light Duty Trucks
     A projection of the number of new light duty vehicles (LDV) and
light duty gasoline trucks (LDGT) was necessary because one of the
control technologies evaluated for automobile refueling would require
control equipment installed on all  new vehicles after a certain model
year.  Estimating costs for this approach required knowledge of the
number of controlled vehicles in each year.  Because of the length of
the projections, the analysis had to take into account the scrappage or
retirement rates for controlled vehicles as they age.  With the projec-
tions of new vehicles and the projections of retirement or scrappage
rates, the number of controlled vehicles in each year up to 2020 could
be estimated.
     The projected number of new LDVs and LDGTs, retirement or scrappage
rates, fuel economies and mileage accrual  rates are all provided within
Appendix C.  Given the fuel economy and mileage accrual rates, the con-
sumption of gasoline by controlled vehicles in each year could be
calculated (see Appendix C, Section VIII).  The gasoline consumption by
controlled vehicles (with and without tampering)  is presented in Table
4-8.  Emission reduction estimates attributable to onboard controls were
based upon the percent of the total  gasoline consumed by controlled
vehicles.  Figure 4-3 presents graphically the gasoline consumed by
                                  4-20

-------
                    Table  4-8.   ONBOARD CONSUMPTION PROJECTIONS
Year   I
Gasoline Consumption
by Controlled Vehicles
   (No  Tampering)
   (109 liters)
  % Consumption
by Controlled Vehicles
  (No Tampering)
Gasoline Consumption
by Controlled Vehicles.
   (With Tampering)
   -  (1Q9 liters)      I
   %  Consumption
by Controlled Vehicles
  (With Tampering)
1988
89
90
91
92
93
94
95
96
97
98
99
2000
2001
2002
2003
2004
2005
2006
2007
2008
2009
2010
2011
2012
2013
2014
2015
2016
2017
2018
2019
2020
33
62
87
110
131
148
164
177
189
198
206
213
218
223
225
227
229
230
231
231
231
231
231
231
231
231
231
231
231
231
231
231
231
11.0
21.3
30.5
39.4
47.6
54.4
61.9
67.8
73.5
78.3
82.7
86.6
89.7
91.8
92.6
93.4
94.2
94.7
95.1
95.1
95.1
95.1
95.1
95.1
95.1
95.1
95.1
95.1
95.1
95.1
95.1
95.1
95.1
32
60
84
105
124
140
153
165
175
182
189
195
199
203
207
210
211
212
212
212
212
212
212
212
212
212
212
212
212
212
212
212
212
10.7
20.6
29.5
37.6
45.1
51.5
57.7
63.2
68.1
71.9
75.9
79.3
81.9
" 83.5
85.2
86.4
86.8
87.2
87.2
87.2
87.2
87.2
87.2
87.2
87.2
87.2
87.2
87.2
87.2
87.2
87.2
87.2
87.2

-------
              Figure 4-3.   TOTAL GASOLINE CONSUMPTION VS.  GASOLINE  CONSUMED
                             BY  ONBOARD CONTROLLED VEHICLES  (1988-2020)
                    -A—A—A—A—  Total gasoline consumption (see  Figure 4-2)
                                    Consumption by all LDVs and LDTs
                   	  Consumption by onboard  controlled vehicles (no tampering)
                   	 Consumption by onboard  controlled vehicles (with  tampering)
  109  1
350
300
250 —
200 —
150 —
100 —
50
          109 gal,
          90 —
          80 —
          70 —
          60
          50 —
          40 —
          30
         20 —
          10 —
                                        ^\    A	 A    A    A

                1990
                          1995
•'  i  !  ;   !  '  i
2000       2005
2010
          2015
                                                                                2020
                                           YEAR
                                          4-22

-------
controlled vehicles, with and without tampering, as well as total  gasoline
consumption for 1988 to 2020.
4.3  REGULATORY STRATEGIES
     Twenty-six regulatory strategies were evaluated for the gasoline
marketing industry.  These regulatory strategies were chosen to delineate
the range of available strategies.  The estimated relative costs of
control and associated reductions in emissions and health risk were
then assessed in order to bound the expected costs and impacts for all
strategies and to compare the relative costs and impacts among the
strategies.  Thirteen of the strategies examined control of all facilities
while eleven of the strategies examined control  of all but certain size
exempted facilities.  The size exemptions were based on statutory
requirements or throughput levels below which control costs might be
expected to be excessive.  The other strategies (benzene reduction in
gasoline)  examine two alternate levels of control while two other
strategies (baseline or current controls and onboard controls for
vehicle refueling)  do not lend themselves to exempting facilities.
     Each regulatory strategy is comprised of a particular control
option for each source category.  The source categories are bulk
terminals, including storage tanks, terminal-owned tank trucks and
for-hire tank trucks; bulk plants, including plant-owned and for-hire
tank trucks; and service stations, including filling or inloading of
the underground storage tank, all vehicle refueling, and as a subset,
self-service refueling.  Section 4.4 describes the control options and
size exemptions considered for each source category.
     The costs and impacts of each regulatory strategy were evaluated
for the time period beginning in 1986 through 2020.  The initial year
of 1986 was chosen because at the time of the analysis it was the first
year that implementation of any control  option could be begun.  The
analysis was extended through 2020 so that all of the light-duty
vehicle fleet would be equipped with onboard controls and so that
Stage II and onboard controls could be evaluated with both options fully
implemented.
                                  4-23

-------
             TABLE 4-9.  GASOLINE MARKETING REGULATORY STRATEGIES
                         STANDARD NUMBERS AND TITLES
Number
Title
Abbreviated
   Ti tl e
I           Baseline (No Additional Controls)

II   A & Ba Stage II - Selected Nonattainment
              Areas

III  A & B  Stage II - All Nonattainment Areas

IV   A & B  Stage I - Nationwide

V    A & B  Stage II - Nationwide

VI   A & B  Stage I and Stage II - Nationwide

VII         Onboard - Nationwide

VIII A & B  Stage II - Selected Nonattainment Areas
            & Onboard - Nationwide

IX   A & B  Stage II - All Nonattainment Areas
            & Onboard - Nationwide

X    A & B  Stage I & Onboard - Nationwide

XI   A & B  Stage II - All Nonattainment Areas and
            Stage I & Onboard - Nationwide

XII  A & B  Stage II & Onboard - Nationwide

XIII A & B  Stage I & Stage II & Onboard - Nationwide


XIV  A & Bb Benzene Reduction in Gasoline
                                       Baseline

                                       Stage II - NA*


                                       Stage II - NA

                                       Stage I

                                       Stage II

                                       Stage I  & Stage  II

                                       Onbd

                                       Stage II-- NA*
                                       & Onbd

                                       Stage II - NA
                                       & Onbd

                                       Stage I  & Onbd

                                       Stage II - NA  &
                                       Stage I  & Onbd

                                       Stage II & Onbd

                                       Stage I  & Stage  II
                                       & Onbd

                                       Gas  Bz Reduction
dA-with size exemptions; B-no size exemptions
 Stage I size exemptions:
      (1) bulk plants with throughputs <4000 gal/d from balance controls  on
          outgoing loads; and
      (2) service stations with throughputs <10,000 gal/mon.
 Stage II size exemptions:
      (1)  all service stations with throughputs <10,000 gal/mon;  and
      (2) all independent service stations with throughputs <50,000  gal/mon.
bBenzene reduction:
       A. removal of 94.5 percent of Bz from reformate  fraction for  total
          reduction of 62.4 percent;
       B. removal of 94.5 percent of Bz from reformate  and FCC  fractions  for total
          reduction of 81.3 percent.
                                     4-24

-------
                  TABLE  4-10.
                                   COMPOSITION  OF  REGULATORY  STRATEGIES

                                   BY SOURCE CATEGORY
                                       Control Option3  for  Each Source Category
     Regulatory
     Strategy
                              Bulk
                            Terminals
 Bulk
Plants
     Service    Stations
 GasoTi ne
Inloading
Vehicle
Refueling
  I .  Basel i ne                  B/L

 II.  Stage  II - NA*
     A. with size exemptions   B/L
     8. no  exemptions          B/L

III.  Stage  II - NA
     A. with size exemptions   B/L
     B. no  exemptions          B/L

 IV.  Stage  I
     A. with size exemptions   C
     B. no  exemptions          C

  V.  Stage  II
     A. with size exemptions   B/L
     B. no  exemptions          B/L

 VI.  Stage  I & Stage II
     A. with size exemptions   C
     B. no  exemptions          C
 VII.  Onboard
                MA*
VIII.  Stage  II -
      Onboard
      A.  with  size exemptions
      3.  no  exemptions

  IX.  Stage  II - MA a
      Onboard
      A.  with  size exemptions
      3.  no  exemptions
                              3/L
                              3/L
                              B/L
                              B/L
                              B/L
  X. Stage I  & Onboard
     A. with  size  exemptions   C
     B. no exemptions          C

 XI. Stage II - NA &
     Stage I  & Onboard
     A. with  size  exemptions   C
     B. no exemptions          C
                                                B/L
                                                B/L
                                                B/L
                                                B/L
                                                B/L
                                                C(E)
                                                C
                                                B/L
                                                B/L
C(E)
C

B/L
B/L
B/L
B/L
B/L
                                                C(E)
                                                C
                                                C(E)
                                                C
             B/L
             B/L
             B/L
             B/L
             B/L
              St. I(E)
              St. I
              B/L
              8/L
 St.I(E)
 St. I

 B/L
 B/L
 B/L
 B/L
 B/L
              St.  I(E)
              St.  I
              St.  I  (E)
              St.  I
                        B/L
                     St.II - NA*(E)
                     St.II - NA*
                     St.II - NA(E)
                     St.II - NA
                     B/L
                     B/L
                     St.II'(E)
                     St.II
                                                                                  St.IKE)
                                                                                  St. 11

                                                                                  Onbd
St.II - NA*(E)  & Onbd
St.II - NA* & Onbd
St.II - NA(E)  & Onbd
St.II - NA & Onbd
                     Onbd
                     Onbd
                      St.II - NA(E) & Onbd
                      St.11 - MA & Onbd
                                  TABLE  4-10 CONTINUED
                                               4-25

-------
                    TABLE  4-10.    COMPOSITION  OF  REGULATORY STRATEGIES

                                     BY SOURCE CATEGORY  (concluded)

XM.
•


XIII.



XIV.




Regulatory
Strategy
Stage II 5 Onboard
A. with size exemptions

B. no exemptions
Stage I 5 Stage II & Onboard
A. with size exemptions

8. no exemptions
Gas Bz Reduction
A. 62.4% benzene Bz:
reduction
3. 81.3% benzene Bz:
reduction
Bulk
Termi nal s

B/L

B/L

C

C

0.376 x B/L

0.187 x B/L

Control Optiona for
Bulk
Plants

B/L

B/L

C(E)

C

Bzi 0.376 x B/L

Bz: 0.187 x 8/L

Each Source Category
Service
Gasoli ne
Inloading

B/L

3/L

St. I(E)

St. I .

Bz: 0.376 x 3/L

Bz: 0.187 x 3/L

Stations
Vehicle
Refuel ing

St. IKE
Onbd
St. II S

St. IKE
Onbd
St. II 5

Bz: 0:376

3z: 0.187




) 5

Onbd

) &

Onbd

x B/L

x B/L

aKey to control option abbreviations:

 3/L - Baseline (Mo Additional Controls)
 C - Controlled
 St. II - HA* - Stage II  in Selected Nonattainment Areas
 St. II - HA - Stage II in All Monattainment Areas
 St. I - Stage I Nationwide
 St. II - Stage II nationwide
 Onbd - Onboard Nationwide
 St. II - HA* & Onbd - Stage II in Selected  Nonattainment Areas and Onboard Nationwide
 St. II - MA & Onbd -  Stage II in All  Nonattainment Areas and Onboard Nationwide
 St. II i Onbd - Stage II & Onboard, Nationwide
 (£) - with size exemptions
 8z: 0.376 x B/L - benzene emissions equal to 0.376 times baseline level;  other pollutants
  unaffected
 3z: 0.187 x 8/L - benzene emissions equal to 0.187 times baseline level;  other pollutants
  unaffected
                                                 4-26

-------
     The  industry-wide regulatory strategies that were evaluated in this
 study are presented in Table 4-9.  These regulatory strategies combine
 the control options for each source category as shown in Table 4-10.
     Baseline  (Regulatory Strategy I) reflects current control under
 promulgated and proposed Federal, State and local  regulations.  Bulk
 terminals and  their storage tanks are controlled under the bulk terminal
 and volatile organic liquid storage new source performance standards
 (NSPS).  Bulk  terminals, bulk plants, tank trucks, and underground
 storage tank filling at service  stations (Stage I) are covered by
 control technique guidelines (CTG's)  for ozone nonattainment areas and,
 thus, many of  the recommended controls have been incorporated into
 State implementation plans (SIP's) and subsequent regulations.  Also,  a
 few localities have implemented Stage II vehicle refueling controls in
 order to attain the ambient ozone standards.
     The nonattainment-area-only strategies (II and III)  affect only
 vehicle refueling emissions by recommending Stage II service station
 controls in those areas while leaving all  "Stage I sources" (bulk
 terminals, bulk plants, and Stage I  - gasoline inloading  - at service
 stations)  at baseline control  levels.  All  Stage II nonattainment area
controls were assumed to be installed during 1986, except for those on
 independent service stations,  which  were phased in over  a 3-year
 period through 1988, as required by  Section 324 of the Clean Air Act.
     The nationwide Stage I and/or Stage II strategies (IV, V, and VI)
 evaluate the implementation of Stage  I  (at all  source categories
throughout the gasoline marketing system)  and Stage II controls (at
service stations),  either singly or  in  combination through a NESHAP.
Nationwide controls under a NESHAP were assumed to be installed equally
within the 2 years  allowed under Section 112 of the Clean Air Act
beginning  in 1987,  except that Stage  II  controls at independent service
stations were assumed to be installed equally over a 3-year period in
accordance with Section 324 of the Act.
     The impacts of onboard controls  alone  through a motor vehicle
requirement under Section 202(a)(6) of  the  Clean Air Act  are assessed
under Regulatory Strategy VII.   Onboard  controls were assumed to be
installed  on new light-duty vehicles  and light-duty trucks beginning in
                                  4-27

-------
1988  (see Appendix C).  Onboard-controlled vehicles gradually comprise
more  of the total fleet as  additional new cars are built and as older
cars  are scrapped.   Onboard-controlled cars are projected to consume
over  one-half of the on-highway  gasoline consumption by 1993 (see Table
4-8).  Essentially all of the light-duty vehicle fleet is expected to
be onboard-controlled by about 2006.  Thereafter, only about 5 percent
of the on-highway gasoli.ne  consumption (used by heavy-duty gasoline
trucks, motorcycles,  etc.)  would not be controlled by onboard.
     Most of the remaining  regulatory strategies (VIII through XIV)
evaluate the impacts  of combinations of the previous strategies and,
in particular, combinations with onboard controls.  One strategy (X)
combines nationwide  Stage I controls throughout the system with onboard.
Several strategies (XVII, IX and XI) assess combinations of the
Stage II in nonattainment area options with onboard.  The rationale
behind these strategies is  to implement controls quickly in metropol-
itan  areas to attain  the ambient ozone standard and reduce hazardous
emissions, while relying on onboard controls (with Stage I controls in
Strategy XI) to reduce nationwide emissions more gradually.   Two
other strategies (XII and XIII) similarly evaluate the combination
of nationwide Stage  II controls with onboard (and with Stage I  in
Strategy XIII).  For  the strategies combining onboard with
Stage II in nonattainment areas or nationwide, the Stage II  controls
were  assumed to be phased out, i.e., not replaced, upon completion of
one useful  equipment life for balance and hybrid systems (15 years)
and two useful  equipment lives for vacuum assist systems (16 years
at 8 years per life) because at that time,  onboard controls  would be
virtually fully implemented.
     The final  regulatory strategy (XIV)  evaluates the effects  of
reducing the benzene content of gasoline during refining.   Two  levels
of reduction were considered, based on 94.5 percent removal  of  benzene
from either the reformate fraction of gasoline alone or both the
reformate and fluid catalytic cracked (FCC)  gasoline.   The refinery
modifications for the benzene reduction  were  assumed to be  installed
over a 3-year period, beginning in 1986,  and  to be completely effective
beginning in 1989.   The removal  of benzene  from gasoline  proportionately
                                  4-28

-------
reduces evaporative emissions of benzene throughout the gasoline market-
ing system, but does not affect the emissions of other pollutants.
Although the reduction of benzene in gasoline apparently would have the
added benefit of reducing automotive emissions of benzene, one study
by the U.S. EPA Office of Mobile Sources (QMS)* suggests that only
evaporative automotive emissions would be reduced.  The benzene exhaust
emissions were found to be approximately constant, regardless of gasoline
benzene content.  Evidently, benzene is formed in the exhaust by crack-
ing of larger hydrocarbon molecules or reforming from smaller molecules.
4.4  SOURCE CATEGORY CONTROL OPTIONS
     This section describes the control options evaluated for each  of
the gasoline marketing industry source categories:  bulk terminals, bulk
plants, and service stations (including self-service).  The considered
levels of control are noted as well  as any size exempted facilities.
The model plant sizes are summarized in Table 4-11.  The numbers of
affected facilities for each source category and facility type are
presented in Table 4-12.  Baseline controls reflect the implementation
of Federal, State, and local regulations as of 1982.  The baseline
control option would require no additional  controls.
     4.4.1  Bulk Terminals
     Two control options were examined for bulk terminals:  no addi-
tional control  measures (baseline controls) and control of all
facilities with no exemptions.  The option requiring control  without
size exemptions would limit VOC emissions from terminals currently  without
controls to 35 mg/liter.  This option would apply only to terminals
without controls.  Terminals that are already controlled under CTG's
are required to emit no more than 80 mg/liter.  In addition to the  35
mg/liter requirement, which would necessitate the use of some type  of
vapor processor for the loading rack emissions, terminal operators
would be required to control fixed-roof storage tanks and to load only
a
 Black, P.M., I.E. High, and J.M. Lang.  Composition of Automotive
 Evaporative and Tailpipe Hydrocarbon Emissions.  Journal  of the Air
 Pollution Control Association.  30_: 1216-1221.  November 1980.
                                  4-29

-------
TABLE 4-11.  GASOLINE MARKETING FACILITY MODEL PLANTS
Facility Type/
Model Plant Number
Bulk Terminals
Model Plant 1
Model Plant 2
Model Plant 3
Model Plant 4
Bulk Plants
Model Plant 1
Model Plant 2
Model Plant 3
Model Plant 4
Service Stations
Model Plant 1
Model Plant 2
Model Plant 3
Model Plant 4
Model Plant 5
1 	
1
1 Model Plant Throughput
1 l 1 ters/ day
378,500
946,400
1,893,000
3,785,000
11,400
24,600
47,300
64,400
liters/month
18,900
75,700
132,500
246,100
700,300
gal Ions/ day
100,000
250,000
500,000
1,000,000
3,000
6,500
12,500
17,000
gall ons/mo nth
5,000
20,000
35,000
65,000
185,000
Throughput Range Represented by Model
Plant
11 ters/ day
0-757,100
757,100-1,514,000
1,514,000-2,271,000
>2, 271, 000
0-15,100
15,100-30,300
30,300-64,400
64,400-75,700
liters/month
0-37,900
37,900-94,600
94,600-189,300
189,300-378,500
>378,500
gallons/day
0-200,000
200,000-400,000
400,000-600,000
>600,000
0-4,000
4,000-8,000
8,000-17,000
17,000-20,000
gallons/month
0-10,000
10,000-25,000
25,000-50,000
50,000-100,000
>100,000
                      4-30

-------
                         TABLE 4-12.    NUMBER  OF FACILITIES AFFECTED

                            BY  GASOLINE MARKETING  CONTROL  OPTIONS
FACILITY TYPE
BULK TERMINALS
BULK PLANTS
- No Exemptions
TOTAL FACILITIES
AFFECTED
500a
8,040t>
Number
MP 1
240
3,400
of Facilities
MP 2
135
4,000
in Each Model
MP 3
105
560
Plant (MP)"
MP 4
20
80
Size
MP
NA
NA
5

    - Exempt Bulk Plants                  8,040C.d
      <  15,140 Ipd

  SERVICE STATIONS
  NATIONWIDE ALTERNATIVES

  Stage  I
    - No Exemptions                      197,900

    - Exempt Stations
      <  37,850 I/mo                      83,100

  Stage  II
    - No Exemptions                      383,200

    - Exempt Stations < 37,850 1/mo
      and Independent
      Stations < 189,300 1/mo            120,600
      (Independents Only)

  NONATTAINMENT ALTERNATIVES

  Stage  I
    -•No Exemption Option                 6,895

    - Exempt Stations < 37,850 1/mo
                                         2,895

  Stage  II
    -No Exemption Option                114,900

    - Exempt Stations < 37,850 1/mo
      gal/mo and Independent
      Stations <189,300 1/mo             36,200
      (Independents Only)

  FIXED  ROOF STORAGE TANKS6                 630

  TANK TRUCKS

    - For-hire Terminal Trucks            5,800 a

    - For-hire Bulk Plant Trucks
      (No Exemptions)                    16,300b

    - For-hire Bulk Plant Trucks
      (<15,140 Ipd BP Exempt)             9,40QC,d
 3,400
114,800
222,100
  4,000
 66,600
4,000






33,600


33,600


65,400



47,500





 1,170


 1,170


19,600



14,300
                            560
 1,035


 1,035


17,300



10,500
                                         80
                                       9,200
                                       (3,150)
                                                  NA
29,700       15,800      4,000


29,700       15,800      4,000


57,500       30,700      7,500
                          34,900       30,700      7,500
                                      (10,500)    (2,700)
  550       140


  550  .     140


9,200      2,200
           2,200
            (800)
aBulk terminal  affected facilities include about 2,870 terminal-owned trucks.

bBulk plant affected facilities with no  exemptions include 16.000  plant-owned trucks.

cBulk plants with  throughput <15,140 Ipd (4,000 gal/day) will  oe  required to use submerged fill  on outgoing loads (account
 trucks) .
dBulk plant affected facilities .under the with exemption option include 9300 plant-owned trucks  requiring vapor
 balance and 6700  plant-owned trucks requiring only submerged  fill.  An additional  6900 for-hire trucks would
 require submerged fill.

eincludes  only  fixed roof. storage tanks  at bulk terminals to oe retrofitted with internal  floating r-oofs.
 The two fixed  roof tanks assumed per bulk plant cannot be retrofitted with internal floating  roofs
 since floating roofs are not compatible with vapor balancing.
                                                        4-31

-------
certified tank trucks.  Fixed-roof storage tanks would be controlled
with internal floating roofs.  Terminal operators would be required to
restrict loading of gasoline tank trucks to those trucks that had
passed an annual vapor tightness certification test.  About 500 terminals
would be affected by the control option.  An estimated 630 fixed-roof
storage tanks would have to be retrofitted with internal floating
roofs.  In addition to the approximately 2870 terminal-owned tank
trucks, 5800 for-hire terminal trucks would have to be controlled.
     4.4.2  Bulk Plants
     For bulk plants, three control options were evaluated: no addi-
tional control measures (baseline controls) and controlled with and
without size exemptions.  The control required would be vapor balancing of
storage tanks with transport trucks (bringing gasoline to bulk plants
from terminals), and either vapor balance or submerged filling of
account trucks (taking gasoline from bulk plants to service stations or
other accounts).  Both transport and account trucks are required to be
vapor-tight.  The bulk plant control  option without size exemptions will
affect 8040 bulk plants, requiring vapor balance of both incoming and
outgoing loads.  The control  option with size exemptions affects bulk plants
loading less than 15,140 liters per day (4,000 gal/day).  Under the
size exemption option, the 3400 small plants (MP1), which have a relatively
greater cost of control and are often exempted by existing State regula-
tions, are required to use only submerged fill  to control emissions
from loading of outgoing (account)  vapor-tight trucks, while using
vapor balance on storage tanks and incoming (transport)  vapor-tight
tank trucks.  The remaining 4,640 larger bulk plants (MP 2-5)  would
still be fully controlled.
     Plant-owned account trucks requiring vapor balance controls are
estimated to number about 16,000 with no size exemptions.  The control
option with the 15,140 liters per day (4,000 gal/d)  size exemption
would require vapor balancing of approximately 9,300 account trucks and
submerged filling of the remaining 6,700 trucks.  Similarly, the for-hire
account trucks requiring vapor balance are estimated to number about
16,300 with no exemptions.  The control option with size exemptions
                                  4-32

-------
requires vapor balancing of approximately 9,400 account trucks and
6,900 trucks would be subject to submerged fill requirements.
     4.4.3  Service Stations
     Gasoline handling operations, emissions, and controls at service
stations are basically divided into two steps, which are commonly
termed Stage I and Stage II.  Stage I refers to gasoline inloading at
the service station, that is, filling of the underground storage tank.
Stage II refers to vehicle refueling at the station.  The control
options for service stations affect one or both of these stages to
varying degrees.
     Stage I.  Stage I, or inloading, controls employ vapor balance
between the tank truck and the underground storage tank at the service
station.  This vapor balance step completes the gasoline marketing
Stage I system that captures and transfers potential emissions from the
service station storage tank, tank trucks, and, if applicable, bulk
plant, to the bulk terminal vapor processor for recovery or destruction.
In addition to baseline, service station inloading was evaluated for
Stage I vapor balance controls on all facilities and on all facilities
with gasoline throughputs greater than or equal to 10,000 gallons per month,
The size exemption reflects the higher relative costs of controls on
smaller facilities and existing size exemptions under State regulations.
The estimated number of affected facilities is 197,900 without exemp-
tions or 83,100 with a size exemption for stations with throughputs less
than 37,850 liters per month (10,000 gal/mo).
     Stage II and Onboard.  The vehicle refueling control options utilize
several different regulatory approaches and control techniques.  One
control technique uses a vapor balance between the service station
underground storage tank and the vehicle gasoline tank and is commonly
termed Stage II controls.  The other control technique uses a vapor
seal in the vehicle fill neck to force the vapors being displaced from
the tank into a carbon canister on the vehicle where they are adsorbed,
or onboard control.  Implementation of Stage II controls, as well as
Stage I control, for service stations, bulk plants, and bulk terminals,
was evaluated on a nationwide basis.  Stage II controls were also
                                  4-33

-------
evaluated for ozone nonattainment areas only.  Implementation of onboard
controls was assumed to occur nationwide.  It should be noted that vehicle
refueling controls not only reduce VOC and hazardous emissions dispersed
from the service station, but also reduce exposure to hazardous pollutants
during self-service refueling.  Reduction of self-service exposure
results in a significant reduction in estimated incidences of cancer (as
shown in Chapter 6) because about 90 percent of the incidences are
attributable to self-service refueling.
     All Stage II options were evaluated for both with and without size
exemptions.  Examined size exemptions include any service stations
generating less than 37,850 liters/month (10,000 gal/mo.) throughput
(based on existing State and local regulations and expected relative
economic impacts) and independent dealers below 189,300 liters/month
(50,000 gal/mo.) sales volume (in accordance with Section 324 of the
Clean Air Act and assuming that Section 324 applies to Section 112
regulations).  These stations, for which size exemptions are examined,
represent the segment of the retail industry that may be most vulnerable
to changes in marketing economics as well as external costs such as
vapor recovery costs.  The basic vehicle refueling control options
(excluding baseline controls and combinations of options) of Stage II
in selected nonattainment areas, Stage II in all nonattainment areas,
Stage II nationwide, and Onboard nationwide, are discussed in the
following paragraphs.
     Stage  II-NA.  Stage II in all nonattainment areas (NA) was chosen
to reflect the development of a CTG recommending Stage II controls as
RACT for vehicle refueling VOC emissions.  These areas are all ozone
nonattainment areas that were granted an extension  (by EPA) up to 1987
for achieving the  standard plus those ozone  nonattainment areas that
EPA identified in  Appendix D of the February 3, 1983, Federal Register
(48 FR  5005) as  "unlikely to attain the  NAAQS" by December 31, 1982.
It should be noted that Table III on page 5026 of the February 3, 1983,
Federal  Register provides a listing of the "extension areas" for ozone
                                   4-34

-------
                                           ) -« r^   CO
                                                                                          f)
                                                                                                   in o '•Q

                                                                                                   O"i ^^ •— »
              t/1  S- O
              rtJ JC *-< •
              Of-—.
                                                                                                    l CO C\J
                                                                                                    *» O CT>
                                                                                                      ^3- CO
                                                                     3 1-
                                                                     1. O   T3   «-
                                                                     «C K ,— 1_ -o O
                                                                       — o o s-
                                                                     QJ 4-> i- V- fO >,
                                                                     C ^- t- J- 3 4J

                                                                     ^ CO O T T* tJ
    o
    UJ

    •=>
    (S)
                   o!  7.
       (O
       z
       o
    <—G£.\—
    < O  <
    z: u_  rs
    •—•     _i
        CD  «t
oo oo i—  >
i— «C O  UJ
 I  UJ
•sf Cti <  C3
    <     Z
        S-  i—i
UJ h- CQ  h-
—1 Z     UJ
CQ UJ Q  i^
< ^Z UJ  Ql
1— 3 1—  <

                                                               o

                                                               3
                                                                                               0;    O
                                                                                               re    £
    U. UJ
                                                        ) —.   —t   C3
                                                                                                             o r-> — • CM ^
                                                                                                                   -
o
       O
       oo
                                                                                                                    cor-COCO^H   r-cor")
                                                                                                                    cr»ovo«3-^r   CM   ro
                                                                                                                    CM — i •— "   •— "   —   —i
                   5-1
                                                                                                     . (I,   ^—
                                                                                                      i&^:
                                                                       &_ eg re   »—
                                                                                                                O)     =      V O O
                                                                                                             .^•DCJUJOJ'—   — i-
                                                                                                              OO_C^ = .—   'CC
                   5.1
                          i5   -^

                                                              C    CJ   r
                                                             4-35

-------

                  *-,    ._;
                  O    Cr
                              . ,_   ,— w. 3 .,_    COIO
                       =    ^  CO
                                      S_  rt3 C 1-   -t— -^- C    ^—    oj —

                                                     •"
                       i   -S
      r*. f*» m co
      LO o .— 4 o ^
      O co co CO •"*
». vo c
a co r
O W f"> CO L
en —^ Cft —^ L
              CO •
              Cl C

                                                                                                  ro —* «sf r-.
                                                                     :  o    >•> (
                                                                     (  4J    C I
      _   J
>V   O =3 )—  (d J

•?    _  « .P p-  !
cj    o = t
                                                                            OJ Crt > r— j
                                             cj —i -J E
                                                                                                =J JC  OJ O

 <9c   2
«;   o
                                              4-36

-------
                                              IO

                                              en
                                                    o
                                                    2   I
    O3
    «C

    oo
    2
    Q
    CQ

    a
    UJ
    i—
    o
  • u_
«*• =C res
•—     oo
  i  oo _i
"d- eC o
    LU CC
CQ     O

VI 3
fl3 a.
O 3
e*»
H-






c
•— r- O
J^ M [^

£ 3
O Z O
a.







o

£

o
a.














C3 cn LA *f?

cn ^H cn cn
i— 4 CM cn












en co CM o

?"^ CO CD









o
^f CO P** *-^
CO «-« CO CO
O i-H CO

' i-H IO U*> **•
CM co r*«. in
•^ CO ^O *

CM" CM cn CM
CM CO P> , CM






0)
, c



V)
fQ


OJ
•o
3

O

f"

V)
4)
^

cn
u-

(U
•a
3
O
] *
IO
c


CD
c



o



VI

s
V
s
o
4->
U


4_>
VI

a

01
.c



VI


ra

VI OO
0)
4-J 4J
| |
l_4



IO

i— CO

c c
0 0
&- +•> VI
o — c
 * 4->
c a.
cu c E
o o 
rf 4J a o
a.' 4J
3 Qt C 41
O X OJ -C
ja LU o +^
0) 14-
>, • cu 0
.am
C CM 4->
4-» t— • C
3 re- -3- 4>
a. cu — . u
3 VI S-
t. 10 >» tu
-C J3 J2 O.
4->

•a 10 3 •
01 CLCO
4-> »-" 3 CM
o >— s-
01 J= >,
 01 3
Ol  CX
5^S2
O t- -C
•a
Ol

VI 1—
i— O
o s-
i. 4->
4-> C
c o.
o u
^ 2



Ol >»


4-> Ol
CO i-

5 "
•r- f—

•0 "*
i2
3
i— X
o
O CO

 3
vi Q.
51
Ol 4-1



•O T3
C C

IO CM


















VI

'o
^5
c
o
u
01
cn
4.9
CO

Ol



VI
•2

e
|


P—

0

• M
a?
tn
VI

0 0
CJ 4->

CM IO
•-4
•o
4-> C
3 IO

CO
•— O
•— O^

en

VI «3-
4-> *
Ol
VI <4-
3 O
"5 C
to O
VI •»—
VI 4->
y ,^
3
C O.
i- O
a.
VI
Ol r- •
•^ IO VI
4-> 4J I—
COO
3 4J S.
O 4J
o ia c

IO *i— i—^
3
•a .01
c. — cn
IO X IO
- t- 4J
>• C CO
i? 01
VI O Ol
   •O    Ol     f-    •—*
   CM    CO    CM    LO

i    =     =     =     =
                              01

                              10


                              Ol
                              s_
                              3
                                                              Ol


                                                              IO
4-"  U
C  I
O  OJ
•a i—
c  cn
01  C
CL-I—
Ol  VI
-a
C  10
 &. O 4-»
   s.
 VI Ol Ol c
^- CL O *
 O    3
 i_ m -a
         4JO

         01
                          3
                          O
                          O
                          o
                          z
                                                                    O  VI
                                                                    o ••-
       3  U
       O  3
       3 TJ
          01
                                                                             3  O i—
                                                                             o .a  o "O
                                                                             o -a
                                                                                           O CM


                                                                                           T- CO VI
   O VI
VI

Ol C Ol
•T- O U


C 4-> '>
3 rtj i-
O i— Ol
O 3 I/I
   CL
IO O C
   CL O
                                                                                           CM
                                                              -r-    VI  IO
                                                                    3  S

                                                                    O i—
                                                                    C  O
                                                   c: o
                                                   o 3
                                                   o
                                                      VI
                                                                                               fQ
                        •S:
                                                                      >f_    -r- 1-4 •—I r—     (fl  3 4->    -^
                                              

                                             LU CO OO
                                                                        -o
                                                            4. 1/1       3
                                                            Q. « J=    r-
                                                                                           O  
-------
           TABLE 4-15
NONATTAINMENT AREAS COMMITTED TO
     OR SCHEDULING STAGE II
VEHICLE REFUELING CONTROLS (NA*)
Area
Connecticut
Fairfield
Illinois
Cook
DuPage
Kane
Lake
McHenry
HUT
Maryland
Anne Arundel
Baltimore
Carrol
Harford
Howard
Montgomery
Prince George's
Baltimore City
New Jersey (Statewide)


a
Gasoline throughput at
Gasoline
Throughput3
1000 gal/yr

319,800

1,664,900
298,600
107,400
158,500
48,800
141,700

196,400
82,000 •
84,900
77,500
69,300
228,700
198,100
453,400
3,133,100


Baseline for National
Area
Mew York
Bronx
Kings
Nassau
New York
Queens
Ri chmond
Rockl and
Suffolk
Westchester
Virginia
Arlington
Fai rf ax
Loudon
Prince William
Pennsylvania
Bucks
Chester
Del eware
Montgomery
Philadelphia
TOTAL
Emissions Data Systems
Gasoline
Throughput3
1000 gal/yr

140,900
276,800
548,600
120,700
382,700
100,200
98,700
483,600
321, 30Q

67,900
197,300
35,800
81,600

176,600
163,200
191,600
254,200
397.900
11,302,700
(NEDS).
              4-38

-------
with the exception that Tennessee-Nashvilie-Davidson County was
inadvertently omitted from that table under Region IV, and subsequently
was omitted from the list of nonattainment areas for the analysis.   The
nonattainment areas that were assumed to be affected by a CTG
recommending Stage II controls as RACT are listed with their baseline
gasoline throughputs in Table 4-13.  This list includes all  the areas
given as "extension areas" or "unlikely to attain the NAAQS" in the
February 3, 1983, Federal Register, except for those areas already
using Stage II controls (26 counties in the Bay Area, San Diego, South
Coast, and Sacramento AQCR's of California plus the District of Columbia,
with a total throughput of about 34 billion liters [9 billion gallons]).
     Stage II-NA*.  Stage II in selected nonattainment areas (NA*)
was chosen to reflect the lowest feasible level of vehicle refueling
control  through a control techniques document (CTD).  The selected
nonattainment areas (NA*) are a subs.et of the ozone nonattainment areas
discussed earlier that were granted extensions (by EPA) up to 1987  for
achieving the ambient standard for ozone.  Table 4-14 presents a com-
parison of the relative sizes of the selected (NA*) and all  (NA)
nonattainment areas.  The NA* areas are listed in Table 4-15 along  with
their baseline gasoline throughputs.  These areas have determined they
will need more than the reasonably available control technology
(RACT) specified in control techniques guidelines (CTG) documents for
various sources as well as control of other 100-ton stationary sources
(plus vehicle inspection and maintenance programs [I/M], Federal motor
vehicle control programs [FMCP], and transportation control  measures
[TCM's]) to attain by the statutory deadline.  Accordingly, the SIP's
for these areas include a commitment/schedule to evaluate and adopt
Stage II as an additional means of achieving necessary reductions in
VOC emissions to provide for attainment by 1987.
     National Stage II.  Stage II nationwide was selected as.a control
option to assess the effect of a NESHAP to reduce benzene and other
hazardous emissions.  The estimated number of affected facilities for
this option is 383,200 with no exemptions or 120,600 with size exemptions
compared with 114,900 for a CTG affecting all nonattainment areas with
no exemptions or 36,200 with size exemptions.  In contrast, a CTD affecting
                                  4-39

-------
selected nonattainment areas would affect an estimated 43,700  facilities
with no exemptions or 13,800 with size exemptions.
     Onboard.  Onboard controls nationwide is the other basic  vehicle
refueling .control  option.  The analysis assumed that onboard controls
would be installed only on new light-duty vehicles  (passenger  cars)  and
light-duty trucks (< 8500 Ib, i.e., vans, pickup trucks,  etc.).  Thus,
the effectiveness of onboard controls increases with time.  For  this
reason, the analysis was extended through the year  2020 so  that  the
analysis could continue after the entire light-duty vehicle fleet  would
be controlled.  In this way, onboard could be compared with Stage  II in
both the short and long term.
                                  4-40

-------
4.5  REFERENCES

1.  Bulk Gasoline Terminals -  Background Information  for Proposed
    Standards.  U.S.  Environmental  Protection  Agency.  Office of Air
    Quality Planning  and Standards.   Research  Triangle Park, N.C.
    Publication No. EPA-450/3-80-038a.   December 1980.  p. 3-1.

2.  Reference 1, Section 6.2.4.

3.  Graver Tanks. Commodity Storage Tank Product Literature. No date
    available.

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

5.  Arthur D. Little, Incorporated.   The Economic Impact of Vapor Control
    Regulations on the Bulk Storage Industry.   Report to U.S. Environ-
    mental Protection Agency,  Research  Triangle Park, N.C. EPA Publication
    No. EPA-450/5-80-001.   June 1979. p. III-9.

6.  Reference 4, p. 3-14.

7.  Lundberg Estimates.  National  Petroleum  News, September 1983.
    p. 12.

8.  Arthur D. Little, Inc., The Economic Impact of Vapor Recovery
    Regulations on the Service Station  Industry.   U.S. Occupational Safety
    and Health Administration, Washington, D.C. and U.S.. Environmental
    Protection Agency.  Research Triangle Park, N.C.  Publication
    No. EPA-450/3-78-029.   July 1978.  p. 47.

9.  National Petroleum News, February 1983.  p. 9.

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

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

12. Reference 8, p. 43.

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

14. Reference 8, p. 32.

15. National Petroleum News.  1983 Factbook  Issue. Mid-June 1983,
    Volume 75, No. 7A. p.  80.

16. U.S. Environmental Protection Agency.  Federal Register, Vol. 47,
    Number 210, October 19, 1982. p. 49329.

17. Telecon.  Scott Osbourn, Pacific Environmental Services, Inc.,  with
    Debra Paxton and Mark Rodekar, U.S. Department of Energy Mid-term
    Forecasting Branch.  September 30,  1983.   Department of Energy
    gasoline projections.

                                  4-41

-------
18. long, Y.P. and D.W. Houser.
    Science.  Oil & Gas Journal
Motor-gasoline Forecasting:   an Embryo
November 14, 1983. p.  114-118.
19. Energy and Environmental Analysis, Inc.  The Highway Fuel  Consumption
    Model Ninth Quarterly Report.  U.S. Department of Energy.   Washington,
    D.C.  Publication No. DOE/PE/70045-1.  February 1983.  Appendix B.

20. Exxon Research and Engineering Company Cost Comparison for Stage II
    and On-board Control of Refueling Emissions.  In:  On-board Control
    of Vehicle Refueling Emissions Demonstration of Feasibility.
    American Petroleum Institute.  API Publication No.  4306.   Washington,
    D.C.  October 1978.  p. 9.

21. American Petroleum Institute.  Cost Comparison for Stage  II and On-
    board Control of Refueling Emissions.  Washington,  D.C.  January
    1984.  p. 12.

22. Peterson, P.R. et _aJL  Evaluation of Hydrocarbon Emissions from
    Petroleum LiquiT~Storage.  U.S. Environmental  Protection Agency.
    Research Triangle Park, N.C.  Publication No..  EPA-450/3-78-012.
    March 1978.

