OCR error (C:\Conversion\JobRoot\00000CDF\tiff\20013J7I.tif): Unspecified error
-------
\ TABLE OF CONTENTS
\
\ ,.*''
Page
1.0 BACKGROUND 1-1
2.0 SUMMARY 2-1
3.0 AIR QUALITY DATA 3-1
3.1 Data Validation 3-1
3.2 Statistical Analysis 3-11
3.3 Meteorological Analysis 3-49
4.0 HYDROCARBON/PHOTOCHEMICAL OXIDANT MODEL 4-1
4.1 Available Models 4-1
4.2 Limitations of Linear Rollback Model 4-6
4.3 Texas Modeling Areas 4-9
4.4 Emission Reduction Required 4-19
5.0 DEVELOPMENT OF REACTIVE HYDROCARBON EMISSION
INVENTORIES FOR TEXAS 5-1
5.1 Introduction 5-1
5.2 Photochemical Reactivity 5-3
5.3 Stationary Source Inventory Development . . . .5-9
5.4 Mobile Sources 5-23
5.5 Ciudad Juarez Emission Estimate 5-26
5.6 Emission Inventory Summaries 5-28
6.0 CONTROL MEASURES CONSIDERED 6-1
6.1 Existing Controls . . . . r 6-1
6.2 Extension of Texas Regulation V 6-1
6.3 Gasoline Marketing Vapor Recovery 6-3
6.4 Control of Emissions from Crude Petroleum . . .6-10
6.5 Ship and Barge Vapor Recovery 6-11
6.6 Solvent Control (Degreasing) 6-12
6.7 Inspection-Maintenance of Light Duty Vehicles .6-13
6.8 VMT Reduction Measures 6-15
7.0 RECOMMENDED CONTROL MEASURES 7-1
7.1 Austin Area Recommended Controls 7-5
7.2 Beaumont/Port Arthur Area Recommended
Controls 7-16
-------
7.3 Corpus Christi Recommended Controls 7-18
7.4 Dallas/Fort Worth Recomnended Controls . . . .7-21
7.5 El Paso Area Recommended Controls 7-24
7.6 Houston/Galveston Area Recommended Controls . 7-26
7.7 San Antonio Area Recommended Controls .... 7-28
7.8 Cost Effectiveness of Recommended Control
Measures 7-32
APPENDICES
11
-------
LIST OF TABLES
Page
1-1 Air Quality Control Regions Exceeding the
Photochemical Oxidant 1-3
3-1 Ozone Levels Used in the Oxidant Control Plan . . . 3-2
3-2 Monitoring Stations for Oxidants
State of Texas 3-13
3-3 Extent of Violations 3-15
3-4 Summary of Data 3-18
thru thru
3-16 3-30
3-17 Table of Model Estimators Highest and Second . . . .3-37
thru Highest Concentrations Observed vs Estimated . . . .thru
3-25 3-45
3-26 Mean and Standard Deviation of Predicted Minus
Observed Values 3-47
4-1 Hydrocarbon Reduction Requirements 4-20
5-1 Non-Reactive Carbon Compounds 5-5
5-2 Weight Percentage Factors for Refinery In-Process
and Product Streams 5-6
5-3 Mobile Source Reactivity Factors 5-9
5-4 Ship and Barge Reactive Hydrocarbon Emissions . . . 5-20
5-5 Reactive Hydrocarbon Emission Inventory for
Austin Area 5-30
5-6 Reactive Hydrocarbon Emission Inventory for
Beaumont/Port Arthur Area 5-31
5-7 Reactive Hydrocarbon Emission Inventory for
Corpus Christi Area 5-32
5-8 Reactive Hydrocarbon Emission Inventory for
Dallas/Fort Worth Area ; . 5-33
iii
-------
Page
5-9 Reactive Hydrocarbon Emission Inventory for
El Paso Area (Juarez Not Included) 5-34
5-10 Reactive Hydrocarbon Emission Inventory for
El Paso Area (Juarex Included) 5-35
5-11 Reactive Hydrocarbon Emission Inventory
Estimate in Juarez 5-36
5-12 Reactive Hydrocarbon Emission Inventory for
Houston/Galveston Area 5-37
5-13 Reactive Hydrocarbon Emission Inventory for
San Antonio Area 5-38
6-1 Reactive Hydrocarbon Emission Reduction in 1977
with Possible Control Measures 6-6
6-2 Reactive Hydrocarbon Emission Reduction in 1980
with Possible Control Measures 6-7
6-3 Reactive Hydrocarbon Emission Reduction in 1985
with Possible Control Measures 6-8
6-4 Employment Data by Size of Facility
Manufacturing 6-20
6-5 Employment Data by Size of Facility
Commercial 6-22
6-6 Employment Data by Size of Facility
Transportation 6-24
6-7 Student and Faculty Population at Educational
Facilities 6-26
7-1A Recommended Additional Controls for Texas Long
Range Oxidant Plan 7-3
7-1B Recommended Additional Controls for Texas Interim
Oxidant Plan 7-4
7-2A Reactive Hydrocarbon Emission Reductions under
Recommended Long Range Stategy for Austin Area . . 7-13
7-2B Reactive Hydrocarbon Emission Reductions under .
Recommended Interim Strategy for Austin Area ... 7-15
iv
-------
Page
7-3A Reactive Hydrocarbon Emission Reductions under
Recommended Long Range Strategy for Beaumont/
Port Arthur Area ................. 7-17
7-3B Reactive Hydrocarbon Emission Reductions under
Recommended Interim Strategy for Beaumont/Port
Arthur Area ................... 7-19
7-4A Reactive Hydrocarbon Emission Reductions under
Recommended Long Range Strategy for Corpus Christi.7-20
7-4B Reactive Hydrocarbon Emission Reductions under
Recommended Interim Strategy for Corpus Christi . .7-22
7-5 Reactive Hydrocarbon Emission Reductions under
Recommended Long Range Strategy for Dallas/
Fort Worth Area ................. 7-23
7-6A Reactive Hydrocarbon Emission Reductions under
Recommended Long Range Strategy for El Paso
(Juarez Included) ................. 7-25
7-6B Reactive Hydrocarbon Emission Reductions under
Recommended Interim Strategy for El Paso Area
(Juarez Included) . ............... 7-27
7-7 Reactive Hydrocarbon Emission Reductions under
Recommended Long Range Strategy for Houston/
Galveston Area .................. 7-29
7-8A Reactive Hydrocarbon Emission Reductions under
Recommended Long Range Strategy for San Antonio
Area ...................... 7-30
7-8B Reactive Hydrocarbon Emission Reductions under
Recommended Interim Strategy for San Antonio
Area ...................... 7-31
7-9A Cost Estimates for Recommended Long Range
Controls .................... 7-33
7-9B Cost Estimates for Recommended Interim Controls .7-34
7-10 Annualized Costs for Long Range Control
Measures in Texas ............... 7-35
7-11 Gas lone Dispensing Facility Ownership ...... 7-38
-------
LIST OF FIGURES
Page
3-1 Availability of Data 3-14
3-2 l-F(x) vs Concentration 3-35
4-1 Priority I Regions For Oxidants 4-10
4-2 Model Area AQCR 212 (Austin Area) 4-12
4-3 Model Area AQCR 106 (Beaumont/Port Arthur Area) . . .4-13
4-4 Model Area AQCR 214 (Corpus Christi Area) 4-14
4-5 Model Area AQCR 215 (Dallas/Fort Worth Area) . . . .4-15
4-6 Model Area AQCR 153 (El Paso Area) 4-16
4-7 Model Area AQCR 216 (Houston/Galveston Area) .... 4-17
4-8 Model Area AQCR 217 (San Antonio Area) 4-18
7-1 Oxidant Levels 1973 - 1985 (Austin) 7-6
7-2 Oxidant Levels 1973 - 1985 (Beaumont/Port Arthur) . .7-7
7-3 Oxidant Levels 1971 - 1985 (Corpus Christi) 7-8
7-4 Oxidant Levels 1974 - 1985 (Dallas/Fort Worth) ... 7-9
7-5 Oxidant Levels 1974 - 1985 (El Paso) 7-10
7-6 Oxidant Levels 1974 - 1985 (Houston/Galveston) ... 7-11
7-7 Oxidant Levels 1971 - 1985 (San Antonio) 7-12
vi
-------
LIST OF APPENDICES
Appendix A - Statistical Analysis Calculations and Figures.
Appendix B - Sample Texas Air Control Board Inventory Questionnaire.
Appendix C - Texas Air Control Board (TACB) Supplemental
Questionnaire.
Appendix D - TACB Worksheets and Data on Stationary Source
Hydrocarbon Breakdowns.
Appendix E - TACB Inventory Worksheets by Region and Industry
for Stationary Sources.
Appendix F - Exxon Ship and Barge Emission Factor Data.
Appendix G - TACB Inventory Worksheets by Region for Mobile
Sources.
Appendix H - TACB Degreasing Inventory Worksheets.
Appendix I - TACB Description of Motor Vehicle Emission Calculations.
Appendix 0 - TACB Emission Reduction Worksheet.
vn
-------
1.0 BACKGROUND
On April 15, 1973, the State of Texas submitted a plan to the
Environmental Protection Agency (EPA) for control of hydrocarbon
emissions and attainment of the national ambient air quality
standard (NAAQS) for photochemical oxidants. The plan was dis-
approved by the Administrator because it did not provide for the
degree of control necessary for attainment of the national stan-
dard. Subsequently, as required by the Clean Air Act, on July 31,
1973, EPA proposed a plan for attainment of the photochemical
oxidant standard in Texas and held public hearings on the plan.
The final plan was promulgated on November 6, 1973. Shortly
thereafter, the State of Texas, Harris County, and 23 other
litigants filed petitions requesting review of the regulations
promulgated by the EPA on November 6.
The U.S. Court of Appeals for the Fifth Circuit rendered its
decision on August 7, 1974. The Court held that the Environmental
Protection Agency determined in a legal and enforceable manner
that the State's plan was inadequate. It further held that
certain of the regulations were valid and enforceable, and that
certain of the Agency's regulations were either invalid or must be
deferred for further Agency considerations.
In September of 1974 the EPA and the Texas Air Control Board
(TACB) entered into a joint study to reevaluate certain aspects of
the plan. At the request of the TACB, the EPA agreed to a
restudy of the entire data base, instead of just those parts of
the control strategy remanded by the Court for restudy. As a
1-1
-------
result, the EPA did not immediately initiate enforcement actions
on those portions of the plan held valid and enforceable by the
Court.
The TACB study was completed in March of 1975 and was pub-
lished in a report entitled, "Reactive Carbon Compound Control
Strategy Reexamination for the State of Texas (Ref. 1-1)."
Updated ambient air quality data and hydrocarbon emission data
were used and published in this report.
On March 4, 1975, staff members of the Dallas Regional
Office of the EPA met with the Chairman and staff of the TACB to
discuss the findings. At this meeting the EPA stated that the EPA
was in substantial agreement with the air quality and emission
data base used by the TACB staff in their restudy. This was a
significant point since there had not been agreement on the
original emission data base, one of1 the major points of the
litigation.
The restudy not only confirmed the original EPA study
(Ref. 1-2) but showed a worsening of the air quality in all seven
areas of the State included in the original study. The restudy
showed that the air quality had either deteriorated in some areas
or had not improved as anticipated, that the photchemical oxidant
standard was being violated in all seven Texas areas covered in
the November 6, 1973 promulgation, and that the projections did
not show attainment of the photochemical oxidant standards by mid-
1977 in any of these areas.
1-2
-------
The technical analysis developed in this report is in support
of hydrocarbon/photochemical oxidant control strategies recom-
mended for counties in each of the air quality control regions
(AQCRs) listed in Table 1-1.
Table 1-1 Air quality control regions
exceeding the photochemical oxidant standard
40 CFR
Region Section c
Aneti n_Uaon__________________A1 1 *?A
Corpus Christi -Victoria 81.136
Dal lac-Fn»«+- Unrth------. _-81 VI
El Paso-Las Cruces-
Al amogordo
; '
UfMictnn_£a! uoc tinn____________Rl ^fl
Can Anf nn
-------
Section 1 References
1-1 TACB Report "Reactive Carbon Compound Control Strategy
Reexamination For the State of Texas," Special Project
Report No. SP-1. March 13, 1975.
1-2 EPA Region VI Report "Background and Support Document For
Texas Photochemical Oxidant Control Strategy 1973".
1-4
-------
2.0 SUMMARY
This report presents the results of a reevaluation of the
Texas Photochemical Oxidant Control Plan and recommends specific
control strategies for each Texas area which exceeds the oxidant
standard. Sections 3.0, 4.0, and 5.0 present the methodology used
by the EPA and the TACB in the reevaluation of air quality and
emissions data. Section 6.0 and 7.0 present the possible and
recommended strategies. Several appendices are also included
which include much of the detailed-supplemental information
developed or used during the reevaluation effort.
Sections 3.0 through 7.0 are summarized as follows:
Section 3.0: This section presents the evaluation of Texas
oxidant air quality data. The results of quality assurance
checks as well as a statistical analyses of the data are presented.
From these evaluations the air quality data used as a basis for
control strategy development are shown to be valid and represent-
ative of oxidant levels in Texas.
Section 4.0: This section discusses the various oxidant
modeling techniques for relating emissions of photochemical
oxidant precursors and expected air quality. As in the previous
control plan, EPA concludes that although there are uncertainties
and limitations, the linear rollback model is still the best
modeling technique available. Explanations of the linear rollback
limitations as well as the details of how this model was applied
in Texas are presented.
2-1
-------
Section 5.0: In this section the details of the reactive
hydrocarbon emission inventory reevaluation are presented. Using
the linear rollback model, the projected emission inventory shows
that additional reductions beyond the reductions from existing
stationary source regulations and the Federal Motor Vehicle
Control Program will be required.
Section 6.0: Possible additional controls are described in
detail in this section. Expected emission reductions for each of
the emission controls in each of the seven problem areas for the
years 1977, 1980, and 1985 are presented.
Section 7.0: This section presents the recommended controls
for each area along with the predicted effects of the strategy on
air quality. Although none of the areas are expected to attain
the standard, the additional control measures will provide sub-
stantial reductions in emissions and improved air quality. The
rationale for the measures recommended along with their expected
costs are also included.
2-2
-------
3.0 AIR QUALITY DATA
The reduction model for developing a control strategy for
each of the seven air quality control regions (AQCRs) under study
depends on the second highest ozone measurement for the year
requiring the greatest percent of emission reductions (beyond
that provided by the existing stationary and mobile source controls)
for attainment of the photochemical oxidant standard by mid-1977.
Data is available for these seven AQCRs frqm 1971 to 1974 (not
continuously in all cases). This year from which reductions in
hydrocarbon emissions are required is termed the baseline year.
Since the reduction model is dependent on a single air
quality measurement for each region it is necessary to determine
(1) the validity of the value, and (2) how representative that
value is of a long-term second high, i.e., is the chosen value
representative or an anomaly. To answer the first question, an
extensive data validation effort was conducted. The second
question is answered by performing a statistical and meteorological
analysis on the available body of data.
3.1 Data Validation
Introduction
The seven high and seven second high ozone values, which are
the basis of the oxidant control plan, and are listed in Table 3-1,
were carefully studied for validity. The ozone monitor which was
used to measure each value was investigated to determine whether
it used the chemiluminescence principle. The data handling
3-1
-------
Table 3-1
Ozone Levels Used in the Oxidant Control Plan
Texas
Region
EPA
AQCR
212
214
216
8 215
217
10 106
Name Year
11 153
Austin-Waco
Intrastate
AQCR
Corpus Christi-
Victoria Intra-
state AQCR
Metro Houston-
Gal veston
Intrastate AQCR 1974
Metro Dallas-
Fort Worth
Intrastate AQCR 1974
Metro San
Antonio Intra-
state AQCR 1971
S. Louisiana-
S.E. Texas
Interstate AQCR 1973
El Paso-Las
Cruces-Alamogordo
Interstate AQCR 1974
High & 2nd
High Ozone
Cone, (ppm)
1973 .160/.160
Agency
Making
Measurement
TACB
1971 .189/.184 EPA, RTP
,277/.234 TACB
197/.187 TACB
.145/.145 EPA, RTP
.380/.325 TACB-Special
Study
.130/.130 TACB
3-2
-------
procedures were reviewed to determine whether the high ozone
values were recorded on a strip chart recorder or data logger or
other permanent record showing the date and time of the observation
and the ozone concentration measured. Operating records were
evaluated to determine whether the one hour ozone values were
bracketed by valid instrument zeros and spans showing that the
instrument was responding properly. Calibration procedures were
reviewed to determine whether the ozone monitor was calibrated
using the neutral buffered potassium iodide calibration procedure
prior to the recording of the subject values. Instrument operator's
records also were examined to investigate calibration data,
possible equipment malfunctions and repairs, and other appropriate
information. Finally, documentation on equipment installation
and siting was reviewed to determine whether each instrument was
located in a temperature stable environment such as an air conditioned
shelter and was located according to good site location principles.
Conclusion
The specific data used in the development of the oxidant
control strategy for Texas have been checked to determine what
procedures were used and what quality control practices were
followed in collecting the data. Documentation of the procedures
and practices used has been submitted (Refs. 3-1 and 3-2). These
documents include strip charts, computer printouts, copies of
operator's records, calibration dates, standard operating procedures,
site descriptions and validation procedures. The review of these
3-3
-------
documents indicates that these data are basically sound for the
purpose of developing the oxidant control strategy.
Discussion
The following is a discussion of the documentation for each
of the seven high and second high ozone values:
Austin-Waco, Texas AQCR 212: 0.160/0.160 ppm high and second
high, both values measured on May 7, 1973 at 1 p.m. and 2 p.m.,
respectively.
The instrument used to measure these two values was a McMillan
Model 1100 chemiluminescent ozone monitor.
The ozone values were recorded on paper tape using a Datagraphics
Dataquire II Datalogger and Teletype Corporation ASR 33 Teletype
Print/Punch. The information punched in the tape included ozone
concentration, time of observation, and was later entered into
TACB's data storage computer in Austin.
The high and second high ozone values were bracketed by zero
values of 0.005/0.006 volts (volts = ppm) and span values of
0.384/0.377 volts showing minimal instrument drift. The instrument
zeros and spans were conducted on May 7, before the high values
were measured, and on May 8, 1973.
The ozone monitor was calibrated against neutral buffered potassium
iodide using an ozone generator as a secondary standard on May 3,
1973 and again on June 20, 1973 giving a calibration bracket of
0.400/.380 volts (volts = ppm). This indicates 5% calibration
drift over the 48 day period.
3-4
-------
The ozone monitor was located in a laboratory model of a Continuous
Air Monitoring Station (Connie) at the TACB headquarters in
Austin and according to the documentation appears to be situtated
according to good site location principles on a paved parking lot
adjacent to the TACB Building.
The operators' zero/span forms were studied and revealed nothing
that would invalidate the data. TACB's malfunction log and
validation form had not been developed at this time.
Corpus Christi. Texas AQCR 214: 0.189/0.184 ppm high and second
high, both values measured on August 17, 1971 at 3 p.m. and 4 p.m.,
respectively, by EPA personnel from the Environmental Monitoring
Branch, Research Triangle Park, N.C.
The instrument used to measure these two values was a Bendix
Chemiluminescent ozone monitor.
The ozone values were recorded on a strip chart recorder showing
that the two high ozone values corresponding to approximately 43%
and 42% of scale occurred on August 17, 1971 at 3 p.m. and 4 p.m.,
respectively.
The high and second high ozone values were bracketed by instrument
calibrations using neutral buffered potassium iodide which were
performed on July 10, and August 18, 1971. The August 18 calibration
shows that the calibration had drifted approximately 7% from the
previous calibration.
The calibration completed on July 10, 1971 occurred 37 days
before the high and second high ozone values were measured. The
calibration was done using neutral buffered potassium iodide.
3-5
-------
The instrument was located in a building called Vector Control
which was located at 3041 Morgan Street.
The station operators correctly filled out calibration forms and
response check forms. There was nothing in the operator's trip
reports that would indicate that the data may be invalid.
Houston-Galveston, Texas AQCR 216: 0.277/0.234 ppm high and second
high, both values measured on July 15, 1974 at 3 p.m. and 2 p.m.,
respectively.
The instrument used to measure these two values was a McMillan
Model 1100 chemiluminescent ozone monitor.
The ozone values were recorded on paper tape using a Datagraphics
Dataquire II Datalogger and Teletype Corporation ASR 33 Teletype
Print/Punch. The information punched in the tape included ozone
concentration data and time of observation and was later entered
into TACB's data storage computer in Austin.
The high and second high ozone values were bracketed by zero
values of -0.001/-0.002 volts and span values of 0.474/0.458 volts
showing minimal instrument drift. The instrument zeros and spans
were conducted on July 15, 1974 before the high values and again on
July 17, 1974 after the high ozone values.
The ozone monitor was calibrated against neutral buffered potassium
iodide using an ozone generator as a secondary standard on July 10,
1974 and again on August 7, 1974 giving a calibration bracket of
0.500/0.540 volts. This indicates 8% calibration drift over the 28
day period.
3-6
-------
The ozone monitor was located in a standard Connie Station at 2701
13th Avenue N in Texas City and was situated on a grass field 50
feet from a gravel road. According to documentation it appears
this site was located in accordance with appropriate site location
principles.
The operators' logs and forms were studied and revealed nothing
that would invalidate the data.
Dallas-Fort Worth, Texas AQCR 215: 0.197/0.187 ppm high and second
high, the 0.197 ppm was measured at 2 p.m. on July 22, 1974 and the
0.187 ppm was measured at 3 p.m. on July 23, 1974.
The instrument used to measure these two values was a McMillan
Model 1100 chemiluminescent ozone monitor.
The ozone values were recorded on paper tape using a Datagraphics
Dataquire II Datalogger and Teletype Corporation ASR 33 Teletype
Print/Punch. The information punched in the tape included ozone
concentration and time of observation and was later entered into
TACIB's data storage computer in Austin.
The high and second high ozone values were bracketed by zero
values of 0.000/0.000 volts. The high value was bracketed by span
values of 0.383/0.379 volts and the second high value was bracketed
by span values of 0.379/0.395 volts. The instrument zeros and
spans were conducted on July 21, July 23, and July 24, 1974.
The ozone monitor was calibrated against neutral buffered potassium
iodide using an ozone generator as a secondary standard on June 12,
3-7
-------
1974 and again on July 24, 1974 giving a calibration bracket of
0.460/0.430 volts. This indicates 7% calibration drift over the 42
day period.
The ozone monitor was located in a standard Connie Station at 12532
Nuestra Drive in Dallas and was situated in a grass field near a
school. This site was in accordance with good site location
principles.
The operators's logs and forms were studied and revealed nothing
that would indicate that the data were questionable.
San Antonio, Texas AQCR 217: 0.145/0.145 ppm high and second high,
both values measured on August 29, 1971, and at 4 p.m. and 5 p.m.,
repectively, by EPA personnel from Environmental Monitoring Branch,
Research Triangle Park, N.C.
The instrument used to measure these two values was a Bendix
Chemiluminescent ozone monitor.
The high ozone values were recorded on a strip chart recorder on
August 29, 1971 but the chart has since been lost.
The high and second high ozone values were bracketed by zeros and
spans according to routine procedures on August 28 and August 30,
1971. The zero bracket was 5.0/5.0% of scale and the span bracket
was 80.5/81.0% of scale which indicates no zero drift and 0.6% span
drift.
The ozone monitor was calibrated using neutral buffered potassium
iodide on July 29, 1971 31 days before the high and second high
ozone levels were recorded. Instrument span from the date of
calibration July 29, 1971 to August 28, 1971 drifted 1.8%.
3-8
-------
The ozone monitor was located in a fire station at Hillcrest and
Bandera in San Antonio, Texas.
The operators logs and trip reports were studied and indicate
nothing that would invalidate the data.
Beaumont-Port Arthur, Texas AQCR 106: 0.380/0.325 ppm high and
second high, both values were measured on March 7, 1973 at 11 a.m.
and 12 noon, respectively.
The instrument used to measure these two values was a Bendix
Chemiluminescent monitor. The standard operating procedure for
this special study was to use a McMillan MEC-1000 ozone generator
as a secondary standard to calibrate the monitor at approximately
monthly intervals. According to the operating procedures the MEC-
1000 was to be calibrated against the primary standard neutral
buffered potassium iodide.
The ozone values were recorded on a strip chart recorder showing
that the two high ozone values occurred'on March 7, 1973 corres-
ponding with approximately 85% and 70% of scale.
The high and second high ozone values were bracketed by zero
values and span values listed on the operators zero/span form as
follows:
March 6 March 7 March 8
Zero values 4.0 4.2 4.0
Span values 83 85 85
These values show minimal instrument drift over the three day
period.
3-9
-------
The ozone monitor was located at the National Weather Service
Station at the Jefferson County Airport terminal south of Nederland,
Texas.
The instrument operators' check sheets were examined and show no
equipment malfunction according to TACB documents.
El Paso. Texas AQCR 153: 0.130/0.130 ppm high and second high,
both values measured on July 11, 1974 at 12 noon and 1 p.m., repec-
ively.
The instrument used to measure these two values was a McMillan
Model 1100 chemiluminescent ozone monitor.
The ozone values were recorded on paper tape using a Datagraphics
Dataquire II Datalogger and Teletype Corporation ASR 33 Teletype
Print/Punch. The information punched in the tape included ozone
concentration and time of observation and was later entered into
TACB's data storage computer in Austin.
The high and second high ozone values were bracketed by zero
values of 0.000/0.000 volts and span values of 0.189/.222 volts
which were measured on July 10, 1974, and July 12, 1974, showing
substantial span drift over the two day period between span checks.
Additional comments indicate span line pressure greater than one
inch water, possibly causing the span drift. According to TACB
documentation this discrepancy was corrected in the calibration of
ppm ozone concentration (Ref. 3-1).
The high and second high ozone values were bracketed by calibration
values of .230/.230 volts which were conducted on June 25, 1974,
3-10
-------
and July 31, 1974, showing excellent agreement over the 36 day
period. The ozone monitor was calibrated against neutral buffered
potassium iodide using an ozone generator as a secondary standard.
The ozone monitor was located in a standard Connie Station at 500
South Campbell Street in downtown, El Paso, Texas, an urban core
site.
The operator's records reveal nothing pertaining to the validity
m
of the ozone data other than the high span line pressure discussed
above.
3.2 Statistical Analysis
The statistical approach taken consists of fitting an averaging-
time mathematical model suggested by Larsen (Ref. 3-3), and then
determining the expected value of second highest observation for
the period of record, typically one year. If the observed value
fell below or relatively close then the observed value was deemed
to be representative and acceptable as input for the rollback
model. In all cases, except for the El Paso 1974 value, the value
predicted from Larsens model exceeded the measured value. This
implies that the second highest measured values recorded during
the baseline years may be low rather than too high.
3.2.1. Description of Data Base
Data from the continuous monitoring sites consists of hourly
averages for oxidants. The hourly averages are computed from 5-
minute averages of continuous data. If there are fewer than seven
of these averages the data is rejected and an hourly average is
not computed.
3-11
-------
Note: An hourly average i§_ computed whenever there are more than
twelve values. Extra values are merely rejected. Table 3-2 and
Figure 3-1 outline the cities and sites for which an analysis was
conducted.
One-hour averages for oxidants are compared with the National
Ambient Air Quality Standards (NAAQS) for violation analysis in
Table 3-3. The standard for photochemical oxidants is .08 ppm,
not to be exceeded more than once per year. A violation analysis
of oxidants consists of determining whether there are two values
in excess of .08 ppm, and if so, how many hourly averages exceed
the value each year. Also included in this table is the percent
of days in which a violation occurred. Care should be exercised
in comparing percentages due to different periods of record.
3.2.2 Data Summary
Various descriptive statistics have been computed for oxidant
concentrations for each year of record. These include arithmetic
mean, geometric mean, standard geometric deviation, highest and
second highest values, percent violations (i.e., number of violations
divided by the number of averages available), percent of days in
violation (number of days with at least one violation divided by
total number of days with record available), and percent of year
for which data is available (number of available averages divided
by 8760).
Only non-zero values were used in the computation of the
arithmetic mean, geometric mean, and standard geometric deviation.
Additionally, cumulative frequency distributions of non-zero
3-12
-------
'
co
I
p3*
-
I i O
( 2:
cf LU
or o
LU cC
o_
o
o
o
:n
I
LU
s:
LU
O
O
O
UJ
ii
co
^M
1
1 t
(_J
<^"
f*S,
"X^
CM
f A
Is"* CO ^J"
"s^ (*». f*%.
CT» ^^
Is*- CO
i
i i
r-~
t~*. co sf
^r-. r-
CO CM
D- CO CO
LU ro ro
X X
. a» cu
CO 1 I
*
^>
CU 0) CU
o u o
c c c
cu cu cu
u u u
CO CO «/>
cu cu cu
c c c
r- »r» "r-
E E E
r r r
r- ! !
E E E
cu cu cu
J= JC J=
co o co
i CM CM
OO O i
000
O O O
CM CM CM
CM CM CM
C
r-
4-J
CO
.
"~ r~
CO
-*s *O
P X
r- CU
OH-
cu cu
0 0
c c
cu cu
u u
CO CO
CU CU
c c
i'i
i i
cC cu
JC J=
0 O
crt in
OO «sl-
o o
0 0
co co
CO
r-«
p.w
ro
o
t--.
^x^
CM
r«
1
I
f*x.
^NK^
o
n <^j-
r fv>
f*%, "N,,^
^, CM
CTi r~
1 1
^~ *4~
i^ r^
co
D. 4-> CO
LU C 03
3 X
. o a>
00 0 1
^3
CU CU
0 U
c c
cu cu
0 0
CO CO
cu cu
c c
E E
,5,5
'i 'i
cu cu
c ^
0 O
r~ (^^
OO CM
0 0
o o
o o
r^ r-.
o
CO
ro
a.
f
UJ
c^o
|*s»
"^s^.
CO
1
CO
ps^
^^
CO
rj- >
r t*~ CO
r^^ ^^ r^»
**^ CM ^^
CXl r r
III
r "CO CM
r-> i^ r~.
vo ^" vo
6- co co
LU ro ro
X X
cu cu
CO h- K-
«
^3
CU CU O)
u u u
c c c
cu cu cu
o o o
r/> co co
cu cu cu
c c c
r- -r- 'r-
§£ E
3 3
'i 'i 'i
CO CU CU
llp" f" _C"
CO O C_3
i si" in
CO OO OO
000
o o o
^o vo ^o
in in tr>
C\J CSJ O
X
cu
K-
cu
o
c~
cu
o
CO
cu
c
^
E
r
cu
x:
CO
co
o
o
o
CO
CO
CO
o
ro
r
s-
cu
-a
cu
z
«^"
f"^.
^x^
C\J
^
r ^*
r**« i***
*«^ N,,^
C5^ O^
1 1
r ^~
f***. p*s.
VD ^J-
^
o.
UJ 10
rO
X
CO CU
h-
ai cu
0 0
C C
cu cu
0 0
CO CO
cu cu
c c
'i'i
r- -i
cu cu
f~ f
CO 0
r VO
CO CO
0 0
0 0
in in
o
c
0
c
^^
J^
ro
CO
^~
fx,.
^^_
en
1
^J-
r~-
^
CO
ro
X
cu
1
cu
o
c
cu
o
CO
0)
c
r~
E
3
1
1
E
cu
c~
CO
CM
o
o
o
in
J>k
+->
*f»
CO
CO
fO
X
cu
h-
<4-
f^
^N^_
CM
^«
1
^J-
f^
LO
CO
ro
X
cu
1
cu
u
c:
cu
o
CO
cu
c
-------
00
o
CVJ CVJ
O
O O
o o
cvj cv)
(M cvJ
O O
«
o o
o o
m in
CO
CO
or
x
o
CO
D
CL
oc
O
O
CO
o
en 10
ro
H
o
CO
X
UJ
o
o
o
CO
CD
U.
-3
O
Z
o
CO
< CVJ
~° r-
-3 CD
<
Q
O
>-
CD
<
I
ro
O
U.
