EP A-600/R-99-010
February 1999
METHANE EMISSIONS FROM
THE U.S. PETROLEUM INDUSTRY
FINAL REPORT
Prepared by:
Matthew R. Harrison
Theresa M. Shires
Richard A. Baker
Christopher J. Loughran
Radian International LLC
8501 N. Mopac Blvd.
P.O. Box 201088
Austin, Texas 78720-1088
EPA Contract No. 68-D2-0160
EPA Project Officer: David A. Kirchgessner
National Risk Management Research Laboratory
Research Triangle Park, North Carolina 27711
Prepared for:
U.S. ENVIRONMENTAL PROTECTION AGENCY
Office of Research and Development
Washington, DC 20460
-------
NOTICE'
This document has been reviewed in accordance with
U.S. Environmental Protection Agency policy and
approved for publication. Mention of trade names
or commercial products does not constitute endorse-
ment or recommendation .for use.
11
-------
ABSTRACT
As concentrations of greenhouse gases increase in the atmosphere, their potential impact
on global climate has become an important issue. Although greenhouse gases, such as carbon
dioxide (C02), methane (CH4), and nitrogen oxides (NOX), occur naturally in the atmosphere,
recent attention has been focused on the increased emissions resulting from human activities.
Methane is the second largest source (after CO2) of anthropogenic greenhouse gas emissions.
Because of the radiative properties of CH4, however, it is more effective at trapping heat in the
atmosphere than CO2, and is therefore a more potent greenhouse gas. This report quantifies CH4
emissions from the U.S. petroleum industry by identifying sources of CH4 from the production,
transportation, and refining of oil. Emissions are reported for the base year 1993 and for the
years 1986 through 1992, based on adjustments to the base year calculations.
An extensive literature search identified 54 reports as having some potential applicability
for estimating CH4 emissions for the petroleum industry. Each report was reviewed and
subjectively ranked based on data quality. Only seven reports were used for this study. Methods
for estimating emissions were developed when data gaps were identified.
For the base year 1993, approximately 98 billion standard cubic feet (Bscf) ± 44% of CH4
emissions are attributed to the petroleum industry. Standard error propagation techniques were
used to determine the precision of the estimate to a 90% confidence bound.
iii
-------
TABLE OF CONTENTS
Page
ABSTRACT ............... .... ...................... . . ..... , ..... .... .......
LIST OFTABLES ..................... . , ............................. . ........ vi
LIST OF FIGURES [[[ vl±
1.0 EXECUTIVE SUMMARY ................................................ 1
2.0 INTRODUCTION ........ . ......... . ............ ... .................. . ..... 4
2.1 PROJECT STRUCTURE ........... ...... .......................... 5
2.2 LITERATURE SEARCH ........... . ....... .... ......... .... ....... 6
3.0 INDUSTRY EMISSION CHARACTERIZATION . ......... ..... ..... . ..... . ..... 8
3.1 SEGMENT DESCRIPTIONS ........ . ............. . ................. 8
3.1.1 Production .......... ......... ..... . .......... ... .......... 10
3. 1 .2 Crude Transportation .... ...... ............... .......... ..... 10
3.1.3 Refining ..... ......... ....... . ...... ........... ........... 10
3.2 EMISSION TYPES ....................... ......... .... ....... .... 10
3.2. 1 Fugitive ..... . . ...... . .......... ............ ..... . ____ .... 12
3.2.2 Vented ............. . ........... . ................ ......... 12"
3.2.3 Combusted ........... . ................................. ... 12
4.0 STATISTICAL METHODS ........ ........... . ......... .... ...... . ..... 13
4.1 PRECISION ........................ ... .......... ....... ......... 13
4.2 PROPAGATION OFERROR .......... ....... ....... ..... . . ____ .... 15
4.3 SCREENING FOR BIAS ...... . ........... .... ........... . ..... ... 18
5.0 ACTIVITY FACTORS AND EMISSION FACTORS— 1993 BASE YEAR ........... 20
5.1 ACTIVITY FACTORS - 1993 BASE YEAR .... ........ . ........ ...... 20
5.1.1 Production Rate and Well Count ...................... ..... .... 20
5.1.2 Production Equipment Extrapolations from Site Visits .............. 25
5.1.3 Miscellaneous Production Activity Factors ....................... 30
5. 1 .4 Crude Transportation ........................................ 32
5.1.5 Refining ...... .......... ..... . . ..... . ..................... 33
5.2 EMISSION FACTORS - 1993 BASE YEAR .......... ..... . ........... 36
5.2.1 Production ........ . .................. . ..... . ..... . . ....... 36
5.2.2 Crude Transportation ..................................... ... 43
5.2.3 Refining ....... .................. . ............ . ........... 43
6.0 RESULTS— 1993 BASE YEAR ............... . ...... . ..... . ...... . ..... .... 46
-------
TABLE OF CONTENTS (Continued)
Page
6,2 CRUDE TRANSPORTATION 48
6,3 REFINING ,....,,, 50
6.4 TOTAL INDUSTRY 52
7.0 METHANE EMISSIONS—1986-1992 54
7.1 METHOD FOR HISTORICAL ESTIMATES 54
7.2 RESULTS 56
8.0 CONCLUSIONS 57
9.0 FUTURE EFFORTS 60
9.1 PRODUCTION SEGMENT KEY ISSUES AND ASSUMPTIONS 60
9.2 CRUDE TRANSPORTATION SEGMENT KEY ISSUES AND
ASSUMPTIONS 62
9.3 REFINING SEGMENT KEY ISSUES AND ASSUMPTIONS 63
9.4 RECOMMENDED FUTURE TEST PLAN 63
9,4.1 Production Segment Improvements 64
9.4.2 Crude Transportation Segment Improvements 66
9.4.3 Refining Segment Improvements 66
10.0 REFERENCES 67
APPENDIX A - Results of Literature Search A-l
APPENDIX B - Site Data B-l
APPENDIX C - Statistical Analysis . C-l
APPENDIX D - Hypothetical Examples of Extrapolation and Bias D-l
APPENDIX E - Production Fugitive Emission Factors E-l
APPENDIX F - Tank Flash Calculations F-l
APPENDIX G - English/Metric Conversions G-l
APPENDIX H - Methane Emissions 1986-1992 H-l
-------
LIST OF TABLES
Page
1-1 1993 Methane Emissions From The U.S. Petroleum Industry 2
2-1 Ranking Of Emission Factor Data Quality 6
2-2 Ranking Of Activity Factor Data Quality 7
4-1 Error Propagation For Addition 16
4-2 Error Propagation For Multiplication 16
4-3 Error Propagation For Division 17
4-4 Values For Example Calculation 17
4-5 Errors For Addition, Multiplication, And Division Of Example Problem 17
5-1 1993 Activity Factors By Segment 21
5-2 Ratio Of Heavy Crude Production to Total Crude Production 24
5-3 Ratio Of Heavy Crude Production Wells To Total Crude Production Wells 25
5-4 Extrapolated Activity Factor Development 27
5-5 Production Extrapolation Parameter Selection 28
5-6 Comparison Of Sample Set To National Values 29
5-7 Final Production Developed Activity Factors 30
5-8 Refinery Throughputs 33
5-9 1993 Refinery Energy Requirements 35
5-10 1993 Refinery Fuel Consumption 35
5-11 Emission Factor Summary 37
6-1 1993 Methane Emissions Estimate Petroleum—Production 47
6-2 1993 Methane Emission Estimate Petroleum—Crude Transportation 49
6-3 1993 Methane Emission Estimate Petroleum—Refining 51
6-4 1993 Petroleum Methane Emission Estimate—Total Of Three Industry Segments 52
7-1 Refinery Engine Activity Factor For 1986-1993 55
7-2 Emission Summary For 1986-1993 56
8-1 Annual Methane Emission Estimates For U.S. Petroleum Industry From Five Different
Studies (Bscf) 58
VI
-------
LIST OF FIGURES
Page
1-1 Sources of Methane Emissions (Including Results From This Study) ..,.,,., 2
2-1 Sources of Methane Emissions (Previous Studies) , 4
3-1 Petroleum Industry Segments 9
3-2 Industry Boundaries 11
6-1 Production Segment Largest Emission Sources 46
6-2 Crude Transportation Segment Largest Emission Sources 48
6-3 Refining Segment Largest Emission Sources , .50
6-4 Emissions by Type 52
6-5 Percent Emissions by Segment 53
8-1 Sources of Anthropogenic Methane Emissions (Updated) , 57
Vll
-------
1.0 EXECUTIVE SUMMARY
As concentrations of greenhouse gases increase in the atmosphere, their potential impact
on global climate has become an important issue. Although greenhouse gases, such as carbon
dioxide (CO2), methane (CH4), and nitrogen oxides (NOX) occur naturally in the atmosphere,
recent attention has been focused on the increased emissions resulting from human activities.
Methane is the second largest source (after CO2) of anthropogenic greenhouse gas
emissions.1 Because of the radiative properties of methane, however, it is more effective at
trapping heat in the atmosphere than carbon dioxide, and is therefore a more potent greenhouse
gas. U.S. anthropogenic methane emissions have three principal sources: emissions from the fuel
cycle of fossil fuels (from production through end use of natural gas, coal, and oil), landfills, and
livestock. This report estimates methane emissions from the U.S. petroleum industry by
identifying sources of methane emissions from the production, transportation, and refining of oil.
Emissions are reported for the base year 1993 and for the years 1986 through 1992, based on
adjustments to the base year calculations.
The goal of this report was to identify the relative magnitude of the emissions from the
petroleum industry, and to identify the likely major sources in the industry. A driving force for
this report was the detailed analysis presented in the report, Methane Emissions from the Natural
Gas Industry.2 The natural gas industry study measured and analyzed methane emissions at an
equipment level of detail, and therefore was more accurate than previous approximations for the
gas industry. Although that report set a precedent of detail and accuracy, the scope of this
preliminary estimate for the petroleum industry was more rudimentary.
The estimated magnitude of petroleum industry emissions presented in this report meets
the initial objectives of a multi-phase approach. This Phase 1 report is limited to analysis of
existing data and studies, and gathered no new field data. Since some of the existing data are
extracted from other industries or have other limitations, the estimates produced in this Phase 1
report should be used only as a guideline for future efforts. Subsequent efforts, which have not
yet been initiated, will further refine the estimate by gathering segment activity factors and
directly measuring petroleum segment field data based on a statistically representative sampling
approach.
This Phase 1 project used the latest available data from published reports and site
measurement efforts. An extensive literature search identified 54 reports as having some
potential applicability for estimating methane emissions for the petroleum industry. Each report
was reviewed and subjectively ranked based on data quality. Only seven reports from the initial
literature search were used for this study. Methods for estimating emissions were developed
when gaps were identified.
This report estimates that 98 billion standard cubic feet (Bscf) of methane emissions are
attributed to the petroleum industry for the base year 1993. This estimate is believed to be
accurate to approximately +/- 100%. While precision of the estimate for 90% confidence bounds
1
-------
was calculated to be only +/- 44% (see Section 6.0), there may be some unqualified biases
resulting from use of the limited data set. Possible contributors to bias are listed in Section 4.3
and Section 9.0 of this report. These biases can be ruled out or corrected in future efforts.
The relative emissions from each segment of the petroleum industry considered in this
study are shown in Table 1-1. The production segment accounts for the majority of methane
emissions. Its largest sources are oil tank venting, pneumatic devices, chemical injection pumps,
and fugitive emissions from large compressors.
TABLE 1-1. 1993 METHANE EMISSIONS FROM THE U.S. PETROLEUM INDUSTRY
Segment
Annual Emissions, Bscf
Production
Crude Transportation
Refining
TOTAL
87 ±48%
1.4 ±85%
9.2 ± 69%
98 ± 44%
Figure 1-1 illustrates how these emissions compare with other anthropogenic sources of
methane emissions in the United States.
Livestock Manure
8%
Fossil Fuel
Consumption
3%
Natural Gas Systems
19%
Other 1%
Coal Mining
13%
Oil Systems
6%
Landfills
31%
Rice Cultivation
1%
Domesticated
Livestock
18%
Figure 1-1. Sources of Methane Emissions
(Including Results from This Study)
-------
In general, previous emission studies for the petroleum industry underestimated total
emissions since they did not include all emission sources. This Environmental Protection
Agency (EPA) study strove to examine all likely methane emission sources and produce initial
estimates for those sources. Statistical analysis of precision is also attempted based on available
data. Additional field data gathering, field measurement programs, and data analysis in later
phases will improve the estimate and reduce potential biases. Key assumptions and data issues
are discussed in this report, and recommendations for future updates are provided.
-------
2.0 INTRODUCTION
Greenhouse gases allow solar radiant energy to pass through the atmosphere to be
absorbed by the Earth's surface, but, due to their radiative-forcing properties, trap in the lower
atmosphere much of the radiant heat emitted from the surface back toward space. The portion of
the energy that is absorbed by the greenhouse gases warms the Earth's surface, creating what is
called the "natural greenhouse effect."1 Naturally occurring greenhouse gases include water
vapor, carbon dioxide (CO2), methane (CH4), nitrous oxide (N2O), and ozone (O3). The current
scientific debate surrounding the greenhouse gas effect on global temperatures focuses on how
sensitive the Earth's climate is to anthropogenic greenhouse gas emissions (those resulting from
human activities),3 On the basis of the belief that greenhouse gas emissions from anthropogenic
activities are contributing to global climate changes, over 133 countries have signed an
agreement under the 1987 Montreal Protocol to work towards limiting climate change and its
effects.1
Energy related activities are the most significant source of U.S. anthropogenic greenhouse
gas emissions, accounting for 88 percent of total U.S. emissions annually on a carbon equivalent
basis.1'4 Atmospheric methane is second only to carbon dioxide as an anthropogenic source of
greenhouse gas emissions. However, a molecule of methane contributes more than a molecule of
carbon dioxide because it is more effective at trapping heat. Sources of anthropogenic methane
emissions include landfills, agricultural activities, fossil fuel combustion, coal mining,
wastewater treatment, and the production and processing of natural gas and oil. Figure 2-1
shows a breakdown of the methane emissions according to a study of greenhouse gas emissions
by the EPA and updated to reflect results from a Gas Research Institute (GRI) and EPA's Office
of Research and Development (EPA-ORD) study on methane emissions from the natural gas
industry1
Livestock Manure
8%
Fossil Fuel
Consumption
3%
Natural Gas Systems
19%
Other 1%
Coal Mining
Landfills
33%
Oil Systems
1%
Domesticated
Livestock
19%
Rice Cultivation
2%
Figure 2-1. Sources of Methane Emissions (Previous Studies)
-------
The purpose of this study was to begin the process of detailed quantification of methane
emissions from the petroleum industry. This was accomplished by providing a level of
magnitude estimate of emissions based on the sum of initial estimates for all likely equipment
sources. Subsequent phases of this study, which have not yet been initiated, will further refine
the estimate by gathering segment activity factors and directly measuring petroleum segment
field data. When all phases are complete, the petroleum industry emission estimate will have a
level of detail that will complement a similar 1996 study on methane emissions from the natural
gas industry conducted by the GRI and EPA-ORD.2
This report presents initial estimates of the methane emissions that result from the field
production, transportation, and refining sectors of the petroleum industry in the United States.
Estimates for the years 1986 through 1993 are shown. This project identifies existing data and
uses those data with extrapolation techniques to estimate U.S. petroleum industry methane
emissions.
This project used data from several existing studies on methane emissions, including
those from: 1) American Petroleum Institute (API);5'6 2) the EPA Office of Air and Radiation
(OAR);4 and 3) the GRI methane project for the natural gas industry.2 The data from the
GRJ/EPA natural gas study, when combined with the data from final phases of this project, will
form a detailed emission inventory for methane from the oil and gas industry as a whole.
The two EPA studies that previously presented petroleum industry methane emissions
data are "Anthropogenic Methane Emissions in the United States—Estimates for 1990"4 and
"Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-1994."1 Results of these reports
are widely published and will be presented and analyzed further in Section 8.
2.1 PROJECT STRUCTURE
This project began with an identification of previous studies on methane emissions from
the petroleum industry. A detailed literature search was performed to identify all sources of
information on methane emissions related to this subject. Once gathered, these studies were
compared in order to determine industry boundary definitions, detail level, representativeness,
comprehensiveness, and data quality. Section 2,2 briefly describes the results of the literature
search and project ranking techniques. Appendix A provides extensive detail on the literature
search.
Section 3 of this report is the industry emission characterization. Descriptions of three
segments of the petroleum industry are presented: production, crude transportation, and refining.
Three different emission types are also discussed: fugitive, vented, and combusted. This
characterization allows a comprehensive structure for the emission estimate that identifies all
potential sources.
Section 4 presents the statistical methods used for this study. Standard error propagation
techniques were used to determine the overall accuracy and precision of the estimate.
-------
Section 5 presents the methods selected to compute methane emissions from the
petroleum industry and the results of the estimate. Section 5.1 summarizes the activity factors
used for the 1993 base year and Section 5.2 presents the 1993 emission factors. The total
methane emission estimate for 1993 is presented in Section 6.
Section 7 presents historical estimates for methane emissions. Estimates for the years
1986-1992 were made by modifying the activity factors from the 1993 base year. Emission
factors were assumed to have remained unchanged over the 1986-1993 time period.
Conclusions from the emission estimates are discussed in Section 8. Section 9 presents
potential future efforts for this type of study. Uncertainties in the 1993 estimate were analyzed to
identify gaps or uncertainties in the data. These weaknesses could be strengthened in the future
through more accurate measurement or research efforts. Section 10 presents a list of references
cited in this report,
2.2 LITERATURE SEARCH
A comprehensive literature search was performed at the start of this project. The purpose
of this search was to determine what type of information was available from previous studies
conducted on the topic of methane emissions from the petroleum industry. A total of 54 reports
were identified from the literature search and reviewed. Results of this search can be found in
Appendix A,
The methodologies used by each previous study for the estimation of activity factors
(equipment populations) and emission factors (average emission rate per equipment type) were
evaluated according to a scale developed for this project. Each reference was subjectively ranked
using generally accepted data quality guidelines to determine the detail level and applicability of
emission factors and activity factors. Tables 2-1 and 2-2 show the ranking values used by this
project.
TABLE 2-1. RANKING OF EMISSION FACTOR DATA QUALITY
DETAIL LEVEL FOR EMISSION FACTORS
EMISSION FACTOR Entire Industry
DATA QUALITY Equipment Level Process Unit Level Segment
Measurements best good not applicable
Field data and very good reasonable not applicable
calculations
Miscellaneous data taken unknown unknown unknown
from other reports
Estimate poor poor worst
-------
TABLE 2-2. RANKING OF ACTIVITY FACTOR DATA QUALITY
DETAIL LEVEL FOR ACTIVITY FACTORS
ACTIVITY FACTOR DATA
QUALITY
Nationally tracked and reported,
well known
Extrapolated from samples/field
data
Miscellaneous data taken from
other reports
Estimate
Equipment
Counts"
best
very good
unknown
pooi-
Process Unit
Activity Data*1
good
reasonable
unknown
poor
Entire Industry
Activity Factors"
not applicable
not applicable
unknown
worst
"Equipment Counts (Counts of specific equipment and/or detailed activities)
bProcess Unit Activity Data (based on unit counts and feed rates)
cEntire Industry AF (based on total oil produced or refined)
These tables, which were developed for this project, present a matrix scale of data quality
for emission factors and activity factors, respectively. Data quality is a function of the detail
level of the calculations and the basis for the emission factors. Emission factors can be
determined from broad estimates, data-based estimates, or field emission measurements. The
method used varied in each segment of the petroleum industry on the basis of available
information and the nature of the segment itself. The tables show a matrix of data quality
ranging from worst to best. The matrix is based on a scale of increasing level of detail. In these
tables, "worst" indicates that the emission or activity factor estimate is from poor or incomplete
background information. "Best" indicates that scientifically valid equipment-level measurements
were performed for the emission factor, and that the equipment-level activity factor is based on a
documented nationally tracked source. For both tables, "unknown" indicates that a ranking could
not be estimated since no documentation was provided.
Of the 54 reports identified from the literature search, seven were used in the emissions
estimate. The remaining reports were determined to be either potentially applicable to specific
emission sources, but with much uncertainty, or were not applicable to this study. Of the
54 reports, none met all of the criteria established for data quality. One-third of the reports were
based on data collected before 1985; none of the reports addressed all of the industry segments of
interest or presented emission data for all sources of interest. On the basis of the results of the
literature review, the project scope shifted from compiling existing emission inventories to
focusing on developing an emission estimate.
-------
3.0 INDUSTRY EMISSION CHARACTERIZATION
The first step in estimating methane emissions from the U.S. petroleum industry is to
identify and characterize each emission source within the industry. This will ensure that all
significant sources are included. To characterize the industry completely, sources were defined
by industry segment and emission type.
The next step is to determine the method to estimate emissions. If emissions could be
sampled from every source in the petroleum industry, then the total national emissions would be
the sum of every source. Unfortunately, because of the size of the industry, measuring emissions
from every source is impractical. Therefore, a method of extrapolating the sampled emissions
from a representative set of sources within the industry is necessary. The activity factor (AF)
extrapolation method was used for this purpose.
The AF extrapolation method is used to scale-up the average annual emissions from a
source to represent the entire emissions from the national population of similar sources in the
industry. The method uses emission factors (EFs) and AFs to do this. An EF for a source
category is a measure of the average annual emissions per source (e.g., emission rate per
equipment or per activity). The EF is a summation of all measured or calculated emissions from
sampled sources divided by the total number of sources in the category that was sampled. AFs
are estimated populations of equipment or estimated frequencies of activities. The national AF is
the total number of sources in the entire target population or source category. An AF is usually
presented as an equipment count, but a few exceptions exist, such as hp-hrs for compressors,
petroleum production rates or throughputs, and events per year for maintenance activities. The
EF and AF are defined so that their product equals the total annual nationwide emission estimate
from a specific source in the petroleum industry. This relationship is shown in the equation
below:
EFS x AFi = AE, (1)
where: I = source type, and
AE; = annual emissions from source type I.
3.1 SEGMENT DESCRIPTIONS
The petroleum industry can be broken down into the following distinct segments for
emission estimates: production, crude transportation, refining, product transportation, and end
use. This study's scope is limited to the first three segments of the industry. Figure 3-1
represents a simplified conceptual diagram of the five industry segments.
-------
Production
End Use
Product Transport
Foreign Oil
Figure 3-1. Petroleum Industry Segments
-------
3.1.1 Production
The production segment covers the exploration and extraction of petroleum from
underground resources in the United States. It does not include foreign production of oil that is
imported to the United States, but does include all U.S. well and surface production equipment
and storage tanks. Because oil and gas can be produced from the same well, the production
segment presents some interesting boundary issues. Some oil well equipment, such as
compressors used to transport natural gas to sales, may be related solely to natural gas production
and should not, for the purposes of this study, be part of the petroleum industry.
Figure 3-2 shows the petroleum sector boundary definitions as defined in the GRI/EPA
project.7 The GRI/EPA study of methane emissions from the natural gas industry is the only
report that deals with the production boundary issue at the equipment level. The present EPA
report elected to remain consistent with the boundaries selected in the earlier GRI/EPA project.
3.1.2 Crude Transportation
The crude transportation segment covers all movement of crude from the production
segment to refineries. Crude transportation includes all truck, marine, rail, and pipeline
transportation of crude; loading and unloading of tank trucks, rail cars, and marine vessels; and
all emissions associated with pipeline terminals and pump stations. It also includes the
transportation of crude oil imported into the United States.
3.1.3 Refining
The refining segment includes all refinery sites that take in crude and produce finished
products such as gasoline. Refining volumes include imported crude oil. Refining includes
crude storage, all refinery units, and finished product tanks. Aromatics and isomerization
processes in refineries also are included. The refinery boundary, however, excludes the
downstream chemical plant operations such as steam cracking ethylene plants, plastic/rubber
operations, and speciality products (even though these operations may sometimes be integrated
within a refinery complex). The refinery boundaries are consistent with those used by the Oil
and Gas Journal for reporting refining activities.8
3.2 EMISSION TYPES
Methane emissions from each piece of equipment in the petroleum industry can be
classified as one of three general types: 1) fugitive; 2) vented; and 3) combusted. Emissions
were analyzed for the facilities and equipment comprising each segment of the industry. Each
source (i.e., piece of equipment) was then examined for different emissions during different
operating modes. Emissions from each source were categorized as fugitive, vented, or
combusted. Some pieces of equipment, such as compressors, may emit gas under all three
10
-------
Cfi
1
DehydratQrJ_
Vapor Recovery
Compressor
Coal Bed
Methane WeH
T_l
I Scparalor ]
Chemical
Injection
Compressar
Meter
Fresh Water
Dehydrator
Dehydralor [ "
Well
(bead
i
('
Lf
j
eparalo
i
t
0
r
1
*
Field Use
Gas
,
M
1
f '
|
^ c v
Vapor Recovery
Compressor
L-®
Figure 3-2. Industry Boundaries
11
-------
categories (fugitive emissions when pressurized, vented emissions when blown down for
maintenance and combustion emissions for the driver engines during normal operations).
Definitions of the three types of emissions are presented below.
3.2.1 Fugitive
Fugitive emissions are unintentional leaks emitted from sealed surfaces, such as packings
and gaskets, or leaks from pipelines (resulting from corrosion, faulty connections, etc). Fugitive
emissions or equipment leaks are typically low-level emissions of process fluid (gas or liquid)
from the sealed surfaces associated with process equipment. Fugitive emissions do not include
periodic vented emissions. Specific fugitive source types of emissions include various
components such as valves, flanges, pump seals, or compressor seals. These components
represent mechanical joints, seals, and rotating surfaces, which tend to wear and develop leaks
over time.
3.2.2 Vented
Vented emissions are releases to the atmosphere by design or operational practice.
Examples of vented emissions include emissions from continuous process vents, such as
dehydrator reboiler vents; maintenance practices, such as blowdowns; and small individual
sources, such as gas-operated pneumatic device vents.
3.2.3 Combusted
Combusted emissions are exhaust emissions of unburned methane fuel from combustion
sources such as compressor engines, burners, and flares. Incomplete combustion of methane fuel
in compressor engine exhaust is the only significant source of methane in this category.
12
-------
4.0 STATISTICAL METHODS
A key part of this project is the accuracy estimation of the overall national methane
emission rate. Accuracy is dependent on precision and bias. In general, precision refers to the
random variability in the measurements. Measurements with low random variability have good
precision and tight confidence bounds. Bias is a systematic error in the measurements. Bias
must be discovered and eliminated, since it is often difficult, if not impossible, to calculate.
For most calculations, bias is assumed to be zero, and this assumption is checked through
tests. If a test shows bias, additional samples are added or the sample set is stratified to eliminate
the bias. Precision can be calculated more directly; namely, by propagating error from each
individual group of measurements into the final numbers. This report used the same statistical
methods for calculating precision and bias (to the extent possible) as described in depth in the
GRFEPA statistical methods report.9
Many EFs and AFs are made up of an average of multiple measurements or calculations.
Therefore, assuming a normal distribution around a mean and error independence, standard
deviations and 90% confidence intervals can be calculated directly for each group of EF or AF
measurements. For this report, many EF and AF confidence bounds were set by engineering
judgment, since no statistical data were available.
The confidence intervals or error bounds can be propagated through the multiplication of
EFs and AFs, and through the addition of multiple emission categories to arrive at a confidence
bound for the total national emission estimate. These generally accepted statistical techniques
are briefly described in the following sections.
4.1 PRECISION
The following basic statistical calculations were performed for EFs. A different and more
complex approach, described later in this section, was used for some AFs. Suppose there are n
individual estimates of a given emission factor. If yit where 1=1 to n, are the individual data
points, then the factor is estimated as the average, y, of the n values:
= (2)
13
-------
The next step is to compute the uncertainty of this value. First, sy> the standard deviation
of the y values, is needed:
\
(3)
A 90% confidence interval is then calculated for the mean value, y. The confidence interval
establishes lower and upper tolerances for the estimate. There is only a 5% chance that the true
value falls below the lower limit of this confidence interval. There is also a 5% chance that the
true value falls above the upper limit of the interval. Thus, there is a combined 10% chance that
the true value falls outside the confidence interval. Since there is a 90% probability that the true
value falls within the interval, it is called a 90% confidence interval. The 90% confidence
interval is computed as follows:
y ± ts/v'n" (4)
The t value in this equation is obtained from a standard table for the t distribution; such
tables are found in most basic statistics books.10 The t value is a function of the confidence level
(90% in this case) and the sample size, n.
Determination of national activity factors is often more complicated than determining
emission factors, and the resulting calculation of the activity factor confidence bound is also
more complicated. A database of emission factor measurements may simply be a set of replicate
measurements, where the national emission factor is simply the average of the measurements and
the confidence bound simply describes the scatter of the replicate measurements. A database for
an activity factor (an equipment count) often requires extrapolation to obtain a national value.