23. Pacific Environmental Services, Inc.  Estimated Nationwide Petroleum
    Storage Tank VOC Emissions for the Years 1983  and 1988.  Report to
    TRW Environmental  Engineering Division.  Research Triangle Park,
    N.C.  Contract No. M23399JL3M.  April  5, 1983.
                                  4-42

-------
            5.0  ENVIRONMENTAL AND ENERGY IMPACTS

     The purpose of this section is to discuss the environmental
and energy impacts associated with the gasoline marketing control
options and regulatory strategies.  The majority of the discussion will
be spent on the methodology used to generate the quantitative analysis
on air pollution emission impacts.  A quantitative analysis of the
energy impacts of the regulatory strategies is also included in this
section.  The scope of the overall analysis did not allow for an  in-
depth evaluation of the other environmental  impacts (i.e., water,
solid waste); however, a qualitative discussion of these impacts  is
included.
5.1  AIR POLLUTION EMISSION IMPACTS
     Estimates of the emission reductions which could be achieved  under
each of the control  options and the regulatory strategies were analyzed
and are discussed here.  The analysis for each industry sector was
basically the same.   The potential emission reductions achievable  in
the base year (1982)  were calculated for each industry sector.  Given
no other changes, the emission reductions would be the same in each
subsequent year of the analysis.  However, there were two factors  which
affected the emission reductions in each year:  1)  phase-in of control
equipment installations, and 2)  change in gasoline consumption with
time.
     Section 5.1.1 discusses the initiation date of the control options
and the phase-in schedules assumed.  Emission reductions were presented
for the study period, both as cumulative and discounted totals.
Section 5.1.2 discusses the rationale for discounting emissions.
Section 5.1.3 discusses the assumptions and methodologies used to
generate the emission reduction estimates, including how gasoline
consumption changes  with time were incorporated into the analysis.
5.1.1  Phase-in Schedules for Control  Options
     It is unreasonable to assume that all equipment required by  the
regulatory strategies would be installed immediately at the time  the
regulatory strategy  takes effect.  It was assumed for the analysis that
                                  5-1

-------
some period of  time or phase-in of controls would take place before
100 percent of  the facilities had controls installed.  This phase-in,
therefore, would affect both the cost and emission reduction analyses.
In all cases a  linear phase-in was assumed which resulted in an even
distribution of installations with time.  In addition, statutory time
frames were used to consider complete phase-in of equipment.
     Under Mational Emission Standards for Hazardous Air Pollutants
(NESHAP) programs, the Clean Air Act requires compliance with regulations
within 180 days; however, variances can sometimes be obtained for up to
2 years.  Therefore, a linear phase-in period of 2 years was selected
for all nationwide options (i.e., terminals, bulk plants, storage tanks,
tank trucks, service station Stage I, and non-independent service station
Stage II).  Further, Section 325 of the Clean Air Act specifically allows
a different phase-in rate (3 years) for independent service stations.
EPA has not determined whether this applies to a Section 112 standard;
however this was assumed for the analysis.  For all  options which
required Stage  II controls in nonattainment areas, a phase-in rate of 1
year was used for all sources with the exception of independent service
stations where  a 3-year phase-in was used.  Figure 5-1 illustrates the
three linear phase-in rates used for the control  options and strategies.
     For purposes of determining capital cost estimates, any costs
spent in a year were attributed to that year (100 percent for 1-year
phase-in, 50 percent per year for 2-year phase-in, and 33 percent per
year for 3-year phase-in).  For emissions estimations and for annualized
cost estimates  the number of facilities which achieved the emission reduc-
tion or incurred the total annualized costs for the  entire year had to be
determined.  This number of facilities was estimated by determining the
area under each curve.  For example,  the following equation was used to
determine the values for the 3-year independent service station phase-in:
            1/3 X = (1/6 XJ; - 1/6 X?)
where
     A = Area under the curve of Y = 1/3X
                                  5-2

-------
                                                                                              re
                                                                                              01
                                                                                       -CM
                                                                                                      CO
                                                                                                      OJ
                                                                                                      O)
                                                                                                     J=
                                                                                                      O
                                                                                                      O)
                                                                                                      CO
                                                                                                      re
                                                                                                      re
                                                                                                      0)
                                                                                                      0)
O
CD
O
CO
O
CM
                                  ijq.LM  saj.q.LL
                                           5-3

-------
Using this formula, the percentage of total  facilities to be considered
for annualized cost and emission reduction in each year of the 3-year
phase-in was determined (year 1-17 percent, year 2-50 percent,  year
3-83 percent, year 4 and thereafter - 100 percent).   This same approach
was used to determine the percentages for a 2-year phase-in (year 1 -
25 percent, year 2-75 percent, year 3 and thereafter - 100 percent)
and a 1-year phase-in (year 1-50 percent, year 2 and thereafter -
100 percent).
     The onboard control option was based upon the installation of
controls on new vehicles.  Therefore, the number of vehicles controlled
in each year of the analysis is based upon the number of new vehicles
and the scrappage rates projected in Appendix C.
     The start date from which the phase-in would begin for the
regulatory strategies was estimated based upon assumptions of when
EPA would decide on the control approach to pursue.  At the time of the
analysis, it was assumed that EPA would decide on an approach in 1984.
Even if this dates slips uniformally for all control strategies,
the relative nature of the impacts between strategies do not change,
only the  actual start dates.  If EPA decided to pursue nationwide
regulatory strategies, it was estimated that the data gathering and
review processes involved with standards development would take 3 years.
Therefore, 1987 was selected as the date when nationwide options (with
the exception  of onboard) would become effective and would be the
initial date of facility phase-in.  If EPA decided to proceed with
nonattainment  area regulatory strategies, it was estimated that a
guideline document could be developed in approximately 1 year and that
it would  take  another year for the  regulatory strategies to be incorporated
into State regulations.  Therefore, 1986 was selected as the effective
date of nonattainment area regulatory strategies.  If EPA were to
pursue onboard regulatory strategies, it was estimated that a vehicle
standard  could be  prepared by 1985; however, the vehicle manufacturers
would need 3 years to incorporate the controls  into the vehicle design.
Therefore, 1988 was selected as the effective date for onboard controls.
Table 5-1 summarizes the effective  dates and the phase-in  schedules  for
onboard,  nonattainment  area, and  nationwide control options and regulatory
strategies.
                                   5-4

-------















GO
LU

=D OO
Q UJ
LU >-H
3= ts
O LU
5
•Z. 01
I-H h-
1 GO
LU
GO >•
< 01
Z O
rt I 	
LJ_ r~~
a _i
21 Z3
< CD
LU
oo o;
LU
1— CS
< z:
a HH
lit lit
LLJ t_LJ
^^ **"^
•-i cc
l— - tO
-p a;
•i- CU >-
3 S.
00 -P
a; c co
•r- LU
•P i.
•t- S- to
r— O <1)
•!-<*->-
O
U_ 0)
i— CM
•P i—
C (O S-
a> -P as
o oo a>
i- C >-
CD HH
00 •— 1
CU r~~
C7> O
(0 S- S.
i- -P (0
a> c cu
> o >-
< 0

c
1— 1
1 CU
CU -P
oo as
as a:
_g^
Q-


00
c
o
01- •!-
O -P
(O
s_ ,—
0) 'tO
>™ ^J
to
i— C
ro HH
•1—
^_j r_
••- o
£= S-
t— 4 -|_>
c
o
0








>^
di
CU
•p
(C
s-
^J
oo



o o
o o
T— t .— i





CO O
CO O
t— 4





O UO
LO r~-






l~- LO
t— 1 CM






S- S-
(O C^
CU CU

1 1
CO CM











r*"-» r*"*
CO OO
er> en
i — * » — i







-P
*— oo c:
cu ~o c cu oo
-OS- O £ CU
•i-  (O 4-> 'f C -i-
3 CU O E 4J -r- CD
£— *^~ *^ CU ^3 'O CU
OC7>£= -O-P S-00 4-»-P
•i-CUO COO CUO) -P(O
•P -P CU -C -I— fO i-
rt3 (0 ^_> r^ CU 4-* 4~* C -p
CS-Q. CUO O-^ OOO
-P CU "O -i- r— 2:
i — OOO C> i — T- a}
i — X i— i i- i — O i — CU
•^ 1 1 1 cu *3C to i — i-
— oo u_ «=C •<


o o
0 O
<— 1 1—1





co o
CO O
1— 1





O 0
LO . O
T— (





r-. o
r-H LO






1_ S-
, >> re
1 I  CTi O^
i— 1 i— < i-H








OO
c:
o
-p •.-
c: -P
CU as oo
-a -p oo cu
£= GO s- cu .-a -i-
CU CU •!— S- CT>
a. cu x: -P re cu
cu o -P •!- o -P
•a T- o i— .a «
c: > -^- e: s_
HH t- ^ 0 ' 0 -P
CU i — as OO
OO 
o
O -E
•P
M- •!-
O S

C CD
l— i C
1 '!—
CU -P
00 S-
fO fO
.c P
Q_ 00
n3
5-5

-------
5.1.2  Discounting of Emission Reductions
     Emission reductions for each year through 2020 were discounted
at 10 percent and then added to obtain the present "value" of
emission reductions, a single number encapsulating both the magnitude
and the timing of the reductions.  Dollar costs and savings were •
handled the same way.  The discounting of dollars is common practice.
However, the discounting of benefits that are not expressed in monetary
terms, such as lives saved, emissions reduced, and visibility improved,
is not done as commonly.  Typically, these other benefits flow more-or-
less evenly over time through the life of the control  equipment.
However, the dollar costs almost never flow evenly over time, and
therefore are discounted to the present and, if costs  are to be compared
with the annual flows of, say, emission reductions, then the costs  are
reannualized into constant dollar cost flow over time.  Then the two
flows, money and emission reductions, for various regulatory strategies,
can be compared.
     The question of whether to discount emission reductions does not
arise in this typical example because there is no need for discounting.
The discounted and reannualized value of any constant  stream, such  as
annual emission reductions, is exactly equal to the original stream,
regardless of what interest rate was used to compute both the present
value of the stream and the reannualized stream.
     The regulatory strategies in this analysis are not typical, because
the emission reductions are not smooth over time in every situation.
For example, the emission reductions from control expenditures made for
Stage II controls for a single service station will decline signifi-
cantly year-by-year if onboard controls are implemented simultaneously.
It is not logical to ignore this effect of time on emission reductions.
One way to summarize the situation for decision makers is to graph  the
flows of money and emission reductions over the years.  Such graphs can
be confusing if a decision is to be made between two close alternatives.
If more concise numerical comparison is to be made, some form of
discounting is necessary, even if the procedure is to  be implicit.
                                  5-6

-------
     Simply adding up the emission reductions  without regard  to  their
flow over time amounts to the use of a zero interest  rate  in   .
discounting.  A zero rate implies that society is indifferent between
a unit of'emission reductions today and a unit sometime in the future.
For example, if a person were given the option of sniffing benzene
today or sniffing the same amount 10 years hence, it  is highly likely
that such a person would opt for the 10-year delay.   This  reasoning
indicates that a positive, non zero interest rate should be used to
discount emission reductions.  Unfortunately,  there  is little theoretical
guidance available for selecting the appropriate rate.  This  analysis
uses 10 percent for all discounting (the effects of  alternative discount
rates on costs are discussed in Chapter 8.0).
     In summary, given a choice between sniffing hazardous emissions
now and sniffing it later, most people would prefer  to wait.   Clean
air today is more important than clean air tomorrow.   This "time
preference" is expressed mathematically by discounting future emission
reductions with a positive interest rate.  The analysis sought to
determine how much emission reduction will result from controlling
service stations and cars.  The answer for service stations depends
on what is done to control cars, and vice versa.  This means  that a
dollar a year spent to control pollution at a service station will
not necessarily yield the same emission reduction in one year as it
does in another year.  Regulatory strategies may be compared with each
other using emission reductions, cost/effectiveness, and a variety
of other measures of costs and benefits.   If the costs or benefits
change over time, the best way, theoretically, of comparing alternatives
is to discount the measures before comparison is made.  Then either
present values or reannualized flows may be used.
5.1.3  Emission Reduction Methodology
     5.1.3.1  Base Year  Emission Reductions.  The methodology for
calculating emission reductions for bulk terminals,  bulk plants, tank
trucks, storage tanks, and service stations was  the same.   The baseline
tables  (Appendix B) were used in estimating emission  reductions.  These
                                   5-7

-------
tables supplied the baseline emissions for the 1982 base year.  Emission
factors associated with the regulatory strategies discussed in Section 4.3
were then assigned to the throughputs in any of the areas in the baseline
table not controlled in 1982.  A controlled emission rate was calculated
and the difference between this rate and the baseline emission rate was
the emission reductions achievable by the regulatory strategies in the
base year of 1982.
     For example, the baseline emission rate of gasoline vapors for
bulk terminals was estimated as approximately 140,200 Mg/yr.  This
baseline included controlled and uncontrolled facilities.  The regulatory
strategies which affect terminals require:  1)  all terminals, not
previously controlled, to install control equipment to meet a limit of
35 nig/liter, and 2) all tank trucks loading at terminals should be
leak-tight (leakage emission factor - 96 mg/liter).  The baseline
emission table for bulk terminals was then adjusted to reflect-these
control measures.  The controlled emission rate in 1982 for bulk
terminals was then calculated to be approximately 55,800 Mg/yr.  The
potential emission reduction in 1982 for the bulk terminal regulatory
strategies was estimated as 84,400 Mg/yr.
     Table 5-2 summarizes the control requirements for each of the
control options, and Table 5-3 summarizes the emission reductions
achievable in 1982.  The emission factors associated with the control
requirements are discussed in Chapter 3.0, Control Technology.  All the
emission factors, with the exception of storage tanks and evaporative
emissions, are in the format of mg/liter and emissions estimates are
generated by multiplying the factor times the gasoline throughput.
This is also true of storage tank working losses.  However, the storage or
seal losses are determined on a per tank basis, so an estimate of the
number of tanks requiring controls had to be made (see Appendix 3, Section
B.2.2).  Excess evaporative emissions are estimated from vehicle miles
traveled by vehicles with onboard controls.
     The vehicle refueling emission reductions for service stations
in  nonattainment areas were calculated as discussed for terminals.  This
control option for service stations further required underground tank
controls (Stage I at service stations) for those areas which did not
have service station Stage I regulations.  The only areas in the "all
nonattainment area" option where no service station Stage I controls
                                  5-8

-------
 TABLE  5-2.   SUMMARY  OF CONTROL  REQUIREMENTS AND AFFECTED EMISSION
              FACTORS  FOR GASOLINE  MARKETING CONTROL  OPTIONS
Control Option
Control Requirements
    Affected
Emission  Factor3
Assigned  For
Controls, ing/liter
Bulk Terminals

  - Loading Racks
  - Storage Tanks
Bulk Plants
  - No Exemptions
  - Size Exemptions
Service Stations
 (Stage I)

  - Nationwide
      o No  Exemptions

      o Size Exemptions
  - All  Nonattainment
    Areas
1) , Vapor processors  required on all           35
    uncontrolled terminals
2)  Leak-tight tank trucks at all             96
    terminals

1}  Internal  floating roof on all exist-        b
    ing fixed-roof tanks at terminals.
1)  Vapor balance on all incoming loads
      a)  Storage Tank Filling                57.5
      b)  Storage Tank Draining               46
2)  Vapor balance on all outgoing loads        96
1)  Vapor balance on all incoming loads
      a)  Storage Tank Filling                57.5
      b)  Storage Tank Draining               46
2)  Vapor balance on all outgoing loads        96
    at bulk  plants 215,100 liters/day
    (72 percent of throughput)
3)  Submerged  fill on outgoing loads at       600
    bulk plants <15,100 liters/day
    (28 percent of throughput)
1)  Vapor balance system for all               40
    underground tanks
1)  Vapor balance systems for all              40
    underground tanks at service
    stations 237,900 liters/month
    (86  percent of throughput)

Same requirements as nationwide
 See discussion of emission factors  assigned for controls  in Chapter 3.0,
 Control  Technology.
b
 Calculated emission factors  for  internal floating-roof tanks were: 2.4 Mg/yr/tank for
 storage  losses and (7.3285 x 10~8 Q) Mg/yr for working losses where Q is  the  product
 throughput in barrels per year (Ref. 7).
                                        5-9

-------
   TABLE 5-2.   SUMMARY  OF  CONTROL  REQUIREMENTS AND AFFECTED  EMISSION
               FACTORS FOR GASOLINE  MARKETING  CONTROL  OPTIONS
                                     (concluded)
Control  Option
Control  Requirements
   Affected
Emission Factors
Assigned For
Controls, mg/Liter
Vehicle Refueling
  - Stage  II
      o Nationwide
           - No Exemptions
1)  Stage  II vapor recovery  on all
    vehicle refueling operations
    at service stations
                   Vehicle  Refueling
                   Underground Tank
                         Breathing
           - Size Exemptions    1)
      o All Monattainment
        Areas

      o Selected
        Monattainment
  - Onboard
    Stage  II vapor recovery  on all
    vehicle refueling operations at
    non-independent service  stations
        37,900 liters/day and at
    independent service stations
        189,000 liters/day
    (71  percent of throughput)
                   Vehicle  Refueling
                   Underground Tank
                        Breathing
Same requirements as nationwide
Same requirement as nationwide
1}  Carbon canister/modified  fillpipe
    system on all new cars  after 1988.
        Vehicle refueling
        Excess evaporative  control
                                                                          54
                                                                          60
                                                                          54
                                                                          60
                                                                          21.6
                                                                           0.13 g/mi
 See discussion of emission factors assigned for controls in Chapter  3.0,
 Control  Technology.
                                         5-10

-------
      TABLE 5-3.    GASOLINE VAPOR  EMISSION  REDUCTIONS  IN 1982
                    ASSOCIATED  WITH  CONTROL OPTIONS
Control Option
 1982 Baseline
Emissions, Mg/yr
1982 Controlled
Emissions, Mg/yr
a
   Emission
Reduction, Mg/yr
Bulk Terminals
- Loading Racks
- Storage Tanks
Bulk Plants
- No Exemptions
- Size Exemptions
Service Stations
- Stage I
o Nationwide
- No Exemptions
- Size Exemptions
o All NAa Areas
- No Exemptions
- Size Exemptions
- Stage II
o Nationwide
- No Exemptions
- Size Exemptions
o All NAa Areas
- No Exemptions
- Size Exemptions
o Selected NAa Areas
- No Exemptions
- Size Exemptions

140,200
56,300

208,000
208,000



222,200
222 ,200

222,200
222,200


406,900
406,900

406,900
406,900

406,900
406,900

55,800
30,700

76,800
90,300



60,300
118,700

215,500
216,400


31,400
149,200

285,500
320,600

364,000
375,700

84,400
25,500

131,200
117,700

-

161,900
103,500

6,700
5,800


375,500b
257,700b

121,400b
86,300b

42,900b
31,200b
 MA = Nonattainment.

 Includes reductions in breathing loss emissions due to Stage II  controls.
                                      5-11

-------
are required are the 11 counties in the Atlanta, Georgia area and
Maricopa County, Arizona.  All of the areas in the "selected non-
attainment area" option already have Stage I controls installed.  For
the Stage I uncontrolled areas, potential emission reductions were
calculated based upon the throughput for these areas and appropriate
emission factors for Stage I controls.  For those control options which
incorporated size exemptions, the percent throughput for the exempt
category was determined from the model plant size distributions (see
Section 4.1).
     The onboard control option analysis differed from that of the
other control  options.  The onboard control option required the
installation of carbon canisters and fill pipe seals on all  new light-duty
vehicles and trucks sold beginning with the 1988 model year.  As more new
vehicles were manufactured and sold, more emission reductions could be
attributed to the onboard control option.  Gasoline consumption by
controlled vehicles for each year was estimated by using the fuel
economy associated with each model year, the annual mileage accrual
rates, and annual scrappage rates (see Appendix C, Section VIII).
Emission reductions achievable with onboard controls were calculated by
multiplying the gasoline consumption associated with controlled vehicles
by an emission factor representing no vehicle refueling controls (1,080
mg/liter), and then multiplying the uncontrolled emissions by a factor
which represented the control effectiveness of onboard controls (98
percent control).  The numbers of new automobiles, and the scrappage
rates, again discussed in Appendix C, were used to determine the number
of controlled vehicles in each year.  Table 5-4 summarizes the results
for the onboard emission reduction analysis.  Emission reductions are
shown for  both control of vehicle refueling only and for the combination
of refueling and expected excess evaporative emission control.  Emission
reductions for vehicle refueling operations are dependent on the gasoline
consumption by onboard controlled vehicles, while the evaporative
emission reductions are dependent on the vehicle miles traveled (VMT)
by onboard controlled vehicles.  Aggregate emission reductions increase
steadily as more new vehicles are required to have onboard controls.
                                  5-12

-------
 Table 5-4.  SUMMARY OF PROJECTED TOTAL ONBOARD EMISSION IMPACTS IN EACH

                      YEAR OF THE STUDY (1988-2020)
Projected
Year
1988
1989
1990
1991
1992
1993
1994
1995
1996
1997
1998
1999
2000
2001
2002
2003
2004
2005
2006
2007
2008
2009
2010
2011
2012
2013
2014
2015
2016
2017
2018
2019
2020
New
Vehicles
(million)
13.35
13.30
13.13
13.48
13.68
13.68
13.60
13.76
13.76
13.76
13.76
13.76
13.76
13.76
13.76
13.76
13.76
13.76
13.76
13.76
13.76
13.76
13.76
13.76
13.76
13.76
13.76
13.76
13.76
13.76
13.76
13.76
13.76
Gasoline
Consumption
by
Controlled Vehicles3
(109 Liters)
33
62
87
110
131
148
164
177
189
198
206
213
218
223
225
227
229
230
231
231
231
231
231
231
231
231
231
231
231
231
231
231
231
VOC
Emission
Reductions3'15
(103 Mg)
35
65
92
117
138
157
173
188
200
210
218
226
231
236
238
240
242
243
244
244
244
244
244
244
244
244
244
244
244
244
244
244
244
VOC
Emission
Reductions
(with E~vap.)a'c
(103 Mg)
61
117
167
214
256
294
327
356
382
404
423
-437
450
455
458
460
462
462
463
464
464
464
464
464
464
464
464
464
464
464
464
464
464
 No  tampering.

)
 From  vehicle refueling  operations  only.


'From  vehicle refueling  operations  and estimated control  of excess
 evaporative  emissions.
                                  5-13

-------
     5.1.3.2  Cumulative and Discounted Emission Reductions.  The
emission reduction analysis evaluated the impacts of the control
options and regulatory strategies for each year from 1986 through the
year 2020.  Emission reductions for each option were calculated for
each year and cumulative and discounted emission reductions were
generated.  The emission reductions in each year of the study period
can be found in Appendix E.
     As discussed in Section 4.2, gasoline consumption was projected to
decrease substantially during the years 1986 to 2000.  Leaded gasoline
was projected to decrease even more rapidly due to lead phase-down
regulations.  Tables 5-5 and 5-6 present the percentage decreases in
total gasoline consumption and leaded gasoline consumption, respectively,
which were projected to the year 2000 from Figure 4-2.  The analysis
further assumed that the gasoline consumption would remain constant
after the year 2000.
     The emission reduction calculations are directly proportional
to gasoline throughput since the reductions are based on consumption
and emission factors.  To develop emission reductions over time,
therefore, the base year emission reductions for gasoline vapors and
benzene were decreased each year at the same rate as the total gasoline
consumption was projected to decrease.  For example, if gasoline
consumption did not change, the gasoline vapor emission reductions from
bulk terminals in 1990 would be the same as for 1982, or 84,400 Mg/yr.
However, the 1990 consumption is only 75.7 percent of the 1982 consumption.
The gasoline vapor emission reduction achievable in 1990 was, therefore,
estimated as only 75.7 percent of the 84,400 Mg/yr, or 63,900 Mg/yr.
This type of calculation was repeated for each control  option for each
year of the analysis.
     EDB and EDC emission reductions can only be achieved from
leaded gasoline, therefore the leaded gasoline projections were
used to estimate EDB and EDC emission reductions in a given year.
The EDB and EDC emission reductions were obtained by multiplying
the gasoline vapor emission reduction for a particular year by the
EDB or EDC emission ratio (see Section 2.2)  and by the percent of
total consumption in that year which was leaded gasoline.
                              5-14

-------
           Table 5-5.  PROJECTED TOTAL GASOLINE CONSUMPTION CHANGES
                              FROM THE BASE YEAR
              Consumption3
           109 Liters   109 Gal
% of 1982 Consumption   % Decrease
1982
1983
1984
1985
1986
1987
1988
1989
1990
1991
1992
1993
1994
1995
1996
1997
1998
1999
2000-2020
389
364
349
338
326
317
309
300
294
288
284
280
273
269
265
261
257
254
250
102. 7a
96. lb
92.3
89.2
86.1
83.8
81.5
79.2
77.7
76C
75
74
72
71
70
69
68
67
66
100
93.6
89.7
86.9
83.8
81.6
79.4
77.1
75.7
74.0
73.0
72.1
70.1
69.1
68.2
67.2
66.2
65.2
64.3
0
6.4
10.3
'13.1
16.2
18.4
20.6
22.9
24.3
26.0
27.0
27.9
29.9
30.9
31.8
32.8
33.8
34.8
35.7
 1982  consumption  from  Reference  1.

)
 1983-1990  consumption  from  Reference 2.


'1991-2020  consumption  estimated  from Figure 4-2
                                      5-15

-------
               Table 5-6.  PROJECTED LEADED GASOLINE CONSUMPTION
Total Gasoline
Consumption
109 Liters 109
1983
1984
1985
1986
1987
1988
1989
1990
1991
1992
1993
1994
1995
1996
1997
1998
1999
2000-2020
364
349
338
326
317
309
300
294
288
284
280
273
269
265
261
257
254
250
96.
92.
89.
86.
83.
81.
79.
77.
76^
75
74
72
71
70
69
68
67
66
Gal.
1*
3
2
1
8
5
2
7
i









Leaded Gasoline
Consumption
109 Liters 109 GAL.
158
134
112
96
84
74
64
56
49
42
34
28
23
19
15
11
8
4
41.79
35.4
29.7
25.3
22.1
19.5
17.0
14.7
13t>
11
9
7.5
6
5
4
3
2
1
% of Total
Gasoline
Consumption
43.4
38.4
33.3
29.4
26.4
23.9
21.5
18.9
17.1
14.7
12.2
10.4
8.5
7.1
5.8
4.4
3.0
1.5
1983-1990 consumption values from Reference 2.
1991-2020 consumption values estimated from Figure 4-2.
                                5-16

-------
     Another  factor which affected the emission reduction in the
 initial years of the analysis was the phase-in of control equipment.
 Obviously, the total emission reduction achievable in a year is
 dependent upon the number of facilities that had control equipment
 installed.  Therefore, the emission reductions for the initial  years of
 the analysis  took into account the percent of facilities which had
 equipment installed in that year (see Section 5.1.1 on facility phase-in
 rates).
     Table 5-7 summarizes the cumulative emission reductions (simple
 sum of all reductions in all years) and the 1986 net present value
 (NPV), or discounted, emission reductions (discounted at 10 percent).
 All discounting was brought back to a NPV in 1986 since this was
 the earliest year any of the regulatory strategies took effect.
     The onboard cumulative and discounted emission reductions were
 developed from the data presented in Table 5-4.  After phase-in,
 all of the other control  options are affected only by the reduction in
 gasoline consumption.  The onboard control option, on the other
 hand, does not cover the entire vehicle fleet until  about the year
2002 or 2003 due to the phase-in of new controlled vehicles and the
 retirement of existing uncontrolled vehicles.   However, once onboard
controls are in place for the entire fleet,  greater emissions
reductions can be obtained with this option  when compared with  Stage
 II options because of the higher control  efficiencies involved.
     The nationwide emission reductions based on theoretical -
efficiencies associated with the regulatory  strategies, discussed in
Section 4.3, are, in most cases,  simply a combination of the emission
reductions for the appropriate control  options.  This is true for all
cases except those which  combine Stage  II and onboard controls.  When
Stage II and onboard coincide at a vehicle refueling operation, the
onboard fill pipe seal  does not allow vapors  to enter the Stage  II
system.  The emission reductions for this operation are then associated
only with the onboard controls.   Therefore,  a  separate analysis was
performed for Stage II  emission reductions when combined with onboard
controls.   Emission reductions for Stage  II  were attributed only to
that percent of the throughput (and therefore that percent of the
                                  5-17

-------




I
» '
fe-
O-
1 =

to
00
as
o
HH
o.
o
1
o
o
m OCJ cor*. 0*0 oo iP^- 2^ . vm oo
PH CM »-* *—«




_
So oo oo oo oo oo S2 S 52
o oo oo entn oo S oo oo
"^ **^ ^,e** » *» •* •• * * »•
fr> ««< Vm U>O tf> ^ CM •-< cn r**

"
g g || |g || ' II §s §| II
VH m r-» ui v^co *r r«- coco io<«f' ID ID

I I §1 II §1 II §| §§ I I
cy\m torn u>to cr* CM couo ou) r* CM ,— « o
Sew oca cnm *r«a- co«5 r^-^ ocw ~S
tr» CM oo r-» a* to CM tn cou> mcsi ««- to^

§o oo oo oo oo oo oo oo
o oo oo oo oo oo oo oo
^> (\j vO CO W ^O ^^ f ^"* ^ *^ *"** *" ^' *" ®
coco com i«csj «-««-i ^ ^L ^^ cnr-». ^^
*-T CM* CM* m^CsT CO U) CM -H VD m

« , 	 l l , ra CT
 cu 4> o.
•** "c oS c Si'«*^iU3 •a^iL**^i"Z^i" Ui-5
J^ nj *^CLO o Q.V o o. a» **- •** o. S << o. £ ^ a. § o> QJ uj
o= ^"^ cxw *32 ouj < OJLU a: <-• c - o •«— o T- •*- cTifl 0*1—^- o *t— a) o<*» o :^ >•
« *o o >— Nat*JZcOf— 3e(/i^a«zzco o
I 1 f- 1 1 t. 1 1 3 ' '
)— GQ 
c: 4J
C3 O *
" -i S
t~ 01
«3 k C
U O
— e
** o *n
 o
u
c: (/)
o « -o
T3 3 W
0 r- O
10 U .a

-------
emission reduction) in each year which was not controlled by onboard
systems.  Furthermore, it was assumed that Stage II systems were
not replaced after one useful life of 15 years (two useful lives of
8 years for vacuum assist systems) because by the time this would
be required, the entire vehicle fleet would be controlled by onboard
systems.  The combination of Stage II and onboard systems allows
for the more immediate emission reduction achievable by installed
Stage II systems while the onboard systems are gradually controlling
the vehicle fleet.  Table 5-8 summarizes the cumulative and discounted
emission reductions associated with Stage II controls when combined
with onboard control options.
     Using Tables 5-7 and 5'-8, the emission reductions associated
with the regulatory strategies can be determined.  Table 5-9 summarizes
these emission reduction impacts.  Emission reductions are presented
for gasoline vapors, benzene, EDB, and EDC.
     As a further comparison between Stage II and onboard control
options, an analysis was performed to evaluate the emission reductions
which could be achieved considering in-use efficiencies rather than
theoretical  efficiencies.  In-use efficiencies take into account
tampering and deterioration which occur during actual  use of the
equipment.   Section 3.7.3 and Appendix D discusses in-use efficien-
cies in greater detail.  Table 5-10 summarizes the emission reductions
for nationwide Stage II control  options based upon in-use efficiencies
associated with annual  inspections (86 percent)  and minimal" enforce-
ment (56 percent).  These emission reductions can be compared to the
nationwide emission reductions obtained from onboard controls,
considering in-use efficiencies (87 percent).
5.2  OTHER ENVIRONMENTAL IMPACTS
5.2.1   Hater Pollution Impacts
     The overall  impact on water resources is negligible.   Only
refrigeration systems for bulk terminal  control,  which cool  and
condense the vapors from the loading operation for liquid recovery,
create a potential water pollution impact.   As the vapor-air mixture
collected at the loading rack is cooled,  a small  amount of gasoline-water
mixture is generated.   The amount of water generated is dependent
                                5-19

-------




*
o
eC
O
ca
o
z
H- <
o
UJ
"Si
CO
y^
O
o

5E
UJ

CO
^^
t—»
S


^^J
H»4
UJ
•c


CO
El
o

U.
CO
ac
0
l_
0

o
UJ
Q£
_
Mb
O
I— 1
CO
CO


Sf?
UJ

*
CO

if)
UJ
_i
CO
h-





















••^
.X
f
• tn
C
o
*J
1

I/I
c
o
I/I
1
UJ




























1
vo
co
en
•-4

g
UJ

0
g
O)
LU



i




a.
z
vo
CO
en


1
N
C
CU
CO


£
U






^^
O-
z
CO
en


cu
5 u
— o
o o. ,
tn IB
IB >•
•°

i









0)
o.
'-
£>
1"^
,—



lfr
o


0 O O O
en vo to PO






o o o o
PO en to «s-

en vo mm




to en 10 ^-
^^




o o o o
CM CM CO O
W CM CM







§o o o
o o o

oi~T vr CM"







o o o o
- 0 0 0 O
P. P. P. P.
co o en —<
PO «r CM o
en vo «a- PO








o o o o
0 § gg
•> * * •»
CM CO «3" CO
vo vo po «r
in o vo «»•
•M 1-4



in en
tn o tn o
C •»- C 1-
O 4-> O 4-> U)
•^ O. •»- O. IB
*J B *• E CU
0.0) Cu CU U
0^ UJ tn oi UJ
X IB X «C
oi oi cu 01 eu 01 z
-O N U N
4- o-- «e 01- -o
3: z 1*0 z c-o oi

O 1 1 Z II U
— cu
4J r— i—
1C r» CU
=e «C 
VO CM
*H *•*






S2

CM >4




CM CM



,

O CM
*•(







§o
u>S

«- T^i







o o
o o
°.°«
CM en
U) O









o o
o o
0 0
•t A
•» CM
CM VO
CM -H




in
in o

O •!->
T- a.
•U S
o. o>
O) UJ
X
Ol Ol
N
o ^


1 1













cu
en
IB
in
cu
IS
n
§•
0.
o.
IB
cu
*»
en
c
•5
to


.0

cu
IB
*->

0
cu
L.
IB
tn
CU
*-> •
IB r^
C 1
ULO
4-> CU
15 3
IB
IB e
O-t-
"c-o
O 1-

tn o
p2"e
o. o


i— i o
V4»
CU
en in
IB CU
<-> 3
IB
E >
o
L. CU
*4" «C

e o
0 *>
u ^«
T3 IB
0) 
*
5-20

-------






.
UJ
as —
i— i o
_1 CM
00
CO CM
< 1
CD ^D
CO
SE: en
O fH
a.*""
CO
CO UJ
O C9
t-H UJ
|— (_•
3 OH
0 1—
UJ CO
^^
Z Of
00
1— 1 ^
lo<
2 =
UJ UJ
UJ
O CD
•-i Z
3 i— i
ZZ J"~"
O UJ
p ee
z=:
•





UJ
«J
CO
H-




































L.
^

en
Z
tA
o
*>
u
3
T3
 Q.
J


""-"••











£-8
O •*•
* s*
=3  l_
0> 4-»


X
X

»
7*


i
LS
1.
o
-i






3
U




»
L>
to
CO
rj





"
3
«i






o
SO
Cft
P*







i
=1
J
i— —



















tOO OO OO OC3 OOO CM IO OO OO OO OO
••* CM en TV. ~* *r en «a- Tl ^* *•* ^i ^ en — • «o- men toen
»-«—* ••* CM CM *•* *•« «•* «•«




§*"*
oo oo oo oo oo o u> o oo oo oo oo
r*. co ^- co ~H T* !£* r-» in co -H »-* CM unr-*- ^N tn IOCM en co
^ »-< CM CM -H CM CO in CM CM CM CO »-H
CM CM tn to <-*co en co o r- o -M CM com — 4 ro ^co toen
-^-N — CM CM —...4 SS






'S ' " '
to co«a- i^^ om r^un r^oo CMCM «>to Oin tn— « enco
— * -H CMCM -*CM COLO CMCM CM CO *H




§o oo oo oo oo oo oo oo oo oo
O OO OO OO OOO OO OO OO OO OO
«>o coo oo oo ooo oo oo oo oo oo
• » •» *• mm **(«•) • * •.» •* mm mm
•HCM *f \o coin om coo * oo CMCM cMin ^"co vio
_« ^N *-4*^ CMcoen »MVI ~-t^« CMCM CMCM »H«H





o oo oo oo oo oo .00 oo CD o oo oo
o oo oo oo oo oo o oo oo oo oo oo
O f^ W O O O O O O O O O O O O O O O • O O O O
to vto «^ot tom r+* ^ co en » *o r^. en^n *HO into CM vo
r^ ^^*4 «»• in co m coo in vv ^ in en o en o in in
^* — « *er — < ^«




OO OO OO OO OO OO OO OO OO OO
oo oo oo oo oo o oo oo oo oo oo
O O O O m^ OO OOO OO OO OO OO OO
SO OO OO OO OO - OO OO OO OO OO
— * tOCM OO OO OOO OO OO OO OO OO
CM co toen —i in toco r* co o into co en in o en «r *-» *r

CMCM *-HCM CO«r * »**-4 *-4«M CO W CO ^^ *^cn om
co ^^ CMCO r*»en inco cor*. * ^.r*. r*^ r^ ^vo into coco
CM -^ .— 1 to -M «-t i— t ^H
*™ •
tO IA 4A I/I i— « to «O (A (A I/) M^tA
c c c c •— ' e= c*ec-oc«*j-oc: s- c
* O to O tn O(A OtA O in * O tn OtAUOtA t_ O CA rtj o co
.^-C • ^- C -^C f-C tt)-^EZ • ^E • —C 10-^-C »«0«*-C 0-^C
< *^ O < 4-» O •*-» O 4->O C04-» O *C 4J O «C 4J O O4^O  CX 03 £X VI Qj CL (U CX QJ CL O tt) CX O (I) CX O) &
1XE 1XE X E x E x E t x E i X E X E > x E •« X E
3)01 4> d> CJ QJ A> 4> *O 4J CJ V (U O fl) *0 O O) ^) O> 4J CJ S
 «^- O O*i—O fQ O) to f— O 0><0-^O O) ••— O 0)  cn co c cntnc cntAC cntACZ cnvic o en o to c cnotAC OIAC cncncnc en en cz

CO CO«CCO CO. * .. *

0-t 1— * >~« > «— 1 1—4 I—I >< I_H

^






«

tn
IA C>

m •*-»
C (0
*° o
0) Q.
C (O

•— o>
0)
tA ^»
jo e
o> •»-
£ £
C T7
•^ (Q
•0 Ol
0) 7
fe =
^3 O
•^ cz
IA •»-
c
O •*->
0 0
O)
tA in
o o
5 a
c *
JZJ O
(A tA
r— C
S 5
i 2
u o»
CX
o o
= »
X d)
o «*-
CM O)
0 I-
CM
o>
S 0
i- .CZ
O) >

(A IA U
tA •*- Z3
•*» E ^
E eg cu
O> 1.
IA
at i- c
C 0) O
*^ ? *tn
O tA Vt
(A C —

ns03 js° *
5-21

-------
UJ
o
to
§
ex
U_ (/)
   UJ
OO »-4
=C O
O Ul
a co
UJ
ce: >-^-
   o:-a
s: o a
o H- *a
•-< «c s
CO _lr-
00 SP O
J-J O C
ac uj o
UJ OS O

UJ CO*"*
Q =
t-* 1—4

=£ UJ
O *£
IT)

UJ
i
i

i


*

-
IA
|
•b
cc
c
IA
' £










!
!

1







i
M
C
CD




i!
n 10

in
13
I
fc.
4-1
(O
o
10
1
tX
L
§
2

B

§
71
i

i
|



i
»
1

1




O O O O
•-H CM


gg 00
ro m

r- Cvj OO
— CM
cO£> °°

§1 i§
cu m csi eg


O o oo
o o o o
cb*-^ ir>*-^
Ol ^M CO ^^
§§ °°
o
S8
co en
O O o o
il
*-• e c
e o o
O) -^ ^ •*-
en IA 4-* *»
•0 C 0) (JO
4J O V) C 33
CO •*- C Cl -O ^3
•*•» O N O O
•Q Q.-*- C i- t_
^ E ** 0*
i-t t. O) O. COO NN
« X E C CO CO
O OJ  ^~
cncotuj =so *r c*-)
 • >
4j iv o QJ re oj *-H
V) «O W Z Di U VO CO
~  -C co
•-* X
X

                                                              5-22

-------





























o
oo
h— I
rf
«C OO
Q. UJ
o a: o
f ^ a^^ *T?>
O UJ
•Z. CD i-<
O 2 O
•— 1 O I-H
H- U-
0 Q U_
Q <
UJ UJ
ce H-. oo
H^ 1H3
Z 1
O UJ 2
i— < <_S i— i
OO — i oo z
S 1-1
uj 32 a;
UJ UJ
• UJ Q
O 3 t-i
~-< t— oo
1 UJ =Z
LO CO O
0
UJ

CO

r— . un o L^) c\j-., «* -
.*-4- «-r —4 ou evj «-«
•
o o o o
csi «~H PO ^r ^3 r*^>
CTt CO Lf> CO CM f**-
*-4 t-l



.




O O O O
'r-. ' u> o uo oo ^tf-
-H ^H —4 OJ CSJ — <






§00 ooo oooo
oo ooo oooo
ro ro ,o , , o ro co PO p** r-. LO
ocn^fi LOCOCO air»-coio
. . ; . - . ,
_



OOO OO-O OOOO
ooo ooo oooo
cst COO co I-HO ^r-x *— icr»
*»« *.* ».*«*
f*1- OOCM «^ C^OJ. tn\jO CMCft
""CM LO^-PO <*CO^ r«fc






O'OO OOO OOOO
o oo o oo oo oo
ooo ooo oooo
o o o ooo •00*0*0'
o o -o o oo " ocr» oo
*J3^CTi CO *— 4 rO ^"VOfO LO
^^ *~* CM CM *— l T— t CM »-4 * CM



ooo ooo oooo
ooo ooo oooo
O OO O_OO O'O O"O

ooo ooo oooo
ooo ooo oooo
f"» . i— 4 PO ro vocTi cr> i— i ^-^-1
LOLOPO COr-"** ^OPOVOCM





• '

U^liO^O LO(JO^ COCOCMCNJ
01 COtT) cn'COLO S S . S! - Si
(tj
Ol CTi
y «^- .,_
i— cn i— a>
 • £ ^ t^ ^~
"oc ccr ccz ooj O)O)
•^ O i— OOJtrtr— OQJ i— Q£3 rv* -s

C4-*o ^-> S QJ O C_) J-> E d> UOOJ (UOJ
0 °"£ r-U3(_>-^'^ ^030 -^-i— C£ »— Ctf
*J <1J flJ OJ =S CX-r- O CL (U OJ =3 CX-^ O OJt-.CX flj.f-.cL

s: UJ o^ cc-^-c:  c^ cc-r-c OGJOJ> 3Cj

•— t a; j= c "^ UJJTC >•* ui ^>->uj^>.>uj
— 'Ml— »-H| 1 H- >— . i | -^j—,1 _,,
•i- o S-
OJOOOO 2:OO TOO O
GO 0







































cn
^
^
OJ
E
*o
o
' W)
*c s-
o
•a
cn 
cn •  u
g OJ
S- U
ll ,^
u-
C OJ
o
•r- C
•P re
= cn
•a o
OJ 5-
i- O.

CZ Ul
O 4->
to OJ
li ^

-------
upon the relative humidity of the atmosphere.  The mixture passes
through a gasoline-water separator, with the gasoline returning to
storage and the water being discharged.  It is estimated that this will
produce only a negligible impact on water quality.
5.2.2  Solid Waste Impacts
     None of the control techniques evaluated generate a solid waste as
a by-product of its operation.  The only solid waste that may be gener-
ated under a worst case assumption would be the carbon used in the bulk
terminal vapor recovery systems and the onboard canister systems.  EPA
has determined in a previous study that, even under the worst case
assumption that the carbon cannot be reused, the resultant solid waste
impact from bulk terminal controls would be negligible.3  If it is
assumed that the carbon in the onboard systems must be discarded after
its useful life, it is estimated that 3 pounds of carbon would be
discarded with every vehicle.4  This is negligible when compared to the
mass of solid waste generated by discarding the vehicles themselves.
5.2.3  Other Environmental Impacts
     Other potential environmental impacts include noise, space
requirements, and availability of resources.  The relative impacts of
the regulatory strategies on these environmental concerns is expected to
be insignificant.  An EPA test showed that the noise level from terminal
vapor processing devices, which created significantly more noise to the
unprotected ear than any other system considered, was less than 70 db
at 7 meters from the noise source.5  The strategies cause only minimum
impacts due to space requirements and resources availability.
5.3  ENERGY IMPACTS
     Energy impacts for the strategies are estimated in the form of
liters of gasoline saved.  Energy savings were estimated from the
recovery credits assigned to the different industry sectors.  Table 5-11
indicates the gasoline savings ratios for bulk terminals, bulk plants,
and service stations.  Gasoline is recovered at terminals where carbon
absorption or refrigeration systems are used.  The ratio of return for
terminals (1.1 liters of gasoline recovered/1000 liters of gasoline
transferred) takes into account the energy consumed by the vapor
processing systems.  Gasoline is recovered, or not lost to evaporation,
at bulk plants where vapor recovery is used on outgoing loads.  When
                                  5-24

-------
Control
Option
                    TABLE 5-11.   GASOLINE  RECOVERY  RATIOS
Net Energy Savings (Liters of Gasoline
  Recovered/1000 Liters Transferred)
Bulk Terminalsa
Bulk Plants'3
Service Station (Stage II)C
Storage Tanks
                      1.1
                      0.6
                      1.5
                       d
 Reference 6.
 Table 7-6 and 7-7.
'Table 7-15.
 Net Energy Savings  = 60,250 liters/year/tank  (Table 7-5)
                                  5-25

-------
gasoline is pumped from storage to fill the trucks, vapors
are returned to the tank, thereby reducing evaporation and saving
gasoline.  This same operation occurs at service stations when
Stage II vapor recovery returns vapors to the service station
underground storage tank.
     Table 5-12 estimates the cumulative energy credits associated
with the regulatory strategies, in units of liters of gasoline
saved, from 1986-2020.  No gasoline is recovered or saved under the
onboard control scenario, even though substantial emission reductions
are achieved.  The gasoline vapors are absorbed in the carbon
canister and burned in the engine.  No energy or recovery credit is
given to this operation because the added weight of the canister is
assumed to offset any added fuel economy (see Section 7.2.5.3 and
Appendix C).
                               5-26

-------
                TABLE  5-12.   ENERGY  SAVINGS  ASSOCIATED WITH
                  GASOLINE MARKETING REGULATORY  STRATEGIES
      Control  Alternative

   I.   Baseline

  II.   Stage II-NA*
       -  Size Exemptions
       -  No Exemptions

 III.   Stage II-NA
       -  Size Exemptions
       -  No Exemptions

  IV.   Stage I
       -  Size Exemptions
       -  No Exemptions

   V.   Stage II
       -  Size Exemptions
       -  No Exemptions

  VI.   Stage I & Stage  II
       -  Size Exemptions
       -  No Exemptions

 VII.   Onboard

VIII.   Stage II-NA* & Onboard
       -  Size Exemptions
       -  No Exemptions

  IX.   Stage II-NA & Onboard
       -  Size Exemptions
       -  No Exemptions

   X.   Stage I & Onboard
       -  Size Exemptions
       -  No Exemptions

  XI.   Stage II-NA & Stage I & Onboard
       -  Size Exemptions
       -  No Exemptions

 XII.   Stage II & Onboard
       - Size Exemptions
       - No Exemptions

XIII.   Stage I & Stage  II & Onboard
       - Size Exemptions
       - No Exemptions
Gasoline Savings (1986-2020)
      (109 Liters)
              1.0
              1.4
              2.8
              3.9
              4.9
              5.1
              8.1
             11.4
             13.0
             16.5

              0
              0.2
              0.3
              0.6
              0.9
              4.9
              5.1
               5.5
               6.0
               1.5
               2.1
               6.4
               7.2
                                5-27

-------
 5.4  REFERENCES

1.   National Petroleum News.  1983 Factbook Issue.  Mid-June 1983,
     Volume 75, No. 7A.  p. 80.

2.   U.S. Environmental Protection Agency.  Federal Register, Vol.-47,
    .Number 210, October 19, 1982.  p. 49329":

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

4.  Cost Comparison for Stage II and On-Board Control of Refueling
    Emissions.  American Petroleum Institute.  Washington, D.C.
    January 1984.  Appendix III, p. 2.

5.  Betz Environmental Engineers, Incorporated.  Gasoline Vapor
    Recovery Efficiency Testing at Bulk Transfer Terminals Performed
    at Pasco-Denver Products Termtnal.  U.S. Environmental Protection
    Agency.  Research Triangle Park, N.C.  Contract No. 68-02-1407.
    Project No. 76-GAS-17.  September 1976.  97 p.

6.  Reference 3, p. 7-8.

7.  Memorandum from Gschwandtner, K.,  Pacific Environmental Services,
    Inc. to Shedd, S.A., Environmental Protection Agency.  March 16,
    1984.  Internal Floating - Roof Emission Factor.
                                5-28

-------
                   6.0  EXPOSURE/HEALTH-RISK ANALYSIS

     This chapter outlines the methodology, assumptions, and results of
the health risk analysis for the gasoline marketing industry.  Exposure
and risk estimates were developed for four pollutants emitted by the
industry:  benzene (Bz), ethylene dibromide (EDB), ethylene dichloride
(EDO, and gasoline vapors (GV).  These substances were chosen for the
analysis based on the availability of quantitative unit risk factors
for carcinogenicity.  The availability of unit risk factors, derived
from human and animal research studies, permits the quantitative
evaluation of health risks.
     Section 6.1 discusses the unit risk factors and types of health
hazards associated with each of the four pollutants.  Section 6.2
outlines the methodology and assumptions used to estimate the annual
nationwide incidence (occurrences) of cancer expected to result from
the projected exposure and to estimate the lifetime risk (probability
of an incidence over a 70-year lifetime) that would result from a
realistically high exposure scenario.  The incidence and lifetime risk
attributable to each of the four pollutants were estimated for each of
the following four source categories: bulk terminals, bulk plants,
service stations, and self-service refueling and then were projected for
the years from 1986-2020.  Section 6.3 presents the cumulative incidence
over 35 years and the lifetime risk from 70 years exposure to expected
emissions from each source category under the implementation of each
regulatory strategy examined.
6.1  UNIT RISK FACTORS
     The unit risk factor for an air pollutant is defined as the proba-
bility of getting cancer as a result of continuous exposure for a
lifetime (70 years) to a unit concentration of the agent.  Derivation
of a unit risk factor permits quantitative estimates of the health
risks for exposed populations.  Table 6-1 lists the four pollutants of
interest with their respective unit risk factors, and a brief health
effect summary.
                                  6-1

-------
                          TABLE  6-1.   UNIT  RISK  FACTOR SUMMARY
  Compound
       Unit
(probability of cancer
 given lifetime exposure
      to 1 ppm)
                                                     Health Effects
                                                        Summary
                                Comments
Gasoline Vapors

  Plausible Upper Limit:b

    Rat Studies
    Mice Studies
      3.5 x 10-3
      2.1 x lO-3
  Maximum Likelihood Estimates;b
    Rat Studies
    Mice Studies
Benzene
Ethylene
  01bromide

Ethylene
  01 chloride
      2.0 x 10-3
      1.4 x 10-3

      2.2 x 10-2
      4.2 x 10-1


      2.8 x 10-2
                              Kidney  tumors  in
                              rats, liver  tumors
                              in mice.
Human evidence of
leukeraogenicity.
Zymbal  gland
tumors in rats;
lymphoid and other
cancers in mice.
Evidence of carci-
nogenicity in
animals by inhalation
and gavage.  Rats:
nasal tumors; Mice:
liver tumors.

Evidence of carci-
nogenicity in animals.
Rats:  ci rculatory
system, forestomach,
skin, and mammary
glands.  Mice:  liver,
lung, mammary gland
and uterus.
                          Gasoline test samples in
                          the animal studies were
                          completely volatilized,
                          therefore may not be
                          representative of ambient
                          exposures.
EPA listed and regulated
as a hazardous air
pollutant.  IARCC:
Sufficient evidence  to
support a causal  associ-
ation between exposure
and cancer in humans.

EPA suspect human carci-
nogen; recent restrictions
on pesticidal uses.
                                                                               EPA:   Suspect human carcinogen.
                                                                               Draft  health assessment document
                                                                               released  for external review
                                                                               March  1984.
 aUn1t Risk Factor is  the probability of a cancer incidence (occurrence)  per 70-yr lifetime exposure
  to 1 PPH.

  In the case of gas vapors, the point estimate unit risk factor represents the slope  of  best fit
  fit linear regression  of  the data through the origin.  The plausible upper limit unit risk' factor
  represents the maximizing of the slope (wwhich is the unit risk factor) while remaining within
  the 95 percent confidence interval.