3-14
-------
«a-
cn
r
CM
f
LO
CM
";
o in «*
CM CM «*
r LO
CM
CM
r CM f-»
LO i «a cn
CM i-s
en
LO
O
»
LO CM r--x
oo vo cn
r CM
^
00
ID 00
f*N CO Si" r
cn r !*>
O
CO
i
'«*«=fU
o o «* s
co r-
*~m
en O CO
LO r CM
CM
10 co cn
»
i r LO LO
CM
to r
00 r 00 O
CM VO
CM
' r CO CO
*»
in LO «a- vo
LO CM in
CM
CO
o
II
CO I
CO
o
cr> o
LO
CM
CM i
^j- r^ co
cr> ro vo KJ-
00 r CM
CO
CO
E re
Q. i
Q. O (U
CO -r- r
C > XI
r- o re
31 - C r-
»-> -r- -r-
"o re re
C i 10 >
o o >, Q re
co +»
M- s- re
> O O Q
C
o
o. re
Q. r CO
' OO O r
C -r- XJ
r- o > re
o re re
c r- to >
o o >><:
o -r- re
O) > Q t- re
** ^5 CD C^
LO i CO *3-
CO r- CM
OL. re
Q. r 0)
00 O r-
C -r- J3
i- o > re
o re re
c i oo >
o o >,<
(J -r- ro
0) > Q re
CO +J
o>- «»- ro
« O O Q
O
CM
i r^. oo
«t
O O r CO
CO LO
CO
i CM CM
CO O i
c o o o
"-4J O
00 CM
3 CM
< O
co
co
f"- j \.
CL re
CL r QJ
oo O i
C -r- JQ
r- o > re
rc£>£
o re re
c oo >
o o >><
u !- re
o» > Q re
oo -(->
M-M- re
- O O Q
CL
Q.
tO
O
re
r re
TD re re
o 'o >>cc
o -r- re
cu > Q re
co +J
M- 4- re
* O O Q
O
a. re
o. i cu
^-- oo o
c: -i- jo
r- o > re
-o re re
c i oo >
o o >, Q re
CO -4->
M- M- re
> O O Q
LO
OO O O
Q.O
i- LO
O i
O r
CO CO
oo O O
re
o
"re co
Q r-
00 O
re
r- 0
re co
Q
o cr> CM
oo o O
re
a. o
o
i r-.
C. CO
3°
00 O
3 VO
O LO
3C CM
3-15
-------
«3-
I
cn
CO
co
CM
co o co
CO
co o cr>
oo
co
-» CM
- m
co
i
CM
CM i
LO
CO O -si"
t* * t
F~. *f CM 00
O CM CO
CM
cn
IT)
co
cn
CM o co
** CM ^J- »~~
*O I LO
CM
CO
<£> VO O
co
, cr> co CM
co r*v
co
CM
cn
10 oo co
^" in ^- vo
CM CM in
CO
to o <
r-. p- co
p- P- CO '
co
o
o
co
i co
co z
o
LU II
I r-
CQ <;
I O
LU
I
X
cn
O
in
in i
in
in
i oo m
co cn
p- CM
CO 00
in CM co in
«=f P- CM
Q. re
Q. I O>
CO O P
o re re
c p- co >
O O >,ec
u T- re
o > Q re
M- 4- re
« O O O
CO
o
CL re
Q. r <1)
* CO O P
C -P- 43
p- o :
CL
Q.
C
O
re
co o i
o > re
CO
c
o
"i" T> e
Q. re Q.
Q. P- O> CL
« CO O i ' CO
E -P- -a E
p- o > re T- o
2-p-
^ re re
C P- CO >
o o >><:
o T- re
a> > Q re
CO 4->
«*- 4- re
« O O Q
T3
O
(J
CO '
re re
co >
o - .
p- re
o re
o re
o i co
E O _
o -p- re
O) > Q
re
re
" O
4- re
o a
o o
o T-
a; >
co
4- 4- re 4-
" O O Q > O
re
4-> QJ
re P
o re
re
>
co re
re
Q
co
aS
"CM
^^^-^
p- in
E CO CO
o o o
CO O
3 VO
O LO
a: CM
"C3
c co
re o
p- O
s_
C^ C3
~O CO
(U CO
z co
r- * '
f~
O P
+J CO
E O
(~~)
re in
to ^t
>>
+->
r- CM
0 0
O
CO
re o
I in
CD
E
re P
s- o
o o
4-> O
CO CO
-------
values are provided for each city and site on an annual basis.
For Houston and Nederland cumulative frequency information is
provided for the periods March 1973 through December 1974 and June
1972 through June 1974, respectively.
Tables 3-4 through 3-16 summarize the available data. The
tables clearly indicate the extent that the national standards are
being violated. Observed oxidant levels have reached peaks in
excess of five times the standard at Houston (August 55 1972;
Clinton Drive site). In 1974, peaks ranged from .125 ppm at
Austin to .277 ppm at Texas City.
3.2.3. Analysis
In this section attention is focused on a statistical eval-
uation of the high air quality values that are used to determine
the needed emission reductions. The approach taken is to use the
second highest observed concentration for the averaging time equal
to the standards. The approach taken is to approximate, as closely
as possible, the distribution of air pollution concentrations.
Larsen has recommended that at least one year of almost
continuous sampling is needed in applying his averaging-time model
to calculate expected maximum concentrations. Since the available
data base varies considerably from one AQCR to another, a more
accurate estimate of the yearly maxima is possible whenever an
AQCR has sampled for an entire year. Also, the probability of
measuring the yearly maximum concentrations decreases as the data
base diminishes. The number of data points used for the analysis
in each region is listed in Column 8 of Table 3-4 through 3-16.
3-17
-------
3
00
UJ
O CD
:r
. LU
a
on i
H^ :
LO i
0 0
r CM
00 o
coo
MOO
10 CM CM
3 CM CM
0:00
>- L' I
I ci
CD _J
s: o
oo f
>- ct
cf.
LU
o
co
o
3-18
-------
I L-°
"LU
o
UJ
_l >-
ca c£
CM
CO
U) OJ
3 CVJ
< O
>- us
I a
-« o
o o
< 1
t 03
§3
a: oo
=3 UJ
oo
_ 'ZL
o a
£ i
u <
ca
s o
-I t/)
2:
o
o 3:
o ca
oo *
r-.
o
CM
cn
in
3-
a:
I co
oo >-<
in
CM
co
CVJ
o
CM
CO
o
o
CO
oo
ro
o
o
CM
a-
o
Lf)
O
CO
co
o
en
o
3-19
-------
|
£
LU
CO C£
t-^ a;
^i
00
UJ
_1
o: CQ
I/> LO r
r- O O
S- f U-
^
CJ i i
oo cn
woo
0.0
s- LO
o i
CJ i
> o
CJ O
o<
LU OO
LU
££
oo
z
o o
oo
Q CD
n:
it OO
=e a
Q
00
OJ
*3- uo cn
OJ OJ l~
OJ
OJ
a-
CO
cn
to
CO OJ
in to
o
CM
* i
O CM
i OO
>~ I
"=f
0
O
o
oo
o
O r- 00
r Cn CO
100
cn i LO
CO r OJ
3-20
-------
O Q
O
< 1
I 03
=C
r- a
en
CO CO
> O O
to CO CO
O T I
> LU
I Q
1-1 O
O CJ
LU
o: co
ID LU
O CD
CO 1
£ O
<3
o
CD :
r> i
=c!
Q
O CO
^. LU
F ro f~~
ro co
r-- r
03
n
UD
Ol
i CO ID
10
CM
LO
CO
ro co
i CM
o
r-.
oj
ir> vo
CM CM
LO CO
CO CM
i CO
cn cn
O
us
CM
CM
ID
O LO O
LO r CM
O O
g
CD
CVJ
CO Q
H: lu
o a
t-t LU
I LU
<_>
X
I LU
_l
^S
LU
ID LU
-------
0<
OO
oS
CO Lt_
o
UJ
o
o
fO
I a
>-« o
<-> CJ
UJ
1
I^
00
a: oo
=D
O
00
z
o o
s: o
"=C
a o
Q OO
~z
o
CD
cn
OJ
O
ui
o °°
"> S
OO Q
2: LU
o a
ct
M O
Q UJ
O <
ID
o
UJ 1-1
a: I
_ uj
13 a_
o
3-22
-------
O O
O LJ-
O CT1 C\J
t/J O O
re
Q- O
o
>- UJ
( Q
i-> O
O <_>
CQ
O <
o:
o it
o
uj in
.
o i
LU
o cu
00
^
O C
UJ <
CO _J
s o
:n
-i i/>
:=:
o
U~l i->
<<
o
o >
<§
O CD
LU 1-1
oo re
cvJ CM
en i
00
CO
in
10
o o
>-< UJ
uj
=> o
OQ X
UJ
01
CJ
o
CVJ
o
CO
o
CO
CM
o
O VO
"* CSJ
o o
o
CO
o
O
CM
CO
o
LT) CO
IT) ^1-
O O
in
hv
o
3
in
*
3-23
-------
CQ
£3
ii
a <
3 O
I-H a:
Cf. O
ro u.
LU°
_J >-
CQ c£
ss;
C CO
o o
o m
n: CM
>- LU
I- a
GS
LU
CO
LU
O CO
CO
O O
o: H-
LU <
O
z >
<
So
o
is
o
o
CO
o
O >-i
Q CO
Z LU
o n:
o
LU
CO I
i CD
LO CO
r CO
CO CO
CM
i
Csl
Ol
<3-
CM
g
CO
cn
LO
CM
CM
cri
r CO
O CM
CM CO
IO «*
* O
CM CM
LO en
LO 1
CM CM
o
LO
CM cn
CO CM
o o
o
CM
en
ro
0
o o
t. LU
CO
} >-i o
CO
iO
0
z
OL
'. LU
O
CM
IO CM
CM IO
CM i
ID CM
LO i
CM CM
3-24
-------
cou-
o
iin
c coco
o oo
4->
l/l O O
3 ID <£>
o in in
3: CM CM
>- UJ
I Q
-i O
O (_>
LD
ct: 03
0
>-
_1
c£ oo
=> UJ
O C9
31 Kf
cz
LL_ UJ
O =>
<
O
z.
oo
u_ z
C^ CD
Oi 1
UJ eC
CQ _J
S 0
r^ >-«
^ >
Z
O
OO 1
>- et
< 1
Q O
SS >
z^
3:
i
K-< OO
r^ ^^
0
t/) 1 1
^* H~
<: <^
O _J
CD
o >
^L
00^
>- =t
2Q
^c
\~~
1
Q OO
^^ 1 1 1
CD 31
c_> C3
Ltl li
OO 3;
HIGHEST
K.
<
UJ
>-
^_
1^.
en
i
^_
r-
vo
in
cii
CM
en
r^
CO
co
in
CM
CO
«3-
co
CO
CO
to
en
o
in
in
in
r^.
en
CM
r-~
^*,
CM
1
CM
r*.
^^
to
CO
to
in
to
in
en
CM
en
«d-
m
r*.
CM
CO
^3-
CM
CO
in
=f
CM
en
t~~
CO
«!
CM
^f
1 1
r^
en
CO
t~~
^*^.
CO
OO
r^.
^
CO
t
CO
^~
^^^
CO
(^
in
CM
C '
en
o
in
^
o
^i-
CD
CO
in
r
CM
en
in
I
io: I-
I CJ
i a:
'. uj
i a.
3-25
-------
LU
_J
ix. cn
S5
O eC
O
LU CO
o: i-<
n Q
3
ro Lu
O
-a
c ro
LU
1 Q
i~i o
o o
C£ CO
_ LU
O CD
CO
2:
o o
C£
LU
ca
1^ O
t» CO
O
3
Q CO
21 LU
O 3C
CM I
r^- co'
CM \i
i 00
O LO CM VD
Lf) CM ro CM
000
I
CM
I
<
CO I
o
CM
LO ID
i-^ an
CO CT! O
ro ro co
o o o
co
CM
ro
ro
CO
OJ
CD CT\ O^ ^"
CM LO LO «3-
O O O
co a
_: LU
O Q
»t LU
. LU
_J (_>
OQ X
CO O
«i- co
CD
CO
CO
O
cn o
r*- co
o o
co o
tD LO
cn
o
co
co
CM
co
CM
ro i
ro
cn
cn
ro
co
t-.
01
3-26
-------
en o;
< <
H- E;
O i U3
+J 00 00
COO
>- uj
I Q
-H O
CJ (J
UJ
Lu
nc 5
Oet
U 1
O. e£
Qi >
0 «£
(_>
LU CO
0 H-
Q §
O
t i :c
c£ o
LU 1-1
ex :c
-^
u_
<§
^
1 e^
O >
<:
CO
c
]
53 ct
5
>
>-
QL I/-.
O CS
n:
*=c
o
2:
CO
o d
2 5
CO 1
s o
2^ >
Z.
0
1 1
>:*
^ o
1 1
ti3 >
z.
\ i
:n
-H t/l
12 ^
o
*" ^
Q Jj
0
O >
f?L
<.
CO I
o
in
. i
~2Z. ~3i
I
Q CO
Z LU
O Z
_> CD
/> o:
t
CO
:n
cu
1C
o:
<
LU
>-
^
J^
CM
V,
i «a-
r~ r^
1
LO P^.
^f r-
'
LO CTt
a- <*
i «3"
en en
CO O
O 0
>- LU
UJ
03 X
1 1 LU
O£
h- CO
/5 1-1
a LU
^3
ys
UJ
ID LU
O-ST
LU i i
X. 1
LL.
U_
UJ O
^
V-* |
~~ ^^r
< LU
IS
CJ
*
O «3-
LO CM
0 *
*
O O
^1- CO -K
O
*
O LO
n co
0 *
*
0 0
CM g
X
*
CD O
^- LO *
O
LU
i
cc
1
1 i
I
1
o
z.
r- L°
O *
*
X
*~~ LO
«* *
*
K
O ^ *
"-;
i ^t-
r^ r^
CTi 01
3-27
-------
I
a: m
°3
si
<
a: o
LU i i
a. n:
§3
i a
"LU
o
LLl
_J >-
ca a;
i- CM
O O
O
CO r
I LT)
>- IJJ
I a
1-1 O
<_> o
LU
OL CO
ZJ LU
O C3
CO
. z
o o
a: _i
oo
O i-i
2: 3:
a co
LU
o n:
<_> ca
UU 1-1
co ac
co a
UJ
o a
H-I LU
1 UJ
ID CJ
OQ X
O i
LO CSJ
c
O O
o
O CT*
CO CO
o
o o
CVJ g
O
LU
ID LU
o-s:
LU >-
CtL \-
u_
LU O
I I
2:
: LU
I O
3-28
-------
Q
p *C
JJ O
00 00
>- LU
I a
i-< O
o u
(~
00
o
<,
U_ LU
O >
OL \
LU c£
CQ I
s o
s
O
O O
«!-
oo'
CO
10
in a
LU
o
Qi
>i O CM
I r
Q LU O
o
CM
O-J
LO
CTv
CO
CM
o
o
=3-
o
CO
00
O I--
CM «a-
o
LU
__ CJ
=) CK
S LU
-^ O.
3-29
-------
g*
O h-
Q
a: o
cu <-<
o; co
ID cu
o o
U. CiJ
o >.
co
L o
u_
o
cu
CO i
o o
Cc. U-
T3
C ^f CO
C IO OO O
O r O O
4-> S-
CO CD O O
3 T3 «3 CO
o cu co co
>- LU
I Q
>-" CD
O O
IU
.
o
to 1
O CO
^L CU
CD m:
c_j o
UJ >-H
a-
CM
co r
oo
-~-.o o
CO r- i
CO
CO
3-
CM
CM
co'
CO O
2: cu
o a
o
CQ X
CU
a:
ID
*3-
CM
O 1
1
1
I- T3 -3-
io co ^« cu c CM r^-
o en i cu i en i
O
in
co
CM
O
O
*
CO
CM
O
a-
co
o
o
co
o
co
o
CM
o
O
CM
O
*
O
lf>
CO
O
O
=> cu
o-s
: cu
1 O
) a:
co
us
CM
co
co
CM
c
o i ii
+-> *3- i- T3
c/) CO r-~ CU C CM ^~
o a* i cu i en en
3-30
-------
Because of the lack of sufficient historical data in some of the
AQCRs and because maximum concentrations vary markedly from year
to year, it is impossible to determine if the second high value
being used comes from a "clean year," a "dirty year," or an
average year. The probability, however, is high that the specific
value chosen is not the all time second high that could be generated
by the worst combination of meteorology, site location, and emissions,
Thus it seems reasonable to assume that the use of the available
data would not overstate the severity of the air quality problem.
The above limitations clearly indicate the need to develop
techniques which may be used to predict maxima and to evaluate
statistically the credibility of the observed peak values. The
method being adopted is attributable to Dr. Ralph I. Larsen of
EPA's National Environmental Research Center in Research Triangle
Park, North Carolina. Larsen suggests the use of the expected
value for the annual maximum concentration as the design value on
which to base control strategies. If control strategies are
developed and applied based on this expected value, then the
standard can be expected to be violated in only one out of every
eight years. Similarily, the use of the second highest value
measured has the expectation that a violation will occur every
other year (Ref. 3-6). By use of Larsen1s methods, an indication
is given whether observed maxima are consistent with, and predict-
able from, the data or should be considered a freak event; outliers
from the population represented by the data base. It should be
3-31
-------
noted that much work has been done on the question of outliers but
the largeness of our sample sizes make these procedures inappli-
cable. Current research is being conducted to treat air quality
data specifically.
3.2.4. The Method
Previous reports by Larsen (3-3, 3-4, 3-5, 3-6) have recom-
mended the use of an empirical mathematical model to describe air
pollution data. One of the characteristics of this model is that
air pollutant concentrations are lognormally distributed for all
averaging times. Potential theoretical reasons for these char-
acteristics have been cited as well (Refs. 3-7 and 3-8).
The lognormal distribution is a left-skewed distribution. A
histogram of lognormally distributed data would have a short left
tail, a hump skewed to the left, and a very long right tail. In
air quality data, this indicates the abundance of lower concen-
trations and a few very high concentrations. This distribution
can be easily transformed to the well-known normal distribution by
considering the natural logarithms of the concentrations rather
than the concentrations themselves. However, a histogram plot of
this nature would give little more than a qualitative description
of concentrations. A simpler evaluation of the data rests on
another feature of the lognormal distribution. When the cumulative
frequency distribution is plotted on log-probability paper, a
straight line is produced when the data are perfectly lognormally
distributed (Ref. 3-3). Let us suppose that there exists an
3-32
-------
infinite population of possible one-hour ozone concentrations. When
viewing each hourly average as an observation from this distribution,
a sample of size 8760 would represent all the hourly averages
possible in a non-leap year. If the sample is representative of
the population sampled, then empirically the data should produce a
plot on log probability paper which is fairly linear. If such is
the case, then the observed concentrations may be used to generalize
the potential population of concentrations.
The lognormal distribution is completely described by two
parameters, the geometric mean and the standard geometric deviation.
The first task with any body of data which appears to fit the
lognormal model is to estimate these two parameters from the
sample values. These estimates can then be used to calculate
expected maxima which may be compared to the observed peaks. We
may then determine whether recorded peaks are typical and expected,
or represent outliers from the total population.
The straight line representing the lognormal distribution on
log-probability paper is used to provide estimates of the two
parameters, since the slope of this line is related to the standard
geometric deviation and the point at which this line crosses the
50th percentile point is equal to the geometric mean. Theoretically,
any chosen set of two points to determine the slope should give
the same result. However, with the data in question this is
clearly not the case in many instances. Oxidant concentrations
seem to depart from lognormality in the extreme upper tail exhibiting
3-33
-------
a tapering-off effect. This is suggestive of a mixture of distribution
functions more adequately representing these concentrations.
Larsen suggests that drawing the lognormal line through the .1 and
30th percentile points should provide an adequate fit for determining
the expected annual maximum and expected annual second high.
Two methods were used in assessing the validity of the lognormal
assumption. One, as indicated previously, is to plot each hourly
average on log-probability paper and if'the resulting plot is
fairly linear this indicates that the distribution of these averages
is lognormal. Another way by which one might assess the validity
of this assumption is to plot l-F(x) versus concentration on semi-
log paper, where concentration in ppm is plotted along the abscissa
and F(x) equals the cumulative frequency. If the resulting plot
bends upward, see Figure 3-2, the distribution is classified as
heavy-tailed; the lognormal distribution is a member of this class
of distributions (Ref. 3-8). The largeness of the data bases
involved made plotting each point prohibitive. Instead only a few
selected points were plotted.
We turn now to how the expected maximum concentrations were
obtained. If the population of concentrations had been subjected
to 100% sampling, i.e., 8760 hourly averages, and were totally
accurate, then a good estimate for the expected yearly maximum for
a particular year would be the observed maximum for the year.
Likewise for the second highest value. However, 100% sampling is
not available for any of the sites, maxima vary markedly from year
3-34
-------
if K)-2
10-3
10-4
V ''
-\
\
1 1 1
1 1 1 1 1 1 1 1 1 1
XK
% N> _
\ N^k ^%,
* ^k Nfeft.
\ \ "<&
1 1 1 M 1 X 1 ^K 1 1
CONCENTRATION, ppm
Figure 3-2 - l-F(x) vs. Concentration
3-35
-------
to year, and as mentioned previously, there is also the possi-
bility that the actual maximum concentrations went unobserved and
unrecorded. This is much more probable whenever 3rd quarter data
is not available (June 1 - September 30) since high values usually
occur more frequently during this period of time. Hence a better
estimate of the 2nd highest value would be its expected value
which is calculated by use of the estimated parameters of the
model. Using this expected value, one can ascertain that the
value chosen for input into the reduction model is consistent with
the body of data and does not represent an anomaly. The expected
values obtained were those that one would expect during the period
of record and not necessarily those that would occur for the entire
year. For example, in 1971, data for San Antonio was available
for the months of June through September, hence the calculated
expected value is that which the model predicts for that period of
record. One would not want to extrapolate over the entire year
because the sample available probably does not represent the true
proportion of concentrations likely to have occurred during the
entire year. Had sampling been for the entire year, the lognormal
line would have resulted in different (and more accurate) estimates
of the population geometric mean and standard geometric deviation.
3.2.5. Results of Analysis
The expected maximum and second high concentrations for each
year were estimated using the method previously described.
Appendix A gives an example of the procedure used in attaining
these values. In Tables 3-17 through 3-25 observed values are
3-36
-------
03
ct
c
CVJ
I
O CO
(Jj LU
1 3:
h- CO
IS
CM ^~
CM i
CO Z
cs: LU
ooo
I 2: uj
< O h-
E: o
-a
c
CM
1
a co
jj iij
S o
LU 1-H
CO 31
CO
O
(/) CO I
UJ LU CO
3: tu
CO i-l
O IE
o rc to
o >
2: o
z o
U_ O LU
O O >
LU K
LU I/) LU
_J 00
CQ O CD
eC 2: O
UJ 1
S I
o =<:
LU K-I
CD :
UJ
i a
2C LU
os:
CM i
CM CO
O O
CM CM
<=>T
oo
o o
CM CM
CM CM
COO
uj nr
: - LU
I1 I--1
CJ to
a'.
cc
3-37
-------
I CO
UJ
±c
CD
/) Zll
O O D
l s: UJ
eC O I
I/) I/) 1
UJ UJ CO
U_ O UJ
O <_) >
UJ ce
UJ CO UJ
_ I CO
CD O CO
oC z: O
I- <.
4-> i CTi
in co i
r-00
V
jC O O
o en en
O-
i-
O UJ
CJ
>- Ul
CO CJ3
CO 14
O re
co t-~ n
i CM «3"
CM i i
CM tr>
co t
CM i
£8 ^
f)
CM
Ol
O
CM
O
3-38
-------
O O O
I :r I.U
cC O h-
I/) CO h-
UJ I.U CO
U_ O LU
O O >
LLJ C£
UJ (/I LU
i on
CO O CD
O O
O O
>- LU
I I-
CO
LU
LU rr;
CO CD
CO t-H
t O
Jj I*
1£ I
O <
LU I-H
=c
>~ LU
O ~~
O
co
U3
O
OJ
10
OJ
OJ
CO
CM
O
00
t
en
LO CO
OJ CXJ
O O
co i
C\J CO
C O
3-39
-------
CQ
_
CO O CQ
«4 2: O
p <
(/>
CC. LU
LU m
O
03 1-1
r r-
00 CM
00
o o
oo
uj en
o
>- UJ
I I
o to
o <
LU I*
O .-J-
LU
O
a
LU r^
s: <:
co
CM
l~. 00
OO CO
O i
LO
LO
cn oo
i CM
o o
i CO
CM CM
O O
3-40
-------
to
z.
o
00 CO t
UJ UJ CO
3; ui
o >
s: Q
z: Q
U- CO '.AJ
o o >
LU a;
uj to uJ
_! to
CCl Q CD
< i: O
h- <
I-2C
< O
Ti
CM
t
JQ CO
LU
H
UJ t-t
co n:
C\J CO
t^ r^
01 01
ro co
o o
gy
o
o
a: ^
i o
O c£
LU ii
ei >
UJ
o
O O O1
OO »3- CM
r CO CSJ
Ol ro 00
O t^ ID
CM CO CM
O
CM
un 01
LT) r
CM CM
LD
Ol
00 ID CM
i CM CM
O O O
CO
O
CM
O
CM
O
CM
O
3-41
-------
CM
CM
I
co
CO
z
o
DC
O
O I-
1 C_> <
CO CO I
UJ LU CO
o: LU
U- O LU
o cj >
LU C£
LU CO LU
_l CO
CD O CQ
a: Z O
CO
LU
m
CD
o o
C CO
ro co
f CO
S-
-> o:
O
xl H-
I CO
LU
o; CD
oj»i
co :c
co
o
CO CD
CO M
o 3:
CO t^ 00
00 CM 00
ro ro i
§ 2
i «3" CO
i CM 1^
ro ro i
o> r-.
r~ oo
CO r-
co r^
i co
co
CM
o
LO
n
o
oo o
n ^-
o o
o
n
o
3-42
-------
ro
CM
o
c
in
;z
O
to z.
OL UJ
O O O
H- Z UJ
eC O I-
21 O et
n: uj
_1 C'J
UJ i-<
D a: 10
o >
s: a
z o
U- O UJ
o o >
LU Q;
LU (^ UJ
_j to
CQ Q CO
< 2: o
p <
-a
c
CM
I
I/)
LU UJ
cc: CD
01 to
o o o
coo
o I*- r*-
+> in uo
o
>- UJ
o: 2;
I o
<*-
o
ai
s_
+j
-------
o
>- UJ
I h-
O l/l
o GO
UJ L'J
1-1 CD
t >-"
GO 3t
a:
C3
LU 1-1
pet
UJ -<
Its >
CM
LO
CM
CO
CM
oo
o
OJ
o
o
m
o
3-44
-------
a
c
O CO
UJ LU
to
z:
o
» i
§
CO ^
o; LJJ
o o a
h- -.2 LU
=C O I
s: o -5
z o
u. o iu
o o >
uj a:
iu oo uj
_i co
co o co
ct ^: o
C3
rc
-o
c
CM
o
ai o
en
- LU
I t-
O CO
CO >-i
o re
o;
^!
-U et
K
Ol
K
^f
«3-
CM
LO
cn
CM
O
"2
o
s
>.t-
i O
°-§
ITJ '-
4-> S_
ro OJ
o a.
i- O)
01 S-
!-> -^
s- +->
10
01
S- !-
i- S-
J= 3
t T3
c -a
o a)
>
-o i-
ai a)
3-45
-------
compared to those predicted by the model. The predicted values
are those that would be expected had 100% sampling during the
entire period of record been performed, hence in some cases
the predicted values are considerably higher. When this extra-
polation is not performed, model estimates more closely approximate
the observed values. For each site, model estimates and observed
values were compared using the points 1(1)5(5)50. It was found
that the mean difference between these values ranged from .014 ppm
underprediction to .006 ppm overprediction, with standard deviations
ranging from .004 ppm to .045 ppm. These results are presented in
Table 3-26. Figures 1 through 18 of Appendix A provide a visual
assessment of the lognormal fit. The points 1(1)5(5)50 as well as
the .01, .1, 10, 20, and 30th precentile points were plotted on
log probability paper and a line drawn through the .1 and 30th
percentile. For the graphs on semi-log paper the points plotted
were, l-F(x) - .0001, .001, .01, .10, .20, . . ., .90. The
theoretical distribution derived from using the estimated para-
meters is plotted alongside the observed frequency for the graphs
on semi-log paper. Thus, the model used appears to be a reason-
able approximation to the distribution of oxidant concentration
and the model estimates of the maximum values lend credibility to
the occurrence of the levels that were measured at each site. The
estimated values appearing in Tables 1 through 18 of Appendix A are
indicative of what could have occurred.
3-46
-------
TABLE 3-26
MEAN AND STANDARD DEVIATION
OF PREDICTED MINUS OBSERVED VALUES
5 » Mean Difference s * standard deviation
SITE YEAR D s
Austin 1973 .005 .015
Austin 1974 .000 .007
Corpus Christi 1971 - .08 .022
Corpus Christi 1974 - .002 .011
Dallas (039) 1973 - .001 .011
Dallas (039) 1974 - .002 .011
Dallas (045) 1974 - .002 .011
El Paso 1971 .002 .003
El Paso 1974 - .003 .004
Houston (034) 1973 - .001 .026
Houston (034) 1974 - .004 .019
Houston (035) 1972 - .014 .045
Houston (035) 1973 - .005 .015
Nederland 1972 .006 .034
Nederland 1973 .001 .007
Nederland 1974 - .004 .011
San Antonio 1971 - .003 .011
Texas City 1974 .004 .013
MINUS SIGN (-) INDICATES UNDER PREDICTION
3-47
-------
The results also indicate that the assumption of lognormality
is a reasonable one and that the observed second high does not
appear to have been an anomaly for any of the sites during the
baseline years. As shown in Table 3-20, El Paso's second high
reading is higher than expected. However, it is close enough
to be used in the rollback model, judging from the variability
of mean differences. This is especially true when one considers
the fact that use of the second high vajue has the expectation
that the national standards will be violated almost every year.
3.2.6. Model Limitations
Some possible limitations in the results should be noted.
Estimated values for maximum concentrations are highly dependent
on the choice of points used for estimation of the lognormal
parameters. The .1 and 30th percentile points were selected since
a lognormal line drawn through these two points adequately represents
the upper 30% of the distribution. It has been noted that air
sample concentrations are not independent of each other, as is
required for model use. Ozone levels have a pattern of seasonal
and diurnal variation, higher levels occurring mostly in the spring
and summer between the hours of 10 a.m. and 5 p.m.. Some dependence
may also exist between sequential observations due to relatively
slow variability in the dominant parameters influencing air
quality. This non-independence of data appears to cause an
overestimate of the expected maxima, which is almost balanced out
by the mathematical approximation used in computationwhich tends
3-48
-------
to underestimate maxima. Neustadter and Sidik have used a Monte
Carlo technique and calculated the underestimate to be about 4 to
8% (Ref. 3-10).
Another limitation lies in the inability to test the goodness
of fit of the model by some criteria. Only a visual assessment of
such goodness of fit was presented.
3.2.7. Conclusions
Although the mathematical model used to describe the distribution
of oxidant levels has some limitations, it is seen that it can
adequately estimate the air quality levels and hence, lends
credibility to the fact that recorded values are consistent with
the body of the data for the period of record and do not represent
anomalies. These estimated values have been presented alongside
the actual observed peak occurrences. Thus the use of the second
high observed during designated baseline years are justified for
use in determination of needed source reductions.
3.3 Meteorological Analysis
A recent study (Ref. 3-13) was conducted for the following
purposes: To evaluate the meteorological conditions which coincide
with high ozone measurements in the seven Texas AQCR's; to determine
the extent to which certain meteorological conditions characterize
the high ozone days; to determine the expected frequency of occurrence
of high ozone days; and to examine the meteorology of the seven
high and second high ozone values which are the basis of this
oxidant control plan.
3-49
-------
3.3.1 Description of Data Base
In order to evaluate and characterize meteorological
conditions associated with high ozone days, groups of meteor-
ological data were collected for the ten highest ozone days
for each year, 1971-1974, at eleven monitoring stations (totaling
24 station years available) representing the seven AQCR's.