The confidence bound determination must take that extrapolation into account. The following
paragraphs briefly describe the techniques used to calculate the confidence bound of an
extrapolated activity factor.
If the activity factor estimate is assumed to be approximately normally distributed, then
the 90% confidence limits for the activity factors can be estimated using Cochran's equation
6.14.10 The equation for the 90% confidence interval (symmetric) for the activity factor is:
iW-oVfi (5)
where: ?(i_0/2,n-i) = me l-a/2 probability of the Student's t Distribution with n-1
degrees of freedom.
u = variance
14
-------
The equation for the variance is:
where:
y; = the number of equipment at site I in the sample set;
X; = the number of wells or amount of production at site I in the sample set;
n = number of sites sampled;
N = the total number of sites nationally;
f = sampling fraction = n/N; and
R = activity factor ratio = (AF/EP)saraple.
The total number of sites (N) is not known nationally. Thus, it must be estimated by the
following equation;
(Production or Wells)
lotal, nationally
(Production or Wells)
tota!,sample
(7)
Either production rate or wells can be used in the equation, depending on which extrapolation
parameter is used.
4.2 PROPAGATION OF ERROR
This section discusses the general techniques used to propagate the error bounds (for
precision) that are calculated in Section 4.1. The error bounds of two numbers can be propagated
to determine the error bound of their sum and/or their product. These techniques are covered in
more detail in the GRI/EPA statistical methods report.9 Multiplication is often used in this study
since the basic extrapolation technique was to take the product of AF x EF to obtain the source's
emission rate (see Section 3.0). Addition is also used frequently since all of the individual source
emission rates are summed to obtain the national annual emissions from the petroleum industry.
Section 4.1 discussed the calculation of 90% confidence half widths for a single term,
such as an EF or AF. These confidence half widths can be substituted into the following
equations (shown in Tables 4-1 through 4-3) to determine the confidence bounds for addition and
multiplication/division.
For uncorrelated values (values not related to each other), the error bound (90%
confidence half width) of a sum is the square root of the sum of the squares of the absolute errors
of the values being summed, as illustrated in the following example. Suppose the following
values, A and B, are to be summed, and the confidence bound of value "A" is expressed as ± "a"
15
-------
(in absolute terms). The bottom cell of Table 4-1 shows the resulting error calculation for the
sum.
TABLE 4-1. ERROR PROPAGATION FOR ADDITION
VALUES TO BE
SUMMED
A
B
Sum = {A + B)
90% CONFIDENCE
Absolute Value
a
b
absolute error of (A+B) =
HALF WIDTHS
Relative Error
(Percent Value)
a% = 100 x a/A
b% = 100 x b/B
square root of {a2 + b2)
For correlated values, the equation for error becomes:
(a2 + b2 + 2rab)y' (8)
where r is the correlation coefficient between A and B. However, r was assumed to be zero since
most categories were derived from different data and were unrelated.
The error bound (90% confidence half width) associated with the product of two numbers
is also calculated with the absolute errors of the terms being multiplied. Suppose that A x B = C,
and that the absolute errors for A and B are expressed as ±"a" and ±"b", respectively, The errors
expressed as a fractional value would be fa and fb, respectively. The bottom cell of Table 4-2
shows the resulting error calculation for the product.
TABLE 4-2. ERROR PROPAGATION FOR MULTIPLICATION
90% CONFIDENCE HALF WIDTHS
VALUES TO BE
MULTIPLIED Relative Error
Absolute Value (Fractional Value)
A
B
Product = (A x B)
a t = a/A
b fh = b/B
relative error of product = square root of [fn2 + fb2 +(fa2 x fb2)]
The error bound for division of two numbers (A -f B) can be expressed in terms of the
absolute errors (a and b). Table 4-3 shows the equation for division of two uncorrelated
quantities.
16
-------
TABLE 4-3. ERROR PROPAGATION FOR DIVISION
90% CONFIDENCE HALF WIDTHS
VALUES TO BE
DIVIDED Relative Error
Absolute Value (Fractional Value)
A
B
Division = (A -f B)
a
b
relative error of
f, = a/A
fb = b/B
(A-rB) = square root of { [(A/B)2 ]x [fa2 +f,,2 ] }
The following example illustrates the use of the statistics equations presented in this
section. The example involves two numbers, A and B, where A is 10 with an absolute error of 5
and B is 6 with an absolute error of 2. This means that A is a number bounded by 5 and 15, and
B is between 4 and 8. Table 4-4 shows A and B in terms of the variables presented in this
section.
TABLE 4-4. VALUES FOR EXAMPLE CALCULATION
90% CONFIDENCE HALF WIDTHS
VALUES
Relative Error Relative Error
Absolute Value (Percent Value) (Fractional Value)
A =10
B = 6
b = 2
a% = 50.0%
b% = 33,3%
4 = 0.500
fb = 0,333
Lastly, Table 4-5 shows the resulting errors when A and B are added, multiplied and
divided.
TABLE 4-5. ERRORS FOR ADDITION, MULTIPLICATION,
AND DIVISION OF EXAMPLE PROBLEM
OPERATION
A+B =
AxB =
A-rB=l
16
60
.67
Absolute Value
5,39
37.4
1.68
90% CONFIDENCE HALF WIDTHS
Relative Error
(Percent Value)
33.7%
62.3%
100.3%
Relative Error
(Fractional Value)
0,337
0.623
1.003
17
-------
4.3 SCREENING FOR BIAS
It is impossible to prove that there is no bias in any data set. Although tests can be
designed that are capable of revealing some bias, there are no tests or group of tests that will
reveal all possible biases. Assuming that a data set has no bias, even after extensive testing, is
only a hypothesis. Such hypotheses can be disproved but cannot be definitively proven. To the
extent possible the data used for this project were checked to identify biases. The basic methods
used to screen for bias included analysis of the data and extrapolation by different parameters
(EPs).
The production site data were analyzed for bias by extrapolating the AFs with multiple
parameters (the site data and extrapolation results are presented in Appendix B). For a subset of
data that is perfectly representative of the crude production industry, equipment counts from the
data set could be extrapolated to national totals by any variable in the data set. Any extrapolation
from the perfect subset of data would result in the correct answer, regardless of the parameter
used. For an imperfect data set, which all data sets are, extrapolation by multiple variables
provides a cross check for bias. For example, in production, the equipment counts can be
extrapolated by production rate or well count. These two methods produced different results that
were averaged to minimize the potential bias from a single method.
Some significant potential biases are believed or known to exist in this report, owing to
the limited nature of the data gathering (no new data or new measurement campaigns were
performed as part of this project). Production site data based on data collected for the GR1/EPA
natural gas study were available for the petroleum industry. A separate site data collection effort
was not part of Phase 1, of this study. It is clear that the small data set has some very large
differences and is not an ideal microcosm for the U.S. petroleum production segment.
Significant problems with the production site database include the following:
• Sites do not represent a random sampling of oil production facilities in the United
States;
• A complete set of equipment counts is not available for all of the sites; and
• Sites do not truly represent a random sampling of oil production facilities in the
United States (but the sites had to be assumed to be similar to the average
facilities);
• Some commonality between the operations and equipment in light oil and heavy
oil service are assumed unless otherwise noted. It is known that these facilities
may actually vary widely.
Bias checks of activity factors used in the production segment were necessary since most
of the production activity factors were developed by this project. Bias checks in the production
segment were simple to perform since the sample database could be compared against some
known national values. Bias checks for activity factors in the other segments, crude
transportation and refining, were not necessary since most of the activity factors were from
published, well defined sources. These published factors, which were not on an equipment detail
18
-------
level, did not depend upon sample (site visit equipment count) databases, so there was no need
nor method for bias checks.
Some biases exist in the data set and they are believed to affect the overall estimate. The
actions taken to minimize the effect of bias in the production data set are discussed in detail in
Section 5.1.1. Some future work will be required to minimize bias. Minimization and/or
elimination of potential biases are discussed as future efforts in Section 9.
19
-------
5.0 ACTIVITY FACTORS AND EMISSION FACTORS—1993 BASE YEAR
Sections 5.1 and 5.2 present the sources and units used for the AFs and EFs, respectively.
Each section is further divided into the three industry segments studied (production, crude
transportation, and refining).
5.1 ACTIVITY FACTORS -1993 BASE YEAR
Two general methods were used to estimate AFs for the 1993 base year. First, national
AFs were taken from existing sources, such as the Oil and Gas Journal?'11 World Oil,12 the
GRI/EPA natural gas study,4 American Petroleum Institute (API) Report 4615,13 the Pipeline
Systems Inc. (PSI) study,14 and other published sources. Second, some production segment AFs
were extrapolated from oil field site visits performed during the GRI/EPA natural gas study, For
some categories, data from published sources had to be modified for use with this study; these
modifications are discussed below. Data taken directly from sources are referenced. Table 5-1
summarizes the AFs used for this report.
5.1.1 Production Rate and Well Count
The two most important AFs in the production segment are total crude oil production rate
and well count. Annual production for the 1993 base year is 6,846,000 barrels of crude per day,
which is taken from a 1995 Oil and Gas Journal article.15 Total number of producing oil wells
for 1993 is 583,879 which is taken from World Oil.12 These values are used to generate other
AFs for the production segment. The confidence intervals for these sources are based on
engineering judgment. Since these are well-documented values, based on credible data that are
nationally published, a confidence interval of ± 5% was assigned.
To correspond to the EF split between heavy and light crude, production rate and well
equipment counts were divided into heavy and light crude. The API Report 461513 designated
heavy crude as having an API gravity of less than 20° and light crude as having an API gravity of
greater than 20° for the purposes of establishing EFs. A 1984 report by the Interstate Oil
Compact Commission was used to determine the volume of heavy crude produced for all states
with heavy crude production except Alaska.16 Alaska's Natural Resources Department was
contacted separately for this information. For the years in which heavy crude production was
reported (1976 through 1981, with the exception of Alaska, which is based on 1993 data), the
total crude production for the same year was determined for each state by referencing the Oil and
Gas Journal.15 A ratio of heavy crude production to total crude production was calculated, as
shown in Table 5-2.
20
-------
TABLE 5-1. 1993 ACTIVITY FACTORS BY SEGMENT
PRODUCTION
Published (Well Known)
Source Category Activity Factor
Developed
Source Category
Activity Factor
Crude oil production rate
Well count
Crude oil completions
Exploratory wells drilled
Well workovers
Burners
Well blowouts
6,846,000 bbUday
583,879 wells
390 completions
390 wells
43,791 workovers/year
3,647,000 bbl/year
2.85 blowouts/year
Heavy/light crude ratio
Heavy/light well ratio
Oil wellheads
Separators
Heater-treaters
Compressors (in light crude service)
Gas lift compressors
Pneumatic devices
Chemical injection pumps (CIPs)
Headers
Tanks
Fields (for sales areas)
Offshore platforms
Pipeline miles
Gas engines
Pressure relief valves
10.7%/89.3%
7.1%/92.9%
41,163 heavy, 542,716 light
9,103 heavy, 113,071 light
77,354 heater treaters
647 small, 1,940 large
2,799 compressors
117,008 devices
125,088 CIPs
15,296 heavy, 47,291 light
54,272 tanks in light service
4,443 fields
1,092 Gulf, 22 rest of U.S.
70,000 miles
17,634 MMhp-hr
422.936 PRVs
(Continued)
-------
TABLE 5-1. 1993 ACTIVITY FACTORS BY SEGMENT
(Continued)
CRUDE TRANSPORTATION
Published (Well Known)
Source Category Activity Factor
Crude pipeline miles
Volume transported by truck
Volume transported by marine
Volume transported by rail car
Volume transported by pipeline
55,268 miles
7.69E+07 bbl/ycar
9.54E+10 gal/year
(2.27E + 10 bbl/year)
8.91E+06 bbl/year
6.71E+09 bbl/year
Developed
Source Category
Pump stations
Volume stored in lanks (total
transported)
Activity Factor
553 stations
9.07E+09 bbl/yr
(Continued)
to
-------
to
TABLE 5-1. 1993 ACTIVITY FACTORS BY SEGMENT
(Continued)
REFINING
Published (Well Known)
Source Category
Total refinery charge of crude
Charge rate to:
Vacuum distillation
Thermal operations
Catalytic cracking
Catalytic reforming
Catalytic hydrocracking
Catalytic hydro refining
Catalytic hydro treating
Alkylation & polymerization
Aromatics/isomeri/Ation
Lube processing
Asphalt production
Activity Factor
13,612,259 bbl/day
5,935,032 bbl/day
1,661,140 bbl/day
4,694,106 bbl/day
3,287,291 bbl/day
1,112,414 bbl/day
1,595,163 bbl/day
7,326,166 bbl/day
1,003,670 bbl/day
693,791 bbl/day
177,624 bbVday
631,440 bbj/day__
Developed
Source Category
Activity Factor
Heaters
Engines
3,200 heaters
20,334 MMhp-hr
-------
TABLE 5-2. RATIO OF HEAVY CRUDE PRODUCTION TO
TOTAL CRUDE PRODUCTION
Heavy Crude
Production16
State
Alabama
Alaska
Arkansas
California
Colorado
Illinois
Kansas
Louisiana
Michigan
Mississippi
Montana
New Mexico
Oklahoma
Texas
Utah
Wyoming
Year (1000 bbl)
1981
1993
1976
521
1,060
4,682
1981 277,825
1981
1981
1981
1979
1980
1981
1981
1976
1976
1981
1976
1981
Total for States with Heavy
197
78
1,247
16,769
10
7,831
341
230
3,394
20,079
1,177
24,275
Total Crude
Production15
(1000 bbl)
20,680
577,913
17,885
384,958
30,303
24,090
65,810
494,575
33,580
34,204
30,813
91,615
190,965
945,132
33,945
130,563
Ratio of
Heavy/Total
Production
0.025
0.002
0.262
0.722
0.007
0.003
0.019
0.034
0.000
0.229
0.011
0.003
0.018
0.021
0.035
0.186
Crude Production
U.S. Total Crude Production,15 1000 bbl
National Ratio
of Heavy Crude to Total
Crude
1993 Total Crude
Production15
(1000 bbl)
18,677
577,430
10,599
293,112
31,211
17,726
49,691
407,340
13,799
22,570
17,431
69,520
96,791
620,210
21,819
87,667
2,355,593
2,498,425
Estimated 1993
Heavy Crude
Production (1000
bbl)
471
1,059
2,775
211,540
203
57
942
13,811
4
5,167
193
175
1,720
13,176
757
16,300
268,350
10.7%
The ratio of heavy crude to total crude production shown for each state in Table 5-2 was
assumed to apply to 1993 production and well counts for the respective states. To estimate the
national heavy crude production for 1993, the ratio of heavy crude to total crude was applied to
the 1993 production rate of each state. The estimated 1993 heavy crude production for each state
was then summed to generate a national heavy crude production of approximately 268 million
barrels, which corresponds to 10.7% of the total crude production for 1993. The confidence
bound (±100%) associated with this estimate was assigned based on engineering judgment. The
error bounds are wide, since 1976-1981 data were used to establish the heavy crude to total crude
ratio for 1993.
The same procedure was used to estimate the number of wells in the United States that
produce heavy crude, as shown in Table 5-3. The ratio of heavy crude production to total crude
production for each state was applied to the number of crude wells in that state,12 resulting in an
estimate of the number of wells that produce heavy crude for that state. The state heavy crude
24
-------
TABLE 5-3. RATIO OF HEAVY CRUDE PRODUCTION WELLS
TO TOTAL CRUDE PRODUCTION WELLS
Heavy Crude
Production16
State Year (lOOObbl)
Alabama 1981 521
Alaska 1993 1,060
Arkansas 1976 4,682
California 1981 277,825
Colorado 1981 197
Illinois 1981 78
Kansas 1981 1,247
Louisiana 1979 16,769
Michigan 1980 10
Mississippi 1981 7,831
Montana 1981 341
New Mexico 1976 230
Oklahoma 1976 3,394
Texas 1981 20,079
Utah 1976 1,177
Wyoming 1981 24,275
Total Crude Ratio of
Production15 Heavy/Total
(lOOObbl) Production
20,680
577,913
17,885
384,958
30,303
24,090
65,810
494,575
33,580
34,204
30,813
91,615
190,965
945,132
33,945
130,563
0.025
0.002
0.262
0.722
0.007
0.003
0.019
0.034
0.000
0.229
0.011
0.003
0.018
0.021
0.035
0.186
Total for States with Heavy Crude Production
U.S. Total Crude Production Wells12
National Ratio of Heavy Crude Wells
to Total Crude Wells
1993 Total Crude
Production
Wells12
886
1,624
8,466
40,231
7,221
31,783
44,000
22,264
4,201
1,631
3,600
18,028
93,192
181,501
1,990
11,287
471,905
583,879
Estimated 1993
Heavy Crude
Production Wells
22
3
2,216
29,035
47
103
834
755
1
373
40
45
1,656
3,856
69
2,099
41,154
7.05%
well counts were summed to give a national number of wells that produce heavy crude, which
was divided by the total number of U.S. crude production wells, resulting in an estimated 7.05%
of the total crude wells in the United States that produced heavy crude during 1993. A
confidence bound of 100% was associated with this estimate based on engineering judgment,
5.1.2 Production Equipment Extrapolations from Site Visits
Equipment populations (separators, heater treaters, etc.) in the petroleum production
segment are not tracked nationally. Thus, equipment extrapolations from site data must be
carried out to estimate the national population. The equipment extrapolations in the production
segment for this study were based on site visit data taken during the GRI/EPA natural gas study.
The sites from the GRI/EPA study that were used in this project were sites in which oil was
produced. There were 26 such sites, as shown in Table B-l of Appendix B.
Production equipment extrapolations for this study were carried out for separators, heater
treaters, pneumatic devices, chemical injection pumps (CIPs), and gas lift compressors. The
25
-------
activity factor for blowdown emissions from vessels was estimated by assuming that the number
of vessels was the sum of separators and heater-treaters.
The equipment extrapolations and statistical methods were carried out in the same
manner as described in the GRI/EPA study. To briefly summarize, an AF ratio was determined
for each equipment type by dividing the site AF, the total number of equipment in the sample
data set, by the site extrapolation parameter (EP), which was the total number of oil wells (well
basis) or oil production (throughput basis) in the whole sample data set. This sample AF ratio
can be designated (AF/EP)^^,.. Next, the AF ratio was multiplied by the extrapolation
parameter (EP)region, which was either the known U.S. oil production or number of wells. This
product yields the extrapolated number of equipment for the well and throughput basis. This is
illustrated in the following formula:
X EPregio,r Region (9)
where:
EP
\ / sample ^> •t-,-n
L E1i
i = 1
n = number of individual sample sites in the data set
Table 5-4 shows the results of the extrapolations. Tables B-l and B-2 in Appendix B
show the detailed site visit data used to generate Table 5-4, and Table C-l (in Appendix C)
shows the corresponding statistical analysis. The following table (Table 5-4) shows that the
results of the well extrapolation basis are very different than the results of the throughput
extrapolation basis. This raises several questions: 1) are the sites representative of the petroleum
industry; 2) is the equipment strictly related to one parameter, so that the extrapolation by the
other parameter produces an erroneous result?; and/or 3) is there bias in the data set that resulted
in the difference? These questions are examined in the following text.
Although the equipment extrapolations are based on data from 26 oil producing sites,
these site visits were conducted as part of the GRI/EPA natural gas industry study,7 The site data
do not truly represent a random sampling of oil production sites, and may therefore introduce
bias and account for some of the difference between the two extrapolation techniques. No new
sites were visited as part of this Phase 1 study,
26
-------
TABLE 5-4 EXTRAPOLATED ACTIVITY FACTOR DEVELOPMENT
Extrapolated Count
Well Basis Throughput Basis
Equipment
Separators
Heater Treaters
Pneumatics
Chemical Injection Pumps
Gas Lift Compressors
Count
217,804
143,491
207,217
125,088
12,523
Confidence
Interval
50.9%
150.5%
71.2%
105.0%
94.1%
Count
26,562
23,873
26,800
24,959
2,162
Confidence
Interval
26.5%
1 16.2%
72.6%
92.2%
87.3%
Selection between the two EPs (wells or throughput) can be done on a technical basis if
there is a clear technical relation between the particular type of equipment and one EP, This is
the case for CIPs, where the pumps are predominantly located at the wellhead. Production
segment technical advisors from the GRI/EPA natural gas study recommended that CIPs be
extrapolated only by well count. For this study, the same recommendation was applied to CIPs
in the oil industry. Logically, methane-powered CIPs could only be used on wells that have
pressured gas available. That operational requirement would exclude many stripper wells. The
sites visited, however, had a higher production per well than the U.S. known production rate per
well. While some of the stripper wells visited for the dataset did have gas powered CIPs, it is
very possible that the CIP extrapolation based on the 16 site visit wells does result in a high CIP
count and methane emission bias in this category.
For other equipment, such as separators, heater treaters, pneumatic devices, and
compressors, a clear technical basis for using one EP over the other could not be determined.
There are cases where equipment count is related only to well count (such as individual, remote
well sites, where equipment must be added for each new well), and cases where equipment count
is related primarily to production rate (such as centralized facilities, where multiple wells are fed
into one separator). The national population of these types of equipment, therefore, is related to
both wells and production rates.
For similar circumstances in the natural gas production segment, technical advisors
recommended combining the two extrapolation techniques by averaging the equipment counts
that result from each method. The same approach is used for the oil production equipment
associated with this study. However, the results from the two EP methods are farther apart for
the petroleum industry than for the gas industry. Table 5-5 shows the selection basis used; the
resulting extrapolation value for each source is highlighted.
27
-------
TABLE 5-5. PRODUCTION EXTRAPOLATION PARAMETER SELECTION
Activity Factor
Extrapolation
Parameter(s)
Selected
Basis
Heater-Treaters
Mean of:
1) Oil wells and
2) Production rate
(83,682 ±78.0%)
Heater-treaters can be related to both production rate and
well count, Heater-treaters can be located at individual
wells or central facilities, where production rate may play
a factor. In the offshore area, the relation is stronger to
production rate.
Separators
Mean of:
1) Oil wells and
2) Production rate
(122,183 ±78.0%)
Separators can be related to both production rate and well
count. Separators can be located at individual wells or
central facilities, where production rate may play a factor.
In the offshore area, separators are more strongly related
to production rate.
Gas-Lift Compressors Mean of:
1) Oil wells and
2) Production rate
(7,342 ±81.2%)
Gas-lift compressors exist within the oil industry to
artificially lift oil. The compressors can be located at each
well site or at a central facility, where the number of
compressors is related to production rate.
Pneumatic Devices
Mean of:
1) Oil wells and
2) Production rate
(117,008 ±78.0%)
Pneumatic devices exist on oil well separators, heater
treaters, gas-lift compressors, and some other equipment.
Therefore, pneumatics are related to the equipment counts
(which, as shown above, are related to both well count and
production rate).
Chemical Injection
Pumps (CIPs)
Oil wells only.
(125,088 ±105%)
CIPs were found primarily at individual well sites, even
where central separation facilities existed. CIPs therefore
have a strong relation to well count. CIPs on gas wells
were not counted in this study.
Table 5-5 also shows the arithmetic mean extrapolated equipment counts for each piece
of equipment. The error bounds were determined using the statistical methods outlined in
Section 4.1, except for the following cases. By using engineering judgment, the error bound for
the separators and pneumatic devices were assigned 78%, since the calculated bound did not
encompass the individual throughput and well extrapolations. This ensures that the error bound
includes the counts given by the well and throughput extrapolations.
In addition to the technical reason for selecting the mean of both methods, a potential bias
that exists in the data set can be corrected by selecting the mean. The well and throughput
extrapolations produce different equipment counts, as Table 5-4 points out. Since it is known
that the site database is less than ideal, bias checks were performed to see how well the collected
data compare with the total U.S. oil production segment. Table 5-6 shows the comparisons
made.
28
-------
TABLE 5-6. COMPARISON OF SAMPLE SET TO NATIONAL VALUES
Site Visit U.S.
Sample Category Database Known" Corrections Made
Production per Well 70,6 11.7 Selected the average of the well count and production
(bbl/d/well) extrapolations (except for chemical injection pumps)
% of Sites with Gas-Lift 23,8% 9,1 % Applied correction factor of 0.381 (9.1/23.8) to gas-lift
Compressors compressor count
% of Oil Wells Offshore 6,4% 1.2% No action taken
a Sources for U.S. data:
Number of Wells: World OH, February 199412
% of Sites with Gas Lift Compressors: JPT, 199317
% of Oil Wells Offshore: GRI/EPA Activity Factor Report, 1992 Data7
Production/Well: Production from Oil and Gas Journal, January 30, ]99515
On the basis of these comparisons, the following biases were identified:
1) The site database results in a much higher production rate per well than the national
average;
2) While the site database contains some stripper wells, it may not accurately represent
the large population of stripper wells in the U.S.;
3) A larger number of gas lift sites are represented in the database than the national
number; and
4) The limited site visit data have more offshore oil wells than the national number.
An attempt was made to analyze the effect of these biases in light of averaging the
equipment counts that result from the two extrapolation techniques. The high production rate per
well would be expected to produce a production rate extrapolation of AFs that was too low, and a
well count extrapolation that was too high, so long as the equipment count was related to both
EPs. The true value will lie between the two estimates. Appendix D demonstrates this point
through some hypothetical examples.
For this project's production site data set, the well count extrapolation was much higher
than the production rate extrapolation in every case, which tends to support the hypothesis that
the equipment is related to both EPs. Unfortunately, it is not possible to determine the exact
relation between equipment type and the EPs. Since it is not known how strongly the equipment
is related to either wells or throughput, the arithmetic mean was used for all equipment except for
CIPs. This same approach was used in the GRI/EPA natural gas study. As explained in
Table 5-5, CIPs were assumed to be related only to wells, since the pumps are primarily located
at the well head.
To correct for the high percentage of gas-lift sites in the database, a correction factor was
applied to the extrapolated count of gas-lift compressors. A factor of 0.381 was developed on the
basis of dividing the percentage of U.S. known gas-lift sites by the percentage in the site database
29
-------
(0.381 = 9.1/23.8), Thus, this bias correction factor effectively adjusts the count of gas-lift
compressors to be more consistent with the true number in the United States. Adding this
correction factor lowers the count of gas-lift compressors from 7,342 to 2,799, The final
corrected values are shown in Table 5-7.
TABLE 5-7. FINAL PRODUCTION DEVELOPED ACTIVITY FACTORS
Separators 122,183" ±78% c
Heater-treaters 83,682a ± 130%
Pneumatic Devices 117,008" + 78%c
Chemical Injection Pumps 125,088h ± 105%
Gas Lift Compressors | 2^99^ ± 81%
" Arithmetic mean of well and throughput extrapolation method,
h Well method only,
c Used engineering judgement to assign confidence bound.
d Lowered from original extrapolation by a factor of 0.381 to account for site visit bias,
5.1.3 Miscellaneous Production Activity Factors
With respect to the count of gas-lift compressors (2,799 total), an assumption was made
on the basis of site data and engineering judgment that 75% of the total compressors are large
and 25% are small, with a confidence interval of 33%. This distinction was necessary since the
fugitive EF varies for small and large compressors. Large compressors are those housed in
facilities where the compressors will have a remote blowdown vent stack. They are similar to
gas transmission compressors, which are located in station facilities. Small compressors are
defined as those with a blowdown vent line located proximate to the compressor. No attempt
was made to relate large and small compressors to horsepower. The distinction here is only
related to the compressor vent arrangement, where the remote blowdown vent lines where found
to have very large fugitive emission rates.13
In the production segment, most oil wells have some type of artificial-lift method in
place. Approximately 85% of artificial lift wells use sucker-rod pumps. Gas lift, mostly
continuous flow, make up about 10% of artificial lift wells. Electric submersible pumps (ESPs)
are used on 4% of the wells. All other lift methods (hydraulic, reciprocating pumps, progressing
cavity pumps, and plunger lifts) represent less than 5% total usage. Eighty percent of total
artificial lift wells are classified as stripper wells that produce small volumes of oil. When the
stripper wells are excluded, of the remaining U.S. oil wells (approximately 100,000 wells) 53%
are gas lifted.17 The majority of these are on continuous gas lift. This information is important to
note because methane emissions are high from gas-lifted wells and compressors.