 CIARC:  International Agency for Research on Cancer.
                                                 6-2

-------
     As noted in the table, benzene is the only pollutant of interest
for which there is sufficient evidence derived from human epidemiolog-
ical studies to support a causal  association between exposure and
cancer.  Although there is limited health evidence from studies of
occupational populations exposed to gasoline vapors in the gasoline
marketing system, the evidence is not sufficient for use in risk estimation,
The unit risk factors for gasoline vapors, ethylene dibromide, and
ethylene dichloride are based entirely upon animal studies.
6.1.1  Credibility of Risk Estimates
     As with all unit risk estimates, these values were derived using
several standard assumptions in the absence of information to the
contrary.  The major assumptions are:
     1)   The agent is a human carcinogen.
     2)   The linear dose-response relationship is plausible as a means
          of estimating the risks associated with the small doses
          typically occurring in the environment.
     3)   In the absence of human data on carcinogenicity:
          a)  Animal bioassays are appropriate for human risk estimation.
          b)  Humans are as sensitive as the most sensitive animal
              species.
          c)  If the route of administration is not appropriate to the
              human exposure situation (i.e., gavage), equal fractional
              uptakes are assumed via the two exposure routes (i.e.,
              inhalation or gavage).
For each of the agents, these general assumptions are discussed in the
following sections.
     6.1.1.1 Benzene.  An association between benzene exposure and
leukemia has been documented in several human studies of occupationally
exposed populations.  Benzene has also been found to be carcinogenic in
both rats and mice by gavage and inhalation routes of exposure.  The
benzene unit risk factor (the risk of cancer resulting from a 70-year
lifetime of exposure to a unit concentration) was derived  from the
average of three occupational studies, assuming a linear dose-response
function.  A unit ractor risk derived from the animal data is very
close  to the value derived from the human studies thereby  indicating a
similar dose-response relationship.
                                  6-3

-------
     6.1.1.2  Unleaded Gasoline.  The evidence of carcinogenicity
comes primarily from the American Petroleum Institute chronic inhalation
study of unleaded gasoline vapor in rats and mice.  The unit risk
estimates for each species based on a linear nonthreshold dose
extrapolation were derived from this study.  Although API studied"
unleaded gasoline, other gasoline grades (e.g., leaded gasoline) are
expected to have as much carcinogenic potency.
     6.1.1.3  Ethylene Dichloride (EDC).  No human evidence of
carcinogenicity is available.  The animal evidence consists of positive
responses at several sites in male rats and mice via gavage.  The unit
risk for EDC inhalation was estimated by two separate methods:  (1) a
direct estimation based on the EDC gavage study, assuming that the
absorption rate by inhalation is one-third that by the oral route;  and
(2) an indirect estimation from the EDB inhalation study.  The potencies
calculated from both approaches are similar.
     6.1.1.4  Ethylene Pi bromide.  No human evidence of carcinogenicity
is available.  The animal evidence consists of positive responses in
mice, in both inhalation and gavage bioassays, as well as nasal  cavity
tumors in rats following inhalation exposure.   The unit risk was obtained
from the rat inhalation experiment using the linear dose-response
extrapolation procedure.
6.2  EXPOSURE AND RISK METHODOLOGY AND ASSUMPTIONS
     This section briefly outlines the methodology and assumptions  used
to estimate the concentrations of benzene, EDB, EDC, and gasoline vapors,
as well as their associated health risks, from each source category
to which the nation's population as a whole and to which selected
individuals subject to high exposures would be expected to be exposed.
Estimates were made and projected to 1986 through 2020 for exposures
due to emissions from each of the three source categories of bulk
terminals, bulk plants, and service stations.   Both ambient exposures
to the public around service stations from all  vehicle refueling and
individual exposures during self-service vehicle refueling were estimated.
(Although the emissions appear to be double counted, the exposures  and
risks are not double counted because the service station estimate
considers the emissions dispersing to area residents off the station
premises while the self-service estimate considers the emissions near
                               6-4

-------
the source when the public comes to the station.)  The emission sources
considered for each of the exposure categories are listed in Table 6-2.
Further details of the methodology and assumptions used to estimate the
various risks are given in the docket (Docket No.  A-84-07).
     The estimates of risk, in terms of individual lifetime risk from
high exposure and aggregate incidence, are applicable to the public in
the vicinity of gasoline marketing sources and those persons who refuel
their vehicles at self-service pumps.  This analysis did not examine
the risk to workers from occupational exposure (e.g., terminal  operators
and service station attendants).  The lifetime risk from high exposure
for these workers is probably substantially higher than for the general
public.  In addition, the estimates of aggregate incidence would be
higher if such worker populations were included in the analysis.  Of
course, any controls to reduce gasoline marketing emissions would reduce
exposure for workers as well  as for the general public.
6.2.1  General Assumptions
     Several basic assumptions underlie the estimation and projection of
exposure and risk for the gasoline marketing industry.  The same
assumptions were made as in the emissions and cost analyses, regarding
baseline control levels, gasoline throughputs for each source category
and gasoline type (decreasing until 2000 and constant thereafter),
exempted throughputs, emission factors, onboard control start-up and
phase-in, phase-in of Stage I and Stage II controls, and phase-out of
Stage II controls when in combination with onboard controls.  Generally,
the risk estimates due to benzene were calculated for a base year.  The
risk due to the other pollutants was then calculated by multiplying by
a factor incorporating the ratio of emissions (from either a storage
tank or a vehicle tank, since the vapor temperature and, thus,  emissions
differ) and the ratio of unit risk factor to that of benzene (see
Table F-l in Appendix F).  This ratioing technique is valid since the
other pollutants are emitted from the same sources only at different
emission rates and different associated unit risk factors.  (The unit
risk factors resulting from a unit volume concentration (ppm) were
used for ratioing as they are thought to better represent the health
response.) In the calculation of EDB and EDC risks, the fraction of
gasoline which is leaded must also be considered, because EDB and EDC
                                  6-5

-------
                                TABLE 6-2.
                       EMISSION SOURCES CONSIDERED
                             IN RISK ANALYSIS
Bulk Terminals
1.   loading racks (including tank truck leakage)
2.   storage tanks
3.   vapor processor (controlled terminals only)
Bulk Plants
1.   loading racks
2.   storage tanks
Service Stations
1.   underground storage tank vents
2.   automobile refueling (tank displacement and spillage in  vicinity
     of gasoline pumps)
Self-Service
1.  automobile refueling (tank displacement emitted from fill  neck)
                               6-6

-------
are contained in and emitted by leaded gasoline only.   The risk due  to
each pollutant was then projected through the year 2020 in proportion
to the total  (for Bz and 6V) or leaded (for EDB and EDO  gasoline
consumption.
     The gasoline vapor to which people are exposed in the environment
contains primarily the most volatile components in liquid gasoline,
whereas in the API animal  experiments, an aerosol  formed from all
components in liquid gasoline was inhaled.  The identity of all the
carcinogenic  components in gasoline has not been determined.  Thus,
there is no way of knowing whether the potency of the  gasoline vapor
to which humans are exposed, is equal  to, more than, or less than
that of the gasoline aerosol used in animal studies.  In the absence
of other data, it was assumed in this analysis that the unit risk
estimate based on total vaporization applies to all  exposures resulting
from emissions from the several gasoline marketing source categories.
6.2.2  Incidence Analysis
     Annual incidences (occurrences) of cancer expected to result
(after some unknown latency period) from estimated exposures to benzene,
EDB, EDC, and gasoline vapors were calculated for the  various control
options examined in each industry segment.  The estimation procedures
considered exposure levels and populations, existing controls, exempted
facilities, and phase-in and phase-out of additional control measures.
Estimates of incidence due to EDB, EDC, and gasoline vapors generally
were calculated from the benzene base year incidence numbers for all or
part of the source category (industry segment).  Base  year estimates of
incidences due to self-service exposures to EDB, EDC,  and gasoline
vapors, however, were based on measured hydrocarbon concentrations.
incidences were projected from the base year estimates to the years
1986 through 2020 in proportion to the total or leaded (for EDB and
EDC) gasoline throughput for the source category.
     6.2.2.1  Bulk Terminals and Bulk Plants.  Since there are about
1,500 bulk terminals and 15,000 bulk plants in the United States
handling gasoline, limited resources would not allow modeling each
plant individually, even if data were available regarding exact location
and throughput.  Therefore, a method was developed to assess the
nationwide incidence due to either bulk terminals or bulk plants.
                                 6-7

-------
This method took into consideration the geographical  distribution of
gasoline throughput and current controls and the varying population
densities around various plant types and sizes.  This method is
outlined in Table 6-3 and the following discussion.
     In general, several cities across the country were selected to
determine an average incidence for each bulk terminal or bulk plant
model plant.  Bulk terminal and bulk plant model plant locations
were selected such that larger facilities were more often assumed to
be placed in larger cities.  An exposure model was used to determine the
incidence due to emissions from each model plant in each of the cities.
The total nationwide incidence due to bulk terminals and bulk plants was
then estimated by:  (1) calculating an average incidence for each model
plant, determined from the incidence in the selected cities, and (2)
multiplying this average incidence by the number of facilities in
each model plant category.
     Ten specific localities were selected for each model plant by
considering:
     (1) the proportion of bulk terminal throughput or bulk plant
         throughput through each of the ten EPA/DOE Regions;
     (2) the proportion of the regional and nationwide throughput for
         each facility type that was controlled or uncontrolled at
         baseline (see Tables B-7 and B-9);
     (3) an assumed locality size distribution;
     (4) as widespread as  possible a geographical distribution of
         cities representing each model plant  size (in order to achieve
         a composite representative of  nationwide climatological
         conditions).
For  each State  and region, the bulk terminal  throughput was assumed to
be equivalent to the total throughput.  The bulk plant throughput for
each State was  estimated based on  the percentage of  total throughput
that passed through bulk plants according to  the 1977 Census (see
Sections B.2.1  and B.2.3).   In most instances  the  localities were
selected from cities where bulk terminals were known to be located or
where  bulk terminals or bulk plants were  likely to be located.  Because
many of  the selected localities reasonably can be  assumed to have
bulk terminals  and/or  or bulk  plants  of varying sizes,  it was  only
                                   6-8

-------
For
                                  TABLE 6-3
                         BULK TERMINAL AND BULK PLANT
                          ANNUAL INCIDENCE ANALYSIS

     Assumed a distribution of localities by size for each of the 4 bulk
     terminal and 4 bulk plant model plants.  For example, the largest model
     plant of (1,000,000 gal/day) was assumed to be located only in cities
     with populations greater than 500,000.
     Ten specific localities were selected to represent each model  plant by
     considering:
          1.   the proportion of total bulk terminal  or bulk plant throughput
              through each of the ten EPA/DOE Regions.
          2.   the proportion of the regional and nationwide bulk terminal  or
              bulk plant throughput that was controlled or uncontrolled at
              baseline.
              the assumed distribution of locality size for each model  plant.
              known or likely locations of bulk terminals and bulk  plants.
      3
      4
example
      •i
              if 10% of the national  gasoline throughput is in Region V,  then
              10% of the localities were chosen from Region V;
          2.   if 80% of the gasoline throughput is within controlled areas,
              then 80% of the localities were selected in controlled areas;
          3.   if 30% of the plants represented by bulk terminal  model plant #2
              are assumed to be in localities with populations greater than
              500,000, then 30% of the localities for model plant #2 had
              populations greater than 500,000.
          4.   if possible, a locality known to have a bulk terminal  or plant
              and that met the criteria of region, control status, and
              population range, was selected.
     Specific coordinates for a bulk  terminal or plant in a locality were
     chosen using features on U.S.G.S. topographical  maps, in the following
     order of preference:
          1.   addresses of known facilities,
          2.   marked gasoline tanks,
          3.   unmarked tanks,
          4.   pipelines,
          5.   transport (major highway or railroad) routes,
          6.   commercial/industrial  areas.
     Human exposure model (HEM) was run for each locality and for each emission
     source (see Table 6-2), assuming different release characteristics and a
     surrogate benzene emission rate  (so the same HEM run could  be used for
     several  model plants with differing emission rates).
     Average  benzene incidence calculated for control 1ed and uncontrolled
     plants under .each control option for each model  plant.  Used proportion
     of throughput controlled, distribution of bulk terminals or bulk plants
     among model plant sizes, and assumed total  number of bulk terminals  and
     bulk plants to calculate total  nationwide incidence due to  benzene
     emitted from bulk terminals or bulk plants in a base year.
     Calculated incidence due to other pollutants using appropriate ratios
     of emissions and risk.
     Projected base year incidence into future based on gasoline consumption
     (except for one type of bulk terminal  emissions proportional to the
     number of tanks).
                                      6-9

-------
 necessary  to use a total  of 41 actual  localities  across the country  to
 provide  10 locations  for  the analysis  for each of the 8 model  plants.
      Specific coordinates for the facility in a locality were  chosen
 using features on U.S.  Geological  Survey topographical  maps.   In  order
 of  preference the features were:   (1)  addresses of known facilities;
 (2) marked gasoline tanks; (3)  unmarked  tanks; (4)  pipelines;  (5)
 transport  (major highway  or railroad)  routes; and (6)  commercial/
 industrial  areas.
      The System Applications Human Exposure and Risk  (SHEAR) version of
 EPA's Human Exposure  Model  (HEM)1  was  used to estimate  annual  incidence
 from  benzene (the model incorporates the unit risk  factor)  for a  surro-
 gate  emission rate of 100 g/s from loading racks  and  storage tanks at
 both  bulk  terminals and plants  and from  vapor processors at controlled
 bulk  terminals.   (The surrogate emission rate was  used  so that a  single
 HEM run  could be used readily for  different model  plants by rat'ioing
 the actual  emission rate  to the surrogate emission  rate.) Each source
 and facility type combination was  assumed to have  different release
 heights, stack (or release)  diameters, and initial  dispersion  parameters.
 Uncontrolled and controlled emission rates were calculated for each
 emission source  at each model plant.   (For the bulk plant size exemption
 control  option,  the smallest  bulk  plant  model  plant was  left uncontrolled.)
      The average incidence  due  to  benzene  from each controlled or
 uncontrolled  model  plant  was  then  calculated  based on the control  option
 and whether  the  locality  is  in  an  area that  is already controlled.  The
 number of facilities  represented by each model plant was  determined
 from  facility  size  distributions given in  Sections 4.1.1  and 4.1.3.
 Thus, the base year nationwide  incidence due  to benzene  could be
 calculated  (by multiplying  the  average incidence by the  emission
 rate  and the  number of facilities  for uncontrolled and controlled
 facilities of  each model  plant  size, and summing for each facility type)
 for each of the  bulk terminal and  bulk plant model plants.
     The incidences due to the  other three pollutants (EDB,  EDC,  and  GV)
were than calculated using their respective ratios (see Table F-l)
of emissions and  risk  relative to those of benzene and using the  leaded
gas fraction for  EDB and EDC.   Incidences were projected to  2000  and
assumed constant  thereafter based on the total (for Bz and GV)  or
                                  6-10

-------
leaded  (for EDB and EDO gasoline throughput for bulk plants (see
Tables  5-5 and 5-6).  The same projection method was used for all
emission sources at bulk terminals except for some types of storage tank
emissions. The AP-42 emission factors for these types of emissions are
proportional to the number of tanks (containing all types of gasoline
for Bz  and GV or containing only leaded gasoline for EDB and EDO,
rather  than the gasoline throughput.
     6.2.2.2  Service Stations.  There are about 400,000 service stations
in the  country.  Because even a single metropolitan area can have a
large number of stations, only a fraction of which could be located and
for which throughput data could be found, even a case study analysis
using actual service stations is infeasible.  Therefore, an area source
approach was used to estimate service station incidence.  It was assumed
that the population in an area was exposed to a uniform concentration
of emissions from service stations in the area.  Service statio'n
emissions were determined by:  (1) assuming all gasoline consumption
(less 3 percent for agricultural  use) in an area passed through a
service station, and (2) using service station emission factors
to estimate the emission rates.  The uniform concentration for the
selected area was then computed by the exposure model, assuming that
all service station emissions were uniformly distributed across the
land area.
     The country was divided into seven ranges of population within
multiple or single county areas (primarily standard metropolitan
statistical areas (SMSA's) or counties outside of SMSA's.  Four to
seven sample areas in each population range were randomly selected to
represent the population in all the areas in the range.  (The population
densities of the selected areas as well  as the populations were checked
to ensure that the range was well  represented.) A total of 35 areas
were chosen to represent the entire nation using the method presented
in Table 6-4.
     The HEM-SHEAR model area source routine4 was run to provide the
ambient benzene concentration that would result in each sample area from
a uniform area-wide emission rate.  Uncontrolled emissions from service
stations in each sample area were calculated for both inloading
gasoline to underground storage tanks and vehicle refueling (full
                                  6-11

-------
                                  TABLE 6-4.
                               SERVICE STATION
                          ANNUAL INCIDENCE ANALYSIS

A.   Divided all areas (primarily standard metropolitan statistical  areas
     (SMSA's) or counties outside of SMSA's) in nation into 7 population
     ranges.

B.   Randomly selected (from 4 to 7) areas to represent the population io all
     the areas in each range (a total  of 35 areas to represent the nation).

C.   Used gasoline throughput and land area for component counties of each
     area selected in B above to develop area source emission rates  for
     inloading and vehicle refueling,  assuming emissions are released
     uniformly over the sample area.

D.   Ran Human Exposure Model for a unit area source emission rate (1.0 kg/
     km2-yr) from each sample locality to estimate resulting ambient
     concentration.

E.   Calculated estimate of incidence  due to uncontrolled inloading  and
     uncontrolled vehicle refueling for each sample area using the area-
     specific ambient concentration, emission rate, and population.

F.   Assigned baseline control levels, appropriate population range, and
     current attainment status to each county in the country.  Then  estimated
     fraction of population in each population range at a given control
     (emission) level under each control option.

G.   Estimated base year benzene incidence from both inloading and vehicle
     refueling under each control option for each population range considering;
          1.  total incidence due to uncontrolled emissions from all sample
              areas in the population  range;
          2.  the fraction of total population in the population range that
              is within the sample areas;
          3.  the population-weighted  emission level summed over all control
              levels under the particular control option.

H.   Estimated national incidence due  to benzene in a base year by taking the
     sum of the incidences for each population range.

I.   Calculated incidence due to other pollutants using appropriate  ratios of
     emissions and risks.

J.   Projected base year incidence into future based on gasoline consumption.
                                      6-12

-------
service and self-service).  The base year on-highway gasoline through-
put from the National Emissions Data System (NEDS) for the applicable
counties was used to estimate emissions.  Emissions from inloading and
refueling were segregated because the release parameters and the ratios
of emissions of other pollutants to benzene emissions vary with the
differing tank temperatures.  Annual incidence from service-stations for
each sample area in a base year was then calculated by: (1) assuming
that all service station uncontrolled emissions from inloading and
vehicle refueling were uniformly spread over the total land area given
in the 1980 census^; (2) using the unit risk factor for benzene;
(3) using the 1980 census population for the counties comprising the
area^; and (4) using the uniform ambient concentration resulting from
an emission rate of 1.0 kg/km2-yr in the area estimated by the HEM-SHEAR
model. A unit emission rate was used, as with terminals and bulk plants,
to allow the flexibility of ratioing the model results for controlled
and uncontrolled emission rates.
    ,The current control level, and the current attainment status (for
assessment of nonattainroent area options) was assigned to each county
in the country and the population of each county.  The fraction of the
population in each population range that was at a given control level
under each service station control option was then estimated.  For the
control options with size exemptions, the percentage of the population
that would be exposed to uncontrolled service stations was assumed to
be equivalent to the percentage of throughput through size-exempted
facilities calculated from the data on service station number of facili-
ties, throughput, and model plants discussed in Section 4.1.5.
     The base year incidences due to public exposures to ambient benzene
emitted from both gasoline inloading and vehicle refueling under each
control option were then calculated.  The incidence for each population
range was calculated by the procedures noted in Table 6-4.  The incidences
for each population range were then summed to calculate the total
national incidence.  The resulting base year benzene incidences for
both gasoline inloading and vehicle refueling under each control option,
when fully implemented, were multiplied by appropriate gasoline throughput
fractions, emission ratios, and risk ratios to estimate risk due to
EDB, EDC, and gasoline vapors and to project risks from 1986 to 2020,
considering phase-in and phase-out of controls with time.
                                  6-13

-------
     6.2.2.3  Self-service Vehicle Refueling.  The self-service vehicle
refueling analysis assesses the risks to the general public resulting
from exposure of individuals to high concentrations near the emission
point (tank fill neck) within the service station during self-service
refueling.  These individual self-service risks are in addition to the
risks resulting from exposure of area residents to much lower ambient
concentrations from dispersion of emissions during both self-service
and full-service refueling, which were assessed by the service station
incidence analysis.  Because of the much higher exposure concentrations
involved, the self-service refueling incidence was found to contribute
about 80 percent of the total nationwide incidence.
     The self-service vehicle refueling incidence analysis was based on
actual concentrations measured in the region of the face of the person
filling the tank.  Exposure data were taken from an API study6 that
measured personnel exposures during multiple (4 to 5)  fillings-of either
unleaded, leaded, or premium gasoline at 13 service stations in 6 cities.
Samples for the study were collected in charcoal  tubes (MSA-type) using
battery-operated pumps and analyzed using gas chromatography and flame
ionization.  Benzene, total hydrocarbons (measured as  n-hexane), and
eight other compounds were measured.
     For the incidence analysis, an average benzene exposure calculated
from the API results was used for each of two fuel types:  leaded and
unleaded (a weighted exposure for premium and regular unleaded was
calculated to represent the entire study period).   The average benzene
exposures during refueling were calculated to be 0.96  ppm for unleaded
and 1.46 ppm for leaded.  Gasoline vapor exposure concentrations in
parts per million were also calculated based on the  measured value of
mg/m3 for total  hydrocarbons, the individual station ambient temperatures
from the API study, and the value for the molecular weight of gasoline
vapors of 66, used in AP-42.7  The calculated gasoline vapor exposures
are 58.8 ppm for unleaded and 72.4 ppm for leaded gasoline.   The assumptions
and calculations for incidence from self-service  refueling are outlined
in Table 6-5.
     Several assumptions had to be made regarding self-service operations
so that the incidence could be computed.   First,  the time duration of
exposure had to be assumed.  The duration of exposure  is  dependent upon
                                  6-14

-------
                                       TABLE 6-5
                                      SELF-SERVICE
                                  INCIDENCE ANALYSIS
A.    Calculated average  benzene  and  gasoline vapor exposure concentrations during
     vehicle refueling for unleaded  (0.96  and 58.8 ppm, respectively) and leaded
     (1.46 and 72.4  ppm,  respectively)  gasoline  from results of monitoring study of
     attendents during multiple  fillings at a number of service stations.

B.    Calculated average  annual risk  due to exposure, e.g., for benzene:

     (2.2 x 10-2 incidences/ppm  -  70-yr. lifetime) / [70 yr/(70 yr. lifetime)]
      = 3.14 x 10~4  incidences/(ppm  - yr)                                    •

C.    Assumed 8 gpm average pumping rate and one  person pumping; therefore, one
     minute of exposure  resulting  from  pumping each 8 gallon unleaded or leaded
     gasoline, e.g.,

     minutes of exposure to 0.96 ppm =  (national  annual unleaoea gasoline
                                       througnput)/ 8 gal/mi n.

D.   'Converted to years  of exposure  resulting from leaded or unleaded gasoline:

     years of exposure = (minutes  of exposure) / (60 min/hr x 24 hr/d x 365 d/yr)

E.    Assumed 70% of all  service  station throughput was self-service, so that
     using equations developed in  C  and D  above,

     years of self service exposure  = 0.70 (national annual gasoline throughput)
                                    8  gal/mi n x 60 min hr x 24 hr/d x 365 d/yr

                                       (national  annual gasoline throughput)
                                    =             6,006,857.1

 F.   Calculated annual  incidence  resulting from self-service refueling during a
      base year, assuming a linear dose-response relationship, e.g., if all gasoline
      were unleaded:

      annual incidence = 3.14 x  10*4 incidences  x 0.96 ppm x
                                     (ppm-yr)

                         (national base year unleaded gasoline throughput)
                                        6,006,857.1
                       =  5.02 x 10"11  (national  base year unleaded  gasoline  throughput)

 G.   Calculated annual  incidence due  to benzene and  gasoline vapors emitted  during
      all self-service refueling based  on:

      1.  projections of onboard-controlled (unleaded),unleaded gasoline  not
          controlled with onboard, and  leaded  gasoline (not controlled with onboard)
          in a given year.
      2.  fraction of throughput controlled with Stage I! under each control  option
          in a given year;
      3.  estimated fraction of  uncontrolled  refueling emissions and resultant
          exposures still emitted out  of  the  tank fill neck with Stage II
          (e.g., 90 percent of 5 percent theoretical  emissions after control)
          and Onboard (50 percent of 2  percent  emissions  after control)  controls.
  H.  Calculated annual incidences due to other pollutants using  appropriate  ratios
      of emissions and risk (see Table F-l), assuming EDB and EDC exposures to  be
      proportional to total gasoline vapor exposures.
  Note:   The impact of  self-service  refueling  emissions on the user population is
         independent of population under  the assumptions made.
                                              6-15

-------
the pumping rate assumed for each self-service fillup.  Gasoline" pumps
operate typically within the range of 8 gallons to 12 gallons per
minute.  The lower value of the range (8 gal/min) was assumed as the
pumping rate because a typical fill up is not conducted at the maximum
pumping rate for the entire fillup period (e.g., "topping off" to get
an even dollar figure or reduced pumping rates toward the end for
prepaid self-service fillups).  Therefore, it was assumed that 8 gal/min
represented a better basis for estimating total exposure duration.
It was further assumed that only one person would be exposed to the
self-service concentrations at each fill up.  Finally, the proportion of
self-service fillups was assumed constant at the level occurring in the
base year (70 percent of consumption was through self-service operations
in 19828).
     Since a nonthreshold, linear dose-response model is the basis  for
the unit risk factor, any exposure (no matter how small) is assumed to
result in some risk of cancer.  The risks across the exposed population
are summed to determine the total  cancer incidence expected.  For self-
service, wherein some person is always pumping fuel, the total annual
incidence is directly proportional to annual self-service gasoline
throughput.  Thus, knowing the leaded and unleaded gasoline throughput,
pumping rate, and pollutant concentrations, the incidences due to
benzene and gasoline vapor uncontrolled emissions in the base (1982)
and subsequent years were calculated.
     The incidences for the study years (1986-2020)  and control  options
were then calculated.  These incidences were based on the projected
amounts of gasoline pumped to onboard-controlled vehicles (all of which
will  use unleaded gasoline) and of leaded or unleaded gasoline pumped
to vehicles not controlled by onboard, either through stations with or
without Stage II controls.  For onboard, one-half of the theoretical
emissions after control  were assumed to be emitted from the fill  neck.
The other half is assumed to be emitted from the carbon canister.   For
Stage II, nine-tenths of the theoretical emissions after control  were
assumed to be emitted from the fill  neck and one-tenth from the  underground
tank  vent.  For in-use incidence calculations,  all  vehicles with tampered
onboard controls are considered totally uncontrolled and assumed to use
one-half unleaded and one-half leaded gasoline.  For Stage II in-use,
                                  6-16

-------
the. emissions from the underground tank vent are considered the same
as for the theoretical case.  All of the additional in-use emissions
are assumed to be emitted from the fill neck.  Onboard and Stage II
controls were phased in and out according to the control option in
proportion the gasoline throughput.
     6.2.2.4  Regulatory Strategies.  The total incidence due to any
pollutant under any given regulatory strategy was calculated by summing
the annual or cumulative incidences for the appropriate control options
for each source category.  (This is the same approach as used for
emission calculations in Chapter 5.0).  For example, the total incidence
under the baseline regulatory strategy is comprised of the baseline
incidence for each of the source categories.  In contrast, the total
incidence under the Stage II - All  Nonattaiment Areas (Stage II - NA)
strategy is comprised of the baseline incidence for bulk terminals and
bulk plants and the Stage II - NA incidence for service stations and
self-service.
     6.2.2.5  Automobi1e Operations.  The incidence due to benzene
emitted during the operation of automobiles was also estimated as a
point of reference for the gasoline marketing evaluation and to assess
the total effect on incidence of reducing the benzene content in
gasoline and, when onboard controls are used, of controlling
evaporative emissions thought to be escaping current evaporative controls.
Benzene emission rates and vehicle miles traveled (VMT) for the various
types of gasoline-powered on-highway vehicles based on the Mobile 2
model were obtained.1-0  Composite emission rates were calculated that
included both evaporative (from fuel tank and carburetor) and exhaust
(tailpipe) emissions or only evaporative emissions from all gasoline-
powered vehicles (light-duty vehicles, light-duty trucks 1 and 2,
heavy-duty gasoline trucks, and motorcycles).  An incidence estimation
approach similar to that for service stations was employed, i.e., an
approach using uniform area source emission rates in each of. a number
of randomly selected sample areas.  The same sample areas, SHEAR-calculated
ambient concentrations, and 1980 Census populations and land areas were
used.  In addition, the on-highway gasoline-powered vehicle miles
traveled from the National Emissions Data. System (NEDS) were used to
calculate the uniform emission rates, and subsequently, the incidence
                                  6-17

-------
 for  each  sample  area.   Annual  benzene  incidences  for  1986  through 2020
 were projected from  the base year  incidences  in proportion to the
 projected vehicle miles traveled.   The VMT projections were taken or
 extrapolated  from DOE's Highway  Fuel Consumption  Model 10,11  The Highway
 Fuel  Consumption model  output  contained VMT values  for 1986 through
 1990,  1995, and  2000.   After 2000,  YMT was assumed  to be constant.
      Benzene  in  exhaust emissions were assumed constant regardless of
 the  gasoline  benzene content,  based on a study using  four carsl2.  The
 study  results show that reductions  of  about 60 or 80  percent from current
 levels of benzene fuel  content (about  2 percent)  would not be likely to
 reduce benzene exhaust  emissions.   Presumably, benzene is formed during
 and  after combustion by cracking larger hydrocarbon molecules and by
 reforming from smaller  molecules.   Evaporative emissions of benzene
 were assumed  to be reduced in  proportion to the reduction of benzene in
 gasoline.
     Evaporative emissions can be further categorized as (1) emissions
 that are  being controlled by current evaporative controls and (2)
 emissions  that are not  being captured  by current evaporative controls.
 A preliminary estimate  of the  average  evaporative emissions that are
 currently  escaping capture of  0.13  grams of benzene per vehicle mile
 traveled was used to estimate the total incidence attributable to
 vehicle operations if these additional  evaporative emissions are
 considered.  These additional evaporative emissions would be reduced
 proportionately if the  benzene content of gasoline were reduced and
would be controlled by  the larger canister that would be  required for
 onboard control  of vehicle refueling.   Benzene reduction  in gasoline
 reduces only the benzene contained in  these additional evaporative
emissions, without affecting the other pollutants.  Onboard controls,
however,  were assumed to control  all of the additional evaporative
emissions, including other pollutants.
6.2.3  Lifetime Risk Analysis
     The lifetime risk analysis estimated  the  probability  that  an
individual subjected to high  exposure  levels throughout a  70-year
lifetime would result in a cancer incidence.   The  term "lifetime risk
from high exposure"  is conceptually similar to the term "maximum lifetime
risk" which has  been presented  in other EPA documents, including those
                               6-18

-------
on benzene sources regulated or considered for regulation under
Section 112 of the Clean Air Act.  The term "lifetime risk from high
exposure" rather than "maximum lifetime risk" is used in presenting the
risk calculations for the gasoline marketing study because EPA is less
certain in this case that the assumptions used result in the maximum
exposure to any single person or group.  The Industrial  Source
Complex (ISC) dispersion model13, capable of estimating individual  and
combined contributions to ambient concentrations at a number of receptor
points from multiple emission sources, was used.  The ISC model calculated
annual concentrations of benzene, gasoline vapors, EDB,  and EDC at  receptors
in the vicinity of a bulk terminal complex, a bulk plant complex, and a
service station complex.  The following three sections describe the methods
used to calculate lifetime risk from high exposures attributable to each
of these three industry segments.  An additional section describes  the metho
used to calculate the lifetime risk from individual self-service fillings.
     6.2.3.1  Bulk Terminals.  Table 6-6 outlines the lifetime risk
analyses for both bulk terminals and bulk plants.  Terminals are
typically clustered together in a location either at a point along  a
pipeline or river.  Therefore, a complex of terminals was used to
estimate the lifetime risk instead of a single terminal  facility.
Figure 6-1 shows the hypothetical layout of the bulk terminal  complex
used as input to the ISC model.  The layout was based on the apparent
centers of individual terminals at a known bulk terminal complex in an
attainment area, shown on a topographical map.  Six bulk terminals  were
assumed to be in the complex, including at least one of each of the
four model terminals of various sizes with gasoline throughputs ranging
from 100,000 gal/day to 1,000,000 gal/day (for a total of 18 loading
racks, 27 storage tanks, and 6 vapor processors in the complex). The
physical dimensions of each source (release height, location,  initial
dispersion parameters, etc.) represent the dimensions of typical sources
within a bulk terminal with throughput comparable to the model terminals.
The ISC model  was executed with varying emission rates (based  on controls
and estimates of unleaded, leaded, and premium gasoline throughput) for
each of the years 1986, 1990, 1995, and 2000 to obtain predicted concen-
trations at each of the receptors shown in Figure 6-1.  The model was
                               6-19

-------
                                  TABLE 6-6.

                         BULK TERMINAL AND BULK PLANT

                            LIFETIME RISK ANALYSIS
A.   Used Industrial Source Complex (ISC) dispersion model  to estimate annual -
     average ambient concentrations due to combined effects of the various
     emission sources (see Table 6-2), assuming different dispersion parameters
     for each emission source at each size model plant.

B.   Ran model using sets of meteorological data from several different
     locations across the country (selected because of known bulk terminal
     locations or suggested high resultant concentrations and because of
     proximity of likely terminal location to residential areas).

C.   Modelled a complex consisting of a number of bulk terminals (6) or bulk
     plants (4) of the various model plant sizes, both for controlled and
     uncontrolled complexes (since the receptor with the highest total concen-
     tration resulting from all emission sources could change with control).
     Configuration of facilities within complexes was based on actual or
     likely arrangements.

D.   Used modelling results for the years 1986, 1990, 1995, and 2000 to
     estimate total risk over a 70-yr. lifetime.

E.   Calculated incidence due to other pollutants using appropriate ratios of
     emissions and risk.
                                     6-20

-------
                              A   A   A
                              .   2
                 A   A •
                 A
                         A   .A
                                                      A   A   A   A
                                                                          3
                                                                  A   A    A   A   A   A    A
                                              6
A   A    A    A   A
A
A  4
                                                          *   A   A
AAAAAAAAAA
                                                     •    *'
                          250 METERS
                           SCALE
                                        A   A   A    A
                   Figure  6-1.   Map  of Bulk Terminal  Complex
                                                                                LEGEND
                                                                              ^ TANKS
                                                                               • VAPOR RECOVERY UNIT
                                                                               • LOADING RACK
                                                                               * RECEPTOR
                                             6-21

-------
 also executed with  three  different  sets  of meteorological  data  represent-
 ing various  parts of  the  United  States;  however, the maximum concentra-
 tion always  resulted  from the  same  set of meteorological  data.  The
 maximum concentration predicted  in  each  of the years 1986,  1990, 1995,
 and 2000 was then used along with the benzene unit  risk  factor  to
 calculate the maximum lifetime risk  for  benzene  (both controlled and
 uncontrolled).  The maximum lifetime risk for gasoline vapors was then
 calculated by using the unit risk factor and emission rate  ratios.
     Because only storage tanks  containing leaded gasoline  emit
 EDB and EDC, and leaded gasoline has a lower and rapidly decreasing
 (with time)  throughput, a different  set  of storage  tank emission rates
 was used with the ISC model to obtain predicted concentrations  for EDB
 and EDC.  The maximum concentration  predicted in each of the years
 1986, 1990,  1995, and 2000 was then  used along with the EDB or  EDC unit
 risk factor  to calculate  the maximum lifetime risk  for EDB  or EDC.
     6.2.3.2  Bulk Plants.  The  lifetime risk analysis for  bulk" plants
 is outlined  in Table 6-6  (p. 6-20).  Figure 6-2 shows the hypothetical
 layout of the bulk plant  complex used as input to the ISC model.  The
 configuration shown was selected to  represent a typical  complex of bulk
 plants, all located in one part  of a metropolitan area.   The bulk
 plants shown include each  of the four model plant sizes with gasoline
 throughputs ranging from  3,000 gal/day to 17,000 gal/day.   (Each of the
 bulk plants in the complex was assumed to have one loading  rack and 3
 storage tanks for gasoline.) The physical dimensions of the sources
 within the complex are representative of typical sources within a bulk
 plant.  The  ISC model was executed with  varying emission rates  (based
 on control and gasoline throughout)  for  each of the years 1986, 1990,
 1995, and 2000 to obtain  predicted concentrations at an array of
 receptors.  The model was also executed with four different sets of
 meteorological data representing various parts of the United States.
 The maximum concentration always resulted from the same set of meteoro-
 logical  data.  The maximum concentration predicted in each of the
years 1986, 1990, 1995, and 2000 was then used along with the benzene
 unit risk factor to calculate the maximum lifetime risk  for benzene
 (both controlled and uncontrolled).   The maximum lifetime risk for
 gasoline vapors, EDB, and EDC was then calculated by using the relative
                                  6-22

-------
      s  1
£  £  3
S3  «C  O
                                     •4        4
                                                            £
                                                            UJ

                                                            §
 X
 
-------
unit risk factors and emission,rates, and in the case of EDB and EDC
the leaded gasoline consumption rate.  All bulk plants were assumed to
emit EDB and EDC from leaded gasoline at a rate proportional to the
leaded gasoline throughput.
     6.2.3.3  Service Stations.  The methodology of the lifetime risk
analysis for service stations is presented in Table 6-7.  Figure 6-3
shows the hypothetical layout of the service station complex used as
input to the ISC model.  This complex configuration was developed to
represent a grouping of service stations at an urban exit from an
interstate highway.  The complex was assumed to be comprised of the
eight service stations shown, which include at least one of each of the
five model stations with gasoline throughputs ranging from 5,000 gal/month
to 185,000 gal/month.  The entire complex was comprised of 14 refueling
islands and 8 underground storage tank vents.  The physical dimensions
of the sources within the complex are representative of typical" sources
at service stations.  The ISC model was executed with varying model
plant-specific uncontrolled emission rates (based on baseline throughput)
for each of the years 1986, 1990, 1995, and 2000.  The model was also
executed with three different sets of meteorological data representing
various parts of the United States.  The maximum concentrations again
always resulted from the same set of meteorological data.  The maximum
baseline concentration predicted in each of the years 1986, 1990, 1995,
and 2000 was then used to calculate the maximum concentration for each
regulatory strategy in each of these years (including the phase-in of
specific alternatives).  These maximum concentrations were then used
along with the unit risk factor to calculate the maximum lifetime risk
for benzene.  The maximum lifetime risk for gasoline vapors, EDB, and
EDC was then calculated by using relative unit risk factors and emission
rates and in the case of EDB and EDC the leaded gasoline consumption
rate.  All service stations were assumed to emit EDB and EDC from
leaded gasoline at a rate proportional to the leaded gasoline throughput.
     6.2.3.4  Self-service Vehicle Refueling.  The baseline maximum
lifetime risk due to individual exposure at self-service filling is a
product of the concentration to which the individual is exposed
(obtained from a report by API6), the unit risk factor  (for either
benzene, gasoline vapors, EDB or EDC), and the estimated time of exposure
                                  6-24

-------
                                  TABLE 6-7.

                               SERVICE STATION

                            LIFETIME RISK ANALYSIS
A.   Used Industrial Source Complex (ISC) dispersion model to estimate
     annual-average ambient concentrations due to both underground storage
     tank unloading and vehicle refueling, assuming different dispersion
     parameters for each emission source at each size model plant.

B.   Ran model using three sets of meteorological data from different
     locations across the country (same locations as for bulk terminals and
     bulk plants).

C.   Modelled a complex of eight service stations of the various model
     plant sizes based on a typical  exit from a major highway.  Uncontrolled
     emissions from both inloading and vehicle refueling were modelled (since
     the receptor with the highest total concentration resulting from all
     emission sources could change with the control option).

D.   Used modelling results for the years 1986, 1990, 1995, and 2000 to
     estimate total risk over a 70-yr. lifetime.

E.   Calculated incidence due to other pollutants using appropriate ratios
     of emissions and risk.
                                       6-25

-------
toll'
           • •

           • •
g.
§
a-
                                                o  ce.
                                                    o
                                       S   (-O   »-H  t_J
                                       UJ   «t   =1  LU
                                       >   d   ea  o;

                                       .   •   B   -
                              ~=\
                                —  o
                                —  o
                                                                          a.

                                                                          I
                                                                         o

                                                                          c
                                                                          o
                                                                         •r-
                                                                         •!-»
                                                                          1C
                                                                          cu
                                                                          o
                               
-------
 per  individual  per  lifetime.  For  this  analysis,  it was assumed that
 high exposure equated  to  a person  conducting self-service fillups
 totaling  40  gallons each  week (e.g., a  traveling  salesman).  The
 maximum lifetime  risk  from individual exposure for benzene gasoline
 vapors, EDB  and EDC under the different regulatory strategies is obtained
 by proportioning  the uncontrolled  lifetime risk in accordance with the
 fraction  emitted  out of the vehicle fill neck as  discussed in Section
 6.2.2.3.  The calculation procedure and assumptions used are outlined
 in Table  6-8.
 6.3   PRESENTATION OF RISK ESTIMATES FOR REGULATORY STRATEGY
      This section presents results of the health-risk analyses.  Both
 the  cumulative  incidence  and lifetime risk from high exposure are given
 for  each  pollutant  from each source category under each regulatory
 strategy.  The  health-risks from EDB and EDC are  reported although they
 are  much  smaller  than those from benzene and gasoline vapors.  The
 lower risks  from  EDB and  EDC are due to the lower pollutant contents in
 gasoline, which result in lower emission rates, and to the dependence
 of EDB and EDC  emissions on only the leaded gasoline throughput, which
 is expected to  decrease significantly.  Appendix F presents annual
 incidences for  each source category and control option as well  as risks
 from high exposure  in 1986, 1990,  1995, and 2000.
     The cumulative incidences from each source category and the total
 from all  source categories are presented in Tables 6-9 and 6-10 for
 benzene and gasoline vapors,  in Table 6-11 for EDB and EDC, and in
 Table 6-12 for benzene and gasoline vapors from vehicle operations.
 The cumulative  incidences are presented for all  regulatory strategies
 (discussed in Section 4.3.2)  both with and without size exemptions.
 For example, the  baseline (no additional controls) estimates of cumulative
 incidence due to  benzene over the 35-year study period (1986-2020)  are
 2.4 from bulk terminals, 1.2  from bulk plants,  6.5 from service stations
 (due to ambient concentrations in the area),  and 113 from self-service
 (due to on-site concentrations during refueling by the public,  i.e.,
 not by paid station attendants).   The total  cumulative incidence due to
benzene emitted from all source  categories is 123.  The reduction in
 incidence from baseline levels is also given  for the strategies assessing
additional controls.
                                  6-27

-------
                               TABLE 6-8
                             . SELF-SERVICE
                         LIFETIME RISK ANALYSIS

A.   Assumed linear dose-response function, i.e.,

     lifetime risk (probability of cancer) = unit risk (probability of
     cancer/lifetime exposure to 1 ppm) x average exposure concentration
     (ppm x fraction of lifetime exposed to that concentration).

B.   Assumed benzene unit risk of 2.2 x 10-2/0ifetime-ppm) (see
     Table 6-1).

C.   Used average benzene (gasoline vapors) exposure concentrations of
     0.98 ppm (58.0 ppm) for unleaded (regular) gasoline and 1.46 ppm
     (72.4 ppm) for leaded gasoline based on monitoring data.

D.   Calculated fraction of lifetime exposed to concentration  assuming

     1.  an average pumping rate of 8 gal/min;
     2.  a weekly gasoline consumption rate for an individual  receiving a
         high exposure under a given control option;
     3.  an equivalent "individual refueling life" of 50 years within a
         70-yr lifetime.

     So that for a weekly gasoline consumption of 40 gal/wk:

     fraction of lifetime exposed = (40 gal/wk)/(8 gal/min x 60 min/hr
                                     x 24 hr/d x 7d/wk) x (50-yr
                                     "refueling life "/70-yr lifetime)
                                  =  0.000354

E.   Calculated lifetime risk due to benzene in uncontrolled emissions,
     e.g., for unleaded gasoline:

     lifetime risk (unleaded, uncontrolled) = (2.2 x 10-2/1ifetime-ppm)
                                              x (0.98 ppm) x (0.000354
                                            .  lifetime)
                                              = 7.63 x 10-6

F.   Calculated lifetime risk to reflect the effects of various control
     options by multiplying by the estimated fraction of uncontrolled
     emissions emitted out of the tank fill neck after either  Stage II
     (0.045, theoretical) or Onboard (0.010) controls, e.g., for Onboard
     alternatives:

     lifetime risk (unleaded, Onboard) = 7.48 x 10~6 (uncontrolled) x (0.010)
                                       = 7.48 x 10-8

6.   Calculated lifetime risks due to EDB and EDC using appropriate
     ratios of emissions and risk relative to gasoline vapors  (see
     Table F-l).

Notes:  (1)  The difference in benzene concentration in the vapor of
             unleaded versus leaded gasoline changes the lifetime risk
             from high exposure (refueling 40 gal/wk) with no  control
             from 7.63 x 10-6 for unleaded to 1.14 x 10-5 for  leaded.
        (2)  A change in the assumed pumping rate from 8 gal/min to 10
             gal/min reduces the lifetime risk from high exposure due
             to uncontrolled pumping of unleaded gasoline from 7.48 x 10-6
             to 6.11 x 10-6.
                                    6-28

-------
o:
o
»_
o
i —
CO
?y
1 —
s:
Z3
>— <
S
— 1
CC. >-
LU CD
Q. LU
Q- i—
C£.
LU 1 —
— ~j CO
CS
\>O £*
*""s t""i
•a: r-
C 	 1
^'cD
CO LU
o _^_
«^ CJ
> <:
LU
LU
•zi a:
>-. LU
_l Q
o -z.
V7 HZ3

CD CO
LU
0 >-.
< 0
CD
LU LU
•z. i—
LU <

LU UJ
CO CJ
CZ.
0 O
OC CO
*"*" CD
LU Z
CJ *—<
z: h-
LU UJ
S ct
CJ <
2: s:
LU
LU ZZ
^* >•— <


1— C
<: co
— j <
=5 CD
SI
^> .^L.
cj o

CD Ll_
| . |
5
5-
^.
CO
LU









^
s_
O
£"
OJ
o
l/J
ID
u
c
CC




o"
CNJ
0
CVJ

1/1
a>
IT)
J_
QJ
o
o
5
."2
c
~~*
>
2
3
E


















e
i"fi •
- t
. t
;, e
4


c
"




t
f
4
&




C
t

e
iH
>




c
c
c.
*
ffi



c
c
c
a
c
t-


~
s








*
- y
c c





r
p>

r
c
rs



tn
S
i
ir
=



r


0
C
n
vC

n
P1

c
c
£
' is
'-'

0.
ps.'
"\
5
CC
\£

x
T






;.
£
c
c
ffl
% £ "" £
25 S?
c c c c
r- — c* c
O IP CS C
IN (N O C1

co r. (N * C


«•« ft,m
7* ff> p* (r
0 ffl W (f
5 c 35
C CD 9 C
£%. ££
xx


(S IS N (V
r-! r-i »C rJ
r* PS ^ fs
00 00
* * 
* C

0 t
£5
rwr,
fv —

5?
0) IX
ss



rt tn
co e
o c
0 C
K /%
::



~ £


0 C
•r P»
o •
- vC
f
 «
«S IT*
"* "*
C C
"! ^
- ^

»n r
O C
tr> Cf
P* (S
m m
c c
- -
r^ rv

"* ""
**. ^




•
=
0 C
— e
a, —
* *E.
X f
t t 1
fie
(£

P- C
5 c
* ^
O <£
N 0
\J) CC

in r
m —
o c



IS
0 C
m S
5Z



- ^
•v

0 C
0 r-
^ (N
ifi ^
P' n


mn
en ff.'
m m
o e
CD tt
(S IN
x x
• •

NM
Pk P>
K r*.
0 C
9 C
•£ <£

x v
TT




C
5 E
& —
1 —
e c.
- X C
* b ft
?^C
tr.
fv C" O
(C — C
0 C C
*x cr rw
g£ ?.
if) 1.1 ff1
M V P*
m ts m

\a * —
p> — n
(DC •*



"^ i
c c e
m so e
fv T N



Ml". •«


zt o
O C O
rt %c *
XX X
»* S"1 (^


no m
co r' tp*
~ — n
0 C 0
r» i£ o
— *s
XX X
•8 g ^
o o *"
*!•
(S P< r*
or •*-
5 c c
— — c
IN fM t£

XX X
"t ^ T


M

C £
o- c e
« — c
i/ C. C.
X t
etc c
5- N C
O — 0 £
V, O

a>c
3 c
ts r.
(ft IT.
— fy
CD CC
ss


mr.
?s



°-
0 0
V 9)
en c
p. «



"" S


0 C
n *•
 S C
S -11
•B ft i

0 X
i .a « a.
f °9 i
tr. a
1- t£
(T. r
0 C
^ it
m m
tr> e
CO (C
cr> cc
o in


IN IN
p- m



ss
0 C
HE
gs



« if
C Po

0 0
n tf
PS
X X


mn
u* e-
m r
0 0
, *> *
M IN
*^ x
N (S



•s P.
5 c
CO C
o «c

X X
* T




c
i £
v "c.

C X
« « i
?- =
B C - C
in j
* o
 «
•> *i
- ™
i. K
0 C
(N vC
«• 0"
"* x
CD O
si in
o c

• •
n r:
0 C

IN rv

X X
-. P.