The meteorological data collected included solar insolation,
temperature, sea level pressure, wind speed, wind direction
range, time of day, season of year, and comparison with stability/wind
frequency distribution ("STAR" data).
3.3.2 Description of Data and Method of Analysis
The following section is a discussion of the method of
analysis and data utilized to perform the analysis. It should
be emphasized that this study is totally retrospective in
nature. Data used in this analysis, both ozone and meteorological,
were recorded and reported for different purposes and were
sometimes not readily compatible. While some parameters known
to be reported would seem easily comparable, frequently other
approaches had to be devised once the data was reviewed.
Solar Insolation
Solar altitude for any hour and day at a given latitude may
be calculated to an accuracy of within one degree. Solar radiation
in langleys per minute cannot be calculated readily without precise
information describing the absorptive characteristics of the atmos-
phere in which the short wave radiation travels. Therefore, the
3-50
-------
accuracy inherent in a calculation of solar altitude would be lost
if calculated solar radiation were the sole parameter used to
characterize solar insolation.
In characterizing solar insolation on high oxidant days in
Texas, both solar altitude and solar radiation classes were used.
Since only afternoon hours for Texas high ozone days were used in
the calculations of solar radiation index, only the indices 0
through 4 would be expected at noon. Further, since most days had
unlimited ceilings and less than 50 percent cloud cover, it would
also be expected that predominately the classes three and four would
result at this hour of peak solar insolation. Maximum daily solar
altitude was tabulated in five degree increments for characterization
of the days of the 10 highest ozone concentrations.
Temperature
For the purpose of characterizing the temperature of the high
ozone day, both the temperature at the hour of recorded high ozone
concentration as well as the average daily temperature, were
tabulated within 10 degree temperature intervals. When the hour
of recorded high ozone occurred between two reported hours, an
interpolation was obtained between these values. Although this
method assumes the temperature was between that reported at the
three hour intervals, the probability is greater that the interpolated
value represents the actual temperature at that hour than if either
of the two surrounding hours were chosen. Further, in light of the
fact that values used to characterize the temperature at an urban
3-51
-------
ozone monitoring site were reported from the meteorological station
within closest proximity, this loss of accuracy due to interpolation
is within the boundaries of allowable error.
Surface Pressure
The synoptic climatology of Texas is such that in the seasons of
high insolation, centers of high pressure over the state occur less
frequently than many locations which have served to define classical
<"*
oxidant formation theory. This observation is further complicated
in the western portions of Texas where the synoptic climatology is
often dominated by a semi-permanent "heat low."
Accordingly, each high value case was classified as to whether
anti-cyclonic or cyclonic curvature prevailed over the area of
interest. Cases where the curvature was ill-defined were classified
as neutral. Cases which were well-defined as having either ridges
or centers of high ozone presure in the area of interest were also
noted as such.
Sea level pressure at 7:00 a.m. was obtained from the surface
weather maps. Where climatological stations were not within close
proximity, interpolations were made between surface pressure values
observed at two or more stations as appropriate. Each surface
pressure value was then compared to the climatological average sea
level pressure for the month when the high ozone day occurred in
order to determine if the pressure was generally higher or lower
than average. This information may be considered supplementary to
the extent that there may be high or low pressures dominating large
3-52
-------
portions of the country, with Texas receiving relatively ill-
defined curvature.
The prevalence of anti-cyclonic curvature over the area on a
high ozone day should be considered exemplary of the classical theory
to the extent that the area is under the influence of high pressure.
It must be reemphasized, however, that stations in the western part
of the state such as El Paso would not have such conditions in view
of the predominace of the spring and summer "heat low" in that area.
Finally, the daily weather maps were carefully checked to determine
if a "peak" synoptic condition was consistently producing the high
Og readings.
Time of Day
The time of day for the 10 highest values were tabulated
according to their occurrence during one of four periods in the day
in local standard time as follows:
(0000 - 0500)
(0600 - 1100)
(1200 - 1700)
(1800 - 2300)
Season
For the seasonal analysis of the 10 highest ozone days, each
case was tabulated as to its occurrence in one of the following
four seasons: (January-March), (April-June), July-September),
(October-December).
In comparing Texas with other locations it should be noted
that the solar altitude will be greater year round in Texas than,
3-53
-------
for instance, the northern East Coast corridor. If other meteor-
ological varibles are held equal, which they are not, the ozone
season would be longer in Texas than most of the locations which
have defined classical theory. Therefore, occurrences of high ozone
concentrations in the afternoon of off-season days may be considered
valid if the other classical meteorological conditions occur.
Wind Speed
The frequency distribution of wind" speed by classes (0-3 mph,
4-7 mph, 3-17 mph, 13-18 mph, 19-24 mph, and ^25 mph) for all high
values in each set was compared to the climatological distribution,
the latter as defined by data obtained from the Climatic Atlas of the
United States, June 1968. Each high value wind speed was defined by
the average wind speed for the two observations surrounding a vio-
lation hour. On those days when more than one high value occurred,
only one wind speed value was extracted; that being the average wind
speed over all hours immediately before or after any of the high
values on that day. Next, average wind speed for all high values
in a set was calculated and compared to the climatological norm as
given by the local Climatological Data Summaries.
Care must be taken in interpreting this tabulation since winds
are generally lighter in summer yet generally stronger in the after-
noon than the climatological norm with the latter effect the more
pronounced.
Wind Direction
A tabulation was made of the range of the wind direction over
the six hour period up to and including the high value hour. Each
3-54
-------
range was then classified according to the following classes which
define wind variability:
0° - 45° Steady
45° - 90° Average Variability
> 90° Extreme Variability
Although these categories are somewhat arbitrary a precise definition
of wind variability in the literature does not exist. This methodology
is loosely based on work by Singer at Brookhaven National Laboratory in
1967. It is recognized that the climatology of New York is significantly
different from that of Texas; however, examination of wind rose and
resultant wind data for four Texas weather stations (Dallas, Houston,
San Antonio, and El Paso) revealed that the wind is apt to be steadier
in Texas than in New York (because the prevailing directions and
resultant speeds are much more pronounced in the former case), so
that if anything the classifications used should be considered a bit
conservative (i.e., perhaps ranges smaller than 90° should be classified
as "extremely variable").
Wind Distribution by Pasquill Stability Class
The final step in the analysis was to classify the combination
of wind direction, speed and stability which occurred at each of
three afternoon hours (1200, 1500, and 1800) on each high value day,
and compare the results for each of the three sets of frequency dis-
tributions (highest values, second highest values, and 10 highest
values) for each station to the appropriate climatological distributions
(as derived from the "STAR" program).
3-55
-------
For the "highest" and "second highest" sets, results are
evaluated in terms of the number of combinations considered which
fell in normally very infrequent categories (0.0 to 0.2 percent
frequency within the given stability class), in rather infrequent
categories (0.3 to 0.5 percent frequency within the stability class),
in "average" frequency categories (0.6 to 1.4 percent frequency within
the stability class), in rather frequent categories (1.5 to 5.0 percent),
and in extremely frequent categories (>5.0 percent). Combinations
or "parts" of combinations (such as a certain wind direction) which
persistently reoccurred for a monitoring site are additionally noted.
An intra-stability analysis was emphasized; that is, a given combination
was classified based on its normal frequency of occurrence compared
to the remaining combinations within the class (i.e., the climato-
logical frequencies of occurrence for all wind combinations had up
to 100 percent for each class). This was felt to be necessary
because of the obvious bias toward unstable Pasquill classes result-
ing from the evaluation of just the afternoon hours, which in turn
would bias a comparison of results with normal frequency of occurrence
relative to all observations. However, it is recognized that there
may be an unusually high percentage of A or B stability cases, even
considering the time of day.
For the data sets representing the 10 highest values, the
complete frequency distributions were collected, for each stability
and for all stabilities combined. The percentages for each stability
class are relative to the total number of cases only within each class.
3-56
-------
Again, unusually high frequencies of A or B stability, even con-
sidering the time of day, are noted where applicable.
In most cases the comparisons were made with the appropriate
seasonal "STAR" frequency distributions, which were usually for
the period June-August. However, individual exceptions to the
latter did occur for high and second high days at various stations,
as well as for the groups of ten highest days at Clute (for which
September-November data were used) and Corpus Christi (for which
annual data were used, since high values were scattered throughout
the year).
The comparison with "STAR" data was only accomplished for
eight of the eleven stations; "STAR" data for Austin and Port
Arthur (Nederland and West Orange stations) were not available at
the time of this report.
The analyses were based on information from several data
sources, which in some cases were not available for the particular
city of interest. In these situations, data from the closest or
most representative climatic station relative to the monitoring
site was used. Thus, the basic meteorological data for the Clute,
Aldine, and Texas City sites were taken from the LCD's (Local
Climatological Data Summaries) for Houston, Texas, while LCD's for
Port Arthur, Texas were used for the Nederland and West Orange
analyses. Concerning the Texas City analysis, LCD's for Galveston,
Texas (the closest major city) were sought out originally; however,
only monthly summaries were available, so it was decided to use
the Houston climatic data instead.
3-57
-------
3.3.3 Analysis of the High Ozone Days
This section summarizes the analysis of the meteorological
parameters defined in the previous section and also discusses the
expected frequency of occurrence of the high ozone days.
3.3.3.1 Meteorological Characteristics
Based on the ten highest ozone days for each year, 1971-1974, at
the eleven monitoring stations, the following conclusions have been
drawn regarding characterization of the meteorological parameters
associated with those days:
Solar Insolation: high solar radiation; high solar altitude;
and less than 50% cloud cover.
Temperature: high ambient temperature.
Surface Pressure: weak pressure gradient; not necessarily
associated with anti-cyclonic curvature.
Time of Day: late morning to early evening.
Season: late spring through autumn; especially summer
months.
Wind Speed: light to moderate.
Wind Direction Range (Variability): not significantly
associated.
The above meteorological parameters were considered characteristic
for a given station year if a good majority (75% or more) of the high
cases exhibited these conditions with respect to a good majority (75%
or more) of the parameters analyzed. Seventeen of the 24 station
years satisfied this condition with respect to all of the 10 highest
3-58
-------
ozone measurements. The following station years were uncharacteristic
as indicated:
Austin-1973: quite a few unusual cases were evident and
only season of year was considered characteristic.
Clute-1974 and Nederland-1972 and 1973: were uncharacteristic
with respect to season.
Corpus Christi-1971: was unusual with respect to wind
speed.
Corpus Christi-1973: was unusual with respect to wind
speed and season of year.
El Paso-1971: was characterized by relatively low
temperature.
3.3.3.2 Frequency of High Values
Based on the eight stations for which comparisons with "STAR"
data were made, it was concluded that the high ozone days are not
generally characterized by extremely unusual meteorological conditions
(i.e., those with less than 0.2% climatological frequency of occurrence
within the stability class). The highest percentages of high ozone
measurements associated with such extremely unusual meteorology
occurred in Dallas (22%), Houston (14%), and Texas City (13%).
The most frequently reoccurring highly unusual combination was
A stability with calm winds which accounted for about one-third of
all the extremely unusual cases. Nearly half of the highly unusual
combinations occurred with A stability and approximately 20% were
combinations of C stability with uncharacteristically low wind speeds.
3-59
-------
Thus it can be seen that the highly unusual conditions character-
ized only a small minority of the weather observations analyzed and
most of the unusual conditions were characterized by unstable atmos-
phere with low wind speeds.
3.3.4 Analysis of the Seven High and Second Highs
3.3.4.1 Austin - 5/7/73
Solar Insolation: high solar altitude (77°); high solar
radiation (index 4); no cloud cover.
Temperature: high (80°)
Sea Level Pressure: moderate cyclonic curvature
Wind Speed: high (13 mph)
Wind Direction Range: steady
Time of Day: afternoon (1300 and 1400)
Season of Year: mid-spring
3.3.4.2 Corpus Christi - 8/17/71
Solar Insolation: high solar altitude (75°); high solar
radiation (index 4); little cloud cover (4/10).
Temperature: high (91°)
Sea Level Pressure: weak anti-cyclonic ridge
Wind Speed: moderate (9 mph)
Wind Direction Range: extremely variable
Time of Day: afternoon (1500 and 1600)
Season of Year: summer
Comparison with "STAR" Data: rather frequent (average
of nearest three hours - 3.1%).
3-60
-------
3.3.4.3 Dallas - 7/22/74 and 7/23/74
Solar Insolation: high solar altitude (77°, both); high
solar radiation (index 4, both); little cloud cover (3/10 and 0/10).
Temperature: high (103° and 105°)
Sea Level Pressure: very weak gradient (both)
Wind Speed: light to moderate (6 and 8 mph)
Wind Direction Range: extremely variable (both)
Time of Day: afternoon (1400 and 1500)
Season of Year: summer
Comparison with "STAR" Data: rather frequent and extremely
frequent (average of nearest three hours - 4.8% and 6.0%).
3.3.4.4 El Paso - 7/11/74
Solar Insolation: high solar altitude (74°); high solar
radiation (index 4); little cloud cover (4/10).
Temperature: high (86°)
Sea Level Pressure: weak cyclonic gradient
Wind Speed: light to moderate (7 mph)
Wind Direction Range: average variability
Time of Day: afternoon (1200 and 1300)
Season of Year: summer
Comparison with "STAR" Data: rather frequent (average
of three nearest hours - 2.5%).
3.3.4.5 Houston - 7/15/74
Solar Insolation: high solar altitude (82°); moderately
high solar radiation (index 3); cloud cover significant (6/10).
3-61
-------
Temperature: high (89°)
Sea Level Pressure: very weak gradient
Wind Speed: moderate (9 mph)
Wind Direction Range: extremely variable
Time of Day: afternoon (1300 and 1400)
Season of Year: summer
Comparison with "STAR" Data: rather frequent (average of
nearest three hours - 3.2%).
3.3.4.6 Nederland - 3/7/73
Solar Insolation: moderately high solar altitude (55°);
moderately high solar radiation (index 3); cloud cover signicant (7/10).
Temperature: average (66°)
Sea Level Pressure: very weak gradient
Wind Speed: very light (3 mph)
Wind Direction Range: calm to steady
Time of Day: late morning to afternoon (1100 and 1200)
Season of Year: late winter
3.3.4.7 San Antonio - 8/29/71
Solar Insolation; high solar altitude (69°); high solar
radiation (index 4); no cloud cover through noon, becoming partly
cloudy (more than 5/10) thereafter.
Temperature: high (89°)
Sea Level Pressure: weak to moderate anti-cyclonic curvature
Wind Speed: moderate (9 mph)
Wind Direction Range: extremely variable
3-62
-------
Time of Day: afternoon (1600 and 1700)
Season of Year: simmer
Comparison with "STAR" Data: average frequency (average
of nearest three hours - 0.9%).
3.3.5 Summary of Meteorological Analysis
As discussed in section 3.3.3.1, the data analysis showed that
the occurrence of high ambient ozone is associated with a character-
istic meteorology. In section 3.3.3.2, it was further pointed out
that the meteorology associated with high ozone measurements does
not represent an unusual event and should be expected to reoccur.
Section 3.3.4 presented an analysis of the data associated with
the basis of the control plan. By comparing this section with the
general findings of section 3.3.3.1, it can be seen that these key
data points fit the expected meteorology extremely well. In fact,
all six parameters fit the expected values in Corpus Christi, Dallas,
El Paso, Houston, and San Antonio. The Nederland data with regard
to season of year and temperature do not meet the expected criteria.
Regarding the Austin case, only half the parameters fit. Although
the May 7, 1973 Austin event included unusual wind speed and pressure
curvature, the preceeding four days all exceeded the ozone standard
and all exhibited an anti-cyclonic pressure influence with more
moderate wind speeds.
It was also shown in section 3.3.4 that the meteorology associated
with the key ozone data occurs rather frequently in most cases. Thus,
it is seen that the key ozone data do not represent an unusual meteo-
rological event and the meteorology is expected to reoccur.
3-63
-------
Finally, in answering the second question posed in section 3.0,
it can be said that the ozone values, from a meteorological stand-
point, are representative and are not anomalies.
3-64
-------
Section 3 References
3-1 The letter dated August 18, 1975 with its enclosures
from Mr. Roger R. Wall is, Director, Air Quality
Evaluation Division, Texas Air Control Board to
Mr. Myron 0. Knudson, Director, Surveillance and
Analysis Division, Environmental Protection Agency,
Region VI.
3-2 The memorandum dated August 8, 1975 with its enclosures
from Mr. F. J. Burmann, Chief, Environmental
Monitoring Branch, Environmental Monitoring and
Support Laboratory, Environmental Protection Agency,
North Carolina to Mr. Myron 0. Knudson.
3-3 R. L. Larsen, A Mathematical Model for Relating Air Quality
Measurements to Air Quality Standards, Publ AP-89, U.S. Environ-
mental Protection Agency, Research Triangle Park, N.C., 1971.
3-4 R. I. Larsen, "An air quality data analysis system for inter-
relating effects standards, and needed source reductions,"
J. Air Poll. Control Assoc. 23:933 (1973).
3-5 R. I. Larsen, "An air quality data analysis system for inter-
relating effects standards, and needed source reductions,"
J. Air Poll. Control Assoc. 24:551 (1974).
3-6 R. I. Larsen, "A new mathematical model for air pollutant
concentration, averaging time, and frequency," J. Air Poll.
Assoc. 19:24 (1969).
3-65
-------
3-7 H. D. Kahn. "Note on the distribution of air pollutants,"
J. Air Poll. Control Assoc. 23:973 (1973).
3-8 N. D. Singpurwalla, "Extreme values from a lognormal law
with applications to air pollution problems," Technometrics
14:703 (1972).
3-9 T. C. Curran and N. H. Frank, "Assessing the validity of the
Lognormal Model When Predicting Maximum Air Pollution Con-
centrations," U.S. Environmental Protection Agency, Research
Triangle Park, N.C., Paper No. 75-51.3 (1975).
3-10 H. E. Neustadter and S. M. Sidik, "On evaluating compliance
with air pollution levels 'not to be exceeded more than once
per year,1" J. Air Poll. Control Assoc. 24:559 (1974).
3-11 N. R. Patel, "Comment on a new mathematical model of air
pollution concentration." J. Air Poll. Control Assoc. 23:292
(1973).
3-12 R. I. Larsen, "Response to comment on a new mathematical model
of air pollution concentration," J. Air Poll. Control Assoc.
23:292 (1973).
3-13 GEOMET Draft Report No. EF-498, October 1, 1975, "An
Analysis of Meteorological Conditions Associated with
The Days In Which The Highest Ozone Values Have Been
Reported At Eleven Monitoring Stations In Texas", EPA
Contract No. 68-02-1442, Task No. 8.
3-66
-------
4.0 HYDROCARBON/PHOTOCHEMICAL OXIDANT MODEL
4.1 Available Models
In order to calculate the necessary emission reductions to
attain ambient air quality standards, a model must be developed
which defines the relationship between emissions and air quality
and which adequately accounts for other parameters affecting
air quality. Several procedures are available for estimating
oxidant air quality based on changes in emissions and atmospheric
dispersion of emissions. The complexity and sophistication of
these procedures range from the few simple calculations of the
linear rollback model that may be made manually, to thousands
of calculations that require a computer for the complex models
involving photochemistry and meteorology. The following is a
discussion of atmospheric simulation models which may be used
to predict ambient oxidant (03) concentrations based on emissions.
Six models are discussed, in order of increasing complexity.
4.1.1. Linear Rollback
Linear rollback assumes that changes in ambient oxidant con-
centration levels are directly proportional to changes in total
emissions. If the existing spatial and temporal distributions
of emissions remain constant, this is a reasonable assumption
(Ref. 4-1). The effects of meteorology on ambient concentrations
are included implicitly in the technique by the dependence on
ambient concentrations. For long time periods or for worst
case values, this lack of explicit dependence on meteorology is
-------
not critical (Ref. 4-2). Advantages of the linear rollback model
are that it is simple to use, it does not require extensive input
data, and it's theoretical basis is readily understood.
4.1.2. Nonlinear Rollback
An alternate approach to the application of rollback to
photochemical oxidants is a consideration that oxidant concen-
trations are related to ambient concentrations of both hydrocarbons
and oxides of nitrogen in a nonlinear fashion. Nonlinear or
modified rollback attempts to account for this effect through an
empirically derived relationship between maximum daily 1-hour
average oxidant concentrations and 6 to 9 a.m. average concentrations
of non-methane hydrocarbons. Hydrocarbon concentrations are then
assumed to depend linearly on emissions, so that a relationship
between changes in hydrocarbon emissions and oxidant concentrations
is achieved. One form of this relationship is the Appendix J
curve of the EPA regulations (Ref. 4-3), which is based on the
maximum observed relationship between hydrocarbons and oxidant
concentrations for selected cities in the United States.
Such curves have also been developed which are specific to a
city. These specific city relationships are an improvement for
the individual locations to which they are applied since other
curves in the literature are composites of data and are based on
data taken before the EPA reference method for oxidants was
defined.
4-2
-------
It may well be that the Appendix J general relationship may
apply in some Texas cities, but such data has not been compiled to
date, and could differ in each urban area. Also, oxidant levels
in some Texas cities are above the limits of the Appendix J curve.
Extrapolation to these high oxidant levels would require reductions
in hydrocarbons of greater than 100 percent. Clearly, Appendix J
is not appropriate for these Texas cities.
4.1.3. Gifford - Hanna Photochemical Model
The Gifford-Hanna model is basically a "box model" in which
average pollutant concentrations within the box are taken to be
proportional to the ratio of the average area source strength to
the wind speed. For purposes of chemical kinetics calculations,
the pollutant concentrations are assumed to be uniform within the
volume defined by the area of the region and the mixing depth.
This technique includes the assumption that predicted 03 concen-
trations are directly proportional to the predicted concentration
ratios of N0£ to NO, but does not allow for dependence on hydro-
carbons.
Although this model may be useful in predicting trends in 03
concentrations, it has to date predicted maximum 03 values which
compare poorly with actual measured values.
4.1.4. Reactive Environmental Simulation Model (REM)
The REM and DIFKIN (described in the following section)
models are trajectory models, based on a Lagrangian or moving
coordinate system. Both chemical and meteorological inputs are
complex and extensive with a resultant large computer requirement.
4-3
-------
The REM model has spatial and temporal detail far in excess
of any of the models considered thus far. However, to date, REM
has only been tested in Los Angeles. The results of that test
show that 03 prediction is within a factor of 2, in 75% of the
comparisons.
4.1.5. Diffusion Kinetics Model (PIPKIN)
As in the case of REM, DIFKIN calculates the trajectory of an
air parcel across an emission grid network and determines time
dependent concentrations of reactive pollutants. Unlike REM,
DIFKIN solves for turbulent diffusion in the vertical direction.
Because DIFKIN ignores horizontal diffusion, the model is
probably more applicable to regions with only small emission
gradients between adjacent grid elements. This effect may cause
an error of about 40% under large gradient conditions (Ref. 4-1).
It should be noted that this may be an important factor when
considering emissions from large stationary sources and industrial
areas which are large sources of photochemicals precursors, such
as in the Gulf Coast area of Texas. Further, neglect of the
vertical wind shear, inherent in trajectory models, may introduce
errors of over prediction of 50% or greater (Ref. 4-5).
DIFKIN has been tested with reasonable success in Los Angeles,
San Francisco, and Denver. However, the model must still be con-
sidered in the development stage.
4.1.6. Systems Applications Inc (SAI)
This model uses a finite difference technique over a grid of
area sources to solve the classical equations of conservation
4-4
-------
of mass, which include local change, advection, diffusion, and
photochemical reaction of emissions.
Application of this model to a given location is made diffi-
cult by the fact that some aspects of the program are quite specific
to Los Angeles. A recent study (Ref. 4-5) concluded that in spite
of the fact that the SAI model is an excellent research tool, the
long running times and extensive data requirements make it difficult
to use the model for analysis of different oxidant strategies
without a considerable investment in time and manpower.
Although this model is one of the better photochemical
models currently available, the reliability of the model has not
been thoroughly examined (Ref. 4-2).
4.1.7. Conclusion
As in the previous oxidant plant, EPA has chosen to use the
linear rollback model to relate hydrocarbon emission to photo-
chemical oxidant air quality. The other simplified modeling
technique, nonlinear rollback using Appendix J curve in 40 CFR 51,
was not used in Texas since it could not be applied in all areas.
The more complex computer modeling techniques Gifford-Hanna,
REM, DIFKIN, and SAI were also not used since their reliability in
predicting oxidant levels has not yet been demonstrated. In
addition, these methods are very complicated and time-consuming,
and require vast amounts of input data that are not generally
available for Texas areas.
4-5
-------
4.2 Limitations of Linear Rollback Model
Although linear rollback is considered the best modeling
method for oxidants currently available due to its simplicity,
uniformity in application, and understandable basis, the model
does have some limitations in predicting maximum yearly oxidant
levels based on changes in hydrocarbon emissions. These limitations
as well as some limitations applicable to all present modeling
techniques are as follows:
1. The first limitation is that the effects of meteorology
on predicted ambient concentrations are not explicitly accounted
for in the linear model. This limitation, however, is more than
offset by the fact that the model uses worst case measured con-
centrations and therefore implicitly includes the effects of worst
case meteorology on ambient concentrations.
2. Another limitation of the linear model is that the model
does not account for the differences in photochemical reaction
processes that can vary with the specific amounts and types of
oxidant precursors in an area's atmosphere. Again, this effect is
minimized by the fact that yearly worst case ambient measurements
are used for each area in the model. A refinement to the model
application in Texas that further minimizes the effects of different
oxidant precursor breakdowns from area to area is that only
reactive hydrocarbon inventories compiled for each area are used
for rollback calculations.
3. Another limitation is that the linear rollback model
does not have a rigorously defined area of influence for the
4-6
-------
locations of oxidant precursor sources. The model assumes that
all hydrocarbon emissions in the chosen model area cause the
ambient oxidant value used for rollback purposes and that this
oxidant value is typical of what could be expected everywhere in
the modeling area. Thus if a modeling area is chosen that includes
both the major populated and industrialized urban counties, (in
which the only air quality measurements are usually taken), and
also includes the surrounding non-urban counties, (with much lower
emissions), the high oxidant values measured in the urban counties
are assumed to be possible in the surrounding non-urban counties
also. Similarly, the emissions from the non-urban areas are
assumed to be contributing to the high oxidant value measured in
the urban area.
The geographical extent of this urban/non-urban interaction
is not clearly defined, and the extent of the modeling area is a
judgement based on the emission air quality, meteorological, and
geographical data available. There is much evidence from other
areas than Texas to suggest that such short-range transport of
oxidants and/or their precursors do occur and that values of
oxidants as great as or higher than observed in the urban counties
can occur in the non-urban counties also (Ref 4-6). The oxidants
in these non-urban areas are due to both the transported oxidants
and precursors from the urban areas and also due to the emissions
in the immediate non-urban areas. Emission controls in both areas
are therefore appropriate to ensure oxidant reductions in both
4-7
-------
non-urban and urban counties. Depending on the wind directions,
the emissions in the surrounding non-urban areas can also contribute
to the measured oxidant problem in the urban areas and controls
can reduce the problems in both areas. Either by obtaining rural
oxidant data in the non-urban areas around the Texas urban areas
or by evaluating the potential non-urban oxidant problems and
transport (rural emissions, typical meteorology, and distance to
*
the metropolitan areas), the geographic limits of the model areas
can be judiciously chosen to minimize the effects of this model
limitation.
4. Another limitation to oxidant models is that the effects
of long-range transport of oxidants and/or oxidant precursors are
not included. EPA is continuing to evaluate the effects of long-
range oxidant transport on the levels observed in both urban and
rural locations, but at present the effects are not completely
known and have not been quantified. Present information indicates,
however, that during the days with meteorological conditions
characterized by stagnant high pressure systems, the same conditions
during which the highest oxidant values are observed, the oxidant
problem is primarily attributed to those mobile and stationary
source emissions generated with the local area itself (Ref. 4-6).
The effects of long-range transport would therefore be minimizied
in the proportional model, since the worst case, episode oxidant
values are generally used.
5. The last limitation of oxidant models to be discussed is
that there is no contribution of emissions of oxidants or their
4-8
-------
precursors from natural sources included. In general, EPA studies
to date indicate that manmade emissions are the predominant source
of the highest levels of oxidant observed and not naturally occurring
sources of oxidants or their precursors (Ref. 4-6). Therefore the
assumption that the naturally occurring sources are zero is con-
sidered to have a small effect on the overall oxidant/hydrocarbon
modeling in a particular area.
Although the proportional rollback model does have limitations
for oxidant projections as discussed above, these limitations can
be minimized and are not so restrictive to make its use as a basis
for developing oxidant control strategies inappropriate.
4.3 Texas Modeling Areas
One of the steps in using the linear rollback model is to
define the extent of modeling area to be used. In the original
plan, EPA chose as the modeling area the entire Air Quality
Control Regions (AQCRs) for the areas of concern. The seven AQCRs
of interest in Texas (each classified in 40 CFR 52.2271 as being
Priority I regions for photochemical oxidants) are shown on
Figure 4-1. Subsequent to the original plan development it
became clear, however, that in several of the regions there was a
low probability of significant interaction of the high oxidant
levels and emissions in the urban areas with some of the outlying
non-urban areas within the same AQCR. Extreme examples of such
situations are the interactions between El Paso and Brewster
4-9
-------
Figure 4-1
PRIORITY I REGIONS FOR OXIDANTS
4-10
-------
counties in AQCR 153 (separated by almost 200 miles), and between
Bexar (San Antonio) and Val Verde counties in AQCR 217 (separated
by almost 150 miles).
In order to improve the modeling technique in the revised
control strategy development, the modeling areas were redefined
and only those counties that had measured oxidant problems, or
that were judged to be contributing to that measured oxidant
problem, or that were judged as having a potential oxidant problem
resulting from short-range transport were included. Figures 4-2
through 4-8 show the seven AQCRs and the model areas chosen for
the revised strategy development. In general, the model areas
chosen by EPA agree with the areas assumed by the TACB in reference
4-7. There were some small exceptions, however, in the Austin,
I
San Antonio, Corpus Christi, and Beaumont areas. In these areas,
EPA added one or two adjacent counties to the counties chosen by
the TACB. Added in each case were the remaining counties in the
Standard Metropolitan Statistical Area (SMSA) for the particular
area of concern. These were determined to have the greatest
potential for short-range transport and interaction with oxidants
and precursor emissions from the adjacent counties. SMSAs, also
exhibit the highest concentrations of population and industry and
are generally the places with existing air quality problems.
Two sub-regions not included in the chosen modeling areas are
the Waco area in AQCR 212 and the Victoria area in AQCR 214.
Based on their hydrocarbon inventories these sub-regions appeared
4-11
-------
to
O)
rO
O)
5-
-o
O
OJ
CM
a:
o
CM
cu
cn
4-12
-------
recommended by T.A.C.B.
Model Area
Area added by EPA
Model Area
AQCR-106 (Beaumont-Port Arthur Area)
Figure 4-3
4-13
-------
Area recommended by
Model Area
Area added by EPA
_
t ^^A^^ ^^^ ^^^ ^^M» «^V
3 Kmos'
Model Area
AQCR-214 (Corpus Christi Area)
Figure 4-4
4-14
-------
Model Area
Model Area
AQCR-215 (Dallas-Fort Worth Area)
Figure 4-5
4-15
-------
S-
«=C
-------
.-.Model Area
Model Area
AQCR-216 (Houston-Galveston Area)
Figure 4-7
4-17
-------
re
cu
s-
c
o
c:
ca
t/i
o;
C_5
cr
a;
s-
ni
'i1
ii!
4-18
-------
to have a high probability of having oxidant problems. Due to the
facts that there was no ambient air quality data for these two
sub-regions and that there was not a clear interaction of these
sub-regions with the identified problem areas in the AQCRs (Austin
and Corpus Christi), they were excluded from the modeling area.
Air quality monitoring data taken in the future in these two sub-
regions as well as other Texas urban areas may identify more
problems and necessitate further changes to the oxidant control
strategy.