The AF for gas engines was derived from the count of compressors combined with an
estimated horsepower per compressor as given in the AF report of the GRI/EPA natural gas
study.7 The GRI/EPA study reported 25,780 ± 134% MMhp-hr for all compressor engine drivers
30
-------
in the natural gas processing segment. The study also reported that there are 4,092 ± 47.7%
reciprocating engines at gas plants. Thus, division of the total engine energy consumption by the
number of engines yields 6.30 MMhp-hr/compressor (± 205.7%) in the natural gas processing
segment. Engines in the natural gas processing segment were assumed to be similar to those of
gas lift compressors in the petroleum industry. Therefore, the AF for combustion from gas-lift
engines in annual MMhp-hr was determined by multiplying 6.30 MMhp-hr/compressor by the
number of gas lift compressor determined from the equipment extrapolation in this study
(discussed earlier in this section), resulting in 17,634 MM hp-hr for 1993. The confidence
intervals for the division and multiplication used in this estimation method were calculated as
described in Section 4.2.
Several other equipment counts were established for the purposes of estimating fugitive
emissions. Total headers (15,296 heavy, 47,291 light) were taken from the API Report 4615
with the assumption of 0.37 headers/heavy well and 0.087 headers/light well.13 These ratios are
based on the equipment counts taken from the API report. The tank AF (54,272 light crude
tanks) was also from the API report, which produced a net ratio of 0.1 tanks/light well. Once
again, "heavy" equipment refers to equipment that is in heavy crude service (API gravity of less
than 20°) and light equipment refers to equipment that is in service to light crude (API gravity, of
greater than 20°).
The number of fields used to estimate sales areas (2,962) was taken from a report by ICF
Resources Incorporated.18 The error bound for the number of fields was assumed to be 30%
based on engineering judgement. There is also an assumption of 1.5 areas per field based on
engineering judgment, with an associated error of 33%. Thus, the total number of sales areas is
4,443 ± 46%.
The number of offshore platforms for the Gulf of Mexico and the rest of the United States
comes directly from the GRI/EPA natural gas study. There were 1,092 in the Gulf of Mexico and
22 in the rest of the United States. The GRI/EPA study presented the total number of platforms
and the number of natural gas platforms; the difference yields the crude platforms. The total
number of oil wellheads was taken directly from World Oil, where the split between heavy
(41,163) and light (542,716) crude production was based on the ratio of heavy crude production
to total crude production, as discussed previously (Section 5.1.1).12
The number of pipeline miles was taken from the GRI/EPA natural gas study.7 This
study reported 140,000 total pipeline miles with an assumed 50/50 split between petroleum and
natural gas pipelines. Using the same assumptions, 70,000 (± 50%) production gathering miles
are associated with the oil industry.
The number of crude well completions (390) was taken from the Energy Information
Administration (EIA) "Annual Energy Review."19 This number is also used as the AF for
drilling as a combustion emission source (exploratory wells drilled). A confidence interval of
10% was assigned by engineering judgment based on the quality of the reported value.
31
-------
An estimate of the number of well workovers per year (43,791) is taken from a PSI
report, which estimated 7.5% of wells are worked over each year based on observations from two
crude production sites.14 Because of the limited sample size, a confidence interval of 100% was
assigned to this AF.
The AF corresponding to burners (3,647,000 bbl/year) is based on the volume of crude oil
consumed by pipelines and on leases as pump fuel, boiler fuel, and so forth. This number is
reported by production companies on EIA Form 813 and published nationally in the Petroleum
Supply Annual,20 The confidence interval for this value was assigned by engineering judgment.
Natural gas is also consumed as plant and lease fuel in crude production. The GRI/EPA natural
gas study considered the total amount of natural gas reported as plant and lease fuel use to be part
of the natural gas industry, where the portion of gas used to run compressors was subtracted from
the total plant and lease gas use, and the remaining amount was assumed to be used in burners.
Methane emissions from burners were negligible for the natural gas industry study, and are also
believed to be negligible for the oil industry. Therefore, it was not necessary to determine the
amount that might be attributed to the oil industry for this study.
The number of pressure relief valves (422,936 PRVs) was developed using the same
methodology as the GRI/EPA natural gas study,7 The GRI/EPA study estimated the number of
PRVs associated with specific equipment types. For similar equipment used in crude production,
the same ratios were used: 2 PRYs per separator (±68%); 2 PRVs per heater treater (±89%),
assuming a heater treater is most similar to a separator; and 4 PRVs per gas lift compressor
(±84%). These ratios were then multiplied by the extrapolated equipment counts (Table 5-7) and
summed to give the total number of PRVs. The confidence interval was calculated (using the
methods described in Section 4.2) on the basis of the confidence intervals associated with the
PRV to equipment ratios and the individual equipment counts.
The number of well blowouts annually (2.85) was estimated on the basis of a total of
57 well blowouts tracked by the U. S. Geological Survey for the years 1956 through 1977.21 A
large confidence bound of 200% was assigned to this estimate because of the age of the data.
5.1.4 Crude Transportation
The AF for pump station emissions is given in units of miles of crude pipelines
(55,268 miles.) The Oil and Gas Journal reports total miles of crude trunk lines for interstate
pipelines." A national source for intrastate crude pipeline miles was not found. Therefore, the
number of miles is underestimated. The AF corresponding to pipeline fugitive emissions
(6.71E+09 bbl/year) is the volume of crude transported by pipelines. The value reported by Oil
and Gas Journal for crude trunk lines was used for this source. This AF is also underestimated
because intrastate pipeline volumes are not included. The confidence interval for these sources
was assigned 100% based on engineering judgment.
32
-------
The EIA Petroleum Supply Annual reports volumes of crude delivered to refineries by
mode of transport [tanker (2.11E+09 bbl/year), trucks (7.69E+07 bbl/year), barge (1.67E+06
bbl/yr), and rail cars (8.91E+06 bbl/year)] for both domestic and imported crude.22
The AF for tanks (9.07E+09 bbl/year) was estimated by assuming each barrel of crude
transported is stored in a tank once. The total number of barrels transported to refineries was
calculated by summing the volumes reported for each mode of transport (i.e., the sum of the
volumes transported by pipeline, marine, rail, and truck). Note that the volume of crude
transported by marine vessels (tankers and barges) is reported in gallons rather than barrels to
correspond to the EF units. The confidence intervals were assigned by engineering judgment for
the individual transport modes.
The AF for pump stations (553 stations) is based on the assumption that one gas operated
pump station exists for every 100 miles of pipeline,14 where the number of pipeline miles is taken
from the Oil and Gas Journal, as discussed above.11 Here also, the AF may be underestimated,
since the Oil and Gas Journal excludes intrastate pipeline mileage. The confidence interval was
assigned to be 100% by engineering judgment.
5.1.5 Refining
All of the AFs for refining emissions, except heaters and engines, are in units of barrels
per day. Two sources of data, both from the Oil and Gas Journal, were used to generate the
crude volumes for each refinery operation.8'23 The Oil and Gas Journal reports crude feed rates
in barrels per calendar day to each refinery process.8 Calendar day throughputs for the individual
refinery process units, which represent the maximum capacity of the unit, were adjusted to actual
refinery still runs based on the total refinery utilization,23 where the total utilization (total refinery
capacity divided by crude runs to stills) was assumed to be applicable to each of the process
units. The resulting throughputs (shown in Table 5-8) represent the actual volume of crude
refined in each process per day.
TABLE 5-8. REFINERY THROUGHPUTS
Process l.OOObbl/d
Vacuum distillation 5,935
Thermal operations 1,661
Catalytic cracking 4,694
Catalytic reforming 3,287
Catalytic hydrocracking 1,112
Catalytic hydrorefining 1,595
Catalytic hydrotreating 7,326
Alkylation/polymerization 1,004
Aromattcs/isomerization 694
Lube processing 178
Asphalt production 63.1
33
-------
The total volume of crude refined (13,612,259 barrels/day) was used to estimate
emissions from tanks, atmospheric distillation, wastewater treatment, cooling towers, system
blowdowns, and flares. The confidence interval for each of these sources was assigned to be
±5% based on engineering judgment.
The AF used for fuel gas system fugitives was the number of refinery heaters. The
number of heaters was taken from a 1993 EPA report entitled Alternative Control
Techniques—NOX Emissions from Process Heaters,24 An estimate of 3,200 is cited as the
number of heaters in the refining industry. An error bound of 50% was assigned based on
engineering judgment.
A number of assumptions were used to estimate the AF for refinery engines (20,334
MMhp-hr). First, the energy requirement for each of the refinery process units (reported in
BTU/bbl crude)25 and the volume of crude refined through each unit (based on the activity factors
shown in Table 5-8) were used to estimate the total energy required by the refinery. Results of
this analysis are shown in Table 5-9. The Petroleum Supply Annual reports the volume of fuels
consumed in refineries,26 which can be converted to energy equivalents based on the heat rate of
each fuel type,27 thus representing the total energy consumed at refineries (shown in Table 5-10).
Assuming that the difference between the total energy consumed at refineries and the
energy requirements of the various refinery processes is attributed to fuels used to power other
engines, the energy input to engines is estimated to be approximately 54E+09 hp-hr (after
converting the difference between the totals shown in Tables 5-9 and 5-10 from MMBtu to hp-
hr). The EF for engines is expressed in terms of energy output, so an engine efficiency of 33%
was estimated on the basis of efficiencies reported in AP-42 for typical gasoline, diesel, and gas
operated engines (AP-42, Tables 3.3-2 and 3.2-2).28 The end result is the energy output from
engines used in refineries (approximately 20E+09 hp-hr). A confidence bound of 100% was
assigned to this value due to inherent problems associated with the difference between two large
values.
34
-------
TABLE 5-9, 1993 REFINERY ENERGY REQUIREMENTS
Refinery Process
Atmospheric Distillation
Vacuum Distillation
Thermal Operations
Catalytic Cracking
Catalytic Reforming
Catalytic Hydrocracking
Catalytic Hydrorefining
Catalytic Hydrotreating
Alkylation/Polymerization
Aromalics/Isomerization
Lube Processing
TOTAL MMBtu/yr
Fuel Usage"
BTU/bbl crude
100,000
74,900
88,000
100,000
320,000
250,000
70,000
75,000
1,100,000
190,000
f 40,000
Crude Feed Rate b
bbl/yr
4,974,001,000
2,168,696,410
606,990,620
1,715,254,720
1,201,195,655
406,482,615
582,882,370
2,677,024,975
366,746,890
253,515,130
64,905,030
MMBTU/yr
497,400,100
162,435,361
53,415,175
171,525,472
384,382,610
101,620,654
40,801,766
200,776,873
403,421,579
48,167,875
9,086,704
2,073,034,168
"Fuel Usage: Radian Corporation, "The Assessment of Environmental Emissions from Oil
Refining," My 198Q.25
b Crude Feed Rate: Oil and Gas Journal, Annual Refining Report, 1993.8
TABLE 5-10.
Fuel Type Heat Rate"
Distillate Fuel Oil 5,825
Residual Fuel Oil 6,287
Still Gas 6,000
Natural Gas 1,030
TOTAL, MMBtu/yr
1993 REFINERY FUEL CONSUMPTION
Units
MMBTU/bbl
MMBTU/bbl
MMBTU/bbl
BTU/scf
Fuel Usage" Units
515,000 bbl
10,460,000 bbl
230,760,000 bbl
735,939 MMscf
MMBTU
2,999,875
65,762,020
1,384,560,000
758,017,170
2,211,339,065
"Heat Rate: Energy Information Administration, Annual Energy Outlook 1995,1995.27
bFuel Usage: Energy Information Administration, Petroleum Supply Annual, 1994,26
35
-------
5.2 EMISSION FACTORS -1993 BASE YEAR
Several of the EFs used in this report were taken from other studies. The GRI/EPA
natural gas study is used often. Other referenced reports include API Report 4615,B AP-42,28
and API's Global Emissions of Methane from Petroleum Sources.5 Sometimes the data had to be
reprocessed to make them apply to the petroleum industry; these corrections will be discussed
below. Data taken directly from existing sources are referenced. Table 5-11 summarizes the
emission factors used by this project.
5.2.1 Production
The EFs for production are presented below under each major emission type.
Fugitive Emissions—
Fugitive EFs for offshore platforms for the Gulf of Mexico and the rest of the United
States (scfd/platform) come directly from the GRI/EPA natural gas study.29 EFs from oil
wellheads (heavy and light), separators (heavy and light), heater/treaters (light crude), headers
(heavy and light), compressors (light crude-small and large), and sales areas, all reported in
scfd/source type, are derived from the January 1995 API Report 4615 Emission Factors for Oil
and Gas Production Operations." A 30% error bound was assumed based on engineering
judgment. The API EFs are split into heavy and light crude, since heavier crude has less methane
and therefore a lower EF. Fugitive EFs for tanks (light crude, scfd/tank) were also taken from
API Report 4615.13 The underground pipeline fugitive EF and error bound came directly from
the GRI/EPA natural gas study.30 More detail on production fugitive EFs can be found in
Appendix E.
Vented Emissions—
Oil tanks emit methane from the flash that occurs when crude oil is lowered to
atmospheric pressure in the tank. Emissions occur through the tank vent to the atmosphere if it is
uncontrolled. This is believed to be a much larger source of methane emissions than working or
breathing losses from the production tanks. The oil tank EF, scf/bbl, and confidence interval
were derived from a 1992 Canadian Petroleum Association (CPA) field measurement study.31
The Canadian Study showed an average tank emission rate of 12.1 scf CH4/bbl. Since tanks are
such a large methane emission source, the Canadian data were compared with emission estimates
predicted using the ASPEN Plus™* process simulator. For the simulations, (details provided in
Appendix F) methane emissions were estimated from fixed-roof atmospheric pressure oil tanks,
assuming that the oil is in equilibrium with a methane stream in a gas/oil separator upstream of
the tank. Methane dissolved in the oil at the temperature and pressure of the separator is flashed
* ASPEN Plus™ is a registered trademark of Aspen Technology, Inc.
36
-------
TABLE 5-11. EMISSION FACTOR SUMMARY
UJ
PRODUCTION
Emissions Source Category
Fugitive Sources:
Offshore Platforms - Gulf of Mexico
Offshore Platforms - Rest of US
Oil Wellheads (heavy crude)
Oil Wellheads (light crude)
Separators (heavy crude)
Separators (light crude)
Heater Treaters (light crude)
Headers (heavy crude)
Headers (light crude)
Tanks (light crude)
Small Compressors (light crude)
Large Compressors (light crude)
Sales Areas
Pipelines
2914
1178
0.83
19,58
0.85
51.33
59.74
0.59
202.78
34.4
46.14
16360
40.55
56.4
Emission Factor
scfd CH4/platform
scfd CH4/platform
scfd CH4/weIl
scfd CH4/weIl
scfd CH4/sep
scfd CH4/sep
scfd CH4/heater
scfd CH4/header
scfd CH4/header
scfd CH4/tank
scfd CH4/compressor
scfd CH4/compressor
scfd CH4/area
scfd CH4/mile
Source
GRI/EPA Study19
GRI/EPA Study29
API 46 15 Report"
API 46 15 Report13
API 46 15 Report13
API 4615 Report"
API 4615 Report13
API 4615 Report13
API 46 15 Report13
API 46 15 Report"
API 4615 Report13
API 46 15 Report13
API 46 15 Report13
GRI/EPA Study30
(Continued)
-------
TABLE 5-11. EMISSION FACTORS BY SEGMENT
(Continued)
oo
PRODUCTION
Emissions Source Category1
Vented Sources:
Oil Tanks
Pneumatic Devices
Chemical Injection Pumps
Vessel Blowdowns
Compressor Starts
Compressor Blowdowns
Completion Flaring
Well Workover
Emergency Shutdown (ESD)
Pressure Relief Valve (PRV) Lifts
Well Blowout
Combustion Sources:
Gas Engines
B urners
Drilling
12.1
345
248
78
8443
3774
733
96
256,888
34
250,000
0.24
0.526
0.052
Emission Factor
scfCH4/bbl
scfd CH4/device
scfd CH4/pump
scfy CH4/vcsscl
scfy CH4/compressor.
scfy CH4/compressor,
scfd CH4/completion
scf CH4/workover
scfy CH4/platfonri
scfy' CH4/PRV
scf CH4/blowout
scfCH4/hp-hr
lbCH4/1000gaI
ton CH4/well drilled
Source
CPA Study"
GRI/EPA Study33
GRI/EPA Study34
GRI/EPA Study35
GRI/EPA Study35
GRI/EPA Study35
GRI/EPA Study36
PSI Report14
GRI/EPA Study35
GRI/EPA Study35
EPA Report21
GRI/EPA Study37
AP-4228
L992 API Report3
(Continued)
-------
TABLE 5-11. EMISSION FACTORS BY SEGMENT
(Continued)
VD
Emission Source Category
Fugitive Sources:
Pump Stations
Pipelines
Vented Sources:
Tanks
Truck Loading
Marine Loading
Rail Car Loading
Pump Stations
Combustion Sources:
Pump engine drivers
CRUDE TRANSPORTATION
Emission Factor
1.06 Ib CH4/yr/mile
0.0 Ib CH4/bbl
4.37e-07 ton CH4/bbl
1.02e-05 tonCH4/bbl
0.5 Ib CH4/1000 gal crude
1.02e-05 ton CH4/bbl
1.56 Ib CH4/y/station
0.24 scfCH4/hp-hr
Source
PSI Report14
PSI Report14
1992 API Report5
AP-4228
PSI Report14
AP-4228
PSI Report14
GRI/EPA Study37
-------
TABLE 5-11. EMISSION FACTORS BY SEGMENT
(Continued)
Source Category
REFINING
Emission Factor
Source
Fugitive Sources:
Fuel Gas System
Wastewater Treating
Cooling Towers
Vented Sources:
Tanks
System Slowdowns
Combustion Sources:
Atmospheric Distillation
Vacuum Distillation
Thermal Operations
Catalytic Cracking
Catalytic Reforming
Catalytic Hydrocraking
Catalytic Hydrorefming
Catalytic Hydrotreating
Alkylation & Polymerization
1.02 MMscf CH4/heater
0.00798 Ib Volatile Organic Carbon (VOC)/bbl
0.01 Ib VOC/bbl
4.37e-07 ton CH4/bbl
580 Ib hydrocarbon (HQ/1000 bbl capacity
0.30 Ib total hydrocarbon (THQ/1000 bbl
0.30 Ib THC/1000 bbl
0.50 Ib THC/1000 bbl
0.43 Ib THC/1000 bbl
0.60 Ib THC/1000 bbl
0.60 Ib THC/1000 bbl
0.18 Ib THC/1000 bbl
0.54 Ib THC/1000 bbl
1.05 Ib THC/1000 bbl
Derived using a 1995 EPA Report35
EPA Report39
AP-4228
1992 API Report5
1977 Radian Report40
1980 Radian Report25
1980 Radian Report25
1980 Radian Report25
1980 Radian Report23
1980 Radian Report25
1980 Radian Report25
1980 Radian Report25
1980 Radian Report25
1980 Radian Report25
(Continued)
-------
TABLE 5-11. EMISSION FACTORS BY SEGMENT
(Continued)
REFINING
Source Category
Aromatics/Isomeration
Lube Processing
Asphalt
Hydrogen
Engines
Flares
0.15
0.0
60
0,0
0.24
0.0008
Emission Factor
Ib THC/1000 bbl
Ib HC/ton
scfCH4/hp-hr
Ib VOC/bbl
Source
1980 Radian Report25
1977 EPA Report40
1977 EPA Report40
1977 EPA Report"0
GRI/EPA Study"
1984 Radian Report39
-------
to the vapor phase in the tank. The resulting emissions ranged from 4 to 15 scf CH4/bbl,
compared with an average of approximately 12 scf CH4/bbl from the Canadian study. The
Canadian measurements were also compared to tank measurements taken at seven sites as part of
a recent API/GRI study.32 The measurements at the seven U.S. sites ranged from 3.5 to 148 scf
CH4/bbl, with a mean value of 47.5 scf CH4/bbl and median value of 8.6 scf CH4/bbl). Since the
mean API/GRI emission factor is much higher than the Canadian emission factor, the more
conservative Canadian value (12.1 scf/bbl) was used. It is recognized that this factor currently
does not account for the use of control devices (such as tank vapor recovery systems on sour gas
tanks).32
EFs for pneumatic devices, CIPs, vessel blowdowns, compressor starts, and compressor
blowdowns (all in scf/equipment type) were taken directly from the respective GR1/EPA natural
gas study reports (Pneumatic Device report,33 CIP report,34 and Blow and Purge report35),
Completion flaring, reported as scfd/completion is also from the GRI/EPA natural gas study
(Vented and Combustion Summary report).36 Well workover (scf/workover) is originally from a
December 1989 PSI report14 and is also used in the GRI/EPA natural gas study (Vented and
Combustion Summary report).36 The confidence intervals for each of these sources are based on
the confidence intervals calculated in the GRI/EPA natural gas methane emissions study.
Upsets are considered a vented emission source. These consist of emissions from
emergency shutdown systems (ESD), PRVs, and well blowouts. The BSD EF, reported as
scfy/platform, and methane emissions from PRV lifts, reported scfy/PRV, are both from the
GRI/EPA natural gas study (Blow and Purge report).35 The confidence intervals for these sources
are also from the GRI/EPA natural gas study (Blow and Purge report).35
The well blowout (scf/blowout) EF is estimated by assuming the quantity of gas released
is comparable to the gas production rate of the well (for the GRI/EPA study, the average well
production rate was approximately 125,000 scfd/well) and by assuming the duration of the well
blowout is 48 hours (a 1977 EPA report provided a range of time from 15 minutes to 5 months
but reported that a few days was a typical duration).21 The confidence bound for this source was
assigned based on engineering judgment.
Combustion Emissions—
The EF from gas engines, reported as scf/hp-hr, and the confidence interval are taken
from the GRJ/EPA natural gas study (Compressor report).37 AP-42 reports a methane EF for
burners in lb/1000 gallons (AP-42, Table 1.3-4).28 A confidence interval of 10% was assigned to
this source. The drilling EF (tons/well drilled) came from a 1992 API report, Global Emissions
of Methane from Petroleum Sources,5 The confidence interval was assigned based on
engineering judgment.
42
-------
5.2.2 Crude Transportation
Crude oil is transported from production operations to refineries by tankers, barges, rail
tank cars, tank trucks, and pipelines. Confidence intervals for the fugitive and vented EF sources
in crude transportation were assigned, based on engineering judgment, for all sources except
pump engine drivers. The confidence interval for this source is carried over from the GRI/EPA
natural gas study (Compressor report).37
Fugitive Emissions—
The fugitive EF for crude transportation pump stations (Ib/mile) is taken from a
December 1989 PSI report, Annual Methane Emission Estimate of The Natural Gas and
Petroleum Systems in the United States}4 This source also reported that fugitive methane
emissions from pipelines are negligible.
Vented Emissions—
The EF for crude transportation storage tanks (tons/bbl) is based on an EF determined for
breathing and working losses of refinery storage tanks from an API project.5 Methane EFs from
storage tanks are not readily available, so the API project simplified some assumptions in order
to utilize AP-42 emission estimates. For the purpose of this study, it was assumed that emissions
from refinery crude storage tanks would be similar to storage tanks in crude transport. The
confidence interval was assigned based on engineering judgment.
Methane emissions for the transportation segment result primarily from the loading of
petroleum crude, since vapors in the transportation carriers are displaced to the atmosphere when
the crude oil is loaded. EFs reported by AP-42 were used for truck loading (tons/bbl) and rail car
loading (tons/bbl).28 Emissions from marine vessel loading and unloading, lb/1000 gallons
crude, are from the same PSI report cited above for pump stations.
An EF for the vented emissions of methane from pump station maintenance, reported in
Ib/year/station, is taken from the above-cited PSI report, assuming one station per 100 miles of
pipeline.14
5.2.3 Refining
Methane emissions are not typically reported for refining operations, since methane is not
a regulated hazardous air pollutant (HAP). In addition, by the time crude oil has reached the
refinery, the volatile hydrocarbons such as methane have already flashed off. Fugitive methane
emissions do result from light-end hydrocarbons produced in some of the refinery operations and
from the use of natural gas or refinery still gas in burners and engines. For the purpose of this
study, reported fugitive emissions of VOC or hydrocarbon were used with assumptions that relate
these emissions to methane emissions.
43
-------
Confidence intervals for all EFs except refinery engines, were assigned based on
engineering judgment. The refinery engine confidence interval is carried over from the GRI/EPA
natural gas study (Compressor report).37
Fugitive Emissions—
The fugitive emissions from refinery fuel gas systems were estimated based on
engineering judgment. The component counts of 90 valves and 200 flanges per refinery heater
were based on the following assumptions:
• 20 burners per heater;
• Each burner has a pipe ran from a blended fuel gas header; and
• Each heater has a fuel gas control valve and metering orifice/differential pressure
cell.
The component EFs for valves and flanges were taken from a 1995 EPA report Protocol for
Equipment Leak Emission Estimates?* A methane content of 80 wt% was used.
The fugitive EF for wastewater treatment, reported in Ib VOC/bbl, is taken from an EPA
test program.39 AP-42 Table 5.1-2 was the source for the cooling tower EF, also reported as Ib
VOC/bbl,28 To convert VOC emissions to methane, the assumption was made that methane
makes up 1% of VOC emissions, based on the AP-42 estimate that less than 1% of total
hydrocarbon emissions are methane. Confidence bounds for these sources were assigned based
on engineering judgment.
Vented Emissions—
Methane emissions from refinery tanks were estimated by using the EF reported for crude
transportation (Section 5.2.2). The system blowdown EF, reported as Ib HC/1000 bbl capacity, is
taken from a 1977 Radian report.40 To convert from hydrocarbon emissions to methane
emissions, a methane composition of 1% was used. A confidence bound was assigned based on
engineering judgment.
Combustion Emissions—
Total hydrocarbon emissions (Ib HC/1000 bbl crude oil feed) from process heater flue gas
emissions were reported for the following refinery processes: atmospheric distillation, vacuum
distillation, thermal operations, catalytic cracking, catalytic reforming, catalytic hydrocracking,
catalytic hydrorefining, catalytic hydrotreating, alkylation and polymerization, and
aromatics/isomerization. These EFs are taken from a 1980 Radian report.25
Total hydrocarbon emissions from combustion sources were converted to methane
emissions by assuming a 51% methane composition. This was calculated based on reported
methane compositions of emissions resulting from natural gas (AP-42 Table 1.4-3)28 and fuel oil
44
-------
combustion (AP-42 Table 1.3-4)28 in boilers, where the methane component of the emissions was
ratioed based on the relative amount of fuel oil versus gas (natural gas or still gas) consumed at
refineries.25
Methane emissions from lube processing and hydrogen production processes were
assumed to be negligible,40
The EF for asphalt processes, reported as Ib HC/ton of asphalt produced, is taken from a
refinery system blowdown emission estimate.40
The methane EF developed for production engines, scf/hp-hr, is also used to estimate
methane emissions from refinery engines (GRI/EPA Compressor report).37 The flare EF,
reported as Ib VOC/bbl, is taken from a 1985 Radian report.39 As with vented emissions, the
methane composition for this source is assumed to be 1% of the reported VOC emissions.
45
-------
6.0 RESULTS—1993 BASE YEAR
Presented below in Sections 6,1 through 6.4 are tables for the 1993 methane emission
estimates for production, crude transportation, refining, and the total petroleum industry,
respectively. All calculated confidence bounds represent a precision basis only. See Sections 4.3
and 9.0 for bias considerations. Each section shows the largest emission sources for the industry
segments considered. Refer to Appendix G for a table that can be used to convert the English
system units to metric units.
6.1 PRODUCTION
The production segment emitted 87 Bscf of methane in the 1993 base year. Figure 6-1
shows the largest sources by percentage within the production segment. As shown in the figure,
oil tank venting, pneumatic devices, fugitives from large compressors, and chemical injection
pumps account for over three quarters of the total emissions in the production segment. The
detailed 1993 methane emissions estimate for the production segment is presented in Table 6-1.