-
•c
M C
a =
c c e
H!
u t i.
X •
• « t
tf N
e - c
« C =

*£
0 C
in ^c
sg
n e


N r.
if) (N



CO C
0 C
— cr
CD te



^ j
P* iTi

0 C
r» -
ZC


ft «
e ri
- —
h W
C 0
fN tf
" *
x "*
4) C
in in
,o c

• •
in in
3 c

*s ri

^ x
p- P-



•o
ti e
c e
< J e
z c c
a c.
— t
t,- V 4
f «"- 0
tn tr.
— >* fi <-
So o -
3 C 0 C
tn t£ r1 IN
a1 r~ "> c
IT. £ ^ C
v C if> p*
ff- O C^ O
n p. o c
v r-, n (\

CD CO J, «
M r. ^ "
SK J;S



S £ S ?
5= 55
5^ -RS
n m •« n

-

^ ff> ,J _|
* '** 5 ?

ss ss
-o nf>
c! s°'


m « in «:
ui oi • o yi
n & — —
00 0 C
« e 's tf
Sr^ - ^
XX ^ ^
^ **! * ?
"" "" o c

tN f%
rx p> »1 m
0 C C C
o « « -
^ u; tv IN

.XX XX
** ^ P-


T

o • e
C C I1 S
? "" " -
^ S a « t
** X * X
v e li £ t
cc h 9- S N
57 -C7
W « li
o o

1. u
0 0
o o
Pv C •
c, e-
v, k.
3 C
f* r
V V
CD «
IT r



it m
CO C
o o .
0 C
ix r. •
m IP
£ w



S S l
M —


C C
o «c
X X i


nn
o (ri
ffl m -
0 C
co c
(S f^*
^ x
r is
0 C

^ •
E
23 "5
£
C tt C
aj i£* **
V V C
x X v
?» -
°° ' Z
e
"c
e e *•
0 C •
- - 0.
c - -> . **
e t f
-55 T
w -C T 13
t 4 * C
£ N N C
EC. ffl ffl S
m T n e
• • t
t




































?
-
i

i
-
—

V
a
•fl S
"a^
ff
3 0
s Z
Oj J,
—
Is
:i
=
cr-.
                                         6-29

-------
 ID >-
   eJ
 a uj
 o h-
 O 
CO CSS

ss
   UJ
Q J—
UJ UJ
2: o
UJ OS
UJ CO
en
U- UJ
   ^x:
UJ O£
UJ
O H-4

2: _l

t-H O
UJ
                    2   8'
                    «   ON
                        PN
                        r* rv   M p,   H 03
   CO
   §
«=C O
O OS
UJ O
uj o:
UJ
_j
ca
                        */ a  x ** o   *• ;
                    ! Si
                    e  £ e c
                            D - 0  0 — C
                                        (N   n *.
                                       - C  c — e
                                       *• c  *. - o
                                       a-*  in &.—

                                       * "c.  a c "S.
                                      H X •   X C
                                      • < e  H < e
                                                            FI pj   n n
                                                                       •. -   cr<
                                                                                   **  0  O O -
                                                      i e  *)  e
                                                            e c -j c c
                                                                     : c *- c
                                                                      c a-
zt-<°* s-S !=   8
                                                                                        §*i **
                                                                                        u o
                                                                                               1-
                                                 6-30

-------






LU
— 1 >-
O CD
OO LU
«=C 1—
CD «SC
o;
O CO
DC.
QZ.
CJ O
Q h-
LU  o
CJ OO
Q CD
| , J ^»

•=C P
2: LU
i— i i£
I— o:
oo <
LU s:

r-'
IO
LU
i
03
1—




t
9
A
CO
Q;


03
e
*



y
_Q in
uj ™*
o
LU

e
— * 0
o -
CM *
O *
1 W
CO 0
I— >
""""* g
U)
1_
V

uo
rn
t_
QJ
> e
° -s
03 -
0 a!
c
OJ ^
-O -M
•r- 3
cj a
QJ
•M
3
o 7
.•
s
b
0
t-
*3
a






Is
a o
*3 0
en b
c *»

m v
N O
o o
0 0
; v m

ss
0 0

N ON
m m N
d do
3 SS
(N N (N
O O O


•4 O ••
N \n ^
(N (N N
0 00
N MM
0 00


f (NO
3 SS
o oo
0 00
CO ON
(N N N

0 O 0



N N N
N N N
N N N
8 88
d do

8S
00

•J) PI
««
d d
o in
o o


m m
o o
Ss
0 0


N <4
SS
P P
0 0
0* *
-I
OP
P 0



N N
N (N
O 0
d d
O.O0558
O.OO330


ss
o o
do
in m
OP
d d


a
< — s
9 O.
H X C
"CO

N N MO
00 00


M M m N
N (N •« O
O 0 00
22 SS
P P O O


P P OP
OP P 0
t* \o C> v
N N — -
PP P P
PO OO



f m N N
88 §8
do do
N ff> in m
o o op
o o o o
dd do*


-- il
o o o o
dp d d
ea eo N N
o o n tn
o o o o
do do*


o e
e s
2S 5
** O *J
O.M a
* 4J •
2S- «S
« * « - «
9 a 0 9 *
0> H en H
e M 0 o -•
*j B c ** a


MO O
m co o
M fS O
do d
^ N O

MM §
do d

M  Ft
So
O O
d d
o.tiosse /
o.oossa /


0 O
. o 0
0,0
r* tv
in ft
o o
d d

,
Jj
0
a a
= e
O ^0
" a
M f
HI 0
1 1
H .

O O O O
fN C* O O
I M 00
00 00


M In o o
MM 0 O
00 00

n eo N N
MM n m
do do
M N -4 M
o o OP


mo N N
o o o o
0 P OP
M 4) MM
M p (N N
do do


T N co eo
(MM m m
OP 00
00 00
c> n «  • •
X X
6-31

-------


2^
§
U.
CO
D;
0
Q-



UJ
Z CO
1-4 UJ
«-J 1— «
O CD
CO UJ

CD «£
CC.
O H—
§ CO

03 CD

E o£
0
U. UJ

UJ O
O UJ
2 _J
UJ UJ
O CO
1 II j
o D;
z ui
§
UJ 3

»-« CO
1— . -Sf
i — <
i i

O UJ
a.
o o
LU

*£ j
Z 0
1 ^ -.p*
CO UJ
UJ >•



CM
rH
1
U3





•°_.
f !_
0
Ow
CO


CD
c
•^*
"o
tn
to
0
1 a>
c
0)
c
 >
1 £
j=|

r- >
3: uj



•tn
c
o
re tn
c in
21
t> UJ
a a)
•a >
4J
U ro
3 i-
0 0
c a.
P CO
S LJ



t/1
C
O

V)
tn
13 "5
§"
i- tU

V)
3;
1

CL
O
a;
^«
U

*•
I

ra
o
r—
ie
«§.
4-> CO





tn
C
O
ra in
C U)
o •*-

a U

4->
"5 Z
J= 0.
4J CO
•»- >
3: uj



UJ "*

_J r


£ 8 £•
i. j- >>
Eo en
4J O
r * *;
S i— ro
i a i;

a o-

^
CM
t- O O
1 O O O
1 — ( O O
U) CM -H
co en CM









o o o
I- f_ S_
j O 00
1 O O O
O CM O
to co




J

» •
c 3E 2£ Z 2
*"-.. ***. **^^*.
\o co r»- ^*-
"•sf PO co ro
fO PO CO OO










^ < <<

z z z z
CO CO CM o
CO CO CO CO
CO CO CO CO





0 O
C •>- 1-
O 4-» 4-»
T- O O
4J 3 3
o -o -c
3  cc N N
c -o ca ca
•»- t- N
"ru o
tn .0 tn ^r co
rO C ro • •
CO O CO CM r-<
UJ CO

•-" > «S CD

"— ' S* X
1111















_—
i
,^
f— 0)
s. . "3
a> 01 «»
a. s- j=
a. o 
a> £ s-
r* CD O

3 - tU
CO T3 4->
r— c re
Q. ro E
t. C 4J
o o tn
•p** QJ
cu tn
4J tn  tn c
V c c
O CO
•O f— U
0 4->
o ra tn
J= S- 0)
•r- a; u
r— a. c:
(U o 1)
•^ QJ >^
1— i—  -0
T- > C
X co
CO 4-> •
^ U 0) tn
— - 0) C C
t*— 
> c at tn
01 o j= tn "E
i^ ^ *^ 1.^
*O r^ 0 re"'a)
V) OS- > O.
CO S- tfv •*—
Ol 4«* QJ fi
C C C •—
S- O O i- ••-
O O T- r— CO
t«- 4-> O 4J
^< O tn -^
>i « 3 re
r— -0 C3 4J
re a; tu in
o en s- 3
CO • re
ro 4-> O dj jz
4J cy> s_ ,_ X
CO U jQ  CO C i- C
CD i- r— O
4-> CL.4J
cn flu , — co co
c en 3 s_
i- re in 4J aj
in 4-» aj o a.
=3 OO 1. Z O
CO JS t_>
6-32

-------
For Stage II-NA* with size exemptions (II.A.),  for example,  the estimated
incidences due to gasoline vapor emissions using the plausible upper
limit unit risk factor (see Table 6-9) are 46.8 or 77.2 from bulk
terminals, 23.8 or 39.3 from bulk plants, 96.5  or 159 from service
stations, and 601 or 996 from self-service, for a total of 768 or"1,271
and a reduction from baseline of 75 or 122.  As outlined previously (in
Section 6.2.2) the incidence estimates consider size exemptions, phase-in
and phase-out of controls, and gasoline throughput projections.  It
should be noted that self-service refueling contributes about 90 percent
of the total incidence due to benzene and about 80 percent of the total
incidence due to gasoline vapors.
     The lifetime risk from high exposure from each source category is
given in Tables 6-13 and 6-14 for benzene and gasoline vapors and in
Table 6-15 for EDB and EDC.  For example, the estimated lifetime risk
from high exposure to benzene at baseline  (no additional controls) is
1.23 x lO-4 for bulk terminals, 6.4 x 1Q-6 for bulk plants, 2.4 x 1Q-6
for service station (ambient concentrations in the vicinity of  facili-
ties), and 1.13 x 10'5 for self-service  (concentrations at a station
near a vehicle tank during refueling).   Thus, the highest lifetime risk
from benzene exposure estimated for any  of the source  categories under
the regulatory strategy is 1.23 x 10~4 for bulk terminals.  In  fact,
the lifetime risk resulting from high exposure to bulk terminal emissions
was found to be higher than the lifetime risk from any other source
category, under all of the regulatory strategies and for all the  pollutants
examined.  The lifetime risk cannot be summed for the  industry  as  a
whole because  it is unlikely that  any one  individual would  be  exposed
to high  exposures from all source  categories.
     Lifetime  risks are given  only for regulatory strategies with  size
exemptions  because the calculated  results  differ only  slightly  for the
with-size exemption and the without-size exemption  options.  Thus, the
presented risks  demonstrate the  effect of  the  regulatory  strategy,
rather  than the  risk  associated  with  any exempted or unregulated  facility.
                                   6-33

-------

 CO
 »—t
 a

 H-
 ea: >-
 LU CJ
 O. LU
 O. H-
 =5 .
 co o;
 => o
 O CS
 Z UJ
 »-( D£
 CO
 =3 ar
 ~_~ cj

 CO UJ
 Z CO


 O D£
 CO O
 < as
 O LU

 o£
 z u

    LU
 LU O
 z o:
 LU =>
 I~W O
 Z CO
 LU
 ea ca
OZ LU
U. ^
   a:
i . en
— | co
VI C O
CJ I-
.cr o> *3"
014J 10
= 5 CM
«. co°
0 CM
U» ~4
4) O
u ( en
"> ! «-
1- O
CU
IA r~
I «*•

a) co
01 ! _;
*-4
2 I •*:
1 ; K
1 4J O
O to
~-l LO
X U CO
vi u • «r
e c« CM
»O
> 1
IS 81 ! •—
U = ~
U f- 1 O
3 ,— Ul CM
MO 4J
0 VI e : fc.
O. (9 Q O
as E -
•ss * -
= s - 0
p
t-S 0
ji  S!
•»• ti« i— ] CO
Lifetime
(probability of e
Bulk Termlr
e 123(2364 or
? r
0 >>  ea
= *
0) 4J —«
a: to
§o
0
cn cn
CO CO
123(2364 or
123(2364 or
~* «— i
«• «•
u •—
o o
in in
CM CM
*-t •**
in in
0 0
i** r*
— i ^*
•9- «•
0 O
co co
in in
CM CM
m co
—4 ~<
CM CM
O O
CM CM
t_ S-
0 0
CM CM
CM CM
•N •*
to to
0 S
§ §
co co
0 0
*r «r
to to
CO CO
CM CM
CO CO
CM CM
»— 1 *"1
*
•< «c
z z
1 1
CU OJ
01 o>
(O fi
4J -W
Ul OO
i— i *-i
— O
CNJ O
*r c^
to co
o o
a* «r
CO to
CO CO
— CM
"°. «
O CM
CM ~*
CS ^
en
° fe
^*
* IT)
^ •

fr> ^H
• in
-H O
C» r-*
9 5
0 O
LCI CO
CO LO
Si CM
to co
~-4 ^^
"m
CO CM
LO O
CM
° fe
CO
• CM
CM CM
CO —<
—4 to
CM O
«• cn
to co
o o
cn *r
CD to
CO CO
•_• CM
CO •— •
CO
O CM
CM -*
V CU
en en
Q (O
JJ 4J
to to
>• >•
— C3
CM 0
«• Cn
to co
0 0
cn w
co to
CO CO
-»• CM
CO • — •
CO
O CM
CM —«
CNJ
— a
o
s-
"! o

to o
0 0
— ^
CO
r^» ^
O 0
«• CO
2 S
CM iO
O -«
in
ro CM
LO O
CM
* *
CM
CM CM
CO ^H
—4 to
— S
CM O
Sen
CO
o o
o> *r
eo to
CO CO
*^ CM
CO —
• CO
O CM
CM -H
4)
cn
(O
•w
to
«o
u
CU (0
cn o
Q J3
JJ C
to o
> 1—
>
o
0
S
0
•a-
'XI
CO
CM
CO
CM
r-t
CM
O
o
o
to
o
0
«r
^-
«r
o
CM
to
CM
«T
•H
CM
g
t.
O
CM
CM
10
I
CT»
C")
O
^
^O
(T>
CNJ
CO
CM
«— 4
•e
-K
<
3
I
""S
0) «O
cn o
K3 J=
4-* C

CO
CM
«n
CM
CM
0
O
o
to
o
o
«r
^
^r
cn
to
CM
w
CM"
O
CM
S_
O
CM
CM
to
i
cn
CO
II - N.A. i
) 123(2364 01
i,
tl> 03
cn o
a .0
4J C

CO
CO
"n
o
CM
I
(t)
O
.n
c
o
•e
0)
cn
4Q
4->
to
X
CM
*T
to
en
CO
CO
CO
G
CM
CM
O
O
o
to
0
o
o
o

to
«
o
CO
CO
in
!_
O
CO
CM
CO
wH
CM
«•
to
§
(T>
S
"1
• ta
<. o
• J3
z c
o
1
•B
to a>
cn cn
« 
00 01
X
cn
CO
«r
to
CO
CM
CO
CM
CM
O
O
o
to
0
o
«3"
W
«•
en
S
*r
^M
CM
S
u
o
CM
CM
**
«o
O
O
en
CO
*r
to
CO
CM
co"
CM
Z
10
o
J2
C
o
•B
ID
cn
a
4^
01
X
CM
^r
to
o
en
CO
CO
co"
0
CM
CM
O
O
o
to
o
o
0
0
o
1-4
IO
*r
CO
o
CO
CO
LO
U
o
CO
CM
CO
CM
"9-
to
O
en
CO
CO
CO
S
•e
01
cn
IO
4J
to
«•£
O
OJ J2
cn c
10 o
4J
01
=
s§
cn 01
m ro
i. u
^* ^r
^O _ ^ 4J
F- j: o
4J 0) 3 01
•F- E O c/1
3 n3 J= tj
« 4J £
•o F- a.
0) (1) 10
IO 4J U
% £-•?
VI 10 t-
« £ u
n QJ o
— . E •*-
I/I V "^
ji -a *J TS
tS— a s-
•*- J tw —
c eg
— cn a; eu
ra > cn
V) 4-> 10 10
a> 01 f 4->
— to
OTO O
0) c vi cn
4J 10 r— S.
i<-o =
v> z t. ja
i it) S
>,— 0 0
i. — J3 0
O C
4^ ai o vi
10 cn c
<— Id J= O
3 4J 4J ^«
CALO ^- 4J
Isks demonstrate effect of re
). Therefore. Stage 1I-NA*.
e II options In combination w
except for service station op
L. vi cn
4) IO *
O> •— 4J 	 •
E +•> 01 -O
4J ^ O> IO
oj ^_ QJ o
«- U 1- J2
^- IO £ 1=
_1 U. 4-1 C
IO
CO
<—4
 I
IO

LU
_J
ca

-------
    cs
 O LU
 O I—
 O «£
 z a.
 1-4 |_
 _i oo
 UJ
 s«i >-
 I-H o;
 _i O
 X UJ
 < cu
 us o
 z <
0£ =9
O
Q_ 00
«=C LU
LU O
Z C9
O <
00 O
^C'
O LU
UJ
•Z.
UJ
   to

   CD
CD
o
OS,
oo _j

az oo

LU C3
I— O
LU C£
Lu U.
<=) O
LU I—
H- O
CO tO
LU i—i
   Q£
 I
vo
CQ s:
0)
>
4-»
*•> 4->
z 3
0
u_

u
>
s_
V
10
u-
*••
0)
00
C
o
•f—
+J
« 5
i GO
o
u
x • •—
>
— u
in 
IB
CU CU
U C
3 i- Ul
*- *>
a. in «
X «3 —
uj en a.
.ea) .*
en C t—
•r- CU 3
z N CD
B
E 01
O -O
i. •
M—— .
*»
.K U
tn cu
i— »*- in
a: u- f—
cu 10
E u_ —
•So E
*j E
u >> cu
w— 4«> h—
_J p— JL
£1 3

.0
O
u
a.
Regulatory
Strategy
0
CM
CM
CM
O
CTv
U>
<•>
CM
~*
^
»•*
ir>
s-
o
CO
IO
00
r-0
v»4
**
CO
o
TT
0
oi
CM
W
CM
to
••<
O
CM
CM
CO
w
10
o
CM
CM
CM
O
1— 1
CT>
10
ro
CM
••H
CD
"«3
t/t
to '
CO
0
CM
CM
CM
0
^H
CJ>
l«
PO
CM
t«H
oo
u
o
r»
•_<
in
o
CO
m
CM
o
r*.
ro
fH
in
•H
v^
O
CM
CM
CO
^
vo
C3
CM
CM
CM
O
~H
en
in
PI
CM
•K
<
•X.
1
cu
C3)
T5
«
OO
O — * O — * O O O • — • — O - — • CM CM
CMinCMLTlCM CM CMLT5 ITICM LTJ CMCM
CMVOCM^OCM CM CM«3 LOOM \O CMCM
CM n CM ro CM CM CM en ?n CM ro
i. S_
00000 O.OO 00 0
I-H i-^
^H C*J ^H CVJ *— I ^ —*. CM OJ*^ CM CT^CT^-
O^VOO^^OO^ O^ CT\ kO ^O O*i iO LDLO "CJQJ+J
tncNjincMuo to iricsj CMIO cv ^-**H «S3
— m •*- • co ^^ — ^ •— * m m -— n mo flt/)^
cn»ro.fn ro m« -m - •• ,__ .^ tn
CNJOCNJOOJ CNJ CNJO OCNJ O *^m 3 0> — ,c 2 x
£^ =
^- ^r^r^^-tr^r^- — '*-» .^ **
^-. • — * — * * « «» *. . *& ^ i- v> O)
eo — 4 ro ro O o OO OO O •• o-»-4-»o»
Ul ^H ^H S. C 
O O O O O OO OO O OO Q.-T- (4- «-
CO romc-orofomn-) s-*J**-o
r*.«r**r*.» • •• •• . coco o»  LD o o OO oo O CM^< -i-J oj rs 1)
• t— (••« • •« •. . *• *^£O ,_» .. •r-.irf.'*-
• • 'C\j • • »uo r** « r^ oo u  in • «m • ^^s- o>«^3
C\JC\JCS1^-CSJ CNJ CNJCO ITJCxJ. IT) i/>ir> * m m 4^ ^, ^^
i^ o «^ o «-« •— » .-HO o «— • o »-*»-H o 
w* O> 
OOOOO O OO OO O OO (Qn3
0> t/1 .C +J
CMP-*.C\ir**CSJ CNJ CNJf** ^.CM P*- CNJ CM *^ CO
CM^-4CMt-HCNi CM CM -H ^HCM ^H CM CM OJCtAO
CO CM CO CM CO CO CO CM CM OO CM COCO +J «3 p— &
U3 — H U3 — 4 UD 1O 10*-* *-H^0 ^ CM -H VtZuIS
1 (0 S
>,*— O O
*—*-*. . W i-^ .Q O
— » — » «—•. — - — » ^^ • O O O C
O — O — » O O O—» -^-»O — * CNJCM 4-> CU O W
CM LO CM IT) CM CNJ CMUO IOCNJ IT) CMCM «O»C:
CM ^O CM VO CM CM CM ^O ^O CM ^O CMCM r— « JS O
CNJPOCMPOCM CNJ CMPO POCM PO 3^-»+J-»-
s- i. cnoo i- *J
OOOOO O OO 0,0 O V. • 0
»-* — * -K e
»MCM»-tr> o 3 — o
U1 CM LO CM un LT> tO CM CMLT) CM — 1 -^ 14^-r-
^-* fO ^^ f*1) ^^ «_• <~~ PO CO ^^ C*0 f"O O D '•^ C *O
ro -ro >ro m ro- -m • ••• a) •—••->
CMOCMOCMCM CMO OCM O IOPO K^CU^U)
^HCM-^CM»^ "^H *-H'CM CM*-H CM' ^* CM >*— o> s
cu  0 U
CD OO -^
EC 4-> C >
oo o o «o M*T— u
i— — s. cu cu
— «e -o— *>+j+ji.S.U CUCUQ.
z *•> z z c z c o *-> u •••o j= o *J
 OO OO 3 M N ' ' 1— O.
1 1 1 log *7cacQV)i— *CU
•O «aaOa6CU J»tf <— • O
•— •-, • . !_i _ ^i»- eK»e»a 10 •• x
i. i. i- i_ sj • . s-uicn-o
cu cu cu cu 10 cut) cua cu cucu cu cu«3 cacM<-< cuia-cu
en en en en o cno eno en cncn cncno ^oco cu •— ** ^-* u
10 re re 10 .a to ,0 ia^ IQ IQIQ ia IQ^ v>- E^OO-OCU
4-> *J +J JJ C 4->C J-J C +J J-> *J 4J 4->C as • • -r= i- S- T3
00 00 00 OO O OOO OOO 00 OOOO 00 OOO UJ«SC3 J-> r— -«>-«—< x x i— i i_i«-. ^» *i- 
-------
TABLE  6-15.
ESTIMATED  LIFETIME RISK  FROM EDB AND EDC FROM  GASOLINE MARKETING
 SOURCE CATEGORIES UNDER EACH REGULATORY STRATEGY

1.
II.
III.
IV.
V.
VI.
VII.
VIII.
IX.
X.
XI.
XII.
XIII.
XIV.
Regulatory
Strategy
Basel 1 ne
Stage II -
Stage II -
Stage I
Stage II
Stage I *
Onboard
Stage II •
Onboard
Stage II
Onboard
Stage. I 4
Stage II
Stage I 4
Stage II
Stage I &
Onboard
Lifetime Risk from High Exposure*
(probability of effect) to EDB/EDC x 10-*
Bulk Bulk Service Self- Highest
Terminals Plants Stations service for Alternative

• N.A.*
• N.A.


Stage II -

- N.A.* &
- N.A. I
Onboard
- N.A. 5
Onboard
& Onboard
Stage II &
Gas Bz Reduction
A. 62.4% Bz. reduction
B. B1.3Z Bz. reduction
347/448
347/448
347/448
68.8/89.0
347/448
68.8/89.0
347/448
347/448
347/448
68.8/89.0
68.8/89.0
347/448
68.8/89.0
347/448
347/448
16.8/21.7
16.8/21.7
16.8/21.7
4.4/5.7
16.8/21.7
4.4/5.7
16.8/21.7
16.8/21.7
16.8/21.7
4.4/5.7
4.4/5.7
16.8/21.7
4.4/5.7
16.8/21.7
16.8/21.7
6.4/8.2
3.5/4.5
3.5/4.5
4.4/5.7
3.5/4.5
0.67/0.86
6.4/8.2
4.3/5.5
4.3/5.5
4.4/5.7
1.6/2.1
4.3/5.5
1.6/2.1
6.4/8.2
6.4/8.2
196/253
8.8/11.4
8.8/11.4
196/253
•8.8/11.4
8.8/11.4
196/253
8.8/11.4
8.8/11.4
196/253
8.8/11.4
8.8/11.4
8.8/11.4
196/253
196/253
347/448
347/448
347/448
196/253
347/448





68.8/89.0
347/448
347/448
347/448
196/253
68.8/89
347/448
68.8/89
347/448
347/448




.0

.0

          •Lifetime risks demonstrate effect of regulatory  strategies (not higher risks associated
           with  exempted or unregulated facilities).  Therefore, Stage II-NA*, Stage II-NA, and
           luge ^nationwide) have the same lifetime risk.  The same three Stage II «P*°»S "
           combination with Onboard also have a common lifetime risk (although Afferent than without
           Onboard).  Since EDB and EDC are emitted only by leaded gasoline, Onboard controls do not
           reduce  EDB and EDC emissions because Onboard controls will be used on cars using unleaded
           gasoline only.  EDB and EDC lifetime risks from  self-service, therefore  apply only to
           those individuals using leaded gasoline during their entire lifetime.  Onboard controls in
           combination with Stage II would increase EDB and EDC emissions after Stage II controls are
           phased  out.
                                                    6-36

-------
IS)
a; oo
o LU
a. >-<
«c o;
>• o
   cs
LU LU

M =1
_J C_)
O
OO LU 
=C O _J
CD a: uu
   =5 :>
o o uj
2: co _j
LU 1— Z
h-J LU LU

LU &^ i*v
ca eC O
   S U_
•Zi    *Z
O LU LU

LL. i—i O

LU O  i
a    1—4

o o LU

i—i U. et

LU .—-I—
> o: oo
—4 O
h- h- >-
«=C <_) a:
—i — OO
s: —i
i—t    CC
t— o; LU
OO LU O
LU a. s:
   Q. =3


UO LU
.—! _J
 i en

   oo
LU =3

CO —I
c
0
tare
<« t
c
c •*
0 i
t 01
*J tf
U f
D 03
CE

C
D
a
0
D
Cre
•H M
-H O
D a

0 1-
X
c
II
N
e
0
a


o a
82
8 >
s in
w2
01
B (0
k
C
e
m
to
i> a
> c
O 0
U «
C 4i
V Eft
•o
15
M >
<»< C
0 CO
a
2


4J
0
_4
a
JC
3
CD





n
._(
' O
c

L.
0
^
•3
CD






CP
9
jj
0
£
10
o
Jj
e
3
S1
B














O
0

in
CO
to
N




5

t-t
0
to
*
to

fH


?

"
g


fv
«^
X
VD
tf

m
to
g

03
S

N



N
1^


U
0

03
VB

N










S)
e
a
o
a
<£ CO
«o -v




0 0


to o

X X
82

co . 03
03 tr>

0 0
CO Ml
N 0*>
in in
O vC

° S
**

m in


0 0


in N
OJtfj
X X
w <<)
v to

m m
to to
0 0

CO 03
to to
(N 
M U
» U b
< 01 0
z av<
6) C
1 C HI
« C
« S -i
IP C C
e c -<
*J < X
in
in o





0 O


^ 0^
» N
X X
ID 0^

CO f
N Oi
0 O
m M
to o
vv
in co




sa


0 0
si

tn M

a> m
V f»


Sn


00


tM t^
N tn
X X
O i
•v n

to n
o> «n
to m
^

03 03
M tM

N (S



N W
N r^


b b
0 0

CO CO
V V

NN




0)
N a
0 m
n ti u
•2 ffl 0
4J Of***
a B) c
C E
•p < Z
in
If »H
N O



h V<
0 0


O V
CT* \0
in m
X X
tO N

§to
O
V CO
0 0

^0 CO
fM V
(0 \0




»4 tS
IM n

0 0
O rt
v «•
rt P)
N tO

S3


* N

03
b k
0 O


t» N
8R
X X
r- 



LI
0


(N
V
X
in

N
0
N
O
tn

*
S




03


0
g
X
N

£


2
d

g


03
03
10
X
CO
m

m
to
t?

CD
^

N.
«-<


w
r^


0

co
§
•s.
V
CM




C '

a i
ft C
t- 0 V
.C 0 ft!
4J "H >
3 •* 4J
T3 < 0
"c * w
O i
°*
to



U
0


UD
in
X
to

g
0
a>
M
•«•
tn




03


0
g
X
^

m


2


g .


CO
03

? S
^ 0
* w
1





































o
a>
v>
0
JC

5
*.»
—
>
I

c
CJ (DO
*O" re *-
4-1 U
Vi t/1 O)
« Cft (/)
4-> C C

C "- (O
i P
JU -D 4-)
i -5s
. £--
>> o <
jD QJ CT
^Z ^CO tn



































».
re
c

4->
•5
•o
re


o
s

=
^
1
'i
?
a*
^
OJ
i-
s
•^
>
CT
C
"^3
ft)
•£
1

Q;
£ S
'o o
re *->
&> re
i-l
§°
ft!
I/I f—
= JC
'o >
c

cS

*— tn
£ -^
"c §
o 01
U QJ
-tJ i-
o c
UJ  e
revs
JCi C
o

O >fc
c a;
o >
tj IE

*£
S "
sf
— (J
ex—
c" ^
^ w
aJ
£ g
a; ai
T; S-
o o
£ 0)
> U

c*"a)
*£ >
Q.- C

i1^

c o
*o w
re *E
WQ,
(/) O)

"u t-
c o
.^ n.
il
O
W c
U —
"c ?
o re
u
4-> C O
O i- i-
-o^1 °

                                                                 6-37

-------
t/o
O
O.
      LU

      LU
O H-I LU
cot— s
-< Q
LU _1 5E
rxi o «c
= to
LU «£ CO
CD O LU
LU--~t—
o ed e/>


o o o:
LU I— 4 CJ3
>• o: LU
>-»    OS

<: HH o.
_i z LU
=3 LU LU
O H- _l
   •a: LU
=3 I- a:
 i  to LU
Z LU O
LU O to
I— 3=LU
< H-4 HH
S _i ct:
•-< LU o
»— :*: CD
CO >-4 LU
LU _l I—
^H i-« O
 I X C£
t
0
£"°t
c
C -4
O i
•~\ c
•a o
u e
3 £0
•o
It
c:





B
1
a
Bra
-4 -1
"o "
C 0
a t-
01
o
B
C
N
B
C
a


O
IN C
o u
M -4
' ^
£ a
cn (n
W *M
O C
a

*>-
n
n

b
a «
> E
O 0

ss
« «
•a
-4 V
U 0
"s

a
z



B
C
Q.
X


0





B
B
E
b
a
X
g
Q








X
0 >•
*> er.
B V
-4 4J
3 0
Dl b
c n

















cn
0
CO
m


CO

•H



CO
CM



b
0

 o
rt


to 7
•r cn
p« 0

b b
0 0

in o

KS
V P)
v r)


» »
N N
CM CM

b b
0 0
O O
VD 10

X X

CM M



cn cn
CO CO
b b
O 0

in m
CO P)

T V
CM CM


X
C

e B E
NEC
~< 0 f
"38
* 0 b
< V 0
z a n
N CM
CM V


7 f
CM CM
CM CM

b b
0 0
0 0
to to

X X

CM N



cn en
CO CO
b b
0 0

n m
PJ pi

v v
CM CM


X
t>
4J
0 B e
0 E
--< o
E *> 0
0 U b
•H O 0
4J 0.1*4
Z C UI
1 M
Stage I
Annual
Minima
""io
cn
PJ

b
0

^
CO
CM
X

ID
10



1
b
0
en
CO
CM

**
(n
in




CM
CM



0

$

 -H > 0
B ^ -<
•TJ b 13
Onboard
- w/o A
Evapo
- w/ Ad
-ON ]
CO t


b 1
0 I

»H 1
CO 1
0) l
1
X 1
1

f"* 1
1
1
1
t
m i
O l
0 i
i
CO 1
M I
l
1

to i



1
£!


I
h i
. 0 I
l
$ !
•H 1
r
M i
* i
«H 1
m i
i
i
i
i

• i
^ i
*D 1
I
b I
0 l
t
t
P) l
• t
\D t
X l
l
CO t
f) t
1
1
1
l
T I
N j

o :
O I
• l
*4 1


CM i
«H 1
1
1
I

cn i
• i
m i
0 i
i
i
CO i
1

• I
N 1
I
t

1
0 I
h i
§ !
O 1
l
0 '
> 1
O I
(4 1
O t
Q, 1
0 1
> 1
U 1
'1
i
1








































































cn
*^
T3
1
O
QJ
•3
1A

u
(*_
QJ
(A


•£
1
_e

o

in
|
£
^
£
C
w V
c
«TS
CO OJ
Gf> in
C C
4J
in »—
*x §
in M

?"£
i<
u z
c «
— (U
OJ U>
M (C
CD to
Q












































(B
J
+j

(O
**-
o

£
§
u
4->
3
•*•>

»
C
1
e
Of

u

1
I

Q)

!
*r- tf)
0§

>»fc
o °
u
V *p»
"u >
c
.£ E
o
c k.

o
1/1 o
*O (A
4-1 E
§s
u o
o 4->
«£
u o
QJ O.
t?
uj Qj












































f
g
^_ j_
a* m
(/) O
ra A
ja c
o
5 >,

•o
C 
"*" —
U JC
= U
~o to
£!o
t. ^
o*
IA
(A
s JT

•^ «
"3 c

& 4J
S- (D
t.
U O
J= QJ


CO.C
C 01
W
"ii
: (/)

il
§> w

II
•§£
C O
i ^

o
CA 15
(A C

cl
O (D
U
O (A
"S £
O) •*-> 4-1
t§§
LU U U
"D
                                                     6-38

-------
Therefore, the lifetime risks associated with Stage II  are the same
whether applied in selected nonattainment areas,  all  nonattainment areas,
or nationwide, both when used alone or when used  in combination with
onboard controls.
     As a further analysis of Stage II and onboard nationwide control
options, the incidence reductions achievable when considering in-use
efficiencies were estimated.  Tables 6-16 and 6-17 indicate the total
annual incidence and the annual  incidence reduction achieved when
considering in-use efficiencies associated with no inspections (minimal
enforcement) and annual inspections for Stage II  and the expected level
of tampering for onboard.
                                    6-39

-------
6.4  REFERENCES

1.   Anderson, Gerald E. and G.W. Lundberg.  User's Manual for SHEAR:
     A Computer Code for Modeling Human Exposure and  Risk from Multiple
     Hazardous Air Pollutants in Selected Regions.  Systems Applications,
     Inc.  San Rafael, CA.  SYSAPP-83/124.  May 1983.  73 p.

2.   Bulk Gasoline Terminals - Background Information for Proposed
     Standards.  U.S. Environmental Protection Agency.  Office of Air
     Quality Planning and Standards.  Research Triangle Park, N.C.
     Publication No. EPA-450/3-80-038a.  December 1980.  p.  6-11.

3.   Arthur D. Little, Incorporated.  The Economic Impact of Vapor
     Control Regulations on the Bulk Storage Industry.  U.S. Environmental
     Protection Agency.  Research Triangle Park, N.C.
     EPA Publication No. EPA-450/5-80-001.  June 1979.  p. II1-9.

4.   Reference 1, p. 53-55.

5.   U.S. Department of Commerce, Bureau of Census.  1980 Census of
     Population, Vol. 1 Characteristics of the Population, Chapter A
     Number of Inhabitants, Part 1 United States Summary, PC 80-1-A1
     April 1983.  Table 17:  Land Area, Population and Population Density
     for Counties:  1960 to 1980.

6.   Clayton Environmental Consultants, Inc.  Gasoline Exposure Study
     for the American Petroleum Institute.  Job No. 18629-15.
     Southfield, MI.  August 1983.

7.   Telecon.  Siebert, Paul, Pacific Environmental Services, Inc. (PES)
     with Coffman, Mike, Clayton Environmental  Consultants, Inc.
     June 14, 1984.  Clayton's gas chromatography-f1ame ionization
     analysis procedures.

8.   How Self-Service Appeals to Motorists.  National Petroleum News
     1983 Factbook.  75_(7a):103.  March 1984.

9.   Telecon.  Siebert, Paul, Pacific Environmental Services, Inc. (PES)
     with Fletcher, Bob, California Air Resources Board.   June 22, 1984.
     Origins of residual vehicle refueling emissions  after Stage II controls,

10.  Office of Mobile Sources.  Mobile 2 Model  Output:  Benzene Exhaust
     and Evaporative Emission Factors for Vehicle Operations, 1975-2001.
     Job No. 665964 University of Michigan Terminal System (Model  CT043).
     Ann Arbor, MI.  November 2, 1983.

11.  Energy and Environmental Analysis, Inc.  The Highway Fuel  Consumption
     Model Ninth Quarterly Report.  U.S. Department of Energy.
     Washington, D.C.  Publication No. DOE/PE/70045-1.  February 1983.
     Appendix B.

12.  Energy and Environmental Analysis, Inc.  Highway Fuel Consumption
     Model Output:  With No Diesels less than 8500 Ib.  U.S. Environmental
     Protection Agency.  Office of Mobile Sources.  Ann Arbor,  MI.  ca.
     1983.
                                  6-40

-------
13;  Black,  P.M.,  I.E.  High,  and J.M.  Lang.   Composition  of Automotive
     Evaporative and Tailpipe Hydrocarbon  Emissions.   Journal  of  the
     Air Pollution Control  Association.  30:1216-1221.  November  1980.

14.  Bowers,  J.F., J.R.  Bjorklund,  and C.S.  Cheney.   Industrial Source
     Complex  (ISC) Dispersion Model  User's Guide.   U.S. Environmental
     Protection  Agency.   Research Triangle Park,  N.C.   Publication  No.
     EPA-450/4-79-030.   December 1979.
                              6-41

-------

-------
                           7.0  COST IMPACTS
 7.1  INTRODUCTION
     Cost data were obtained and developed on a per facility basis for
each model plant size of each source category within the gasoline
marketing network.  These per facility costs were then combined with
data on the number of facilities requiring controls within each source
category in order that nationwide costs could be determined.  This
section presents source category costs for bulk terminals, storage
tanks, bulk plants, tank trucks and service stations.   Service station
costs were further divided into Stage I, Stage II and onboard analyses
on a per facility basis.
     Nationwide capital and annualized costs were calculated for each
control option and regulatory strategy on both a cumulative and a net
present value basis.  Net present value (NPV) costs were developed to
compare the strategies and to take into account the opportunity costs
associated with spending money now or in the future.  Discounting costs
(using the NPV of costs) was performed because, like emissions and
incidence reductions, control cost for the regulatory strategies are
not uniform over the time period analyzed due to 1) the slower phase-in
rate of onboard controls compared to Stage I and Stage II controls,  2)
the varying equipment life of onboard, Stage I, and Stage II control
equipment, and 3) the declining gasoline consumption which directly
influences the recovery cost credits.  All costs are presented as
constant 1982 dollars.
     For each control option, the capital cost spent and the annualized
cost incurred in each year was determined for 1986-2020.  These costs
per year (time lines) for each control option can be found in Appendix G.
Capital costs over the 35 years of the analysis were incorporated
in the initial years of the analysis and then repeated in the years in
which the economic life of the equipment ended and replacement equipment
should be required.  The capital costs and annual costs time lines took
into account phase in of equipment, as well.  Annualized costs were
                                  7-1

-------
adjusted each year, as appropriate, to reflect reduced recovery credits.
Since the number of facilities were assumed constant and the gasoline
consumption declined with time, the consumption decline was reflected
as a reduced throughput at each facility and therefore reduced recovery
credits.  Reduced recovery credits result in higher annualized costs.
     Nationwide capital and net annualized costs are presented in
Section 7.3 for each of the gasoline marketing control options.
Enforcement costs are not included in these estimates.  Net annualized
costs were combined with the emission reductions (discounted and
non-discounted) given in Section 5.0, resulting in cost-effectiveness
determinations for each of the gasoline marketing options.   Similarly,
as per the gasoline marketing options of Section 7.3, costs and cost-
effectivenesses of each gasoline marketing regulatory strategy were
developed and are presented in Section 7.4.
7.2  INDIVIDUAL FACILITY COSTS
7.2.1  Bulk Terminals
     Capital expenditures and net annualized operating costs for the
control of emissions from bulk gasoline terminal loading operations
have been estimated for four model plant sizes.  The costs presented
for bulk terminals are updated costs from the 1981 costs presented in
the New Source Performance Standard (NSPS) background information
document for bulk terminals.!  Further, capital charges reflect a
10 percent interest rate.  The complete list of model plant parameters
is presented in Table 4-1.
     The control strategy recommended for bulk terminals requires a
35 mg/liter control level be applied only to the approximately
500 terminals that are currently uncontrolled.  Those terminals currently
with controls are operating below 80 mg/liter or 35 mg/liter (see
discussion on baseline emissions).  As in a previous EPA study,2 it was
not considered cost effective to try to upgrade control systems meeting
80 mg/liter to a level of 35 mg/liter.  As in estimating baseline
emissions, it was assumed that approximately 90 percent of the uncontrolled
facilities (450 terminals), regardless of model plant size, currently
practice bottom loading and 10 percent (50 terminals) practice top
splash  loading.  Therefore, within each model plant size, both top and
bottom  load control costs for the base year were determined for three
                                  7-2

-------
   types of vapor processors  (carbon  adsorbers, thermal oxidizers,  and
   refrigeration systems) and presented in Tables 7-1 and 7-2.
        Average bulk terminal control  costs are summarized  in Table 7-3.
   Comparison of net annualized  control  costs for each vapor processor
   type indicates that the thermal  oxidizer would not be a  cost-effective
   alternative for the two larger model  plant sizes in Tables 7-1 and  7-2.
   In these cases, average costs were based upon only the carbon adsorber
   and refrigeration systems to  obtain representative capital and net  annualized
   operating costs.  Average cost estimates for the two smaller model  plant
   sizes are based upon the average of all three vapor processor types.
        Bulk terminal average weighted facility costs are presented in Table 7-4.
   This breakdown by model plant and  loading configuration, used in conjunction
   with the average bulk terminal control  costs in Table 7-3, provides a  weighted
   average net annualized cost in the base year of $34,600  per facility,  a
   weighted average capital cost of approximately $342,000  per facility and a
   value of $63,700 for the weighted  average recovery credit.

               TABLE 7-4.  BULK  TERMINAL AVERAGE WEIGHTED COSTS
                  (THOUSANDS OF  FOURTH QUARTER 1982 DOLLARS)
Model Plant % of Facilities
Size Each MP
(103 liters/day) Category
380 48
950 21
1,900 21
3,800 4
Average value
(weighted by MP)
Top/Bottom Adjustment0
Total
Net Annual 1 zed Costs3
Bottom Top
Load Load
55.9 141.2
41.7 167.6
-39.5 73.1
-139.5 0.8
24.2 128.4
21.8 ' 12.8
34.6
Capital Costs*
Bottom Top
Load Load
252 689
289 949
325 996
444 1,368
285 851
257 85.1
342
Recovery Credit3
Bottom Top
Load Load
18.8 21.9
47.0 54.8
141.2 164.7
282.4 329.4
62.7 73.1'
56.4 7.3
63.7
Average costs per model plant taken from Table 7-3.
Obtained by multiplying the average values weighted according to model plant by
the 90% of terminals assumed to practice bottom loading and 10.% assumed to be in
the top load configuration.
        Gasoline consumption  was  projected in Section  4.2  to decrease with
   time and this had  a corresponding inverse effect on  net annualized cost
   figures.  As recovery  credits  decrease, net annualized  costs increase.
   The weighted average recovery  credit was decreased  annually by a percentage
                                      7-3

-------
                    Jl
                    -I
m
                                                   CO I   I  CD IT> Crt
                                                                     ^4 VO VO CM
                                                                     in to .-H co
                                                           o «• o   r-» i
                                                   *-- en  i  \o in en
                                     • o n
                                     J en \o
                                                   co  i CM en tr> en
                      i  u.
                                                                         • .-*      CT»
CO

to
                                       ^H   o f*-en CM
C£  CO
o  o
o o
                                                            . «^ r*.   ^ «r o VD
                                                       i  i UD in
                                                                       > *f «H r*.   CM
o cr>
                                                   r** in  i in in C
     O)
CO
                                                              <• t^    *^ co o *
    cr
                                                                        10 <•« P-   01
UJ  O
                                                                «r    —« to O e
                                                                          - —4 CM   tn
                     -g
CQ  cn
  I  "inmr"
                                                         ^tlOlf)<-4   O CM O CO
 CO
                           I    I

                           S.
                                          g.
                                     •s-se    s.
                           — 
                              o-ja c
                             ' O  U T-
                                             ^- jj  trt t—    M

                                             	J •*•  V O    "T—
                                      ffl 3   ^- I
                                    , O. W      U  - -
                                    :o H-   f-o t-o
                                                                        7-4

-------
vo    m *r o   m  r*) *
U.
u.

||
0 C
O t-
o"
CO
m

<
c_

u
u
Q
1
^•1
0 C
^
* *irt §1
SI- ~!
Or—
O O - «•
Q
O CM
C£ 05
z 2 L
O !
o fc. '
o>
Q •«-> i
•* *- x
<^ (O *°
/•y _«
c^ ; E
O JC g
^- o
"^ g
«C <*-
z o
S to '
UJ C
1— <0
t/i

-i 0 t
=> -^ t
03 t— , c
CM !
1 ^
^ -el
o - »
o
f-"*

"§
1— i

i






*j
3
Q.
•§,
3
0
w
J=
H-
i
^
O
to
CD
; r^, o>r\i *>
CO in tf> m
• ^^ —4 CO --

; 0--CM VO
~«-4 CO -^




! "^ O CM VO
CM en vo cn
CM •-* CO -^
m
f*- vo en -N
J ^^ CM PO VO
: — » ^M vo
CO
*
3 223*


pn
n^ p^ o* ••<
«! r— «-fnio
J, _,—..£>


«
! r*^ tA O> O
^ CM <*> ^

C




°I
3 «Jr- 0>O
- esj o «*> ^"




"1
51 «§S°




1 I T
L. m c^teo
J •*• CSJ CM C*

: i— v
3 tnr-Iio o
- -JOCVIKJ



•o;
i i vo vo VO O
S f> wt CM CM
J . ^4 ^4 W

MO) -^
0 4-> Ut
 *rt ^ (/» U O
(U 4-> 
o aj u 4J > c en
u E u w» c O c
3u *3 3 C O y •—
Vt Q.^- u ^
1_ OJ Jrf «
a> -M 4-> ^£ U t-
e -^-*- o 3 «
ra — • c c m i- a
> 3 =D ai h- o
*> ^3
^- 3
^1 I

co i i co ui cn -i n* men rC

CM (*)
m o o ^- o r-* ^^
CM f-t VO O «• ' — '
CM <*•>



ro -^ eo «r o irt -^ ^ ^r
co icMcnmen vcovocn en
CM . VO CM VO CM
CM ro
.
—4t ivounvr r^. co co «r o
CM f*> tO CO VO f*


* _^ * ._ ... ^ _^ _^ .^^ j — .
•-« CM


-4 co r*- *r »- O CO r» CM
in i «M ^^. vo «r «er vo O •* vo
^* —4


vo r^^pw v -4m
.-i 1 1 VO iO CM VO VO CD CM ^*>
CM i i m u> m co *•






p*. «r i eo *• P*. o in
f^ tn in in CM r^. en vo i CM
CM v ro —
OM CM



CO^CDW^ -* ^m
•* '-^"^ sisss 5




"7 1 1 **^^^ *7 ^°1
en r < vo ir> *^ CM vo eo CM m

eo i CM er»^- o
en *— «o^7 « c o. i- u *•*
>,»- i— 6*—^- o  &_ (Q  fc

^ §•
trt
1 - ft
^ *m
0 Z *-
1- -x» (t)
--»**• c=
° " i
U M C
(A V
g £ '-
> 9) JX
> *-
"a w m
•»•* u
0 01
h- *»- ••
H- 41
ua u

4>* 3
0 °
tJ *
          7-5

-------
                       Footnotes for Tables 7-1  and 7-2
i
 Carbon Adsorption Unit.

 Thermal  Oxidation Unit
r*                                                               '
"Refrigeration Unit.
d
 Cost of installing vapor collection  equipment  on  existing bottom
 loading tank trucks,  $3,180 per  truck.
a
"Cost of converting top loading  racks to  bottom loading  and vapor
 recovery,  $200,000 per rack.

 Cost of retrofitting  existing top  loading  tank trucks with bottom
 loading and vapor collection equipment,  $6,784 per tank truck.
i
 Electricity costs are based on  average consumption rates reported by
 manufacturers.
i
 Propane for pilot burner estimated at 12.5  liters per hour, a.t $0.18
 per liter.

 Estimated  activated carbon  replacement period  is  10 years, at $3,85
 per kilogram carbon cost.

 Estimated  as 4  percent of unit  purchase  cost,  plus annual rack vapor
 collection  maintenance of $200  per rack  and $200  per terminal.

 Daily system inspections at one  hour per day,  plus a monthly inspection
 for liquid  and  vapor  leaks  in the  vapor  collection and  processing
 systems.

 Cost to perform annual  vapor tightness testing, including one-half
 day downtime, $450 per truck.

 Total  capital investment x  (capital  recovery factor + 0.04), where interest
 rate = 10  percent, equipment economic life  = 10 years (0.163 capital recovery
 factor).
i
 Amount recovered per  year,  at $0.29  per  liter  assuming  a density of
 0.67 kg/liter.