4.4 Emission Reduction Required
Using the proportional rollback model and the yearly second
highest ozone levels observed in the regions, the percent hydro-
carbon reductions required to meet the standard of 0.08 ppm are
presented in Table 4-1. The percent reductions are calculated as
follows:
% reduction = ° " s x 100%
where 0 = observed second high yearly ozone value in the baseline
year
S = the oxidant standard of 0.08 ppm
In general, the baseline year chosen is the year between 1971
and 1974 that the largest second high yearly ozone value was
observed. Only in the Houston area is there an exception to this
general situation in that the baseline year chosen is 1974 while
the largest second high yearly ozone value was observed in 1972.
The baseline years were all chosen to be the worst combination of
4-19
-------
Table 4-1 - Hydrocarbon Reduction Requirements
Baseline 2nd High
Area Year Values % Reduction
(ppm)
Austin (AQCR 212) 1973 .160 50
Beaumont (AQCR 106) 1973 .325 75
Corpus Christi (AQCR 214) 1971 .184 57
Dallas/Ft. Worth (AQCR 215) 1974 .187 57
El Paso (AQCR 153) 1974 .130 38
Houston (AQCR 216) 1974 .234 66
San Antonio (AQCR 217) 1971 .145 45
4-20
-------
both the baseline year air quality and the expected emission
reduction from the baseline year to 1977. For all cases except
Houston, the combination of air quality and emission reductions
yielded a baseline year the same as the year when the largest
second high ozone value was observed. For Houston the combination
of air quality and expected emission reductions gave 1974 as the
worst case baseline year even though the air quality value was
lower than other years. This effect is shown in the inventory and
reduction tables of reference 4-7 where the additional reduction
required for the 1974 baseline year is clearly the greatest of the
potential baseline years evaluated.
4-21
-------
Section 4 References
4-1 Systems Applications, Inc., Development of a Second Generation
of Photochemical Air Quality Simulation Models. Progress
Report Prepared for EPA, Contract No. 68-02-1237, 1974-
4-2 Robert M. Patterson, et. al., Photochemical Oxidant Modeling;
Volume I : Techniques Applicable to Highway System Evaluation.
Bedford, Mass.: GGA/Technology. July 1975. (EPA-450-3-75-069a,
October 1975.)
4-3 Appendix J to Title 40, Code of Federal Regulations. Part 51.
4-4 Office of Air Quality Planning and Standards, U.S Environmental
Protection Agency. Guideline for Air Quality Maintenance Planning
and Analysis, Volume 12; Applying Atmospheric Simulation Models to
Air Quality Maintenance Areas. EPA-450/4-74-013. Research
Triangle Park, N.C.; EPA September, 1974.
4-5 Robert M. Pattersort, et. al., Photochemical Oxidant Modeling;
Volume II: Detailed Technical Report. Bedford, Mass.: GCA/
Techology, April 1975. (EPA-450/3-75-0696, October 1975.)
4-6 Office of Air Quality Planning and Standards, U.S. Environmental
Protection Agency. Control of Photochemical Oxidants - Technical
Basis and Implications of Recent Findings (EPA-450/2-75-005).
Research Triangle Park, N.C.: EPA July 15, 1975.
4-7 Texas Air Control Board Report SP-1, Reactive Carbon Compound
Control Strategy Reexamination for the State of Texas,
March 13, 1975.
4-22
-------
5.0 DEVELOPMENT OF REACTIVE HYDROCARBON EMISSION INVENTORIES
FOR TEXAS
5.1 Introduction
In its August 1974 decision on the Texas Oxidant Control
Plan the U.S. Court of Appeals for the Fifth Circuit remanded to
EPA for reconsideration the reactive hydrocarbon emission inventory
for petroleum refineries in Texas and the reactivity factor
(0.12) used by EPA to estimate the photochemically reactive
portions of refinery emissions. In September 1974 EPA and the
Texas Air Control Board (TACB) entered into a joint study to
reevaluate the Texas reactive hydrocarbon emission inventory. It
was agreed that a restudy of the entire emissions data base and
not just the refinery reactivity factor, questioned by the court,
should be performed to ensure that the entire inventory for the
Texas Oxidant Plan is computed on a consistent basis.
The basic ground rules established for development of the
revised hydrocarbon emission inventory were as follows:
1. The entire inventory for stationary point sources, area
sources, and mobile sources for the baseline year would be redone
in the Dallas-Fort Worth, Austin, San Antonio, Houston-Galveston,
Beaumont-Port Arthur, Corpus Christi, and El Paso Areas.
2. Stationary point source emissions would be based on the
1972 and 1973 Emission Inventory Questionnaires submitted to the
TACB by industry.
3. Mobile source emissions would be computed based on the
latest vehicle data available (1973 and 1974 data).
5-1
-------
4. The inventory would include only photochemically reactive
hydrocarbons (defined by joint agreement).
5. Specific information would be developed on the extent of
industry controls in effect prior to implementation of TACB
Regulation V to ensure that correct credit for emission reductions
due to Regulation V would be made.
6. Projections of the expected emissions in 1977, 1980, and
1985 would be made, accounting for the effects of growth, TACB
Regulations V and VI, and the Federal Motor Vehicle Control Program
(FMVCP).
The TACB portion of the inventory reevaluation was conducted
from September 1974 until March 1975. Results of the TACB effort
are summarized in a preliminary report dated March 13, 1975
(Ref. 5-1). Sections 5.2 through 5.5 of this document describe
how the revised inventory was developed and are based on the TACB
report, supporting data supplied to EPA by TACB, and information
developed by EPA during the course of the reevaluation effort.
The resulting projected inventories for 1977, 1980, and 1985
presented in Section 5.5 assume only the emission controls in
existence now (TACB Reg. V [Control of Hydrocarbons from Existing
Sources] and Reg. VI [Control of Emissions from New or Modified
Sources] and the Federal Motor Vehicle Control Program) and do
not include the effects of any additional controls. Possible
additional controls are discussed in Section 6.0 and the proposed
additional controls are discussed in Section 7.0.
5-2
-------
5.2 Photochemical Reactivity
5.2.1. Basis
Reactivity as used in this control plan means the tendency
of hydrocarbons and organic compounds to photochemically react
with other atmospheric pollutants and contribute to the production
of photochemical oxidants. The use of a reactive hydrocarbon
inventory instead of a total or a nonmethane hydrocarbon inventory
is a refinement of the proportional rollback model which is
designed to make the predicted reactive hydrocarbon reductions
resulting from the applied controls more representative of their
actual effects on ambient air quality.
It should be noted, however, that the use of reactive
inventories is just a modeling tool to better predict the effects
on air quality and that the controls applied reduce all hydro-
carbon emissions and not just reactive emissions. Very few
hydrocarbons are truly nonreactive. There is evidence that
control of all nonmethane hydrocarbons might prevent oxidant
production many miles downwind from the source where the low
reactive (or slow reacting) compounds have had time to react
(Ref. 5-2). Based on the proportional modeling technique used
in this plan, however, the highly reactive (or fast reacting)
compounds are of greatest concern and are considered to have the
predominant effect on the ambient air quality measured in the
major Texas metropolitan areas under consideration. For this
reason the reactivity approach was retained for the revised
inventory development. Should ongoing studies show a different
5-3
-------
reduction model is more appropriate in the future, a recalcu-
lation may be necessary at a later date.
Hydrocarbons and other organics vary widely in reactivity
depending on chemical composition and reaction conditions.
Present scientific knowledge does not permit making precise
measurements of the reactive emissions in a particular metro-
politan area, but an approximation of the reactive emissions can
be made by a careful study of the sources of emissions in a
particular area. To do this, a list of individual compounds that
are nonreactive (or slowly reacting) are removed from the total
hydrocarbon emissions for each source in that area. This can be
accomplished by obtaining individual breakdowns by compound for
each source and eliminating the nonreactive compounds or by
developing reactivity factors based on a typical compound breakdown
for a particular type of source.
For this revised inventory development, the list of nonreactive
compounds (Table 5-1) published as part of the original EPA plan
for Texas in the Federal Register on November 6, 1973
(F.R. 30626) in Section 52.2292 was agreed upon and used by the
TACB and the EPA. This list is almost identical to the EPA list
given in Appendix B to 40 CFR Part 51 for guidance in developing
state implementation plans.
5-4
-------
Table 5-1
Non-Reactive Carbon Compounds
C -Cr n-paraffins
saturated halogenated hydrocarbons
perchloroethylene
benzene
acetylene
acetone
cyclohexanone
ethyl acetate
diethyl amine
isobutyl acetate
isopropyl alcohol
menthyl benzoate
2-nitropropane
phenyl acetate
triethylamine
5.2.2. Stationary Source Reactivity
The source breakdown by compound for industries could not
come directly from the TACB industrial emission inventory question-
naire (sample shown in Appendix B), since it lists total hydrocarbons
and some breakdown on combustion emissions, but no breakdown on
the amounts of specific hydrocarbon compounds emitted. For this
reason, a more detailed questionnaire specifically requesting
information on breakdown of nonreactive and reactive hydrocarbons
5-5
-------
emitted was sent to all Texas industries with major hydrocarbon
emissions (500 tons/year or greater). A sample questionnaire is
included in Appendix C. In addition to responses to the supple-
mental questionnaire, various companies furnished detailed analyses
for typical emissions from in-process streams and final products.
For petroleum refineries, average weight percentage values
of reactive emissions (reactivity factors) from 23 in-process and
product streams (Table 5-2) were developed based on the information
available to the TACB (see Appendix D for additional breakdown of
refinery emissions).
Table 5-2
Weight Percentage Factors for Refinery
In-process and Product Streams
Wt. Percent
Stream Reactive
Finished Gasoline 59
Crude Oil 40
Kerosene 67
Jet Fuel 67
Cat Cracker (CC)
Lt. Gasoline 93
Hydroformate 80
Alky late
undebutanized 65
debutanized 85
depentanized 100
Platformate 77
Reformer Feed 85
Naptha Base Stock 68
5-6
-------
Heavy C.C. Naphtha
undebutarn" zed 60
debutanized 80
Raffinate 70
BTX Feed 80
Distillate Fuel
n Diesel 70
Heating Oil 70
Crude Heavy Slop 67
V.MEP Naphtha 90
Cat. cycle stock
coker
decant oil 90
cat. cracker feed
Lt. cycle stock 80
Heavy Diesel
Bunker C
No. 6 Fuel Oil 60
Based on the total hydrocarbon information obtained from the
regular TACB industry questionnaire, the refinery factors were
applied to obtain the reactive emission from each process in the
refinery. For chemical plants, a similar process was applied
using the list of nonreactive compound with the data supplied in
both the regular and supplemental questionnaires on a process-by-
process basis to determine the reactive portions of the emissions.
For the smaller industrial sources with emissions less than 500
tons/year, either reactivity factors developed from similar large
processes were used to estimate the reactive portions or the list
of nonreactive compounds in Table 5-1 was applied to specific
process emissions to eliminate nonreactive compounds from the
inventory.
5-7
-------
The resultant industry-wide inventories for refineries and
chemical plants in Houston using this process turned out to be
61% and 57% reactive respectively as compared to an assumption of
12% reactive for refineries and 60% reactive for chemical plants
used by the EPA in the previous plan. Taken alone, this relatively
large increase in the petroleum refinery reactivity and small
decrease in the chemical industry reactivity would tend to add
more reactive emissions to industry; therefore, more reactive
reductions would be obtained from industry due to Texas Regu-
lation V and less reductions needed from other sources such as
transportation. This did not turn out to be the case, since the
added effects of little improvement in air quality and, as will
be discussed in Section 5.3 through 5.7, the reductions that were
credited to Regulation V and FMVCP were not nearly so large as
estimated in the previous plan.
5.2.3. Mobile Source Reactivity
Reactivity factors were used to establish the reactive
inventory for all mobile hydrocarbon sources. The factors used
are based on a typical breakdown by compound from the source of
emissions and application of the definition of nonreactive com-
pounds give in Table 5-1 to establish the reactivity factor for
the source. In most cases the reactivity factors are about the
same as those used by the EPA and the TACB in the earlier plans.
The reactivity factors used are given in Table 5-3.
5-8
-------
Table 5-3 Mobile Source Reactivity Factors
Mobi1e Source Type Factor
Gasoline Vehicle Exhaust 0.74
Gasoline Vehicle Evaporative
and Crankcase 0.60
Diesel Vehicle Exhaust 0.90
Aircraft 0.90
Other Transportation 0.80
Vehicle reactivity data and breakdowns of emissions by compound
that were taken from data presented in Reference 5-3, 5-4, and 5-5.
Specific breakdowns for the last three categories were not readily
available but the available information seems to support the relatively
high factors chosen. Due to the small contribution from these three
emission categories, no further adjustments were made.
5.2.4. Oth er Sources
For trade paints, all water-based paints were considered
nonreactive and all oil-based paints were considered to be reactive.
Reactive gasoline marketing emissions and ship/barge/tank car refinery
loading emissions were calculated using the reactivity factors
developed from petroleum refinery information and are listed in
Table 5-2.
5.3 Stationary Source Inventory Development
5.3.1. Industiral Point Source
5.3.1.1. Base Year 1972 Company Reported, Adjusted, and
Reactive Inventories.
5-9
-------
A detailed evaluation of the emission inventories for industrial
point sources was made by ishe TACB on a source-by-source basis. The
1972 hydrocarbon inventory, which is based on information from
questionnaires sent to individual industries, was used as the basic
source document of emissions from existing industrial sources.
Although the 1972 inventory is the source document, the 1973 industry
questionnaires were used extensively by TACB in updating 1972 emissions
known to exist in 1972 but not reported until 1973.
A blank questionnaire is included in Appendix B to illustrate
the kinds of information requested by the TACB. The basic categories
cf hydrocarbon data obtained from the questionnaires and evaluated
by the TACB are as follows:
1. Storage
a. Type of Storage
(1) Fixed Roof
(2) Floating Roof
b. Storage Tank Capacity
(1) < 1,000 gallons
(2) 1,000 - 50,000 gallons
(3) > 50,000 gallons
c. Vapor Pressure
(1) < 1.5 psia
(2) 1.5 - 11 psia
(3) 11+ psia
5-10
-------
2. Loading and Unloading Facilities
a. Ships and Barges
b. Cars and Trucks
3. Water Separators
4. Process Vents Not Flared
5. Fugitive and Miscellaneous Losses
Tank emissions comprise a major source of hydrocarbons emitted
from Texas Gulf Coast petroleum processing industries. Individual
product emissions were reported in the questionnaires on a tank-by-
tank basis for storage and loading operations. TACB made detailed
checks of company-reported emissions with values calculated by TACB
using emission factors from EPA document AP-42 (Ref. 5-6). Factors
from AP-42 were used with an adjustment for vapor pressure at typical
Texas storage temperatures. The TACB calculated values were used
rather than the company reported values when significant differences
were found. TACB also adjusted the company-reported values by
eliminating methane and other combustion emissions since these
emissions are all considered to be non-reactive. The process emissions
from point sources were as reported by the companies except when
refineries failed to report emissions from processes known to TACB.
In such cases the TACB used an appropriate emission factor from AP-42
and calculated the emissions based on the known operating rate for
each process unit.
Emissions for ship and barge loading operations were also
calculated by the TACB based on AP-42 emission factors, but the
5-11
-------
emissions from these sources have been recalculated by EPA based
on more current emission data. The details of the ship and barge
emission calculations are presented in Section 5.3.2.
Since the control strategy is based on reactive hydrocarbons,
the TACB took the adjusted inventory for each company and eliminated
the non-reactive compounds by the methods previously discussed in
Section 5.2.2.
5.3.1.2 Projections of Reactive Emissions to 1973, 1974,
1975, 1977, 1980, and 1985
Projections of emissions to years 1973, 1974, 1975, 1977,
1980, and 1985 were made by the TACB. These projections changed
the 1972 base year point source emissions by including the effects
of industrial growth and stationary source emission controls
required by Texas Regulation V.
The TACB used EPA estimates for SMSA Indices of Productions
for Selected Industries (Ref. 5-7) to project industrial growth
for stationary point sources. An overall index was calculated
from the actual industry indices and applied to the 1972 base
year emission inventory total to project emissions in 1977, 1980,
and 1985 for each Texas area of concern. For years between 1972
and 1977, actual growth of point source emissions was known to
the TACB from construction and operating permits and these emission
growth values were used to project the 1973, 1974, and 1975
emissions.
TACB Regulation V (Control of Air Pollution from Volatile
Carbon Compounds) became effective in 1972; therefore, after that
date, substantial hydrocarbon emission reductions from industrial
5-12
-------
stationary sources were assumed to occur within the counties
covered. Growth after 1972 is controlled by TACB Regulation "VI
(Control of Air Pollution by Permits for New Construction or
Modification) which requires best available control technology,
and which, in general, is more restrictive than the requirements
for existing facilities. Thus, growth in emissions after 1972 is
at a lower rate than the normal production growth rate.
Regulation V, which applies to existing point sources in
certain counties, has rules which reduce emissions from these
point source categories. Regulation V requires compliance with
the applicable rules by certain dates which were determined with
the required attainment date in mind. The compliance schedules
of the larger emitters were reviewed by the TACB and used to
determine the date the resulting reductions occurred.
The expected percentage reduction in emissions assumed by
the TACB by applying controls required by Regulation V rules for
each source category are as follows:
Category % Reduction
Tank
1-50,000 gal. 96
> 50,000 gal. 97
Loading Facilities 85
Water Separators 98
Process Vents Not Flared 98
5-13
-------
The above reduction factors were applied to the existing point
source emissions for base year 1972 based on the individual company
compliance schedules in order to project the reductions due to
Regulation V in years 1973 through 1985. It should be noted that in
reviewing the point source inventories for petroleum refineries, the
TACB found that many companies had already installed controls on
large petroleum storage tanks by 1972, that the effects of these
controls had already been included in the inventories, and that much
of the reductions predicted by the TACB in their previous plan due
to Regulation V (90% reduction assumed by the TACB in 1973 plan)
would not be obtained due to previous control of the tanks. Due
primarily to this effect, the reductions resulting from Regulation V
decreased from 90% in the 1973 plan to about 56% based on the current
information in the Houston area. Similar changes in the effectiveness
of Regulation V were observed in all areas.
Appendix E contains the TACB worksheets on a county and company
basis for the point source inventory. The worksheets include the
1972 base year company-reported emissions, 1972 TACB adjusted emissions,
1972 TACB estimated reactive emissions, and 1973-1985 projected
emissions with the effects of Regulation V included.
5.3.2 Emissions from Ship and Barge Loading Operations
5.3.2.1 Background
During the loading of petroleum products in tank ships and
barges at terminals in the Houston, Beaumont, and Corpus Christi
areas, significant amounts of hydrocarbons are emitted to the
5-14
-------
atmosphere through the tank vessel vents. Gasoline vapors are the
principal reactive hydrocarbon emissions from such loadings due to
the high volatility of gasoline and the large quantities of gasoline
produced and shipped by the Texas Gulf Coast petroleum processing
industries. Gasoline shipments in the Houston-Galveston area on
ships and barges are the greatest of the three areas, and they
amount to about 100 million barrels per year. The Beaumont area has
the second largest quantity of gasoline shipments with about 60
million barrels/year, while the Corpus Christi area gasoline loadings
amount to about 20 million barrels/year.
In order to calculate the amount of emissions from ship and
barge loadings in Texas, an emission factor O'.e., the number of
pounds of vapor emitted per volume of product loaded) typical of
gasoline loading conditions found in Texas is necessary. By having
such a factor and by knowing the volume of gasoline loaded, expected
emissions during gasoline loading can be estimated. In addition to
the gasoline emissions, emission factors for other products can be
established by adjusting for the vapor pressure and molecular weight
of the product loaded. Annual product loading volumes are not
difficult to establish and are published in reference 5-8. In
addition, industries have submitted to both the TACB and the EPA
total product volumes by product and by vapor pressure that have
been loaded at their terminals.
The emission factor for each loading will not be constant,
however, but will depend on many variables including the concen-
tration of hydrocarbon vapor in the tank before loading, the true
5-15
-------
vapor pressure of the product being loaded, loading rate, the
geometry of the tank, and how full the tank is loaded. The diffi-
culty in the ship and barge emissions calculations lies in obtaining
emission factors which have been weighted to be representative of
all the above variables.
In the initial control plan, TACB used an emission factor of
11 pounds/1000 gallons of gasoline loaded into ships and barges.
This factor was obtained from the second edition of AP-42
(Ref. 5-6) and was the factor given for working losses from fixed
roof storage tanks. There was no emission factor for ship and barge
emissions included in this edition of AP-42. This factor was based
on a 100% saturation for the fixed roof storage tanks.
In Supplement No. 1 to AP-42 (Ref. 5-9) emission factors for
ship and barge loading and unloading operations were included and
were used by the TACB to estimate emissions from ship and barges for
the revised inventory. The factor used by the TACB from AP-42 was
2.9 pounds/1000 gallons loaded. The TACB adjusted this factor up to
4.0 pounds/1000 gallons to account for expected higher loading
temperature and higher true vapor pressures in Texas. Although the
emission factor used for the revised inventory from Supplement No. 1
was specifically for ship and barge loadings, there were still
several uncertainties in the factor. First of all, the revised
factor was based on a very limited data base taken in the late
1950's for loadings on a few ships. In fact, the document containing
the data and correlation (Ref. 5-10) on which the factor was based
states that the marine loading correlations were not based on
5-16
-------
sufficient data and that more testing was needed. Secondly, the
factor did not take into account the amount of vapor existing in the
tank before loading. Thirdly, the data supplied to the EPA by
several Houston area oil refiners indicated that the value of 4.0
pounds (1000 gallons) may be high by at least a factor of two.
Based on these uncertainties, the EPA requested by letter (Ref. 5-11)
copies of emission data collected by industry in order to develop a
gasoline emission factor for ship and barge loadings.
To date the EPA has received twenty (20) responses from oil or
petrochemical companies containing ship and barge emission data. In
most cases the data seems to support the contentions that the original
factor used (11 lb/1000 gal.) was much too high and that the revised
factor used (4.0 lb/1000 gal.) may also be somewhat high. It is not
clear from the data, however, just what a truly typical factor
should be. What is typical for one company with one mode of loading
may not be true for another company with a somewhat different mode
of operation, and may even vary for the time of the year due to
seasonal vapor pressure changes in gasoline loaded. In order to
obtain a truly representative emission factor, much more data gathered
on a consistent basis would be necessary.
5.3.2.2. EPA Ship and Barge Emission Estimate
In order to develop a revised ship and barge inventory on the
best data available at this time, a new ship and barge emission
factor was used by the EPA. The EPA used the data supplied by one
oil company, Exxon, to be typical of all gasoline ship and barge
loadings in Texas. The Exxon information was used since the company's
5-17
-------
Baytown refinery conducts more ship and barge loadings than any
other refinery in the area and since the company submitted a sub-
stantial base of data. Exxon developed an emission factor for
gasoline of 2.1 pounds/1000 gallons loaded based on measured factor
of 1.8 lb/1000 gal. for 65 ship loadings of gasoline and 3.8 lb/1000
gal. for 20 barge loadings of gasoline at the Exxon terminal in
Baytown. Development of these factors along with data sheets was
reported to the EPA in references 5-12 and 5-13. Pertinent sections
of the Exxon information on development of their emission factor is
included in Appendix F.
Subsequent to submittal of the emission factor, Exxon in
reference 5-14 developed their own inventory of hydrocarbon emissions
from all volatile hydrocarbon products for all companies conducting
ship and barge loadings in the Houston-Galveston area. The Exxon
inventory yielded an average emission factor of 2.01 lb/1000 gallons
for emissions of all products loaded from petroleum refineries and
1.2 lb/1000 gallons for emissions of all products loaded from
chemical plants. Using these two emission factors and assuming that
they are typical of refinery and chemical plant loadings in Texas,
EPA developed a new emissions inventory for ship and barge loadings
in the Houston, Beaumont, and Corpus Christi areas.
For the new inventory, the volumes of products loaded reported
by the companies for 1974 loadings in Houston in response to the
reference 5-11 request by the EPA were used to the maximum extent
possible. Since no company volume loading data was submitted
directly to the EPA for the Corpus Christi or Beaumont areas, the
5-18
-------
volumes reported to the TACB in the 1972 emissions inventory question-
naire were used for refinery loadings in the Corpus Christi and
Beaumont areas. The volumes used were as follows:
Refi nery Loadings Chemical Loadings
Houston-Galveston 93.1 million BBL 23.8 million BBL
Beaumont-Port Arthur 64.0
Corpus Christi 20.6
These figures were checked against summaries on waterborne commerce
given in reference 5-8. The comparison showed good agreement. For
chemical loadings in Beaumont and Corpus Christi, the emissions
calculated by TACB were reduced by the ratio of the Exxon to the
TACB/AP-42 gasoline emission factors (2.1/4.0) since they were such
a small portion of the emission relative to refineries.
For reactivity factors, as discussed in Section 5.2.2. the
average point source factors for petroleum refineries of 0.61 and
for chemical plants of 0.57 were used. Since ship and barge emissions
are predominantly from gasoline loadings, the reactive emissions
must be controlled by the reactivity factor of 0.59 for gasoline
evaporative emissions. This factor was based on the industry break-
downs included in Appendix D.
Projections of growth for ship and barge emissions were based
on the petroleum refining and chemical industry growth rates listed
in reference 5-7. The factors used were as follows:
Base Year to 1977 1980 1985
1971 (Corpus Christi) 1.139 1.252 1.426
1973 (Beaumont) 1.100 1.211 1.378
1974 (Houston) 1.085 1.193 1.358
5-19
-------
For petroleum refinery and chemical plant, 1974 ship and barge
emissions in Houston, the calculation went as follows:
Refinery
(93.1 x 106 bbl/year)(42 gal/bbl)(2.01 lb/1000 gal)(ton/2000 Ib)
(.61) = 2,397 tons/year
Chemical
(23.8 x 106 bbl/year)(42 gal/bbl)(2.01 lb/1000 gal)(ton/2000 Ib)
(.57) = 342 tons/yr.
Total 1974 ship and barge reactive emissions for the Houston area
thus equaled 2,739 tons/year. Results for similar calculations
for all three areas are summarized in Table 5-4.
Table 5-4
Ship and Barge Reactive Hydrocarbon Emissions (tons/yr)
Baseline Year 1977 1980 1985
Corpus Christi
(1971 Baseline) 553 630 692 789
Beaumont
(1973 Baseline) 1,677 1,845 2,031 2,311
Houston
(1974 Baseline) 2,739 2,972 3,268 3,720
Note: About 90% of the above emissions are due to gasoline
loadings.
5.3.3. Emissions from Gasoline Marketing Operations
As service station storage tanks and automobile tanks are
filled, the gasoline vapors are expelled by displacement to the
atmosphere. In the metropolitan areas of Texas, these emissions
5-20
-------
can be quite significant due to the large density of vehicles and
their fueling requirements. Marketing emissions that occur during
loading operations for tank trucks at bulk terminals are not
included in this gasoline inventory but have been separately
accounted for in the point source estimates discussed in Section
5-3.
To calculate emissions from such operations, the TACB used
emission factors developed by the EPA and published in Supplement
No. 1 to AP-42 (Ref. 5-9). Based on their knowledge of the petroleum
marketing industry in the State, the TACB assumed that for loading
the service station storage tank 45% of the loadings would be
splash fill, 45% would be submerged fill, and 10% would have vapor
recovery equipment. The emission factors given in AP-42 are:
(1) 11.5 lb/1000 gal. - splash loading
(2) 7.3 lb/1000 gal. - submerged loading
(3) 0.8 lb/1000 gal. - vapor recovery loading
(4) 11.0 lb/1000 gal. - vehicle tank loading
(5) 0.67 lb/1000 gal. - spillage loss during loading
Based on the assumed percentage breakdown during loading and
the above emission factors, TACB calculated the following net
emission factor for gasoline marketing operations:
EF - (.45)(11.5 + 11.0 + .67) + (.45) (7.3 + 11.0 + .67) +
(0.1) x (0.8 + 11.0 + .67) - 20.2 lb/1000 gallons marketed.
Using an average vehicle gasoline mileage of 13.3 miles per
gallon and a reactivity factor of 0.6, the 20.2 lb/1000 gallons
converts to a factor of .00046 tons of reactive carbon compounds
5-21
-------
emitted per 1000 vehicle miles. This factor, multiplied by the
appropriate vehicle mileage (VMT) gives the emissions from this
source for each area and year of concern. TACB assumed that half
of the additional gasoline necessary for any increase in VMT is
supplied from existing gasoline stations, and half from new
stations. Since stations constructed after 1973 are subject to
new source controls on gasoline storage (usually consisting of
facilities for submerged loading), a smaller emission factor
(.00036 tons of reactive carbon compounds/1000 vehicle miles) is
applied to gasoline dispensed from these stations. Because of
this adjustment the average emission factor varies from .00046 in
1972-73 to about .00044 in 1985. The emissions from gasoline
marketing activities were computed for each county and year of
concern as part of the overall vehicle emissions calculations
described in Section 5.4. Results are summarized in the TACB
worksheets contained in Appendix G.
5.3.4. Area Sources (Solvent Evaporation)
Various manufacturing and maintenance processes result in
the evaporation into the atmosphere of reactive hydrocarbon
solvents. These processes include manufacturing and architectural
coating operations, drycleaning of garments, degreasing and
cleaning of components, etc. The revised inventory prepared by
the TACB specifically includes an estimate of solvent emissions
from trade paints (architectural coatings), and drycleaning
operations in Texas. The development of this inventory was the
same as for the original 1973 Texas plan and is documented in
5-22
-------
reference 5-15. Emissions from these categories are relatively
small and due to changes in the need for such processes the
reactive emissions are decreasing with time.
Industrial solvent emissions such as manufacturing coatings
and solvent degreasing have not been specifically identified in
the TACB inventory, but significant sources should be included by
industry in their submittals of point source inventory questionnaires,
and are included in the point source emissions inventory totals
contained in this document. The TACB did make an estimate of the
amount of industrial degreasing emissions in order to calculate
the possible effects of controls on degreasing operations. The
estimates were made only for the three largest manufacturing
areas, Dallas-Ft. Worth, San Antonio, and Houston. The TACB
based the estimates on inquiries with solvent manufacturers and
distributors in the State. The TACB worksheets for the degreasing
estimate are included in Appendix H. The estimates discussed
above are used for the revised inventory.
5.4 Mobile Sources
Emissions from mobile sources were completely recalculated
by the TACB for the reexamination effort. In reference 5-1 the
TACB discusses in detail how the recalculations were performed
(applicable pages of reference 5-1 are included in Appendix I).
Sources included in the mobile category are exhaust, crankcase,
and evaporative emissions from both gasoline-powered and diesel-
powered vehicles, aircraft emissions, and other transportation
emissions (off-highway, railroads, and vessels).
5-23
-------
By far the most important, due to the large number of vehicles
in this category, are emissions from light-duty gasoline-powered
vehicles. Emissions in this category average about 80% of the
mobile emissions in the Texas areas evaluated in 1972. For this
reason TACB went into the most detail in their evaluation of
light-duty (LDV) emissions.
The LDV emission analysis performed for the reexamination was
in general the same type of analysis as performed by the TACB and
the EPA in 1973 for the original oxidant/hydrocarbon control plans
(Refs. 5-15 and 5-16). The results of this analysis give projected
LDV emissions for years 1971 through 2000 which include the effects
of expected growth in the number of vehicles and emission reductions
as a result of the Federal Motor Vehicle Control Program (FMVC).
Differences between the original calculations and the reexamination
calculation are as follows:
a. More current vehicle registration and vehicle miles
traveled data that reflect the effects of the energy crisis are
incTuded.
b. Revised emission factors contained in Supplement No. 5
to AP-42 are included.
c. Revised emission factors to reflect the additional one-
year extension granted auto manufacturers in June 1974 in meeting
the 90% reduction (a reduction from 4.1 grams/mile in 1970 to
0.41 grams/mile in 1977} established by the 1970 Clean Air Act
are included.
5-24
-------
Including the effect of growth in the number of vehicles on
the road, the TACB calculations show that the FMVCP obtains about
20% reduction from 1971 to 1977. These reductions are not nearly
so great as the previous reductions of about 50% in the initial
plan. This lower reduction is due primarily to the change in
emission factors for evaporative emissions from controlled LDV.