Gas Engines
5%
Oil Tank Venting
35%
Large Compressor Fugitives
13%
Pneumatic Devices
17%
Figure 6-1. Production Segment Largest Emission Sources
46
-------
Table 6-1. 1993 METHANE EMISSIONS ESTIMATE
PETROLEUM - PRODUCTION
Emission Source
Annual Production
% Heavy Crude (API<20°)
Total Producing Oil Wells
% Heavy Wells (AP!<20°)
Fugitives:
Offshore Platforms
Gulf of Mexico
Rest of US
Oil Wellheads (heavy crude)
OH Wellheads (light crude)
Separators (heavy crude)
Separators (light crude)
Heater/Treaters (light crude)
Headers (heavy crude)
Headers (light crude)
Tanks (light crude)
Compressors (light crude)
Small
Large
Sales Areas
Pipelines
Venting:
Oil Tanks
Pneumatic Devices
CIPs
Vessel Slowdowns
Compressor Starts
Compressor Slowdowns
Completion Flaring
Well Workover
Casinghead Gas
Upsets:
ESD
PRV Lifts
Well Blowout
Combustion Sources:
Gas Engines
Burners
Drilling
Flares
Total
Emission
Factor
2914
1178
0.83
19.58
0.85
51.33
59.74
0.59
202.78
34.4
46.14
16,360
40.55
56.4
12.1
345
248
78
8443
3774
733
96
256,688
34
250,000
0.24
0.526
0.052
Methane
Emissions Units
scfd CH4/p!atform
scfd CH4/platform
scfd CH4/well
scfd CH4/well
scfd CH4/sep
scfd CH4/sep
scfd CH4/heater
scfd CH4/header
scfd CH4/header
scfd CH4/tank
scfd CH4/comp
scfd CH4/comp
scfd CH4/area
scfd CH4/mile
scf CH4/bbl
scfd CH4/device
scfd CH4/pump
scfy CH4/vessel
scfy CH4/comp.
scfy CH4/comp.
scfd CH4/cornpletion
scf CH4/workover
scfy CH4/plat
scry CH4/PRV
scl CH4/blowout
scfCH4/HPhr
lbCH4/1000gal
ton CH4/well drilled
Confidence
Interval
27%
36%
30%
30%
30%
30%
30%
30%
30%
30%
100%
100%
30%
97%
88%
40%
83%
266%
157%
147%
200%
200%
200%
252%
200%
5%
10%
100%
Activity
Factor
6,846,000
10.7%
583,879
7.1%
1,092
22
41,163
542,716
9,103
113,071
77,354
15,296
47,291
54,272
647
1,940
4,443
70,000
6,846,000
117,003
125,088
205,870
2,799
2,799
390
43,791
1,114
422,936
2.85
17,634
3,647,000
390
Activity
Units
bbl/d
wells
platforms
platforms
wells
wells
separators
separators
heater treaters
headers
headers
tanks
small g.l. cornp.
large g.l. comp.
sales areas
miles
bbl/d
pneumatics
CIPs
sep. and h.t.
gas lift comp.
gas lift comp.
completions
w,o./year
platforms
PRV
blowouts/yr
MMhp-hr
bblfyear
expl. wells
Confidence
Interval
5%
100%
5%
100%
10%
10%
100%
100%
78%
78%
131%
109%
109%
109%
92%
92%
46%
50%
5%
78%
105%
70%
81%
81%
10%
421%
10%
103%
200%
277%
5%
10%
Emissions
(Bscf)
1.161
0.009
0.012
3.879
0.003
2.11B
1.687
0.003
3.500
0.681
0.011
11.585
0.066
1.441
30.235
14.734
1 1 .323
0.016
0.024
0.011
0.104
0.004
0.286
0.014
0.001
4.232
0.002
0.001
87.14
Confidence
Interval
29%
38%
109%
109%
87%
87%
140%
118%
118%
118%
164%
164%
57%
119%
88%
93%
160%
333%
218%
206%
201 %
962%
201%
376%
490%
277%
11%
101%
48.3%
-------
6.2 CRUDE TRANSPORTATION
The crude transportation segment emitted 1.4 Bscf of methane in the 1993 base year.
Figure 6-2 shows the largest sources by percentage within the crude transportation segment.
Marine unloading and tank venting account for the majority of emissions. The detailed 1993
emission estimate for crude transportation is shown in Table 6-2.
Tank Venting
14%
Other
3%
Marine Unloading
83%
Figure 6-2. Crude Transportation Segment Largest Emission Sources
48
-------
TABLE 6-2. 1993 METHANE EMISSION ESTIMATE
PETROLEUM - CRUDE TRANSPORTATION
Emission Source
Fugitives:
Pump Stations
Pipelines
Metering
Venting:
Tanks
Loading
Truck
Marine
Rail Car
Maintenance:
Pump Stations
Combustion Sources:
Pump engine drivers
Heaters
Total
Emission
Factor
1.06
0.0
4.37E-07
1 .02E-05
0.5
1 .02E-05
1.56
0.24
Methane
Emissions Units
Ib CH4/yr/mi!e
ib CH4/bbI
ton CH4/bbl
ton CH4/bbl
ibCH4/1 000 ga! crude
ton CH4/bbl
ib CH4/y/station
scf CH4/HPhr
Confidence
Interval
100%
10%
100%
100%
100%
100%
100%
5%
Activity
Factor
55,268
6.71 E+09
9.07E+09
7.69E+07
9.54E+10
8,91 E+06
553
Activity
Units
miles
bbl/yr
bbl/yr
bbl/yr
gal/yr
bbl/yr
stations
Confidence
Interval
100%
5%
4%
10%
10%
10%
100%
Emissions
(Bscf)
0.0014
0.0000
0.188
0.037
1.132
0.004
0.000
1.362
Confidence
Interval
173%
11%
100%
101%
101%
101%
173%
85.1%
-------
6.3 REFINING
The refining segment emitted 9.2 Bscf of methane in the 1993 base year. Figure 6-3
shows the largest sources by percentage within the refining segment. Engine exhaust emissions
and fugitive emissions account for the majority of emissions. The detailed 1993 methane
emissions estimate for the refining segment is presented in Table 6-3.
System Slowdowns
7%
Other
4%
Fuel Gas System Fugitives
35%
Engines
54%
Figure 6-3. Refining Segment Largest Emission Sources
50
-------
TABLE 6-3. 1993 METHANE EMISSION ESTIMATE
PETROLEUM - REFINING
Emission Source
Fugitives:
Fuel Gas System
Pipe Stills
Wastewater Treating
Cooling Towers
Venting:
Tanks
System Slowdowns
Process Vents
Upsets
PHVs
Combustion Sources:
Process Heaters:
Aim, Distillation
Vacuum Distil.
Thermal Operations
Cat. Cracking
Cat. Reforming
Cat. Hydrocraking
Cat. Hydrorefining
Cat. Hydrolreating
AlkyI & Polymer.
Aromatics/Isomeration
Lube Processing
Asphalt
Hydrogen
Coke
Engines and Flares
Engines
Flares
Total
Emission
Factor
1.02
0.00798
0.01
4.37E-07
580
0.30
0,30
0.50
0.43
0.60
0.60
0.18
0.54
1.05
0.15
0.0
60
0.0
0
0.24
0.0008
Methane
Emissions Units
MMscf CH4/heater/yr
tb VOC/bbI
Ib VOC/bbI
ton CH4/bbl
# HC/1000 bbc capacity
Ib THC/1000 bbl
IbTHC/IOOObbl
Ib THC/1000 bbl
IbTHC/IOOObbl
IbTHC/IOOObbl
IbTHC/IOOObbl
IbTHC/IOOObbl
IbTHC/IOOObbl
IbTHC/IOOObbl
IbTHC/IOOObbl
# HC/ton
Included in Thermal Ops
scf CH4/hp-hr
Ib VOC/bbI
Confidence
Interval
100%
100%
100%
100%
100%
100%
100%
100%
100%
100%
100%
100%
100%
100%
100%
100%
100%
100%
100%
5%
1 00%
% Methane
in THC*
1.0%
1.0%
1.0%
51.0%
51.0%
51.0%
51.0%
51.0%
51.0%
51.0%
51.0%
51.0%
51.0%
51.0%
1.0%
Activity
Factor
3,200
13,612,259
13,612,259
13,612,259
13,612,259
13,612,259
5,935,032
1,661,140
4,694,106
3,287,291
1,112,414
1,595,163
7,326,166
1 ,003,670
693,791
177,624
631,440
20,334
13,612,259
Activity
Units
heaters
b/d
b/d
b/d
b/d
b/d
b/d
b/d
b/d
b/d
b/d
b/d
b/d
b/d
b/d
b/d
b/d
MMhp-hr
b/d
Confidence
Interval
50%
5%
5%
5%
5%
5%
5%
5%
5%
5%
5%
5%
5%
5%
5%
5%
5%
100%
5%
Emissions
(Bscf)
3.26
0.009
0.012
0.103
0.684
0.018
0.008
0.004
0.009
0.009
0.003
0.001
0.018
0.005
0.000
0.000
0.167
0.000
4.880
0.001
9.191
Confidence
Interval
122%
100%
100%
100%
100%
100%
100%
100%
100%
100%
100%
100%
100%
100%
100%
100%
100%
100%
100%
69.2%
% Methane in VOC (volatile organic compounds) taken from AP-42 (Reference 28)
% Methane in HC (hydrocarbons) for system blowdowns is taken from AP-42
% Methane in THC (total hydrocarbons) calculated based on data from AP-42
% Methane in HC for asphalt calculated based on data from AP-42
-------
6.4 TOTAL INDUSTRY
The total petroleum industry (production, crude transportation, and refining) emitted 98
Bscf of methane in the 1993 base year. Presented below is Table 6-4, the 1993 methane
emissions estimate for all three industry segments combined. Figure 6-4 shows the emissions by
type, and Figure 6-5 shows the total industry percentage of emissions attributable to each
segment. As shown in Figure 6-4, vented emissions are the largest type of emission. When
emissions are presented by segment, the production segment accounts for the vast majority of all
emissions.
TABLE 6-4. 1993 PETROLEUM METHANE EMISSION ESTIMATE—TOTAL
OF THREE INDUSTRY SEGMENTS
Segment
Annual Emissions, Bscf
Production
Crude Transportation
Refining
TOTAL
87.1 ±48%
1.36 ±85%
9.19 ±69%
97.7 ± 44%
Combustion
10%
Fugitive
30%
Vented
60%
Figure 6-4. Emissions by Type
52
-------
Crude Transportation
1%
Refining
9%
Production
90%
Figure 6-5. Percent Emissions by Segment
53
-------
7.0 METHANE EMISSIONS—1986-1992
After the base-year emission estimate was constructed for 1993, estimates were made to
cover the years 1986 through 1992. It should be noted that the 1986 through 1992 estimates have
the same limitations that apply to the 1993 estimate. Section 7.1 covers the methods used to
make the historical estimates, and the results are presented in Section 7.2.
7.1 METHOD FOR HISTORICAL ESTIMATES
Activity (AFs) and emission factors (EFs) were examined for potential changes that could
have occurred between 1986 and 1993. AFs, such as well count and crude production rates, were
known to have changed during the period. However, potential changes to EFs required analysis.
Potential EF changes could have resulted from maturing domestic production fields, technology
changes in all segments, operating and maintenance practices in all segments, and applied
emission controls. An analysis of these potential EF effects reveals that none of these factors had
a significant impact on emissions. Therefore, EFs are assumed to have remained unchanged
from 1986 to 1993, as explained in the following paragraphs.
Minimal technology changes or changes due to maturing fields are assumed to have
occurred from 1986 to 1993, since the oil production and domestic refining industry was already
very mature (over 50 years old) in 1986. Some maturing oil fields may have required additional
artificial lift over this period, where an increased use of gas lift would contribute to higher
emission rates. However, there is no method available to estimate those changes over the period,
and they were assumed to be negligible. In general, most capital investment of energy
production companies in production and refining has been overseas since the mid-1980s, which
reduces the application of new technologies domestically, especially technology that would affect
emission rates. Therefore, these potential technology factors are believed to have had little or no
impact on EFs.
The primary operating practices that could affect EFs are leak detection and repair
(LDAR) programs, which minimize fugitive emissions, and maintenance changes that affect gas
equipment blowdowns. LDAR programs did not exist in the production or crude transportation
segments during the 1986 to 1993 period. LDAR programs began in refineries in the 1980s,
since many were located in non-attainment urban areas. However, refinery LDAR programs do
not target methane, so refinery fuel gas systems, the primary source of refinery methane
emissions in this study's estimate, would not be affected. Maintenance practices, the primary
element of which is compressor blowdown, are also assumed to have remained constant during
this time period. Therefore, these potential operating practice changes had little or no impact on
EFs.
For this study, it was assumed that applied emission controls were not in use or had no
effect during the 1986 to 1993 period. Although EPA's Maximum Achievable Control
Technology (MACT) standards will soon require control technologies for some of the petroleum
industry, they were not in effect during the 1986 to 1993 time period. Even when MACT
54
-------
standards do become official and enforced, their effect on methane emissions is not certain, since
the MACT is aimed at other hazard air pollutants (HAPs), not methane.
On the basis of this analysis, EF changes during the 1986 to 1993 time period are
negligible. The primary method for reverse estimates to 1986 was to use known changes in AFs,
In the production segment, the primary AFs are oil wells and oil production rate, which are
known for each year in the 1986 to 1993 period.12'15 The well count and production rate are also
used to extrapolate the production segment equipment counts from the site visit data. These
activity factors therefore changed the equipment counts over the period, even though the site visit
data set remained unchanged. Completion wells and wells drilled were also known and changed
over that time period,19'41 The number of well workovers was based on the total number of
wells,12
In the crude transportation segment, the AFs for the volume of crude transported by mode
of transportation was known for each year and was therefore adjusted over the 1986 to 1993
period.11-22
In the refinery segment, the crude charge rate and the utilization factor were known for
each year and were therefore adjusted over the 1986 to 1993 period,15'42 The refinery engine AF
was not adjusted for each year. Instead, the value reported for 1993 is based on an average of the
values that resulted for each year, accounting for the energy requirements for the various refinery
processes and the energy equivalent of the fuel consumed at refineries for each year (shown in
Table 7-1 and based on calculations presented in Section 5.1.5).
TABLE 7-1. REFINERY ENGINE ACTIVITY FACTOR FOR 1986-1993,
Year
1993
1992
1991
1990
1989
1988
1987
1986
Refinery Energy
Requirements,
MMBtu
2.071E+09
2.007E+09
1.964E+09
2.080E+09
2.058E+09
2.021E+09
1.911E+09
1.877E+09
Average Refinery Engine Activity
Refinery Fuel
Consumption,
MMBtu
2.211E+09
2.213E+09
2.175E+09
2.199E+09
2.159E+09
2.135E+09
2.049E+09
2.101E+09
Factor, MMhp-hr
Refinery Engine
Activity Factor,
MMhp-hr
18,221
26,780
27,410
15,441
13,118
14,814
17,857
29,032
20,334
This AF is based on the difference between two large numbers, such that a small
difference in either the refinery fuel usage or refinery energy usage for a particular year has a
large impact on the estimated engine fuel use AF, Owing to the uncertainties resulting from the
calculation approach for this AF and because this is a large emission source, the results were
55
-------
misleading. For example, the AF difference between 1986 and 1993 would result in a decrease
in refinery emissions of approximately 2.6 Bscf from this single source. Since the year to year
change in hp-hrs is believed to be due only to year to year errors in fuel use, the hp-hr value was
held constant across the years examined.
These AF changes were used in the estimation of methane emissions for 1986 through
1992, the results of which are shown in the following section.
7.2 RESULTS
Table 7-2 provides a summary of the total emissions by industry segment for each year of
this study. Appendix H (Tables H-l through H-21) show the detailed emission results for the
years 1986 through 1992 of this study. The net emissions changed very little over the period:
There were 110.1 Bscf of emissions in 1986, compared with 97.7 Bscf of emissions in 1993.
Production segment emissions were actually higher in 1986, due to the larger number of
domestic oil wells and oil production rate in 1986.
TABLE 7-2. EMISSION SUMMARY FOR 1986-1993.
Methane Emissions, Bscf
Year Production Transportation Refining Total
1993 (Base Year)
1992
1991
1990
1989
1988
1987
1986
87.1
89.6
92.6
91.3
92.7
96.1
97.8
99.8
1.36
1.32
1.30
1.28
1.30
1.26
1.18
1.11
9,19
9.19
9.12
9.19
9.19
9.18
9.13
9.12
97.7 + 43.6%
100.1 ±43.8%
103,0 ±44.0%
101,8 ±43.9%
103.2 ±44.0%
106.5 ±44.3%
108.1 ±44.5%
11 0.0 ±44.7%
56
-------
8.0 CONCLUSIONS
As presented in Section 6, the total methane emissions estimate from the U.S. petroleum
industry is 98 Bscf for the base year 1993. This estimate is believed to be accurate to
approximately +/- 100%. Accuracy, which is comprised of precision and bias components,
cannot be rigorously calculated, given the limitations of the data. While precision of the estimate
for 90% confidence bounds was calculated to be only +/- 44%, there may be some unquantified
bias resulting from use of the limited data set. Possible contributors to bias are listed in
Section 4.3 and Section 9.0 of this report. This bias can be ruled out or corrected in the
following phases of effort. Figure 8-1 shows the relative contribution of the petroleum segment
to the total anthropogenic emissions of methane in the United States, based on methane emission
estimates from EPA1 and GRI/EPA sources.2 According to this 1998 EPA study, petroleum
sources could account for 3 to 4 times as much methane as estimated previously by EPA (both
the April 1993 and November 1995 reports).4'1 The updated higher emission estimates presented
here still only account for less than 1% of total greenhouse gas emissions, when CO2 emissions
are considered.1
Fossil Fuel
Livestock Manure Consumpdon
8%
3%
Natural Gas Systems
19%
Other 1%
Coal Mining
13%
Landfills
31%
Oil Systems
6%
Rice Cultivation
1%
Domesticated
Livestock
18%
Figure 8-1. Sources of Anthropogenic Methane Emissions (Updated)
Table 8-1 shows methane emission estimates for the U.S. petroleum industry from four
previous studies compared to this 1996 EPA-ORD study.
57
-------
TABLE 8-1. ANNUAL METHANE EMISSION ESTIMATES FOR U.S. PETROLEUM
INDUSTRY FROM FIVE DIFFERENT STUDIES (Bscf)
Base Year Crude
of Report Production Transportation Refining Total
API, 19925
API, 19966
EPA, 19934
EPA, 19951
EPA, 1998
1987-1989
1990
1990
1993
1993
0.6
38.9
6.1-25,3"
6.1-25.3"
87.1
0.8
0.5
0.3
0.3
1.4
4.5
0.7
0.5
0.5
9.2
5.9
40.1
6.9-26.1
6.9-26.1
97.7
a Production segment includes field fugitive emissions, field routine maintenance emissions, crude oil
storage facility emissions, and venting and flaring.
The 1992 API study provided a global estimate using the base years 1987 to 1989.5 The
1996 API report provided an updated estimate for 1990 methane emissions.6 The 1996 study is
higher primarily due to adding tank emissions. Both studies included these three segments:
• Production;
• Crude transportation; and
• Refining.
The 1993 EPA study presented an estimate for all U.S. sources of manmade methane
emissions.4 Of these sources, the study estimated petroleum emissions to be approximately 1.1%
of total methane emissions, or between 6.9 and 26.1 Bscf per year. The study accounted for six
sources of petroleum emissions:
• Production field fugitive emissions;
• Production field routine maintenance emissions;
• Crude oil storage facility emissions;
• Refineries;
• Marine vessel operations; and
• Venting and flaring.
The EPA numbers came from various sources, including a 1991 draft report of the GRI/EPA
natural gas study. For comparative purposes, the six sources listed above were regrouped into
production, crude transportation, and refining.
The 1995 EPA study presented all greenhouse gas emissions and sources.1 The
1995 EPA study directly used the results of the 1993 EPA study.4 The report states that
anthropogenic methane constitutes approximately 11.3% of total greenhouse gas emissions.
Petroleum emissions were divided into the same six categories as in the 1993 EPA study.
58
-------
This 1998 EPA-ORD study provides an initial estimate that is more detailed than other
previous efforts. While it may have some biases, the previous reports were also biased in the use
of broader, more general estimates that did not address all possible emission sources. In fact, the
previous studies did not perform data gathering nor measurements, and none used an equipment
level of detail. Instead., broad segment-wide emission factors were used, which tend to
underestimate emissions.
Although measurements were not performed in this study, this report does draw on new
measurements unavailable prior to this effort, such as measurements made for the GR1/EPA
natural gas study. This 1998 EPA report also uses new detailed data, such as the production site
visit database, that allow equipment level AF estimates that were not possible previously,
The 1992 API study was limited to the use of a few broad assumptions, and identified
only a few of the sources of methane emissions in the industry. The 1996 API study primarily
used the United Nations Intergovernmental Panel on Climate Change (IPCC) Greenhouse Gas
Protocol,43 which has a generic, undetailed method of estimating methane emissions that ignores
some known sources. In fact, the major difference between the estimates in the first and second
API study is that the second study added production tank emissions to the IPCC protocol.
The 1993 and 1995 EPA studies are identical, since the 1995 study relies entirely on the
1993 study. These two EPA studies were not performed on an equipment detail level; thus,
many sources were overlooked, such as production tank flash emissions, compressor fugitive
emissions, CIPs, refinery fuel gas systems, compressor exhaust emissions, etc. The reports also
used a "vented and flared" term that has since been shown to have some data quality concerns
(see the Vented and Combustion Summary Report of the GRI/EPA natural gas study).36
The conclusion from comparison of previous efforts to this effort is therefore that the
emissions from the petroleum industry may be much higher than previously estimated. Further
study will be required to verify this initial conclusion. The results of this project show a
confidence bound of ± 43.6%. In reality, this confidence bound represents precision only. As
discussed in Section 3, there is an assumption in any project that the bias term in accuracy is
zero. However, there are some potential biases that have been identified, which, if real, would
change the emission estimate. Many of these potential biases are discussed in Section 9 on future
efforts.
If these emission data are ultimately used to analyze the global warming impact of
emissions associated with domestic consumption of oil, it may be necessary in the future to add
an analysis of foreign emissions from the production of oil imported into the United States, This
would raise the total emissions associated with U.S. oil consumption. In addition, it may be
necessary to add emissions from downstream segments, refined product transportation,
marketing, and end use, so that a total life-cycle analysis is included in the global warming
analysis.
59
-------
9.0 FUTURE EFFORTS
This report attempts to improve upon earlier methane emission estimates for the
petroleum industry by examining emission sources on an equipment level of detail. However,
the basis for this Phase I estimate can be improved in future efforts. This section outlines key
assumptions and key data issues, and provides recommendations for future updates. As with any
analysis, if key assumptions are incorrect, the estimate could be biased to some degree. All the
assumptions are believed to be reasonable and correct, but the estimates should only be used as
guidelines for further study. Additional field data gathering, field measurement programs, and
data analysis could eliminate potential bias issues and lead to improved accuracy in future
refinements.
Key data issues are also identified where the data set is small, and where a larger data set
would add confidence to the overall estimate. This is the case for many items in this report, since
no measurement efforts were conducted for this study.
Sections 9.1 through 9.3 serve as a sensitivity analysis on the key issues for each segment
of the industry. In general, the current estimate is highly sensitive to the assumptions listed in the
following sections. The recommendations in Section 9.4 can be used to develop future
improvement projects.
9.1 PRODUCTION SEGMENT KEY ISSUES AND ASSUMPTIONS
Production segment key issues are described below. Each paragraph presents a new issue
and its potential impact. There are key assumptions that, if incorrect and then corrected, would
increase the emission estimate, and others that would decrease the emissions estimate. Some
issues require further efforts, while others do not. Recommendations regarding these key issues
are summarized in Section 9.4.
Production emissions resulted primarily from four major sources:
1) Oil tanks;
2) Pneumatic devices;
3) Large compressor fugitive emissions; and
4) Chemical injection pumps.
Therefore the estimate is very sensitive to assumptions that affect these categories.
Production segment AFs were assumed to be bounded by the definition of the petroleum
segment industry boundaries shown in Figure 3-2. This report used the identical production
segment boundaries defined in the GRI/EPA natural gas study. This approach ensures that when
the results of this study are combined with the results from the GRI/EPA natural gas study, all
production segment emissions are counted and none are double-counted. Selection of these
boundaries directly affects the equipment counts that are attributed to the petroleum industry.
60
-------
The limited site visit data set used to generate production equipment count AFs is not
assumed to be completely representative of the United States petroleum production segment. For
example, the production rate per well is high in the sampled data set compared to the known
production rate per well, and there is an over-representation of gas-lift sites and offshore oil
wells. Major limitations in the existing site database include the following:
1) A complete set of equipment counts was not available for all of the sites;
2) The limited database lacks information to stratify data based on regional differences
or operational differences (e.g., differences between equipment associated with
heavy versus light crude production or differences in equipment associated with
stripper wells);
3) The production data set was not generated by a random sample, but instead from oil
sites coincidentally visited during the GR1/EPA natural gas industry study.
A more detailed site data collection effort was not conducted in this Phase 1 study. Therefore,
adjustments were made to the existing production data for use in this study (described further in
Section 5.1.2). For most equipment, the data set extrapolation was corrected by using an
arithmetic mean of the equipment counts determined by well count versus production rate. In
addition, a correction factor was applied to the extrapolated count of gas-lift compressors, which
if determined to be not appropriate, could increase the total emission by 25.7 Bscfy for the base
year 1993.
Other potential production site sampling biases could significantly lower the estimate.
For example, the majority of oil wells in the United States are low production rate, marginally
profitable wells called stripper wells. The field activity factor data set did include stripper wells
with pneumatic devices and chemical injection pumps supplied by natural gas pressure. If these
were not representative of typical stripper wells, then future phases would estimate lower
emissions from these two sources.
Most production AFs (equipment types) are assumed to be related to both well count and
production rate. Extrapolating by well counts produces a much higher AF than the production
extrapolation. If future technical analysis could prove the exact relation between the equipment
counts and well counts or production rate, the AF estimates would change. ,
This report has extrapolated production AFs using the "ratio method" specified in the
GRI/EPA natural gas study (Statistical Methods report).9 In order to be consistent with the
GRI/EPA natural gas study, this project has used the identical extrapolation method. Although
this method is believed to be the appropriate technique, the method weights large sites (sites with
many wells and more production per well) more than small sites. If this technique were
incorrect, the AF estimates and the emissions could change.
This project has assumed that the available data on heavy crude versus light crude
production from 1976-1981 are applicable to the years 1986 through 1993. Fugitive equipment
61
-------
counts (for wellheads, separators, heater/treaters, headers, tanks, and gas lift compressors) were
split between those with heavy crude production and light crude production, since there are
different published emission factors for each type (i.e., higher EFs for light crude production). If
the ratio of heavy crude production to total production for the time period of this project is
significantly different than that during 1976-1981, then the emission estimates would change.
Large gas compressors in the petroleum industry (particularly production) were assumed
to have the same characteristics as compressors in gas transmission, which were measured in the
GRI/EPA natural gas study. Large compressors (those with similar equipment setups as
transmission compressor stations) were found to have very high fugitive emissions in the gas
industry. However, no emission measurements of large compressors in production are readily
available. If large production compressors were not similar to transmission compressors,
emissions could be more than 11.5 Bscf lower. In addition, this report has assumed that 75% of
the compressors in the production segment are large compressors, based on the fact that most of
these are gas lift compressors, and several industry sources believed that all were large, housed
stations, or gas plants. If these assumptions were incorrect, the emission estimate would be
affected.
This report's production data set contained compressors primarily associated with gas lift.
If there is a large number of compressors associated with other artificial lift methods, such as
CO2 flood, then the combustion and fugitive emissions associated with compressors would
increase.
The 30 Bscf of oil tank emissions are the largest single source of emissions in production,
so any bias in this category will have a very large effect. In fact, the tank-vented EF is based on a
Canadian program consisting of only five measurements.31 If this was an inaccurate sample, the
emission estimate would change.
Some miscellaneous emission sources were assumed to be negligible and are not
currently accounted for by any national system. Negligible sources in production include vented
casinghead gas, vented oil well gas production, and burner and flare flame-out (these sources are
listed in Table 6-1, but no EFs or AFs were estimated). If these assumptions were incorrect,
emissions would increase.
9.2 CRUDE TRANSPORTATION SEGMENT KEY ISSUES AND ASSUMPTIONS
Transportation is a small contributor to methane emissions. Therefore the key issues in
transportation are relatively minor compared with production. Recommendations regarding
some of these issues are summarized in Section 9.4.
No AF was estimated for pump engine drivers, owing to lack of data. Combustion
emissions from gas engines in the production and refining segments were significant. Therefore,
this source for transportation could also be significant.
62
-------
The pipeline station count AF was based on an assumption of one station per 100 miles.14
If this were incorrect, the emission estimate would have to be changed. In addition, the number
of intrastate pipeline miles and the volume of crude transported by intrastate pipelines are not
included in the estimated number of miles, number of pump stations, and volume of crude
transported by pipelines. Accounting for the intrastate pipelines would increase the emission
estimates.
Pipeline fugitive methane emissions are assumed negligible based on past reports and oil
industry experience. Metering and heaters were also identified as emission sources in Table 6-2,
but are believed to be negligible. If this were incorrect, the emission estimate would increase.
9.3 REFINING SEGMENT KEY ISSUES AND ASSUMPTIONS
This project assumed that there are no significant fugitive emissions of methane in the
refinery except in the fuel gas system. This is based on an assumption that the only significant
concentration of methane in the refinery is in the fuel gas system. However, other units that
handle or generate light ends (such as pipe stills and light end units) may have methane in
concentrations high enough to generate measurable methane emissions.