 Difference  between uncontrolled  and  controlled emission level.
3
 Cost effectiveness not calculated  because  net  annualized cost is a
 negative quantity (cost credit).
1
                                     7-6

-------









n
e
X-. ' Q.
.a w» o
«• s- »-
re a
I— ^
co o
0 0
o
CM
—1 00
O Oi
f^* ^.
^b ^^
G3 Gi
_j *
«C 3
z o-
£ f
LU ^f "re
1— o
M— — *

•»! iR
rs w t>
CD T3 O
c 03
LU re
||

LU .C

<«—
,
CO
1
r*.

0>

.a
1C
^


















CO
o to in CM to oo o — i
§o r«* m co to to
oj ~* CO -H ro r*.
o o to en *-4 to co »-*
o to co co to en r*» co
cn^ r-t — < to en co
en
o to«j-eno oeoo
in «9-cMPO«- intoen
en ** *-* 10 en -H — t


*^ «^
§-^ CM »o o en «-i «o
CO «^ CM CM CO ^* !"*>
CO ^^ ** ^- tO ^^





en in
i s • s s ? a a
eo^ CM -^ «• ^4 to

f* in
o| o to eo in en ic
o to co CM CM co CM
en^ — 4 _. ro i PO
•-^1




CM t^
o to ^ en en — < co
in . «; CM -< co «r to
en »-4 «^ , CM *•*



to en CM
o — < CM en CM m in
co co ^H ' in in to
CO 4-4 r* CM



^.
u
^
en
^-

in 4-> ^
O *> in >» it)
u in 0 C ^.
— 4-> o U Ol i—
>, in c u > o
re o o c o i.
•0 0 -- C O U M4->
--- 4J O ^- (U M4JC
f— 4J O) « T- U5 L. +t ino
c in >— ML. M oo
o a? re ,— i- >o o c_>
O •i3 U 4J > C Q. "O O
*»^ inL-incoos i" 01^
O13COU> re N
in > O. «- o +J ^. r-

3 i— . 4-> J-) ji O O Q. 
CL -^i-U33 IO 3O
J= i— C C « S. t. o Cl—
cnio=5=3o:(—i— c
3 ** • — *
O -4— IO
S- Q. 4-> +J
f IO O O)
H- «J (- Z

^



2

_^
CO
co



in
in
oo
^





1
1
1


1
1
1




in
in
CM




to
in
CO










en
«*>

in
in
O)

^j

01
i«_
14—
LU

in
0







<8
o
CL
5
e

•o
IO
o

— ^_*
O in
4>* in
0 0
J3 U
eno
in 01
3-0
CLV
to en~"
~* i1^
1 5§
c  w-
re u o
p— O
I— >. 01
o CL en
•o w
g "S I
2 M- *
"* o*»
o e:
*» o 5
en in
TS <0 Ol
0) t- u
4-* O) CL
re > 01
CSL
= €5
Oi in
• in o
4-1 Ol U
j_
Ol CL L.
U O) Ol
c ^ ^
oi i>
« 43 °
M- tn^-
Ol Or—

o 40 ja - ~
7-7

-------
equal to the percent decrease in gasoline consumption.  Revised annualized
costs were then calculated for each year of reduced gasoline consumption
(see Table G-2).  The projected consumption of gasoline is assumed to
be constant from the year 2000 to the year 2020, therefore recovery
credits and net annualized costs remained constant from the year 2000 on.
     Data on the number of bulk terminals requiring control installa-
tions in each year were necessary since nationwide cost information was
generated from per facility costs.  It was assumed (see Section 5.1.1)
that 50 percent of the facilities incurred capital  costs in the initial
year of implementation (1987) and 50 percent the following year.
This 2-year phase-in of the capital costs was repeated every 10 years
(1997-1998, 2007-2008, 2017-2018) since 10 years was estimated as
the useful life of the control equipment.  Top to bottom loading
conversion costs were assumed only for the initial  installation on
those applicable facilities.  Bottom loading costs were used for all
facilities when the equipment had to be replaced.  Net annualized
operating costs were incurred by 25 percent of facilities in 1987, 75
percent in 1988, and 100 percent each year thereafter.  This was
determined in the same manner as the emission reduction percentages for
bulk terminal phase in over 2 years.
     Total cumulative net annualized costs were determined to  be approxi-
mately $789 million.  Net annual ized costs were discounted at  a rate  of
10 percent to obtain an NPV of $214 million.   Capital  costs were also
determined on both a cumulative and net present value  basis and approxi-
mated $598 million and $221 million, respectively,  during  the  period
1986-2020.
7.2.2  Storage Tanks
     The cost analysis for storage tanks is based upon the control
approach of requiring the installation of an  internal  floating roof on
all existing fixed-roof gasoline tanks at terminals.   The  costs and cost
assumptions used in the storage tank analysis  were  obtained from a draft
Volatile Organic Liquid Storage Tank NSPS3  and a draft CTG for Control
of Volatile Organic Compound Emissions from Volatile Organic Liquid
Storage in Floating and Fixed Roof Tanks.4  These draft reports presented
costs in fourth quarter 1982  dollars so that updating  the  costs was not
necessary.  The capital  and annual ized costs  to install  an internal
                                  7-8

-------
floating roof with a liquid-mounted seal on an existing fixed-roof tank
are presented in Table 7-5.  Some of the capital  cost estimates are a
function of the storage tank parameters which in this case are assumed
to be:  1) a volume of 2,680 m3 (16,750 b,bl),  2)  a diameter of 15.2
meters (50 feet), and 3) a height of 14.6 meters (48 feet).
     The largest single capital cost component involved in the installa-
tion of an internal floating roof is the basic cost of the roof at
$17,188.  To make modifications needed to control emissions from existing
tanks, an estimate from two vendors was used to establish a relationship
for the costs to clean and degass the storage vessel.  The "controls"
for deck fittings are gaskets for covers and sleeve seals for support
columns.  To account for retrofit cost, the cost of additional work
(i..e., cutting roof vents, etc.) was estimated as 5 percent of the
installed capital cost of new construction.
     The annualized cost without product recovery credits is calculated
by adding the annualized capital charges to the costs for taxes,
insurance, and administration (4 percent of the capital costs) and the
operating costs.  Operating costs include the yearly maintenance charge
of 5 percent of the capital cost and an inspection charge of 1 percent
of the capital cost.   Including product recovery credits, the net
annualized cost per facility was determined to represent a savings of
approximately $11,832.
     The baseline analysis indicated that there were approximately 683
fixed-roof tanks at gasoline terminals (see Appendix B, Section B.2.2).
Total  capital  and net annualized costs were determined by using the
same method as for bulk terminals and bulk plants.
7.2.3  Bulk Plants
     Capital  investment as well as annualized costs are calculated for
loading operations at bulk plants for four model  plant sizes.  A
complete list of model plant parameters is given in Chapter 4.0.
Costs were estimated for each of the two bulk plant control options.
Capital cost estimates were developed for installation of a vapor
balance system with submerged loading of storage and truck tanks, for
both incoming and outgoing transfer of gasoline.   Several past
studies5'5'7'8 were reviewed to obtain a range of costs for installing a
vapor balance system and submerged fill on incoming and outgoing loads
                                  7-9

-------
              Table 7-5.  COST OF INSTALLING A BOLTED INTERNAL
              FLOATING ROOF ON AN EXISTING FIXED-ROOF
                       (Fourth Quarter, 1982 Dollars)
Capital Cost & Installation

     Degassing0
     Basic Roof Costd
     Liquid-Mounted Primary Seal6
     Controlled Deck Fittings
        Retrofit Adderf
                               Total Capital Cost
                                                           $ 7,515
                                                           $17,188
                                                           $    124
                                                           $    250
                                                           $    859

                                                           $25,936
 Annualized Cost

     Maintenance (5%)
     Taxes, Insurance, G&A (4%)
     Inspections (1%)
     Capital Charges9
                               Total Annualized Cost
                               Product Recovery Credit*1
 Net Annualized Cost  (Savings)
                                                            $1,297
                                                            $1,037
                                                            $   259
                                                            $3.046

                                                            $5,640
                                                           $17,473

                                                          ($11,832)
  Tank parameters  for fixed roof tank:  Volume   = 16,750 bbl = 2,680m3
                                        Diameter =     50 ft = 15.2m
                                        Height   =     48 ft = 14.6m

  References 3 & 4.
  •*
  "For an existing  tank  the first step  is to clean and degass the storage
  vessel, before installation of the floating-roof.  Cost for this procedure
  is based  on the  following relationship:

       Total Cleaning Costs = 130.8  (tank capacity = 2,680)0.5132
Estimated from the equation:
D 3 tank diameter in meters.
                                 Cost  ($)  =  1,069  D + 939 where
                                 (References 3  & 4).
  "The  cost of  the  liquid mounted  primary  seal  is  estimated  to be
   $2.60  per linear meter of circumference.

   Retrofit costs  (installing an internal  floating roof  in an existing
   fixed-roof tank) are the cost of  additional  work (i.e., cutting
   roof vents)  and  is estimated as 5 percent  of the installed capital
   cost of  new  construction.

   Capital  recovery factor of 0.1176 is  based on a 10  percent interest
   rate and 20  year lifetime for the tank  and floating roof.

   Amount recovered per year, at $0.29 per liter assuming a  density  of
   0.67 kg/liter.   Based on emission factors  between controlled  [internal
   floating-roof losses of (7.3285 x 1Q-8Q) + 2.4  Mg/yr, where Q is  the
   weighted throughput for all  model plants]  and uncontrolled (fixed-roof
   breathing loss  of 8.8 Mg/yr and working loss of 34.2  Mg/yr).
                                   7-10

-------
 for a typical bulk plant.  Cost estimates were obtained for both top
 loading and bottom loading systems for outgoing loads,  A past survey
 of bulk plants9 showed that an average of 91 percent of bulk plants use
 top loading on outgoing transfer of gasoline and that the remaining 9
 percent of bulk plants use bottom loading.  This percent distribution of
 top load vs. bottom load bulk plants was used to estimate a weighted
 average capital cost for control  of outgoing loads at a typical  bulk
 plant.
     7.2.3.1  No Exemption Option.  The "no exemption" control  option
 for bulk plants requires vapor balancing of storage tanks, and trans-
 port and account trucks at all bulk plants.  Uncontrolled plants will
 incur the cost of balancing both  the incoming and outgoing loads as
 shown in the top half of Table 7-6.  Weighted costs are presented to
 account for both top and bottom loading systems for outgoing loads.
 Some bulk plants already have a vapor balance system installed for
 incoming loads.  These plants would incur lower capital  and annualized
 costs in order to balance outgoing loads only as shown in the lower
 half of Table 7-6.
     Capital costs are assumed the same for each model  plant size in
 this analysis.  The physical  layout of the bulk plant (number of
 storage tanks for gasoline, distance to loading rack, number of  loading
 arms for gasoline) does not vary  significantly between large and small
 bulk plants.  Therefore,  one control  system and installation cost was
used to represent vapor systems installed at bulk plants.   However, for
estimating net annualized costs,  the throughput of the model  plant was
used to calculate gasoline recovery credits for a balance  system on
outgoing loads only (or on storage tank draining losses).   A gasoline
recovery credit was not calculated for a balance system on incoming
loads  (or on storage tank filling losses).   This credit is realized by
the bulk terminal  which receives  the vapors during subsequent transport
 truck  fillings.
     In order to calculate the nationwide cost impact of the "no
 exemption" control  option, it was necessary to determine the number of
 uncontrolled bulk plants.  The number of uncontrolled bulk plants was
estimated to be 8,040 (refer to Table 4-10 or Table 7-19).  It was
 assumed that 50 percent of the facilities incurred capital costs in the
                                  7-11

-------
TABLE  7-6.   AVERAGE  CONTROL COSTS  FOR BULK PLANTS  (NO  EXEMPTIONS)
                          (4th Quarter 1982 Dollars)

         Model Plant Mo.                 1234
         Throughput (liters/day)        11,400     24,600     47,300     64,400
         Weighted Average Top & Bottom Loading Costs
         Balance Incoming &
         Outgoing Loads on"
         Uncontrolled Plants*
             Capital  Costb>c
             Annual 0 5 H (3%)
             Capital  Charge (13.IS)
             Taxes, Ins. (4%)
             Recovery Credit^
             Net Annual1zed
             Emission Reduction (Hg/yr)
             Cost Effectiveness (5/Mg)
        Balance Outgoing Loads on
        Plants with Incoming
        Load Balanced"
             Capital  CostM          21,240    21,240
             Annual 0 &  M (31)            637       637
             Capital  Charge (13.15)     2,793     2,793
             Taxes Ins.  (45)              850       850
             Recovery Cred1td            613       1,322
             Het Annuallzed            3,666     2,957
             Emission Reduction (Mg/yr)  •  4.4        9.4
             Cost Effectiveness ($/Mg)    839       314
28,540
856
3,752
1,142
613
5,137
8.1
634
28,540
856
3,752
1,142
1,322
4,428
17.5
253
28,540
856
3,752
1,142
2,543
3,207
33.6
95
28,540
356
3,752
1,142
3,462
2,288
45.3
50
21,240
   637
 2,793
   350
 2,543
 1,737
    18.1
    96
21,240
   637
 2,793
   350
 3,462
   817
    24.7
    33
         Includes the cost of retrofitting  two account trucks for  use in vapor
         balance service.
        b
         Top Load Cost - $19,490 (915),  Bottom Load Cost - 538,970  (95).
         References 5, 6, 7.
        d
         Recovery credits are based on a control  efficiency  of 90 percent on outgoing
         loads  from a balance system (or storage tank draining losses).
                                       7-12

-------
base year 1987 and 50 percent the following year.   This 2-year phase-in
of the capital .costs was repeated every 15  years,  since 15  years  was
estimated as the useful  life of the control equipment.   Net annualized
operating costs were incurred by 25 percent of facilities  in 1987,
75 percent in 1988, and 100 percent each year thereafter.   As with
terminals, revised recovery credits, and hence revised  annualized
costs, were calculated for the control  facilities  for each  year to
account for the decrease in gasoline consumption.
     Total cumulative net annualized costs  were determined  to be  approx-
imately $1,260 million.   Net annualized costs were discounted at  a rate
of 10 percent to obtain an NPV of $337  million. Capital  costs were
also determined on both a cumulative and a  net present value basis and
approximate $653 million and $252 million,  respectively,  during the
period 1986-2020.
     The calculated cost effectiveness  values for  this  control, option
are shown in Table 7-6 for each of the  four model  plant sizes.  Cost
effectiveness is presented in units of  dollars per megagram of emission
reduction and was determined by dividing the total annualized cost by
the emission reduction achieved by the  control technique (i.e., balance
incoming and outgoing loads).
     7.2.3.2  Size Exemption Option. The second control  option would
require bulk plants loading less than 15,100 liters (4,000  gallons)  of
gasoline per day to install vapor balance on incoming loads only
for storage tanks and transport trucks.  An uncontrolled model plant
number 1 (
-------
Table  7-7.   ESTIMATED CONTROL COSTS  FOR BULK PLANTS  (EXEMPT  <  4,000  gal/day)
                                (4th Quarter  1982  Dollars)
          Model  Plant No.                 1         2         3         4
          Weighted Average Top and Bottom Loading Costs
          Throughput (liters/day)        11,400     24,600    47,300    64,400
Balance Incoming Loads
and Install Outgoing Submerged
Fill on Uncontrolled Plants
with < 4,000 gal/day*
Capital Costb«c
Annual 0 & M (32)
Capital Charge (13.1%)
Taxes Ins. (4%)
Recovery Credit
Net Annual 1 zed
Emission Reduction (Hg/yr)
Cost Effectiveness ($/Mg)
To Install Outgoing
Submerged Fill on Plants
with Incoming Load
Balanced < 4,000 gal/daya
Capital Cos1^»c
Annual 0 5 M (3D
Capital Charge (13.1%)
Taxes Ins. (4%)
Recovery Credit
Net Annual 1 zed
Emission Reduction (Hg/yr)
Cost Effectiveness ($/Mg)
8,500
255
1,118
340
1,243d
469
6.6
71
1,200
36
158
48
1,243d
-1,002
2.9
—
28,540
356
3,752
1,142
1 ,322®
4,428
17.5
253
21,240
637
2,793
350
1,322®
2,957
9.4
314
28,540
356
3,752
1,142
2,5436
3,207 '
33.6
95
21,240
637
2,793
850
2,543*
1,737
18.1
96
28,540
856
3,752
1,142
3,462e
2,288
45.8
50
21,240
637
2,793
850
3,4626
817
24.7
33
          a Includes  the cost of retrofitting  two account trucks for use  in vapor balance service.
          b Top Load  Costs (91%)  - $19,490.  Bottom Load Costs (9%) - $38,970.
          c References 5, 6, 7.
          d Recovery  credit is based on control efficiency of 58 percent  for conversion
            from top  splash loading to submerged fill.
          e Recovery  credits are based on a  control efficiency of 90 percent on
            outgoing  loads from  a balance system (or storage tank draining losses).
                                               7-14

-------
throughput less than 15,100 liters of gasoline per day.   From  a  previous
study, it was determined that 42 percent of bulk  plants  have a throughput
less than 15,100 liters per day.10'11  Therefore,  out of the 8,040
uncontrolled plants, 42 percent, or 3,400 of these bulk  plants load
less than 15,100 liters of gasoline per day.  Again,  it  was assumed
that 50 percent of these facilities incurred capital  costs in  the base
year 1987 and 50 percent the following year.  This 2-year phase  in of
the capital costs was repeated every 15 years, as 15  years was assumed
to be the useful life of the control equipment.  Net  annualized  operating
costs were incurred by 25 percent of facilities in 1987, 75 percent  in
1988, and 100 percent each year thereafter.  Again, annualized costs
were adjusted each year to reflect the change in  gasoline consumption.
     Total cumulative net annualized costs were determined to  be approxi-
mately $780 million.  Net annualized costs were discounted at  a  rate of
10 percent to obtain an NPV of $208 million.  Capital costs were also
determined on both a cumulative and a net present value  basis  and
approximate $462 million and $177 million, respectively, during  the
period 1986-2020.
     The calculated cost effectiveness values for this control option
are also shown in Table 7-7 for each of the four  model plant  sizes.
7.2.4  For-Hire Tank Trucks
     Independent owners or operators of gasoline  tank trucks  transporting
gasoline from bulk terminals and bulk plants will  also incur  costs as  a
result of the control options.  The costs to these companies  will
include the capital investment required to convert tank  trucks to
bottom loading (where necessary) and to install vapor recovery equipment.
Annualized costs include maintenance of the vapor recovery equipment
and yearly vapor-tight tests on the trucks.  A gasoline recovery credit
is not given to the for-hire tank trucks for installing vapor recovery
equipment.  The credit is realized  by the bulk terminal  or plant that
receives the vapors during subsequent tank  truck  fillings.  The  costs
to the oil companies and bulk plant owners  which  operate  tank trucks at
their own terminals and/or plants have  already been  included  in the
cost analysis for bulk terminals and bulk  plants.
     7.2.4.1  For-Hire Tank Trucks  at Terminals.   The average cost of
converting tank trucks to bottom loading and  adding  vapor recovery
                                  7-15

-------
equipment would be the same for a for-hire tank truck company as for a
tank truck owned by a bulk terminal.  Bottom loading conversions average
about $3,604 per tank truck and the addition of vapor recovery provisions
requires an average expenditure of about $3,180 per tank truck as shown
in Table 7-8.12
     In order to calculate the cost impact on the for-hire tank truck
industry, it was determined in Chapter 4.0, Section 4.1.4 that 5,800
for-hire tank trucks would be affected.  Since it was estimated that
10 percent of the uncontrolled terminals use splash filling, 10 percent,
or 580 of the 5,800 affected for-hire tank trucks using splash-fill would
require both bottom loading and vapor recovery retrofitting.  The capital
investment for these tank trucks would be $3,934,720 as shown in the upper
half of Table 7-8.  The remaining 5,220 vehicles would already use sub-
merged fill and would thus require vapor recovery provisions only.  The
total capital cost for these tank trucks would be $16,599,600 .as shown in
the lower half of Table 7-8.  The total capital cost accruing to the
for-hire tank truck industry would be $20.5 million.
     The annualized cost due to retrofitted tank trucks includes the
cost of maintaining the vapor recovery equipment and of performing an
annual vapor-tight test.  Capital charges on the initial  investment on
the equipment are also included.  With maintenance costs at $1,000 per
year and tank truck testing at $450 per year,12 the total  annualized
cost for 5,800 trucks would be $11.1 million.
     In summary, the initial capital investment required by for-hire tank
trucks at terminals will be $20.5 million.  The annualized cost for
these tank trucks will be $11.1 million per year.  Fifty percent of the
capital costs were assumed to be spent in 1987 and 50 percent in the
following year.  This 2-year phase-in, consistent with terminal  controls,
was repeated every 15 years (the useful life of the equipment).   After
the phase-in period, the annualized cost stayed constant throughout
the analysis since no recovery credits are associated with tank  truck
controls.
     7.2.4.2  For-Hire Tank Trucks at Bulk Plants.   The average  cost of
adding vapor balance equipment to tank trucks would be the same  for a
for-hire tank truck company as for a tank truck owned by a bulk  plant.
The addition of vapor recovery equipment for top loading systems at bulk
                                   7-16

-------
     Table 7-8.  COST FOR THE FOR-HIRE TANK TRUCKS AT TERMINALS*
                        (4th Quarter 1982 Dollars)
 No.  of Affected Trucks
 Bottom Loading &
 Vapor Recovery
   580
5,220
      Capital  Investment per Truck
      Total  Capital  Costs
      Capital  Charges (13.1%)
      Annual Maintenance/Testing
        (@ $1,450 per truck)
      Product  Recovery Credit
      Net Annualized Cost
    6,784
3,934,720
  517,312
  841,000

     NA
1,358,312
Vapor Recovery Only

      Capital Investment per Truck
      Total Capital Costs
      Capital Charges (13.1%)
      Annual Maintenance/Testing
        (® $1,450 per truck)
      Product Recovery Credit
      Net Annualized Cost
                       3,180
                  16,599,600
                   2,182,412
                   7,569,000

                       NA
                   9,751,412
 TOTAL CAPITAL COSTS FOR 5,800 FOR-HIRE TANK TRUCKS AT TERMINALS =
                                                        $20.5 million.

 TOTAL ANNUALIZED  COSTS FOR 5,800 FOR-HIRE TANK TRUCKS AT TERMINALS =
                                                        $11.1 million.
   Reference 10  shows  that bottom  loading  conversions  average  about
   $3,604  per tank  truck  and that  addition of vapor  recovery is  about
   $3,180  per tank  truck.   Also, maintenance costs average  $1,000  per year
   and tank truck testing averages $450  per year.
                                    7-17

-------
plants averages about $3,180 per tank truck.  Bottom loading conversion
costs for bulk plant trucks were assumed to be the same as for terminal
trucks ($3,604 for bottom loading conversion and $3,180 for vapor
recovery equipment).  It was assumed that the percentage of trucks requiring
bottom loading conversion was the same percentage as the bulk plants
converting to bottom loading (9 percent).  Therefore a weighted cost per
truck was generated ($3,180 x .91 + $6,784 x 0.09 = $3,500).  In Chapter
4.0, Section 4.1.4, it was determined that of the total 58,700 tank
trucks at bulk plants, 26,400 have vapor balance, and 16,000 are owned by
bulk plants and have been included in the bulk plant cost estimates.  The
remaining 16,300 affected for-hire trucks will require vapor balance
provisions under the "no exemption" control option for bulk plants.
Therefore, the total capital cost accruing to the for-hire tank trucks
under this control option would be $57.1 million as shown in the upper
half of Table 7-9.
     The annualized cost due to retrofitted tank trucks includes the
cost of maintaining the vapor recovery equipment and of performing an
annual vapor-tight test.  Capital charges on the initial investment on
the equipment are also included.  Thus the total annualized cost in the
base year for 16,300 tank trucks under this control option would be
$31.1 million.
     Of the 32,300 tank trucks without vapor balance, it is assumed
that 58 percent, or 18,700 tank trucks, load at bulk plants with a
throughput greater than 15,100 liters per day, since 58 percent of the
bulk plants have a throughput greater than 15,100 liters per day (see
Table 4-2).  Of the total 18,700 tank trucks loading at bulk plants-
with a throughput greater than 15,100 liters per day, 9,300 vehicles
are owned by bulk plants (58 percent of the 16,000 uncontrolled tank
trucks owned by bulk plants).  The remaining 9,400 affected for-hire
trucks will require vapor balance provisions under the control option
which exempts bulk plants with a throughput less than 15,100 liters per
day.
     The initial capital cost accruing to the for-hire tank trucks under
this control option would be $32.9 million as shown in the lower half
of Table 7-9.  The annual!zed cost due to retrofitted tank trucks
includes the same maintenance and annual testing costs as outlined in
                                  7-18

-------
      Table 7-9.   COST  FOR  THE FOR-HIRE TANK TRUCKS AT BULK PLANTS
                        (4th Quarter 1982 Dollars)
 NO  EXEMPTIONS
 No. of Affected Trucks
 16,300
Weighted Average Costs - Top or Bottom Loading and Vapor Recovery
     Capital Investment per Truck3
     Total Capital Cost
     Capital Charges (13.1%)
     Annual Maintenance/Testing
       (@ $1,450 per truck)
     Total Annualized Cost

EXEMPT BULK PLANTS < 4,000 gal/day

No. of Affected Trucks
     3,500
57,050,000 or $57.1 million
 7,500,579
23,635,000

31,135,579 or $31.1 million
     9,400
Weighted Average Costs - Top or Bottom Loading and Vapor Recovery
     Capital  Investment per Truck3
     Total  Capital  Cost
     Capital  Charges (13.1%)
     Annual  Maintenance/Testing
       (@ $1,450  per truck)
     Total  Annual ized Cost
     3,500
32,900,000 or $32.9 million
 4,325,487
13,630,000

17,955,487 or $18.0 million
 Top Load Costs  (91%)  -  $3,180,  Bottom Load Costs  (9%) - $6,784.
                                 7-19

-------
Section 7.2.4.1.  The total annualized cost for 9,400 tank trucks would
be $18.0 million as shown in the lower half of Table 7-9.   As with
previous analyses, a 2-year phase-in of controls was assumed.  After
phase-in, capital costs were repeated every 15 years based on the
useful life of the equipment.  The annualized cost remained constant
since no recovery credits are involved.
7.2.5  Service Stations
     7.2.5.1  Service Station Stage I.  Capital costs and  installation
costs of Stage I controls were obtained from two sources.13'14  Average
capital and net annualized costs are presented in Table 7-10 as $1,698
and $341, respectively (4th quarter 1982 dollars).  Since  the number of
underground storage tanks and the amount of piping at service stations
does not vary considerably with throughput (storage capacity would vary
more), costs to comply with Stage I at facilities were assumed to be
independent of facility size.  A gasoline recovery credit  is not given
to the service station for Stage I control.  The credit is realized by
the bulk plant or terminal receiving the vapors upon subsequent tank
truck fillings.
     Data on the number of service stations requiring Stage I control
installations in each year were necessary since nationwide cost infor-
mation was generated from per facility costs.  Four control scenarios
were developed for installation of Stage I systems at service stations
- two nationwide options (with and without size exemptions) and two
nonattainment area options (with and without size exemptions).  Under
both the nationwide and nonattainment options, service stations with an
average throughput of 37,900 liters per month were exempted from Stage
I requirements.  For the nationwide options, it was assumed (see Section
5.1.1) that 50 percent of the facilities incurred capital  costs in the
base year 1987 and 50 percent the following year.  This 2-year phase-in
of capital costs was repeated every 15 years (2002-2003, 2017-2018),
since 15 years was estimated as the useful life of the control equipment.
Net annualized operating costs were incurred by 25 percent of facilities
in 1987, 75 percent in 1988, and 100 percent each year thereafter.
This was determined in the same manner as the emission reduction
percentages for  a Stage I service station nationwide phase-in over 2
years.  The nonattainment options assumed a 1-year phase-in for capital
                                  7-20

-------
               Table 7-10.
           SERVICE  STATION STAGE  I CAPITAL AND
           MET ANNUALIZED COST ESTIMATES3
            (4th Quarter 1982 Dollars)
Capital Cost and
  Installation
                                     1,698
Annualized Costs
  Maintenance (3%)
  Taxes, Insurance
    and G & A (4%)
  Capital Chargesb
    (0.131)
                                        50.9
                                        67.9

                                       223
  Annualized Cost
  Recovery Credit
  Net Annual ized Cost
                                       342
                                        MA
                                       342
$/Mg, Cost Effectiveness0
      MP1 (18,950 liters/mo.)
      MP2 (75,800 liters/mo.)
      MP3. (132,650 liters/mo.)
      MP4 (246,350 liters/mo.)
      MP5 (701,150 liters/mo.)
                                     1,380
                                       345
                                       197
                                       106
                                        37.2
 References 11,  12.
D
 Capital  charges are based on a 10 percent interest rate and  on
 equipment life  of 15 years.
 Since the number of underground storage  tanks  at service stations
 do not vary considerably  with throughput (storage capacity would
 vary more), costs to comply  with Stage  I at affected  facilities
 were assumed to be  independent of facility  size.
 Sample emission reduction calculation:
 (1130-40) ing
    TTter
246,350 liters
      mo.
12 mo.       Mg
  yr.      109 mg  =  3.22 Mg/Yr
                                  7-21

-------
costs beginning in 1986.  Net annualized operating costs were incurred
by 50 percent of facilities in 1986 and 100 percent each year thereafter.
     Capital  costs and net annualized costs were calculated for the
four Stage I  control options described.  Costs were determined both
as a cumulative total and as a net present value (NPV)  of total  costs
incurred from 1986 to 2020, discounted at a rate of 10  percent.   Stage I
control costs are presented in Table 7-11.
                   TABLE 7-11.  STAGE I CONTROL COSTS
                 (Millions of 4th Quarter 1982 Dollars)
Regulatory
Strategy
Nationwide (NO EX.)
Nationwide (EX.)a
NAb areas (NO EX.)
NAb areas (EX.)a
Capital
Cumulative
1,008
423
35
15
Costs
NVP thru
2020
378
159
14
6.4
Net Annual
Cumulative
2,234
938
81
34
ized Costs
NVP thru
2020
590
248
24
10
a
 Size exemptions to service stations with a throughput  <37,900 liters
 per month.
 NA s Nonattainment.
     7.2.5.2  Service Station Stage II.  Cost data were obtained from
three references15'16*1? for service stations ranging in size from two
nozzles  (19,000 liters/month throughput) to 16 nozzles (680,000 liters/month
throughput).  These costs are representative of each of the three Stage II
control  systems currently in use (i.e., vapor balance, hybrid, and vacuum
assist)  and were updated to fourth quarter 1982 dollars using cost
indices.16 The data were grouped according to the service station model
plant characteristics detailed in Table 4-4 (i.e., number of nozzles and
monthly  throughput range).  Estimates of service station control costs
by  type  of Stage II system are based on the average of these data and
presented in Table 7-12.  Capital and net annualized cost ranges from
which these averages were obtained are also presented in Table 7-12.
                                  7-22

-------










1
(S
1— *—
CO "*
oo ce
f*
UJ "o
D. O
co CM
h-CO
CO CTi
Oi-«
o
t.
H* «

»
UJ 
t r** CTt
CO « •
to o








10
CM 10
O^CM

s






_
>t
•—' —
• o
sz
e ^*"
-X.
M C
(/» i. O
^!«J«
• N -P- O
O Mr- 3
z o — -o
4J Z 4-> OJ
C W- 3
« O a. c
r** .O O v>
•S i-fei
0 Z t— UJ
o"
2 -~-
- . tn r-» in co O *o 9)
co eo ^-i f-t ^-H -H o
CM O
*-< cn





CO P^
en ^ ^ ^«j ^
en i co r*- CM r*- ~+ i J •* "^^, ^
in CM ~* en *-*
4* VO







ol^O^r^S
O 1 in (O CM ••* CM^^-t CM^

CM f-
«• en





•^*"
2 ««

S^-o c
w u S>H:5"S -^

o— «OJ= «ACJ*—*-— a>
goa)oc«e» s 2» «*-
4Jr-C: r- ^ U C C UJ
10 "o.*"* C "S. X S -M *^ O
c •
10 v^
CD **
o
CM — •*

\ocM"WCMPom-m«-<
CO CO ^-t CM d O Cft
— « cn^ CM m
U)
s
CM 10
o
en ~-< en «tr o co 10 CM m
H^O • " *^D
••H CM O
o* *—
v4


_
o
*^ ^^
10 If)
• en
m o m 

^~— c
v> S>S5'S —

o«— «e 5 viol^— >=S> t
5 "ro «i— "<8 - cu = « *"
V) 4^ CC (B 4^ VI > ^ C£ 4>*
>«^*^>' 3 <*v V O ^" (A
t/VO. CQ.X04-> O
10 c ic « o a) o
7! o>
1 S
S -
CM ^O
* U)
•I'vigisl1? s
—4 o *-« PO CM m PO
CM O^ P^
CO* —
CM in

»o i ^ in PO *o vo i r*.
«eo * *«r *
in to^ CM PO CM *-«
PO *-*



J_.
U [ *•»
in^ 10
vo CM «-< *o m to ^- ^4
to CM * *o *
•-4 • CM

^




3
s. s
to co
o ieM«fentocn«r. in
CM^CMCOOPnr** t CM
-2 ~ "S







' .8 ~
to o
• ^*
co — 4 r» en co 10 o» m »o
i*r ^eo ""j
CM «• CM MIO M
»-* en o — *
o —
^




r^
2
O VI
S.<~ a)
e »ft 4-> C
4^ (A O**— *O OJ >

(S* o •— «e j= w o ^ *— u
** ° ^*«= ° "^ b 1 I1 S
J2'«fl3f-.'« *aie(o uj
v»^ ^— s v a> o •— **
^ A c O. X U •** IA
0 c  *^-












,
^
0)
c
CO -p-
X J3
_g o
e a>
*/>
O «A
CJ 0)
+•» F—
c «
« >
Z at
en
Ul 0
u *-
a> >
J= 10
„ |
1 1
**• in
e> «-<

CM 2
2 S
u
L. C

4J fc.
w at
1 £
.c e
t.

*£ £
i-
2 a.

^ L.
S V,
-§ U
0. 0

• o

- c"

'
" 5

u c
U i-
«j" *











•o .
i
5
VI
>)
o
u
«3
p
1^
<0
a>
>>
k
0)
B.
01
 ^2
*" S
a> 
i2 o

£ v>
uj a

o
" 5

°!
3 41
•C O











1
Vi
€
4-*
>»

•o
"C
*
* «2
01 4J
-P
* £
•S£
A e
0) """
je^--
fel
ll

in PO

«s

•^ 0)
u in
| J

*o w
*> V?
e
0 4->
U VI
e *v)
O VI
IO
"S E
u =
A U
o) 2
10 V

+>
^£
^
iS
s a
•0"°
7-23

-------
     The cost of installing vapor recovery systems at service stations
can vary greatly, depending on the station throughput, station layout
and the number of nozzles and pumps (see Table 7-12).  The average
capital cost for equipment and installation of balance systems ranges
from $5,088 to $13,709 depending on the size of the service station.
The average cost of equipment for vacuum assist also varies with station
size.  The cost of these systems is usually the highest due to the
secondary processing unit required and ranges from $12,231 to $21,082.
The hybrid system is basically a vapor balance system with liquid
aspirator or jet pump equipment added.  The additional cost for
the equipment puts this system in the middle between vapor balance
systems and vacuum-assisted systems.
     In calculating net annualized costs, a credit of $0.29 per liter
of gasoline recovered during refueling was assumed.  The procedure
followed for determining Stage II recovery credits is detailed in Table
7-13.  Capital  charges were based on an interest rate of 10 percent and
an equipment life of 15 years for vapor balance and hybrid systems and
8 years for vacuum assist systems.  Annual operating and maintenance
(O&M) costs reflect a nozzle replacement charge of $100 each, once per
year for balance systems and once every two years for vacuum assist and
hybrid systems.  Installation costs for the balance system include the
installation of liquid blockage sensors with an average cost of $100  to
$150 per dispenser.  Costs do not include installation of over-head
hose retractors.
     The weighted average Stage II control costs, presented in
Table 7-14, were determined by using an average cost weighted by the
population of Stage II systems currently installed - 80 percent vapor
balance systems, 15 percent hybrid systems and 5 percent vacuum assist
systems.^  Emission reductions (Mg/yr) and cost effectiveness figures
($/Mg) per model plant size are given both for Stage II costs per
system and for weighted average system Stage II costs in Tables 7-12
and 7-14,  respectively.
     Additional  cost data were obtained from the California Air Resources
Board (ARE)20 and the American Petroleum Institute (API).21  These  cost
estimates  reflect assumptions which differ slightly from those used in
                                  7-24

-------
            Table 7-13.   STAGE II  RECOVERY  CREDIT CALCULATIONS
Balance and Hybrid Stage II  Systems


    Emission factors:   Displacement  losses  =  1080 mg/liter
                       Breathing losses     =   120 nig/liter

    Assume 95% recovery of displacment loss and  50%  recovery  of
    breathing loss

Recovery factor = (1080 mg/liter)(.95) + (120 mg/liter)(.50)  = 1086 mg/liter

Example:

1,086  mg   x 37,900 liter x    kg__ x  liter x  12 mo.  x  $0.29
     liter       mo.          10° mg    .67 kg  .yr       liter
                                = $214 per year recovery  credit.
Equation reduces to:
    [Throughput (liter/mo)] x 0.0056 $ - mo.   =  $  per year
                                     liter-yr
Vacuum Assist Stage II System


    Assume 50% recovery of displacement loss and 50% recovery
    of breathing loss

Recovery factor = (1080 mg/liter)(.50)  + (120 mg/1iter)(.50)  =600 mg/liter

Example:

600    mg   x 37.900 liter x    kg    x   1iter x 12 mo.  x  $0.29
      liter      mo.           iuo mg    TBTTg   yr      liter
                                = $118 per year recovery credit
Equation reduces to:
    [Throughput (liter/mo)] x 0.0031 $ - mo.  = $ per year
                                     liter-yr
                                7-25

-------
                  Table 7-14.   WEIGHTED AVERAGE STAGE II  COSTS3

                                (Fourth Quarter 1982  Dollars)
Model Plant No.
     Throughput (liter/mo.)
18,925
75,700    132,475    246,025   700,225
     Emission Reduction (Mg/yr)     0.25
             0.99
            1.73
3.21
9.13
Capital Costs
Annual 0 & M
Capital Charges13
Taxes, Ins.
Recovery Credit
Net Annual i zed
$/Mg, Cost Effectiveness
5,654
495
778
226
104
1,394
5,653
6,098
621
837
244
417
1,285
1,302
6,604
826
906
264
731
1,266
734
9,766
1,068
1,328
391
1,357
1,430
446
14,790
1,789
2,004
592
3,861
522
57
 Stage II control costs for all three systems were determined by using an average
 cost weighted by the population of Stage II systems currently installed (i.e.,
 80% balance, 15% hybrid, and 5% vacuum assist).17

 Capital charges are based on a 10% interest rate and an equipment life of 15 years
 for balance and hybrid systems, and 8 years for vacuum assist.
                                           7-26

-------
this analysis.  The API costs,  given in 1983 dollars,  include use of
high-hang retractors, installation and annual  permit fees,  certification
fees, and annualization based on a 7.7 percent interest rate.  Comparison
of capital  costs is difficult because representative throughput ranges
are not given; however, the API estimates appear to be much higher than
those presented in Table 7-12.   Net annualized costs were generated by
API on a nationwide basis and a determination of per facility costs
from available data was not possible.  For  comparison, ARB  data are
presented in Table 7-15.  ARB costs are based on a 12  percent interest
rate and are given in 1979 dollars.  A comparison between these costs
and the average costs presented in Table 7-12 is difficult  because the
nozzle configurations and throughput ranges representative  of the model
plants chosen by ARB are not entirely consistent, with  the model  plant
characteristics of this analysis.   Given the corresponding  throughput
ranges, the ARB costs given in Table 7-15 for the 6 nozzle  stations
should be compared with model plant 3 of Table 7-12; the 9  and 12
nozzle cases closely approximate the parameters of model plant 4.
ARB capital  costs for all systems  range from approximately  5 percent
lower to 20 percent higher than the costs presented in Table 7-12.  The
ranges of net annualized costs reported by  ARB closely approximate the
ranges presented in Table 7-12.
     For ease of calculation, average costs were developed  so that
future costs could be generated based on loss of throughput due
to decreasing gasoline consumption.  The average weighted Stage II
system costs per model plant size  (Table 7-14) were adjusted by
percentages representative of each of the five model plant categories
given in Table 4-4 (i.e., 58 percent, 17 percent, 15 percent, 8 percent,
and 2 percent for model plants 1 through 5, respectively),  thus obtain-
ing an average weighted net annualized cost, capital cost,  and gasoline
recovery credit per facility.  Table 7-16 presents these average weighted
costs and credits on a per facility basis for calculation of proposed
control options with and without size exemptions.  Service stations
exempted from Stage II control  requirements were independent dealers
averaging less than or equal to 189,000 liters per month throughput and
all service stations with an average throughput of 37,900 liters per
month or less.
                                  7-27

-------
           TABLE 7-15.  CALIFORNIA AIR RESOURCES BOARD (ARB) STAGE II
                                    CONTROL COSTS
                                   (1979 Dollars)16
Model Plant No.