Based on more extensive measurements of evaporative emissions in
1973 from controlled vehicles, the evaporative emission factor
ranged from 0.2 grams/mile to 1.76 grams/mile. This makes the
evaporative hydrocarbon emissions for the 1977 model year and
later model year vehicles about as large as the exhaust emissions,
and is the primary cause of lower credited reductions to the
FMVCP in the revised plan. EPA is presently working with the ./
automobile manufactures to revise the test method for evaporative
loses. This should result in better control for evaporative
losses.
Normal engine deterioration from wear, etc., is included in
the calculations and is determined from in-use vehicle testing
performed by EPA. The assumption is also made that the control
devices installed on new vehicles as part of the FMVCP remain
installed and working properly on the vehicle throughout the life
of the car. If this is not the case, as might occur with removal
of control devices by owners or repairmen, or with poor maintenance,
then even the 20% calculated reduction will not be realized and
emissions from LDV will not decrease as rapidly as expected and
could conceivably increase.
5-25
-------
It should be noted that on March 5, 1975, the Administrator
announced his decision to delay for one year (to 1978 model year
vehicles) the requirement for 90% reductions in hydrocarbon and
carbon monoxide exhaust emissions. In addition, the Administrator
proposed that those statutory standards be further delayed until
the 1982 model years. The effects of the one-year delay and the
effects of the possible delay to 1982 have not been incorporated
into the TACB emission calculations for LDV. The effects on
calculated 1977 LDV emissions would not be significant, but
vehicle exhaust emissions for 1980 and 1985 could increase 10 to
20% as a result of these delays.
5.5 Ciudad Juarez Emission Estimate
Since Ciudad Juarez and El Paso are adjacent to one another
and share a conmon air shed, hydrocarbon emissions from Ciudad
Juarez should be considered with the El Paso emissions in order
to properly model the effects of hydrocarbon emissions on oxidant
levels in El Paso. The TACB in Addendum 1 to reference 5-1 made an
estimate of Ciudad Juarez hydrocarbon emissions by assuming the
emissions for years 1972 through 1977 would be equal to the 1972
El Paso emissions. Based on additional information, the EPA made
a similar estimate of Ciudad Juarez emissions for years 1974
through 1985. The assumptions used by the EPA are listed below:
1. Area sources and point source hydrocarbon emissions were
assumed to be 1/2 that of El Paso in 1974 due to the expected
fewer hydrocarbon-producing industries in Ciudad Juarez. These
emissions were assumed to remain constant through 1985.
5-26
-------
2. Gasoline marketing emissions were assumed to be 0.44
that of El Paso based on the ratio of numbers of vehicles in
Ciudad Juarez and El Paso.
3. Light-duty gasoline vehicle emissions for Ciudad Juarez
in 1974 were assumed to be 1.55 times the 1974 El Paso value.
This factor was derived assuming that all Ciudad Juarez vehicle
emissions were equal to typical levels for precontrolled vehicles
(pre '68) in the U.S. plus 20% to account for less maintenance of
Juarez vehicles. This value was then reduced by the ratio of the
number of vehicles in the two cities. With typical pre-'68
exhaust emissions equal to 8.8 grams/mile and 1974 El Paso exhaust
emissions equal to about 3.0 grams/mile, the factor was cal-
culated as follows: (8.8)(1.2) ^ 44 = ] 55
3.0
Based on Mexico's emission control program (discussed in reference
5-17) it was assumed that controls would obtain similar percentage
reductions as in El Paso for years 1977, 1980, and 1985.
4. Heavy-duty vehicle emissions were assumed to be 1/2 that
of El Paso.
5. Diesel vehicle emissions were assumed to be equal to El
Paso diesel emissions.
6. Aircraft emissions were assumed to be 5% of those in El
Paso based on less scheduled commercial traffic and comparatively
little military air traffic.
5-27
-------
7. Other transportation emissions were assumed to be 0.44
as great as El Paso based on the ratio of vehicles in each area.
The emission inventory summaries in Section 5.6 include
combined El Paso-Ciudad Juarez emissions based on the EPA estimate
along with individual estimates for El Paso and Ciudad Juarez.
Although values for Ciudad Juarez are based on several gross
assumptions, the combined El Paso-Ciudad Juarez estimate is
considered much more representative of the area emissions than
the more accurate El Paso inventory if taken alone. The combined
area estimate should allow a much better prediction of how controls
developed and implemented in El Paso affect the oxidant levels in
the El Paso area.
A detailed evaluation of Ciudad Juarez hydrocarbon emissions
would better define the specific contributions of their emissions
to the oxidant levels in El Paso and identify what possible
international control strategies might be considered.
5.6 Emission Inventory Summaries
A summary of the reactive hydrocarbon emission inventories
is presented in Tables 5-5 through 5-13. The projected emissions
for years 1977, 1980, and 1985 reflect the anticipated emission
levels with no controls beyond the TACB Stationary Source Regu-
lations V (existing sources) and VI (new sources), and the Federal
Motor Vehicle Control Program (FMVCP). In addition, the tables
present the required emission reductions and allowable yearly
emissions to meet the oxidant standard based on the reduction
requirements established in Section 4.
5-28
-------
As discussed in Section 4.3, the model area chosen by the
EPA is slightly different from that chosen by the TACB in that
emissions from five additional counties are included for modeling
purposes. Based on the TACB worksheets supplied to the EPA the
inventories in Tables 5-5 through 5-13 include the emissions from
these additional counties and, therefore, differ slightly from
the values reported by the TACB in reference 5-1.
In all Texas areas evaluated, the projected 1977, 1980, and
1985 inventories exceed the calculated allowable emissions
necessary to attain the oxidant standard. Emissions exceeding
allowable range from as little as 14% in 1985 for El Paso and as
great as 134% in 1985 for Houston. It should be noted that the
projected emissions for 1980 and 1985 are more uncertain than the
1977 projections. This increased uncertainty is due to the
limited accuracy of specific area growth trends in industry and
transportation, and also to possible changes in the FMVCP between
1977 and 1982.
5-29
-------
Table 5-5--Reactive Hydrocarbon Emission Inventory
with FMVC and Existing Stationary Source Controls
for Austin Area
1973* 1977 1980 1985
tons/yr tons/yr tons/yr tons/yr
Stationary Sources
Area (trade paints &
drycleaning) 632 423 374 328
Point Sources 1,921 897 897 983
Ship & Barge
Gasoline Marketing
Storage Tank Loading 405 441 486 569
Vehicle Tank Loading 607 662 729 853
3,565 2,423 2,486 2,733
Mobile Sources
Gasoline Vehicles
LDV 11,034 8,809 7,248 6,079
HDV 704 562 463 388
Diesel Vehicles 155 141 155 177
Aircraft 223 287 338 438
Other Transportation 977 1.010 1,045 1,137
13,093 10,809 9,249 8,219
Grand Totals 16,658 13,232 11,735 10,952
Required Reduction = 50% x Baseline Inventory = 8,329 tons/year
Allowable Emissions = 8,329 tons/year
* Baseline Year
Austin Area Counties: Travis and Hays
5-30
-------
Table 5-6--Reactive Hydrocarbon Emission Inventory
with FMVC and Existing Stationary Source Controls
for Beaumont-Port Arthur Area
1973* 1977 1980 1985
tons/yr tons/yr tons/yr tons/yr
Stationary Sources
Area (trade paints &
drycleaning) 685 469 420 377
Point Sources 160,404 48,028 48,828 50,376
Ship & Barge 1,677 1,845 2,031 2,311
Gasoline Marketing
Storage Tank Loading 378 367 457 533
Vehicle Tank Loading 568 551 683 800
163,712 51,260 52,419 54,397
Mobile Sources
Gasoline Vehicles
LDV
HDV
Diesel Vehicles
Aircraft
Other Transportation
10,334
660
354
121
1,735
13,204
176,916
7,461
476
225
130
1,736
10,028
61,288
6,802
434
277
140
1,736
9,389
61 ,808
5,666
362
317
154
1.736
8,235
62,632
Grand Totals
Required Reduction = 75% x Baseline Inventory = 132,687 tons/year
Allowable Emissions = 44,229 tons/year
*Baseline Year
Beaumont-Port Arthur Area Counties: Hardin, Jefferson, and Orange
5-31
-------
Table 5-7--Reactive Hydrocarbon Emission Inventory
with FMVC and Existing Stationary Source Controls
for Corpus Christi Area
1971* 1977 1980 1985
tons/yr tons/yr tons/yr tons/yr
Stationary Sources
Area (trade paints &
drycleaning) 573 327 283 241
Point Sources 63,527 37,028 36,865 40,419
Ship & Barge 553 630 692 789
Gasoline Marketing
Storage Tank Loading 270 330 365 427
Vehicle Tank Loading 404 496 547 641
65,327 38,811 38,752 42,517
Mobile Sources
Gasoline Vehicles
LDV 8,934 6,857 5,583 4,598
HDV 570 438 356 294
Diesel Vehicles 63 76 82 94
Aircraft 2,767 2,641 2,659 2,697
Other Transportation 1.202 1,180 1,180 1,180
13,536 11,192 9,860 8,863
Grand Totals 78,863 50,003 48,612 51,380
Required Reduction = 57% x Baseline Inventory = 44,952 tons/year
Allowable Emissions = 33,911 tons/year
*Base1ine Year
Corpus Christi Area Counties: Nueces and San Patricio
5-32
-------
Table 5-8--Reactive Hydrocarbon Emission Inventory
with FMVC and Existing Stationary Source Controls
for Dallas/Fort Worth Area
1974* 1977 1980 1985
tons/yr tons/yr tons/yr tons/yr
Stationary Sources
Area (trade paints &
drycleaning) 4,428 3,625 3,284 3,010
Point Sources 13,782 13,107 13,552 14,289
Ship & Barge
Gasoline Marketing
Storage Tank Loading 2,714 3,041 3,370 4,005
Vehicle Tank Loading 4,076 4,563 5,057 6,006
25,000 24,336 25,263 27,310
Mobile Sources
Gasoline Vehicles
LDV 63,598 55,800 46,757 40,638
HDV 4,060 3,562 2,984 2,594
Diesel Vehicles 1,516 1,676 1,837 2,108
Aircraft 4,381 5,003 6,555 8,996
Other Transportation 6.746 7,238 7,667 8.719
80,301 73,279 65,800 63,055
Grand Totals 105,301 97,615 91,063 90,365
Required Reduction = 57% x Baseline Inventory = 60,022 tons/year
Allowable Emissions = 45,279 tons/year
*Baseline Year
Dallas-Fort Worth Area Counties: Coll in, Dallas, Denton, Ellis
Hood, Johnson, Kaufman, Parker, Rockwall, Tarrant, and Wise
5-33
-------
Table 5-9--Reactive Hydrocarbon Emission Inventory
with FMVC and Existing Stationary Source Controls
for El Paso Area (Juarez Not Included)
Stationary Sources 1974* 1977 1980 1985
tons/yr tons/yr tons/yr tons/yr
Area (trade paint &
drycleaning) 616 480 411 343
Point Sources 4,202 2,196 2,237 2,289
Ship & Barge
Gasoline Marketing
Storage Tank Loading 300 336 373 443
Vehicle Tank Loading 450 504 559 665
5,568 3,516 3,580 3,740
Transportation
Gasoline Vehicles
LDV
HDV
Diesel Vehicles
Aircraft
Other Transportation
8,486
542
191
676
737
10,632
16,200
7,532
481
213
699
808
9,733
13,249
6,277
401
236
832
743
8,489
12,069
5,185
331
276
994
762
7,548
11,288
Grand Totals
Required Reduction = 38% x Baseline Inventory = 6,156 tons/year
Allowable Emissions = 10,044 tons/year
*Baseline Year
El Paso Area County (Juarez not included): El Paso
5-34
-------
Table 5-10--Reactive Hydrocarbon Emission Inventory
with FMVC and Existing Stationary Source Controls
for El Paso Area (Juarez Included)
1974* 1977 1980 1985
tons/yr tons/yr tons/yr tons/yr
Stationary Sources
Area (trade paints &
drycleaning) 924 788 719 651
Point Sources 6,404 4,398 4,439 4,491
Ship & Barge
Gasoline Marketing
Storage Tank Loading 432 484 537 638
Vehicle Tank Loading 648 726 805 958
8,408 6,396 6,500 6,738
Mobile Sources
Gasoline Vehicles
LDV 21,639 19,206 16,006 13,221
HDV 813 752 672 602
Diesel Vehicles 382 426 472 552
Aircraft 710 734 874 1,044
Other Transportation 1,061 1.164 1,070 1,097
24,605 22,282 19,094 16,516
Grand Totals 33,013 28,678 25,594 23,254
Required Reduction = 38% x Baseline Inventory = 12,545 tons/year
Allowable Emissions = 20,468 tons/year
*Baseline Year
El Paso Area County (Juarez included): El Paso
5-35
-------
Table 5-11--Reactive Hydrocarbon Emission Inventory
Estimates in Juarez
1974* 1977 1980 1985
tons/yr tons/yr tons/yr tons/yr
Stationary Sources
Area (1/2 El Paso in '72) 308 308 308 308
Point Sources (1/2 El Paso '74) 2,202 2,202 2,202 2,202
Ship & Barge
Gasoline Marketing
Storage Tank Loading (.44 x
El Paso) 132 148 164 195
Vehicle Tank Loading (.44 x
El Paso) 198 222 246 293
2,840 2,880 2,920 2,998
Mobile Sources
Gasoline Vehicles
LDV (1.55 x El Paso 13,153 11,674 9,729 8,036
HDV (1/2 El Paso '74) 271 271 271 271
Diesel Vehicles (- to El Paso) 191 213 236 276
Aircraft (5% of El Paso) 34 35 42 50
Other Transportation (.44 x
El Paso) 324 356 327 335
13,973 12,549 10,605 8,968
Grand Totals 16,813 15,429 13,525 11,966
*Baseline Year
5-36
-------
Table 5-12--Reactive Hydrocarbon Emission Inventory
With FMVC and Existing Stationary Source Controls
for Houston-Galveston Area
1974* 1977 1980 1985
tons/yr tons/yr tons/yr tons/yr
Stationary Sources
Area (trade paint &
drycleaning) 4,065 3,299 2,962 2,676
Point Sources 215,302 139,887 152,157 175,047
Ship & Barge 2,739 2,972 3,268 3:720
Gasoline Marketing
Storage Tank Loading 2,469 2,743 3,021 3,533
Vehicle Tank Loading 3,704 4.114 4,533 5,296
228,279 153,015 165,941 190,272
Mobile Sources
Gasoline Vehicles
LDV 61,884 55,311 45,073 36,469
HDV 3,950 3,531 2,877 2,328
Diesel Vehicles 1,635 1,804 1,973 2,253
Aircraft 1,739 1,913 2,384 3,065
Other Transportation 7,009 7,351 7.712 8,682
76,217 69,910 60,019 52,797
Grand Totals 304,496 222,925 225,960 243,069
Required Reduction * 66% x Baseline Inventory * 200,967 tons/year
Allowable Emissions * 103,529 tons/year
*Baseline Year
Houston-Galveston Area Counties: Brazoria, Chambers, Fort Bend, Galveston,
Harris, Liberty, Matagorda, Montgomery, and Waller
5-37
-------
Table 5-13--Reactive Hydrocarbon Emission Inventory
with FMVC and Existing Stationary Source Controls
for San Antonio Area
1971* 1977 1980 1985
tons/yr tons/yr tons/yr tons/yr
5,936
4,568
Stationary Sources
Area (trade paints &
drycleaning)
Point Sources
Ship & Barge
Gasoline Marketing
Storage Tank Loading
Vehicle Tank Loading
Mobile Sources
Gasoline Vehicles
LDV
HDV
Diesel Vehicles
Aircraft
Other Transportation
Grand Totals
Required Reduction - 45% x Baseline Inventory
Allowable Emissions * 20,558 tons/year
2
2
1
,004
,054
751
,127
1
1
,142
956
988
,482
1
1
1
989
,072
,099
.647
1
1
1
842
,232
,301
,952
4,807
5,327
22,783
1,454
545
4,483
2,178
31 ,443
37,379
19,308
1,232
641
3,299
2,172
26,652
31,220
16,445
1,050
710
3,473
2,167
23,845
28,652
14,051
897
814
3,816
2,249
21,827
27,154
16,821 tons/year
* Baseline Year
San Antonio Area Counties: Bexar, Comal, and Guadalupe
5-38
-------
Section 5 References
5-1 Texas Air Control Board Report SP-1, "Reactive Carbon Compound
Control Strategy Reexamination for the State of Texas," March 13,
1975.
5-2 EPA Office of Air Planning and Standards, Technical Report;
Control of Photochemical Oxidants - Technical Basis and
Implementation of Recent Findings, July 15, 1975.
5-3 EPA Office of Air and Water Programs, Emission Control Tecnology
Division. Aldehyde and Reactive Organic Emissions From Motor
Vehicles. Part I and II. APTD-1568a & b, March 1973.
5-4 Hurr, R. W. Air Pollution (A. Stern, Ed), Volume III, 2nd
Edition (1968).
5-5 Air Resources Board. State of California, "Light Hydrocarbons in
Engine Exhaust and the Los Angeles Atmosphere", Fall 1969.
5-6 EPA Office of Air Quality Planning and Standards, Report
AP42; Compilation of Air Pollutant Emission Factors; Second
Edition, April 1973.
5-7 EPA/U.S. Department of Commerce Report, Population and
Economic Activity in the United States and Standard Metro-
politan Statistical Areas, July 1972.
5-8 Department of the Army Corps of Engineers Annual Report,
Vlaterborn Commerce of the United States, Part 2, Calendar
Year 1972.
5-9 EPA Office of Air Quality Planning and Standards Report
AP42, Supplement No. 1. July 1973.
5-39
-------
5-10 American Petroleum Institute Bulletin 2514, Evaporation
Loss from Tank Cars, Tank Trucks, and Maine Vessels, 1957.
5-11 EPA Region VI Letter to 22 Houston Area Industries, from
James Doyle, Dated January 29, 1975.
5-12 Exxon Company, USA, Maine Vapor Recovery Questionnaire,
(2 Volumes) April 17, 1975.
5-13 Exxon Company, USA, Letter to EPA from H. H. Meredith, Jr.,
dated April 23, 1975.
5-14 Exxon Company, USA, Letter to Texas Air Control Board from
J. M. Johnson dated June 9, 1975.
5-15 Texas Air Control Board Report, Hydrocarbon Control Strategies
for the State of Texas, April 13, 1973.
5-16 Federal Register, Volume 38, No. 213, November 6, 1973.
5-17 Garcia, Gaspar, "Automotive Air Pollution? (Mexico Department of
Environmental Improvement.) In, Proceedings Third North American
Motor Vehicle Emissions Conference, September 25-27, 1974.
5-40
-------
6.0 Control Measures Considered
A number of measures designed to reduce emissions of reactive
hydrocarbons have been considered. Several of these affect stationary
sources while others affect mobile sources (light-duty vehicles,
LDV). The stationary source control measures considered are the
extension of Texas Regulation V to additional counties, gasoline
marketing vapor recovery, ship and barge vapor recovery, control of
emissions from crude petroleum, and control of emissions from
degreasing operations. Mobile source controls considered are
inspection/ maintenance and reduction in vehicle miles traveled
(VMT).
6.1 Existing Controls
Existing controls consist of Texas Regulation V for control of
volatile carbon compounds from stationary sources in specific Texas
counties and motor vehicle emission controls from the Federal Motor
Vehicle Control Program. These two categories, in fact, contribute
the greatest to hydrocarbon reductions in each area in the Texas
plan. The present restudy updated expected emissions from both
stationary and mobile sources (see Section 5.0) as a result of:
(1) additional sources coming into compliance; (2) addition of
sources not included in the original inventory; (3) an update of
vehicle population; (4) the effect of the interim Federal motor
vehicle emission standards; and (5) an updating of emission factors
for in-use vehicles.
6.2 Extension of Texas Regulation V
Texas Regulation V, "Control of Air Pollution from Volatile
Carbon Compounds," limits hydrocarbon emissions from storage
6-1
-------
tanks, loading and unloading facilities (except crude oil and ships
and barges), water separation units, and waste gas disposal from
processing facilities. The regulation presently applies in sixteen
counties in Texas which include most of the counties with oxidant
problems. Additional reductions can be obtained by extending Texas
Regulation V to Hardin County in Beaumont-Port Arthur area and to
Tarrant County in the Dal las-Fort Worth area. Requirements of the
Texas Regulation V include the following:
1. Storage tanks of greater than 25,000 gallons capacity
containing any volatile carbon compounds must be pressure tanks
capable of maintaining working pressures and preventing vapor or
gas loss to the atmosphere or, otherwise, equipped with a floating
roof or vapor recovery device.
2. Loading and unloading facilities for volatile carbon
compounds having 20,000 gallons or more throughput per day must be
equipped with a vapor recovery system.
3. Water separation units receiving 200 gallons or more of
volatile carbon compounds per day must be controlled by tight
enclosure, floating roof, or vapor recovery system.
4. Waste gas streams containing specified volatile carbon
compounds must be burned in a smokeless flare or direct-flame
incinerator at a temperature of at least 1300° F.
The effectiveness of Texas Regulation V in terms of reduction
of emission of hydrocarbons where the required controls are applied
is about 96-97% for storage tanks, (floating roofs or vapor recovery
system), 85% for loading and unloading operations (vapor recovery
6-2
-------
systems), 98% for water separators (tight enclosures or other
devices), and 98% for waste gas streams (incineration). The expected
additional reductions in emissions of reactive hydrocarbons by
application of Texas Regulation V to Hardin County in the Beaumont-
Port Arthur area and to Tarrant County in the Dallas-Fort Worth
area shown in the Tables 6-1, 6-2, and 6-3.
6.3 Gasoline Marketing Vapor Recovery
Two types of gasoline marketing vapor controls were promulgated
on November 6, 1973. These were termed Stage I (filling of storage
tanks at service stations and bulk terminals) and Stage II (filling
of individual vehicle tanks). In the August 7, 1974 decision, the
Court upheld the use of Stage I vapory recovery for some areas,
remanded it back to EPA for further consideration for other areas,
and invalidated it for still other areas. The original Stage II
vapor recovery regulation was remanded back to EPA for further
consideration of the need for the regulation in specific areas and
technical issues concerning the degree of control required by the
regulation.
A typical gasoline distribution pattern consists of delivery
by pipeline, barge, or tanker to intermediate storage, then by
pipeline or barge to bulk terminals. From the bulk terminal, the
gasoline is delivered by tank truck or trailer to the service
station or to a bulk plant and then by tank truck to the service
station. At the service station, the gasoline is transferred from
the tank trucks into storage tanks. Automobiles are refueled from
the storage tanks via the service station pump island.
6-3
-------
In uncontrolled bulk terminal tank truck loading operations,
service station deliveries and automobile refueling, the vapors
displaced by the gasoline are normally released directly to the
atmosphere. The amount of hydrocarbons which are contained in
these vapors is highly variable depending on the manner in which
the tank truck is loaded at the terminal; the amount of gasoline
contained in the vapors carried by the returning truck; the manner
in which the gaoline is delivered to the service station; operator
practice in automobile refueling; and a number of other factors
such as physical properties of the gasoline; geographical and
seasonal effects; and meteorological conditions.
As exhaust emission control standards for motor vehicles
become more stringent, the proportional share of hydrocarbon
emissions from evaporative losses will increase. Gasoline marketing
vapor recovery is designed to reduce hydrocarbon vapor emissions
during the filling of tank trucks at bulk terminals, storage tanks
at service stations (referred to as Stage I controls) and during
the filling of automobile gasoline tanks from these storage tanks
at service station pumps (referred to as Stage II controls). The
regulation requires that such tanks be equipped with a submerged
fill pipe and a system of vapor recovery to prevent release of
hydrocarbon vapor to the atmosphere during loading. The vapor
recovery system must have an efficiency of at least 90% and can
employ a vapor tight vapor return to the tank truck or a refrigeration-
condensation system to achieve control. The delivery tank truck
must be compatible with the service station system and can only be
6-4
-------
refilled at facilities with vapor recovery systems. Application of
this measure to all seven areas results in reductions shown in
Table 6-1 through 6-3.
Two concepts have been developed for Stage II vapor recovery
during vehicle fueling; namely, displacement (the balance system)
and vacuum assist with secondary recovery. The balance system
depends upon the displacement of air and hydrocarbon vapors as a
result of pumping gasoline into the automobile fuel tank. Pressure
in the tank created by the incoming fuel forces vapor out to the
atmosphere under uncontrolled conditions. The balance system is
designed to provide an alternative route for those vaporsthrough
a vapor return line to the underground storage tank where it replaces
the liquid gasoline which is pumped to the vehicle. Potential
points of leakage with the balance system are (1) the vehicle vent,
(2) the storage tank vent, (3) the automobile fill neck/fill nozzle
interface, and (4) any unforeseen leaks in the system. Automobiles
produced since 1970 are equipped with carbon canisters which eliminate
losses through the vehicle vents. The bulk of hydrocarbon emissions
are released at (1) the automobile fill neck and, to a lesser
degree at (2) the storage tank vent (Ref. 6-1).
The vacuum assist system adds two features to the balance
concept, i.e., a blower which develops a suction at the nozzle/
fill neck interface and a processing unit to recover or otherwise
reduce hydrocarbon emissions to the atmosphere. The purpose of the
blower is to force displaced vapors to enter the vapor return line
rather than leak to the atmosphere. The processing unit condenses,
6-5
-------
^s
s
O S-
(O C. "^^
to o co
^> c;
c o
=C -M
1
1
1
, .
^«
o
O CO >]
M CO
3 C
CD f0 CD
^ CD -M
1
1
1
1
^ N
^5
o -
O E
i O O
s- o
-o
>> 0
0 Q-
(O
O)
oc
.
^
1
VO
o>
1
,
r (A
LU c:
o
4_J
,^-s.
C" £^
1 +* ** *
s-
co O 1-
10 3 >!
I C/)
to +-> c
o u. o
1
1
1
1
^ -*
cy^
.
r
^_^
CM
in
0
f>
CM
r-
4-J ^*~*s
CO ^5
r ^ *
s-
.c s-
O >i
CO CO
3 c
O. O
S- 4J
o
o
1
1
1
1
_^
CM
^~ ^""^*
1 3 ^^
^~ **__-.
-M -!-»
C $- 5-
o «a: >i
E -^
3 +-> 10
as S- c.
OJ 0 0
OQ O- -M
sS
? s-
-4-) ^
to ->^
3 CO
CO
0)
O CO
S Q)
3 a:
o
to «»_
o
i- C
O)
(O +->
+J X
to uj
in cn
CM CM
O> CM
OO
CT» O
#i
r
CO O
O r-^
CO
in o
in o
» «
CM CO
OO CO
O CM
^ ***
CM OO
r^ vo
CM CO
*^^»^"^
r~. CM
CM CO
» '^^
00 r-
CM CO
oo co
»> tft
CM CO
co in
O 0
> '^^
r-. CM
vo vo
CM CO
s-^^
CM CM
*
0 0
~~-*^^
1^- CM
o-» o
CM «*
^-^.* ^
r- d»
CM CM
'- -
f^* CO
in oo
CO «3~
cn
c
i
+->
co to tn
(U
cn
s- >>
CO !~
co C O
c i- s-
O CO 3
<_> cO O
CD (X)
c cn ai
> TO -
i - 43
0 O
to s:
d-
CM
o
CM
cr>
a?
o
~ '
VO
CO
VO
»
CM
"^
1
*
CT>
in
CO
^ *
in
C-J
^
cr>
in
VO
>
CM
^
o
* . ^
r-.
CM
CO
. .
CM
O
^
VO
m
re
* *
in
CM
* '
o
CM
^J-
^^^
C
o c
u
O
3
-a
0)
or.
t
s
>
6-6
-------
rx CM
.
6«
0 S-
C >,
«a c -<
«/> O tfi
i j; f"
d O
1 CM OO
s « -
CM CM
CM O
1 O CM
1 *. »
i r
STP
^,
^«
o
E 4-» s-
o v> >,
-u cu ~-~.
tn > in
3 i E
^> to o
,
^5>
^-^
o
U> S-
"£
r-
^
^-
o
4_>
O
5
-o
O)
a:
c
O in
-i- CU
i/l 5-
10 3
1 C/1
c (O
LU CU
s:
c
O r
-0 0
0 S- J-
CO !O +->
en o c:
. 00
E- CJ
-o
>v to
i o
O D-
5- S-
0 0 >,
ref 3 \
1/1
c
tO -M O
O U. 4->
1 O
1 '
O cn
C r- 0
i oo co
« r>
IO - -
r S3
S v '
c~
CJ S-
>^
t/) -^
3 c/l
EL C
S.- 0
O -M
CJ
^~^^.
i ^a- in
i
0 0
* f^ f
i vo cn
1 CTi CTl
CM CO
. . ^, ,
o-a CM co
&_ * «
1 .",3 ^5
r~ < f
4-^ -4-^
C S- S-
O i
E
3 +J c^
(C i. C
O) 0 O
CD a. +->
*
CD O O
^ ^ '^ ^
oj o cn
co r*^ 0^1
co ro «3-
^-^. .
st- CM
t
^
c;
4J >.
tn -,'
3 to
<: c
0
i CNi CO
3- CM
1 CTi CO
i co in
i
1
i
i
£
^
CM
cn
CM
k
CM
"~
1
1
1
1
, ,
CO
o
^
i
CO
CO
,
o
CM
- "
in
in
in
»
r
, s
O
m
CO
* ^
Lf '
CO
CM
ft
CO
i
i
1
1
0
1 CO
1
i-.
CM
1 i
i **
i
<£> LO
c o
' -
in r-
CTl CTl
oo ^i-
* *«
, 1
1 1
1 1
1 1
1 1
, .
1 -vl-
1
1
»_^
i l^~~
i en
|CT
«%
r
. ,
LO 1
. |
O
>^^«
f 1
O 1
^J-
^^
r^. !
I
O
* *
oo
t-v i
* i
9V
I '
i 1
1 1
1 1
1 I
cn
,-1
^"^
O
O
r-
«r>
o
CO
r
O%
A
r-~
^.^
OO
0
'
r-.
^o
CM
, .
Cf>
i
-
o
CTi
CTl
\
i
^~.
CO
o
-. "
CO
CO
CM
^-^
CM
CD
*~-^
CTi
00
CM
. ^
CTI
oo
o
CO
I
CO
"
LO
^~
fl
^
^
._!
'
in
in
i~
»
<^o
,__^
CO
r
'
CTl
CO
r^-
, .
r
OO
V ^
CO
c^.
CM
*
CO
.
in
.
o
^
^~
CTl
CO
, .
CO
.
CD
*~-^
to
r-.
^j-
. .
O
.
CO
1^
0
I/I
CU
O
s_
3
O
to
*>^
i.
ro
C
O
r
^_>
(O
CO
D)
CU
CC
M-
O
E
O
i
to
r~
tu cu
cn cn
fl3 (O
4-} 4-*
CO CO
CO
cn
to
S.
o
.40
CO
,
r
O
O!
-o
s_
CO
-
o
s_
4_>
C
0
o
i-
o
Q.
(O
^>
CU
cn
S- >>
(0 S-
CQ CU
>
o5 O
U
a. cu
f C£
r~
CO
r
O *
S- cn
4-> c:
C -r-
O LO
c.^ to
cu
+-> S-
E cn
CU CO
> T3
r~- ^~^
o
CO
^>^
l/>
cu
o
t_
^3
O
CO
CO
p
1
o
o
2:
c
o
r-
+->
0
cu
O-
(/>
c
1 1
~^»
Q
CO
U
C
ITS
c
cu
4-3
C
1
03
s:
^'^
r-«-
N S
t/1
^
0
I
^_>
^J
n
"O
OJ
p^
1
>
6-7
-------
,_,
2.
o s-
E -r- >,
as E ~v.
co o t/>
»-> E
E O
,
1
1
1
,_,
^^
1 E
O
E 4-> i-
O in >i
M CO -^
3 i E
O 03 O
31 CD -M
-^o"
. s
O
-C in X-
-M
3
c
O
«^~
O
3
T3
CD
QC
E
O in
r- CO
in S_
t/> 3
i- in
E to
LU CO
E
O f
-Q 0
in s- $-
co ra +->
CTi O E
r O O
S- 0
a
>> CO
JC
CO -i
> in
c in
--> O
o a.
to
0)
CXL
.