This project has assumed that there is no methane in atmospheric process vents at the
refinery. Therefore, no methane EFs or AFs were estimated for pipestills, process vents and
PRVs, If this were incorrect, the emission estimate would increase.
The largest estimated source of methane emissions in the refinery is from gas engine
exhaust (unburned fuel). This project has assumed that engine fuel use in the refinery can be
calculated by an energy balance of all fuel driven process equipment. Currently the estimate is
based on fuel usage by refineries minus the amount accounted for by heat input to process
heaters. By difference, an estimated fuel use for compressors results. On a national basis, the
accuracy of this estimate is limited by the inherent problems associated with the difference
between two large values. Since this is the largest emission source in refineries, an error in this
method could lead to a significant difference in the estimated emissions.
A national composition of methane in the refinery was estimated for fugitive emissions.
The assumption that methane comprises 1% of VOC emissions was used, which could be
conservatively high since AP-42 suggests that the methane component of total hydrocarbon
emissions is less than 1%.28 The methane composition of combustion sources was estimated to
be 51%, based on the relative quantities and compositions of the various refinery fuels.28 If
more exact average compositions were determined in the future, this might decrease emissions.
9.4 RECOMMENDED FUTURE TEST PLAN
Future efforts aimed at improving the estimate presented in this report should center on
data gathering, measurement, and data analysis. An initial approach to data gathering might be to
63
-------
establish a voluntary industry review panel that would provide data, provide sites for
measurement, and occasionally meet to review the underlying assumptions and work produced.
As was mentioned in Section 8, a general trend observed in methane emission estimates
is that estimates that lack supporting data tend to underestimate emissions. Over time, as
detailed activity data and emission measurements are taken, the estimates rise and plateau when a
more accurate answer is reached. This is similar to a learning curve effect. This has been the
experience with the projects estimating methane emissions from the gas industry, and this trend
is also reflected in methane emissions for the petroleum industry as shown in Table 8-1.
This project shows that the production segment has emissions an order of magnitude
higher than the combined emissions from refining and transportation. Although this relative
comparison is probably accurate, it should be noted that the production segment has the most
data available, and therefore may be further along the learning curve of emission estimates.
Although it is tempting to concentrate all future efforts in the segment of the industry showing
the highest current estimate, such action may prevent the project from reaching a reasonable
degree of accuracy in all the segments. The sampling philosophy established for the GRI/EPA-
ORD natural gas industry project was to focus on large emission categories and large
uncertainties. The philosophy even involved establishing target accuracies for every single
source category. This requires an assumption that all of the sources are well known and that all
that is required is refinement of precision. This petroleum industry project currently has
considerably less data than the gas industry project, and still has some bias concerns. Therefore
the petroleum project cannot adopt a detailed target accuracy sampling approach.
This report recommends that some additional work be conducted in each petroleum
industry segment. The following subsections identify specific AF and EF data gathering efforts
for each segment of the industry. The additional work will allow unknown sources to be
discovered,
9.4.1 Production Segment Improvements
The production segment of the industry has the highest emissions of methane. The
following descriptions list recommendations in order of importance.
Activity Factor Site Visits—
One of the most important issues in the production segment is to eliminate potential
biases in the production site visit database. This can be accomplished by collecting additional
production site data, based on randomly selected sites across the nation. These sites may be
provided by the volunteer participants in the industry panel mentioned earlier, or may be directly
solicited by a future project team. Company databases, if available from other efforts such as air
permit emissions programs, could also be employed where offered. Another sampling goal
would be to add regions of the United States as a strata for the AFs, and sample oil production
64
-------
sites within those regions. The key regions identified during the gas industry production segment
could be used for this analysis.
A sampling plan can be established from the regional approach or by using a recognized
oil industry database. Although using a database could be an expensive approach, it would be
the most robust. This would require further investigation.
Compressor Measurements—
Another key issue is the large compressor EFs and AFs. Since emissions from large
production compressors have never been measured, a recommended future action is to conduct a
production field fugitive sampling effort for compressors using screening and direct
measurement devices, In addition, site visit efforts can concentrate on additional oil production
sites and/or use company databases that verify whether the assumed fraction of large compressors
is correct.
Tank Measurements—
Production tank vented emissions, which account for approximately 31 % of the total
1993 industry emissions estimate, should be refined in the future. While this can be
accomplished through additional field measurements, sampling programs are very expensive and
time intensive, and the wide variability resulting from field characteristics could make
representative sampling difficult. Therefore, tank vented emissions can best be updated by a
modeling and activity data gathering effort.
To support a future tank emission update effort, national activity data on crude production
should be gathered as input to a tank emissions modeling program.32 The activity data should
include the following:
1) Stratification of crude production in the U.S. into homogeneous groups (or regions)
of similar API gravity, Reid Vapor Pressure, and sweetness;
2) Average separator pressures and temperatures for the same groups; and
3) Average tank controls applied for each group.
The results of such a modeling effort could then be used to replace the current estimate of
national methane emissions from oil tanks.
Miscellaneous—
For improved fugitive estimates, the split of light and heavy crude production should be
updated. Some national databases on production parameters exist, but are proprietary and often
require the user to purchase the database. Alternatively, state data can be examined. The Texas
Railroad Commission reports similar data for their oil leases, which could be used to determine a
65
-------
light and heavy crude split. Likewise, if the other states with heavy crude production track this
information, each state's data could be analyzed to update the data or validate the assumptions.
9.4.2 Crude Transportation Segment Improvements
Crude transportation is estimated to be a small contributor to methane emissions, but
some additional work is recommended. First, a better characterization of the crude transportation
segment should be made. For example, the characterization should define exactly what types of
equipment are associated with crude transportation terminals, pipeline pump stations, vessel and
car loading, and unloading terminals. These data are not currently available. Also, a more
accurate count of pipeline miles (including intrastate mileage) may be available from a national
Geographic Information System (GIS) database, such as the Pennwell Map database.
9.4.3 Refining Segment Improvements
Current estimates indicate that refining is a relatively small contributor to methane
emissions. However, very little information of methane emissions from refineries exists, and
therefore most emission estimates from refineries are based on simplifying assumptions
identifying one or two potential sources of methane. Future efforts should center on
characterizing refinery units for potential fugitive emissions as well as potential point source
emissions.
If refineries participate in future efforts, they may be able to provide component
speciation data for individual process vents in refineries. Although methane is not a regulated
pollutant, and therefore may not be measured directly, methane concentrations might have been
measured by difference, since total hydrocarbon (THC) and non-methane hydrocarbon (NMHC)
concentrations are often determined.
Future efforts can update the fugitive emissions estimate by validating this project's fuel
gas system component count assumptions. Obtaining actual fuel gas system counts from
participating companies may be possible, since these companies may have produced counts for
their air permit emission inventories. In addition, overlooked sources for fugitives may be added
if participating companies add methane to fugitive emissions data gathering efforts for refineries
in specific areas where methane is expected (fuel gas, light ends, pipe stills).
Finally, data could be collected from participating refiners on gas-driven compressor
counts and compressor fuel usage to validate the assumptions made by this report. Data for
estimating methane emissions from internal combustion engines used at refineries may also be
available through a national emissions inventory database used to track other air emissions.
66
-------
10.0 REFERENCES
1, U.S. Environmental Protection Agency. Inventory of U.S. Greenhouse Gas Emissions
and Sinks: 1990-1994, EPA-230/R-96-006 (NTIS PB96-175997). Office of Policy,
Planning and Evaluation, Washington, DC, November 1995.
2, Harrison, M.R., L.M. Campbell, T.M. Shires, and R.M. Cowgill. Methane Emissions
from the Natural Gas Industry, Volume 2: Technical Report, Final Report, EPA-600/R-
96-OSOb (NTIS PB97-142939). U.S. Environmental Protection Agency, Air Pollution
Prevention and Control Division, Research Triangle Park, NC, June 1996.
3, Hileman, B. "Climate Observations Substantiate Global Warming Models," Chemical &
Engineering News, November 27,1995, pp. 18-25.
4. U.S. Environmental Protection Agency. Anthropogenic Methane Emissions in the United
States Estimates for 1990, Report to Congress. EPA-430-R-93-003. Office of Air and
Radiation, Washington, DC, April 1993.
5. Radian Corporation. Global Emissions of Methane From Petroleum Sources. American
Petroleum Institute, Health and Environmental Affairs Department, Report No. DR140,
February 1992.
6. Harrison, M.R. and T.M. Shires. Methane and Carbon Dioxide Emission Estimates from
U.S. Petroleum Sources, Final Report. API Publication 4845. American Petroleum
Institute, January 1997.
7. Stapper, B.E. Methane Emissions from the Natural Gas Industry, Volume 5: Activity
Factors, Final Report, EPA-600/R-96-080e (NTIS PB97-142962). U.S. Environmental
Protection Agency, Air Pollution Prevention and Control Division, Research Triangle
Park, NC, June 1996.
8. Bell, L. "Worldwide Refining - Survey of Operating Refineries in the U.S. (State
Capacities as of January 1, 1994)." Oil and Gas Journal, December 20, 1993, p. 50.
9, Williamson, H.J., M.B. Hall, and M.R. Harrison. Methane Emissions from the Natural
Gas Industry, Volume 4: Statistical Methodology, Final Report, EPA-600/R-96-080d
(NTIS PB97-142954). U.S. Environmental Protection Agency, Air Pollution Prevention
and Control Division, Research Triangle Park, NC, June 1996.
10. Cochran, William G. Sampling Techniques, Third Edition, New York, NY: John Wiley
& Sons, 1977.
11. True, W.R., "U.S. Interstate Pipelines Ran More Efficiently in 1994." Oil and Gas
Journal, November 27, 1995, p. 56.
67
-------
12. World Oil, "Producing Oil Well Numbers Still Dropping." World Oil, February 1994,
p. 70.
13. Star Environmental, API Report No. 4615. Emission Factors for Oil and Gas
Production Operations. American Petroleum Institute, January 1995.
14. Tilkicioglu, B.H. and D.R. Winters. Annual Methane Emission Estimate of The Natural
Gas And Petroleum Systems in the United States. Pipeline Systems Incorporated (PSI),
December 1989.
15, Beck, R.J, "Economic Growth, Low Prices to Lift U.S. Oil and Gas Demand in 1995."
Oil and Gas Journal, January 30, 1995, pp. 54-64.
16. Interstate Oil Compact Commission (IOCC). Major Tar Sand and Heavy Oil Deposits of
the United States, IOCC, Oklahoma City, OK, 1984.
17. Clegg, Joe D., S. Mike Bucaram, and N.W. Hein, Jr. Recommendations and
Comparisons for Selecting Artificial-Lift Methods. Journal of Petroleum Technology,
pp. 1128-1131, 1163-1167. 1993.
18. ICF Inc. Estimation of Activity Factors for Gas E&P Facilities, Final Report.
U.S. Environmental Protection Agency, Office of Air and Radiation, Washington, DC,
July 12, 1995.
19. Energy Information Administration (EIA). Annual Energy Review 1994, "Table 4.5 Oil
and Gas Exploratory Wells, 1949-1994." EIA, Office of Oil and Gas, U.S. Department of
Energy, DOE/ELV0384(94), Washington, DC, July 1995.
20. Energy Information Administration (EIA). Petroleum Supply Annual 1993 Volume 1,
"Table 2. U.S. Supply, Disposition, and Ending Stocks of Crude Oil and Petroleum
Products, 1993." EIA, Office of Oil and Gas, U.S. Department of Energy, Washington,
DC, May 1994.
21. Braxton, C., R.H. Stephens, and M.M. Stephens. Atmospheric Emissions from Offshore
Oil and Gas Development and Production. EPA-450/3-77-026 (NTIS PB 272-268). U.S.
Environmental Protection Agency, Office of Air Quality Planning and Standards,
Research Triangle Park, NC, June 1977.
22. Energy Information Administration (EIA). Petroleum Supply Annual 1993 Volume 1,
"Table 46. Refinery Receipts of Crude Oil by Method of Transportation by PAD District,
1993." EIA Office of Oil and Gas, U.S. Department of Energy, Washington, DC, May
1994.
68
-------
23. Beck, R.J. "Economic Growth to Raise U.S. Oil Products, Natural Gas Demand." Oil
and Gas Journal, January 31, 1994, p. 55.
24. Sanderford, E.B. Alternative Control Techniques—NOX Emissions from Process Heaters.
EPA-453/R-93-034 (NTIS PB94-120235). U.S. Environmental Protection Agency,
Office of Air Quality Planning and Standards, Research Triangle Park, NC, September
1993.
25. Wetherold, R.G., and D.D. Rosebrook. Assessment of Atmospheric Emissions from
Petroleum Refining. Volumes 1, 2, 3, 4, 5. EPA-600/2-80-075a-075e (NTIS PB80-
225253, 80-225261, 80-225279, 81-103830, and 80-225287), 1980.
26. Energy Information Administration (EIA). Petroleum Supply Annual 1993 Volume 1,
"Table 47. Fuels Consumed at Refineries by PAD District, 1993." EIA, Office of Oil and
Gas, U.S. Department of Energy, Washington, DC, May 1994.
27. Energy Information Administration (EIA). Annual Energy Outlook 1995, Appendix I,
p. 173, Washington, DC, 1995.
28. U.S. Environmental Protection Agency. Compilation of Air Pollutant Emission Factors:
Volume I: Stationary Point and Area Sources, AP-42 (GPO 055-000-005-001), Office of
Air Quality Planning and Standards, Research Triangle Park, NC, Fifth Edition,
January 1995.
29, Hummel, K.E., L.M. Campbell, and MR. Harrison. Methane Emissions from the Natural
Gas Industry, Volume 8: Equipment Leaks, Final Report, EPA-600/R-96-080h (NTIS
PB97-142996). U.S. Environmental Protection Agency, Air Pollution and Prevention
Control Division, Research Triangle Park, NC, June 1996.
30. Campbell, L.M., M.V. Campbell and D.L. Epperson. Methane Emissions from the
Natural Gas Industry, Volume 9: Underground Pipelines, Final Report, EPA-600/R-96-
080i (NTIS PB97-143002). U.S. Environmental Protection Agency, Air Pollution and
Prevention Control Division, Research Triangle Park, NC, June 1996.
31. Picard, D.J., B.D. Ross, and D.W.H. Koon. A Detailed Inventory ofCH4 and VOC
Emissions from Upstream Oil and Gas Operations in Alberta Volume III: Results of the
Field Validation Program, Canadian Petroleum Association, March 1992.
32. Radian International LLC. Evaluation of a Petroleum Production Tank Emission Model,
Final Report. GRI-97/0117, American Petroleum Institute and Gas Research Institute,
May 1997.
69
-------
33. Shires, T.M. and M.R. Harrison, Methane Emissions from the Natural Gas Industry,
Volume 12: Pneumatic Devices, Final Report, EPA-600/R-96-0801 (NTIS PB97-143036).
U.S. Environmental Protection Agency, Air Pollution Prevention and Control Division,
Research Triangle Park, NC, June 1996.
34. Shires, T.M. Methane Emissions from the Natural Gas Industry, Volume 13: Chemical
Injection Pumps, Final Report, EPA-600/R-96-080m (NTIS PB97-143044). U.S.
Environmental Protection Agency, Air Pollution Prevention and Control Division,
Research Triangle Park, NC, June 1996.
35. Shires, T.M. and M.R. Harrison. Emissions from the Natural Gas Industry, Volume 7:
Blow and Purge Activities, Final Report, EPA-600/R-96-080g (NTIS PB97-142988).
U.S. Environmental Protection Agency, Air Pollution Prevention and Control Division,
Research Triangle Park, NC, June 1996.
36. Shires, T.M. and M.R. Harrison. Methane Emissions from the Natural Gas Industry,
Volume 6: Vented and Combustion Source Summary, Final Report, EPA-600/R-96-080f
(NTIS PB97-142970). U.S. Environmental Protection Agency, Air Pollution Prevention
and Control Division, Research Triangle Park, NC, June 1996.
37. Stapper, CJ. Methane Emissions from the Natural Gas Industry, Volume 11: Compressor
Driver Exhaust, Final Report, EPA-600/R-96-080k (NTIS PB97-143028). U.S.
Environmental Protection Agency, Air Pollution Prevention and Control Division,
Research Triangle Park, NC, June 1996.
38, Epperson, D.L. Protocol for Equipment Leak Emission Estimates, 1995. EPA-453/R-95-
017 (NTIS PB96-175401). U.S. Environmental Protection Agency, Office of Air Quality
Planning and Standards, Research Triangle Park, NC, November 1995.
39. Wetherold, R.G., G.E. Harris, F.D. Skinner, and L.P. Provost. Model for Evaluation of
Refinery and Synfuels VOC Emission Data, Vol. I, EPA-600/7-85-022a (NTIS PB85-
215713). U. S. Environmental Protection Agency, Industrial Environmental Research
Laboratory, Research Triangle Park, NC, May 1985.
40. Burklin, C.E. Revision of Emission Factors for Petroleum Refining. EPA-450/3-77-030
(NTIS PB275-685). U.S. Environmental Protection Agency, Office of Air Quality
Planning and Standards, Research Triangle Park, NC, October 1977.
41. Energy Information Administration (EIA). Natural Gas Production Responses to a
Changing Market Environment. EIA, Office of Oil and Gas, U.S. Department of Energy,
DOE/EIA-0532, Washington, DC, May 1990.
42. Bell, L. Survey of Operating Refineries in the U.S. Oil and Gas Journal, Annual
Refinery Report, December 1986-1992. (One issue for each year).
70
-------
43. Intergovernmental Panel on Climate Change (IPCC). Greenhouse Gas Inventories: IPCC
Guidelines for National Greenhouse Gas Inventories. United Nations Environment
Programme, the Organization for Economic Co-operation and Development, the
International Energy Agency, and IPCC, Vols 1-3, Braknell, U.K. 1995.
71
-------
-------
APPENDIX A
Results of Literature Search
A-l
-------
APPENDIX A
LITERATURE SEARCH RESULTS
A comprehensive methodology review was conducted for this project to identify all
previous studies that have produced estimates or studies that have described methodologies for
estimating methane emissions for the petroleum industry. Information was gathered from
internal sources, an extensive on-line literature search, and contacts with key experts. The
literature search covered the time period from 1975 to the present. The keyword search strategy
was formed using combinations of the following:
Oil/petroleum industry Methane emissions
Oil/petroleum refineries/refining Greenhouse gases
Exploration/production VOC (volatile organic compound) emissions
Oil/petroleum transportation Hydrocarbon emissions
Emissions
As shown, the keywords chosen were fairly general, such that as many possible sources remotely
related to emissions from the petroleum industry would be identified. Extensive abstract listings
were reviewed to identify all sources applicable to this study. A total of 54 reports (listed in
Table A-l) were identified as potentially having some applicability to emissions from the
petroleum industry.
TABLE A-l. LITERATURE SEARCH RESULTS
References/Resources
Arthur D. Little. Methane Emissions from the Oil and Gas Production Industries. Final Report. Reference No.
63193. Ruhrgas A.G., July 1989.
American Petroleum Institute. Atmospheric Hydrocarbon Emissions from Marine Vessel Transfer Operations,
American Petroleum Institute Publication 2514A, Washington DC, 1981.
American Petroleum Institute, Basic Petroleum Data Book: Petroleum Industry Statistics. Volume VII, Number
3, Washington DC, September 1987.
American Petroleum Institute. "Evaporative Loss from Fixed-Roof Tanks", API Bulletin 2518, API Manual of
Petroleum Measurement Standards. First Edition, Washington DC, March 1993.
American Petroleum Institute. Hydrocarbon Emissions From Refineries. Publication No. 928. American
Petroleum Institute, Committee on Refinery Environmental Control, Washington DC, July 1973.
Bell, L. "Worldwide Refining - Survey of Operating Refineries in the U.S. (State Capacities as of January 1,
1990)," Oil & Gas Journal, March 26, 1990, p. 78.
Burklin, C.E. and R.L. Honercamp, Revision of Evaporative Hydrocarbon Emission Factors, Final Report.
EPA-450/3-76-039 (NTIS PB 267-659), U.S. Environmental Protection Agency, Office of Air Quality Planning
and Standards, Research Triangle Park, NC, August 1976.
(Continued)
A-2
-------
TABLE A-l. (Continued)
References/Resources
Burklin, C.E, Revision of Emission Factors for Petroleum Refining, EPA-450/3-77-030 (NTIS PB 275-685),
U.S. Environmental Protection Agency, Office of Air and Waste Management, Research Triangle Park, NC,
October 1977.
DeLuchi, M.A., "Emissions from the Production, Storage, and Transport of Crude Oil and Gasoline," Journal of
Air and Waste Management Association, November 1993, Volume 43, pp, 1486-1495.
Dubose, D.A., J.I. Steinmetz, and G.E. Harris, Onshore Production of Crude Oil and Natural Gas: Fugitive
Volatile Organic Compound Emission Sources - Data Analysis Report Frequency of Leak Occurrence and
Emission Factors for Natural Gas Liquid Plants. EMB Report No. 80-FOL-l, U.S. Environmental Protection
Agency, Office of Air Quality Planning and Standards, Research Triangle Park, NC, July 1982.
Energy Information Administration. Emissions of Greenhouse Gases in the United States 1987-1994.
DOE/EIA-0573(87-94), Energy Information Administration, Office of Integrated Analysis and Forecasting,
Washington DC, October 1995.
Energy Information Administration. Natural Gas Production Responses to a Changing Market Environment,
1978-1988. DOE/EIA-0532. Energy Information Administration, Office of Oil and Gas, Washington DC, 1989.
Energy Information Administration, Petroleum Supply Annual, 1993, Volumes 1 and 2, DOE/EIA-0340(93)/1,
Energy Information Administration, Washington DC, June 1994.
Espenshade, S.A. and G. Rhines. "The Oil and Natural Gas Producing Industry in Your State," Petroleum
Independent. Independent Petroleum Association of America, Washington DC, September 1993.
Frazicr, N.A., D.L. Maasc, and R. Clark, Offshore Oil and Gas Extraction - An Environmental Review, EPA-
600/7-77-080 (NTIS PB 272242), Industrial Environmental Research Lab, Resource Extraction and Handling
Division, Cincinnati, OH, July 1977.
Gibbs, M.J., P. Hathiramani, and M. Webb, Fugitive Methane Emissions From Oil and Gas Production and
Processing Facilities: Emissions Factors Based on the 1980 API-Rockwell Study. ICF Consulting Associates
and STAR Environmental, Universal City, CA, April 1992.
Intergovernmental Panel on Climate Change. IPCC Guidelines for National Greenhouse Gas Inventories:
Volume 3 Greenhouse Gas Inventory Reference Manual. United Nations Environment Programme, the
Organisation for Economic Co-operation and Development, the International Energy Agency, and the
Intergovernmental Panel on Climate Change. Braknell, UK, 1995.
Interstate Oil & Gas Compact Commission. History of Production Statistics: Production and Reserves 1971-
1991. Oklahoma City, OK, 1992.
Klett, M.G., and J.B. Galeski, Flare Systems Study, EPA-60Q/2-76-079, (NTIS PB 251-664), U.S. Environmental
Protection Agency, Industrial Environmental Research Laboratory, Research Triangle Park, NC, March 1976,
Kuuskraa, V,A. and M.L, Codec, A Technical and Economic Assessment of Domestic Heavy OH, Interstate oil
Compact Commission, Oklahoma City, OK, 1987.
(Continued)
A-3
-------
TABLE A-l. (Continued)
References/Resources
LaFlam, G.A , Revision of Marine Vessel Evaporative Emission Factors, Pacific Environmental Services, Inc.,
Durham, NC, November 1984.
LaFlam, G.A., S. Osbourn, and R.L. Norton, Revision of Tank Truck Loading Hydrocarbon Emission Factors,
Pacific Environmental Services, Inc., Durham, NC, May 1982.
Lemlin, J.S., and I. Graham-Bryce, "The Petroleum Industry's Response to Climate Change: The Role of the
IPIECA Global Climate Change Working Group," UNEP Industry and Environment, January-March, 1994, pp.
27-30.
Lipton, S,, "Fugitive Emissions," Chemical Engineering Progress, June 1989, Volume 85, Number 6, pp. 42-47.
Mussig, S., Possibilities for Reduction of Emissions - in Particular the Greenhouse Gases C03 and CH4 - in the
Oil and Gas Industry, Society of Petroleum Engineers, Presented at the European Petroleum Conference,
Cannes, France, November 16-18, 1992.
Pennwell Publishing Company. Oil and Gas Journal Databook, 1990 Edition, Tulsa, OK, March 1990.
Picard, D.J., B.D. Ross and D.W.H. Koon. A Detailed Inventory' ofCH4and VOC Emissions From Upstream Oil
And Gas Operations in Alberta. Canadian Petroleum Association, Calgary, Alberta, March 1992.
Radian Corporation. Assessment of Atmospheric Emissions from Petroleum Refining: Volumes 1-4. EPA-600/
2-80-075a through EPA-600/2-80-075d (NTIS PB 80-225253, PB 80-225261, PB 80-225279, and PB 81-
103830). U.S. Environmental Protection Agency, Industrial Environmental Research Laboratory, Research
Triangle Park, NC, July 1980,
Radian Corporation. Development of Fugitive Emission Factors and Emission Profiles for Petroleum Marketing
Terminals, Volumes 1 and 2. American Petroleum Institute, Washington DC, May 1993.
Radian Corporation. Global Emissions of Carbon Dioxide From Petroleum Sources, American Petroleum
Institute Health and Environmental Affairs, Departmental Report Number DR 141, Washington DC, July 1991.
Radian Corporation. Global Emissions of Methane From Petroleum Sources, American Petroleum Institute
Health and Environmental Affairs, Departmental Report Number DR 140, Washington DC, February 1992.
Radian Corporation. Oil and Gas Field Emission Survey, Final Report, EPA-600/R-92-083, U.S. Environmental
Protection Agency, Air and Energy Engineering Research Laboratory, Research Triangle Park, NC, May 1992.
Radian International LLC. Methane Emissions from the Natural Gas Industry, Volumes 1-15, Final Report.
EPA 600/R-96-080a through EPA 600/R-96-080o (NTIS PB 97-142921, etc.). Gas Research Institute and U.S.
Environmental Protection Agency, June 1996.
Reister, D.B. An Assessment of the Contribution of Gas to the Global Emissions of Carbon Dioxide (February-
December 1983), Final Report. Oak Ridge Associated Universities, Institute for Energy Analysis, GRI-84/0003.
Gas Research Institute, Research and Development Operations Division, Chicago, IL, June 1984.
(Continued)
A-4
-------
TABLE A-l. (Continued)
References/Resources
Rosebrook, D.D., R.G. Wetherold, and L.P. Provost, The Development of Fugitive Emission Factors for the
Petroleum Refining Industry, Presented at the 72nd Annual Meeting of the Air Pollution Control Association,
Cincinnati, OH, June 24-29, 1979.
Rosenberg, E.S., Impact of the "EON" Rule on the Petrochemicals and Refining (Industries). National
Petroleum Refiners Association. Presented at the 1993 NPRA Annual Meeting, San Antonio, TX, March 21-23,
1993.
Shah, A. And P. Pope, Methods for Estimating Atmospheric Emissions from E&P Operations. Report No.
2.59/197, E&P Forum, London, UK, September 1994,
Braxton, C, R.H. Stephens, and M.M. Stephens. Atmospheric Emissions from Offshore Oil and Gas
Development and Production, Energy Resources Company, EPA 450/3-77-026 (NTIS PB 272268),
Environmental Protection Agency, Office of Air and Waste Management, Research Triangle Park, NC, June
1977.
Taback, H.J., G, Lauer, L.K. Gilmer, and K. Ritter, Strategies for Improving HAP Emission Factors and
Profiles for the Petroleum Industry, Presented at the 85th Annual Meeting and Exhibition of the Air and Waste
Management Association, Kansas City, MO, June 21-26, 1992.
Taback, H.J. and K. Ritter, 1994 Fugitive Emissions Estimating Data for Petroleum Industry Equipment Leaks,
Presented at the Air and Waste Management Association and California Air Resources Board 5th Annual West
Coast Regional Specialty Conference, Sacramento, CA, November 9-10, 1994.
Tilkicioglu, B.H, Annual Methane Emission Estimate of the Natural Gas Systems in the United States. Radian
Corporation, EPA Prime Contract No. 68-02-4288. Austin, TX, September 1990.
U.S. Department of Commerce. 7992 Census of Mineral Industries; Crude Petroleum and Natural Gas. MIC92-
1-13A, U.S. Department of Commerce Economics and Statistics Administration, Washington DC, April 1995.
Harris, G.E., K.W. Lee, S.M. Dennis, C.D. Anderson and D. L. Lewis. Assessment ofVOC Emissions from Well
Vents Associated With Thermally Enhanced Oil Recovery, EPA-909/9-81 -003 (NTIS PB-82-134750), Region 9,
San Francisco, CA, September 1981.