  No. of nozzles

  Throughput range
    (1i ters/mo.)
        3

        6
4

9
 4

12
(113,700-379,000)    (189,500-379,000)     (265,300-379,000)
Capital Costs
Balance System
Hybrid System
Vacuum-Assist
Net Annual ized Costs3
Balance System
Hybrid System
Vacuum Assist

7,372
9,618
15,502

11-1,324
327-1,641
2,826-3,397

9,133
11,962
17,813

653-1,591
1,043-1,981
3,585-3,992

10,844
14,332
20,074
-
1,287-1,850
1,762-2,325
4,340-4,584
  aNet annualized costs are given as a range to reflect the throughput range
   (113,700 - 379,000 liters/mo) of the costs reported.
                                         7-28

-------
 TABLE 7-16. SERVICE STATION STAGE  II  WEIGHTED  AVERAGE COSTS
                  (Fourth Quarter  1982  Dollars)
Cost Factor
1
Throughput 18,925
(liters/mo.)
NO EXEMPTIONS
% each MP category 58
Net Annual i zed Cost 1,394
Recovery Credit 104
Capital Costs*
1) 5,654
'2) 12,231
3) 5,308
EXEMPTIONS (< 37,900 liters/mo.'
% each MP category
NIb
I
Net Annual 1 zed Cost
MI
I
Recovery Credit
NI
I
Capital Costs - NI
1)
2)
3)
Capital Costs - I
1)
2)
3)
Model Plant Number
234
75,700
17
1,285
417

6,098
12,720
5,749
132,475 246,025
15
1,266
731

6,604
13,661
6,233
all stations and
40 36
1285
417
6098
12,720
5749
-
1266
731
6604
13,661
6233
-
8
1,430
1,357

9,766
15,768
9,450
Weighted Av.
5 Value
700,225
2
522
3,861

14,790
21,082
14,459
189,500 liters/mo.
19 5
80 20
1430
1430
1357
1357
9766
15,768
9450
9766
15,768
9450
522
522
3861
3861
14,790
21,082
14,459
14,790
21,082
14,459
NA
NA
1,342
427

6,384
12,989
6,036
for independents (I))
NA
NA
1270
1249
872
1858
7391
14,035
7042
10,771
16,831
10,452
Indicates installation of 1) balance, hybrid,  and vacuum assist systems,
 2) vacuum assist system alone,  and 3) balance  and hybrid systems.

bNI denotes non-independents, I  denotes independents.
                                     7-29

-------
     Gasoline consumption is expected to decrease with time and this
will have a corresponding inverse effect on net annualized cost figures.
As recovery credits decrease, net annualized costs will increase.   The
average recovery credit (with and without size exemptions) was decreased
annually by a percentage equal to the decrease in gasoline consumption
and the amount of decrease in each year was added to the average weighted
net annualized cost.  The projected consumption of gasoline is assumed to
be constant from the year 2000 to the year 2020, therefore recovery cre-
dits and net annualized costs will remain constant from the year 2000 on.
     In this manner gasoline recovery credits and net annualized costs
were determined on a "theoretical" basis, that is, assuming that the
Stage II equipment is capable of achieving control efficiencies
approaching theoretical rates.  In order to determine credits and costs
at actual  "in-use" efficiencies, it was necessary to take the recovery
credits previously adjusted  for the decrease in gasoline consumption
and apply  the following equation:
                                           In-use efficiency
In-use credits = Theoretical credits  x  Theoretical efficiency

The in-use efficiencies for  Stage II systems under State and Federal
enforcement scenarios are detailed in Chapter 3.0.  The theoretical
control efficiency of Stage  II systems is 95 percent.
     Stage II recovery credits and net annualized costs were also
adjusted  for control options combining both Stage II and onboard
technologies.  When these technologies are used in combination, the
onboard  system dominates and is credited with vapor recovery (note that
no monetary value  is assigned to  the onboard recovery  credit as it  is
assumed  that any recovery credit  is more than offset by the additional
burden of carbon canister weight  on the vehicle).   In  cases where
Stage  II  systems are used in combination with onboard, the recovery
credits  previously adjusted  for  decrease in  gasoline consumption were
further  adjusted by the  percentage of onboard controlled  consumption.
In other words,  if onboard  technology were estimated  to control
10 percent of  gasoline consumption in a  given year, then  the Stage  II
recovery credit  would  be 90  percent of  the value  determined from
application  of Stage II  technology alone.
                                   7-30

-------
      In summary,  recovery credits and corresponding  net annualized  costs
 were adjusted on  a per year basis, (see Appendix  G, Table G-2)  in
 consideration of  the following  factors:
      1)  Gasoline consumption experienced  a  gradual  decline  until the
          year 2000 when a constant consumption was assumed,
      2)  Control  scenarios in which Stage  II and onboard control
          technologies  are combined reflect the percentage of onboard-
          controlled consumption  in each year and result  in lower Stage II
          recovery credits and higher  net annualized  costs, and
      3)  Actual "in-use"  Stage II  system control efficiencies are less
          than theoretical  levels,  resulting  in lower recovery credits and
          higher net annualized costs  than  theoretical cases.
      Since  the equipment  lives of  Stage II systems vary  (15 years for
 balance and hybrid systems  and 8 years for vacuum assist), it was
 necessary to  determine  capital cost weighted averages from three phase-in
 scenarios:   (1) initial capital  cost  estimates would include installation
 of all  three  types of Stage II systems, (2) after each 8-year cycle
 only  vacuum assist systems  (5 percent of total  facilities) would require
 replacement and (3)  each  15-year period from the date of  initial instal-
 lation, both  balance and  hybrid systems (95 percent of total facilities)
 would be  in need  of  replacement.   All  weighted average costs, with and
 without size  exemptions,  are presented in Table 7-16.
     The number of  affected facilities under each of the control options,
 as well as nationwide costs, is detailed in Section 7.3.  The facility
 phase-in schedule, used to determine capital  costs and net annualized
 costs under both nationwide and nonattainment control scenarios, is
 similar to the schedule presented  for service station Stage  I control
 installations.
     7-2.5.3  Onboard Controls.   This  section summarizes the appropriate
 cost sections on onboard control  costs presented in Appendix C  of  this
 report.  As  shown   in Table 7-17,  an onboard vapor recovery system  is
 expected to  carry  a consumer cost of $13.32 for light-duty vehicles
 (LDVs) and $18.19  for light-duty trucks (LDTs).   Those LDTs  using  dual-
 fuel  tanks (approximately 20 percent)  may  require two separate  onboard
control  systems for a total cost of $36.38.  This is  a conservative
                                  7-31

-------
TABLE 7-17.  ONBOARD VAPOR CONTROL HARDWARE COSTS
                  (1983 dollars)
Incremental Costs
Component or Assembly
Charcoal Canister LDV/(LDT)
Purge Control Valve
Liquid Vapor Separator
Fill pipe Seal
Pressure Relief Valve
Hoses/Tubing
Miscellaneous Hardware
Vehicle Assembly
Systems Engi neering/Certification
LDV Totals: Vendor
LOT Totals: Vendor
Vendor
$3.99/(7.83)
0.74
0.71
1.12
0.44
1.90
0.40
—
—
$9.30
$13.42
Retail Price
$5.077(9.94)
0.94
0.91
1.42
0.56
2.41
0.51
1.00
0.50
Retail $13.32
Retail $18.19
                     7-32

-------
 assumption  since  costs  would  likely be  reduced by using one large
 charcoal canister rather than  two separate canisters.
      A  fleetwide  estimate  for  all LDVs  and LDTs can be determined by
 sales weighting the costs  given above.  Using the projected sales for
 1988  from Appendix C, and  assuming 20 percent of LDTs have dual-fuel
 tanks,  the  fleetwide average cost is calculated to be $15.08 as shown
 below.  For future calculations, this cost will be rounded to $15 per
 vehicle.
 (10.45  M) ($13.32) + (.8)  (2.778 M) ($18.19) = (.2) (2.768 M) ($36.38)
                   = $15.08
      The $15 per  vehicle is believed by EPA to be the average
 incremental cost  above  the cost of the present evaporative control
 systems.  However, there are reasons why the cost of onboard control
 systems could be  somewhat  greater (approximately $25), see Appendix
 C, page C-23.
      Note that these total  cost estimates per vehicle include fill pipe
 seal  and pressure relief valve modifications.  The net annualized
 onboard control cost was obtained by multiplying the capital  cost by
 a capital  recovery factor.   This capital recovery factor represents a
 10 percent  interest rate and an onboard control system equipment life
 of 10 years for light-duty vehicles and 11 years for light-duty
 trucks  (the lifetime of the vehicle).   A gasoline recovery credit is
 not given.  Available data indicate that fuel economy is not enhanced,
 as any  potential  fuel  credit will  be offset by the additional  weight,
 of the onboard control   equipment.
     The control  strategy proposed assumes that onboard controls be
 implemented beginning in 1988.  Data on the number of light-duty
 vehicles (LDVs) and light-duty trucks  (LDTs)  requiring onboard control
 in each year were necessary since  nationwide cost information was
 generated from per vehicle costs.   The number of new vehicles per
year, as well as vehicle scrappage rates,  were determined in  a manner
 similar to onboard emission reductions (see Section 4.2,  Gasoline-
                                  7-33

-------
Consumption and Model Plant Projections).  The nationwide costs of
onboard controls are presented in Section 7.3.
     7.2.5.4  In-Use Costs.  As a further comparison between Stage II
and onboard control options, an analysis was performed to evaluate the
costs incurred when considering in-use efficiencies rather than  -
theoretical efficiencies.  Section 3.7.3 discusses in-use efficiencies
in detail.  In-use efficiencies affect only the recovery credit component
of net annualized cost estimates (i.e., recovery credits decrease and
net annualized costs increase); capital costs would remain unchanged
from the theoretical case.
     Table 7-18 summarizes the cost and cost-effectiveness for nationwide
Stage II control options based upon in-use efficiencies associated with
annual inspections (86 percent) and minimal enforcement (56 percent).
These costs and cost-effectivenesses can be compared to the nationwide
estimates obtained from onboard controls, considering in-use efficiencies
(92 percent).  Table 7-18 also evaluates the effect on cost and cost
effectiveness if the additional emission reduction associated with
control of excess evaporative emissions with onboard controls are
considered (see discussions in Section 3.7.3).  Cost-effectiveness is
shown in Table 7-18 for both cumulative values (cumulative dollars/cumulative
emissions) and discounted values (NPV dollars/NPV emissions).
7.3  NATIONWIDE COSTS OF CONTROL OPTIONS
     Costs were derived for each of the industry sectors under both size
exemption and no exemptions options.  These control options were then
used to develop the costs for the regulatory strategies discussed in
Chapter 4.0.  The nationwide costs of the control options were
determined* by using the individual facility costs discussed earlier and
the total number of facilities requiring control in each of the years
evaluated (1986-2020).  As discussed in Section 4.2.2, the number of
facilities was considered constant throughout the analysis; however,
there were different phase in schedules for different control options
(see Section 5.1.1 for a more detailed discuss of facility phase in).
Table 7-19 illustrates the number of facilities requiring controls for
each of the industry segments.
                                   7-34

-------
CO
1
o
i—
o
o
Q
2
o
CD
O

o
Z CO

M *«/>
« 0 S-
^•4 SS ^O
UJ P-
C3 O O
•f. t-i Q
CO Ll_ CM-—
UJ CO O
z en CM
LU LU ^H O
i— i a> \o
ui z f»co
CO t— i fc. O>
(C «-4
Z CD =3— •
o z o>
CO H-4
O£ LU 4*

01
^
U
0>
UJ
1
+>
o
CJ



s
s
C/l
«
O

•o
s
«T*
^
a
e
+j
jg



^
a.
Z£
Id
CO
en
I-M




CJ


TT
IO
§
3


•5

o  cfi mr~ «a-p~
-H-H-^CMCMCM -^ -^






o oo o o o oo oo
CM «— 1 O *O Cn 10 O^ CMO
«-CM ^rr». meo

•H ^^ — < CM CM f> , ^H — <






eo r— 10 en-«rto oo oo
cMeno icto icie
ro m m r— o to «oi0 «oio
10 10 IA co en in en en en en
^•1 »"4 «^






inioiomieio enei CM CM
eneoineneou) enen enen




44 4^
M c «i e
CO) C CD O) CA 9 Cn
op o E cc ce
4^CJ **O ^"'^" ^^
O) UU UU O) CJ H) S)
•O O) O CO O 33 S3
^- 0.1k 0.1k «.lk M- Ik
% cne ME; cuoi« oioi*
C r«- C LU «> ^ CUJ r*L.S*Q. WkO>
OIBM C 10 •— <0 « «5
_ u r— O tj I— UOICU> OIOI>
4-> in— r— IB —~- i— •> — >— >— CJ 1— I— CJ
IO C 4J «3 E *•> *-" «J E -I-1 O O O U
Z OO1O1 3^ O-CUOI 3-^- O)-r- >,i- .C O> -^- >,-^ f
~- t- tn c c EWM c c v. .c r— f 4J w ^ .— .c *J
i 4JO=> e: — cj o =3 c — o cj c cj — =5 cj c o> -.-
O.OII *C. X X CJ 1 «C £ CJ > O >. S t>O>X
i-^ E.CCUJJSC: .c c
•— i 35 H- i-* 1 1 H-*— * 1 I^H-I 1 k4i I
X 0 i.
OIUJOO SEOO IO O O
en 0
re i i .0
-u c
co o

















_













c
Ol
0)
0.
o

C IB
O
"- -0
4 •
in Q)
•»- 3

V IB

CJ


CJ
e in
O 01
L.
-o o.

in *>
IB 0)
ea z
IB ^3
7-35

-------
                   Table 7-19.   NUMBER OF  FACILITIES  REQUIRING' CONTROLS
                                 FOR  GASOLINE MARKETING  OPTIONS
 FACILITY TYPE
TOTAL FACILITIES
                                                    MP 1
                                                               MP 2
                                                                          MP 3
STORAGE TANKS AT TERMINALS               630
F0R-HIRE TANK TRUCKS
   - Teralnal Trucks                  5,800
   - Sulk Plant Trucks
    (Ho Exemptions)                 16,300
   - 3ulk Plant Trucks
    (< 4000 Gal/Day 8P Exempt)         9,400
                                                                                     MP 4
                                                                                               MP 5
TERMINALS
8UU PLANTS
- No Exemption
- Exempt Bulk Plants
< 4000 Gal /Day
SERVICE STATIONS
;(ESHAP Alternatives
STAGE I
- Ho Exemption Option
- Exempt Stations
< 10,000 Gal/Month
STAGE II
- Ho Exemption Option -
- Exempt Stations < 10,000
Sal /Mo and Independent
Stations < 50,000 Gal /Ho
-Independents Only
CTG Alternatives
STAGE I
- NO Exemption Option
- Exempt Stations < 10,000
Gal /Ho
STAGE II
- Ho Exemption Option
- Exempt Stations < 10,000
Gal /Ho and Independent
Stations < 50,000 Gal/Ho
- Independents Only
500
3,040
3,040
197,900
33,100
383,200
120,600

6,895
2,395
114,900
36,200

240 135 105 20
3,400 4,000 560 80
3,400 4,000 560 80
114,800 33,600 29,700 15,800
— 33,600 29,700 15,800
222,100 65,400 57,500 30,700
47,500 34,900 30,700
10,500
4,000 1,170 1,035 550
1,170 1,035 550
66,600 19,600 17,300 9,200
14,300 10,500 9,200
3,150
—
--
—
4,000
4,000
7,500
7,500
2,700 .
140
140
2,200
2,200
800
                                               7-36

-------
     Table 7-20 presents the nationwide costs ,for each of the control
 options.  Cumulative costs represent the simple summation of the total
 capital and  annualized costs over the period 1986-2020.  The net present
 value  (NPV)  costs represent the 1986 present value of all costs over
 the analysis period, discounted at a rate of 10 percent.  The year 1986
 was used as  the basis for comparison of NPV costs between options
 because this was the first year in which any of the regulatory strate-
 gies took effect.  NPV costs were developed to compare the strategies
 and take into account the opportunity costs associated with spending
 money  now or in the future.  In this analysis, nationwide options (with
 the exception of onboard) were assumed to begin taking effect in 1987,
 all non-attainment area options were assumed to begin in 1986, and the
 onboard option was assumed to begin in 1988.
     These nationwide costs for the options were combined with the
 cumulative and discounted emission reductions found in Table 5-7 to
 calculate the cost effectiveness of each option.  Cost effectiveness is
 presented based on both the cumulative emission reduction and cost
 values and the discounted (NPV) emission reduction and cost values.
 Cost effectiveness is expressed as dollars spent per megagram of emissions
 reduced ($/Mg).  Table 7-21 presents the cost effectiveness values for
 each option  and for each pollutant (gasoline vapors, benzene, EDB,
 EDO.  Cost  effectiveness values could not be calculated for the tank
 truck  options because, although costs for the tank truck controls were
 calculated,  the emission reductions due to these controls were included
 in the bulk  terminal  and bulk plant emission reduction calculations.
 Since  the EDB emission reductions were the lowest, the cost effectiveness
 associated with these emission reductions are the highest.   The most
 cost effective of the nationwide control options appears to be the
 option requiring controls on bulk plants (while exempting bulk plants
with a throughput of less than 15,100 liters/day).  The least cost
effective option appears to be the requirement for Stage II equipment
 nationwide on all  service stations,  regardless of size (no exemptions).
7.4  NATIONWIDE COSTS OF REGULATORY  STRATEGIES
     The nationwide costs associated with the regulatory strategies
 discussed in Section 4.3 are a combination of the control options costs
                                  7-37

-------
TABLE  7-20.   NATIONWIDE  COSTS OF GASOLINE  MARKETING CONTROL  OPTIONS
Costs (Millions of 1982 Dollars) (1986-2020)
Facility Type
Terminals3
- Loading Racks
- Storage Tanks
Bulk Plants3
- Ho Exenptions
- Size Exemptions
Capital
Cumulative

598
33
653
462

1986 MPV

221
16
252
177
Net
.Cumulative

789
(247)
1,255
781
Annual i zed
1986 NPV

214
(65)
337
208
 Tank Trucks
   - For-H1re Terminal Trucks       57
   - For-HIre Bulk Plant Trucks
     o Ho exemptions              161
     o Size Exemptions             93
 Service Stations (Stage I)
   - Nationwide
     o Ho exemptions             1,008
     o Size Exemptions             423
   - All HA Areas
     o No exemptions               35
     o Size Exemptions              15
 Vehicle Refueling
   - Stage II
     o Nationwide
 23
 63
 36
378
159
 14
  6
  361
1,014
  585
2,234
  938
   81
   34
 96
270
156
590
248
 24
 10
- No exemptions
- Size Exemptions
o All NA Areas
- No exemptions
- Size Exemptions
o Selected HA Areas
- No exemptions
- Size Exemptions
o Onboard
7,823
2,977

2,350
895

893
340
6,836
2,847
1,086

977
373

371
142
1,787
18,710
6,323

5,841
1,973

2,220
750
9,666
4,869
1,628

1,672
558

636
212
1,921
  Includes the costs associated with trucks owned by the facilities.
                                           7-33

-------








CO
o
£



o
h-
o
o
ts
p
UJ

**
to
I



fiu
vo
en
••*

w

V
•i
^

o







0)
a.
£
U
10
u.

o
g
8
r-.
f-.


88 88 gg
o o «a: o o o o
|g ,
p-Tin"


o o o o
o o o o
o ou «a- vo
rt t-T i»Tfn"
~*

§00 o o o o
O O O 0 0 O
o oo << oo oo
o"
s
IT}
—1
i
1
CO



1
o"
8
in


g
CO
CO
vo





0
s
o



o







I










in
in e
c cZ
1 £
oi a
1— CO
00
O 0 1
in o
— i rt
8g
°-°. *
§"o" i
m°
cn w
f*«- in


00
0 0
o"o"
88. '
§s


gg
cocn^ i «s
S5





00
8S , <
Kg



CM cn i







38 '•".
<





Q
C
i

O)
=? «
O IS
t! i «
O) ^ IB ^/
X 0. 1— U
UJ E 3
01 01 U
OX 01 1—
Z UJ IO
II 0 *C
i
f> ^ VO P^

•^

o o o o
00 00
H it
— < CM ,*en
mia vot~
CM -H i-4


§o o o
o oS
ceTin" ^ToT
cn vo com





§0 00
o o o
w 10 fn
SCO VO CM
^ IO COW



SI? gS
mm ^- CM







o o . o o
vo 5? SS





*•*
a,
cn

t/> *rt M
— C C
o o cn
« •»• M VI ^ (A C
c ** c: to *» c -t-
o v a. o o a. o •—
I* "° 1^ u E"^ **
. C <1> E  z a; a:
01 4J ° K _ ° X
u * m z ^ »2
•^ z oo< oo u
5. ' ' i
trt ^
§0 00 00
o o o o o
oo" oo" o"o"
o o o o ooo
. vo^ r— r- Km
m f*. vo CM co m
PO ~* CM —4 CM -*

00 00 00.
c? o o o o o
o o o o o o
o o o o ooo
m vo ^- p«» *H co
i-» m ' in CM vo CM
o o o o o o
00 00 00
o^o^ o o o o
§o o o o o
o o o o o
m co CM cn m to
vo P-. P-. c\i cn m-
f» •"* CM t^ CM ^H

go o o o o
o o o o o
o o o o o o
80 o o o* o* u
o .00 ooo
SCM W O CO W
F- CO CO CM Cn


o o oo ooo
o o o o ooo
o^o^ o o^ ooo
«r co cTco" m"in" vo"
CM to cnm — i«r o
m— CM— m— CM




o o oo ooo
o o oo ooo
o^o^ °°, ooo
in o io"o cTco" m"
m— m lf> *^ in —4



00 00 OO 0
*2X S!S j-io 10
^H cp cn Ol O Cfi M
« » * * »






gg gg Sg g
CM^*-^ °.Ch CM O ^
CM -4 CM CM ^ -H*





M
03
a*
u

c vi e c
^o e o «c o
•o X»c: £1 *»= z ^c
*-a.o < 0.0 -^ cxo
X E=<>~ E^~ VE*^
i— » C +•» «C O>4^ 4J C>4-*
"-"^o xo.z x a. u xo.1^
g^^ Of— § f— " § «
*^ c
W? O II O 1 ! 6 1 1 O
t 1











0)
c
t"-
o
tA
s.
•o
01
!l
i— O>
"« *«!
h- 3
VI*
at
a> «n
ix> O
(J
c a)
•^ >


v»,g
s*
u *»
4-1 ai
ai <*-
C ib>
•9
C O
*»- c

2E
(A
2^
O ^
^- o
Q. in
O 0)
e *o -P-
a> u .j=
o ** ai
u. c >
2S =
4-> S 0
« CJ
O) "C3
•O JB Ol
e x 13
3 0 *»
o JZ in
in •*•
•«> in
•o in O)
Ol &-

>resent value,
>st-effectlvei
ird controls i
u o
zzo
v .a u
7-39

-------
that were just discussed in Section 7.3.  In many cases, the nationwide
costs for the strategies could be determined by simply adding the
nationwide costs of the appropriate options.  This is true for all
cases except those which combine Stage II and onboard controls.
     As was discussed in Section 5.1.3.2, when Stage II controls -and
onboard controls coincide at a vehicle refueling operation,  the onboard
controls will dominate.  This means that no vapors are returned to  the
service station which results in no recovery credits and higher annualized
costs associated with the service station vapor control equipment
(Stage II equipment).  The strategies also assumed that Stage II equipment
would be installed for only one useful life of 15 years (two useful
lives of 8 years each for vacuum assist systems).  At that time, Stage
II would not be replaced since onboard equipment, alone would result in
greater emission reduction.  This reduces the total capital  cost burden
of the regulatory strategies.  Therefore, the analyses of all -Stage II
system options, when combined with onboard in a regulatory strategy,
were calculated separately.  The costs for Stage II options, when
combined with onboard, are shown in Table 7-22.  Both the cumulative
and NPV of the annual and capital  costs for these options are much
lower than the costs shown in Table 7-20 because the analysis incorporates
only one useful life of the equipment (two useful  lives for  vacuum
assist.)
     Table 7-23 illustrates the estimated cumulative and NPV nationwide
costs of the regulatory strategies.  Generally, as the regulatory
strategies get more complex, the costs increase.  The lowest cost
regulatory strategies are those which would require Stage II in non-
attainment areas only and the highest cost strategies are those which
require the combination of Stage I, Stage II, and onboard controls
nationwide.
     Table 7-24 presents the cost effectiveness of the gasoline market-
ing regulatory strategies.  The cost-effectivess values were generated
from the nationwide costs in Table 7-23 and the nationwide emission
reductions presented in Table 5-9.  In terms of VOC, the most cost-effective
strategy was Strategy IV - Stage I controls.  The least cost effective
strategy was the no exemptions option of Strategy XII - Stage II  and
Onboard controls nationwide.
                                  7-40

-------
        TABLE 7-22.  COSTS FOR STAGE II  OPTIONS WHEN  COMBINED  WITH  ONBOARD
                             Costs (Minions of 1982  Dollars)  (1986-2020)
Stage II Options
                                 Capi tal
                                     Net Annualized
Cumulative
1986 NPV     Cumulative
1986 NPV
Nationwide
- Size Exemptions
- No exemptions
All NA Areas
- Size Exemptions
- No exemptions
Selected NA Areas
- Size Exemptions
y*K :.
- No exemptions

1,022
2,695

307
808

hiW
-•»' '*'"•"
307
.- •- ;• ' ••
842 vf; . 3, 589 --;, ;' = ";;" 1,575
2,207 9,521 : ^ ;-r:4,261

'"'" 289 ''"';/'" 1,040 T;: r"" -> 518
-;::^': 757 :;,: '. , • 2;803;7i'l:;!;vi:u; -J;l,429
f •::;,:, i.-;-i...5 yi*(«?hii ,r-y
;,;.-; 110 ;:-^c i395;«^ ^;7^ 197
ix'J.S ' 4i*i,'' - ' 9fWrff««iiKS: on . -fi
288 1,Q65.; , i: lgs?£ ,-543
                                       7-41

-------
TABLE 7-23.  NATIONWIDE COSTS OF GASOLINE MARKETING REGULATORY STRATEGIES
Costs (Millions of 1982 Dollars) (1986-2020)
Regulatory
Strategy
I. Baseline
II. Stage II - M.A.*
A. size exemptions
B. no exemptions
HI. Stage II - M.A.
A. size exemptions
B. no exemptions
IV. Stage I
A. size exemptions
B. no exemptions
V. Stage II
A. size exemptions
B. no exemptions
VI. Stage I & Stage II -
A. size exemptions
B. no exemptions
VII. Onboard
VIII. Stage II - H.A.* 4
Onboard
A. size exemptions
B. no exemptions
IX. Stage II - N.A. &
Onboard
A. size exemptions
B. no exemptions
X. Stage I 4 Onboard
A. size exemptions
B. no exemptions
XI. Stage II - N.A. &
Stage I & Onboard
A. size exemptions
B. no exemptions
XII. Stage II & Onboard
A. size exemptions
B. no exemptions
XIII. Stage I & Stage II &
Onboard
A. with exemptions
B. no exemptions
XIV. Gas Bz Reduction
A. 62A% Bz reduction
B. 81.3% Bz reduction
Capital
Cumulative

340
890
910
2,390
1,670
2,510
2,980
7,820
4,640
10,300
6,840
6,953
7,140
7,160
7,680
8,500
9,350
8,820
10,200
7,860
9,530

9,530
12,000
3,200
8,500
Net Annual i zed
1986 MPV

140
370
380
990
630
950
1,090
2,850
1,720
3,800
1,790
1,900
2,080
2,080
2,560
2,420
2,740
2,720
3,510
2,630
3,990

3,260
4,950
1,800
4,900
Cumulative

750
2,220
2,010
5,900
3,210
5,410
6,320
18,700
9,530
24,100
9,670
10,010
10,700
10,700 •
12,600
12,900
15,100
13,900
18,000
13,300
19,200

16,500
24,600
30,100-33,300
75,000-87,400
1986 NPV

210
640
s
570
1,700
860
1,440
1,630
4,870 .
2,, 490
6,310
1,920
2,120
2,460
2,450
3,370
2,780
3%360
3,310
4,820
3,500
. 6,180

4,350
7,630
7,390- 8,050
18,600-21,700
                                  7-42

-------
a.
SC
1
^•4

CB S
t—t
l__ _
u3 |
OS

S
UJ 1 1
_J *
O ==
OO 10

S 2


OS
£ 3
oo
UJ 3=
<— 1 g
CD
UJ
^™
OS
t—
CL.
0 | I
< 01 01
	 i «rt C
__; eu 01
§||
UJ - BO
-QS S
a
U- t z
OL*J 5
U
//* M
to S
UJ

UJ
S 1

Co* CO
UJ 2
u_
U.
LU
1— g
00
o
o
3
•ef
CVJ
I


UJ

CO



£>>


IO 4->
o»**

cc

• §|
o"o
ss
en"eo"
OH CM

§O
0
§o
coS
co"*-*1
CM tO


o o
88
ii

to fo
ro cn
«•* CM


§O
8
||

•o- eo
en CM
CM to

o o
8.8.
(A in
^f ••«



ii
co"o"
in *r
~* p*>





Is

CM





O O
V CO
0 C^
~* CM



IA
C
*.-2S
«c ** o
• Q.^-
z e 4-*
Q> Q.
1 Si
  *^ O
0) OKA e
«n to
«c *» • »
» to«teo
-H M



ii
88
~4 CO
CM m
•^ CM

II
o o
o o
m m
to in
CM in


o o
8.8.
oo
80
0
•, *
^f PO
CM tO
•-* CM


ii
||

eo T
to to
CM in

oo
33
Ss"



II*
^•It)
*f O






§8
CO C0_
^





o o
CM



(A
C
0 V)

s.S-5
QJ Q.
1 « i
1-4 X
»•* Cl CU

a> *^ o
cn « c


CO *t CO
t_^



O O O O
88 88
11 l§
r-«. O r*- to
-H ~H PO

|| 8|
§"o o"o"
o o o
v \n in co
LO -H (O  a. d> ex
x E x E
OJ QJ CJ tt»
X ^H X
i—iwo; •»• o> 4)
N N
O> •*- O 4) ••• O
cn in c cn «A e
IO (0

CO «C CO Crt -S CO
" ^



o o •
0 O
§o
o o
CM f
-H CM

o o
gg
II °
in r^.
CM ^


o o
33
8*8
°*°» °

m to

*^ CM


00
o o
°.p.
o"o"
38 o
* •>
in eo
CM V

o o o
8.8. 3.
cn en to
o o o
—•CM CM



o o o
8.8. 8.
in"w< f*

*H(M CM





33 g
to ro^ co^
*-T -7





CM CO O
f^« PO ^
•^ ^


1

i— i C
o «
cn*^ o
« a.-*-
sl§-
x
i-» CU O* T3

cn «n c o

4j . . e:
CO «CCO 0
1-4 «-H



88
P.P,
88
<— i ^.t
CO in

O O
0 0
p. p.
o o
§o
.°
oeo
in en
we-

oo
8.8.
II

o o

CO U)
-Z-*

38
00
gg"
o o

§§

to in
i|
CM CM



O O
8.8.
to" en"
^4 CM
CM CM





go
en in
— <"— T





00
CM O
OH «-<



•0 VI
S
*. ££
•c *> o
=E E *»
HI CL
1 g g
1-1 X
1-1 T3 O) O>
L* K
eu «•••: o
cn o in c

-4-> C • •
oo o «C aa
••4


*
0 O
o o
oo
0 0
r— CM
cn co
to to

o o
0 O
o o
O CD
P.O.
coo"
CM cn
CM OH

o o
§0
.p.
oo
33

Oto

SlO


oo
38
II

o o
CMW

CM OH
2°
8.8.
CM CM



oo
8.8.
co'eo"

CM CM





o o
cn eo
enr-
OH'OH





5S
oHoH



1A
•e e
O 1A

X E 5
tu eii
— X


O  C . .

CM CM

§O
O
-°«
oo
§0
C7*
-Hen"
to tn


§0
0
p. p.
§0
0

en en
in in
CM CM


o o
§0
.p.
o"o
83
• •
en CM
tow

o o
8.8.
CM en



o o
§0
„=>„
CM O

oHoH





SO
in







0 0
SS




«A
-o e
L. O (A
0 4J O
C E 4J
•0 Q) CU
X


Qt **"• O
cn «A c
#e
4J * •
CO < CO
X



0 O
8.8.
oo
ss.
CM UD
CM CM

88
P. P.
oo
o o
r— in
•a- to"
tn in


33
P.P,
o"o*
83

Ch in
CM VO
CM CM


00
O 0
00.
o"o
CD CD
00
• *
cn I*-
in r-*
in in

§o
o
55



o o
8.8.
P*TO
•*• f^






§0
o
^





o o
eSS
^^



(A
*«3 -0 =
t- 0 (A

I X E
•e eu o>
i— i X
»•* »-^ cu 
N
0} O> •*- O
cn cn iA c


co to -a; ca
*-*
x


||
o o
eo o*
in r-~

o o
0 O
0 O
88
CMO"
in in
o4 oH

00
33
s"§
p. p.

-4 en
OCM
tor~


O O
8.8.
oo
o o
P.P,

oo"
»r en
m in
OH OH
O O
eo en
in e*
CM en



o o
go
.0.
en in
in w
CM en





§0
CM
r-*> to
OH CM





o o
to to
tO CM
OH CM




b. C
a o vi
O -o- c

OJ 0.
•e x E
U V
NH X

N
W-o- o
en IA e
«

V> < CO
^

X

o o
o o
0.0.
80
0
in or
CM en

oo
og_.
oo
00
in «r"
in to


§0
0
o o
g"g"
°.°.

eh" en
in oj>
CM en


oo
00
o"o"
88
M •
to to
in to

II
r-Teo
ss



II
CO CM
tO CM
^CM





gg
O in^
•-T^T





o o
to cn
— t — *


( t


g, V,
a c
+* O IA
CO -o- C
 0
^ S *J
OH l- a> a.
•a x E
o a> «
cn'c  — 0

"-H < CO


X

3=5=




=!-!





«<:
3E SC






«




11
°»s.
o o
CO CO



§o
o
in'o* •
r*. PO M
cu
c
CU
^
U
OJ
u-
0)
2 z **
o
u
o
|Q
=
15
0
e
4J
gg ^
1
0)
cn
a
e c t.
0 0
*A ^- «— --
« ** 4-» (O
Cfl U U ^™
= || |
CO M N O
CO CO

u »4 »« c
=j ^-n •»-
•o • • o
o> CM *-• ^
Q£ 10 CD ^
> «C CO
1-4 «
x

7-43

-------
7.5  COST PER INCIDENCE REDUCTION
    An analysis was performed to determine the residual costs expended
per cancer incidence avoided for selected nationwide and nonattainment
area regulatory strategies.  The residual costs were determined by
obtaining the annualized costs of the controls associated with the
regulatory strategy and subtracting several assumed benefit values
per megagram of VOC emissions reduced.  The annual VOC benefits are
those in addition to cancer prevention, such as non-cancer health effects
and agricultural damage due to ozone.  The residual cost per incidence
is then simply the residual costs divided by the appropriate amount of
cancer incidences avoided by implementation of the regulatory strategy.
    In Table 7-25, several of the regulatory strategies are presented
with their corresponding emission reductions.  The emission reductions
are presented as the net present value of all the annual emission
reductions over the study period and as a reannualized value representing
equal emission reduction for each year of the study period.  Calculations
were then conducted to determine the residual costs after assuming
several different dollar values for the benefit of reducing each megagram
of VOC emissions.  For example, in Table 7-25 the reannualized emission ,
reduction associated with Stage I is 0.218 million Mg.  Multiplying
this emission reduction by each of the assumed VOC benefit values
yields the annual ized VOC benefit in dollars.
    Table 7-26 presents annualized cost (including control  equipment
costs and enforcement costs) and annualized incidence reduction due to
benzene exposure associated with several of the regulatory strategies.
The cost per cancer incidence avoided, assuming no additional benefits,
is calculated by dividing the annualized costs by the annualized incidence
reduction.  Table 7-27 takes this one step further by incorporating the
annualized VOC benefits into the analysis.  The values presented represent
the residual cost, assuming varying benefits for reducing VOC emissions,
of reducing cancer incidences due to benzene exposure.
    Table 7-28 contains a similar analysis compared to that used in
Table 7-27, except that Table 7-28 was developed using the  sum of
the incidences due to benzene and gasoline vapors.  It is assumed that
                                  7-44

-------

























OO
I—
1— 1
u.
UJ
•sy
UJ
03

O
0


O
^»
^

00
(—
0
o
of
0
H-
«j
S-
o
4-J
fl3
^—
3
co
cu
C£












• •
M-
O
CU
3
1o
^
•f-3
•r"
M-
CU
C
CD
CQ

O
0

















c
0
•(••
CO
• VJ
•f—
UJ






M-»
cu
3
fw>
rt3
>•
































































C
O
•r-
4^
O
•a
s





to
c
o
•r"
to
to
•r-
3
JS
UJ


















cn
2T
•^"^^
o
o
0
CM


CD
^r
^*^
O
o
uo
f"H

CD
^
O
o
o
t-H



CD
s:
0
o
uo




CO
3?"
^%^
o
uo
CM



^^
CD
s:

c
o
2-
^
S



_
en
^g*
O
•p.
!•«•
l_
•p—
S

>*
U
CU

o

<4«
t^

00


in
!**«,







to
LO





r*.
oo








cn









cn








^^
CO
o
o








10
co
O






•— ••
s
LO
^_^

<
•z.
1
1 — 1
I — 1

cu
co
(O
^_)
oo


LO
cn
CM






r-H
CM
CM




p^.
^»
t-H







j*









f^
CO







p^,.
^^
t-H
O








CM
sr
»-H




^
CO

c
o
•r-

to
ry?
1
t— i
i— i

cu
CO
(C
] *
oo


, 	 i
cn
t-H






OO
•=!•
t-H




LO
cn








CO









^>
CM







LO
cn
o
0








cX
0




to
LO
•~—
£.
o
•r-
•»->
its
^»
1
>— 1
l_4

cu
cn
to
4_>
OO






































































^— .
&Q
CM
CT)
^ -

^3
S.
(O
o

c
o


^" CT^
1 — .— <
CM LO






LO cn
O CO
CM 00




p^- cn
OO LO
T-H CM







CO O
to co
t-H








«* uo
co to







p>- cn
oo uo
t-H CM
o o








CM 0
OO LO
»-r CM





cu
> cu
'•*-> -f—
(O -r*9
^ (13
0 S_
0. O
fO CL
> 03
CU >
CU
0

3 3



7-45

-------
      TABLE 7-26.  BENZENE REGULATORY COSTS AND INCIDENCE REDUCED
Regulatory Strategy
(with size cutoffs)
(In-use-efficiency)
Stage I
Stage II-NA (87%)
Stage II-NA (56%)
Stage II-Nation (86%)
Stage II-Nation (56%)
Onboard (92%)
w/o evaporative
w/ evaporative
Annuali zed
Costs
($ Millions)3
91
62
52
183
146

199
199
Annual ized
Benzene
Incidence
Reduction'3
0.06
0.83
0.41
1.92
1.13

1.44
1.66
Costs
($ Mil lions
per Benzene Cancer
Incidence Avoided)
1,564
75
126
95
128

138
120
a Includes control equipment and annual  enforcement costs.

blncidence reduction after controls.  Before-control  annualized incidence:
 Stage I = 0.18, Vehicle Refueling =4.09.
                                   7-46

-------










LU
O

LU
a
1—4
O
^z
H- *

r**
LU
O

*^ **"^
O 00
(—
LU U_
Cu LU
•z.
OO LU
1— CQ
oo
O O
0 0
^»
^
o 2:
l^" l_^
t ^"**
^C -^
_l 3
:D oo
CD OO
LU <
Qi —

LU Q
Z LU
LU O
M l-H
z o
LU >
CO eC


r^
CM
1
f^

LU
l
CQ
^£
1—

























-a
CU
•o
o
>
^

CD
O ••
£Z ti—

to "O O
3 T- 3
s: o -a
c c a>
«c i— i a:






^•^
cn
O)
•p
n3
i-
+-*
oo

>ij
^
o _

fO
r^
3
cn
0)
C£.


cn
•^_
o
0
o

VJ-

cn
2:

O
0
LO
r-t


e ri
^r*
^v^
O
O
O
<— t


CD

^,
O
o
LO
4&"


cn
s:
*^**»
0
LO
C^I
•feO-











^— H
*J
O
£Z
 O O







O <^- LT> CT» ^ LD O
CO t-4 ^ ^t-_






O O^ T-H P**» ^D CO OO
co co in co cr» ^t-







c^ «^^ *^^ v,o ^^ c^ co
CM LO O r— CD CM CO















^ CO t-l CM CO •* tO
O CO -tf- O1 ,-H <* to
• • « • • • •
O O O I— 1 i— 1 I— I r-)









s? &s
to to
OO LO  cu
&5 &9 T- >
r-. to c c +J -r-
OO LO O O (O •»->
~ ^ ^ ^ (pv >p— ^— ^ t__ fQ
+-> 4J &S O S-
^ «^ (B (O CM CX O
•Z. -Z. Z. Z CTi n3 Q.
1 1 1 1 — > 
1— ( 1—4 1 — 1 1 — 1 1 — t "O QJ
S- O
CLJ CU  4^ £3
OO OO OO OO OO O









in
CM
1



II 1 —

"O rt3
 S-
03 q—
CU *~*
O CO
t— ' 1 s
CU CO
•a o
•r- 0
c -o
•i- cu

^ •r^


i *

c~
cn
• (-I*
i
-M
3
0

CO

>r_
<4_
01
O)
-Q

03
a

•a
c


CO
o

cu
M

7-47

-------











cet
O
UJ
a
O

«t
UJ
o
367
UJ
S
O
5e
UJ
0
o

C£
UJ
Qu

CO
f—
a
CO
os
ex.
ic£
5*
UJ
se
H-4
—1
O
CO


o
§
^
3E
UJ

J3E
UJ
CO
.
*!*
r^*

UJ
_J
«
*^c
h-
































*-».
CO
C£.
CZ3
§E
>•
CO
S
Of
£
a:
0
&
— «
_J

a;
L^
a.
o.
UJ
ca
>™4
CO
S3
W2C
_J
a.
^w{

~""J


































u
•o
o

o

.
C43
IO •
Q) CT
O.C
g§
•v* l/>
r- in
z
^^
in
4J
O
U




^^
Q) et
M U
1?
5-S





•g«
N U
||
11







>>
W

IQ
t.
s>
^*
o
"S


en
6>





O)

o"
o
Si
*^





§

g
in
•-i




en
^•4



O
^
1



CD
£
g
Si




I-" ^S
55
=1 N
•^ co
-O
*»3
0 f
"§S







^^
&

QJ
•r-
O
«^—
*
0)
in


C
H-l

M
a>
ai —
i o-o
1 **~ 3
mv>
in
Ql
« Tj
i- S
vl


in
a —
— 3

in
43 
^l
g 43
V)
in
«n
in
O)
•E 3
in
in
w5
I- 3
tn



in
15
V)





in
at
i > "Q
IO 3
t- 43
in
























o o o o o oo





o o o o o o o




o o o o o o o



o o o o o oo




O O XT <*> «3 » O
O O CO CM •» * O


o tn o co CM cMin


u
O co vo tn co co co


-H f*. ro o m men
fO *^ *^ *"*^ **^



en tn co i«. en O vo





CO CO r» VO CM CM O
— < XT CM C71 CO COCO
^4 VO CO CO CO O ^







O P«- CM f»- X*1 CM"*
en o »— < r** co ch ro
-H o ID -H evj vo co
^4 C\J -H •-< CM






»« »7
VO 'C*
co in 4) a>

r- S c: c 43 43
co vn o o " £ £
«C «c  >
_i i— , _( i— i ai o>
1- O
0) O) 0) O> O) 1Q -^-^
o en en O) CTI o ^ f
arsiaaajs
CO CO  fcO I/I O
7-48

























in


c
CO
u
o

^
01
^J
f—
IO
i
c

1
to

•

CJ
i
a i—
•e s
0) O
"o —
G in

0) •
o o
S °
•o -o
•^ o>
C3 N
C •?
 c
c %.
a
u

t.
Ol


in
O
«°







































w
u
3
Ol
ct
O)
u
01
•o
u
c


^
|g
^
•
C?
•t-

N
CQ
Q




























-















is
1
01

43
^
en

0)
3
g
if
£•
Ol


Ol
43
^J
10
5

M
01
U
^K
c
"^
o
i.
VV
u


-------
the incidences due to benzene exposure and the incidences due to gasoline
vapor exposure are additive since the respective exposure results  in
different types of cancer incidences (leukemia in the case of benzene
exposure and liver or kidney tumors in the case of gasoline vapor
exposure).
                                   7-49

-------
7.6  REFERENCES

1.   Bulk Gasoline Terminals - Background Information for Promulgated
     Standards.  U.S. Environmental  Protection Agency.   Office of Air
     Quality Planning and Standards. Research Triangle Park,  NC.   Pub-
     lication No. EPA-450/3-80-038b.  August 1983.   p.  .B-3 through B-5.

2.   Reference 1, p. 2-42.

3.   VOC Emissions From Volatile Organic Liquid Storage Tanks - Background
     Information for Proposed Standards (Preliminary Draft).   U.S.
     Environmental Protection Agency.  Research Triangle Park, N.C.
     EPA-450/3-81-003a.  June 1983.   p. 8-2 through 8-7.

4.   Control of Volatile Organic Compound Emissions from Volatile
     Organic Liquid Storage in Floating and Fixed Roof Tanks  - Guideline
     Series.  U.S. Environmental Protection Agency.  Research Triangle
     Park, N.C.  (Draft) August 1983.  p. 5-1 through 5-6.

5.   Arthur D. Little, Incorporated.  The Economic  Impact of  Vapor
     Control Regulations on the Bulk Storage Industry.   Report to U.S.
     Environmental Protection Agency.  Research Triangle Park,- N.C.
     Publication No. EPA-450/5-80-001.  June 1979.   p.  II-6.

6.   Pacific Environmental Services, Inc.  Study of Gasoline  Vapor
     Emission Controls at Small Bulk Plants.  Report to U.S.  Environ-
     mental Protection Agency, Region VIII.  EPA Contract No.
     68-01-3156, Task No. 5.  October 1976.  p. 6-1 through 6-10.

7.   Control of Volatile Organic Emissions from Bulk Gasoline Plants  -
     Guideline Series.  U.S. Environmental  Protection Agency.
     Research Triangle Park, N.C.   December 1977.   p.  4-1 through 4-11.

8.   Pacific Environmental Services, Inc.  Evaluation of Top  Loading
     Vapor Balance Systems for Small Bulk Plants.   U.S. Environmental
     Protection Agency, Washington,  D.C.  EPA Contract No. 68-01-4140,
     Task Order No. 9.  June 1977.   p. V-2, Table V-l.

9.   Pacific Environmental Services, Inc. Stage I Vapor Recovery  and
     Small Bulk Plants in Washington, D.C., Baltimore,  Maryland,  and
     Houston/Galveston, Texas.  U.S. Environmental  Protection Agency,
     Washington, D.C.  April 1977.   p. II-2, Table  II-l.

10.  Reference 5, p. 111-19.

11.  Reference 6, p. 3-4.

12.  Reference 1, Appendix B.

13.  Arthur D. Little, Inc.  The Economic Impact of Vapor Recovery
     Regulations on the Service Station Industry.   Report to  U.S.
     Environmental Protection Agency. Research Triangle Park,  N.C.
     Publication No. EPA-450/3-78-029.  July 1978.   p.  H-8.
                                   7-50

-------
 14.  Pacific Environmental Services, Inc.  Hydrocarbon Control
     Strategies for Gasoline Marketing Operations.  Report to U.S.
     Environmental Protection Agency.  Research Triangle Park,  N.C.
     Contract No. 68-02-2606, Task No. 13.  April 1978.   p.  6-4.

 15.  Luken, Ralph A.  Cost and Cost-Effective Study of Onboard and
     Stage II Vapor Recovery Systems.  U.S. Environmental  Protection
     Agency.  Research Triangle Park, N.C.  August 1978.  39 p.

 16.  Reference 12, p. 6-9.

 17.  Reference 11, p. H-6.

 18.  Economic Indicators: CE plant cost index.  Chemical Engineering.
     Vol. 90, No. 6.  March 21, 1983.  p. 7.

 19.  Telecon.  Norton, Robert,  Pacific Environmental  Services,  Inc.,
     with Simeroth, Dean, California Air Resources Board.
     August 23, 1983.

20.  Memorandum from Norton, R.L., Pacific Environmental Services, Inc.
     to Shedd, S.A., Environmental Protection Agency.   Decembe-r 20, 1983.
     Trip report to California Air Resources  Board.

21.  American Petroleum Institute.  Cost Comparison  for  Stage II
     and Onboard Control of Refueling Emissions.   API  Publication
     No. 4306.  Washington, D.C.   January 1984.   Appendix  V.
                                   7-51

-------

-------
           8.0  ECONOMIC IMPACT OF THE REGULATORY STRATEGIES

     This chapter examines some of the economic impacts associated
with regulatory strategies that apply nationwide and, in addition,
Regulatory Strategies VIII and IX.  (See Section 4.3 for a description
of the regulatory strategies.)  This chapter presents a sensitivity
analysis of the total cost of each of these regulatory strategies,
where total cost is defined as the 1986 net present value (NPV) of
both control and enforcement costs.  Price increases are then estimated
for each of the nationwide strategies.  The price increase per liter
of gasoline is based on average annualized control cost and average
annual gasoline consumption.   Price increases for light duty vehicles
(LDV's) and light duty trucks (LDT's) are assumed to equal estimated
unit cost increases.  Given these price increases, resulting declines
in quantity of gasoline, LDV's, and LDT's consumed are estimated.   The
chapter concludes with an examination of some of the impacts of the
regulatory strategies on facilities of different sizes and regions.  A
report to EPA contained in the docket provides a more detailed discus-
sion of the methodology, analyses, and results discussed in this
chapter.1
8.1  METHOD AND SCOPE
     The total  cost of a regulatory strategy is the principal measure
of economic impact presented in this chapter.   Total cost is the 1986
NPV of the control  cost incurred by firms and the enforcement cost
incurred by government agencies.   Total  cost is a "net"  cost in the
sense that "recovery credits" from reduced vaporization  of valuable
gasoline stocks are subtracted.   Other benefits of the regulatory
                                  8-1

-------
strategies, such as improved health and reduced crop damage, are not
considered here.  Consequently, there is no attempt in this chapter to
identify a regulatory strategy that balances marginal total cost with
marginal air quality benefits in monetary terms.  The total cost of
each of the 10 regulatory strategies considered is computed using the
control cost assumptions of Chapter 7 and the annual inspection
assumptions of Chapter 9.  This provides base case total cost
estimates, which are then broken out by industry sector.
     The sensitivity of total cost projections to the underlying
assumptions employed in Chapters 7 and 9 is illustrated by  changing
key variables and computing revised total cost for comparison.  These
variations cover the following:  constant gasoline consumption at the
projected  1986  level, declining gasoline consumption incorporated
through declining numbers of facilities with constant throughput as
opposed to constant numbers of facilities with declining throughput
(base  case), 5  percent and 15 percent  real  rates of  discount, and a
unit cost  of onboard control of $25 per vehicle tank (20 percent of
LDT's  have dual  tanks).   In addition,  the  effect of  varying levels of
enforcement effort  and compliance  on the total  cost  of  nationwide
Stage  II  strategies  is examined.
      In a perfectly competitive market, the control  costs  required of
 firms  will be  passed on  to  their  customers  in  the  form  of  higher
 prices.   In  this analysis,  it  is  assumed that  the  gasoline marketing
 industry is  competitive  and that  control cost  will  be passed along   •
 through the  stages  of marketing to the purchasers  of gasoline.  The
 gasoline price increases are  estimated by  computing a "unit cost
 increase."  The unit cost increase of  gasoline is  the control  cost  for
 the entire gasoline marketing  portion  of  a nationwide regulatory
 strategy, annualized over the  period of analysis  (1986-2020) and
 divided by projected average annual  gasoline consumption for the
 period.  Gasoline price impacts  are not assessed  for regulatory
 strategies that affect only nonattainment areas because such
 assessments would require consideration of local  market conditions.
                                   8-2

-------
 Average unit cost increases for LDV's and LDT's are inputs  into the
 analysis.   The unit cost increases are used to estimate reductions  in
 annual  consumption likely to occur in response to price increases
 equal  to the unit cost increases.   These reductions are termed "quantity
 impacts" and are based on a range  of pure elasticity values  published
 in the  professional  literature.
      The last section  of this chapter identifies  some of the distribu-
 tive  economic impacts  of Stage I,  Stage  II,  and benzene removal  control
 options.   In particular,  the differential  control  costs per  unit of
 output  for firms of  different size and for different regions of the
 country are considered.
 8.2   SUMMARY OF  COST COMPARISONS                               .   •   ..
 8.2.1   Total  Cost by Regulatory Strategy
     The NPV of  control  cost,  enforcement  cost, and total cost for
 each nationwide  strategy  are given in  Table  8-1.   The  base case  assump-
 tions used  in  developing  the economic  impacts  in  this  chapter  are the
 same as  those  used in  Chapter  7, including the  assumption that decreases
 in gasoline  consumption over time  would  result  in  declining  throughput
 at a constant  number of facilities.  Declining  facility  throughput
 results  in declining recovery  credits  over time, which materially
 affects both the magnitude and distribution  of  aggregate and sectoral
 costs.  Accordingly, an alternative method of incorporating  projected
 declines in gasoline consumption is considered  in  Section 8.2.3.
     The numbers in Table 8-1 may be interpreted as the amount of
 national income that must be committed to control  equipment and enforce-
ment to support a 35-year program of reduction  of  benzene emissions
 from gasoline.  The total cost values  in Table 8-1  range from $0.88
 billion to $21.7 billion.  Selecting regulatory strategies with exemp-
 tions decreases total cost--in most cases by one-half to two-thirds.
     As previously stated, the total cost of a  regulatory strategy
must include enforcement cost.  This cost represents real resources
 used even though the cost is not borne by the firms or industry sectors
 impacted.   Based on the Chapter 9  assumptions for annual inspections,
enforcement cost comprises 0.5 to  4.3 percent of total cost for
                                  8-3

-------
                 TABLE 8-1.   1986 NPV OF THE COSTS OF THE REGULATORY STRATEGIES
                                       (109 1982 dollars)
Regulatory strategies Option
IV.
V.
VI.
VII.
VIII.

IX.

X.
XII.
XIII.
XIV.
Stage I — nationwide
Stage II — nationwide
Stage I and Stage II —
nationwide
Onboard — nati onwi de
Stage II — selected nonattain-
tnent areas and onboard —
nationwide
Stage II — all nonattain-
inent areas and onboard —
nationwide
Stage I and onboard —
nationwide
Stage II and onboard —
nationwide
Stage I, Stage II, and
onboard — nationwide
Benzene reduction in
gasoline
A
B
A
B
A
B

A
B

A
B

A
B
A
B
A
B
A
B
Control cost
0.86
1.44
1.63
4.87
2.49
6.31
1.92
2.12
2.46

2.45
3.37

2.78
3.36
3.50
6.18
4.35
7.63
7.39-8.05';!
18. 6-21. 7°
Enforcement
cost
based on
annual
inspections
0.02
0.04
0.07
0.22
0.09
0.27
0.001
0.01
0.02

0.02
0.06

0.02
0.04
0.06
0.18
0.08
0.22
—
Total
cost
0.88
1.48
1.70
5.09
2.58
6.58
1.92
2.13
2.48

2.47
3.43

2.80
3.40
3.56
6.36
4.43
7.85
7.39-8.05'j
18.6-21.7
aEsti(«ates assume declining gasoline consumption, declining recovery credits, and a constant
 number of facilities.
 For Strategies IV through XIII, Option A exempts certain facilities from the regulatory
 strategies and Option B allows no exemptions.  For Strategy XIV, Option A requires the removal
 of 94.5 percent benzene from the reformate fraction for a total reduction of 62.4 percent;
 Option B requires the removal of 94.5 percent of benzene from reformate and fluid catalytic
 cracker fractions for a total reduction of 81.3 percent.
 The total cost is control cost plus enforcement cost.