CO
1
VO
CO
_Q
*o
1
10 >,
o. \
Vt
i E
uj o
»-»
^^
^§
c~ *. »
1 ->
S~ S-
1/1 o >,
10 3 ---.
i in
r . E
(O 4-» O
Q Lu -M
1
1
1
1
1
1
1
1
^
r
^_*r
CM
in
o
A
CM
f
\ * ^ ^
(/> ^5
f ^^*
i.
^ $_
CJ >.
in c/>
3 E
a. o
S. 4->
o
CJ
1
1
1
1
^^
CM
1L ^^-^
1 3 S«
f~ *^^'
4-^ 40
E S- S-
O ^C ^i
E -^
3 4-> in
.
o
^ '
CM
CO
CO
1
*^^J
s«
c
r- ^_
M >i
Ul ^
3 in
1
1
1
CMOO
CO CO
o in
i CM
CM «*
M f*
r f~~
f CO
,_! ,_I
^-^^.^
VO vo
OO VO
CM 00
M «
CO CO
** -^""v
f O
. CO
* '* ^
en in
in oo
co *3-
VO CM
* *
CO «*
*w_^^ ^
in ^J-
CM OO
t-~ oo
A f*
co *d-
^-«*^^
=1- vo
.
0 0
** X«« X
VOOO
vT VO
CO si-
.. .
CM CO
.
O 0
- ^*-
CM «*
CO OO
*j- in
, ., .
oo i^
CM 00
>
r- OO
VO CM
^- VO
1
1
1
I
0^
=3-
**_^»
CM
CM
»
CM
r
,
1
1
1
OO
o
V ^
t
CO
CO
*- ^
o
CM
N-^
in
in
in
9*
r~
^
0
.
CO
^ ^
in
oo
CM
*
in
i
i
i
i
i
i
i
i
P
0
"« ^
00
in
r
*»
OJ
1
I
1
1
1
1
1
1
* -*
<£>
0
» '
oo
in
^~
. ,
00
o
^ "
o
^~
co
»
t
i
i
i
i
o
^
0
in
*
r
^T
0
*fc-x
!
00
0
**
CM
1
1
1
1
"erT
i -
«~- *
VO
CO
o
«
CM
1
1
1
1
1
1
1
1
1
1
1
1
*
,_!
^*
00
CO
in
in
o
* *
r~-
CTi
OO
H
r
*^~^-
IO
0
*^-x
o>
»
r^
^^^
CM
O
*^-^
to
r^
r~~
^-^
^~
o
* *
f*^.
f-~
CM
.^-^
**
r
^ *
OO
co
CM
vo
CO
'
^J.
CO
en
oo"
o
~ ^
CO
in
in
*
CM
^"~^
'
,_!
^-^
CO
VO
CO
r^T
CM
* -
in
^~
oo
*
CM
^ N
«^-
O
* ^
CM
CM
CO
,^^
CM
0
^ "
r~~.
en
CO
, .
vo
C-J
vo
CM
«d"
CO
,
i.
at
C
o
r
M
ca
-M
CO
01
QJ
a:
M-
0
c
0
in
c
cu
-*->
X
UJ
O)
c
r
-»->
ai
.^
j-
(O
s:
CO
c
r .
0
vn
m
O
i i
i 1 1 i
CO CO
en en
to to
!-> -M
CO CO
a>
CD
(O
i.
0
4->
CO
r r~
r- O
o s-
4->
CO C
-o o
30
S-
tJ
S-
o
Q.
fO
>
CO
en
$-
to
CQ
o3
O.
r
.E
CO
>,
S-
01
>
0
O
CO
ct
,^~
0--^
s- en
+-> E
C -r-
O in
o s-
E en
CO CO
> -o
t>^ --^
o
CO
^^^
tn
in
CO
o
S-
3
O
CO
CO
1
f-
.a
0
s:
E
0
4->
O
CO
D_
CO
E
i i
>
Q
_1
CO
(J
E
(O
E
CO
+>
E
i^
fO
s:
&e
r--
%^^'
in
E
O
r
+J
6-8
-------
absorbs, or incinerates hydrocarbon vapors which are not retained
in the storage tank. With vacuum assist systems, there is always
an influx of additional air caused by the "vacuum sweeper" char-
acteristic of the system. Vacuum assist systems are composed of
appreciably more mechanical and other working components than
balance systems. However, they are more effective than balance
systems since they have the potential of preventing leakage at the
nozzle/fill neck interface and of recovering excess hydrocarbons
created by "vapor growth" in the vehicle tank. For a more detailed
discussion see reference 6-2.
Stage II, for control of hydrocarbon vapors during filling of
vehicle tanks, specified that service stations may make such
transfer of gasoline only when the dispensing facility is equipped
with a system which meets a specific emission limit. This limit
depends on the amount of gasoline dispensed by a facility (service
station) during a representative calendar year. For a service
station which dispenses 360,000 gallons or more of gasoline per
year, the emission limit is 0.40 gram of total hydrocarbon vapor
per gallon of gasoline dispensed, requiring a vacuum assist vapor
recovery system; for a station which dispenses 120,000 gallons or
more but less than 360,000 gallons per year, the emissions limit
is 0.80 gram of total hydrocarbon vapor per gallon of gasoline
dispensed and can usually be obtained with a balance system.
Service stations dispensing less than 120,000 gallons of gasoline
per year would not be subject to this measure.
6- 9
-------
Expected reductions in emissions of reactive hydrocarbons for
Stage I and Stage II controls for all of the areas for which these
measures were considered are shown in the Tables 6-1 through 6-3.
Stage I was assumed to provide about 93% reduction and Stage II was
assumed to provide about 73% reduction.
6.4 Control of Emissions from Crude Petroleum
Evaporative emissions of hydrocarbons from storage tanks occur
in all phases of the petroleum industry, from production of crude
oil through final storage of finished products in the distribution
and marketing system. Hydrocarbon emissions from storage tanks
depend essentially on three basic mechanisms: breathing loss,
working loss, and standing storage loss. Breathing and working
losses are associated with cone or fixed roof tanks, variable vapor
space tanks, and low pressure tanks. Standing storage losses are
generally associated with floating roof tanks. The magnitude of
emissions depends on factors such as the physical properties of the
material being stored, climatic conditions, and the size, type,
control and condition of the storage tank.
Texas Regulation V is effective in control of hydrocarbon
emissions from storage tanks and other facilities and processes to
which it applies. However, the regulation contains an exemption
for storage tanks containing crude petroleum. Emission inventories
show that crude-petroleum storage vessels are a significant source
of reactive hydrocarbon emissions in certain areas (see Tables 6-1
through 6-3). This measure requires control of emissions from
storage vessels with crude petroleum, having a capacity greater
6-10
-------
than 100,000 gallons. Such storage vessels, except those located
at a drilling and production facility, would be required to be
equipped with a floating roof or a vapor recovery system with an
efficiency of at least 85%. Expected reductions in reactive hydro-
carbon emissions due to application of these controls on crude-
petroleum storage in the areas with significant amounts of crude
storage are shown in Tables 6-1 through 6-3.
6.5 Ship and Barge Vapor Recovery
On November 6, 1973 (38 FR 30626), the EPA promulgated a
regulation for control of reactive hydrocarbon emissions from ship
and barge loading and unloading facilities in the Houston-Galveston
area. This regulation eliminated an exemption for ships and barges
in Rule 503 of Texas Regulation V. This meant that ships and
barges were subject to the same control as stationary facilities
with reference to control of hydrocarbon emissions. Regulation V
requires hydrocarbon vapors to be controlled to 1.5 psia Reid vapor
pressure. This is equivalent to concentrations in the controlled
stream of about 10 percent by volume. In lieu of Texas Rule 503,
measures have been designed specifically for application to ship
and barge gasoline loading operations in Texas. During unloading
operations from a ship or barge to an on shore storage tank, pipe-
line, or other unit, controls are required by Texas Regulation V
(results in over 90 percent recovery}. The loading measure regulation
requires that, whenever a ship or barge is loaded with gasoline,
the emissions of total hydrocarbon vapor must never exceed 5% by
volume. In general, this requires that the compartment being
6-11
-------
filled must be connected to a vapor recovery system. The vapor
recovery system is required to have the capability of reducing the
total hydrocarbon emissions to the emission limit. Some of the
reduction methods that could be applied are refrigeration, incin-
eration, or absorption. This control measure is expected to accomplish
a 58% reduction in the ship and barge uncontrolled emissions. This
value was based on data submitted by the Exxon Company, Baytown,
Texas (Ref. 6-3) for Houston which shows that 90% of the emissions
from all ship and barge loading operations in Houston/Galveston are
due to gasoline. Refrigeration control equipment is presently
available for control to 1.5 percent concentration (at 110 degrees
Fahrenheit) and to 5.0 percent concentration Cat -70 degrees
Fahrenheit). Expected reductions in emissions which would result
from this measure are shown in Tables 6-1 through 6-3.
6.6 Solvent Control (Degreasi'ng)
Degreasing is a term broadly applied to the use of various
organic solvents to dissolve and remove grease and soils from metal
objects. If the solvent is near or at room temperature, the process
is termed "cold solvent cleaning"; if the solvent is maintained at
its boiling point, the process is referred to as "vapor degreasing."
This measure is designed to limit the reactive hydrocarbon
emissions from degreasing operations. Emissions from degreasing
operations must be controlled by at least 85%. Operations using
perch!oroethylene and saturated halogenated hydrocarbons are
specifically exempt, as are operations emitting less than specified
minimal amounts of organic compounds C8 pounds per hour and 40
6-12
-------
pounds per day). The regulation for control can be applied in the
Dallas-Fort Worth, Houston-Galveston, and San Antonio areas where
degreasing operations have been identified as sources of reactive
hydrocarbon emissions. Expected reductions for these three areas
are shown in tables 6-1 through 6-3.
6.7 Inspection-Maintenance of Light-Duty Vehicles
Inspection/maintenance programs aim at reducing emissions from
in-use vehicles by ensuring that the emission levels of those
vehicles are not permitted to deteriorate due to inadequate or
improper maintenance. The programs include two phases: (1) An
inspection phase used to screen the vehicle population to determine
which vehicles have excessive emissions and require maintenance;
(2) a maintenance phase in which appropriate corrective maintenance
to achieve low emissions is performed on vehicles with excessive
emissions which have failed to pass the inspection test. Recent
studies have shown that mandatory inspection-maintenance programs
can result in significant reductions in light-duty vehicle emissions
(Ref. 6-1). The effectiveness of a program depends primarily upon
the type of inspection test and the fraction of the vehicle popu-
lation induced to receive corrective maintenance.
Two different types of vehicles emission tests have been used:
(1) the "idle mode" test, and (2) the "loaded mode test." The idle
mode test measures emissions while the engine is running with the
transmission in neutral such that there is no load on the engine.
The "loaded mode" test measures emissions while the engine is
running with the transmission in gear and the drive wheels trans-
6-13
-------
nritting power to a treadmill-!ike deyice called a dynamometer. A
test based on the idle procedure will give an indication of high
emissions for vehicles that are poorly maintained or have major
mechanical problems. Tests based on the loaded mode procedure
determine emissions during simulated driving conditions and in
general can single out vehicles having more sophisticated emission
related problems, since some poorly performing or malfunctioning
engines may show problems only under load. On a technical basis
the "loaded test" is more accurate and will single out more high
emission vehicles. However, it is considerably more costly to
operate, requires a large initial capital investment, and needs a
more expert operator to perform the tests. In addition, the loaded
mode test is more suited to a high volume centrally located (such
as a state inspection station) operation. The idle mode can be
either centrally located or under the operation of individually
licensed garage owners. Since Texas presently has a vehicle safety
inspection program operated by licensed privately-owned garages, it
appears more practical to use the idle mode test operated as part
of the privately licensed safety inspection program. Therefore,
the inspection/maintenance program has been set up for an idle mode
test.
The two main provisions of the inspection/maintenance program
are the requiement for emission tests using the idle mode and the
requirement for the State to set emission tests pass-fail criteria
such that an overall 8 percent reduction in light duty vehicle
exhaust hydrocarbon emissions is obtained in the first test cycle.
6-14
-------
The provision for an average of 8 percent reduction in light-
duty vehicle exhaust emissions during the first inspection cycle
requires the State to design emission test pass-fail criteria. Any
type of program which meets the 8 percent reduction will satisfy
this requirement. Typically, programs designed by other states have
allowed higher emission levels for older uncontrolled or less-
controlled vehicles and lower levels for well-controlled newer
models. For example, in New Jersey, vehicles are grouped into four
model year categories; pre-1968, 1968 to 1969, 1970 to 1974, and
1975. Hydrocarbon emission limits by category range from 1,600 to
200 parts per million, the philosophy being that a well-tuned older
vehicle (uncontrolled or partly controlled) could not meet stringent
limits for a well-tuned newer, controlled vehicle.
The estimated reductions in emissions of reactive hydrocarbons
are shown for each area in the accompanying Tables 6-1 through 6-3.
6.8 VMT Reduction Measures
Average automobile occupancy in the U.S. is about two persons
per car. Average occupancy for work trips is about 1.4 persons per
car. Since most cars are capable of carrying at least four persons,
there is considerable room for reducing automobile use and emissions
through carpooling. In order for carpooling to be successful,
carpoolers must have trip origins and destinations that are close
to one another, must desire to travel at the same times of day,
and, to minimize the problems of locating carpool partners, must
make trips that are repetitive from day to day. As a result, the
greatest potential for increased carpool use is in connection with
peak-period work trips. These trips are responsible for about 25
6-15
-------
percent of urban area automobile emissions. The present low
occupancy factor for work trips indicates that such travel is
especially conducive to carpooling, and a prime potential area for
VMT reduction if adequate incentives to encourage carpooling are
presented. Preliminary indications are that programs to encourage
carpooling should be capable of reducing total urban area automobile
emissions by 5 to 10 percent (Ref. 6-4).
The speed and reliability of bus service can be increased
substantially by allocating road facilities preferentially to buses
through the use of reserved lanes on existing highways, specially
constructed busways, or other means. Certain forms of priority
treatment can be applied to carpools so as to increase their speed
and, thereby, their attractiveness. Within the last decade, over
200 bus priority treatments have been implemented or proposed in
the United States and elsewhere (Ref. 6-5).
Examples of bus priority treatment on freeways include:
The San Bernardino Busway in the Los Angeles area and the
Shirley Busway in the Washington, D.C. area;
Contra-flow bus lanes on the Long Island Expressway (New York
City), 1-495 in New Jersey, the Southeast Expressway in Boston, and
U.S. 101 in Marin County, California;
A special bus ramp for Seattle's Blue Streak express bus
service, and the bus-carpool bypass lanes at the San Francisco-
Oakland Bay Bridge toll plaza.
Examples of priority treatment on surface streets include:
Dedicated bus streets in Chicago and Minneapolis;
6-16
-------
Contra-flow bus lanes in San Juan, Louisville, and San Antonio;
Median bus lanes in Chicago and New Orleans, and curb bus
lanes in most major cities.
Priority treatment can increase bus freeway speeds by a
factor of two or more without significantly affecting auto speeds
(Refs. 6-5, 6-6). Speed increases of up to 50 percent have resulted
from surface street priority treatment (Ref. 6-7).
The prospective measures for Texas consist of three regulations
(VMT reduction measures) aimed at reducing average daily VMT in
certain problem areas where the oxidant problem is especially acute
and where automobile transportation accounts for an appreciably
high percent of total reactive hydrocarbon emissions. These
regulations deal with employer incentive programs to encourage use
of mass transit and carpooling, preferential bus/carpool treatment,
and comrouter/carpool matching. These measures for VMT reduction
are designed to supplement and reinforce one another. Their
effectiveness derives from the three measures acting concurrently.
When fully implemented and applied as an entire group, it is
intended and anticipated that these measures would result in
areawide VMT reductions of about 7%, with a corresponding reduction
in emissions of hydrocarbons from light-duty vehicles. The expected
reductions in emissions of reactive hydrocarbons for 1977, 1980,
and 1985 due to application of this group of measures in each of
the areas where considered are shown in Tables 6-1, 6-2, and 6-3.
6.8.1. Employer Incentive Programs
These programs are designed by employers to provide incentives
to the use of mass transit and carpools by employees in their
6-17
-------
regular commuting to and from work. Under this approach, an
employer has the flexibility to develop his own program to reduce
average daily WIT applicable to a facility. The concept is based
on known existing programs which have been developed by employers
to encourage multiple-passenger automobile use in regular work-
related commuting, and thereby to conserve fuel. The employer
incentive regulation in the present action calls for a program to
reduce average daily VMT applicable to an affected facility by 15%,
or alternatively, to attain a commuter/vehicle occupancy ratio,
applicable to all commuters to the facility by private motor vehicle,
of 1.75 or greater. The measure applies to employment facilities
with 250 or more employees and to educational facilities of college
level or of vocational training above the secondary level with
1,000 or more employees and students. Based on employment data by
size of facility in five counties (Dallas, Tarrant, Harris, Galveston,
and Bexer) in which the employer incentive measure would apply, the
250-employee figure for employment facilities would result in the
following percentages of employees being included within employer
incentive programs: 57% of employees in manufacturing industries,
33% of employees in commercial activities (insurance, banking,
retail), and 44% of employees in transportation industries. For
all three types of employment (manufacturing, commercial, and
transportation), the average figure of employee coverage is 44% -
of more than one million employees included in the data on which
these figures are based (Refs. 6-8 and 6-9). With reference to
educational facilities (colleges and universities), approximately
6-18
-------
95% of employees and students are located at facilities with popu-
lation of 1,000 or more [employees and students! and would thus be
included within employer incentive programs tin the same five
counties)(Ref. 6-10). Tables 6-4, 6-5, and 6-6 show employment
data by size of facility for manufacturing, commercial, and trans-
portation industries. Table 6-7 provides population data for
educational facilities.
In the event an employer or educational institution does not
achieve either of the above goals (reduction in average daily VMT
of 15% or attainment of a commuter/vehicle occupancy ratio of 1.75
or greater) for a facility by the applicable compliance date,
certain measures must be adopted to encourage use of mass transit
and carpools. These are as follows:
(1) Providing a commuter/carpool matching program.
(2) Establishing park and ride facilities.
(3) Subsidizing the costs of carpool and transit trips by
commuters.
(4) Establishing shuttle bus service and vanpool programs.
(5) Publicizing schedules, rates, and routes of all forms
of mass transit which serve the facility.
(6) Cooperating and negotiating with authorities in charge
of mass transit for improved service to the facility.
The compliance date for a facility depends on whether it is an
employment or an educational facility, and its size. Compliance
dates extend from October 1, 1976, to December 1, 1976. The
regulation includes a requirement for reporting to the Regional
6-19
-------
s_
o cu
»-> m CT
<: CM s-
10
t/> Mi
CU O
CU "O
>, 00 C
O CU rtj
D-4-> 10
E -r- 0)
UJ i CU
U O
O to i
Z U- O.
o
o
o
in
oo
en
o v*
z cu
cu
ro o
»-> i
o o-
en
o
o
O
o
O
o
CD
as
o o ro
. o CM
LO
as
CM O O
CM O r
in
UD
N
ii
oo
ro
x
_ cu
: 4_>H-
o
o co
o i
o
8
co
o
CVJ
cu
o
e
cu
s-
OJ
o
o
CM
CO
as
CM
CO
o o
CM o
CM in
co
CM
CM
^a-
LT)
CM
r^.
1C
00
i un
10 o
co
CO
el-
O
si-
CM
CD
O
>,
c
o
S-
ro
X
cu
CO
cu
1
"o.
E
LU
O
O
c
o -o
r- CU
+J 4->
ro ro
o »
i +j
UJ
cu
cu
o
Q.
E
UJ
ro
1
0
+->
c
(U
o
S-
cu
Q.
U
t- -r-
0 r -0
i- C
(-> (J ra
c ro
CU U_ CU
to O N
CU S- +J-r-
cu cu ro oo
o oo -a
i CU CU CU
O- > CU 4-J
E -*"- >-> ro
UJ 4->
ro
O 3
Z E
o
0 +->
i 00
a.
EM-
UJ O
i.
cu
01
s_
ro
i
O i T3
r- c
i-> o ro
c ro
i CU LL. CU
to 00 O N
4-> CU S- +> !-
o cu cu ro oo
oo O oo XJ
CU t- i CU CU CU
OO-OCUOCL>CO-I-J
^-r-CU>, E-r->,roS-
CL+J-U O+->UJJ-> O4-> CU
E ro ro r c ror-oooi
UJCJEQ.CU -. a- s-
O-«-Eoo3EH-ro
I-PUJ &-2: EUJ O I
O 00 CU 3
Z UJ Q_ O
6-20
-------
xi
c
o
o
i
10
CO
S
ro
1
>_
4J
C
3
O
O
I
tf)
(U
ro
CD
0
101 1
I O 1 1
uo
"""
en
in
1 CO 1 1
i * i i
t
iii i
iii i
^ o cn cn
o vo 10
CM
s-s s^
o o o cn
IO
&« &*
co o t-~ 10
o r oo
o
CO
co m o cn
CM CM CO
LO
a« ^
CM 0 O «=T
i o i cn
cn
r^ r^ u O N
+J 4-> CU S- <-> !-
C O CU CU ro CO >i
co o l >>a- !->
Emzin o in xi c
S> C CU <4- , CU CU CO 3
OOTJCUoa.>CO-tJ O
i -r- CU >> E - >> rO S- O
O.4-* +* O 1 % LU +J O +J CL)
E rO (O i C ro *~" LO cn 10
LU O_J i.
Z LU O. U 3C
in
1 CM 1 1
1 10 1 1
cn
CO
m
cn
1 CO 1 1
CM
10
1
iii i
iii i
tf- O IO IO
i O CM CM
O
CM
&S fcS
cn o co cn
CM m i ro
,_r
CM
&« S«
cn in 10 in
10 CM r in
r- C
+J 0 (0
C ro
t CULL. CU
ro in o N
+J -tJ CO J. +J -i- >)
C O CU CO ro in 4->
tO t >>Q- C
in z to o in x) 3
c cot-. cocucu o
^° > E -i- >> ro S-
E ro *O * C rO r ry> C7> C
0"-EOO*3Et-ro i-
o *r
C O CU CO ro t/>
cu o i >>a.
E in z in O in xi
>>C CO D 4->
i M- CU>> E'"->>roS-
CL+J -MO4->UJ4->O4->CU
Ero rOi C ro< c/)cn
LU O EQ.CU.r- O. S-
O r-EOO3Evt-ro
_l MLUS-ZELUO-J
o in
i "cx
1 O
in
1 CM
i "<-
o
1
in
1 CU
i +->
i £
1 U
ro
1 U-
l +*
I
in
1 CU
CO
i >>
o
"a.
i E
^_^LU
in
CO O
c
C_> 0
1
in
o
LO C
_J CO
f "
O CO
1 0-
6-21
-------
in i~
01 1
r- CT
[ « j^
r- re
i _i
u xi
re
u.
in >,
dj o
0) i
in
i r-.
o
CM
ID
r-.
i oo
i "
E O
LU in
CM
O <4-
Z O
re i
-)-> Q.
O
ro
vo
o
01
CM
i co
I 1
o
CO
>-
h-
_l
££
Ll_
U.
O
UJ
M
OO
>-
CO
et
1
<:
CD.
H
Z
UJ
s:
>-
0
_i
O-
E:
UJ
in
i
l£>
LU
CQ
<:
p
in
re
X
i
10
_ s
1
re
4->
cu
ce
n
i CT
re c
o -x
s- c
cu re
E CQ
O
O CU
o
c
i-
D
(/)
c
cu
4->
re
0.
in
(/i
cu
c
^
10
13
CQ
>,
C
13
o
CJ
ro
r^
en
A
ID
T
1
O
CD
t-
01
XI
c
ro oo CM o
in oo LO o
«3- CD
in o
d- O
o
LO
sf
1 LO
=f
CM
CM
in i-.
T ^~
*
r-
CM
i ro
in i~^
0 ^
*> *
en co
CO
- o
o
o
CM O
r- O
i CM
CO UD
r CO
CD
O
in
c m
o cu
r QJ
4-3 ^) *
ro o o
<-) r- z
O CL
_l E i
LU re
c . o
to i
»)
O XI O
i CU
CX+J 4->
E re c
LU E CU
.££
O tfl Q)
Z UJ CL
O 4->
i re
CLi
iS|
o
i_
cu
in en
cu S- in
+j i o
r- J
in xi c
CU CU =3
CU -M O
>i ro <->
O !->
Q. re
EM- i
LU O <
0 4J
i re
S-
cu
(/I CO
cu s-
^_> 1
1^
i XI
r- C
o re
re
U_ CO
Nl
-M -i-
ro oo
I/) XI
cu cu
CU (->
O -t-J
CL
in
c
o
4_>
ro
u
o
c
o
c~
O
tft
cu
E "= E 1- >
LU E UJ O c
o
ro
es
i
4_3
C
CU
E
O
f»
E
UJ
o
2:
CO
cu
cu
0 0
CL
E i
LU ro
+J
o
0 1
z
XI O
cu
re c
E cu
r- O
+-> i-
l/> CU
LU a.
i^
o
c:
cu
o
i.
cu
a.
in
cu cu
cu >
O 4->
i re
CLi
E =
LU E
t_>
S-
0)
in en
cu s-
(- re
!-> 1
^ ~^}
r- C
u re
re
U- CU
N
*-> -f-
ro oo
cu
>> ro
O J->
r 00
CL
E M-
LU O
6-22
-------
o
in
j
10
CD
>!
+->
C
3
O
0
t/l
S-
10
in
i co i i
1 « 1 1
CM
ID
en
en
00
i id i i
1 00 1 1
S-3 &«
t--. o in in
** O CM CM
O
r-T
&S &9
r- in OO CO
si- r-~ co
CO
s
in
en in en CM
en CM i in
i oo
^
CO
*« s*
r*. en oo o
10 en ^ o
in 10 r
ft ft
o en
oo
i.
cu
(/) C7>
c/) 9) i-
C _l
i- 0) H- -r-
4-> >, . O i -0
to O O -i- C
CJ i z 4-> O ro
O O. C f8
1 E 0) U. CU
LU ro U IM
4J 4J S- 40 -i- >,
C O CU «JCO 4->
cu o t a. c
e z: w ui -a 3
>, <»- cu cu cu cu o
OT3OCU+J O
r- CU >> -i- >» ro
Q.4-> 4_> o +-> O 4-> 4->
E ro Cr ro i CO C
LU E CU O-r O. 1-UJ E UJ O S-
O
U) CU S-
C > Oi TJ
ro O O -i- C
O Z +> O ro
O Q. c as
_!£ cuu- eu
LU ro U N
+J 4-> S_ 4-> -r-
C O CU 1C CO
to i a.
ts ui ~o
<*- eu cu cu eu
O "O O CU > CU +>
i CU >vr- >, rO
Q.+J +J o 4-> O 4->
E <0 Ci rOr CO
LU E CU OL.I Q.
<- O E 3 E M-
. 4J S- LU £ UJ O
o 10 eu 3
z: LU o. o
o
o
in
*
in
in
r
CO
CO
«* 1
r»-
CO 1
in" i
1
1
1
1
1
1
1
1
1
1
1
1 J_
1 CU
cr>
1 i.
ra
I _1
1 T3
c
1 ro
cu
1 CU
1 O
i a.
1 UJ
1 O
in
1 CM
1 t-
o
1
1 CU
i £>
1 1
r-
1 U
ro
1 U-
1 4->
1
1 CU
cu
"o.
I E
LU
1
*^~» o
z:
eu
4J ro
3 O
0 1
(^
in o
Xv*
+->
co c:
_l
-------
oo s-
0) CU
r- Ol
M S-
U -O
ra C
U. OS
+J U)
10 <1J
03
,
cu o
Q.
E O
LU LO
O il-
ls: O
LO
ro
i *
i OJ
00
O CU
Z 0)
8?
r
O
LU
l-sl
§
00
C
ra
i-
01
c
3
O
O
OJ O
o
o
O
CVIO
o
CM
OJ
LO LO r
I OJ i
Vi
in tn co
r~- ro
co
CL
5
8
<3- O
LO
CO
OJ
- O OJ
)LO r
cn
LO LO
r^
00
CO
et
co ^r
co r-
oo crv
S
o
SCO
vo
co r~
i.
cu
00 O>
00 _J
.^ CU t- -
+J >> Or- -0
CO O O -i- C
U r Z -M 0 CO
o o. c: ra
1 E r- CU Lu CU
LU rO U N
+J -P £- 4-> -r-
C O CU ro CO
£
o
o
S-
ro
X
ai
CO
, H- CO Q) CO CO
o -o o cu >
r >!- >, CO
Q-4-> +J O +J O 4J
E ro C i ro i CO
LU £ CU Q.I Q.
r- 0 E 3 E <»-
4-> &- LU E LU O
O oo OJ 3
Z LU Q- CJ
00
c
o
4^
CO
U
O
_J
>J
c
o
CJ
oo
ra
CO
c
0
a.
E
LLJ
.
0
z
00
cu
cu
J^J?
"a.
LU "ro
0
0 1
M-
^ °
+-> !->
ro c
E cu
r= 0
oo cu
LU CL,
M-
o
+J
c
i-
cu
oo cn
CU i-
r- ro
r
r -O
r- C
(J ra
ra
OO
C 00
O CO
r- CU
t-1 >>
,
(J CM +J
S.
cu
CL.
cu
0 -M
r- ra
O.r-
E 3
LU S
o
ra co
« -a
cu cu
cu +>
>> (0
0 +J
r- CO
a.
E "4-
LU O
c
o
o
c
o
00
cu
ra
0
JL§
C
(U O
1*
r CU
CL4->
E ca
LU g
+J
O 00
Z LU
O
CO
4J
O
t-
t-
o
!->
CU
e
cu
a.
00
cu
cu
o
"a.
LU
t-
ai
oo cn
CO S-
i- tO
M _i
0 ^ T3
r- C
+-> U ra
C ra
CU LL. CU
U N
S- +J !-
0) ro OO
CL,
oo "O
CU QJ CU
> , CO
4-> O 4->
ra r OO
i Q.
3 E M-
E LU 0
CJ
6-24
-------
8
VO
CM
LO
CO
O
LO
r
°
TJ
4J
<§
VO
VO
Ol
c
o
o
in
si
S-
ns
o
VO
*e s«
en o CM CM
r O CO CO
O
Ol
%^ ^^
o o CM «*
CM O i *
LO
^
CO O 1^ r
LO LO i VO
'
O
0
CO 00 O"> O
CO CO CO O
0 VO i
i CO
CM
5_
Ol
en en
t/1 01 S-
C l/t !- fO
O Ol 4-> _J
+->>! O i T3
IO O O -i- C
O Z 4-> O IO
O Q. C 10
_J E i 0) U. 0)
UJ IO U N
C O O) IO CO 4-a
>- >> 10
Q.4-> 4J O -^ O 4-* -^
ElOCi LO C
UJ E Ol CLr- Q. TO
r- U E 3 E 1- 1-
+J S- LU E LU O S-
O ul O> 3 IO
Z LU O. O H-
,^
B9 &«
^O^ si-
0 CO CO
^% %^
i LO co r~-
r- co
CO
O O LO CM
r- LO 1 LO
f-
,
10 en co o
r^ sf « o
CM LO r
LO
S-
o>
(A cn
in Ol s_
c in -r- 10
O Ol (->_!
^ Ol <*--
*J >> O i T3
IO O O - C
O i Z 4-> O IO
O 0. C IO
_l E Ol Ll- Ol
LU IO U N
+J +J S- 4-> -r-
C O Ol IO (/)
tO 1 Q-
z in in -o
V)_ Ol 01 0) 0)
O T3 O Ol > Ol 4J
1 Ol >)!- >) 10 i I/)
LU E Ol O.I Q.
r- U E 3 E "<-
4-> S- UJ E UJ O
O Ol 3
Z LU CX O
01 1
CO
r- 1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
i.