U.S. Environmental Protection Agency. Compilation of Air Pollutant Emission Factors: Volume I Stationary
Point and Area Sources, AP-42, (GPO 055-000-005-001), U.S. Environmental Protection Agency, Office of Air
Quality Planning and Standards, Fifth Edition, January 1995.
U.S. Environmental Protection Agency. Development Document for Effluent Limitations Guidelines and
Standard for the Petroleum Refining Point Source Category (Proposed), U.S. Environmental Protection Agency
Effluent Guidelines Division, Washington DC, December, 1979.
U.S. Environmental Protection Agency. Estimating Air Toxics Emissions From Coal and Oil Combustion
Sources, EPA-450/2-89-001 (NTIS PB89-194229), Research Triangle Park, NC, April 1989.
(Continued)
A-5
-------
TABLE A-l. (Continued)
References/Resources
U.S. Environmental Protection Agency, Emissions of Producing Oil and Gas Wells, EPA-908/4-77-006 (NTIS
PB 283-279), Region 8, Denver, CO, November 1977.
U.S. Environmental Protection Agency. Independent Quality Assurance of Refinery Fugitives Testing By
Western States Petroleum Association. Final Audit Report. EPA-454/R-93-033 (NTIS PB 94-131232), U.S.
Environmental Protection Agency, Office of Air Quality Planning and Standards, Research Triangle Park, NC,
September 1993.
U.S. Environmental Protection Agency. Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-1994.
EPA-230-R-96-006 (NTIS PB 96-175997). U.S. Environmental Protection Agency, Office of Policy, Planning
and Evaluation, Washington DC, November 1995.
Viswanath, R.S., and J. H. Van Sandt, Oilfield Emissions of Volatile Organic Compounds, EPA-450/2-89-007
(NTIS PB 89-194286), U.S. Environmental Protection Agency, Office of Air Quality Planning and Standards,
Research Triangle Park, NC, April 1989.
U.S. Environmental Protection Agency. Options for Reducing Methane Emissions Internationally Volume 1:
Technological Options for Reducing Methane Emissions, Report to Congress, EPA 430-R-93-006,
Environmental Protection Agency, Office of Air and Radiation, Washington DC, July 1993.
Webb, M. And P. Martino, Fugitive Hydrocarbon Emissions from Petroleum Production Operations, Presented
at the 85th Annual Meeting and Exhibition of the Air and Waste Management Association, Kansas City, MO,
June 21-26, 1992.
Wetherold, R.G., G.E. Harris, F.D, Skinner, and L.P. Provost. A Model for Evaluation of Refinery and Synfuels
VOC Emission Data: Volumes 1 and 2. Final Report. EPA-600/7-85-022a/b (NTIS PB 85-215713 and
PB 85-215721) U.S. Environmental Protection Agency, Air and Energy Engineering Research Laboratory,
Research Triangle Park, NC, September 1984.
World Meteorological Organization. Climate Change, 1990. Intergovernmental Panel on Climate Change,
United Nations Environment Programme, 1990.
Each report was reviewed for industry boundary definitions (i.e., which equipment types
of emission sources were considered part of the petroleum industry), level of detail,
representativeness, comprehensiveness, and data quality. As a secondary goal of this project, the
studies were also searched for VOC emission estimates; however, the scope of the project was
later revised to focus only on methane emissions. A summary worksheet and emission source
checklist were used to simplify the report review process, so that the information from each
report could be summarized in a consistent format. A blank summary worksheet and emission
source checklists for each industry segment are shown in Tables A-2 and A-3, respectively.
A-6
-------
TABLE A-2. METHODOLOGY REVIEW - SUMMARY SHEET
REPORT/STUDY
Report Title:
BOUNDARIES
US Specific?
Petroleum Industry
DETAIL LEVEL/COMPREHENSIVENESS
EF and/or AF Estimate
Industry Segment (Production,
Transportation, Refinery)
Equipment Type
REPRESENTATIVENESS
Based on Petroleum Industry?
Specific to Methane
Specific to VOCs
Other (THC) - What's required to generate
methane EF?
DATA QUALITY
Year(s)
Data Basis - Measurements, guess?
Are modifications required to use data?
Accuracy - Can it be calculated, guessed?
Data Quality Ranking or Estimate
COMMENTS
A-7
-------
TABLE A-3. EMISSION SOURCE CHECKLIST/SUMMARY
PRODUCTION
Overall Site Emission Estimate
Tanks
Flares -Well completion
Fugitives
Separators
Heaters
Compressors
Metering/Sales
Wells
Pipeline
Pumps
Offshore Platforms
Heaters
Burner
Vent
Pneumatic Devices
Chemical Injection Pumps
Compressor Exhaust - Gas Lift
Maintenance
Vessel Blowdown
Well Workovers
Compressor Starts
Metering/Sales
Heaters
Pumps
Upsets
Pressure Relief Valves
ESD/EBD
Other Engines
Other
(Continued)
A-8
-------
TABLE A-3. Continued
TRANSPORTATION
Overall Emission Estimate
Tanks
Heaters
Pumps
Fugitives - Pipeline Pump Station
Components
Loading/Unloading
Tank Cars
Rail Cars
Barges
Pneumatic Devices
Maintenance
Upsets
Fuel Consumption
Other
Mobile source or end use - not considered here.
Overall Site Emission Estimate
Atmospheric Crude Distillation
Vacuum Crude Distillation
Naphtha Hydrotreating
Middle Distillate Hydrotreating
Gas Oil Hydrotreating
Vacuum Resid. Hydrodcsulfurization
Catalytic Reforming
Aromatics Extraction
Catalytic Cracking
Hydrocracking
Thermal Cracking
Delayed Coking
Fluid Coking
Light Ends Recovery & Fractionation
Other Fractionati on
REFINING
(Continued)
A-9
-------
TABLE A-3. Continued
Alkylation
Polymerization
Isomerization
Lube Oil Processing - Solvents
Other Lube Oil Processing
Asphalt Production
Hydrogen Production
Gasoline Treating
Other Product Treating
Olefins Production
Other Volatile Petrochemicals
Low Volatility Petrochemicals
Blowdown System
Wastewater Collection & Treating
Sludge/Solids Handling
Storage - Fixed Roof Tanks
Storaee - Floatins Roof Tanks
Cooling Towers
Loading
Combustion Sources
Boiler
Flares
Heaters
Compressor/Engine
Tower
Maintenance
Pneumatics
Other
A-10
-------
After completing the worksheets and checklists for each report, a database was set-up to
rank the reports on the basis of the criteria listed above. The database facilitated maintaining a
record of any emission factors and activity factors for each source category available in each
report. This provided a mechanism to quickly scan the available resources and identify data gaps
where emission factors or activity factor data did not exist. By using the database, the reports
were sorted by industry segment (production, transportation, and refining) and ranked based on
their applicability. The database results are shown in Tables A-4 through A-7, where Table A-4
presents a summary of the reports, and Tables A-5 through A-7 are the more detailed databases
corresponding to each industry segment.
On the basis of the review of existing literature, it became clear that emission factors did
not exist in sufficient form to fully characterize methane emissions from the petroleum industry.
Many sources simply did not report methane emissions. As a result of this analysis, the project
scope shifted from an emission inventory compilation to one in which methane emission factor
and activity factors were developed and estimated.
A-ll
-------
TABLE A-4a. SUMMARY OF REPORT DATABASE
Report Title
"Atmospheric Emissions from Offshore Oil and Gas
Development and Production", EPA 450/3-77-026, June 1977,
"Atmospheric Hydrocarbon Emissions from Marine Vessel
Transfer Operations," Publication 251 4A, American Petroleum
Institute, Washington, DC, Sept. 1981.
"Global Emissions of Carbon Dioxide from Petroleum Sources,"
Prepared by Radian for API, July 1991.
"Worldwide Refining," Oil and Gas Journal, Dec. 21, 1992, p
84.
Radian Corporation, "Assessment of Atmospheric Emissions
from Petroleum Refining: "Volumes I - IV" Prepared for U.S.
Environmental Protection Agency. Research Triangle Park, NC.
July 1980.
Radian, Tier 2 Report for the GRfBPA Methane Emissions
Project, June 1995.
Wctherold, R.G., G.E. Harris, F.D, Skinner, and L.P. Provost
(Radian Corporation). "A Model for Evaluation of Refinery and
Synfuels VOC Emission Data; Volumes 1 and 2. Research
Triangle Park, NC.
"Development cf Fugitive Emission Factors and Emission
Profiles for Petroleum Marketing Terminals," Volume I and II,
API, Washington DC, May 1993.
"Global Emissions of Methane From Petroleum Sources,"
Prepared by Radian for API, February 1992.
"Offshore Oil and Gas Extraction - An Environmental Review,"
Battelle Columbus Labs, Prepared for Industrial Environmental
Research Lab, July 1977.
U.S.
Specific?
Yes
Yes
Worldwide
Yes
Yes
Yes
Yes
Yes
Worldwide
Yes
(Mainland
and Alaska)
Petroleum
Industry
Based?
Oil and Gas
Yes
Yes
Yes
Yes
Natural Gas
Industry
Yes
Yes
Yes
Oil and Gas
EF and/or
AF Estimate
EF
EF
Both
AF
Some of both
Both (AFs for
Natural Gas)
EF, Default
AFs
EF
Both
EF
Indus try
Segment
Production
Transportation
Combination
Combination
Refinery
Production
Refinery
Transportation
Combination
Production
Emission Source
Offshore
Marine Loading
Various
Production and
Refineries
Various
All production
equip, except
tanks
Fugitive
Components
Fugitives
Various
Exploration and
Drilling
Emission
Type
Total
Hydrocarbon
(THC)
THC
CO,
None
Non-methane
Methane
Volatile
Organic
Carbon
(VOC)
THC
Methane
Methane
Year(s) Data
Gathered
1975, 1985
1977
1987-1990
1993
1975-1978
Base year 1992
(1991-1995)
19S5
7
1987-1990
1965-1975
(Continued)
-------
TABLE A-4a. SUMMARY OF REPORT DATABASE
(Continued)
Report Title
Petroleum Supply Annual 1993, Volumes 1 and 2, Energy
Information Administration, US DOE, Washington DC, June
1994.
Picard, D.J., B.D. Ross, D.W.H. Koon. "A Detailed Inventory
of CH4 and VOC Emissions From Upstream Oil And Gas
Operations in Alberta," Canadian Petroleum Association,
Calgary, Alberta, 1992.
Radian Corporation. "Study of Refinery Fugitive Emissions
from Equipment Leaks," API Health and Environmental
Sciences Dcpt. and Western States Petroleum Assoc., Volumes
1 and 2, April 1994.
Rosebroak, D.D., R.G. Wetherald, and L.P. Provost, "The
Development of Fugitive Emission Factors for the Petroleum
Refining Industry", Presented at the 72nd Annual Meeting of
the Air Pollution Control Association, June 1979.
Tilkicioglu, E.H., "Annual Methane Emission Estimate of the
Natural Gas Systems in the United States, Phase 2," Pipeline
Systems Inc., Sept, 1990.
"Assessment of VOC Emissions from Well Vents Associated
With Thermally Enhanced Oil Recovery", EPA 909/9-81-003,
Sept. 1981,
"Compilation of Air Pollutant Emission Factors (AP^2)," US
EPA,
"Emissions of Greenhouse Gases in the United States 1987-
1994," Energy Information Administration, November 1995,
"Emissions of Producing Oil and Gas Wells", EPA 908/4-77-
006, November 1977.
"Fugitive Methane Emissions from Oil and Gas Production and
Processing Facilities. Emission Factors Based on the 1980 API-
Rockwell Study," Prepared for U.S. EPA, Prepared by STAR
Environmental, April 1992.
U.S.
Specific?
Yes
Canadian
Data
Yes
Yes
Yes
Yes
Yes
Yes
Yes, Colorado
Yes
Petroleum
Industry
Based?
Yes
Oil and Gas
Yes
Yes
Some Oil
Fields
Yes
Yes
Oil and Gas
Oil and Gas
Yes
EF and/or
AF Estimate
AF
EF
EF
EF
Site ER
EF
EFs
EF (ER)
EF
EF discussed,
no numbers.
•
Industry
Segment
Combination
Production
Refinery
Refinery
Production
Production
Combination
Combination
Production
Production
Emission Source
Supply and
disposition data
Various
Fugitives
Fugitives
Well vents
Various
Not Specific
Wells
Fugitives
Emission
Type
NA
Methane,
VOC
THC
Non-methane
HC
Methane
Methane,
VOC
TOC, THC
Methane
THC
THC, non-
methane HC
Year(s) Data
Gathered
1993 and earlier
1989
1993
1979
1989
1978-1980
Most recent
version
1987-1992
1976
1980's
(Continued)
-------
TABLE A-4a. SUMMARY OF REPORT DATABASE
(Continued)
Report Title
"Oilfield Emissions of Volatile Organic Compounds", EPA-
450/2-89-007, April 1989.
C.E. Burkliii, and R-L. Honercamp, "Revision of Evaporative
Hydrocarbon Emission Factors," EPA-450/3-76-039, U.S.
Environmental Protection Agency, Research Triangle Park, NC,
August 1976.
C.E, Burklin, et ah, Oil and Gas Field Emission Survey,
USEPA, ContractNo, 68-DI-0031, April 1992.
DeLuchi, M.A,, "Emissions from the Production, Storage, and
Transport of Crude Oil and Gasoline," Journal of Air and Waste
Management Association, Nov. 1993, pp. 1486-1495.
DuBose, D.A,, J.I. Steinmetz, and G.E. Harris, "Frequency of
Leak Occurrence and Emission Factors for Natural Gas Liquid
Plants," Prepared for EPA, July 1982.
Houghton, J.T., G.J. Jenkins, and J.J. Ephraums. 1990.
"Climate Change: The IPCC Scientific Assessment. Report
prepared for IPCC by Working Group I," Intergovernmental
Panel on Climate Change, Press Syndicate, University of
Cambridge.
M.G. Klett and J.B. Galesti, "Flare Systems Study," EPA-
600/2-76-079, U.S. Environmental Protection Agency,
Research Triangle Park, NC, March 1976.
N.F. Suprenanl, et al., "Emissions Assessment of Conventional
Stationary Combustion Systems, Volume V: Industrial
Combustion Sources," EPA Contract No. 68-02-2197, GCA
Corporation, Bedford, MA, October 1980.
"An Assessment of the Contribution of Gas to the Global
Emissions of Carbon Diexide Final Report (February-December
1983)", GRI, June 1984.
"Anthropogenic Methane Emissions in the United States;
Estimates for 1990," Report to Congress, EPA 430-R-93-003,
April 1993,
U.S.
Specific?
Yes
Yes
Yes
Yes
Yes
Global
Yes
Yes
US and
worldwide
Yes
Petroleum
Industry
Based?
Yes
Yes
Yes
Yes
No, gas plants
All sources
Yes
Yes
Natural Gas
and Oil
Some data
EF and/or
AF Estimate
EF?
EF
Both
EF
EF
ER
EF
EF
EF for Carbon
Some of both
Industry
Segment
Production
Combination
Combination
Combination
Production
Production
Combination
Refinery
Production
Combination
Emission Source
Wellhead &
tanks
Fugitives
Overall
production ER
Flares in
production and
refining
Combustion
Sources
Emission
Type
Non-methane
HC
Hydrocarbon
(HC)
Total Organic
Gas
VOC
VOC, non-
methane/non-
ethane
Methane and
other GHGs
THC
THC
Carbon
Methane
Year(s) Data
Gathered
1984-1985
1976
Not clear
2000
i
1988
1958, 1973
1978
1984
Base year 1990
(Continued)
-------
TABLE A-4a. SUMMARY OF REPORT DATABASE
(Continued)
Report Title
"Evaporation Loss from Fixed-Roof Tanks", API Bulletin 2518,
June 1962, Reaffirmed August, 1987.
"Hydrocarbon Emissions from Refineries," API Publication No,
928, American Petroleum Institute, Washington, DC, July
1973.
"Independent Quality Assurance of Refinery Fugitives Testing
by Western States Petroleum Association," EPA Office of Air
Quality Planning and Standards, September 1993.
"Methane Emissions from the Oil and Gas Production
Industries," Ruhrgas A.C., July 1989.
"Options for Reducing Methane Emissions Internationally
Volume 1: Technological Options for Reducing Methane
Emissions," Report lo Congress, EPA 430-R-93-006, July
1993.
Basic Petroleum Data Book, Petroleum Industry Statistics Vol.
VII, No 3, API, Sept. 1987.
C.R Burklin, et al., "Revision of Emission Factors for
Petroleum Refining," EPA-450/3-77-030, U.S. Environmental
Protection Agency, Research Triangle Park, NC, October 1977.
Estimating Air Toxics Emissions From Coal and Oil
Combustion Sources, EPA-450/2-89-001, U.S. Environmental
Protection Agency, Research Triangle Park, NC, April 1989.
Kantor, R.H., "Trace Pollutants from Petroleum and Natural
Gas Processing," Prepared for EPA by M.W. Kellogg Co., June
1974.
Lemlin, J.S., I. Graham-Bryce, "The Petroleum Industry's
Response to Climate Change: The Role of the IPIECA Global
Climate Change Working Group," UNEP Industry and
Environment, Jan-Mar. 1994, pp 27-30.
U.S.
Specific?
Yes
Yes
Yes
Worldwide
Worldwide
Yes
Yes
Yes
Yes
Global
Petroleum
Industry
Based?
v
Yes
Yes
Yes
Oil and Gas
Oil and Gas
Yes
Yes
Oil burned by
users
Oil and Gas
Yes
EF and/or
AF Estimate
EF
EF
EF
EFandAF
Some of both
Some AFs
EF
EF
Some AFs
Neither
Industry
Segment
Refinery
Refinery
Refinery
Combination
Production
Combination
Refinery
Production
Emission Source
Tanks
Refining
Equipment
Fugitives
Well testing,
emergencies in
Prod, and
Refining
Fugitives,
pneumatics,
compressors
Boilers
Emission
Type
Loss reported
in barrels
HC
Non-methane
Organic
Carbon, Air
Toxics, THC
Methane
Methane
None
THC
Air Toxics,
trace metals
None
None
Year(s) Data
Gathered
1961
1973
1992-1993
I9S7
1991
1947-1986
1972-1977
1986
1972
(Continued)
-------
TABLE A-4a. SUMMARY OF REPORT DATABASE
(Continued)
• !!••••!• IIMI1 •••III ll .1 1^^^-^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^—
Report Title
Lipton, Sydney, "Fugitive Emissions," Chemical Engineering
Progress, Vol. 85, No. 6, pp. 42-47, June 1989,
Mussig, S., etai., "Possibilities for Reduction of Emissions - in
Particular the Greenhouse Gases CO2 and CH4 - in the Oil and
Gas Industry," Presented at the European Petroleum
Conference, Nnv. 1992.
R,F. Boland, et al., Screening Study for Miscellaneous Sources
of Hydrocarbon Emissions in Petroleum Refineries, USEPA,
Dec. 1976.
Rosenberg, E.S., "Impact of the "HON" Rule on the
Petrochemicals and Refining (Industries)," Presented at the
1993 NPRA Annual Meeting, March, 1993.
Taback, HJ,, G. Lauer, L.K. Gilmer, and K. Ritter, "Strategies
for Improving HAP Emission Factors and Profiles for the
Pelroleum Industry," Presented at the 85th Annual Meeting and
Exhibition of the Air and Waste Management Association, June
1992.
Taback, HJ., and K. Ritter, "1994 Fugitive Emissions
Estimating Data for Petroleum Industry Equipment Leaks,"
Presented at the Air and Waste Management Association and
California Air Resources Board 5th Annual West Coast
Regional Specialty Conference, Nov. 1994.
U.S. EPA, "Development Document for Proposed Effluent
Limitations Guidelines and Standard for the Petroleum Refining
Point Source Category (Proposed)," John Lum - Project Officer,
Effluent Guidelines Division, U.S. EPA, Washington, DC,
20460, December, 1979.
US Crude Oil, Natural Gas and Natural Gas Liquid Reserves,
1993 Annual Report, Energy Information Administration, US
DOE, Washington DC, October 1994.
U.S.
Specific?
Yes
No
Yes
7
Yes
Yes
Yes
Yes
Petroleum
Industry
Based?
Chemical and
Refining
Yes
Yes
Yes
Yes
Yes, all
segments
Yes
Yes
EF and/or
AF Estimate
EF
EFfor
Germany
EF from AP-
42
Neither
EFsforHAPs
EF
Neither
Neither
Industry
Segment
Refinery
Production
Refinery
Refinery
Refinery
Combination
Emission Source
Refining
Fugitives
Not specific
Fugitives
Burners
Valves, flanges,
connections
Emission
Type
THC?
Methane
HC
None
HAPs
THC?
Year(s) Data
Gathered
1989
1985
1993
1992
1994
1979
(Continued)
-------
TABLE A-4a. SUMMARY OF REPORT DATABASE
(Continued)
Report Title
Webb, M. and P, Martina, "Fugitive Hydrocarbon Emissions
from Petroleum Production Operations," Presented at the 85th
Annual Meeting and Exhibition of the Air and Waste
Management Association, June 1992.
U.S. Environmental Protection Agency. Inventory of U.S.
Greenhouse Gas Emissions and Sinks: 1990-1994, EPA-230-
R-96-006 (NTIS PB 96-175997). U.S. Environmental
Protection Agency, Office of Policy, Planning and Evaluation,
Washington DC, November 1995.
U.S.
Specific?
Yes
Yes
Petroleum
Industry
Based?
EF and/or
AF Estimate
Yes ] EF
Oil and Gas
EF
Industry
Segment
Emission Source
Production j Fugitive
| Components
Combination
Model Facilities
Emission
Type
Methane, HC
Year(s) Data
Gathered
1980
CO, and
Methane
1990-1993
-------
TABLE A-4b, SUMMARY OF REPORT DATABASE
Report Title
"Atmospheric Emissions from Offshore Oil and Gas
Development and Production", EPA 450/3-77-026, June
1977.
"Atmospheric Hydrocarbon Emissions from Marine
Vessel Transfer Operations," Publication 2514A,
American Petroleum Institute, Washington, DC, Sept,
1981,
"Global Emissions of Carbon Dioxide from Petroleum
Sources," Prepared by Radian for API, July 1991.
"Worldwide Refining," Oil and Gas Journal, Dec. 21,
1992, p 84.
Radian Corporation. "Assessment of Atmospheric
Emissions from Petroleum Refining: Volume 2,
Appendix A" Prepared for U.S. Environmental Protection
Agency. Research Triangle Park, NC. July 1980.
Radian, Tier 2 Report for the GR1/EPA Methane
Emissions Project, June 1995.
Wetherold, R.G., G.E. Harris, F.D. Skinner, and L.P,
Provost (Radian Corporation). "A Model for Evaluation
of Refinery and Synfuels VOC Emission Data:
Volumes 1 and 2V Research Triangle Park, NC,
"Development of Fugitive Emission Factors and Emission
Profiles for Petroleum Marketing Terminals," Volume I,
API, Washington DC, May 1993.
"Global Emissions of Methane From Petroleum Sources,"
Prepared by Radian for API, February 1992.
Methodology
Measurements
& Guesses
Measurements
EPA & Industry
Documents
Industry
reports/survey
Measurements
&AP-42
Measurements,
Surveys
Literature
Search
Measured
EPA & Industry
Documents
Modifications
THC is 83.6%
OMby
volume, convert
toCH4
Need methane
composition.
AFs only
Might be able
to scale the
non-CH4.
Need methane
composition.
Need methane
composition.
Usefulness
Excellent
Excellent
Excellent
Excellent
Excellent
Excellent
Excellent
Good
Good
Accuracy?
Unknown
Confidence
Intervals and
STD
provided.
Compares
well w/ other
estimate
sources
Good, can be
calculated
for
companies
listed.
Good for
data given.
90%
Confidence
Intervals
Not
Available
95%
confidence
intervals for
most
Low
Comments
Great source for offshore emission factors.
Report provided emission factors for gasoline and crude oil for a
general case or based on specific information (type of vessel,
prior cargo, compartment treatment, volume). No statistically
significant correlation could be developed for gasoline loading.
End uses (beyond the scope of this study) have the largest
emissions. Emission factors given for individual refining
processes based on bbl throughput. Data for exploration and
production are very general.
Lists production/well and # wells for production and capacity of
various refining operations.
Calculates non-CH4 EF based on refinery throughput. Main
contributors are fugitives and heaters. Representative
components counts for fugitives are provided for each unit.
Assume CH4 emissions from gas industry equipment are
applicable to oil industry. Project is still in progress, so
numbers may change slightly.
Great source for emissions factors. Need to relate unit
operations presented here to unit capacities reported in Oil and
Gas Journal.
Primarily overall emission estimates for each industry segment.
Production excludes venting and flaring, and couldn't separate
oil from gas.
(Continued)
-------
TABLE A-4b. SUMMARY OF REPORT DATABASE
(Continued)
Report Title
"Offshore Oil and Gas Extraction - An Environmental
Review," Batlelle Columbus Labs, Prepared for Industrial
Environmental Research Lab, July 1977.
Petroleum Supply Annual 1993, Volumes 1 and 2, Energy
Information Administration, US DOE, Washington DC,
June 1994,
Picard, DJ., B.D. Ross, D.W.H. Koon. "A Detailed
Inventory of CH4 and VOC Emissions From Upstream
Oil And Gas Operations in Alberta." Canadian Petroleum
Association, Calgary, Alberta, 1992,
Radian Corporation. "Study of Refinery Fugitive
Emissions from Equipment Leaks," API Health and
Environmental Sciences Dept. and Western States
Petroleum Assoc., Volumes 1 and 2, April 1994.
Roscbrook, D.D., R.G. Wetherold, and L.P, Provost, "The
Development of Fugitive Emission Factors for the
Petroleum Refining Industry", Presented at the 72nd
Annual Meeting of the Air Pollution Control Association,
June 1979.
Tilkicioglu, B.H., "Annual Methane Emission Estimate of
the Natural Gas Systems in the United Slates, Phase 2,"
Pipeline Systems Inc., Sept, 1990.
"Assessment of VOC Emissions from Well Vents
Associated With Thermally Enhanced Oil Recovery", EPA
909/9-81-003, Sept. 1981.
"Compilation of Air Pollutant Emission Factors (AP^-2),"
US EPA.
"Emissions of Greenhouse Gases in the United States
1987-1994," Energy Information Administration,
November 1995.
Methodology
Estimates
Industry reports
Measurements
Measurements
from 3 sites
Samples and
study
Extrapolated
based on 3 sites
Measurements
from 58 sites
Various
Other studies
Modifications
72.4% CH4 in
natural gas
NA
Convert THC
losses to CH4.
CH4
composition
provided in raw
data
Need methane
composition.
Data provided
to calculate EF
Need methane
composition.
Usefulness Accuracy?
Good Poor
Good Good
Good 95%
| confidence
intervals are
provided.
i
Good | 95%
I Confidence
| Intervals
Good i 95%
| confidence
1 interval
Good j Poor, could
I provide site
| specific data
| for two sites.
Potential | Can be
| calculated
| from data
Potential | Data quality
| estimates
| provided.
Potential Precision of
CH4
estimates 30-
1 50%.
Comments
Relies on AP42 emission factors: O.IWbanel (fires), 3S#/barrcl
evap. Oil production is generally thought to have a low potential
for contribution to air pollution. Control of evaporation from
tanks is required.
Data reported on a monthly basis by all but 4 states.
Some methane specific EPs others reported as VOC Raw data
provided with methane and carbon dioxide spcciation.
Reviews EPAs 1993 protocol equations for refinery fugitive
emissions. The five facilities measured had O&M programs to
reduce the number of leaking components.
Great source for nonmethane VOC EFs
Data reported from two sites of interest: oii field w/ gas
utilization for sale, and gas/oil field. Project collected data from
an oil field w/ no gas production but data were not reported.
Cyclic THEOR wellhead casing vents.
Some of the data presented are taken from Radian's API study
on global emissions. It is likely that the actual CH4 emissions
are higher than shown.
(Continued)
-------
TABLE A-4b. SUMMARY OF REPORT DATABASE
(Continued)
Report Title
"Emissions of Producing Oil and Gas Wells", EPA 908/4-
77-006, November 1 977.
"Fugitive Methane Emissions from Oil and Gas
Production and Processing Facilities. Emission Factors
Based on the 1980 API-Rockwell Study," Prepared for
USEPA, Prepared by STAR Environmental, April 1992.
"Oilfield Emissions of Volatile Organic Compounds",
EPA-450/2-89-007, April 1989.