 These control costs include expenditures required to boost gasoline octane and volume after
 benzene reduction.
                                                   8-4

-------
 regulatory strategies involving Stage I and Stage II controls.   Enforce-
 ment cost is only 0.07 percent of total cost for onboard controls.   No
 estimate of enforcement cost is made for benzene reduction in gasoline.
 8.2.2  Total Cost by Sector
      The following paragraphs briefly describe base case cost impacts
 by sector.   The estimates presented assume projected declines in
 gasoline consumption, declining recovery credits,  and a 10-percent
 real  discount rate.   The sectoral  impacts of alternative assumptions
 are discussed in Section 8.2.3.   (For a more detailed discussion of
 sectoral  impacts,  see Reference 1.)
      Regulatory Strategies XIV.A and XIV.B would impact petroleum
 refiners  most directly.   Strategy XIV.A would remove 94.5 percent of
 benzene  from the reformate fraction for a total  reduction of  62.4
 percent,  and Strategy XIV.B would remove 94.5 percent of benzene from
 both  the  reformate and the fluid catalytic cracker (FCC) fractions  for
 a  total  reduction  of 81.3 percent  (see Table 8-2).   Table 8-3 gives
 the 1986  NPV of the  control  costs  of benzene reduction required  of
 gasoline  producers.
      Also included in Table 8-3  are estimates of two additional  costs
 of benzene  removal.   The  cost  of octane  loss reflects  the cost of
 gasoline  additives needed to maintain  gasoline octane  levels  at  roughly
 pre-benzene-reduction levels.  The  cost  of volume  loss  is  the cost of
 additional  gasoline  production needed  to  make up for the volume  of
 gasoline  lost  due  to  benzene removal.
      Table  8-4 presents the  1986 NPV of  Stage I  control  and enforcement
 costs under  base case  conditions for bulk  terminals, bulk plants, and
 for-hire  truck operators.  Both  bulk terminal and  bulk plant  NPV's
 already include the cost  to  equip trucks owned at  these  facilities;
 therefore, for-hire truck cost is treated  separately.  Bulk terminal
 cost  also includes cost to control  fixed-roof storage tanks.
      For the bulk  terminal sector,  the NPV of Stage  I cost, including
 enforcement, is $246 million.  Enforcement cost  represents approximately
 0.4 percent of total cost.   Storage tank control cost is negative
while for-hire truck control cost is positive and substantial.
                                  8-5

-------
      TABLE 8-2.   ESTIMATED REDUCTION IN BENZENE CONTENT OF GASOLINE
                  RESULTING FROM REGULATORY STRATEGY XIV
Option
A
B
Estimated
volume of benzene
before reduction
(percent)
1.37
1.37
Estimated
volume of benzene
after reduction
(percent)
0.52
0.26
Reduction
in benzene
content of gasoline
(percent)
62
81
a
 Source:  Reference 2.
           TABLE 8-3.  1986 NPV OF THE COSTS OF BENZENE REDUCTION
                            (109 1982 dollars)
Costs
Control costs required
of gasoline producers
Investment requirements over
analysis period
Operating costs
Cost of octane loss
Cost of volume loss
Total cost
Option A
5.77
1.80a
3.97
0.28-0.55
1.34-1.73
7.39-8.05
Option B
14.27
4.77a
9.50
2.57-5.
1.75-2.
18.59-21




15 .
26
.68
Investment costs include onsite and offsite construction, equipment
  and materials cost, and working capital for a 3-year construction phase  .
  and a 20-year plant life.
  Total costs for benzene reduction do  not  include an estimate of enforcement
  costs but are the  sum of control costs to gasoline producers and the costs
  of octane loss and volume  loss.
                                      8-6

-------
     TABLE 8-4.   1986 NPV OF  STAGE  I  CONTROL  AND  ENFORCEMENT COSTS FOR
             BULK TERMINALS,  BULK PLANTS,  AND FOR-HIRE  TRUCKS
                             (106 1982  dollars)
Facility, option
Bulk terminals3
Bulk terminal costs
Bulk terminal storage tanks
Independent bulk terminal trucks
Total
Bulk Plants— Option A (exemptions)
Bulk plant costs3 -
For-hire bulk plant trucks
Total
Bulk Plants—Option B (no exemptions)
Bulk plant costs3
For-hire bulk plant trucks
Total
Control cost
213.0
-65.0
96.4
244.4
206.9
155.8
362.7
337.9
270.2
608.1
Enforcement
cost
0.8
0.3
—
1.1
6.3
—
6.3
6.3
—
6.3
Total
cost
213.8
-64.7
96.4
245.5
213:2
155.8
369.0
344. 2
270.2
614.4
Includes cost to equip trucks owned at the facility.
                                   8-7

-------
     For the bulk plant sector, costs vary by option.   Under Option A
(exemptions), the total cost is $369 million.   Enforcement cost contrib-
utes nearly 2 percent of this total while for-hire truck controls
contribute 42 percent.  Under Option B (no exemptions), the total cost
is $614 million.  Enforcement cost and cost to control for-hire trucks
represent 1 and 44 percent, respectively, of the bulk plant sector
total cost.
     Table 8-5 presents the 1986 NPV of Stage I and Stage II control
and enforcement costs  for service  stations.  Costs for the Stage II
program with onboard  controls are  also presented.
     The nationwide Stage I total  cost for service stations is $262.2
minion under Option  A and $624.4  million under Option B.  Enforcement
cost constitutes  5 percent of total  cost for each option.  When  Stage I
is  instituted in  all  nonattainment areas only, the total cost  is $10.5
million for  Option A  (enforcement  cost is 6 percent)  and $25.1 million
under Option B  (enforcement  cost is 5 percent).   (These costs  are
relatively low  because almost  all  nonattainment  areas are assumed  to
have Stage I controls at  baseline.) The  nationwide Stage II total
cost under Option A js $1.69 billion.  Enforcement cost constitutes
4 percent  of this total.  - Under Option B,  the total cost is $5.08  bil-
lion,  of which  4 percent  is  enforcement  cost.
     Costs are  calculated separately for options  that require  both
Stage  II  and onboard controls  because  in these  cases  Stage  II  equipment
is installed only for one lifetime for balance  systems (^15 years)  and
two lifetimes  for vacuum  assist (a total  of 16  years) as  onboard
equipment is phased in.   For Stage II  with onboard, decreasing recovery
 credits are assigned to  the stations as  the onboard population increases
 and the vapors  are therefore no longer returned to the stations.
 Thus,  while the annual capital cost for  Stage II is  lowered when
 Stage II is combined with onboard, operating cost increases over time
 due to the loss of the recovery credits.
      As a result, the total cost for Stage II with onboard differs
 from that for Stage  II alone.   The 1986 NPV of control and enforcement
 costs for nationwide  Stage II with onboard presented  in Table 8-5 is
 $1.6 billion under Option A and $4.4 billion under Option B.   Enforce-
                                   8-8

-------
        TABLE 8-5.   1986 NPV OF STAGE I AND STAGE II CONTROL AND
                 ENFORCEMENT COSTS FOR SERVICE STATIONS3
                           (106 1982 dollars)
Facility, option
Stage I — nationwide
Option A (exemptions) . -
Option B (no exemptions)
Stage I--all nonattainment areas
Option A (exemptions)
Option B (no exemptions)
Stage II — nationwide
Option A (exemptions)
Option B (no exemptions)
Stage II — nationwide with onboard
Option A (exemptions)
Option B (no exemptions)
Stage II — selected nonattainment areas
with onboard
Option A (exemptions)
Option B (no exemptions)
Stage II — all nonattainment areas
with onboard
Option A (exemptions)
Option B (no exemptions)
Costs pertain only to service stations
Control cost
248.0
590.6
9.9
23.8
1,622.2
4,859.9
1,565.9
4,259.8
195.8
543.3
515.0
1,429.1
and do not incl
Enforcement
cost
14.2
33.8
0.6
1.3
170.7
224.0
70.7
178.3
7.3
23.2
19.4
61.2
ude costs of
Total
cost
262.2
624.4
10,5
25.1
1,692.9
5,083.9
1,636.6
4,438.1
203.1
566.5
534. 4
1,490.3
onboard
controls.
                                    8-9

-------
merit cost is 4 percent of each total.  Note that Table 8-5 shows only
costs to service stations and excludes costs of onboard controls.
     In addition to the nationwide regulatory strategies, two additional
strategies covering nonattainment areas only were considered.  The
total cost for Stage II in selected nonattainment areas with onboard
is $203 million under Option A, with 4 percent of this total attributed
to enforcement cost.  Under Option B, total cost is $567 million,  of
which 4 percent is enforcement cost.  Total cost for Stage II in all
nonattainment areas with onboard is $534 million under Option A and
$1.5 billion under Option B.  Enforcement cost is 4 percent of total
cost for both options.
     Although Stage II costs change when combined with onboard control,
onboard cost is the same whether or not it is combined with Stage II.
The analysis uses unit cost estimates of $13.32 per LDV and $21.83 per
LOT (based on a weighted average of control costs for single- and
dual-tank trucks.)  Given these inputs, the 1986 NPV of onboard control
costs is $1,922 million.  This figure includes an enforcement cost
estimate of $150 thousand per year, which has a 1986 NPV of $1.3 million
or 0.07 percent of total cost.
8.2.3  Sources of Variation in Cost
     For several reasons, the total cost, unit cost, and quantity
impact of these regulatory strategies might deviate from the estimates
presented above and in Section 8.2.4.  This section investigates the
sensitivity of total cost estimates to certain variations in base case
assumptions.  The meaning of each variation is discussed below and the
impact of each variation on sectoral and total cost is presented.
Table 8-6 summarizes the results of this portion of the sensitivity
analysis.  Base case results are included for purposes of comparison.
     8.2.3.1  Constant Gasoline Consumption.  The base case cost
estimates presented in Table 8-1 incorporate a projected decline in
annual gasoline consumption of 23.3 percent over the period of analysis.
(Gasoline projections are discussed in Section 4.2.1.)  However, there
is always considerable uncertainty attached to projecting values over
a 35-year period.  Cost estimates are sensitive to these projections.
Accordingly, Table 8-6 provides estimates of the total cost associated
                                  8-10

-------











to
o
1— t

a.
yr

I/)
to


UJ
t—
•z.
oc
UJ

_J


C£.
UJ
a
z
rj
CO
UJ
s
UJ
1-

cc co

,_(

o: c
O i-
§ .2
UJ r-
ce o
•o
UJ
X CM
1— CO
cn
U. rH
0
01
r— O
CO rH
O *•>
_l
r-
g
UJ
X

u_
o


a.
z
CD
CO
cn




CD
1
CO

UJ

CO

1—











p
IA
o
u

•o
to
o

o




























0)
c

E
12
u
ftl
a


c
10
p
(A
C
o
o











































01
"u

j=
flirt
> '-/
X. C
in to
CM P
«•
t-
O

01
p
P ro
c c.
01
0 P
i. C
01 3
Q- O
I O
in in
•5
j.,
01
p
P ro
c s-
01
0 P
i. s:
ft) 3
a. o
1 U
in in
•5


u
14- in
o oi

s- -P
ft) •!—
o r— i

3 U
C ro
It-
c
0
ft) .,-
C P
•r- O.

0 3
tA IA
ro c
Ol 0
U



.a
ai oi
in in
ro ro
m u



ro
c
o

p
a
o






IA
ftl
Ol
01

ro

p
IA


i.
o
p
10

's
Ol
ftl
CfL





CO
CO
o









o
IX
O









in
CM
rH









CM
r~
o




px
f^

o








co
CO
o






^








01
•a

I
o


ro
c
1
1


ftl
Ol
to
p
CO


I-H





CO

rH









r^
H
rH









CD

CM









0
CO
r-i




co
ro

r-i








co
<3-
rH






CO


































cn
CO
,_H









CM
ro
rH










in
CM









CM
ro
pH




o


l-i








o
rH






<







01
-o


o

p
to
c
I
1
HH
rH

01
O)
10
p
CO








CO
o
in









CO
co
ro










CO
p^









CM

^.




f^.
ID

^








cn
o
in






CO


































i*-
in
CM









CM
o
CM









CO
r~
ro









in
o
CM




CD
rH

CM








CO
tn
CM






<





t
1
rH
rH

ai
Ol
to
P
CO

•a.
c
10

rH

01
Ol
10
P
CO


HH-





f^
CO
CD









in
0
in









ro
CO
cn









ro
IX
in




in
0

ID








8
CO






m
















01



c
o

p
ro
c








rH
ro
ro








•o

ro
iH








T3
O
rH
ro









CM
cn
rH




CM
cn

rH








CM
cn
rH















Ol
•a

I
o


10
c
1
1
•o

to
o
JD
c
o


rH





CM
in
CO









CM
in
rH










ro
ro









rH
rH
CM




p_4
H

CM








ro
rH
CM






<




1
C
O
c


Ol
u
01

ftl



rH
rH

U
Ol
to
p
CO


»





CO
co
ro











rH










IX
CO









f»^
1-
CM




r^
•3-

CM








?
CM






CO







•o
c
ro ft)
TJ
(A •!-
ro 3
01 C
S- 0
ro -r-

P ro
c c
01 1

C S-
•r- 10
10 O
P JQ
-P C
ro o








VD
CO
ro












3
rH












in
IX
co











ro
CM







CO
*
CM







£
CM







<









1
C
o
c

r—

ro

I
rH
I— 1

01
Ol
ro
P
CO






r- '





ro
CO
^












cn
ID
CM












cn
CO
4-°











cn
ro
ro







cn
ro
ro







CO
ro







CO




ftl



c
o

in p
ro ro
Ol c
S- I
(0 I
TJ
P S-
C 10
01 O
§J3
C
•r- O
to
p -a
p c
ro ro












cn o
rH CO
*^ <3*











•a T3
•3" rH
o in
CM CM











•o -a
in co
ro CM
^ in











•3- CM
CO CM
CSJ CO







cn o
co ro
CM ro







O 0
CO «»•
CM CO








C
O

•o
c
ro Oi
•a

i-i 3
C
01 O
Ol-r-
ro P
P ro
CO c




rH
rH
X





rH -3-
CO CM
in cn












cn CM
ro CM
ro co.











•o -a
rH in
«*• co
ID O
rH










rx cn
rH CO
•31 IX







rH 0
CM CO
^> f*.







CO IT)
«t CO








< CO





•o
c
ro
ftl
-•o
rH -r-

C
01 O
01 .r-
ro P
P ro
CO C
I

rH TD
S-
01 ro
Oi O
ro JD
p C.
CO O






X






I
'










CM
ro
in
1
CM
cn

•3-






cn
o

c^
rH
1
m
co

CM
rH







1
1








1







in
o
CO
1
cn
CO

rx





<







c


c
o


u

•o
01
s-c

01
c
ftl
N
c
01
m





rH
X






1
1









rH
ro

"3-
rH
CM
^

CM
H





in
cn

rs^
ro
CO
rH

CM
ro







l
i








l







^
rH
CM
CO
CO
rH





CO


















D
01
C


o
IA
10
01







01

ro u
P ro
o o s-
c p u

3 .;_
o s- P


to ro
c p
co o ro
•r- U
C P
O O "O
•f- ro -i-
P t- 3
a. i- •—
O H-
01
•0 P T3
C ro C
ro E to

IA O 0>
11 It- p
•r- ftl ro
Ol t. S
01 £-
p 01 O
to jr. i+-
S- p ftl
P S-
IA E
O E
>,£ 0
O 1-
p ftl
ro c ai
r- ftl C
3 N Ol
Ol C N
01 ftl C
s- a 01
a
ai P
J= C 14-
P ftl O
sS-
O ftl C
I- D. ftl
<4- U
in s-
IA . 01
Q) ^* o
•r- cn
p in

r- O *3-
— cn
O r-
it- > O


•i- oi ro
ro s- >
t a. 1
ftl JZ ftl
O P t
in in ot •
ts- P c
•r- 0)
oi 3 in u
x. cr oi s.
ftl ft) c. ,01
t- -i- O.
< 3
•4 O" CO
C ftl .
0 C S- rH
•r- 0 CO
p .r- CO

o a. c o
o o

rH - P O
r-l > 0.-1-
rH rH O P
XX 0
•- 3
jr >, p -o
Ol Ol C 01
3 ftl 01 S-
O P 0
S- to S- i—
c i- Ol ro
P P D. P
CO O
> «3- P
rH £-
O CM 10
(A U. CD
Ol i-
•i- It- O
O) • O »t-
fti in
p c c in
to o o c
S- -r- -r. 0
p p p •(-
CO D, U P
E 3 U
i- 01 T3 ro
O X 
•D
S
u
>J

01

o
fj
ftl
s.
Ol
c

c

13
01
•o

o
p
D.
E
3
in
c
o
u
Ol
c

'o
IA
ro
Ol

Ol
C

E


u
ftl


01
E
3
in
tA
to

tA
01
P
10
E

P
(A
UJ
a
in
0)
c


o
ftl

c
o

p
ro

"D ,
a.
o
a

E

O
E
3

^3
C
ro

C
0

p


3
IA
O
u

ftl
c


o
(A
to
Ol

Ol
c
c

"u
ftl


IA
ftl
p
ro
o
a.
i.
o
u
c

p
ro

g)
O

*o
•a
o

P
QJ

QJ



O U)
l4_ Q)



(0 0)
QJ
•a P
QJ JO
•P U
c
QJ -P
w c
QJ CO
S- r-
a a.

W r—
0) QJ

3 O
CD E


C
QJ O
-C E

U
















QJ
U)

JZ

>»

•a
QJ


Q.
E


QJ
•P
ro
s_

•p
c
. 3
O
u
in

"a


(0
QJ
S-
QJ

C


in
0)
u
c
QJ
S_
*P


•D

O

"D
QJ
3

(TJ


O
C

in

in
O

•P
C
o
u


S-
fO
o

c
o


o

^_)

o
u

p
c
3

0}
.c

D
O)
c


U) C
QJ 3
S- O
3 U
•P UJ

•5 -5
C
QJ QJ
a. ^
X -P
QJ
c •

to
p *•
•r— S- '
a. QJ
fO >
u CD
t- o
o .c

c -
o ~o

P P
(t) O
N QJ

i — 4—
fO QJ
3 S-
C
C QJ
ft) i-
to
c
to in
QJ
C P
o to
£-

0) -P
in c
m 3
A O
u
t/) in
"D QJ
P -P
ro -P aj
JZ C E
+J QJ -r-
S- -P

O) C QJ
JC 3 J=
P O -P
u
w in -P
o T- co

P in
C (U P
QJ £ in
C -P O
Re U
E "- -a
0 £-
U QJ CO
D) O
>, C jQ
C (0 C
< .c o


CO fO

in > c
CD .D C
in co

i — QJ QJ
CO 4-> -P
c u to
tO QJ Ol
**- QJ

P CO D)

> QJ CO
•P QJ
••- TJ JT
U5 r— +J
C 3
QJ O «*-
in .5 o
















C
o

•P

3
"D
QJ
S_

QJ
C
QJ
- N
C
QJ
.2


QJ
+j

TJ
QJ

3

o


"O
. c
to

QJ
C
CJ
O
QJ
C


O

to
O)
•p
in
o
o
.Q
O

•D
QJ
3
a*
QJ
t.

in
Q)
S-
3
-P

C
QJ
Q.
X
QJ

QJ
T3
3

u
c


in

in
O
u

,_
o

_j_>
c
o
u

QJ
W)
QJ
.C
H-
QJ
8-11

-------
with the regulatory strategies when annual gasoline consumption is
assumed to remain constant at the projected 1986 level of 326 billion
liters (86 billion gallons) per year.
     When constant gasoline consumption is assumed, the total cost
estimate for bulk terminals decreases 55 percent if storage tank costs
are included and 38 percent if storage tank costs are excluded.  For
bulk plants, constant gasoline consumption decreases estimated total
cost by 14 percent under Option A and 7 percent under Option B.  For
service stations, nationwide Stage II total cost decreases 17 percent
when exemptions are allowed and 8 percent without exemptions.  (Stage I
costs for service stations are unaffected because service stations
receive no Stage I recovery credits.)  The total cost of nationwide
Stage II with onboard, which still incorporates some decline in recovery
credits reflecting the phase-in of onboard controls, decreases only
about 6 percent under Option A.  Total cost for Stage II with onboard
strategies in all nonattainment areas and in selected nonattainment
areas decreases 7 percent under Option A.  Decreases are only 3 percent
for Option B under Stage II with onboard both nationwide and in non-
attainment areas.
     As shown in Table 8-6, constant gasoline consumption decreases
the total cost associated with the nationwide regulatory strategies by
2 to 17 percent depending upon the individual strategy and option
examined.  For strategies that affect only nonattainment areas, total
cost decreases 0.4 to 2 percent.  Constant gasoline consumption also
affects the NPV ranking of regulatory strategies.  Under Option A,
Regulatory Strategies VI and IX exchange rankings, while under Option B,
Regulatory Strategies VI and XII exchange rankings.
     8.2.3.2  Declining Gasoline Consumption with Declining Numbers
of Facilities.  Impact estimates are sensitive to the manner in which
the gasoline marketing industry is assumed to accommodate projected
declines  in gasoline consumption.  Two alternative methodologies—one
varying recovery credits and the other varying the number of facilities
and only  indirectly varying total recovery credits—can be used to
estimate  the cost impact under projected declines in gasoline consump-
tion.  The declining recovery credit methodology that is incorporated
                                   8-12

-------
in the base case assumes that projected declines in gasoline consumption
are shared by all existing facilities,  with no facilities closing.
The result is a constant number of model  plants with throughput
declining over time in each model  facility.  The declining number of
facilities method is based on the  assumption that declines in gasoline
consumption would result in closure of marketing facilities, which
decreases the number of model plants over time with each facility
maintaining a constant throughput.  Two variations of this assumption
are considered:  uniform decline and skewed decline.  Under the former
variation, the decline in number of facilities is assumed to be uniform
across model plant categories.  Under the latter variation, smaller
model plant categories experience greater declines.  This skewing
reflects recent industry trends toward larger, more efficient facilities.
     For bulk terminals, when the percentage decline in number of
facilities is assumed to be uniform across model plant categories, the
total cost decreases 71 percent from the base case if storage tank
costs are included and 49 percent if storage tank costs are excluded.
When the decline in facilities is restricted to the three smallest
model plants (skewed decline), these total cost estimates decrease 60
and 42 percent, respectively.  For bulk plants, if reductions in
facilities are assumed to be uniform across facility size categories,
Stage I costs decrease 18 percent under Option A (exemptions) and
11 percent under Option B (no exemptions).  If the same decline in
throughput is restricted to the two smallest model plant size cate-
gories, Stage I total cost decreases 18 percent under Option A and
13 percent under Option B.
     For service stations, the total cost for Stage I programs (nation-
wide and in nonattainment areas)  decreases 6 percent when the station
population is decreased in uniform proportions as consumption declines.
Under the uniform decline, Stage  II with onboard total cost decreases
6 percent under Option A and 3 percent under Option B for Stage II
control applied nationwide.  For  Stage II with onboard controls in all
nonattainment areas or in selected nonattainment areas, total cost
decreases 7 percent under Option  A and 3 percent under Option B.  The
                                  8-13

-------
total cost for Stage II programs without onboard control decreases 22
percent under Option A and 13 percent under Option B.  When it is
assumed that population decreases would be greater among smaller model
plant groups rather than larger ones (skewed decline), the total cost
for Stage I decreases 7 percent under Option A.  All other changes in
total cost to service stations are the same under the skewed population
decline as they are under the uniform decline.
     Table 8-6 shows the total cost figures associated with the
regulatory strategies assuming that the projected decline in gasoline
consumption results in comparable declines in numbers of marketing
facilities.  The figures presented are for reductions in facilities
that are uniform across facility size categories rather than skewed
toward smaller model plants.  For regulatory strategies that include
either Stage I or Stage II control (all except Regulatory Strategies
VII and XIV),.the total cost decreases 0.4 to 22 percent when the
facility population decreases over time.  The smallest declines are
associated with regulatory strategies that include onboard control.
Regulatory strategies that include only Stage I and/or Stage II control
exhibit 13- to 22-percent declines in total cost when a decline in the
number of facilities is posited.
     Clearly, the total cost impact of declining gasoline consumption
depends upon the manner in which the industry accommodates this decline.
If the decline is to be absorbed through decreases in the number of
facilities rather than decreased throughput at all facilities, cost
estimates decline.  Moreover, the cost ranking of regulatory strategies
depends on the methodology chosen to incorporate declining consumption.
The  relative positions of Regulatory Strategies VI.A and IX.A are
affected as are the rankings of VLB and XII.B.  In  recent history,
much of the gasoline marketing industry has been characterized by
declining numbers of facilities and increasing average  facility size.
This pattern supports the assumption that declining  consumption would
result in declining numbers of facilities rather than decreases in
average throughput.
     8.2.3.3  Discount Rates.  Varying the discount  rate applied to
streams of costs  is tantamount to changing the economic weight given
                                   8-14

-------
to the future.  A lower discount rate gives more weight to future
costs, while a higher rate gives less weight to these streams.  However,
because streams of costs may be negative due to recovery credits, the
net effect of changing the discount rate is not easily predictable.
     For bulk terminals, when storage tank control cost is included,
reducing the discount rate to 5 percent increases the total cost
13 percent and increasing the rate to 15 percent reduces this cost
7 percent.  When storage tank costs are excluded, decreasing the
discount rate increases the total cost 36 percent and increasing the
discount rate reduces it 17 percent.  The direction of these changes
is dictated by recovery credit streams for bulk terminals and bulk
terminal storage tanks,  for bulk plants, reducing the discount rate
from 10 to 5 percent increases the total cost 35 percent under
Option A and 39 percent under Option B.  Increasing the discount rate
from 10 to 15 percent reduces the total cost 17 percent under Option A
and 19 percent under Option B.
     With the 5-percent rate, the total cost at service stations for
Stage I controls nationwide is raised 46 percent above the base case
under both Options A and B.  Total cost for strategies involving
Stage I in nonattainment areas only increases 37 percent.  Nationwide
Stage II without onboard total cost increases 48 percent under Option A
and 51 percent under Option B.  Increasing the discount rate from 10
to 15 percent decreases the total cost for nationwide Stage I 22 percent
under both Options A and B.  Total cost for Stage I in nonattainment
areas is decreased 16 percent under both options.  Under Option A,
total cost for nationwide Stage II without onboard controls decreases
22 percent; under Option B it decreases 24 percent.
     For both the 5- and 15-percent rates, the costs for nationwide
Stage II with onboard vary less dramatically than for Stage I and
Stage II without onboard.  This is because most installations of
Stage II in combination with onboard involve only one equipment
lifetime, and no further Stage II costs are incurred after this point.
At a 5-percent discount rate, the total cost for nationwide Stage II
with onboard goes up 27 percent under Option A and 26 percent under
                                  8-15

-------
Option B.  Total cost for Stage II with onboard in all or selected
nonattainment areas increases 18 percent under both options.  At a
15-percent rate, the total cost for nationwide Stage II with onboard
decreases 17 percent under Option A and 16 percent under Option B.
The decrease is 11 percent under both options for Stage II with onboard
in all or selected nonattainment areas.
     In the case of onboard controls, reducing the discount rate from
10 to 5 percent increases the total cost over 60 percent.   Raising the
discount rate to 15 percent reduces that estimate approximately 30 per-
cent.  Note that in the case of onboard controls, the per-vehicle cost
is not adjusted for differences in the real discount rate.  (Any
component of the unit cost estimate that is based on an annualization
of capital expenditures would be affected by a change in the discount
rate.)  The different discount rates are reflected only in the discount-
Ing of the aggregate annual onboard costs at the given unit cost
estimate.
     Based on the estimates provided in Table 8-6, lowering the discount
rate from 10 to 5 percent increases the total cost associated with the
regulatory strategies by a low of 37 percent (Regulatory Strategy
XII.B) to a high of about 74 percent (Regulatory Strategy XIV--benzene
reduction).  Most regulatory strategies exhibit an increase of 40 to
50 percent.  Reducing the discount rate changes the ranking of Regula-
tory Strategies IX and X on a cost basis.   Raising the discount rate
from 10 to 15 percent reduces the total cost associated with the
regulatory strategies by a low of 20 percent (Regulatory Strategy
IV.A) to a high of approximately 34 percent (Regulatory Strategy
XIV—benzene reduction).  Under Option B,  raising the discount rate
affects the cost ranking of Regulatory Strategy VI.   Under the base
case, Regulatory Strategy VLB is more expensive than Regulatory
Strategy XII.B; under the 15-percent discount rate,  these strategies
are equally expensive.
     8.2.3.4  Onboard Control Costs.   While EPA's best estimates of
onboard control costs are used in the cost analysis,  it is instructive
to consider what effect substantially higher onboard costs would have
                                  8-16

-------
 on  the  rankings of  regulatory  strategies.  Accordingly, a higher
 per-unit cost  is also considered.  The total cost of strategies
 including onboard control  rises by $1.4 billion  (72 percent) when the
 cost  is $25 rather  than $13 per vehicle tank.  The effect of this
 variation on the total cost rankings of the  regulatory strategies
 depends upon the option selected.  Under Option  A, Regulatory Strategy
 VI  becomes.cheaper  than Regulatory Strategies VII, VIII, and IX when
 the higher onboard  unit cost is used.  Under Option B, the higher unit
 cost  of onboard control reverses the total cost  rankings of Regulatory
 Strategies VI  and XII.
      8-2.3.5   Enforcement  Cost.  Total costs are sensitive to assump-
 tions concerning the level of  enforcement required to maintain various
 levels of in-use efficiency.   The enforcement costs in this chapter
 are based on annual inspections; however, scenarios assuming more or
 less  frequent  inspections would result in different costs.   Two alterna-
 tive  enforcement scenarios are considered in this section.   In keeping
 with  the in-use efficiency analysis in Chapter 7, these variations
 involve only Stage  II controls.  The first sensitivity analysis assumes
 quarterly inspections of Stage II facilities.  In the second sensitivity
 analysis, "minimal  enforcement" assumptions are  incorporated.   As
 discussed in Chapter 3, a minimal enforcement situation is one in
which no State or Federal  resources are allocated to Stage II program
 enforcement.
     When quarterly inspections of Stage II facilities are assumed,
 nationwide Stage II total  cost is increased 7 percent under both
Options A and B.   Total cost for Stage II with onboard is increased
about 6 percent under both options.   It should be noted that compliance
 levels under quarterly enforcement are assumed to be the same as under
annual enforcement  in this analysis.
     When the minimal  enforcement scenario is applied,  enforcement
cost is eliminated,  but several other changes also occur.   Therefore,
the net effects of such a scenario are difficult to assess.   Since
fewer facilities  comply,  lowered capital  cost can also be assumed.   As
in the "in-use efficiency" analyses  of Chapter 7 (Section 7.2.5.4),
                                  8-17

-------
it was assumed that, due to noncompliance, minimal enforcement control
cost would be 20 percent lower than the cost under full enforcement.
Also, since fewer facilities comply, credits for vapor recovery are
correspondingly reduced.  Total cost for nationwide Stage II is reduced
15 percent under Option A and 20 percent under Option B.  Stage II
with onboard control total cost decreases 19 percent under Option A
and 21 percent under Option B.
8.2.4  Unit Cost and Quantity Impacts
     Substantial cost variations translate into widely varying unit
cost and quantity impacts for the  regulatory strategies.  These values
are shown in Table 8-7.   (As discussed  in Section 8.1, unit  cost and
quantity impacts are estimated only for nationwide strategies.  Conse-
quently, Regulatory Strategies VIII and IX have been omitted from
subsequent tables.)  The  estimated increase in unit cost  of  gasoline
ranges from 0.034$/liter  to 1.17t/1iter (0.ISC/gal Ion  to  4.43
-------





to
UJ
(3
LLJ
t
g
H-
>
1

— J
S
UJ



Q
rH
O
rH
1-


O 1-H
U. l~
CL

O CO
UJ CO
U- <
u.
UJ UJ
£5
^— UJ
=E to
Z> CO
o*
ce
O UJ
z o
< z

i —
CO
0
o
1—
z


LLJ
CD
ce
LU



i
LU
.J
CO








•o
c
O ^^
o E n
-J 3 O
I/I rH
C C ^
O O
•P
o
3
01 C
S- O
r- 4J i-
5 > i- >>
C Q E \
O -J 3 CT
•P C rH
10 o ^-*
C U
a
CT
(0

> U S-
** £ -?>
o \.
Ol •!- S-
C 4-> Ol

O S r—
in in
10 C (D
o o o
U iH
^•^


J3
If- S-
4J 0 01
in oi 4->
O 01 C -r-
O l/l -r- r-
10 1— \
•r- S- U)
C U 10 CM
n c oico

i-H
'*•'


C
O

&









>1
D}
P
(0
S-
•p
>>

o
p
10


Oj
01
ct:





it ii








ii ii
ii ii







00 rH rH CM
H CM CM CO








* co in *3-
ro in ic cri
O O O rH
O O O O





< m < co







O)
Ol T3
•O •!-
'r- 2
2 C
C 0
O i-

P «
ro c
C 1
i i
rH HH
O> Oi
O) Ol
ro to
CO CO





>-« >



II ro to ro
in in in







ii r>.° i-: i>;








•3- «3- 1 ro rH
rH Lf) 1 ^- «*
*f O rH CM

rH






 01
*O TD ro T3 T3
C -r- C C •!-
ro 3 i ro 3
C 1 C
rH O -O rH O
01 4-1 ro oi P
01 ro o 01 ro
ro c JD ro c
4-> C 4->
CO O CO



«
rH rH
> > X



ro ro
in in







r-: f>:








0) Ol
ro to
«3- CM
CO O

i-H






Ol 01
CM in
00 «3-
O CM
o o





< CO






t
"O
s-
10
o
c
o

•o c
c -a
"•5
l-l C
rH O
01 P
01 ro
10 C
CO



^
rH
X



ro ro
in in
















01 01
ic co
00 CO
*fr CM

rH






01 01
CD CM
rH 0
rH ro
0 0





<. CD








rH
I
a> i
O1T3
ro s- 01
4J ro -o
CO O •!-
ri s
» c c
rH O O
01 c ro
ro 10 c
p
CO



«

X



1 1
1 1







1 1
1 1






o o
en ic
I-H *(•
1 1
O 0
CD O
CD CM
rH ti-





ro r~
* rH
O rH
1 1
O rH
11^ O
o' r-!





< CO











01
c
o <—
4-> U!
01 O 10
C 3 Ol
N 01 C
C S- -r-
co



..;

X


4J
C
10
4->
in
c
o
u
10
•a
c
10
„
I/I
4J
?
S-
o

&
01
•>
o
u
01
£-
Ol
c
c


u
01
•o

«
c
o
4J
in
c
o
u
Ol
c
, —
o

ro
Ol

Ol
c
c

"o


Ol
•o
_2
u
c

ul .
C I/I
O 01
4J P
D.-I-
E r-
3 -r-
in u
in ro
ro !r-

01 n-
01 O
10
u s-
01
01 JD
CO C
ro





































CO
ro

>,
>,

"a

4J
3
E

O

ro
O)
\

S

L.
01
>
c
o
u
r-





HI
4-1
I/I
ai
i. .
S 01
in u
01 £
E a
*o c
10
E 01
•o ro
ai
14- S-
o u
c
>>•"-

•r- P
U C
•r- 01
P U
l/l t-
10 01
r- O.
0) 1
rH
01
U 10
S- O
a. P

01 01
1— "D
01
in.5
in p
O P
1- C
0 3
Ol • t-
C 0)
•r- O.
'o -o
in oi
ro "o
01 C

o a>
&1
•r- 4-1
U V-

P C
in ro
r- O-
01
c

u

'£ o
ft5
ro o
Ol "O
c a.-
•i- S-
m

3 o>
•o ro
01 P
P C
3 01
Ife

U 1



^
o
i-
p
c
o
u
•o
i.
ro
&
c
o •
rH
!- rH
01 •
•O rH
j;
3 in
10
1/1 1
01
in ~o
10 01
0) in
S- 3
o
c in
•^ 01
01 U
O 'r-

i- 01
a. >
Ol 4J

0 01

.C r—
> »-
O
££
(U U
^ U)
r ,2
4J OJ
p
re a)
u
o ••-
S- S-
*0 CL
O £
••- 1—
4_J
t .
3 in
in ai
C in
o ro
U 01
r- O
0 C

•o oi
c u
ro *^
> Q.
a
c
c •.-
o

C (0
O Ol

P O
O P


3 .






























3
O
1
01
ro
a.
, t

01
Ol
ra
P
CO

o
t(_
01

in
4-)
u
01
14—
01

p

p
c
ro

•o
ro

P
in
o
u

P

a>
8-19

-------
         I         *>
         I      4J  C
         *      C  30
         I   I  Cl  O
         I'   U) U  U  <
         I   wi t-  U>
         I      Cj 1-
         '      C_ Tl
s
               *>  e
o o  CM o
                                      O r-i     O CM
O O  O O
                                                                          O O  CM O
                                                                                          •r-H  O CO O
                                      O  00      C
                                                                                                                                      01    0)
                                                                                                                                    I •!-    t-
                                                                                                                           (J  4->    t.  S-
                                                                                                                     !—•(-(/}   0)  -P
                                                                                                                     f—   r-  C   -P  fO   U)   •
                                                                                                                      m    o  o   co   
                                                                                                                      O)   v)  u   
-------
these estimates are discussed in terms of deviations from the base
case, which assumes declining gasoline consumption, declining recovery
credits, and a 10-percent real discount rate.
     When constant gasoline consumption at the 1986 level is assumed,
unit cost impacts decline 29 to 35 percent under Option A and 26 to 27
percent under Option B.  These declines are attributable to different
recovery credit streams and a different level of production over which
costs are spread.   When declining gasoline consumption accompanied by
a declining number of facilities of constant throughput is assumed,
unit cost impacts decline 18 to 23 percent under Option A and 13 to 14
percent under Option B.  These declines are attributable principally
to different recovery credit streams.                      ....       >
     The unit cost impacts under various discount rates are also
presented in Table 8-8.  If the discount rate used to evaluate future
streams of expenditures is 5 rather than 10 percent, unit gasoline
cost impacts generally decline 9 to 16 percent.   (The phasing of
Stage II and onboard controls under Regulatory Strategies XII and XIII
produces some differences in impact patterns.)  The largest decline is
experienced under Regulatory Strategies IV and X, where the unit cost
impact decreases 15 percent under Option A and 16 percent under Option B.
Assuming a 15-percent discount rate, the unit cost impact increases 0
to 18 percent depending upon regulatory strategy and option selected.
The smallest increases occur under Regulatory Strategies XII and XIII.
All other strategies exhibit increases of 8 percent or more.
     Certainly, the magnitude of average unit cost impacts depends
upon the costing assumption chosen.   Moreover, the ranking of regulatory
strategies by average unit cost impact sometimes changes when assump-
tions are varied.   (Regardless of option, Regulatory Strategies VI and
XII switch rankings when a 5-perceht discount rate is substituted for
the 10-percent discount rate.)  It should be noted, however, that such
switches occur only when the impacts associated with the regulatory
strategy are close in magnitude.
     Quantity impacts are estimated by applying price elasticities of
demand for gasoline and automobiles to estimated average unit cost
                                  8-21

-------
increases.   Table 8-9 presents the sensitivity of gasoline quantity
impact estimates to differing costing assumptions.  Quantity adjustments
are not used in this preliminary study to refine cost estimates of the
regulatory strategies.  Such refinement is difficult to make and is
not likely to have significant impact on the costs.  Assuming constant
gasoline consumption, quantity impacts decline 13 to 18 percent under
Option A and 7 to 8 percent under Option B.  Under the assumption of a
declining number of facilities, quantity impacts  decline 18 to 23
percent under the exemption options and 12 to 13  percent under the
nonexemption options.
     Table 8-9 also presents  estimates of quantity reductions under
different discount  rates.  When a 5-percent  discount rate  is used, the
effect on quantity  impacts ranges from a 5-percent increase for Stage II
with  onboard under  Option B to a  decrease of 16  percent for Stage  I
alone under Option  A.   The wide variation  is primarily due to the
different recovery  credit streams implied by the strategies.  Using  a
15-percent  discount rate, the quantity  impacts  either  remain constant
 (Regulatory Strategy XII.B)  or increase  from 2  to 17 percent.
      For the purposes of the analysis above, a  long-run gasoline
 demand  elasticity of 0.55 was used,  based on Department of Energy
 estimates.3  It should be noted,  however,  that  no consensus  exists as
 to a specific, national, long-run elasticity of demand for gasoline,
 and that 0.55  is a mid-range estimate.   For sensitivity,  low-  and
 high-range estimates that bracket the published estimates were substi-
 tuted.3  Table 8-10 demonstrates the sensitivity of gasoline quantity
 impact estimates to variations in price elasticities of demand for
 gasoline.  Using a higher estimate for gasoline price elasticity
 nearly triples the quantity  impact estimates calculated using the base
 elasticity.  Using the  lower price elasticity for gasoline more than
 halves initial quantity impact estimates.
      Table 8-11 shows  the sensitivity of automobile quantity impacts
 to changes in per-unit control cost  assumptions.  Increasing the  unit
 cost for onboard controls to $25 per vehicle tank (about  a 90-percent
 increase in unit cost) increases LDV quantity reduction 88 percent  and
 LOT  reduction  38 percent.

                                   8-22

-------



00
LU
to
Ul
u
TZ
r— I
I—
00
o

1-
z
3
LU
111
>

O Z
r- O
O r-
Ul a.

^ ~^
CQ V) (•—*
l-l 00 (-
o: < IB
ti-Sl
^

3 «0
O LU ^->

— ••
3.

rH Hi
Si
g =
z
hH

W
§
a
LU
ce
en
i
CO
LLI
1—






+>
4J C
C 3 01
1 01 O *>
in u u ID
rH S- Ul S-
a. TJ


*> 'c
C 3 01
01 O P
1 O O 10
in c- in s.
01 •*-
Q. TJ

.Q
Ul
Ol«t- 01
C 0 i-
C i. -r-
•r— 01 r—
u e u
01 3 10
a c «-



«..£
C C -P
ID *r- D.
•P r- E
Ul O 3
C in in
one
o 01 o
u



ai oi
in in
IB IB
CO U






c
o
p
o.
o


01
10
s-
m
o
10
"3
01
01
oc





rH CM



CM O
rH CM





rH rH
rH CM








•J- «!•
CM CM
H CM




CO rH
rH CM







< CO




01
TJ
1
-natib
i
o>
01
10
•p
00



t-l

eo co
0 0
ro en



ID CM
CO CM
CM r**





O «3-
i-l 0
CM r-.








CM PO
CM *
CM r»




rHCM
CM oo







< CD




.1
o
.p
03
C
i
i
o
D)
(0

UO



>

m m
r*. co

co en




rH in
«*• o

rH






10 TJ
"'I
01 O
O1T-
10 <->
•P ID
00 C



>

1
1



1





1








1
1




,
1













01
-natio
i
•o
£_
ID
O
a
c
0

.

>

rH CM



o r-
CM O
rH CM





f- rH
rH rH
rH CM








- *f *3-
CM CM
rH CM




PO rH
rH CM







< m




i
i
•o
s-
o
c
o
•a
c a>
10 TJ
01 O
Ol-r-
ID -P
•P ID
OO C



>C

CTI o ID o in r*» PO f**
CO O ~~" in rH CO CM
rH ""rn""'


en o IT>^ O CM ID CM
CM O It rH rH CM
rH




COOVOO ID 00 CO rH
CO O ^f rH CM CM

rH rH





< m < CD



1 'O
•a IDO
ID -TJ
O HH •!-
.Q rH 3
o ai o
O) 'r.
"O 10 +J
CO 4J 10
ID 0) 00 C
"O 1
KH *r- . |
l-l S 1— T5
C t,
ai o at ID
Ol'r- O! O
P ID +J C
OO C OO O

H^
rH HH
X X

1 1
OOOO
co en in CM
rH rH «• in


OOOO
o vo co in

rH rH PO *



Illl
Illl








i i i r
il il




OOOO
ID Cn O ID
ID r- CM CO

rH rH «• 


>
5





















in
m
o

4-
O
0)

^
o
in
ID
Ol
0
2"
•r-
U
•f-
.p
tn
01
01
u
i-
O-
ID
Ol
C
in
3

^j
3
&
O



p
C
2
a
1

O!
C
o
E
10
in
0)
c
u
a>
TJ

g
o
<*-
c
3

"
ID
O
Q.
O
U
c

^
ID
£
<->
oi

r^
O

o
£
•p
i
.c
*>

i.
£
g
resent*
a.
u
2

•?
'+-
0)

3

u>
C
O
. u
3
•a
2
>>

p
c
C7
4->
C
01
in
s-
a.
0)

IA
a>
in
.c
P
c

m
O.
C
Ul
1

Dl
Lu

•£,
o
1
01
in
Q
o.
rH
rH
Ol
Dl
ID
p
00
2 .
O P
<*- 3
01 O
.a i
in ui
C ID
O S.
5 a
U t-t
3 t-t
•D
2 S,
ID

^_
P W
c ai
ID P
3 Ir-
o- w
U
8-23

-------
  TABLE 8-10.   GASOLINE QUANTITY IMPACTS:   AVERAGE NATIONAL REDUCTIONS IN
              CONSUMPTION UNDER CONSTANT GASOLINE CONSUMPTION
                    WITH VARIOUS ELASTICITY ASSUMPTIONS3
                            (106 liters/year)

IV.

V.

VI.

VII.
X.

XII.



XIII.



XIV.



Regulatory strategy
Stage I — nationwide

Stage II — nationwide

Stage I and Stage II —
nationwide
Onboard — nat i onwi de
Stage I and onboard —
nationwide
Stage II and onboard —
nationwide


Option
A
' B
A
B
' A
B

A
B
A

B

Stage I, Stage II, and A
onboard — nationwide


Benzene reduction in
gasol ine



B

A

B

Base
elasticity
,(n = 0.55)
143
241
271
812
414
1,054
—
143
241
343
(0)
1,026
(0)
486
(143)
1,268
(241)
1,660-
1,790
4,200-
4,860
Higher
elasticity
(n = 1-59)
413
697
784
2,349
1,197
3,046
—
413
697
990
(0)
2,967
(0)
1,404
(413)
3,664
(697)
4,800-
5,160
12,100-
14,000
Lower
elasticity
(H = 0.24)
62
105
118
355
181
460
,
62
105
149
(0)
448
(0)
212
(62)
553
(105)
725-
779
1,830-
2,120
aThe price elasticity of demand (n) measures the percentage reduction in
 quantity demanded per unit time due to a 1-percent increase in price.
 Quantity reductions before Stage II phase-out.   Figures in parentheses
 represent quantity reductions after Stage II phase-out.
                                     8-24

-------
      TABLE 8-11.   REDUCTIONS IN VEHICLE CONSUMPTION ATTRIBUTABLE TO
          UNIT COST INCREASES UNDER ALTERNATIVE COST ASSUMPTIONS
LDV

Unit price =
$13/vehicle tank
Unit price =
$25/vehicle tankc
103/yra
17.7
33.3
Percent
0.16
0,31
LOT
103/yra Percent13
5.3 0.18
7.3 0.25
 Computed using a price elasticity of light vehicles of 1.11.
 Percentage of average LDV or LOT consumption over the period of the analysis
 (10.78 x io6 LDV's per year; 2.91 x io6 LDT's per year).
'Represents about a 90-percent increase in unit control cost.
     TABLE 8-12.   REDUCTIONS IN VEHICLE CONSUMPTION ATTRIBUTABLE TO A
  UNIT COST INCREASE OF $13/VEHICLE TANK WITH VARIOUS PRICE ELASTICITIES'
                                  LDV
                                   LOT
                          103/yr
            Percent
             103/yr
         Percent
Base elasticity
(H = 1.11)

Higher elasticity
(H = 1-70)

Lower elasticity
(p = 0.22)
17.7
27.2
 3.5
0.16
0.25
0.03
5.3
8.1
1.1
0.18
0.28
0.04
 The price elasticity of demand (r)) measures the percentage reduction in
 quantity demanded per unit time due to a 1-percent increase in price.