1 01
cn
1 S-
IQ
i _j
1 T3
c
1 IO
1 l/l
01
1 Ol
1 O
I CL
1 LU
1 0
LO
1 CM
1 t-
O
in
i Ol
1 4->
, ^
1 U
IO
1 U.
10
1
in
1 Ol
Ol
i >,
o
1 1
Q.
'LU
i
^^ o
v> Z
Ol
4-* fO
3 O
01-
M-
m o
+->
to c
1 01
-------
TABLE 6-7. STUDENT AND FACULTY POPULATION AT EDUCATIONAL FACILITIES
Reference: Texas Almanac, 1974-1975
Student and Faculty
Population at Educational
Total Student and Facilities with 1,000 or
Faculty Population More Students and Faculty
Bexar County
(7 Colleges and Universities)
Dallas County
(10 Colleges and Universities)
Galveston County
(2 Colleges and Universities)
Harris County
(13 Colleges and Universities)
Tarrant County
(7 Colleges and Universities)
TOTALS (5 Counties)
32,844
40,087
2,622
56,636
39,406
171,595
32,146
37,178
1,651
53,489
38,876
163,340
Percent of Total Student and Faculty Population at Facilities
of 1,000 or More Students and Faculty 95%
6-26
-------
Administrator on the incentive program within one month after the
compliance date, and annually thereafter. For a facility which has
not achieved either goal stipulated above by the applicable com-
pliance date, the employer or educational institution is required
to report to the Regional Administrator with a complete description
of all measures which it has applied in the incentive program,
indicating the consideration given to measures which it has not
applied, and providing an explanation of the probable reasons why
neither of the goals has been accomplished.
6.8.2. Preferential Bus/Carpool Treatment
This consists of measures such as exclusive bus/carpool lanes
designed to speed and facilitate the flow of bus and carpool
traffic, and thereby make such forms of transportation more attractive
to commuters. The present action requires selected cities to
develop and implement a plan to provide preferential bus/carpool
treatment and to reduce average daily VMT by 8% from that existing
during the base period of July 1 - September 30, 1974. Such measures
may consist of any one or combination of the following:
(1) Exclusive bus/carpool lanes or other preferential
bus/carpool treatment on main thoroughfares.
(2) Park and ride facilities.
(3) Dial-a-bus system.
(4) Restricted use of ramps leading to and from thoroughfares.
(5) Restriction of on-street parking in congested areas.
(6) Vehicle-free or bus-only zones.
(7) Other measures designed to reduce average daily VMT.
6-27
-------
Each measure may be restricted to certain hours and/or days of
heaviest traffic flow. For the purpose of determining compliance
with the required reduction in average daily VMT, comparison of the
predicted average daily VMT for the period of March 1 - May 31,
1977, is to be made with average daily VMT for the base period.
The plan is to be fully implemented by May 31, 1977.
A report describing the plan and the measures to which consid-
eration was given, indicating those measures to be instituted, and
containing estimates of the VMT reductions expected, is to be
submitted to the Regional Administrator by May 31, 1976. The
report is to contain compliance schedules, with the compliance date
for each measure to be not later than May 31, 1977. Following
receipt of the report, the Regional Administrator shall make a
determination of the approvability of the plan by July 31, 1976,
and may require additional measures if he determines that the city
is not utilizing all feasible measures to meet the required reduction
in average daily VMT. He may also require revisions in a compliance
schedule if he determines that implementation of a measure will not
be made as expeditiously as practicable.
Each affected city is to strive for cooperation and joint
effort by other cities and governmental entities adjacent to its
boundaries, to encourage such cities and entities to take such
action as may directly improve the effectiveness of the measures to
be implemented. Each affected city also is to publicize its plan
of preferential bus/carpool treatment and VMT reduction so as to
gain the understanding, cooperation, and support of the public.
6-28
-------
The cities to which this regulation applies have already completed
elaborate transportation and traffic studies, the results of which
may be utilized in considering and adopting measures for preferential
bus/carpool treatment. Measures contained in any city plan not
yet placed in effect may be part of the plan presented to comply
with this regulation. In estimating city-wide VMT reductions, the
affected cities should consider the reductions in VMT which are
expected due to application of the employer incentive regulation.
6.8.3. Commuter/Carpool Matching
The present action contains a measure for the establishment
of a commuter/carpool matching system by selected cities, to
encourage commuters to facilities in each respective city to form
and utilize carpools. The system is to be designed to facilitate
initial contact between commuters with similar travel patterns to
the same or neighboring facilities. The system is to utilize
computer processing equipment for the purpose of matching names,
travel origins and destinations, and travel times of commuters to
facilities in each affected city. The measure is considered to be
one of general support to reduction of average daily VMT in the
effected cities and, as such, is designed to serve commuters to
facilities in each affected city whether or not they are included
in employer incentive programs.
The commuter/carpool matching system for each affected city
is to provide for the following functions and services to commuters
and facilities in that city:
6-29
-------
(1) Preparation and distribution of a manual to each facility
describing the commuter/carpool matching system and providing
directions and guidance for obtaining information from commuters
on travel patterns and for the proper preparation of computer
cards to be processed in the commuter/carpool matching system.
(2) Distribution to any commuter, upon request, of directions
and forms to be completed which can be used by the city to enter
appropriate information into a computer program for the purpose of
commuter/carpool matching.
(3) Processing of any set of properly prepared computer
cards for 250 or more commuters to an employment facility, and for
1,000 or'more commuters to an educational facility, submitted by
a facility to show results of commuter/carpool matching in terms
of commuters who have similar travel patterns.
(4) Processing of properly prepared information submitted by
any commuter to show results of commuter/carpool matching.
(5) Distribution of results of commuter/carpool matching to
each facility which submitted a set of properly prepared computer
cards.
(6) Distribution of results of commuter/carpool matching to
any commuter who submitted properly prepared information.
(7) Technical assistance on request of a facility with
respect to initiating and maintaining a carpool program and to
improving its effectiveness.
6-30
-------
The final compliance date for implementation of the commuter/
carpool matching system by each affected city is July 1, 1976,
with a final report on the system and the status of compliance due
one month thereafter.
6-31
-------
Section 6 References
6-1 Control Stategies for In-use Vehicles,, U.S. Environmental
Protection Agency, Mobile Source Pollution Control Program,
Washington, D.C. 20460, November 1972.
6-2 EPA Report 450/375-046-a, A Study of Vapor Control Methods
for Gasoline Marketing Operations; Volume I - Industry
Survey and Control Techniques, April 1975.
6-3 Exxon Company, USA, Letter to Texas Air Control Board, from
J. M. Johnson, dated June 9, 1975.
6-4 Transportation Controls to Reduct Automobile Use and Improve
Air Quality in Cities. U.S. Environmental Protection Agency,
Office of Air and Waste Management, Washington, D.C. 20460,
November 1974.
6-5 Bus Use of Highways: State of the Art, Report No. 143,
National Cooperative Highway Research Program, Highway
Research Board, 1973.
6-6 Gerald K. Miller and Keith M. Goodman, The Shirley Highway
Express-Bus-on-Freeway Demonstration Project-First Year
Results, Interm Report No. 2, prepared for the Urban Mass
Transportation Administration, November 1972.
6-7 Downtown Distribution Plan San Bernardino Freeway Express
Busway, prepared by Wilbur Smith and Associates for the
Southern California Rapid Transit District, March 1973.
6-32
-------
6-8 Directory of Texas Manufacturers, 1973, Bureau of Business
Research, University of Texas at Austin.
6~9 1973 County Business Patterns, Texas, U.S. Department of
Commerce, Bureau of the Census.
6-10 Texas Almanac, 1974-75. Dallas Morning News.
6-33
-------
7.0 Recommended Control Measures
In the EPA restudy of the hydrocarbon/photochemical oxidant
control plan for Texas, the Agency reviewed the specific strategies
promulgated on November 6, 1973, and considered some additional
measures not used in the original plan. , Jn trying to determine
what control measures were needed versus the reasonableness of the
measures and their socio-economic impact, a general priority
ranking was set up according to the following scheme: Application
of existing controls was ranked at the top; extension of existing
stationary source controls to areas not covered by the State-
submitted plan (April 15, 1973) was considered next; subsequently
additional source controls were applied; this was followed by
moderate controls on motor vehicles; and finally some methods
which could have significant socio-economic impact were examined
but not applied. Added to the November 6, 1973 promulgation is a
proposal to control hydrocarbon vapor losses from storage of crude
oil. Substantial reductions of hydrocarbons can be obtained in
some areas by this additional measure. Measures included in the
original plan that are not recommended for the present strategy
are light-duty vehicle retrofit, limitation of new reactive carbon
compound emission sources, gasoline limitations, and parking
management.
Although the recommended amendments to the plan would take no
specific action on existing controls the effects of existing
controls were considered in development of the plan. Existing
controls consist of Texas Regulation V for control of volatile
7-1
-------
carbon compounds from stationary sources in specified Texas counties
and control of motor vehicle emissions as required by the Federal
Motor Vehicle Control Program. These two categories, contribute
the greatest to hydrocarbon reductions in each area in the Texas
plan.
It is also worthy of note that all of the measures being
proposed are conservative of petrochemical products in addition to
reducing emissions of hydrocarbons to the atmosphere. A consider-
ation has been given to the socio-economic impact of both the
recommended and other potential hydrocarbon emission reduction
measures. None of the measures presently being recommended are
considered to have a disruptive socio-economic impact. In selecting
the recommended measures, the highest priority was put on stationary
source controls with mobile source controls coming next. Measures
recommended include control of crude oil vapor losses, control of
gasoline marketing vapor losses, control of solvent (degreasing)
vapor losses, control of gasoline vapor losses from loading of
ships and barges, light-duty vehicle inspection and maintenance,
and some measures to reduce vehicle miles traveled.
The recommended measures for each area in Texas are listed in
Tables 7-1A and 7-1B. Table 7-1A presents the EPA Long Range
Oxidant Plan and includes the minimum reasonable measures that are
recommended for ultimate application in each of the seven Texas
oxidant problem areas. From the list of minimum reasonable measures
given in Table 7-1A, another list of measures has been chosen for
7-2
-------
o
c <->
(O C
CO <
I X X I
c
(O
c
o
C 4->
O to
4_> CO
00 >
3 i
O
Q U.
XXX
X X
S-
o
10
"o
+J
o
o
to
c
o
(U
o
(U
o
o
a)
Q£
I
I
3 CO
O.T-
s- s-
o ^:
<_3 O
t.
3
M ^:
C 4J
O S-
(O
OJ -M
CQ Q.
XXX
I I
X X X X
I X X
1 1 1 1
I/)
CO CO r
>fc
s-
03
C"
0
+3
oj
+j
co
f
03
O
i
-P
^
"O
-o
> CO CO O
O3 03 S-
. 4J +J 4J CO
COCO CO C t/> CO
to
^^
0
i.
4_>
c
o
o
-r- 4->
O !- «i- CO C to CO
»->*-> S- O (O r
CCOCOOSCJCOiO
O ^ ^ CQ S- -r- i.
r- i- S- -t-> COO 4->
to - -00
)-> to to M- i 3
X OJ OJ ^1 O S-
LU CD CO CO CO O
CO
O
S-
3
O
co
CO
^~
^
r.(^
o
r
03
O
r-
+J
i
-o
-o
CO
o
c:
03
C
+J
C
I
03
s: to
1 CO
C S-
O 3
i- tO
4-> 03
O U CO
s- cos:
-M a.
c to i
o c 2:
O M >
-o
OJ
to
o
Q.
o
Q.
O)
3
to
to
II
X
CU
to
o
Q.
O
CL
7-3
-------
c
o
c. -u
(O C
CO <
I LU I
co
i co
c
o
c
(O
to
CU
>
3 i
o TO
31 CJ3
CO
CO
LU CO
c
IB
x
o
S-
CU
O
(/I
r (O
LU O.
I 00
co
X
Ol
s-
o
CD O.
co co i i
CO
I CO I I
>,
s-
to
c
o
r
(0
+->
CO
1
n
c
o
r
4^
i
-o
>
CT>
i
e
+j
d
o
o
cu
o
S-
3
O
CO
cu
ce.
M-
o
c
o
1
to
c
cu
+>
X
LU
cu
cn
to
±}
CO
1
cn
c
r-
^ *
CU
*/
S-
(O
s:
to
(T3
05
CU
0>
rd
4_>
co
i
i
cn
c
r
M
CU
s-
<0
s:
to
c
o
o
cu
en
i_
c
o
o
^^^
cn
c
to
CO
1
1
(-> cno
oB
Q.
f
£Z
CO
c
cu
>
p
o
CO
cu
o
cu
o
3
s-
CJ
r_
o
S-
I-)
c
0
o
cu
o
3
O
CO
cu
cu
o
c
(O
c
cu
I CU
c t-
O 3
I- tO
+J tO
u cu
cu 2:
Q.
-»->
fO
4->
CO
"O
cu
to
o
0.
o
O-
O)
3
to
<0
OJ
II
CO
o
O)
to
o
Q.
O
Q.
to
a;
o
o
to
+>
c
o
o
C
(O
JO
o
cu
-4->
re
-(->
co
*
7-4
-------
implementation now as an interim plan. These interim measures are
presented in Table 7-1B. The implementation responsibilities
shown for TACB in Table 7-1B are based on a resolution by the TACB
to adopt such measures. In most Texas areas, additional controls
beyond those recommended in either Table 7-1A or Table 7-1B will
have to be implemented before the control plan can provide for air
quality levels even approaching the standard and before the plan
can be considered final. This approach will result in a series of
promulgated regulations. The interim control measures will comprise
the first set of regulations in the series, the remaining long range
measures will comprise the second set of regulations, and regulations
requiring other control measures will be added as identified.
Figures 7-1 through 7-7 show the expected effects of the
existing and long range controls on air quality in each of the
areas from the baseline year through 1985. Detailed discussions
of the interim and long range measures chosen and their effects in
each area are presented in Sections 7.1 through 7.7. The cost
effectiveness of the plan is discussed in Section 7.8.
7.1 Austin Area Recommended Controls
The additional long range controls recommended for the Austin
area consist of requirements for vapor recovery for gasoline
transfers to storage tanks (Stage I) and to vehicle tanks (Stage II).
The effects of these measures are presented in Table 7-2A. As
shown in the table, there will remain further reductions required
to meet the standard even in 1985, but substantial progress toward
the required 50% reduction will be made.
7-5
-------
0.3
El
-J
k.
i
0.2
I
O
1 °-«
I
.160
LEGEND
li'ililil INDUSTRY
TRANSPORTATION
| PREDICTED OXIDANT LEVELS WITH EXISTING CONTROLS,
(PREDICTED OXIDANT LEVELS WITH EXISTING CONTROLS
I PLUS ADDITIONAL REASONABLE CONTROLS.
^H0.08 ppm STANDARD
.127
.113
.105
.093
1973
BASELINE YEAR
1977
1980
1985
FIG. 7-} OXIDANT LEVELS 1973-1985 (AUSTIN)
7-6
-------
0,4
1
Q.
0.3
0.2
a:
o
0.1
.325
i i
LEGEND
11,'ij11 INDUSTRY
TRANSPORTATION
HJ PREDICTED OXIDANT LEVELS WITH EXISTING CONTROLS,
m PREDICTED OXIDANT LEVELS WITH EXISTING CONTROLS
&&IPLUS ADDITIONAL REASONABLE CONTROLS.
H^IO.08 ppm STANDARD
.113
114
1973
BASELINE YEAR
1977
I960
1985
FIG. 7-2 OXIDANT LEVELS 1973-1985 (BEAUMONT/PORT ARTHUR)
7-7
-------
0.3
el
i
§
i
0.2
O.I
si
i
.184
LEGEND
Ei'iy.1 INDUSTRY
TRANSPORTATION
Hgg PREDICTED OXIDANT LEVELS WITH EXISTING CONTROLS.
f*m PREDICTED OXIDANT LEVELS WITH EXISTING CONTROLS
LMjPLUS ADDITIONAL REASONABLE CONTROLS.
\>4 0.08 ppm STANDARD
.113
.120
971
BASELINE YEAR
1977 1980 1985
FIG. 7-3 OXIDANT LEVELS 1971-1985 (CORPUS CHRISTI)
7-8
-------
0.3
0.2
Q
§
8
1
I
I
.187
LEGEND
|;|;|;i;| INDUSTRY
TRANSPORTATION
PREDICTED OXIDANT LEVELS WITH EXISTING CONTROLS.
mm PREDICTED OXIDANT LEVELS WITH EXISTING CONTROLS
fcg&l PLUS ADDITIONAL REASONABLE CONTROLS.
I 4 0.08 ppm STANDARD
.162
.132
1974
BASELINE YEAR
1977
1980
1985
FIG. 7-4 OXIDANT LEVELS 1974-1985 (DALLAS/FORT WORTH)
7-9
-------
LEGEND
0.3
Q.
Q.
ti
i
§
0.2
I
§ a.
.130
I INDUSTRY
TRANSPORTATION
IHH PREDICTED OXIDANT LEVELS WITH EXISTING CONTROLS
^m PREDICTED OXIDANT LEVELS WITH EXISTING CONTROLS
PLUS ADDITIONAL REASONABLE CONTROLS.
0.08 ppm STANDARD
110
.101
,.098
092
.088
1974
BASELINE YEAR
1977 1980
FIG. 7-5 OXIOANT LEVELS 1974-1985 (EL PASO)
1985
7-10
-------
0.3
Ui
-J
0.2
§
i
01
LEGEND
FijTjT] INDUSTRY
TRANSPORTATION
.234
[PREDICTED OXIDANT LEVELS WITH EXISTING CONTROLS.
E?m PREDICTED OXIDANT LEVELS WITH EXISTING CONTROLS
fiffiiflPLUS ADDITIONAL REASONABLE CONTROLS.
t 4 0.08 ppm STANDARD
.187
.174
.153
1974
BASELINE YEAR
1977
I98O
1985
FIG. 7-6 OXIDANT LEVELS 1974-1985 (HOUSTON/GALVESTON5
7-11
-------
0.3
ti
I
I
UJ
I
I
I
0.2
O.I
LEGEND
Fliil INDUSTRY
TRANSPORTATION
HH! PREDICTED OXIDANT LEVELS WITH EXISTING CONTROLS.
"***"
mm PREDICTED OXIDANT LEVELS WITH EXISTING CONTROLS
EilSiiJ PLUS ADDITIONAL REASONABLE CONTROLS.
LJo.06 ppm STANDARD
.145
.121
.III
.105
1971
BASEUNE YEAR
1977
1980
1985
FIG. 7-7 OXIDANT LEVELS 1971 -1985 (SAN ANTONIO)
7-12
-------
Table 7-2A--Reactive Hydrocarbon Emission Reductions under
Recommended Long Range Strategy for Austin Area
1977 1980 1985
tons/yr (%) tons/yr (%) tons/yr (%)
Present Controls
Plus Growth 3,426 (20.5) 4,923 (29.6) 5,706 (34.3)
Stationary Sources
Gasoline Marketing
Stage I 410 (2.5) 452 (2.7) 529 (3.2)
Stage II 483 (2.9) 532 (3.2) 623 (3.7)
Crude Oil Storage
Control
Ship & Barge Vapor
Recovery
Solvent Control
(degreasing)
Mobile Sources
LDV Inspection/
Maintenance
VMT Reductions
Total Reductions 4,319 (26.6) 5,907 (35.4) 6,858 (41.2)
Deficit 4,010 (24.0) 2,422 (14.6) 1,471 ( 8.8)
Austin Area Counties: Travis and Hays
7-13
-------
The requirement for gasoline marketing vapor recovery was
recommended because this was the only additional stationary source
control available in Austin and since this control provided more
reductions than any of the other possible measures. This control
is considered by EPA to be one of the more cost-effective ways of
providing hydrocarbon emission reductions needed beyond those
provided by the existing stationary and mobile source controls.
Vapor recovery for gasoline marketing operations is in general
being adopted nation-wide in areas with oxidant problems. EPA
will be proposing these regulations for all areas in the State of
Texas needing additional hydrocarbon reductions. However, the
proposal for Stage II in Austin and in four other areas less
severely polluted than Dallas and Houston will not be part of the
interim measures proposed in the initial plan amendments. Stage II
for Dallas and Houston will be proposed as part of the interim
plan in the initial amendments. Stage II in Austin, Beaumont,
Corpus Christi, El Paso, and San Antonio will be proposed in the
second series of regulations which will complete the Long Range
Plan. Table 7-2B shows the effects of the interim plan on emissions
in Austin.
Two additional control measures, Inspection/Maintenance (I/M)
and VMT reductions, were available to apply in these areas, but
were not included as part of either the Long Range or Interim
control strategy. I/M will definitely be needed in this area due
to the large contribution of mobile sources to the hydrocarbon
7-14
-------
Table 7-2B--Reactive Hydrocarbon Emission Reductions under
Recommended Interim Strategy for Austin Area
1977 1980 1985
tons/yr (%) tons/yr (%) tons/yr (%)
Present Controls
Plus Growth 3,426 (20.5) 4,923 (29.6) 5,706 (34.3)
Stationary Sources
Gasoline Marketing
Stage I 260 (1.6) 287 (1.7) 336 (2.0)
Stage II --
Crude Oil Storage
Control
Ship & Barge Vapor
Recovery
Solvent Control
(degreasing)
Mobile Sources
LDV Inspection/
Maintenance
VMT Reductions -- --
Total Reductions 3,686 (22.1) 5,210 (31.3) 6,042 (36.3)
Deficit 4,643 (27.9) 3,119 (18.7) 2,287 (13.7)
Austin Area Counties: Travis and Hays
7-15
-------
inventory. I/M would provide substantial reductions as well as
ensuring that the reductions assumed from controls installed on
new cars as part of the Federal Motor Vehicle Control Program are
realized. However, until the State has developed an I/M program
for the more severely polluted areas of Dallas/Fort Worth and
Houston, I/M in Austin is not recommended as a part of the proposed
control strategy. First priority should be given to developing
the I/M program for the two most heavily polluted areas and then
the strategy can be modified to include the other areas of the
State, like Austin, where the automobile is causing the major
portion of the oxidant problem. VMT reduction measures were not
applied to Austin, since it is doubtful that a city of this size
would accomplish significant VMT reductions without substantial
social disruption.
7.2 Beaumont/Port Arthur Area Recommended Controls
The additional long range controls recommended for the Beaumont/
Port Arthur area consist of extension of TACB Regulation V to one
additional county, gasoline marketing vapor recovery, crude oil
storage controls, and vapor recovery for ship and barge loading
operations. The effects of the long range measures are presented
in Table 7-3A. As shown in the table there will be further
reductions required in 1977 and slightly more in 1985 to meet the
standard. Substantial progress should be made toward attainment of
the standard with the expected accomplishment of 69% reduction out
of a required reduction of 75%.
7-16
-------
Table 7-3A--Reactive Hydrocarbon Emission Reductions under
Recommended Long Range Strategy for Beaumont-Port Arthur Area
1977 1980 1985
tons/yr (%) tons/yr (%) tons/yr (%)
Present Controls
Plus Growth 115,628 (65.3) 115,108 (65.0) 114,284 (64.5)
Stationary Sources
Extension of Reg. V 332 (0.2) 332 (0.2) 332 (0.2)
Gasoline Marketing
Stage I 341 (0.2) 425 (0.2) 496 (0.3)
Stage II 402 (0.2) 499 (0.3) 584 (0.3)
Crude Oil Storage
Control 5,235 (3.0) 5,235 (3.0) 5,235 (3.0)
Ship & Barge Vapor
Recovery 1,070 (0.6) 1,178 (0.7) 1,340 (0.8)
Solvent Control
(degreasing)
Mobile Sources
LDV Inspection/
Maintenance
VMT Reductions
Total Reductions 123,008 (69.5) 122,777 (69.4) 122,271 (69.2)
Deficit 9,679 (5.5) 9,910 (5.6) 10,416 (5.8)
Beaumont-Port Arthur Area Counties: Hardin, Jefferson, and Orange
7-17
-------
Since the primary source of hydrocarbon emissions in the
Beaumont/Port Arther area is from stationary sources, all available
stationary source controls have been recommended for this area.
The mobile source controls, I/M and VMT reduction, have not been
applied as part of either the Long Range or Interim Plan primarily
due to the fact that mobile sources are not a major contributor to
the oxidant problem in this area. In addition, it is doubtful
that cities the size of Beaumont, Port Arthur, or Orange could
accomplish significant VMT reductions without substantial social
disruption. I/M may later be proposed for this area after the
State has developed a system for Dallas/Fort Worth and Houston.
Stage II vapor recovery and ship and barge vapor recovery
will not be part of the Interim Plan, but will be proposed with
the second series of regulations in the Long Range Plan. Table 7-
3B shows the effects of the interim plan on emissions in Beaumont/
Port Arthur.
7.3 Corpus Christi Recommended Controls
The additional long range controls recommended for the Corpus
Christi area consist of gasoline marketing vapor recovery, crude
oil storage controls, and vapor recovery for ship and barge loading
operations. The effects of these measures are presented in Table
7-4A. As shown in the table, there will remain substantial reductions
required in 1977 as well as in 1985. Substantial progress will be
made with the expected reductions of 40% out of a required reduction
of 57%.
7-18
-------
Table 7-3B--Reactive Hydrocarbon Emission Reductions under
Recomnended Interim Strategy for Beaumont-Port Arthur Area
1977 1980 1985
tons/yr (%) tons/yr (%) tons/yr (%)
Present Controls
Plus Growth 115,628(65.3) 115,108(65.0)114,284(64.5)
Stationary Sources
Extension of Reg. V 332 (0.2) 332 (0.2) 332 (0.2)
v»
Gasoline Marketing
Stage I 217 (0.1) 270 (0.2) 314 (0.2)
Stage II
Crude Oil Storage
Control 5,235 (3.0) 5,235 (3.0) 5,235 (3.0)
Ship & Barge Vapor
Recovery
Solvent Control
(degreasing)
Mobile Sources
LDV Inspection/
Maintenance
VMT Reductions
Total Reductions 121,412 (68.6) 120,945 (68.4) 120,165 (67.9)
Deficit 11,275 (6.4) 11,742 (6.6) 12,522 (7.1)
Beaumont-Port Arthur Area Counties: Hardin, Jefferson, and Orange
7-19
-------
Table 7-4A--Reactive Hydrocarbon Emission Reductions under
Recommended Long Range Strategy for Corpus Christi Area
1977 1980 1985
tons/yr(%) tons/yr (%) tons/yr (%)
Present Controls
Plus Growth 28,860 (36.6) 30,251 (38.4) 27,483 (34.8)
Stationary Sources
Extension of Reg. V
Gasoline Marketing
Stage I 307 (0.4) 339 (0.4) 397 (0.5)
Stage II 362 (0.5) 399 (0.5) 468 (0.6)
Crude Oil Storage
Control 1,555 (2.0) 1,555 (2.0) 1,555 (2.0)
Ship & Barge Vapor
Recovery 365 (0.5) 401 (0.5) 458 (0.6)
Solvent Control
(degreasing)
Mobile Sources
LDV Inspection/
Maintenance
VMT Reductions
Total Reductions 31,449 (39.9) 32,945 (41.7) 30,361 (38.5)
Deficit 13,503 (17.1) 12,007 (15.3) 14,591 (18.5)
Corpus Christi Area Counties: Nueces and San Patricio
7-20
-------
As was the case in Beaumont/Port Arthur, the main source of
hydrocarbon emissions in Corpus Christi is from stationary sources.
For this reason all available stationary source control measures
have been applied. Mobile source controls have not been applied
as part of either the Long Range or Interim Plan due to small
reductions from this source and because significant VMT reductions
measures would probably not be accomplished without social disruption.
I/M may later be applied to the area,after the State has developed
such programs for Dallas/Fort Worth, Houston, and San Antonio.
Stage II vapor recovery and ship and barge vapor recovery
will not be part of the Interim Plan, but will be proposed with
the second series of regulations in the Long Range Plan. Table 7-
4B shows the effects of the interim plan on emissions in Corpus
Christi.
7.4 Dallas/Fort Worth Recommended Controls
The additional controls recommended for the Dallas/Fort Worth
area consist of extending TACB Regulation V to one additional
county, vapor recovery for gasoline marketing, crude oil storage
controls, solvent controls for degreasing, I/M, and VMT reductions.
The Long Range and Interim Plans for Dallas/Fort Worth are identical.
The effects of these measures are presented in Table 7-5. Although
significant reductions are obtained, the table shows that there
will still be substantial reductions required even in 1985 in
order to attain the standard.
Essentially all available controls have been applied to the
Dallas/Fort Worth area due to the severity of the oxidant problem
7-21
-------
Table 7-4B--Reactive Hydrocarbon Emission Reductions under
Recommended Interim Stategy for Corpus Christi Area
1977 1980 1985
tons/yr (%} tons/yr (^T tons/yr (%)
Present Controls
Plus Growth 28,860 (36.6) 30,251 (38.4) 27,483 (34.8)
Stationary Sources
Extension of Reg. V
Gasoline Marketing
Stage I 195 (0.2) 215 (0.3) 252 (0.3)
Stage II
Crude Oil Storage
Control 1,555 (2.0) 1,555 (2.0) 1,555 (2.0)
Ship & Barge Vapor
Recovery
Solvent Control
(degreasing)
Mobile Sources
LDV Inspection/
Maintenance
VMT Reductions
Total Reductions 30,610 (38.8) 32,021 (40.6) 29,290 (37.1)
Deficit 14,342 (18.2) 12,931 (16.4) 15,662 (19.9)
Corpus Christi Area Counties: Nueces and San Patricio
7-22
-------
Table 7-5--Reactive Hydrocarbon Emission Reductions under
Recommended Long Range and Interim Strategy for Dallas/Ft. Worth Area
1977 1980 1985
tons/yr (%) tons/yr (%) tons/yr (%)
Present Controls
Plus Growth 7,686 (7.3) 14,238 (13.6) 14,936 (14.1)
Stationary Sources
Extension of Reg. V 2,052 (1.9) 2,052 (1.9) 2,052 (1.9)
Gasoline Marketing
Stage I 2,828 (2.7) 3,134 (3.0) 3,725 (3.6)
Stage II 3,331 (3.2) 3,692 (3.5) 4,384 (4.2)
Crude Oil Storage
Control 331 (0.3) 331 (0.3) 331 (0.3)
Ship & Barge Vapor
Recovery -- --
Solvent Control
(degreasing) 1,217 (1.2) 1,497 (1.4) 2,036 (1.9)
Mobile Sources
LDV Inspection/
Maintenance 2,659 (2.5) 1,990 (1.9) 1,556 (1.5)
VMT Reductions (7%) 3,906 (3.7) 3,273 (3.1) 2,845 (2.7)
Total Reductions 24,010 (22.9) 30,207 (28.7) 31,865 (30.3)
Deficit 36,012 (34.1) 29,815 (28.3) 28,157 (26.7)
Dallas-Ft. Worth Area Counties: Coll in, Dallas, Denton, Ellis, Hood,
Johnson, Kaufman, Parker, Rockwall, Tarrant, and Wise
7-23
-------
in the area. Due to the large deficit of reductions, even after
application of the recommended controls, additional control
measures beyond those recommended here will probably have to be
developed and implemented in order to allow attainment of the
oxidant standard.
7.5 El Paso Area Recommended Controls
The additional long range controls recommended for the El
Paso area consist of vapor recovery for gasoline marketing, and
crude oil storage controls. The effects of these measures are
presented in Table 7-6A. The reductions shown are based on appli-
cation of the regulation in El Paso only, but the calculated
percent reductions are based on the total area emissions from both
El Paso and Ciudad Juarez. As shown in the table, substantial
reductions are obtained in emissions between 1977 and 1985, but
there will remain further reductions required in 1985 to allow
standard attainment. It should be noted that the projected
percent reductions for El Paso/Juarez are probably the most
uncertain of all seven Texas areas evaluated due to the rough
approximation made of Ciudad Juarez emissions. A more detailed
inventory for Ciudad Juarez emissions could result in significant
changes in the expected reductions, particularly in 1980 and 1985.