C.E. Burklin, and R,L. Honercamp, "Revision of
Evaporative Hydrocarbon Emission Factors," EPA-450/3-
76-039, U.S. Environmental Protection Agency, Research
Triangle Park, NC, August 1976.
C.E. Burklin, et al,. Oil and Gas Field Emission Survey,
USEPA, Contract No. 68-D 1-0031, April 1992.
DeLuchi, M.A., "Emissions from the Production, Storage,
an Transport of Crude Oil and Gasoline," Journal of Air
and Waste Management Association, Nov. 1993, pp.
1486-1495.
DuBose, D.A., J.L Steinmetz, and G.E. Harris,
"Frequency of Leak Occurrence and Emission Factors for
Natural Gas Liquid Plants," Prepared for EPA, July 1982.
Houghton, J.T., G.J. Jenkins, and J.J. Ephraums. 1990.
"Climate Change: The IPCC Scientific Assessment.
Report prepared for IPCC by Working Group I,"
Intergovernmental Panel on Climate Change, Press
Syndicate, University of Cambridge.
Methodology
Sampled
Sampling
Measurements
Measured and
Guessed
Paper study
Measurements
Not clear
Modifications
Need methane
composition.
Can be
calculated
based on non-
CH4HC
Need THC or
methane
composition.
Need methane
composition.
Need methane
andCO2
composition.
Need methane
composition
related to VOC
Need methane
composition.
Convert Global
ER to US.
Usefulness
Potential
Potential
Potential
Potential
Potential
Potential
Potential
Potential
Accuracy?
Diurnal
variation
mean 95
Could be
calculated
based on raw
data.
Poor
7
Unknown
95%
confidence
interval.
Unknown
Comments
Good source for well emissions, but old data.
Important parameters affecting EF's: % of components that leak
and distribution of leaks into various size categories. Old
emission factors may over predict current operations based on
recent industry changes. Reports says that fugitive emissions
factors were developed for production, but none are reported.
For the gas-drive well, gas was vented to the atmosphere. The
well produces a maximum of 10,000 cfd.
Speciation of non-methane gas components is provided for
sources tested. However, compositions are scaled to 100%
without CH4.
Updates AP-42 emission equations.
Referenced other resources.
Estimate VOC emissions for 2000, EFs in terms of grams
VOC/gallon of fuel consumed.
Report summarizes measured fugitive emissions from gas plants
at crude oil petroleum and natural gas onshore production
facilities. Perhaps components common to both oil and gas can
be used for oil study-
Reported production emission range of 25-50 Tg/yr for gas
drilling, venting, and transmission.
(Continued)
-------
TABLE A-4b. SUMMARY OF REPORT DATABASE
(Continued)
1 Report Title
MG. Klett and J.B. Galeski, "Flare Systems Study," EPA-
600/2-76-079, U.S. Environmental Protection Agency,
Research Triangle Park, NC, March 1976.
N.F, Suprenant, et al., "Emissions Assessment of
Conventional Stationary Combustion Systems, Volume V:
Industrial Combustion Sources," EPA Contract No. 68-02-
2197, GCA Corporation, Bedford, MA, October 1980.
"An Assessment of the Contribution of Gas to the Global
Emissions of Carbon Dioxide Final Report (February-
December 1983)", GRI, June 1984.
"Anthropogenic Methane Emissions in the United States:
Estimates for 1990," Report to Congress, EPA430-R-93-
003, April 1993.
"Evaporation Loss from Fixed -Roof Tanks", API Bulletin
2518, June 1962, Reaffirmed August, 1987.
"Hydrocarbon Emissions from Refineries," API
Publication No. 928, American Petroleum Institute,
Washington, DC, July 1973.
"Independent Quality Assurance of Refinery Fugitives
Testing by Western States Petroleum Association," EPA
Office of Air Quality Planning and Standards, September
1993.
"Methane Emissions from the Oil and Gas Production
Industries," Ruhrgas A.G., July 1989.
"Options for Reducing Methane Emissions Internationally
Volume 1: Technological Options for Reducing Methane
Emissions," Report to Congress, EPA 430-R-93-006, July
1993.
Methodology
Estimates, min.
field testing
Not reported.
Reported
emissions as kg
C
Other sources
Survey of
available data
Measurements
& estimates
Measurement
Measurements
from 18 plants
Other sources,
some measured
Modifications
Need methane
composition
Need methane
composition.
Can carbon
emissions be
related to CH4?
Need methane
composition.
Need to scale
from worldwide
data
Usefulness
Potential
Potential
Poor
Poor
Poor
Poor
Poor
Poor
Poor
Accuracy?
Poor, AP-42
rankC
Not reported.
Poor
Poor, mostly
estimates
+/- 10% for
calculations
within the
range of
data.
Could be
calculated
9
No estimates
Comments
Total flare emissions annually <1% average yearly plant
emissions. Very little experimental data on flare emissions exist
due to sampling difficulty. Compare with GRI/EPA flare
summary table.
Can boiler emissions be related to other combustion sources
(e.g, heater/treater)?
Report assesses the contribution from the future consumption of
fuel gas to global emissions of CO2. Emissions are given for
carbon based on energy usage.
Total methane emissions from petroleum production and
refining were estimated to range from 0,1 to 0.6 Tg/yr in the
U.S. The majority of this is associated with venting during oil
production.
This bulletin is the result of a study of available test data on
evaporation losses from cone-roof tanks. Test data did nol
include crude containing significant amounts of methane or
ethane; therefore, the equations may not be applicable to
production lease tanks.
Report estimates the major sources of HC emissions from
refineries. Costs of methods and facilities for reducing HC
losses were developed . Extremely difficult to follow emission
estimates.
This report is an audit of the data presented in "Study of
Refinery Emissions from Equipment Leaks." Report mainly
discussed the errors associated with Ihe data. No new data are
presented.
Oil industry ~ 6x gas industry methane emissions. 14 Bcf CH4
enters the atmosphere each year.
Some EFs are given for illustration purposes. Deals primarily
with options to reduce CH4 emissions.
(Continued)
-------
TABLE A-4b. SUMMARY OF REPORT DATABASE
(Continued)
Report Title
Basic Petroleum Data Book, Petroleum Industry Statistics
Vol. VD, No 3, API, Sept, 1987.
C£. Burklin, etal., "Revision of Emission Factors for
Petroleum Refining," EPA-450/3-77-030, U.S.
Environmental Protection Agency, Research Triangle
Park, NC, October 1977.
Estimating Air Toxics Emissions From Coal and Oil
Combustion Sources, EPA-450/2-89-001, U.S.
Environmental Protection Agency, Research Triangle
Park, NC, April 1989.
Kantor, R.H., "Trace Pollutants from Petroleum and
Natural Gas Processing," Prepared for EPA by M,W.
Kellogg Co., June 1974.
Leralin, J.S., I. Graham-Bryce, "The Petroleum Industry's
Response to Climate Change: The Role of the IPIECA
Global Climate Change Working Group," UNEP Industry
and Environment:, Jan-Mar. 1994, pp 27-30.
Liplon, Sydney, "Fugitive Emissions," Chemical
Engineering Progress, Vol. 85, No. 6, pp. 42-47, June
1989.
Mussig, S., et al., "Possibilities for Reduction of
Emissions - in Particular the Greenhouse Gases CO2 and
CH4 - in the Oil and Gas Industry," Presented at the
European Petroleum Conference, Nov. 1992.
R.F. Eoland, et al.. Screening Study for MisceDaneous
Sources of Hydrocarbon Emissions in Petroleum
Refineries, USEPA, Dec. 1976.
Methodology
No emissions
data.
Literature
Search
Literature
search,
interviews
7
AP-42
Data from
another report.
AP-42
Modifications
Reported that
overall
emissions were
0.3 wt% CH4
Need methane
composition.
Assume
Germany
emissions are
same for US
Need methane
composition.
Usefulness
Poor
Poor
Poor
Poor
Poor
Poor
Poor
Poor
Accuracy?
Poor due to
age of data
and
unknown
stream
comp.
Low
Poor
i
Unknown
AP-42 basis
Comments
Same AFs, but from 1987. Economic data primarily.
Data are old. Waste stream emissions reported as HC, no
spcciation.
Not related to scope of this project. Boiler data only.
Identifies continuous vs. intermittent emitters. Some activity
factors provided.
Discusses IPIECA's role and approach in understanding the
global climate change issue. Does not report EFs or AFs.
AP^t2 fugitive emissions.
Article discusses methods to reduce CO2 and CH4 emissions.
AP-42 factors were refined based on the extent of BACT control
estimated from NSPS regs. Controlled and uncontrolled EFs
were weighted to arrive at an average EF, which could be
ratioed by throughput- Good descriptions of various refining
operations.
to
to
(Continued)
-------
TABLE A-4b. SUMMARY OF REPORT DATABASE
(Continued)
Report Title
Methodology I Modifications
Usefulness
Accuracy?
Comments
Rosenberg, E.S., "Impact of the "HON" Rule on the
Petrochemicals and Refining (Industries)," Presented at
the 1993 NPRA Annual Meeting, March, 1993.
Taback, H.L, G. Lauer, L.K. Gilmer, and K. Ritter,
"Strategies for Improving HAP Emission Factors and
Profiles for the Petroleum Industry," Presented at the 85th
Annual Meeting and Exhibition of the Air and Waste
Management Association, June 1992.
Source testing
Taback, H,J,, and K. Ritter, "1994 Fugitive Emissions
Estimating Data for Petroleum Industry Equipment
Leaks," Presented at the Air and Waste Management
Association and California Air Resources Board 5th
Annual West Coast Regional Specialty Conference, Nov.
1994.
Poor
I No CH4 EFs or AFs. Overview of the HON, its interaction
j w/the rules for modification of the sources of toxic air
j pollutants, controls of VOCs, and EPA's operating permit rule.
Measurements
Need methane
composition.
Poor
Need methane
composition.
Poor
Not given w/
data.
No CH4 emissions. Lists research projects underway in the
areas of component leaks, process vents, transfer operations,
wastewater and others.
Symposium paper.
to
UJ
U.S. EPA, "Development Document for Proposed Effluent
Limitations Guidelines and Standard for the Petroleum
Refining Point Source Category (Proposed)," John Lum -
Project Officer, Effluent Guidelines Division, U.S. EPA,
Washington, DC, 20460, December, 1979.
Poor
US Crude Oil, Natural Gas and Natural Gas Liquid
Reserves, 1993 Annual Report, Energy Information
Administration, US DOE, Washington DC, October 1994.
EIA surveys
Summarizes EPA's review of petroleum industry with respect to
discharge of toxics in US waters.
Poor
Good
Presents data only on reserves.
Webb, M, and P. Martino, "Fugitive Hydrocarbon
Emissions from Petroleum Production Operations,"
Presented at the 85th Annual Meeting and Exhibition of
the Air and Waste Management Association, June 1992,
Published and
field study
Poor
Not available
Analysis of field data for fugitive equipment leaks. Indicates
that existing EFs over predict HC emissions from petroleum
production operations with directed maintenance programs. No
data given.
U.S. Environmental Protection Agency. Inventory of U.S.
Greenhause Gas Emissions and Sinks: 1990-1994. EPA-
23D-R-96-006 (NT1S PB 96-175997). U.S.
Environmental Protection Agency, Office of Policy,
Planning and Evaluation, Washington DC, November
1995,
Emission
Inventory &
Models
Potential
Poor due to
lack of
supporting
data for oil
Combines oil and gas production. Report states high level of
uncertainty in the data.
-------
TABLE A-5a. PRODUCTION EMISSION DATABASE
Report Title
"Anthropogenic Methane Emissions in the United States:
Estimates for 1990," Report to Congress, EPA 430-R-93-003,
April 1993.
"Assessment of VOC Emissions from Well Vents Associated
With Thermally Enhanced Oil Recovery", EPA 909/9-81-003,
Sept, 1981.
"Atmospheric Emissions from Offshore Oil and Gas
Development and Production", EPA 450/3-77-026, June
1977.
"Global Emissions of Methane From Petroleum Sources",
Prepared by Radian for API, February 1992.
"Methane Emissions from the Oil and Gas Production
Industries," Ruhrgas A.G., July 1989.
"Offshore Oil and Gas Extraction - An Environmental
Review," Ralte'le Columbus Labs, Prepared for Industrial
Environmental Research Lab, July 1977,
DuEose, D.A., J.I. Steinmetz, and G.E. Harris, "Frequency of
Leak Occurrence and Emission Factors for Natural Gas
Liquid Plants," Prepared for EPA, July 1982.
Petroleum Supply Annual 1993, Volumes 1 and 2, Energy
Information Administration, US DOE, Washington DC, June
1994.
Picard, D.J., B.D. Ross, D.W.H. Koon. "A Detailed
Inventory of CH4 and VOC Emissions From Upstream Oil
And Gas Operations in Alberta." Canadian Petroleum
Association, Calgary, Alberta, 1992.
Radian, Tier 2 Report for the GRUEP A Methane Emissions
Project, June 1995.
Tilkicioglu, B.H., "Annual Methane Emission Estimate of the
Natural Gas Systems in the United States, Phase 2" Pipeline
Systems Inc., Sept, 1990.
Overall Site Emission
Estimate
Methane emission estimates
kg/weuVyr
Well casing emissions 64.9 Ib
CH4/d (+/- 29%)
HC emissions
U.S. 12,074 ton CH4/y
excluding Venting & Flaring
10.5 bcm CH4/yr worldwide,
1.7 bcm/y N. America
Reported THC and non-
methane/non-ethane as
kg/day/source
Production for 1988-1993 in
comments
90% Confidence Interval
calculated
Emissions in scfd/device
Data from 2 Facilities: #1 Oil,
#2 Oil/gas
Tanks
367MgHC/10*6bbl
Some estimates
provided
7.94 mA3 gas/hr (+/-
88%)
Flares - Well
Completion
10kg/10*6cfgas
flared
98 Bscf gas flared in
US
Lasts 5-10 days, 95%
test gas flared
8lbCH4/MMcf(AP-
42) (per well?)
1993 # completions
7994(6%<1992)
Well Blowout
Wells in general: 72
kg/well/y
20 M g /well/day
Some estimates
provided
Combustion 0.1,
Evaporation 38
(Ib HC/bbl)
Well Workovers
#1 cf
CH4/hr
-------
TABLE A-5b, PRODUCTION EMISSION DATABASE
Report Title
"Armospherie Emissions from Offshore Oil and Gas Development and
Production", EPA 450/3-77-026, June 1977.
C.E, Burklin, et aL, Oil and Gas Field Emission Survey, USEPA, Contract No.
68-DI-0031, April 1992.
Picard, D .}., B.D. Ross, D.W.H. Koon. "A Detailed Inventory of CH4 and
VOC Emissions From Upstream Oil And Gas Operations in Alberta."
Canadian Petroleum Association, Calgaiy, Atberta, 1 992.
Radian, Tier 2 Report for the GRI/EPA Methane Emissions Project, June
1995.
Tilkicioglu, B.H., "Annual Methane Emission Estimate of the Natural Gas
Systems in the United States, Phase 2," Pipeline Systems Inc., Sept, 1990.
Heaters
0.05 M g HC/lO^fi bbl
31bCH4/10'16flA3
Pneumatic Devices
0. 1996 mA3 gas/hr
(+/- 52%)
493 (+/- 55%)
#1 946, #2 25,871
#CH4/y
Chemical
Injection Pumps
Fugitives 0.004
Ib/day-well
0.39446 mA3
gas/hr O/-30%)
439 (W- 91%)
Compressor
Starts
14.3(+/-74%)
Compressor
Blowdowns
12 (+/- 52%)
to
-------
TABLE A-5c. PRODUCTION EMISSION DATABASE
Report Title
"Emissions of Greenhouse Gases in the United Slates 1987-1994," Energy Information
Administration, November 1995.
'Emissions of Producing Oil and Gas Wells," EPA 908/4-77-006. November 1977.
"Options for Reducing Methane Emissions Internationally Volume 1: Technological
Options for Reducing Methane Emissions," Report to Congress, EPA 430-R -93-006,
July 1993.
C.E. Burklin, and R.L. Honercamp, "Revision of Evaporative Hydrocarbon Emission
Factors," EPA-4-50/3-76-039, U.S. Environmental Protection Agency. Research
Triangle Park, NC, August 1976.
CE. Burklin, et ah, Oil and Gas Field Emission Survey, USEPA. Contract No. 68-D1-
0031, April 1992.
DuBose, D.A., J.I. Steinmetz, and G.E. Harris, "Frequency of Leak Occurrence and
Emission Factors for Natural Gas Uquid Plants," Prepared for EPA, July 1982.
Picard, D.J., B.D. Ross, D.W.H. Koon. "A Detailed Inventory of CH4 and VOC
Emissions From Upstream Oil And Gas Operations in Alberta." Canadian Petroleum
Association, Calgary, Alberta, 1992.
Overall Site
1992, 5.97 trillion fl«3 natural
gas withdrawn from oil wells
16.5 Ib THC/day (Rod pump
well), 0.008 Ib/d (dec. subm.)
Reported as kg CH4/day,
number of components, plant
emissions
346,000 ton HC/yr (oil), 544,
000 ton HC/yr (gas)
Reported as 1) THC 2)non-
methane/non-ethane
Other Fugitives
Vented = 640,000
metric tons CH4/yr
Connections & Flanges
0.021,3000,62
Some estimates provided
0.026
0.011
Gas/Vapor 0.0079,
Light Uquid 0.0001 9
Open Ended Lines
0.53
0.34
-------
TABLE A-5d. PRODUCTION EMISSION DATABASE
Report Title
"Anthropogenic Methane Emissions in the United
Slates: Estimates for 1990," Report to Congress, EPA
430-R-93-003, April 1993.
"Global Emissions of Methane Erom Petroleum
Sources", Prepared by Radian for API, February 1992.
"Methane Emissions from the Oil and Gas Production
Industries," Ruhrgas A.G., July 1989.
C,E, Burklin, et al.. Oil and Gas Field Emission Survey,
USEPA, Contract No. 68-D1-0031, April 1992.
Picard, D.J., B.D. Ross, D.W-H. Koon. "A Detailed
Inventory of CH4 and VOC Emissions From Upstream
Oil And Gas Operations in Alberta." Canadian
Petroleum Association, Calgary, Alberta, 1992.
Radian, Tier 2 Report for the GRI/EPA Methane
Emissions Project, June 1995,
Fugitives -
Overall
76.7
kg/well/yr
Wells -0.1 735
t/y per well
WeUs 5 ton/yr
per oil well
19,151 kt
THC, 10.533
let VOC
Fugitives -
Separators
122 (+/- 33%)
Fugitives -
Heaters
57.7 (+/- 40%)
Fugitives -
Compressors
0.07 Ib ROG/day
per well
Small 267.8, Large
16360
Fugitives -
Meter/Sales
52.9
Fugitives -
Pipeline
57.8(+/-97%)
Fugitives -
Platforms
Gulf 2914, Other US
1178
>
to
-------
TABLE A-5e. PRODUCTION EMISSION DATABASE
Report Title
"Anthropogenic Methane Emissions in the United States: Estimates for 1990,"
Report to Congress, EPA 430-R-93-003, April 1993.
"Atmospheric Emissions from Offshore Oil and Gas Development and
Production", EPA 450/3-77-026, June 1977.
"Compilation of Air Pollutant Emission Factors (AP-42)," US EPA.
"Emissions of Greenhouse Cases in the United States 1987-1994," Energy
Information Administration, November 1995.
"Global Emissions of Methane From Petroleum Sources", Prepared by Radian for
API, February 1992.
"Offshore Oil and Gas Extraction - An Environmental Review," Bartelle
Columbus Labs, Prepared for Industrial Environmental Research Lab, July 1977.
C.E, Burklin, et al.. Oil and Gas Field Emission Survey, USEPA, Contract No.
68-D 1-0031, April 1992.
Radian, Tier 2 Report for the GRMEPA Methane Emissions Project, June 1995.
Tilkicioglu, B.H., "Annual Methane Emission Estimate of the Natural Gas
Systems in the United States, Phase 2," Pipeline Systems Inc., Sept, 1990.
Vessel Blowdown
0.375 (+/- 67%)
Other Engines
Turbine 0.14 g/hp-hr, Gas
Recip 4.86, Oil Recap 0.43
Turbine(E/kWhr):0. 1 1 7(gas),
0.083(oil)TOC as CH4(D)
le^t ton CH4/yr/well
Diesel Eng.: 0.16 Ib HC/hr
or 37.5 Ib HC/1000 gal
Diesel 0.07, Dual fuel 4.7 Ib
CH4/1000 hp-hr
PRV
0.337 ±
112%
ESD/EBD
704 ± 200%
Other
Maintenance 0.15
kg/well/yr
Pump seals 0. 1 M
g/10A6bbl
Oil wells - 0.072 metric
tons CH4/well
Well heads 0.01 Ib
ROG/well day
Fugitive Comp. Station
8247
Upsets: #1 - 34776 #
CH4/y
to
oo
-------
TABLE A-5f. PRODUCTION EMISSION DATABASE
Report Title
"Emissions of Greenhouse Gases in the United Scales 1987-1994," Energy
Information Administration, November 1995.
"Emissions of Producing Oil and Gas Wells," EPA 908/4-77-006,
November 1977.
"Options for Reducing Methane Emissions Internationally Volume 1:
Technological Options for Reducing Methane Emissions," Report to
Congress, EPA 430-R-93-006, July 1993.
C.E. Burklin, and R.L. Honercamp, "Revision of Evaporative
Hydrocarbon Emission Factors," EPA-450/3-76-039, U.S. Environmental
Protection Agency, Research Triangle Park, NC, August 1976.
C.E. Burklin, et aL, Oil and Gas Field Emission Survey, USEPA, Contract
No. 68-D1-003I, April 1992.
DuBose, D.A., J.I. Steinmetz, and G.E. Harris, "Frequency of Leak
Occurrence and Emission Factors for Natural Gas Liquid Plants," Prepared
for EPA, July 1982.
Picard, DJ., B.D. Ross, D.W.H. Koon, "A Detailed Inventory of CH4 and
VOC Emissions From Upstream Oil And Gas Operations in Alberta."
Canadian Petroleum Association, Calgary, Alberta, 1 992.
Pump Seals
5.12, 12,30
(seals in general)
S Ib HC/day per
seal
1.5
1,2
0.02139
kg/hr/source
THC
Compressor Seals
41bHO1000bbl
crude
4.6
1.0
0,80488 kg/hr/source
THC
Valves
0.384, 750, 288
Some estimates
provided
0.48
0.18
G/V 0.01417, LL
0.00121
Pressure Relief
Valves
3.6, 12,43
2.4 Ib HC/day-valve
4.5
0.33
0.12096 kg/hr/source,
0.0019mA3/hr
,
Other
Wells = 0.04e6 metric
tons CH4/yr (1987-1992)
THC is 47.6% methane
from 5 measurements
Compressor Exhaust:
recip 500, turbine 6-12 kg
CH4/MMcffuel
Separators 8 lb/1000 bbl,
Pumps 75 Ib/lOOOhbl
95% Conf. Int. provided.
>
I
to
-------
TABLE A-6a. TRANSPORTATION EMISSION DATABASE
Report Title
"Anthropogenic Methane Emissions in the United States: Estimates
for 1990," Report to Congress, EPA 430-R-93-003, April 1993,
"Atmospheric Hydrocarbon Emissions from Marine Vessel Transfer
Operations," Publication 2514A, American Petroleum Institute,
Washington, DC, Sept. 1981.
"Compilation of Air Pollutant Emission. Factors (AP-42)," US EPA.
"Emissions of Greenhouse Gases in the United States 1987-1994,"
Energy Information Administration, November 1995.
"Global Emissions of Methane From Petroleum Sources", Prepared
by Radian for API, February 1992.
"Hydrocarbon Emissions from Refineries," API Publication No.
928, American Petroleum Institute, Washington, DC, July 1973.
C,E. Burklin, and R.L. Honercamp, "Revision of Evaporative
Hydrocarbon Emission Factors," EPA^50/3-76-039, U.S.
Environmental Protection Agency, Research Triangle Park, NC,
August 1976.
C.E. Burklin, et al., Oil and Gas Field Emission Survey, U.S. EPA,
Contract No. 68-D1-0031, April 1992.
DeLuchi, M.A., "Emissions from the Production, Storage, an
Transport of Crude Oil and Gasoline," Journal of Air and Waste
Management Association, Nov. 1993, pp. 1486-1495.
Wetherold, R.G., G.E. Harris, F.D. Skinner, and L.P. Provost
(Radian Corporation). "A Model for Evaluation of Refinery and
Synfuels VOC Emission Data: Volumes 1 and 2." Research
Triangle Park, NC.
Truck/Car Loading
Calculation methods and
average values given.
12,400 rpy, 7.9e-6 ton
CH4/bbI
Splash 700, Sub. 225 1 HC/y
0.4-7 lb/1000 gal transferred
2.8, 4.7 Ib VOC/IOOO gal
Values provided based on
loading style, and petroleum
product.
Barge Loading
Average Factors given.
typical=3.4, unclean=3.9
3500 tpy, 2.55e-6 ton
CH4/bbl
0,007 % load vol,/psia true
vp
1.2-4 lb/1000 gal transferred
0.024 (oil), 0.23 (gasoline)
Values provided based on
loading style, and petroleum
product.
Marine Tanker Loading
llbHC/1000 gal crude.
HC contains 20% CH4
General EF 1.0 Ib HC/1000
gal
1.8 typical
2.55e-4 short tons CH4/bbI
Oii:0.027 (AK), 0.004 lower
48
Values provided based on
loading style, and petroleum
product.
Rail Car Loading
700 tpy, 7.9e-6 ton CH4/bbl
6.6, 4.7 Ib VOC/IOOO gal
-------
TABLE A-6b. TRANSPORTATION EMISSION DATABASE
Report Title
"Atmospheric Emissions from Offshore Oil and Gas Development and
Production", EPA 450/3-77-026, June 1977.
"Compilation of Air Pollutant Emission Factors (AP-42)," US EPA.
"Emissions of Greenhouse Gases in the United States 1987-1994," Energy
Information Administration, November 1995.
"Global Emissions of Methane From Petroleum Sources", Prepared by Radian for
API, February 1992.
C.E. Burklin, and R,L Honercamp, "Revision of Evaporative Hydrocarbon
Emission Factors," EPA-450/3-76-039, U.S. Environmental Protection Agency,
Research Triangle Park, NC, August 1976.
CJE. Burklin, et al., "Revision of Emission Factors for Petroleum Refining," EPA-
450/3-77-030, U.S. Environmental Protection Agency, Research Triangle Park,
NC, October 1977.
C.E. Burklin, et. al_, Oil and Gas Reid Emission Survey, U.S. EPA, Contract No.
68-D 1-003 1, April 1992.
DeLuchi, M. A., "Emissions from the Production, Storage, an Transport of Crude
Oil and Gasoline," Journal of Air and Waste Management Association, Nov. 1993,
pp. 1486-1495.
Petroleum Supply Annual 1993, Volumes 1 and 2, Energy Information
Administration, US DOE, Washington DC, June 1994.
Wetherold, R.G., G.E. Harris, F.D. Skinner, and L,P. Provost (Radian
Corporation). "A Model for Evaluation of Refinery and Synfuels VOC Emission
Data: Volumes 1 and 2." Research Triangle Park, NC.
Overall Site
AFs for transport from offshore
platforms
VOC EF as lb/1000 gal transferred
1992 - 83,000 metric ton CH4
from marine vessels
US 16,703 ton CH4/yr
538,000 ton HC/yr
VOC emissions: l)submerged
loading 2} splash loading
Emissions reported as g VOC/gal
fuel consumed by motorists
Total US Imports in 1000 BPD:
For 1993 = 6787;
For 1992 = 6083
EFs Ib VOC/ 1 000 gal
Fugitives
Pipeline fugitives, negligible, most
crude transported by pipe
0. 15 Ib HC/day per valve
Field storage EF 0.056, bulk plant
0.27-0.55
Other
14 Barge systems, 66 pipeline
commingling systems
Heaters: 42 Ib HC/1000 bbl, 0.003
lb/1000 ftA3
Gasoline tanker loading 0,047,
treating crude 0.022
>
-------
TABLE A-7a. REFINERY EMISSION DATABASE
Report Title
"Anthropogenic Methane Emissions in the United
Stales: Estimates for 1990," Report to Congress,
EPA 430-R-93-003, April 1993.
"Compilation of Air Pollutant Emission Factors
(AP-42)," US EPA,
"Development of Fugitive Emission Factors and
Emission Profiles for Petroleum Marketing
Terminals," Volume I and n, API, Washington DC,
May 1993.
"Emissions of Greenhouse Gases in the United
Stales 1987-1994," Energy Information
Administration, November 1995.
"Evaporation Loss from Fixed-Roof Tanks", API
Bulletin 2518, June 1962, Reaffirmed August,
1987.