 Percentage of average LDV or LOT consumption over the period of the analysis
 (10.78 x io6 LDV's per year; 2.91 x io6 LDT's per year).
                                     8-25

-------
     Table 8-12 demonstrates the effect on the quantity Impact for
vehicles of varying price elasticities.  Assuming a higher price
elasticity of demand for light vehicles increases estimated quantity
reductions over 50 percent from the base estimate -for both LDV's and
LDT's.  Assuming a lower price elasticity reduces estimated quantity
reductions approximately 80 percent.  Clearly, quantity impact estimates
are very sensitive to price elasticity assumptions.  Moreover, estimates.
of the price elasticities of demand for light vehicles vary consid-
erably from study to study.4
8.3  DISTRIBUTIVE IMPACTS
     Imposing a standard, whatever its design, will affect the profit-
ability and competitive position of firms and the well-being of con-
sumers.  Some industries may benefit from increased demand (e.g.,
carbon canister producers) if onboard controls are selected, and some
firms may be better positioned than others to respond to a standard's
requirements (e.g., a new service station may already have installed
Stage I and Stage II systems in anticipation of a standard requiring
them).  There will, -in short, be "winners" and "losers," but these
gains and losses are not necessarily economic costs in the strict
sense of the term.  The distribution of these outcomes over classes of
firms and individuals as well as over time is, however, part of the
economic impact.
     As consumers and firms adjust to the new set of prices and costs
resulting from a standard, a number of transitory conditions are
likely to occur.  Holders of fixed capital may find that their rate of
return is less than what led them to inves.t in, that line of business.
Because of the nature of the fixed capital commitment, however, the
facility will continue to produce until operating costs cannot be
recovered.  In the short run, then, the firm is selling its output at
less than a normal rate of return.  Similarly, a firm may find that,
because it has production facilities in place, it  is in a good position
to take advantage of the new cost and market conditions and that its
profits are higher than expected.  Over time, capital depreciation and
reinvestment opportunities permit firms to leave or enter various
                                  8-26

-------
 lines of business and new, normal profit levels of activity and industry
 structure will be established.  Some of these adjustment gains or
 losses and the longer term effects of the new cost conditions on
 industry structure are discussed below.
     The regulatory strategies covering Stage I, Stage II, and benzene
 removal undoubtedly would impact the competitive balance and structure
 of gasoline production and marketing.  The data summarized in Table
 8-13 show that under the regulatory strategies the average cost of
 smaller distributors would increase more than that of large distrib-
 utors.  Where measurable, the ratio of additional control cost per
 unit of throughput for larger facilities to additional cost per unit
 of throughput for smaller facilities ranges from 3 to 76.  (For top-
 loading bulk terminals this ratio is undefined because the unit cost
 increase for the largest model facility is negative, but close to
 zero.)
     The qualitative analysis presented in Reference 1 shows that such
 cost increases would most likely result in fewer, larger facilities.
 Because of economies of scale, this reduction in the number of facili-
 ties would be more than proportional to the quantity reductions
presented in Table 8-7.   This situation would amplify a trend already
well established in the gasoline marketing industries:  a 9-percent
decline in bulk terminals between 1972 and 1978, a 30- to 40-percent
decline in bulk plants between 1972 and 1982, and a 36-percent decline
 in conventional service stations between 1972 and 1982.   Use of exemp-
tions reduces but does not eliminate the cost differential between the
 largest and smallest facilities that have to control.  Exemptions
would actually improve the competitive position of the exempt firms and
 increase competitive pressure on firms just above the cutoff levels.
     Differential regional impacts also would be associated with the
 regulatory strategies involving gasoline marketing.   Rural and attain-
ment areas would experience relatively greater cost increase because
these areas generally have smaller average facility sizes and do not
 have baseline Stage I or Stage II controls.   For similar reasons,
benzene removal would result in about twice the average cost increase
                                  8-27

-------
TABLE 8-13.  DIFFERENCE IN AVERAGE UNIT COST FOR SMALL AND LARGE FACILITIES
                   UNDER DECLINING GASOLINE CONSUMPTION
         Facility
Smallest
  model
 facility.
(
-------
for refineries in Petroleum Administration for Defense District (PADD)
IV (northern Rocky Mountains) as in PADD III (Gulf Coast).
     It is difficult, on this preliminary basis, to estimate whether
regulatory strategies involving onboard control would result in differ-
ential cost impacts or changes in industry structure.  If,  on an
international basis, automobile producers are competitive and U.S.
producers are the marginal producers, most of a sales reduction would
accrue to U.S. producers.
8.3.1  Petroleum Refineries
     The distributive impacts of the benzene reduction regulatory
strategies can be analyzed by both facility size and region of the
United States.  According to ADL's 1978 report,5 the control cost of
benzene reduction varies dramatically with the scale of the refinery.
For small refineries (10,000 barrel/stream-day), ADL's scaling factors
result in direct control costs of reduction per liter of gasoline that
range from three to six times the average direct cost of reduction for
U.S.  refineries as a whole.   For a small refinery with both reforming
and FCC capacity, the cost multiple is closer to three.  Differential
unit costs, based on the national averages of Table 8-3, are displayed
in Table 8-14.  With an average wholesale gasoline price of about
$0.26/liter ($0.99/gallon) (as stated in Chapter 7), these figures
suggest greater than a 3.6- to 9.0-percent rise in the average unit
cost to small refineries.  The U.S. average unit cost increase would
be about 1.2 to 2.9 percent, depending on the regulatory option.
     A substantial number of small producers operate in the petroleum
refining industry and, to the extent that each of their refineries is
also small, these figures indicate they would be put at a substantial
competitive disadvantage relative to gasoline producers operating
large refineries.  Based on an analysis of differential cost shifts,
these regulations would tend to reduce the number of small  refineries,
especially if, as suggested by scale factors and age distribution,
small refineries are the marginal producers of gasoline.  The regula-
tion, operating in conjunction with forecast declines in gasoline
demand over the next decade, would likely accelerate this industry
                                  8-29

-------
    TABLE 8-14.   INCREASED COSTS TO PETROLEUM REFINERIES DUE TO BENZENE
                REDUCTION IN GASOLINE FOR A 10,000-BARREL/
                 STREAM-DAY REFINERY VS. THE U.S. AVERAGE
          Costs
Option A
Option B
Increased unit costs for a 10,000-
barrel/stream-day refinery with
reforming and FCC capacity, 
-------
 "shakeout."   Furthermore,  cost  differentials  could  increase  if  the
 cost of capital  for small  refineries  were  higher  than  that for  large
 refineries.
      The cost of benzene  reduction  varies  considerably with  geographic
 region.   When price indices  are applied  to results  of  ADL's  1978
 report,6 the  distribution  of estimated average  unit costs for each
 PADD is  obtained;  these are  shown in  Table 8-15.
      Other markets,  especially  chemical  markets,  may be affected by
 benzene  reduction  regulations.   The ADL  report,7  particularly,  cites
 the  possibility  of a benzene  glut, which it considers  a cost of regula-
 tion.   Environmental considerations aside,  it is  economically and
 commercially  more  correct  to  consider the  additional benzene produced
 by benzene reduction as a  credit or benefit of  the  regulation.  This,
 indeed,  is what  ADL  did when  it used the difference  between benzene's
 value in  gasoline  and the  next  best alternative use  of benzene  to
 estimate  the  cost  of volume  loss.  Chemical benzene  was wholesaled at
 $0.41/1 Her ($1.55/gal) in May  1982,8 a  price 50  percent greater than
 the  prevailing wholesale price  of gasoline.  If benzene reduction
 doubles the benzene  supply and  if the price of benzene drops as a
 result of this increased supply, industrial chemical users and  producers
 may  find  it profitable^to switch to benzene in their processes.   As an
 upper bound,  the benzene credit as an industrial  chemical  may exceed
 the  direct-volume-related loss  of benzene  in gasoline.   If this is the
 case, a "volume credit"  rather  than a "volume loss"  should be incorpor-
 ated into the indirect cost analysis.
     Markets for other chemicals, especially those used to boost
 octane, also would be affected by the two benzene reduction strategies.
 If their prices increase as a result of an increase  in demand for
 octane boosters,  the. cost of octane loss  due to benzene reduction
would have been underestimated by the ADL analysis.   In particular,
 the price per octane-gallon would tend to be higher than estimated.
8.3.2  Bulk Terminals and Bulk Plants
     As noted in  Chapter 4 of this report,  an estimated 67 percent of
total gasoline throughput is handled currently at controlled  terminals,
                                  8-31

-------
so 67 percent of the gasoline terminal population is assumed to be
controlled.  The remaining 33 percent of gasoline terminals would be
controlled under several of the regulatory strategies considered here.
In particular, Regulatory Strategies IV, VI, X, and XIII include
Stage I controls nationwide.  Gasoline bulk terminals would be impacted
under each of these strategies, regardless of option, and the impacts
would be the same in each case.  No differences in coverage are proposed
and no exemptions apply.
     Cost variations across model terminals are significant.  Top-
loading facilities experience the largest impacts.  Within each plant
size category, both the total cost and the annualized cost of control
over the projection period for a top-loading terminal are higher than
the corresponding costs for a bottom-loading terminal.  For the three
smallest plants, the ratio of such costs for a top-loading facility to
the corresponding costs for a bottom-loading facility ranges from 2 to
57.  For the largest size plant, the bottom-loading facility exhibits
negative 1986 NPV of control cost and annualized control cost while
the top-loading facility incurs positive costs.
     Within loading classifications, the two larger model terminals
always generate lower 1986 NPV of control cost and annualized control
cost than do the two smaller terminals.   The largest model terminal
always shows the lowest such cost.  The smallest model terminal
generates the highest such cost for bottom-loading facilities while
the next-to-smallest terminal realizes the highest cost for top-loading
facilities.
     In summary, larger terminals experience lower NPV of control and
annualized control costs, whether viewed on a total or per unit of
throughput basis.  Thus, there are economies of scale in Stage I
control for terminals.  These economies will reinforce any economies
of scale in production.  Accordingly, the process of market rationali-
zation, which would probably continue to exert pressure on smaller
terminals even in the absence of additional control, can be expected
to continue.
                                  8-32

-------
      As Regulatory Strategies IV, VI, X, and XIII include Stage I
 control nationwide, bulk plants would be impacted under each strategy.
 However, each of these strategies includes two options.   For each
 strategy,  Option B requires incoming and outgoing vapor balance systems
 for all bulk plants.   Option A exempts small  bulk plants (throughput
 less than  15,000 liters/day) from outgoing balance requirements but
 retains the requirement- of submerged filling  of account trucks.
 Consequently, the number of bulk plants affected would  be the  same
 under each option (8,040), but costs assessed for the smaller  plants
 (Model  Plant 1)  would vary by option.
      Because cost impacts are not uniform across model  bulk plants,
 industry organization would experience some impact.  Generally,  NPV of
 control  cost and annualized control  cost decrease as plant size
 increases.   Thus,  control  costs  for  bulk plants exhibit  economies of
 scale as do basic  production costs.   Accordingly,  additional regulation
 would reinforce  the industry trend toward larger,  more  efficient
 plants.  Regulatory cost would put additional  pressures  on small
 facilities,  many of which  are already marginal.   However,  because of
 exemptions  considered, potential  impacts  would vary by regulatory
 option.
      Under  Option  A,  some  of the pressure of  control cost  on the
 smallest facilities would  be eased.   However,  for  plants with  through-
 put  greater  than 15,000  liters/day,  economies  of  scale would still be
 associated with  control  cost as well  as production cost.   The  industry
 trend toward  larger, more  efficient  bulk  plants, therefore, is  likely
 to continue  under  Option A.   Under Option B,  Stage I regulation com-
 pletely  reinforces this  trend.
     Geographic distribution of impacts is difficult to assess.
 Impacts would be restricted  to attainment areas, which are not concen-
 trated on the East Coast and  in the Midwest (PADD's I and  II).   However,
the bulk plant population is concentrated on the East Coast and in the
Midwest, only portions of which are attainment areas.   Consequently,
 impacts would be spread across much but not all of the United States.
It is not possible given present data to be more specific about regional
impact variations.
                                  8-33

-------
8.3.3  Service Stations
     Costs of potential regulations do not affect all service stations
in the same manner.  Small stations in particular are subjected to
disproportionately higher unit costs than are larger firms since
larger firms are able to distribute the slightly higher fixed cost
over a much larger throughput.  Stage II unit costs for stations with
high throughput (Model Plant 5) are only 1.0 percent of those of
low-throughput (Model Plant 1) stations.  For Stage I, high-throughput
unit costs are 2.7 percent of low-throughput costs.  Some low-throughput
stations will likely have difficulty competing under this skewed cost
burden, particularly given the shrinking margins in today's market, and
might close, allowing larger stations to take up the sales volume.
     The difference between the costs of nationwide Stage II programs
with and without onboard controls  also  affects the distribution of
costs across model plants.  The 1986 NP.V of control costs per station
for the various model plants under nationwide Stage I and Stage II
with and without onboard are presented  in Table 8-16.
     Differences in distributional  effects of the two programs are
evident in the table.  Smaller model plants bear a higher per-station
control cost as a  rule when Stage  II without onboard  is  implemented
largely due to higher  recovery credits  realized by the larger model
plant groups.  When onboard is combined with Stage II, however, the
larger  stations are assumed to lose the recovery credit  advantage over
time, and  thus bear a  higher  per-station control cost.
     The  assumptions  incorporated  in the sensitivity  analyses can
cause variation  in the distribution of  per-station costs.   Under the
constant  gasoline  consumption  case, the differences  between per-station
costs over the model  plant sizes  for Stage  II with and without  onboard
are more  extreme.   Recovery credits remain  constant  in the  constant
gasoline  consumption  case,  causing annualized control cost  per  station
 to be  as  much  as  70  percent lower than  in  the base case  for the large
model  plant for  Stage II  without  onboard.   The  decrease  for the smallest
 model  plant is  less  substantial:   2 percent less  than in the base
 case.   The small  model plant  thus has  a much  larger  annualized  control
                                   8-34

-------
    TABLE 8-16.   1986 NPV OF CONTROL COST PER STATION BY MODEL PLANT'
                             (1982 dollars)

Stage I
Stage II — without onboard
With exemption
No exemption
Stage II — with onboard
With exemption
No exemption
MP1
2,984
12,399
10 , 134
MP2
2,984
12,278
12,093
10,770
10,629
MP3
2,984
12,923
12 , 647
12,018
11,809
MP4
2,984
15,669
15,669
15 , 685
15,685
MP5
2,984
14,260
14,260
20,449
20,449
Enforcement cost is not included because it is not borne by the service
station sector.
                                    8-35

-------
cost per station than the large model plant with constant recovery
credits when Stage II without onboard is imposed.  For Stage II with
onboard, which includes declining recovery credits to reflect Stage II
phase-out, the changes in per-station control cost relative to the
base case are less dramatic, ranging from a negligible decrease for
the smallest model plant to a 16-percent decrease for the largest.
Stage I per-station costs do not vary in the sensitivity analysis
since no recovery credits are assigned.  The per-station costs for the
declining facility cases are not directly comparable to the base case,
due to the variation in station populations over time.
     The service station industry is currently undergoing structural
change.  The number of stations is decreasing as small stations close
and are replaced by larger, more profitable stations.  This trend is
due in part to declining gasoline consumption in recent years, which
has lowered price margins and has forced the higher cost stations to
leave the industry, and in part to the "stand alone" economic philosophy
that has induced oil companies to take advantage of large-station
economies of scale, which is discussed in more detail in Reference 1.
     The higher unit cost estimated  for small service stations would
exacerbate this existing market trend.  Although some small firms will
still have a market niche in the industry, others would be forced to
leave the industry more quickly and  in greater numbers if the additional
cost of regulation were imposed upon them.  As increased costs are
passed along to customers and retail prices rise, quantity demanded
decreases.  Station owners will seek means to maintain volume and cut
costs.  These means might include moves toward high-volume locations
and might tend to speed the trend toward self-serve stations.
     The market implications of regulations exempting small and inde-
pendent stations are twofold.  First, the exemption of small and
independent stations effectively subsidizes them.  If, as may be the
case, smaller stations are marginal  in the industry,  such exemptions
would protect the less efficient facilities at the expense of the
mid-size  stations.  Second, with exemptions, the middle throughput
range stations would bear higher increases in unit cost.  They would
                                  8-36

-------
be less competitive with both the largest and smallest exempt segments

of the industry.

8.4  REFERENCES

 1.  Research Triangle Institute.  Preliminary Economic Impact Analysis
     of the Regulatory Strategies for the Gasoline Marketing Industry.
     Draft report to U.S. Environmental Protection Agency.  July, 1984
     Docket No. AT84-07.                   ,        .
 2.

 3.
 6.

 7.

 8.
 National Petroleum News.  Factbook Issue.  July 1983.  p. 91.

 Memorandum from Aitken, Mary, Research Triangle Institute, to
 Morris, Glenn, Research Triangle Institute.  April 26, 1984.
 Gasoline Demand Elasticities for Use in Benzene/Gasoline Marketing
 Analysis.

.Memorandum from Giberson, Linda, Research Triangle Institute to
 Robson, John, EPA.  December 21, 1983.   Price Elasticity of
 Demand Estimates for Automobiles.

 Arthur D.  Little, Inc.   Cost of Benzene Reduction in Gasoline to
 the Petroleum Refinery Industry.  U.S.  Environmental Protection
 Agency.  EPA Contract No. 450/2-78-021.  April 1978.  pp. 5-25
 through 5-31.

 Reference  5.   pp. 5-17 through 5-22.

 Reference  5,   pp. 6-11 through 6-17.

 Chemical Marketing Reporter.   May 10, 1982.  p.  37.
                                  8-37

-------

-------
 9.1
    9.0  ENFORCEMENT STRATEGIES AND COST CONSIDERATIONS
ENFORCEMENT STRATEGIES
 9.1.1   Stage -II  Programs
       Generally speaking,  enforcement  of  an  air  pollution  control
 program occurs  in two  phases.   The  first  phase consists  of assuring
 installation  of acceptable control  systems and the  second, applying to
 cases where the control  systems are  subject  to damage or other malfunction,
 consists of assuring that  the  installed systems  are properly used and
 maintained.   [Hereafter, this  latter process shall be referred to as
 "in-use enforcement"].
       9.1.1.1    In-Use Enforcement.  In cases where control technology is
 subject to malfunction,  the preferable method of in-use enforcement,
 all  things being  equal,  is to  assess system performance by some sort of
 in-use emissions  test.   EPA's  Field  Operations and Support Division
 (FOSD)  has endeavored in the past to develop an accurate and expeditious
 in-use emissions  test.   EPA's  primary effort has been the  development
 of the so-called  "Short" Test.  This test was first proposed in EPA's
 November 1, 1976  notice of rulemaking proposing amendments to already-
 promulgated federal Stage  II regulations.  The Short Test measured
 vapor  recovery  system emissions occurring during actual  vehicle
 refuel ings.  Vapors emitted at the nozzle-fill pipe interface were
 captured by a flexible sleeve  and fed into recording instrumentation.
     The Short  Test required two people—one to conduct the
 vehicle-refueling/vapor-recollecting process and one to handle the
 instrumentation.  As conceived, the test required that emissions from
 100 cars be measured in order  to establish a violation.   Experience
with the short  test indicated that about 75 cars could be measured at
 high throughput stations in an average eight hour workday.   Thus,
 roughly  2.7 person-days would be required to perform a "Short"  Test.
 EPA estimates that, on average, state inspection effort yields  about
 195 person-days (260 days per year x 75% available field time)  per
year.  The number of stations at which short tests could be performed
per inspector man-year would be about 72 (195 person-days /2.7  person-
days per test).  Methodology presented later in this chapter estimates
                                  9-1

-------
that Stage II programs will achieve about 56 percent efficiency in a
minimal enforcement scenario but that this figure may be raised to about
86 percent if regulated outlets are inspected annually.  EPA estimates
that Strategy V.A (Stage II nationwide, with size exemptions) would
cover about 120,000 outlets, and that strategy III.A (Stage II in all  ozone
nonattainment areas, with size exemptions) would cover about 36,000 outlets.
Accordingly, the person-years of effort which would be required to inspect
each outlet annually in the Strategy V.A and Strategy III.A scenarios,
using the Short Test, would be roughly 1,660 and 500 respectively.
      A second approach to in-use emissions testing which has been
investigated by EPA/FOSD is the so-called REST (Refueling Emission
Simulator Tank) procedure.  This procedure would, not measure emissions
during  actual vehicle refuel ings.  Rather, emissions would be determined
by measuring the vapors escaping when  gasoline was dispensed into
portable  fuel tanks  equipped with a  number of interchangeable fill necks
representative of fillnecks in the on-the-road vehicle population.
       As  in  the case of the Short Test, two persons are required  to
perform the  REST procedure.  By eliminating the  need for  actual vehicle
refuel ings,  however,  REST  achieves a time advantage over  the Short
Test.   It is estimated that about three hours would be required to
perform the  test at each  station.  In addition,  a  daily calibration of
the  equipment  requiring about  45 minutes  would  be  necessary.  Adding  in
travel  time  to  and from stations,  the most  reasonable  estimate presently
is  that,  using  REST, two  service stations could  be tested per  day by
each team of two  inspectors.   This means  that one  team could inspect
 about 390 stations per year (195 field inspection-days x  2  tests  per
 day).   Accordingly, about 195  tests  can be  performed  per  person-year.
Thus,  the resources necessary  to  inspect  each outlet  once annually  in
 the Strategy V.A and Strategy  III.A  scenario's,  using  the REST
 procedure, would be about 600  person-years,  and 200  person-years,
 respectively.
                                   9-2

-------
       It appears that there is an enforcement strategy  capable  of
 achieving given in-use inspection frequencies with  substantially fewer
 inspection resources than would be required  with  either the  Short Test
 or REST procedure.   Examination of Tables  D-2 and D-3  (Appendix D)
 setting forth  the frequency of occurrence  of each type  of  Stage II
 system defect  in a  minimal-enforcement  scenario,  along  with  the average
 effect on efficiency of each type of  defect,  suggests a strategy.  From
 the tables it  can be seen that most Stage  II  system defects  are visually
 observable—e.g., nozzle and hose defects  for all systems, tampering
 for all  systems.  Indeed, the only  system  defects which  are  not directly
 visually detectable are misinstallations (all  systems)  and miscalibrated
 aspirators, or  jet  pumps, in the  case of Hybrid systems.
       The visually  observable defects constitute  by  far the  dominant
 portion  of all  system efficiency  losses.   For  example,  programs
 requiring balance systems,  which  have been certified at  95%  efficiency,
 are estimated  later in  this  chapter to be  54  percent efficient  on
 average  in a minimal-enforcement  scenario.  The majority of  this
 efficiency loss  is  attributable to  visually observable  defects  and a
 high rate of noncompliance.   The  percentages of system efficiency
 losses attributable  to  visually observable defects  for other systems
 are comparable.
      That such  a high  proportion of system efficiency losses would be
 attributable to  visually  observable defects and a high rate of non-
 compliance suggests  that  periodic visual inspections of  in-use control
 systems  should serve  as the cornerstone of any Stage II enforcement
 strategy.  Operators  of gasoline-dispensing outlets determined to be
 without controls or  to be using one or more defective or
 tampered-with system  components could be served with "notices of violation"
 or  some  type of citation, depending on the type of enforcement
 authority  available, as well  as on the Stage II enforcement policy
 chosen in  each jurisdiction.  In either case, the visual inspections
would ultimately lead to correction of observed violations.
      Success  of a Stage  II program whose enforcement strategy was
premised on visual inspection of in-use systems presupposes several
                                  9-3

-------
factors.  First, there would have to be a procedure for certifying,  on
a generic basis, that permissible-to-use systems were capable,  if
properly installed, maintained and handled, of achieving the recovery
efficiencies upon which the Stage II program's emission reduction
credits would be premised.  Second, some form of quick check for system
deficiencies not visually detectable would need to be developed—at
least to ensure that losses from such deficiencies remained as low as
theoretical analysis appears to indicate.  The vapor recovery system
certification process operated by the California Air Resources Board
can serve as a model in these respects:  the process requires that
systems  achieve high efficiency levels when tested under actual operating
conditions; in  addition the procedure requires the sponsor of each
control  system  to  submit  an operating manual "identifying critical
operating parameters affecting system operation...[and  identifying] the
operating  range  of these  parameters  associated with  normal,in-compliance
operating  of the control  system..."  Checks of these Stage  II operating
parameters  have been  developed which are simple  and  expeditious  enough
to serve as supplements  to  visual  in-use system  checks.
       9.1.1.2   Installation Monitoring.   Some  of the principal  system
 defects not subject to visual  detection are  installation  defects—parti-
 cularly, in the case of Stage II systems, defects in the  installation
 of the underground piping.   These deficiencies can be detected  by the
 parameter checks just referred to.
       The checks can be performed after a system has become operational.
 However, there is a substantial  advantage to structuring Stage II
 enforcement efforts so that performance of these checks occurs prior  to
 a system's becoming operational.  If the test is performed before the
 earth-and-concrete covering over the pipes is replaced, a service
 station owner will incur substantially  less expense in correcting a
 misinstallation problem.  Thus, during  the installation phase of a
 sensibly-run Stage II program, enforcement resources will be concentrated
 on  assuring that  control systems are properly installed.
                                    9-4

-------
9.1.2   Onboard Control Programs
      The modified fill pipe program scenario could be enforced, at
probably modest incremental cost, by including the onboard vapor
recovery feature among those emission-control features already
monitored within the established EPA certification and in-use mobile
source testing programs.  The regulatory standard would be a performance
standard couched in a form capable of being tested in conjunction with
the Federal Test Procedure's evaporative emissions tests.
9.2   RESOURCES REQUIRED TO PURSUE ENFORCEMENT STRATEGIES AT VARIOUS
      LEVELS OF EFFORT
      In the immediately preceding subsection, the general  approach to
enforcement of Stage II was discussed.   The present subsection sets
forth the specifics of Stage II enforcement and describes the
methodology for determining the amount  of resources needed to enforce
the various gasoline marketing program  options.
9.2.1  Installation Monitoring Resources
      It was assumed that the resources applied to the task of monitoring
installation of systems during the phase-in of a Stage II program would
be the same as the resources applied to performing in-use inspections.
9.2.2   In-Use Inspections Resources
      The resources necessary for performing in-use inspections at the
various frequencies may be determined from the following formula:
      R = S x i x (TI  +  T?  x r)   x.~B
where
   R
   S
   r
   M
   B
                        M
resources,.expressed in man-years of effort
number of outlets covered by the control  option or alternative
number of times per year each regulated outlet is inspected
time necessary to perform each in-use inspection (hours)
time necessary to perform follow-up inspection at outlets
found in violation upon first inspection (hours)
percentage of outlets reinspected after initial inspection
field time available in each inspector man-year (hours)
multiplicative factor to account for supervisory overhead
                                  9-5

-------
These individual factors were determined as follows:
S  -  The number of facilities covered by each control  option is
      shown in Table 7-18.
i   - 0.5 (bi-annual enforcement)
    - 1.0 (annual enforcement)
    - 4.0 (quarterly enforcement)
Tl> T2 " The tl'me Pen'ods to perform each in-use inspection (T]_) and
each reinspection (T£) were developed based on field inspection
experience for each of the industry sectors except for Stage II.
Table 9-1 indicates the inspection and re-inspection times assumed
for this analysis.
   The time to perform each'in-use inspection (Tj_) for Stage II
facilities was computed as the weighted average of the times to
conduct inspections at typical (9-nozzle) stations utilizing each
type of control technology.  The inspection times for stations with
the various types of control were computed as follows:
   Common Elements
      -  Conduct explosimeter checks during refueling occurring
         while inspector  is at station
                                                    5 minutes
      -  Visual checks of nozzles/hoses
                                                   15 minutes
      -  Time for recording observations and discussions
         with station personnel
                                                   20 minutes
      -  Travel Time per station

         Total, Common Elements
      Individual Elements
       Balance System Stations
         -  Check No Seal-No Flow features

                               9-6
10 minutes

50 minutes
    5 minutes

-------
Table 9.1  INSPECTION AND RE-INSPECTION TIME ASSUMED FOR
                ENFORCEMENT-COST ANALYSIS
                      Inspection Time,a
                         Hours	

                          2.5

                          2.0

                          0.5C
Re-Inspection Time
       Hours

       1.5

       1.0

       0.5C
                          0.5
                          1.26
       0.33
       0.5
 Industry Sector

 Bulk Terminals'3

 Bulk Plants'3

 Storage Tanks

 Service Stations'^
  - Stage I only
  - Stage II only


 aBased on EPA estimates.

 blncludes inspection of trucks at the loading racks.

Inspection time per tank.

dCombination of Stage I and Stage II inspections could result in a time
 savings of approximately 0.17 hours per inspection.  The enforcement
 impacts analysis assumes a worst case in which Stage I and Stage II
 inspections would be performed independently.

e Weighted average of inspection times for balance system (80% @ 1 2
  hours ,  Hybrid systems (15% @ 1.3 hours), and vacuum assist (5% @ 1.1
  hours).
                          9-7

-------
  -  Check  Back Pressure-Shut Off
    Features  (All Nozzles)
  -•  Conduct Liquid  Blockage Test
     (20 minutes per test,  1 dispenser
     in 50,  at random)
  Total, Indiv.  Balance Syst.  Elements
Aspirator-Assist System Stations

  -  Check Nozzle Vacuums
     (3 nozzles per station at random)
  -  Check Back-Pressure-Shutoffs
     Check Aspirator Calibrations
     (Check 5 nozzles at 1 station
     in 20, at random, 15 minutes
     per nozzle)
     Liquid Blockage Test
     Came as bal. syst. stns.)

     Total, Individual Aspirator
     Assist Station Elements
 Vacuum-Assisted System Stations

   -   Gross  Check  of Nozzle Vacuum
      (1  nozzle per station, at random)
      Check  Gross  Function  of
      Incinerator, Blower,  Underground
      Storage Tank Pressure Gauge
      Total, Individual  Vacuum-Assist
      Station Elements
                                              15 minutes
  .4 minutes


20.4 minutes
10 minutes


15 minutes
 3.75 minutes


   .4 minutes



 29.15 minutes
                                                3 minutes
                                               10 minutes
                                               13 minutes
                         9-8

-------
 Total  Inspection Times (Common and Individual Elements)
         -  Balance                 70.4  minutes
         -  Aspirator               79.15 minutes
         -  Vacuum-Assist           63.0  minutes
Weighted Average
Total Inspection
(Assumes 80%/15%/5%  throughput  coverage
 distribution  evidenced  in  South  Coast
 and San Diego,  CA,  areas)
   = (70.4 x .8) + (79.15 x .15) + (63.0 x .05) = 71.5 minutes

T£ for Stage II was determined to be 25 minutes (10 minutes travel
time; 15 minutes to record observations and to discuss matters
with station personnel)

r - The percentage of stations reinspected is expressed as (a x v)
where "a" is the fraction of .initial  violators reinspected and "v"  is
the percentage of facilities initially determined to be in violation.
The fraction of initial violators reinspected depends on a judgement
respecting the need to ensure that initial  violators, who at some  point
agree to correct system violations, actually do correct the violations.
Where enforcement is done on a quarterly basis, it is judged that  the
succeeding quarter's inspection can generally serve as a follow-up
inspection and, accordingly, "a" is assigned the value zero for that
enforcement scenario.  Where enforcement is performed only biannually,
it is judged that a 100% follow-up rate is desirable.  For annual
enforcement, it is believed that a 50% follow-up rate would be adequate.

"v",  the percentage of facilities which would be determined to be
initially in violation upon an in-use inspection, has been determined
to be 15 percent for Stage I equipment based on experience in over
4,000 Stage I inspections.  No Federal  experience is available for
Stage II systems, therefore it was assumed that the initial  violation
rate was a function of the percentage chance (probability)  that an
individual installed control system will  have a defect.   A curve
                                 9-9

-------
expressing the relationship between the probability that an individual
in-use control unit would.have some defect and the probability that
any given outlet would have one or more defective units--i.e., be
initially in violation—was developed by the following procedure.
     First, a curve expressing the relationship between these two
probabilities, using only the mathematical laws of chance, was plotted.
(See Curve A in Figure 9-1).  The mathematical probability "P" that
any station would have one or more defective units is expressed by
the following formula:
                          P = 1 _ (1 _ p)m
where "p" is the probability than an individual control unit will
have a defect and "m" is the number of nozzles-(estimated to be 9)
at an average station.
   The mathematically determined curve would be a correct potrayal
only if individual control system defects were randomly distributed.
Presumably, however, at least some dealers would be conscientious
about maintaining control equipment, and 'thus the actual  distribution
curve would deviate from the mathematical model.  It was decided to
use available data to generate several data points for such an actual
distribution curve, extrapolating the remainder of the curve based on
a comparison with the mathematically-predicted curve.I/
    Analysis of empirical data collected in EPA's survey of service
stations in the District of Columbia revealed that the rate of unit
defects and the rate of stations exhibiting one or more defective
units varied considerably from one type of system to another.  Among
stations using one type of nozzle, the percentage of unit defects was
roughly 44%.  The percentage of stations with one or more defective
units was roughly 90%.  By contrast, units at stations using another
type of nozzle had only a 16% unit defect rate and "only" 50% of the
stations had one or more violations.  These two points were plotted
  Basing the extrapolation on the mathematically-predicted curve
  assumes that randomness will be the dominant factor in determining
  the distribution of defects among stations, and since it is believed
  that most service station operators will  not voluntarily maintain
  vapor recovery equipment, this assumption seems reasonable.
                                 9-10

-------
                                                                                                    -_  O
                  03
O)
>
 CJ
 cu
O)
Q
(13
3
•o
s=
t—<

c
4-
O
 O

0_
 O)

 3
 CT>
                                                                                                    -.  o
                                                                                                   .. a
            +J


             C
...  O      ,_
    "^      13

            -a
                                                                                                               T3
                                                                                                       o
                                                                                                       CO
            C
            (13
    O
    CM
-Q
 (T3
J3
 O

a_
                                                                                                  -- o
s s ° c
O CO 10 ^
3
r
1 "
o
CM
                                     99
                                                     9-11

-------
  and from them was extrapolated a curve (Curve B, Figure 9-1) express-
  ing the empirically determined probability that any one station
  would have one or more defective control units as a function of the
  probability that an individual unit would be defective.
     The rates (probabilities) of individual unit defects, including
  the weighted average of such defects for each level of enforcement
  effort, were determined from the information in Tables D-8 through.
  D-10 of Appendix D and are set out in Table 9-2.
       Little EPA data have been compiled for violation rates at bulk
  terminals and bulk plants.  Results of a recent survey by EPA
  Region IX of terminals in California,! where a relatively active
  inspection program exists, indicated 68 percent of the terminals
  inspected resulted in violations either for truck leaks or for
  loading rack operations.  Therefore, the same percentage of facili-
  ties in violation for Stage II was assumed for the bulk terminals
  and bulk plants.  Table 9-3 contains violation rates assumed for
  bulk terminals, bulk plants, and Stage II at service stations.

  M - Experience with pollution control program enforcement at the Federal
  level  using contractor assistance for inspections indicates that an
  inspector, on average, spends about 75% of a man-year in the field
  doing  inspection work.  Enforcement officials in the South Coast and
  San Diego areas of California indicate that their inspectors spend
  75%, and almost 100%, of  their time in the field, respectively.
  For purposes of this study, M was determined to be 1560 hours per
  man-year at both the Federal and State "level (260 days x 0".75 x 8 hours
  per day).

  B - South Coast and San Diego enforcement officials  indicate that the
   supervisor-to-inspector ratio's  planned for in-use enforcement of Stage
   II in  their districts  are 1 to 6 and  1 to 5, respectively.  For this
   study,  the  supervisory overhead  factor,  'B1, was set as 1.2.
      Using  the  formula  set  out  at  page  9-7, along with the above-
derived  values  of  the  independent variables, the inspection resources
required  to  enforce  the  various  control  options at the  various levels
of efforts were  calculated.
                                9-12

-------
             TABLE 9-2.  PROBABILITY THAT INDIVIDUAL CONTROL UNIT WILL
                         HAVE AT LEAST ONE DEFECT, AS A FUNCTION OF
                         ENFORCEMENT EFFORT
Inspection
Frequency

Quarterly
Annually
Bi -Annually
Bal
Dir.
1%
21%
37%
ance
N.O.V.
10%
30%
50%
Aspi
Dir.
5%
22%
33%
rator
N.O.V.
8%
33%
42%
Vac.
Assist
Dir. I
2%
8%
16%

J.O.V.
5%
15%
27%
Weighted
Average
Dir. N.O.
6.5% 9.
21% 29%
35% 48%

V.
5%


     Using Curve B,  Figure 9-2,  values of "v"  for  the various  weighted  average
rates of individual  unit defects (i.e.,  for the various  levels of  enforcement
effort), were then determined.   The results appear in Table  9-3.
                                       9-13

-------
Frequency of
Inspections
                 • Table 9-3

   PERCENTAGES OF FACILITIES (STEADY-STATE
   AVERAGE) WHICH WOULD BE IN VIOLATION AS
A FUNCTION OF FREQUENCY OF IN-USE INSPECTIONS

                                 Percent Violators
                              (Steady-State Average)
Quarterly

Annual

Bi-annual
                             State
                          'Enforcement3/

                               28

                               65

                               85
Federal
Enforcement3/

     34

     72

     90
aThe Federal figures are based on 100% N.O.V.-type enforcement-.  Some
 States have N.O.V.-type and some have direct enforcement mechanisms.
 The State  figures presuppose a 50%-50% split between direct and N.O.V.
 enforcement.
                                   9-14

-------
9.2.3   Test Observation Resources
      For the  analysis, it was  assumed  that  an  initial  performance  test
of control equipment would be required  only  at  bulk terminals.  The
level of resources  necessary to  observe a performance test was determined
according to the  following formula:
      R = S(Tj +  T2 x r) x B
where
   R  = Test observation resources, person-years
   S  = Number of controlled facilities
   TI = Time to observe initial  test, hours
   T2 = Time to observe re-test, hours
   r  '= Fraction  of facilities requiring retest
   B  = Overhead  factor                           ,
   M  := Field time available, man-years                        .
      The number  of facilities,  the overhead factor, and the field time
available was the same as used in the previous analysis.  The observation
time for both the initial  test and the  re-test was assumed to be 16 hours
(8 hours to observe the test and 8 hours to review the  report).  This
assumes that the  testing for bulk terminals would be conducted in the same
manner as that required by the NSPS for bulk terminals  (40 CFR 60, Subpart XX
- Standards of Performance for New Bulk Gasoline Terminals).  The fraction of
facilities requiring re-test (one-half) was taken from  EPA's analysis of
reporting and recordkeeping burdens performed for the NSPS.
9.2.4   Legal-Clerical  Resources                            - -  .
      The level of legal resources necessary to process detected
violations of regulations  imposed under the regulatory  strategy was
determined according to the following formula:
      R  =  S x i  x v x (Ti + T? x c)
                       ^
where

   R = legal-clerical  resources, in person-years.
   S,i,,v = same as in Section 9.2.1.
                               9-15

-------
 TI = total time, including legal, secretarial, and supervisory, to
 prepare either a notice of violation or a complaint, depending on the
 enforcement mechanism utilized.  Based on the Federal experience
 enforcing Stage I vapor recovery, TI is estimated at 3 hours. 'It is
 assumed that State enforcement processes would not be substantially
 more expeditious than Federal processes.

 T2  s total additional time required  to process a case where the
 violation is  not "voluntarily" corrected and  the case settled  as a
 result of the  N.O.V.  or complaint.   Based on  the Federal experience,
 the typical  case is  settled  upon issuance of  an administrative
 order, a  process consuming 5 hours  of  legal,  clerical and  supervisory
 time.   It is assumed that the "difficult" case would be  handled  with
  the same  effort at the State level.

  c ~ the percentage of cases  which could be  expected to  go  beyond the
  complaint or N.O.V.  stage ("difficult" case). As  most  violations
  will be relatively uncomplex and, compared  to the  cost  of  a legal
  proceeding, relatively inexpensive (to correct),  it is  anticipated
  that the rate of "difficult" cases will be small.   EPA  estimates a
  2.% rate  for quarterly enforcement, 5% for biannual enforcement,  and
  4% with  annual enforcement.

  M  * the  time  available in a  legal-clerical  person-year, expressed in
  hours.   There is assumed to  be  no difference between the federal and
  state figures.  EPA  estimates the average amount of legal-clerical time
  per year which would actually be spent on processing cases at 1920
  hours per-year  (240  days per work year x 8 hours/day).

      Using the  formula set out above, this page, and  those  values of
the independent variables just derived,  the legal-clerical resources
required to enforce  the various control  options and  regulatory
strategies  at the various  levels  of enforcement efforts were calculated.
                                  9-16

-------
 9.2.5   Onboard Control  Inspection Resources
      The onboard-control, modified-fillpipe option can be enforced
 through the existing certification and in-use testing programs.  It
 appears that the amount of personnel and equipment needed to effectively
 enforce this type of onboard control would be much less than that
 required for enforcing service station controls.  It was estimated that
 incorporating a test of the onboard control system would add only about
 1/2 man-year to the EPA certification program, and would require only
 about $50,000 in equipment costs.  Monitoring onboard controls would
 probably not require any additional personnel.  The increase in the
 testing budget necessary to accommodate the monitoring of onboard
 controls would be modest, running at most 1Q% of the FY79 budget of
 $1.5 million—i.e., $150,000.  At a rate of $30,000 per person-year
 (see section 9.3), this equates to about 5 person-years.
 9.3   ENFORCEMENT COSTS
     In preparing budget estimates, EPA estimates the annual  cost of
 its "average" employee (legal,  technical, secretarial-clerical  all
considered)  at $30,000 per year.   A check of enforcement costs with
personnel  in the South Coast and  San Diego, California areas  reveals
that this figure is probably a  reasonable estimate on the state level
as well.  Accordingly, the annual costs of enforcing the various Stage
 II and onboard control program  options can be determined by multiplying
$30,000 by the number of person-years involved.   The results  appear in
Tables 9-4 and 9-5.  The costs  of equipment, being a minor portion  of
the overall  cost,  have not been estimated.
9.4  ENFORCEMENT COST EFFECTIVENESS ANALYSIS
     A final  analysis concerning  enforcement costs was performed to
determine the effect of enforcement costs on the cost effectiveness of
the regulatory strategies.  Table 9-6 contains a comparison of the
different levels of enforcement for the Stage II nationwide control
option and the Stage II in all  nonattainment area control  option.
Cost effectiveness of strategies  based upon theoretical  control
efficiencies  and cost effectiveness based upon in-use efficiencies,
given different levels of enforcement,  are presented both with and
                                  9-17

-------
without Including enforcement costs.  As indicated, enforcement costs
do not significantly effect the cost effectiveness of the strategies.
In addition, Table 9-6 indicates that annual inspections are the most cost-
effective approach for any enforcement level under the no exemption
option and that annual or quarterly inspections are roughly equivalent
under the size exemption option.
                                  9-18

-------
         +•» S. H-

         S«2S
         o c t*
O
O
O

00
o
o

1—0
Z CM
LU O
O IO
a: co
o CTI
U_ i-H
    oo
LU
 <-
u  tu
v- a.
o —
                                  I rf    —( O O
                              to co CM   CM *-H *^
                                                           O O LO   O O IT)
                                                           t^* c\j ^H   r** CM •—(
                                                                                              o o o
                                                                                            o o o
                                                                                            m co r—
                  §§g
o o o
00 «3- CM
o <
o <
CM <
                                                                      CM <


                                                                      CM
> o
I O
i tn
• o o o
 o o o

                               !_
                               
                              C
                              o
c
o
+J
01
c
o
•t—

a i—

00 O)
en

T- OO
> •*-*
(U
00
s
oj <0
•a x


C (U
O M

• ^ oo

z o

1


.a

o
+j
' o.
S
0)
X
UJ

o


o




                                                                 9-19

-------
               •si!

               4-> Sl^
               « U-r-

               0 F— c     ia e   *>
               s.  3 
                                 c
                                 o

                               10 4J
                               ut a.
                               
-------
a.

C c
0 §£
•*•* {_> i—
	 • U) fc. •*•
TSr O O £
tJ O **—
Qi C **
H- LLJ— ^
32 3
0 o
O
Lu
0
CO
oo
o
o
h-0
IZ CM
LU O

LU^^O S
O IO CU u
OS CO T3 ^_
O CTi 3 o  S.
32*— U .2 S
LU c ":£
CO O *i i
LU 2Z 0 | §
1—4 HH V ^
i— i— £ S.
 «
e
at >, o
50 o •»-
r— e +J
« l°l
r— aj «
t. c
U- i— •













e
0

i *
Q.
O
UVO» ^O O O O r-4
CM r~- m CM




o o .0 ' o o o u)
CM »H















0 O O O O 0 O
o of o o o r**
US O «3 C\l CM O r-t .
H ft ft f, fl
CM •-< 00 CO CM












S- 3 S- 3
0) r— = 0) i— C
•4-> «9 C +•>  o
Q. •!-
•< £ 4-*
^K m i-i
x e
o x
4-> 0) LU ^3
U M -0
CD •"- O S-
• — CO ^. - iQ
03 O
 4J
i. Q.
(U O
o.
 s-
^_>
Js
2
cj 15
E 4J
0 «
M- e
s- o
0) C

+•> >>
c^
§g
0
0 1-
s- o

•a c
C V
t) S

«0
• IQ o

s- a
0 •!->
Q. CO















.
"a!
c
o
tn
i.
0)
a.
ia
s.


-------
                                                                        o o
                    .
                   . OS
                                        o o o   o o o
                                        co co CM   in o in
                                                               o o o    o o o
                                                                                       o o o
to
fc
S LU
LU I—I
O CJ
CC LU
O \- *~*
u. «c o
sz: oi CM
LU I— O
   C-0 CM
LU     I
>> >- VO
•—« oi co

!5gS
                                                      §<
                                                      (
SS8
I O
I O
                                        CM *-4     CO CO CM
                                                               O t-. r»   to 10 «*_

                                                               t— CM ••*   CM CO LO
I 0
i o
S CD
O UJ
ID
 i
cn
 
                                                           9-22

-------
                          e e
                       «- cu o
                       o e i-
                           -  W3 4->
•-• GC.  CO  =
H- O    O
=3 CD
O LU
    Cf
LO
 I
CT)


 O)
 <0
 s.
CD <9
c£ 01
S in
cu s_
o  •* CM
        o o o
        en CM «a-
        LO CM ^H
> O O
> to co
§00   O
r^ r^.   o
                                                                         O O
                                                                         m co
                                                                       CM co in
                                                             o o
                                                             10 en
                                                             CM
                                                             o o o
                                                             CO lf> CM
                                                      oo o
                                                      «3- «3 IO
                                                      r~- CM •-<
                                                                           O •-*
                                                                        co o 10
                 o o o
                 000
                 CO CM CO
                   So o    o o o
                   o o    o o o
                fl- O 10    -^ ui v^

                f> en in    •* co r<
                CM         r-~ CM *-t
   §o o
   o o
oo^r-O^

CM -H r».
                                        §0
                                        o
                                     en t~
                                                                              00
                                                                                                   CM oo o
                                                                                                   cnco CM
                                            O-«£ eo
                                            X
                                            UJ

                                            o
                                  r-    (O   r—    Iw
                                   S.    3    S.    3
                                   0) r— =    0) I— C
                                  *» Id C   *J  fO C
                                 •  V. 3     CL
                                                                                   
-------
                        *•>«-»
                        c  c
                     U. gj  O
                     O M *^
O

CO
O
O
SSR3
LU CO
s: LU
LU 1—1
O O
o£ LU

u. «c c?^
S: o: CM  . >- to 4->
»-« OS CO  C
I— O CFt  O
«CI— *-<  O
LO

O»

 (U
                                                      o m tr>
                                                      to CM •-<
                                    o o o
                                    ir> 10 f
                                            o of>   oo o
                                            ^> CM «—<   ^^ u) en
                          o o o   o o o
                          01   oo
                                     o o o
                                     o o o
                                     r- oo in
  SO O
  O O


en CM >->
                                                       i—     id  e
                                                                                                        I
          u>
          e
0.
I
                                              01
                                              N
                            O.

                    i      I
                            X
                     I       UU

                    H-.      01
                    1-1 T3   M

                     0) ID   OO
                     O) O
                     tO -O
                    4J C    •
                    00 O   «S
                                                                                   w>
                                                                                   e
                                                                                   o
                                                    O"«S I

                                                    tn
                                                    c
                                               •«   o
                                                                                                   X
                                                                                                  Ul
                                                                                              I
                                                              9-24

-------
                       U> i. -r-
                       002:
                       o <<-
 U_
 O

 00

 oo
 O
 o
LU 00
s: LU
UJ i—i
O CD
0£ UJ
O I— ^—•
Ll_ < O T3
Z C£ CM  O)
LU I— O  3
    oo CM  s=
LU      I  •!-
> >- yO -M
i—< C£ CO  C
I— O CT>  O
< I— ^M  O
O LU
LO
 I
(Ti

 (U
                                                          o o to
                                                          ro