Although the crude oil storage control regulation was applied
to the El Paso area, to be consistent with the strategies for the
rest of the State, no reductions were credited to this regulation
because the TACB inventory shows all sources in this area to be
already controlled. Gasoline marketing vapor recovery was applied
7-24
-------
Table 7-6A--Reactive Hydrocarbon Emission Reductions under
Recommended Long Range Strategy for El Paso Area (Juarez Included)
1977 1980 1985
tons/yr (%) tons/yr (%) tons/yr (%)
Present Controls
Plus Growth 4,335 (13.2) 7,419 (22.6) 9,759 (29.6)
Stationary Sources
Extension of Reg. V
Gasoline Marketing ""
Stage I 313 (0.9) 347 (1.0) 412 (1.2)
Stage II 368 (2.3) 408 (2.5) 485 (3.0)
Crude Oil Storage
Control
Ship & Barge Vapor
Recovery
Solvent Control
(degreasing)
Mobile Sources
VMT Reductions
Total Reductions 5,016 (15.2) 8,174 (24.7) 10,656 (32.2)
Deficit 7,529 (22.8) 4,371 (13.3) 1,889 (5.8)
El Paso Area (Juarez included) County: El Paso
7-25
-------
because this is the only other available stationary source control
in El Paso, the control provides more reductions than any of the
other possible measures, and the control is being applied statewide
in areas with oxidant problems. However, the proposal for Stage
II vapor recovery will not be part of the interim measures proposed
in the initial plan amendments, but will be proposed with the
second series of regulations in the Long Range Plan. Table 7-6B
shows the effects of the interim plan on emissions in El Paso.
Two additional control measures, I/M and VMT reductions were
considered available to apply in this area, but were not included
as part of the control strategy. I/M is definitely needed in this
area due to the large contribution of mobile sources and would
provide significant hydrocarbon as well as carbon monoxide reductions,
However, until the State has developed an I/M program for the more
severely polluted areas of Dallas/Fort Worth and Houston, I/M is
not recommended to be apart of El Paso's control strategy. First
priority should be given to the heavily polluted areas and then
the strategy modified to include other areas of the State like El
Paso with less severe problems. VMT reduction measures were
similarly not applied to El Paso nor in Corpus Christi, Beaumont,
and Austin since it is doubtful whether significant reductions
could be accomplished without substantial social disruptions.
7.6 Houston/Galveston Recommended Controls
The additional controls recommended for the Houston/Galveston
area consist of gasoline marketing vapor recovery, crude oil
storage controls, ship and barge vapor recovery, controls for
7-26
-------
Table 7-6B--Reactive Hydrocarbon Emission Reductions under
Recommended Interim Strategy for El Paso Area (Juarez Included)
1977 1980 1985
tons/yr (%) tons/yr (%) tons/yr (%)
Present Controls
Plus Growth 4,335 (13.2) 7,419 (22.6) 9,759 (29.6)
Stationary Sources
Gasoline Marketing
Stage I 198 (0.6)** 220 (0.7) 261 (0.8)
Stage II
Crude Oil Storage
Control
Ship & Barge Vapor
Recovery
Solvent Control
(degreasing)
Mobile Sources
VMT Reductions
Total Reductions 4,533 (13.7) 7,639 (23.1) 10,020 (30.4)
Deficit 8,012 (24.3) 4,906 (U.X9) 2,525 (7.6)
El Paso Area (Juarez included) County: El Paso
7-27
-------
solvent degreasing, inspection/maintenance, and VMT reductions.
The Long Range and Interim Plans for Houston are identical. The
effects of these controls are presented in Table 7-7. Although
significant reductions will be obtained, the table shows that
there will still be substantial reductions required in order to
attain the standard.
Essentially all reasonably available controls have been
applied to the Houston/Galveston area due to the severity of the
oxidant problem there. Due to the large deficit of reductions,
even after applications of the recommended controls, additional
controls beyond those recommended here will probably have to be
developed and implemented in order to allow attainment of the
oxidant standard.
7.7 San Antonio Area Recommended Controls
The additional long range controls recommended for the San
Antonio area consist of gasoline marketing vapor recovery, solvent
control for degreasing, inspection maintenance, and VMT reductions.
The effects of these controls are presented in Table 7-8A. As
shown in the table there will remain further reductions required
to meet the standard even in 1985, but most of the required 45%
reduction will be accomplished.
Stage II vapor recovery and inspection maintenance will not
be part of the Interim Plan, but will be proposed with the second
series of regulations in the Long Range Plan. Table 7-8B shows
the effects of the interim plan on emissions in San Antonio.
7-28
-------
Table 7-7--Reactive Hydrocarbon Emission Reduction under
Recommended Long Range and Interim Strategy for
Houston-Galveston Area
1977 1980 1985
tons/yr {%) tons/yr (%) tons/yr (%)
Present Controls
Plus Growth 81,571 (26.8) 78,536 (25.8) 61,427 (20.2)
Stationary Sources
Extension of Reg. V
Gasoline Marketing
Stage I 2,551 (0.8) 2,810 (0.9) 3,286 (1.1)
Phase II 3,003 (1.0) 3,309 (1.1) 3,866 (1.3)
Crude Oil Storage
Control 12,292 (4.0) 12,292 (4.0) 12,292 (4.0)
Ship & Barge Vapor
Recovery 1,724 (0.6) 1,895 (0.6) 2,158 (0.7)
Solvent Control
(degreasing) 1,217 (0.4) 1,497 (0.5) 2,081 (0.7)
Mobile Sources
LDV Inspection/
Maintenance 2,636 (0.9) 1,918 (0.6) 1,397 (0.5)
VMT Reductions (7%) 3.872 (1.3) 3,155 (1.1) 2,553 (0.8)
Total Reductions 108,866 (35.7) 105,548 (34.6) 89,216 (29.3)
Deficit 92,101 (30.3) 95,419 (31.4) 112,751 (36.7)
Houston-Galveston Area Counties: Brazoria, Chambers, Fort Bend, Galveston,
Harris, Liberty, Matagorda, Montgomery, and Waller
7-29
-------
Table 7-8A--Reactive Hydrocarbon Emission Reductions under
Recommended Long Range Strategy for San Antonio Area
1977 1980 1985
tons/yr (%) tons/yr (%) tons/yr (%)
Present Controls
Plus Growth 6,159 (16.5) 8,727 (23.4) 10,225 (27.4)
Stationary Sources
Extension of Reg. V
Gasoline Marketing
Stage I 919 (2.5) 1,022 (2.7) 1,210 (3.2)
Stage II 1,082 (2.9) 1,202 (3.2) 1,425 (3.8)
Crude Oil Storage
Control
Ship & Barge Vapor
Recovery
Solvent Control
(degreasing) 932 (2.3) 1,127 (3.0) 1,510 (4.0)
Mobile Sources
LDV Inspection/
Maintenance 920 (2.4) 700 (1.9) 538 (1.4)
VMT Reductions (7%) 1,352 (3.6) 1.151 (3.1) 984 (2.6)
Total Reductions 11,364 (30.4) 13,929 (37.3) 15,892 (42.5)
Deficit 5,457 (14.6) 2,892 (7.7) 929 (2.5)
San Antonio Area Counties: Bexar, Coma! and Guadalupe
7-30
-------
Table 7-8B--Reactive Hydrocarbon Emission Reductions under
Recommended Interim Strategy for San Antonio Area
1977 1980 1985
tons/yr (%) tons/yr (%) tons/yr (%)
Present Controls
Plus Growth 6,159 (16.5) 8,727 (23.4) 10,225 (27.4)
Stationary Sources
Extension of Reg. V
Gasoline Marketing
Stage I 919 (2.5) 1,022 (2.7) 1.210 (3.2)
Stage II
Crude Oil Storage
Control
Ship & Barge Vapor
Recovery
Solvent Control
(degreasing) 932 (2.3) 1,127 (3.0) 1,510 (4.0)
Mobile Sources
LDV Inspection/
Maintenance
VMT Reductions (7%} 1,352 (3.6) 1.151 _ [3.1) _ 984 (2.6)
tal ^ducHon; 9,362 (25.0) 12,027 (32.2) 13,929 (37.3)
, . 7,^59 (20.0) 4,794 (12.8) 2,892 (7.7)
o
- ,,-, <\ntoni-.-. Area Counties Bexar, Comal and Guadalupe
V 31
-------
All reasonably available controls have been applied to the
area. Additional controls beyond those proposed here may have to
be developed and implemented in order to ensure attainment of the
oxidant standard. VMT reduction measures have been applied due to
the large contribution to the oxidant problems caused by the
autombile and due to the moderately severe oxidant problem being
observed.
7.8 Cost Effectiveness of Recommended Control Measures
7.8.1 Introduction
Both positive and negative economic impacts can be expected
from the proposed control measures in the Texas areas where they
are to be applied. On the positive side there will be reductions
in air pollution related health cost, reductions in automobile use
costs to commuters, new jobs created in implementing the controls,
and reduction in the consumption and evaporation of petroleum
products. On the negative side these control measures will
require expenditure of significant funds for installation and
operation.
Estimates of capital and operating costs for most of the long
range controls are presented in Table 7-9A. Table 7-9B summarizes
the costs for the interim controls. Also included where appli-
cable in Table 7-10 are annualized costs (direct operating expenses
plus annual capital charges, minus any savings from recovered
petroleum products) in order to better put in perspective what the
costs for improving air quality will be on an annual and unit
basis. Details of these cost estimates are discussed in the
following paragraphs for each control measure.
7-32
-------
*
in
'o
i.
p
s=
O
o
O)
CT
C
ID
a:
cr
c
o
'
T3
O)
-D
C
i
o
o
0)
a:
o
4-
O)
+J
S
-p
t/}
UJ
.!_>
(/)
o
0
«^
en
i
UJ
CO
«=C
i
00
£
1
OO
^£
UJ
a:
0
c
o
-p
c
^^
c
ID
00
C
o
C -P
O
P «^-
>*«« n | «\ f.
in CM en oo en i r~ co
«j- CM i~~ co r i en r-»
«o r» in «s- in o
CM CO i VO
oo oo o oo oo oo
oo oo o oo oo oo
i*~ ^- CO f"v f^ CD CD "^ f^O ^f CO
VOUD i co cr»i Ocvj \O r-» ^D^
C\4»O OOr-. «d-| CNJLf) CMLO ^O i
t/> cr>to ID Lnror^ covo
r^. ^j- r ro ^j~ LO ^~ vo
CM ^t
O O O O
o o o o
^ 0^ oo r^»
r^ o <^- CT>
un i r^ rv
CM r C\J
A «
^ ^*
OO OO OO O OO OO
oo oo oo o oo oo
*d~ in en en o oo r^ o 10 en o
ooo oio oooo cni r^r mo
uscn OOCM r-~ui *! i m oooo
<- r^oocn in r toco
* > * * * 1 1 1
o «3- CM ^j- r^ s
r OO r t- 1
1-
oo
HH
OO OO OO OO
OO OO OO OO 1
OOCMUOCM sj-^t-^j-io «t
ooi~-T a\*& «a^i~-T ooco uj
CM «d- co r-% in in _i
co r r u> en co ca
CM en _i
i i
>
<£
I
o
OOOOOO OOOO r
OOOOOO OOOO OO
lO^t-OOr OCM COCOCOOO O
*t«l «|1 AM ** vin (_3
OOi OCM OCh CQCM lOf^
i i locnini CMOCMOO o
oo oo CM oo o o r*^ ~z.
r ^- o ro n
00
O O O O
O 0 0 O
ID en oo r^
^j- r CM in
1 i M3 O
oo in oo
A A
i *d~
at
o
en en i* cr> en en en ro en en
cnc c c c c ccc c
-P -Pro -Pro CU-Pro i -Pco -Pro n3-Pco C4_)ro -Pro
cu T- s- T- s- Q; T- s- o- !- s- a) T- s- s_ T- s- -r- -i- s- in-i- s-
-^ Q. ai i 10.01 a. a> s_ en Q. aj cnQ.cu o n_ CD 100.01 c a. a>
S- i i t8 Q. i itOQ. S-cOQ. -PC o> o;
o ai > o a. TD o.
1/1 +J, T-3CJ1(
CD X O .C S- C Z
ID UJ OO OO O i i >
o o
o o
CM in
r CM
00 CO
CO CO
ID «^
r in
$*<*
en
c
re -P
p re
r- S-
a. ai
CD D.
0 O
(O
re
o
r-
-o
c
re
S-
iS
o
p
Ol
3
a
c
o
-p
re
re
o
ai
o
c
a
c -a
re 3
t/i o
i- c
re *r-
r C
0 0)
-a o>
JO
in
en re
"~~ c
M- UO
o r-~
en
C/) 1
s- -o
p o
c a>
O) C
s- O
re-r-
10 re
p r
01 4-
o c
O -r-
*
7-33
-------
TABLE 7-9B Cost Estimates for Recommended
Interim Control*
Measure
Gasoline Marketing
Stage I
Capital
Operating
Stage II
Capital
Operating
Extension of Reg. V
Capital
Operating
Solvent Controls
Capital
Operating
Ship and Barge
Capital
Operating
Crude Oil Storage
Capital
Operating
Inspection/Maintenance
Capital
Operating
VMT Reductions
Capital
Operating
Totals:
Capital
Operating
Estimated Cost
23,401,200
202,500
59,712,200
4,000,600
1,323,000
72,500
1,518,400
43,520,000
4,352,000
10,468,700
104,700
9,052,300
34,444,300
Not Available
$148,995,800
$ 43,176,600
*Costs are in terms of 1975 dollars and no escalation
due to inflation beyond 1975 has been included
7-34
-------
o
s
(/)
fO
X
01
I
CO
01
i.
3
CO
ro
r
r
O
+J
c
o
o
o> oo
Ol UJ
c cc.
c£ oo
Ol LLl
C S
0
_J _1
o
i- ex
O r
<+- 5
O
CO O
1 *
CO
O
O
-n
O)
N
*
r*«
ro
3
C
C
O
1
r-,
UJ
i
CD
Ol
01
oe
^ 01
C 0
o c
r- ro
O) +J C
r O
r- O. c
i^^ CO 'i"~
Ol C ro
0 01 r-
0) O
Ol (O i-
-a s- -P
3 O C
S- 4-> O
O 00 CJ
01
s_
ro
OQ >,
°3 Ol
s. >
Q. O O
i- 0. U
J= ro Ol
oo > o:
**+
CT
C
CO !-
4-> i CO
C O ro
O) S- +-> i-
r C Ol
O O CD
00 0 Q
M- S"
O
C
c o
O T-
r- +J
co ra
C r
O) 3
+> 01
x ai
11 1 c£
*~~1 ^1
lt ^
ai
ai s- >
o) o o
ro a. o
4-> ro O>
oo > a;
>.
11 i-
S- >
O) O O
ra a. u
+J ro O>
oo > o;
ai
JQ
ra
ra
<
o
z.
o
o
01
o
ro
CNJ
o -
O CO
oo en
« c
CNJ -i-
LO >
uo ro
O
O
oo
f,
oo
00
<£>
oo
ty)
o
o
vo
CVJ
oo
t/3
o
o
o
r
0
0
ID
A
Lf)
CO
oo
r--
"^
o
o
ai
f,
00
Ol
oo
OJ
ai
i^t
rO
;
ra
<
*-*
o
S- T3
ai a>
0.+J
u
01 ai
vo a.
fc£
1
1
S. <4_
a> o ai ~o
CL CO)
C !- 0
>*. O r 3
CO i O T3
O i CO O
ra ra i.
O Ol Ol Q.
1
1
1
1
i. M-
o) o ai
Q. C
C !- T3
-W. O i OJ
CO i"" O ^
oo i co E
ro ro 3
o 01 o> a.
s- t-
Ol O Ol
D. C
c -i- -a
W. O i O)
VD r O Q.
Or CO E
ra ro 3
O Ol O> D.
ai
4D
ra
J
ro
>
4-*
o
z:
to
o
o
00
o
o
o
en
oo
to-
o
=»
o
uo
f-^
CTl
uo
CVJ
CM
-a
en
ra
3
4J C
O) C
i- CO
c o
3 (J
rjj
O 3 >
co c i E
o o o ai
co t a. o;
7-35
-------
7.8.2 Cost for Gasoline Marketing Vapor Recovery (Stage I)
Capital costs per facility to install Stage I vapor recovery
equipment vary from $1,000 to $3,000 with an average cost of about
$2,000 (Ref. 7-1). There are approximately 19,000 gasoline
dispensing facilities (Refs. 7-2 and 7-3) in the areas of concern
and of these facilities the Stage I regulation would require about
57% (11,000) of the largest to install vapor recovery equipment.
The capital cost for the 11,000 affected facilities will be $25,409,000.
Also included in this cost is about $3.4 million to modify about
1,100 tank trucks for vapor recovery.
Assuming an annual capital charge of 16.3% (10 years at 10%
interest) and an operating/maintenance (0/M) cost of $20/year/facility,
the total annualized cost would be $2,893,900. On a cost per
gallon pumped basis, the cost would be 0.06
-------
Boston (Ref. 7-3), 37% of the 19,000 facilities in the affected
Texas areas are expected to require the displacement systems and
20% are expected to require the vacuum assist system. The remaining
43%, representing small facilities, will require no controls due
to the higher cost buren they would have to bear if controlled.
The capital cost for these affected facilities amounts to $84,339,200.
Assuming an annual capital charge of 16.3% (10 years at 10%
interest) and an operation/maintenance (0/M) cost of 5% of the
capital cost for displacement systems and 8% of the capital cost
for vacuum assist systems, the total annualized cost would be
$17,385,500. In terms of cost per gallon pumped at Stage II
facilities the annualized cost would be about 0.36<£ per gallon. In
terms of cost per ton reduced annually the cost would be $975/
ton. The annualized cost and unit costs all include a credit for
the 5.7 million gallons saved annually through Stage II vapor
recovery.
The responsibility for the initial capital costs for both
Stage I and Stage II vapor recovery equipment are expected to be
split in accordance with the estimated breakdown of facility
ownership shown in Table 7-11. From the table it is seen that
about 45% of the initial capital costs will be borne by the oil
companies and the remaining 55% by individual dealers, jobbers, or
commercial users.
7.8.4 Costs for Extension of TACB Regulation V to Hardin
and Tarrant Counties
7-37
-------
OJ O>
S- 4->
0) G
J3 O
.0 O
O
^> S-
O
O
LO
LO
OJ
oo
CO
r--
vo
10
A
CO
1/5
s_
0)
cu
u
fO
Dl
Q.
CO
S
0)
O
00
ro
c 01
vo
cr>
LO
o>
A
*d-
CQ
+J to
O E
d) O
3 fO
CO i
i- cn
O) OJ
JO C£.
lo
z: *->
o
o
CO
o
LO
LO
o
LO
r OJ
o
o
o
o <-
> (O
fO
(U
o;
O) C
o o
S-
en oo
ro
(U 4-*
U 0)
S- E
QJ O
OO -r-
E fO
O -(->
z: on
o
i.
CD CO
E s-
t OJ
o to
o ^
7-38
-------
Extension of Regulation V to Hardin and Tarrant Counties will
affect primarily one petroleum refinery in each county. Based on
testimony (Refs. 7-4 and 7-5) by these sources at TACB public
hearings in April 1975 capital costs, 0/M costs and annualizes
costs have been computed. Based on the company statements the
capital costs for complying with TACB Regulation V is $1,323,000.
This cost is primarily for modification to storage tanks and
loading racks at the refineries.
Assuming an annual capital charge of 16.3% (10 years at 10%)
and an 0/M cost of 1% of the capital for storage tank controls and
8% of the capital for loading rack controls, the total annualized
costs would be $288,200. The net cost with a $287,300 credit for
the 957,538 gallons/year of products recovered, is about $1,000
per year.
7.8.5 Cost for Solvent Controls (Degreasing)
Compliance with the degreasing regulation can be accomplished
by either switching to a non or low reactive solvent or by vapor
recovery systems. Since the solvent switch is by far the least
expensive method of compliance, it is expected that the bulk of
the degreasing sources will comply by this method. However, all
sources may not be able to comply by switching solvents due to the
nature of their degreasing operations and will have to use vapor
recovery techniques.
In order to develop a cost estimate for this control it is
assumed that 90% of the degreasing sources will comply by solvent
switches and 10% by vapor recovery. Using the values developed by
7-39
-------
the TACB in reference 7-6 for solvent switching and an average
vapor recovery (by adsorption) cost of $24,000 per degreaser
(Ref. 7-7) the capital costs will be $1,518,400.
Assuming that the solvent switching costs will not be amortized
(since they are a small one time expense), that the annual capital
recovery costs for the vapor recovery equipment will be at 16.3%
(10 years at 10%), and that the 0/M costs for the vapor recovery
equipment will be offset by the savings in solvent use (260,000
gal.), the annualized cost will be $326,600 for the three areas of
concern. On a cost per ton of total hydrocarbon reduced basis the
cost will be $38/ton.
7.8.6 Cost for Ship and Barge Vapor Recovery
At present there are no cost estimates for most of the
facilities that will be required to install ship and barge vapor
recovery equipment. One facility, Exxon Baytown Refinery, has
made a preliminary capital estimate for compliance with the
regulation of $12,000,000 for the refinery equipment and $4,000,000
for the ship/barge modifications (Ref. 7-8). In order to extra-
polate this estimate to all facilities in the three areas of
concern an assumption was made that the costs for facilities other
than Exxon would be proportional to the estimated emissions from
ship and barge loading operations developed in Section 5.3.2.
Using this technique for the petroleum refineries in the Houston/
Galveston area, the Beaumont/Port Arthur area, and the Corpus
Christi area results in a total capital cost of $83,123,200.
Assuming an annual capital recovery cost of 13% (15 years at
10%), an annual 0/M cost of 10% of the capital, and 30jf credit per
7-40
-------
gallon for the 1.7 million gallons of gasoline recovered annually,
the net annualized cost is $18,683,300 per year. In terms of cost
per gallon shipped by ship or barge in Texas the cost is 0.25£
gallon. In terms of cost per gallon of gasoline produced along
the Texas Gulf Coast, the cost is O.OS^/gallon. Also, in terms of
cost per ton of total hydrocarbon reduced the cost is $3,500/ton.
7.8.7 Cost for Crude Oil Storage Controls
Based on the emission inventory worksheets provided to EPA by
the TACB, there were 27 large-fixed roof crude oil storage tanks
at the Exxon Baytown Refinery in 1972. The average new tank
construction cost differential between large-fixed roof and
floating roof tanks (55 to 600 m bbl) as taken from reference 7-9
is about $88,700. By adding 25% to cover the costs due to retrofit,
the average per tank cost to retrofit the Exxon tanks becomes
about $110,900. Using these values the total capital cost for the
Exxon refinery would be $2,994,000. In order to extrapolate a
cost for the oil companies it is assumed that the costs are
proportional to the fixed roof crude oil storage tank emissions
called out on the TACB worksheets in Appendix F. Using this
method the capital costs for controlling large crude oil tanks at
refineries in Houston, Beaumont, and Corpus Christi is $10,468,700.
Assuming an annual capital recovery cost of 13% (15 years at
10%), an annual 0/M cost of 1% of the capital, and a 15?! credit
per gallon for the 13.5 million gallons recovered annually, the
net value is a cost savings of $552,300.
7-41
-------
7.8.8 Cost for Vehicle Inspection and Maintenance (I/M)
Cost estimates were made in reference 7-10 for implementation
on I/M program as a part of the existing Texas Safety Inspection
Program. Estimates were included for the expected increased
inspection fee, repair costs, and savings from improved fuel
economy. The estimates presented in reference 7-10 form the basis
for the I/M costs presented in Table 7-9A but have been adjusted to
apply only to the three Texas areas being recommended for the
program. In addition the reference 7-10 estimates were adjusted
to use a three year instead of a one year payback by inspection
stations for the emission instruments and the gasoline mileage
improvements have been increased from 1% overall to 1-1/2% (Ref. 7-11).
Using the adjusted estimates the total capital costs for I/M
in the three areas will be $10,649,700. The operating costs
(primarily vehicle repair costs) will be $40,522,600 per year.
Taking credit for the 38.4 million gallons of gasoline saved and
spreading the capital costs for instruments over three years, the
net annual costs for I/M in Texas would be $24,376,900. In terms
of cost/vehicle inspected the cost is $7.69/vehicle. In terms of
costs per ton of pollutant removed the costs is about $3,000/Ton*.
7.8.9 Costs for VMT Reduction Measures
Until specific programs are outlined by cities and employers,
cost estimates can not be made for this control.
*Note: The cost/ton are expected to be much lower than this value
(by as much as a factor of two) when the effects of higher deterioration
of cars with emission controls and of tampering are included. At
present, however, there is not sufficent data to quantify these
effects.
7-42
-------
References for Section 7.0
7- 1 Telephone conversation with E. J. Vincent, EPA Office of Air
Quality Planning and Standards.
7- 2 Texas Department of Agriculture Weights and Measures Computer
Printout (C157) of Retail Gasoline Purflps by County, dated 9/24/75.
7- 3 Arthur D. Little, Inc., Letter with Attachments dated September 4,
1975, which documents retail and non-retail gasoline outlet
distributions for Denver and Boston.
7- 4 Testimony by Mr. E. Sager representing Winstein Refining
Company, at the public hearings held by the Texas Air Control
Board in Fort Worth on April 18, 1975.
7- 5 Testimony by Mr. Jack Dana, Executive Vice President of South
Hampton Company, at the public hearings held by the Texas Air
Control Board on April 18, 1975.
7- 6 Texas Air Control Board Report No. SP-3, Documentation for Cost
Estimates for Implementing Controls Required by EPA, dated
August 4, 1975.
7- 7 EPA Report - 45/2-74-006, Systems and Costs to Control
Hydrocarbon Emissions from Stationary Sources, dated September
1974.
7- 8 Exxon Company, USA, letter dated September 12, 1975, from
Mr. J. M. Johnson on the expected capital costs of ship and
barge vapor recovery equipment.
7-43
-------
7- 9 Process Plant Contraction Estimating Standards. Volume 4,
Division 100, Published by Richardson Engineering Services,
Inc., 1975.
7-10 Report by RADIAN Corporation, A,Program for Implementation of
Vehicle Inspection/Maintenance in Four Regions of Texas.
dated August 19, 1975.
7-11 Telephone conversation on September 4, 1975 with Mr. H.
Davis, EPA Ann Arbor.
7-44
-------
APPENDICES
There are 465 pages of appendices to this report and consist
primarily of TACB data worksheets. Due to the number of pages
and the fact they are most handwritten, the appendices have not
been printed and published as a part of this document. However,
copies of the appendices will be available for public inspection
during normal business hours at the following locations:
Environmental Protection Agency
Region VI
1600 Patterson, Suite 1100
Dallas, Texas 75201
Environmental Protection Agency
Houston Facility
6608 Hornwood
Houston, Texas 77036
Environmental Protection Agency
Public Information Reference Unit
Room 2922, EPA Library
401 "M" Street S.W.
Washington, D.C. 20460
A-l
-------
TECHNICAL REPORT DATA
(Please read Instructions on the reverse before completing)
1 Rt PORT NO
.__EPA-906/9-76-001
4 ri I Ll-ANO SUII tl I I.L
2.
TECHNICAL SUPPORT DOCUMENT: HYDROCARBON/PHOTOCHEMICAL
OXIDANT CONTROL STRATEGY FOR THE STATE OF TEXAS
/ AUTHOHIS)
3. RECIPIENT'S ACCESSIOI*NO.
s. REPORT DATE
January 1976
6. PERFORMING ORGANIZATION CODE
8. PERFORMING ORGANIZATION REPORT NO.
9 PE RTORMING ORG ^.NIZATION NAME AND ADDRESS
U.S. Environmental Protection Agency
Region VI
1600 Patterson, Suite 1100
Dallas, Texas 75201
10. PROGRAM ELEMENT NO.
11. CONTRACT/GRANT NO.
12. SPONSORING AGENCY NAME AND ADDRESS
13. TYPE OF REPORT AND PERIOD COVERED
Final
14. SPONSORING AGENCY CODE
15. SUPPLtMFNTARY NOTES
16. ABSTRACT
This report presents the results of a reevaluation of the Texas Photochemical
Oxiidant Control Plan and recommends specific control strategies for each Texas
area which exceeds the oxidant standard. The purpose of the reevaluation was
to !resolve technical questions that had arisen on the original November 1973
EPA plan as a result of litigation. The reevaluation was a joint effort of
the Environmental Protection Agency, Region VI and the Texas Air Control Board.
17.
a
KEY WORDS AND DOCUMENT ANALYSIS
DESCRIPTORS
Photochemical Oxidant
Ozone
Reactive Hydrocarbons
Mobile Sources
Stationary Sources
b.IDENTIFIERS/OPEN ENDED TERMS C. COSATI Field/Group
1B [JIT, rHIBUTION STATEMENT
Releasable to public
19. SECURITY CLASS (This Report)
Unclassified
21. NO. OF PAGES
221
20 SECURITY CLASS (Thispage)
Unclassified
22. PRICE
EPA Form 2220-1 (9-73)
B-l
-------
INSTRUCTIONS
1. REPORT NUMBER
Insert the F.PA report number as it appears on the cover of the publication.
2. LEAVE BLANK
3. RECIPIENTS ACCESSION NUMBER
Reserved for use by each report recipient.
4. TITLE AND SUBTITLE
Title should indicate clearly and briefly the subject coverage of the report, and be displayed prominently. Set subtitle, if used, in smaller
type or otherwise subordinate it to main title. When a report is prepared in more than one volume, repeat the primary title, add volume
number and include subtitle for the specific title.
5. REPORT DATE
l.ach report shall carry a date indicating at least month and year. Indicate the basis on which it was selected (e.g., date of issue, date of
approval, date of preparation, etc.).
6. PERFORMING ORGANIZATION CODE
Leave blank.
7. AUTHOR(S)
Give name(s) in conventional order (John R. Doe, J. Robert Doe, etc.}. List author's affiliation if it differs from the performing organi-
zation.
8. PERFORMING ORGANIZATION REPORT NUMBER
Insert if performing organization wishes to assign this number.
S
9. PERFORMING ORGANIZATION NAME AND ADDRESS
Give name, street, city, state, and ZIP code. List no more than two levels of an organizational hirearchy.
10. PROGRAM ELEMENT NUMBER
Use the program element number under which the report was prepared. Subordinate numbers may be included in parentheses.
11. CONTRACT/GRANT NUMBER
Insert contract or grant number under which report was prepared.
12. SPONSORING AGENCY NAME AND ADDRESS
Include ZIP code.
13. TYPE OF REPORT AND PERIOD COVERED
Indicate interim final, etc., and if applicable, dates covered.
14. SPONSORING AGENCY CODE
Leave blank.
15. SUPPLEMENTARY NOTES
Enter information not included elsewhere but useful, such as: Prepared in cooperation with, Translation of, Presented at conference of,
To be published in, Supersedes, Supplements, etc.
16. ABSTRACT
Include a brief (200 words or less} factual summary of the most significant information contained in the report. If the report contains a
significant bibliography or literature survey, mention it here.
17. KEY WORDS AND DOCUMENT ANALYSIS
(a) DESCRIPTORS - Select from the Thesaurus of Engineering and Scientific Terms the proper authorized terms that identify the major
concept of the research and are sufficiently specific and precise to be used as index entries for cataloging.
(b) IDENTII IERS AND OPEN-ENDED TERMS - Use identifiers for project names, code names, equipment designators, etc. Use open-
ended terms written in descriptor form for those subjects for which no descriptor exists.
(c) COSAT1 MELD GROLP - Field and group assignments are to be taken from the 1965 COSATI Subject Category List. Since the ma-
jority of documents are multidisciplinary in nature, the Primary Field/Group assignment(s) will be specific discipline, area of human
endeavor, or type of physical object. The application(s) will be cross-referenced with secondary Field/Group assignments that will follow
the primary postmg(s).
18. DISTRIBUTION STATEMENT
Denote releasability to the public or limitation for reasons other than security for example "Release Unlimited." Cite any availability to
the public, with address and price.
19. & 20. SECURITY CLASSIFICATION
DO NOT submit classified reports to the National Technical Information service.
21. NUMBER OF PAGES
Insert the total number of pages, including this one and unnumbered pages, but exclude distribution list, if any.
22. PRICE
Insert the price set by the National Technical Information Service or the Government Printing Office, if known.
EPA Form 2220-1 (9-73) (Reverse)
------- |