"Global Emissions of Methane From Petroleum
Sources", Prepared by Radian for API, February
1992.
"Hydrocarbon Emissions from Refineries," API
Publication No. 928, American Petroleum Institute,
Washington, DC, July 1973.
"Methane Emissions from the Oil and Gas
Production Industries," Ruhrgas A.G., July 1989.
DeLuchi, M.A., "Emissions from the Production,
Storage, an Transport of Crude Oil and Gasoline,"
Journal of Air and Waste Management Association,
Nov. 1993. pp. 1486-1495.
Overall Site
AFs given for oil refinery
with 330000 bbVd capacity.
EF reported as Ib THC/hr
Emissions reported as short
tons CH4/bbl crude
Losses calculated as bbVyr.
U.S. 96,508 tons CH4/y
0.4 bcm CH47yr from
refineries worldwide
Emissions reported as g
VOC/gal fuel consumed by
motorists
Fixed Roof
Tanks
Equations given
for breathing and
working losses.
All tanks -2,100
ton CH4/yr
(based on
throughput)
Gasoline 6700,
Crude 5200 ton
HC/yr
0.4% of refinery
throughput will
evaporate
0.035
Floating Roof
Tanks
Oil storage tank
emissions given.
Tank farms: 4.37e-7
ton CH4/bbl
throughput
4.37e-7 ton CH4/bbi
0.05% of refinery
throughput will
evaporate
0.133
Other Fugitives
Non-methane
emission factors
given for fugitive
components
Gas -0.001 4,
Light Liquid -
0.00025
1 .635e-6 tan
CH4/bbl capacity
92,100 ton CH4/yr
equip, leaks
Flange
Connectors
600 #VOC/d,
46500 flanges
Gas - 0.000067,
Light Liquid -
0.000023
Non-flanged
Connectors
1
Flanged/not
specified
UJ
to
(Continued)
-------
TABLE A-7a. REFINERY EMISSION DATABASE
(Continued)
Report Title
Lipton, Sydney, "Fugitive Emissions," Chemical
Engineering Progress, Vol. 85, No. 6, pp. 42-47,
June 1989.
Petroleum Supply Annual 1993, Volumes 1 and 2,
Energy Information Administration, US DOE,
Washington DC, June 1994,
Radian Corporation. " Assessment of Atmospheric
Emissions from Petroleum Refining: Volume 1,"
Prepared for U.S. Environmental Protection
Agency. Research Triangle Park, NC, Contract
No, 68-02-2147. July 1980.
Radian Corporation, "Study of Refinery Fugitive
Emissions from Equipment Leaks," API Health and
Environmental Sciences Dept. and Western States
Petroleum Assoe.. Volumes 1 and 2, April 1 994.
Rosebrook, D.D., R.G. Wetherold, and L.P.
Provost, "The Development of Fugitive Emission
Factors for the Petroleum Refining Industry",
Presented at the 72od Annual Meeting of the Air
Pollution Control Association, June 1979.
Wetherold, R.G., G.E. Harris, F.D. Skinner, and
LP. Provost (Radian Corporation). "A Model for
Evaluation of Refinery and Synfuels VOC Emission
Data: Volumes 1 and 2." Research Triangle Park,
NC. EPA Contract No.
Overall Site
Emissions reported as #
THC/hr/source
Provides # of refineries and
volume of crude received.
Non CH4 HC EFs Ib/hr-
source
Emissions reported as
Ib/hr/source
EFs Ib VOC/day/source
Fixed Roof
Tanks
Floating Roof
Tanks
Other Fugitives
Sampling 0.033
THC/hr/
source
Flange
Connectors
0,00056 # THC/hr/
source
0.00056
4.9e-7
Gas/Vapor, Light
Liquid - 0.0005,
Heavy Liquid -
0.0007,
0.013
Non-flanged
Connectors
1.7e-6
All -0.00058
(general flanges)
Flanged/not
specified
'o
OJ
-------
TABLE A-7b. REFINERY EMISSION DATABASE
Report Title
"Compilation of Air Pollutant Emission Factors (AP-42)," US EPA.
"Global Emissions of Methane From Petroleum Sources", Prepared by Radian for API,
February 1992.
"Hydrocarbon Emissions from Refineries," API Publication No. 928, American
Petroleum Institute, Washington, DC, July 1973.
"Worldwide Refining," Oil and Gas Journal, Dec. 21, 1992, p 84.
C,E. Burklin, and R.L. Honercamp, "Revision of Evaporative Hydrocarbon Emission
Factors," EPA-450/3-76-039, U.S. Environmental Protection Agency, Research
Triangle Park, NC, August 1977.
Petroleum Supply Annual 1993, Volumes I and 2, Energy Information Administration,
US DOE, Washington DC, June 1994.
R.F. Boland, et al., Screening Study for Miscellaneous Sources of Hydrocarbon
Emissions in Petroleum Refineries, USEPA, Dec. 1976.
Radian Corporation. " Assessment of Atmospheric Emissions from Petroleum
Refining; Volume 1." Prepared for U.S. Environmental Protection Agency. Research
Triangle Park, NC. Contract No. 68-02-2147. July 1980,
Wetherold, R.G., G.E. Harris, F,D. Skinner, and L.P. Provost (Radian Corporation).
"A Model for Evaluation of Refinery and Synfucls VOC Emission Data: Volumes 1
and 2." Research Triangle Park, NC.
Overall Site
THC emissions as lb/1000 bbl feed
Separation processes EF = 1 .635e-5 ton
CH4/bbl
Model refinery based on 100,000 bbVd
U.S. Crude Capacity 15,209,853 b/cd;
Throughputs by process also given.
21 01000 ion HC/yr
Operable capacity of process units,
1000 bbl/d
Component counts for process units
given.
EFs Ib VOC/1000 bbl fresh feed
Catalytic Processes
Fluid - 220, moving bed - 87
Catalytic regeneration: 220 Ib HC/1000
bbl feed to FCC;
87 Ib HC/1000 bbl feed to TCC
220 Ib HC/1000 bbl feed
9259 (includes hydrocrack & thermal
crack units)
No Emission Control - 220
Fluid Coking
ND
Negligible
No Emission Control
-135
-------
TABLE A-7c. REFINERY EMISSION DATABASE
Report Title
"Anthropogenic Methane Emissions in the
United States: Estimates for 1990," Report to
Congress, EPA 430-R-93-003, April 1993.
"Compilation of Air Pollutant Emission Factors
(AP-42)," US EPA.
"Development of Fugitive Emission Factors and
Emission Profiles for Petroleum Marketing
Terminals," Volume 1, API. Washington DC,
May 1993.
"Emissions of Greenhouse Gases in the United
States 1987-1994," Energy Information
Administration, November 1995,
"Global Emissions of Methane From Petroleum
Sources", Prepared by Radian for API, February
1992.
"Hydrocarbon Emissions from Refineries," API
Publication No. 928, American Petroleum
Institute, Washington, DC, July 1973.
"Methane Emissions from the Oil and Gas
Production Industries," Ruhrgas A.G., July
1989.
C.E. Burkliri, et al., "Revision of Emission
Factors for Petroleum Refining," EPA-450/3-77-
030, U.S. Environmental Protection Agency,
Research Triangle Park, NC, October 1977.
Open Ended
Lines
Gas - 0.0067,
Light Liquid -
0.0065
Pump Seals
1300#
VOC/d;
350 seals
Light Liquid
- 0.00093
200 ton
HC/yr
Compressor
Seals
HOOffVOC/d, 70
seals
Valves
6800
#VOC/day,
1 1500 valves
Gas -
0.00016,
Light Liquid
-0.00015
Pressure Relief
Valves (PVR)
500#VOC/d,100
PRVs
Gas -0.0014,
Light Liquid -
0.00025
75-350 ton HC/yr
Flares
4e-7 ton
CH4/bbl
capacity
2,300 1
CH4/y (4e-7
ton/bbl)
0.15-0.5%
feedstock to
flare
0.8 Ib
HC/lOOObbl
Other
Waste gas stream 10.4 kg
CH4/yr from 1 0 refineries
650 drains, 1000#VOC/d;
Cooling tower 1600; Separator
32100
Loading arm valves:
Gas - 0.045,
Light Liquid - 0.00087
Reciprocating Compressor:
1.41bHC/1000ftA3,
Turbine 0.02 Ib HC/1000 ft*3
(Continued)
-------
TABLE A-7c. REFINERY EMISSION DATABASE
(Continued)
Report Title
DeLuchi, M.A., "Emissions from the
Production, Storage, an Transport of Crude Oil
and Gasoline," Journal of Air and Waste
Management Association, Nov. 1993, pp. 1486-
1495,
Lipton, Sydney, "Fugitive Emissions," Chemical
Engineering Progress, Vol. 85, No. 6, pp. 42-47,
June 1989.
M.G. KJett and J.B. Galeski, "Flare Systems
Study," EPA-600/2-76-079, U.S. Environmental
Protection Agency, Research Triangle Park,
NC, March 1976.
R.F. B aland, et aL, Screening Study for
Miscellaneous Sources of Hydrocarbon
Emissions in Petroleum Refineries, USEPA,
Dec. 1976.
Radian Corporation. " Assessment of
Atmospheric Emissions from Petroleum
Refining: Volume 1." Prepared for U.S.
Environmental Protection Agency. Research
Triangle Park, NC. Contract No. 68-02-2147.
July 1980.
Radian Corporation. "Study of Refinery Fugitive
Emissions from Equipment Leaks," API Health
and Environmental Sciences Dept. and Western
States Petroleum Assoc,, Volumes I and 2,
April 1994.
Open Ended
Lines
0.005
5.7e-7
Pump Seals
Light Liquid
- 0.25, Heavy
Liquid -
0,046
17 lb HC/
1000 bbl
Light Liquid-
0.25;
Heavy
Liquid-0.046
Heavy Liquid
4.3e-7
Compressor
Seals
1 .4 # THC/hr/
source
51bHC/1000bbl
HC-1.4
Valves
Gas - 0.059,
Light Liquid
-0.024,
Heavy Liquid
-0.00051
Valves and
flanges = 28
lb HC/1000
bbl capacity
Gas -0.059
Light Liquid-
0.024
Heavy
Liquid-
0.0005
6.5e-6
Pressure Relief
Valves (PVR)
0.36 # THC/hr/
source
1 lib HC/1000
bbl capacity
0.19
1.9e-8
Flares
5 lb HC/1000
bbl refining
capacity
Other
General storage @ refineries =
0.155
Drains - 0.07 Ib/hr/source
Drains & wastewater separators.
200 lb HC/1000 bb] water
Drains 0.07
(Confidence bounds given)
Light Liquid Pump Seal: 7.3e-6
(Confidence bounds given)
us
ON
(Continued)
-------
TABLE A-7c. REFINERY EMISSION DATABASE
(Continued)
Report Title
Rosehrook, D.D., R.G, Wetherold, and L.P,
Provost, "The Development of Fugitive
Emission Factors for the Petroleum Refining
Industry", Preseoted at the 72nd Annual Meeting
nf the Air Pollution Control Association, June
1979,
Wetherold, R.G., G.E. Harris, F.D, Skinner,
and L.P. Pravosl (Radian Corporation), "A
Model for Evaluation of Refinery and Synfuels
VOC Emission Data: Volumes 1 and 2."
Research Triangle Park, NC.
Open Ended
Lines
0.12
Pump Seals
Light Liquid-
0,26;
Heavy
Liquid-0.045
Light Liquid-
6;
Heavy
Li quid- 1.1
Compressor
Seals
Service:
HC - 0.98,
Hydrogen -0.1
HC Gas - 34,
Hydrogen - 2.6
Valves
G as/Vapor -
0.047,
Light Liquid-
0.023,
Heavy
Liquid-0.007
Light Liquid
- 0.58, Heavy
Liquid -
0.01 2, Gas -
1,4,
Hydrogen -
0.43
Pressure Relief
Valves (PVR)
Gas/Vapor - 0.36,
Light Liquid -
0.013, Heavy
Liquid -0.01 9,
all -0.19
Gas -8.6, Liquid -
0.37
Flares
Flares =
O.glb/1000
bbl crude
Other
Drains:
Light Liquid 0.085,
Heavy Liquid 0.029, A1J 0.07
Drains 1.7; Cooling Towers 6
Ib/lO^ gal water; Wastewater
treatment EFs given;
Slowdowns = 0.8 lb/1000 bbl
crude
-------
TABLE A-7d. REFINERY EMISSION DATABASE
Report Title
"Compilation of Air Pollutant Emission Factors (AP-42)," US
EPA,
"Hydrocarbon Emissions from Refineries," API Publication
No. 928, American Petroleum Institute, Washington, DC, July
1973.
C.E. Burklin, and R.L. Honercamp, "Revision of Evaporative
Hydrocarbon Emission Factors," EPA^50/3-76-039, U.S.
Environmental Protection Agency, Research Triangle Park,
NC, August 1977.
C.E. Bnrklin, et aL, Oil and Gas Field Emission Survey,
USEPA, Contract No. 68-D1-0031, April 1992,
Petroleum Supply Annual 1993, Volumes 1 and 2, Energy
Information Administration, US DOE, Washington DC, June
1994.
R.F. Boland, et al., Screening Study for Miscellaneous Sources
of Hydrocarbon Emissions in Petroleum Refineries, USEPA,
Dec. 1976.
Radian Corporation. " Assessment of Atmospheric Emissions
from Petroleum Refining: Volume 1." Prepared for U.S.
Environmental Protection Agency. Research Triangle Park,
NC.' Contract No. 68-02-2147. July 1980.
Wetherold, R.G., G.E, Harris, F.D. Skinner, and L,P. Provost
(Radian Corporation). "A Model for Evaluation of Refinery
and Synfuels VOC Emission Data: Volumes 1 and 2."
Research Triangle Park, NC.
Vacuum Distillation
Negligible
Vacuum gas disposal
6570 ton HC/yr
SOlbHC/lOOObbl
15034
Condenser Emission
Control (EC)- 18
Asphalt
asphalt blowing 165 ton HC/yr
60 Ib HC/ton asphalt
No EC - 60, Incinerator - 1.2
Ob VOC/ton asphalt)
Slowdown System
305-580 IbHC/l 000
bbl capacity
300 Ib HC/1000 bbl
capacity
Other
Reciprocating -1.4, Turbine - 0.02 #/1000
ftA3 burned
Wastewater 5.2 lb/1000 gal;
Also gives EFs for: boiler, compressor
engine, cooling tower, vacuum jet, and
overall refinery fugitives.
Compressors: Reciprocating - 9.7, Turbine
0.2 Ib HC/1000 hp-hr
Isomerization, reforming and alkylation
processes = 5459
Cooling Tower 0.00087S(+/-0.00165)n>
HC/1000 gal
OJ
oo
-------
APPENDIX B
Site Data
B-l
-------
APPENDIX B
SITE DATA
Table B-l shows the sample data set for the site visits conducted for the GRI/EPA natural
gas study of methane emissions from the natural gas study. This data set formed the basis for the
oil production extrapolated equipment counts. See Section 4.1.1 for an explanation of the
national equipment extrapolations methodology.
B-2
-------
TABLE B-l. EPA PETROLEUM METHANE SITE DATA SUMMARY
Site
Number
1
2
3
4
5
6
td 7
OJ
8
9
10
11
12
13
14
15
State
TX-OFF
LA-OFF
LA-OFF
CA-OFF
LA
LA
TX
TX
TX
TX
TX
TX
OK
OK
MT
Oil
Wells
3
150
40
22
50
3
3
300
155
127
1345
120
55
11
4
Throughput
(1000 B/D)
0.3
-
-
-
3.0
12.5
3
50
5.7
3.2
52
4.3
3.55
2.95
15
Separators
3
8
3
5
39
0
1
200
39
36
227
80
10
3
4
Heater
Treaters
0
-
-
-
43
9
0
500
-
-
9
0
140
5
-
Pneumatic
Devices
11
0
7
0
68
0
0
375
15
36
175
160
179
36
0
Chemical
Injection
Pumps
4
0
0
0
98
0
0
60
0
0
0
173
0
0
0
Gas Lift
Compressors
1
5
0
-
0
0
0
0
0
15
0
14
0
0
-
(Continued)
-------
TABLE B-l. EPA PETROLEUM METHANE SITE DATA SUMMARY
(Continued)
td
Site
Number
16
17
18
19
20
21
22
23
24
25
26
State
CA
CA
CA
CA
CA
CA
CA
CA
CA
OH
OH
Oil
Wells
913
18
8
10
15
20
7
728
4
163
418
Throughput
(1000 B/D)
71
0.3
0.03
0.03
0.07
0.143
0.09
47.9
0.0125
-
-
Separators
200
8
4
2
3
19
3
-
1
163
418
Heater
Treaters
0
1
2
0
0
0
0
-
0
-
-
Pneumatic
Devices
0
13
3
0
0
0
5
0
1
163
418
Chemical
Injection
Pumps
666
0
1
0
0
0
1
0
0
0
2
Gas Lift
Compressors
37
0
0
0
0
0
0
-
0
-
-
Note: The data for compressor starts and blowdowns are based on the number of gas-lift compressors. This information was gathered through follow-up calls to
sites visited during the natural gas study. For one of the sites, electric-driven reciprocating compressors were used for artificial lift (C02 injection for this
particular site). Since the compressors are electric driven, there are no methane emissions associated with compressor startup or blowdown. However, the
compressors still handle a gas stream, so fugitive methane emissions could result. For this particular site, however, the gas concentration was over 80% CO2 (the
compressors were used to move gas to and from a CO2 gas plant), which means that less than 20% of the gas being recovered could contain methane and/or other
tracecompounds. Because the percentage of methane is so small for this site, the extrapolation was simplified by not including the gas-lift compressors from this
site in the extrapolation of compressors for fugitive emissions.
-------
APPENDIX C
Statistical Analysis
C-l
-------
APPENDIX C
STATISTICAL ANALYSIS
Tables C-l and C-2 show the statistical analysis for the oil production equipment
extrapolations. The equipment was shown previously in Table 5-4. The statistical analysis is
based on the methods presented in Sections 4.1 and 4.2.
TABLE C-l. STATISTICAL ANALYSIS FOR EXTRAPOLATED ACTIVITY FACTORS
WELL BASIS
Equipment
Separators
Heater Treaters
Pneumatic
Devices
Chemical
Injection Pumps
Gas-Lift
Compressors
Sample
n Wells/n
25 158.5
17 169.7
26 180.4
26 180.4
21 159.9
N f t
3,683 0.00679 1.711
3,441 0.00494 1.746
3,236 0.00804 1.708
3,236 0.00804 1.708
3,653 0.00575 1.725
Equipment
Separators
Heater Treaters
Pneumatic
Devices
Chemical
Injection Pumps
Gas-Lift
Compressors
A
_ _ R
158.5 59.1 0.373
169.7 41.7 0.246
180.4 64.0 0.355
180.4 38.7 0.214
159.9 3.4 0.021
90%
YR u (YR ) tsqrt(u) Confidence
Interval
217,804 4,198,966,996 110,864 50.9%
143,491 15,303,125,160 215,976 150,5%
207,217 7,457,682,234 147,511 71.2%
125,088 5,912,326,706 131,342 105.0%
12,523 46,651,936 11,780 94.1%
Note:
x and y are the average number of wells and equipment per site in the sample data set, respectively.
A
YR is the extrapolated equipment count.
t is the Student's t Distribution with n-1 degrees of freedom.
i) = variance
n
N
f
A
R
number of sites sampled;
the total number of sites nationally;
sampling fraction = n/N; and
activity factor ratio = (AF/EP)sam]lll.
C-2
-------
TABLE C-2. STATISTICAL ANALYSIS FOR EXTRAPOLATED ACTIVITY FACTORS
THROUGHPUT BASIS
Equipment
Separators
Heater Trcaters
Pneumatic
Devices
Chemical
Injection Pumps
Gas-Lift
Compressors
Sample
n Wclls/n
20 11.4
17 12.0
21 13,1
21 13,1
19 11.2
N f t
603 0.03319 1.729
572 0.02970 1.746
523 0,04019 1.725
523 0.04019 1.725
613 0.03100 1,734
Equipment
Separators
Heater Treaters
Pneumatic
Devices
Chemical
Injection Pumps
Gas-Lift
Compressors
x y R
11.4 44.1 3.880
12,0 41.7 3.487
13.1 51.3 3.915
13.1 47.8 3.646
11.2 3.5 0.316
90%
A A
YR u(YR) tsqrt(u) Confidence
Interval
26,562 16,516,733 7,027 26.5%
23,873 252,393,777 27,737 116.2%
26,800 127,148,788 19,448 72.6%
24,959 178,189,572 23,023 92.2%
2,162 1,183,252 1,886 87.3%
Note:
x and y are the average throughput (1000 bbl/day/site) in the sample data set, respectively.
YR is the extrapolated equipment count.
C-3
-------
-------
APPENDIX D
Hypothetical Examples of Extrapolation and Bias
D-l
-------
APPENDIX D
HYPOTHETICAL EXAMPLES OF EXTRAPOLATION AND BIAS
If an equipment type were related to a single extrapolation parameter, yet
extrapolated by another parameter, the extrapolation will be correct or biased to the degree that
the relation between the extrapolation parameters at the site is correct or biased. See the simple
cases below for examples.
Case Example 1
Hypothesis: Equipment is related only to production rate, yet was extrapolated by well count.
Data: Sample set has a high production rate per well.
E = Equipment, P = Production, W = Wells (D-l)
I El
True, accurate extrapolation = - x Pnalional = Enatbnal (D,2)
\ / site
Actual extrapolation used: ^ — I x Wnational * EMtional (D.3)
P| I El
— where — is accurate,
E\ P\
Therefore, — is biased by — , which is HIGH (D-5)
WJ . I Wl .
/site \ / site
D-2
-------
Conclusion: Well count extrapolation will overestimate the actual value in this case.
Case Example 2
Hypothesis: Equipment is related only to well count, yet was extrapolated by production.
Data: Sample set has a high production rate per well.
I E I
True, accurate extrapolation - — x Wn = En (D-6)
I El
Actual extrapolation used: — x Pn = En (D-7)
E E W , E .
P J W T W 1SaCCUrate' (D-8)
I El I Wi
Therefore, — is biased by — , which is LOW (D-9)
\ / s \ " / site
Conclusion: production rate extrapolation will underestimate the actual value in this case.
D-3
-------
-------
APPENDIX E
Production Fugitive Emission Factors
E-l
-------
APPENDIX E
PRODUCTION FUGITIVE EMISSION FACTORS
The purpose of this appendix is to briefly discuss the data sources and development of
fugitive emission factors from leaking valves and fittings. The component method used here
involves using an average emission factor for each type of fitting that comprise a facility. The
average emission factor for each component type was determined by measuring the emission rate
from a large number of randomly selected components from similar types of facilities across the
country. The component emission factor is next combined with the average number of
components associated with major equipment or facilities to determine the average estimate of
emissions per equipment/facility.
Two reports were analyzed to determine fugitive emission factors for production. The
first report was API Report 4615,' The second was the 1995 EPA Protocols document2, which
includes EPA-approved fugitive component emission factors for oil and gas production. The API
report was chosen because it contained information not in the EPA Protocols. The API report
contained the methane fraction for light and heavy crude and the component count data.
Table E-l presents the resulting fugitive methane emission factors for oil industry equipment.
Only onshore fugitive equipment is presented.
TABLE E-1. CH. FUGITIVE EMISSION FACTORS (OIL INDUSTRY)
Emission Source Equipment EF Units
Oil wellheads (heavy crude)
Oil wellheads (light crude)
Separators (heavy crude)
Separators (light crude)
Heater/Treaters (light crude)
Headers (heavy crude)
Headers (light crude)
Compressors (light crude)
Small
Large
Sales Areas
0.83
19.58
0.85
51.33
59.74
0.59
202.78
46.14
16,360
40.55
scfd CH4/well
scfd CH4/well
scfd CH4/separator
scfd CH4/separator
scfd CH4/heater
scfd CH4/header
scfd CH4/header
scfd CH4/compressor
scfd CH4/compressor
scfd CH4/sales area
Star Environmental. API Publication No. 4615. "Emission Factors for Oil and Gas Production
Operations." American Petroleum Institute, January 1995.
2U.S.EPA. EPA-453/R-95-017. "1995 Protocol for Equipment Leak Emission Estimates." U.S.
Environmental Protection Agency, November 1995.
E-2
-------
The next step in estimating the national emissions from equipment leaks from oil industry
production equipment is to estimate the number of each type of equipment. This was done by
ratioing the other equipment as a function of the number of wellheads. Table E-2 lists the
relative population of other production equipment versus well counts according to the API field
counts.
TABLE E-2. RELATIVE POPULATIONS
PER API DAT A
Equipment Type
Wells
Separators
Headers
Tanks
Scrubber
Sales
Meters
Instruments
Heaters
Compressors
Light Crude
241
47
21
24
10
10
12
8
34
6
OF FIELD EQUIPMENT
Heavy Crude
183
5
68
--
-
--
-
-
..
For example, the ratio of light crude headers to light crude wellheads is 21/241 = 0.087
(or, 8.7 headers per 100 light crude wells). As explained in the body of this report, the national
population of light crude and heavy crude wellheads was estimated separately because both the
equipment emission factors and the field equipment populations were significantly different for
light crude and heavy crude. These reasons include: 1) different component emission factors-the
light crude component emission factors are one or two orders of magnitude higher; 2) different
component counts-the light crude separators and headers have significantly more components;
and 3) there are up to 12 field equipment types listed for light crude, versus only 3 for heavy
crude.
The underlying basis for the equipment emission factors are the counts and component
emission factors. As described in the API and EPA reports, there are several component types
such as connections, flanges, open-ended lines, valves, and other miscellaneous fittings.
Comparisons of the emission factors between the API and EPA reports were made to ensure that
the use of API or EPA factors would produce similar results for a given component type. Table
E-3 provides a comparison of THC fugitive emission factors between API 4615 and EPA
Protocols.
E-3
-------
TABLE E-3, COMPARISON BETWEEN LIGHT CRUDE AND HEAVY CRUDE
EMISSION FACTORS IN API/STAR 4615
Component Emission Factor
Service
Light Crude
Heavy
Crude
Connection
8.66E-03
(20 limes higher)
4.22E-04
Flange
4.07E-03
(3.5 times higher)
1.16E-03
Open-Ended
Line
6.38E-02
{7.8 times higher)
8.18E-03
(lb THC/day)
Valve
7.00E-02
(102 times higher)
6.86E-04
Other
3.97E-01
f 107 times higher)
3.70E-03
COMPARISON BETWEEN API 4615 AND EPA PROTOCOLS
(FEBRUARY 1996)
Component Emission Factor (lb THC/day)
Service
Light
Crude
Heavy
Crude
Basis
API
4615
EPA
API
4615
EPA
Connection Flange
8.66E-03 4.07E-03
1.1E-02 5.8E-03
4.22E-04 1.16E-03
3.97E-04 2.1E-05
Open-Ended
Line
6.38E-02
7.4E-02
8.18E-03
7.4E-03
Valve
7.00E-02
1.32E-01
6.86E-04
4.4E-04
Other
3.97E-01
3.97E-01
3.70E-03
1.7E-03
The data in Table E-3 show that the light crude component emission factors are higher
than the heavy crude factors. While some differences exist between API and EPA factors, in
most cases the values are similar and if EPA factors were used instead, the fugitive estimates in
this report would not differ drastically.
E-4
-------
APPENDIX F
Tank Flash Calculations
F-l
-------
APPENDIX F
TANK FLASH CALCULATIONS
Methane emissions from a fixed-roof crude oil gathering tank were calculated using a
process flow model and standard design parameters for oil field equipment. It was assumed that
the crude oil in the gathering tank was separated from well head gas in a standard gas/oil
separator upstream of the gathering tank. It was also assumed that the crude oil leaves the gas/oil
separator in equilibrium with a pure methane stream that has been separated from the oil. The oil
leaving the separator contains dissolved methane in equilibrium with the temperature and
pressure of the separator. When the crude oil enters the gathering tank which operates at near
atmospheric pressure, the dissolved methane is flashed to the vapor phase and vented from the
tank. The assumption of being in equilibrium with pure methane results in conservatively high
estimates for gathering tank methane emissions.
The ASPEN Plus process simulator model was used to calculate the oil field gathering
tank emissions. The process flow configuration modeled in ASPEN is shown in Figure F-l. The
separator temperature and pressure were varied across the standard range of process conditions
experienced in the oil field to test the sensitivity of the methane emissions to standard process
conditions. The crude oil was assumed to be an East Texas Intermediate Crude, (however
analysis done later proved that the choice of the crude oil had minimal impact on the emissions
from the gathering tanks). The crude oil entering the gathering tank was assumed to be at the
same temperature as the gas/oil separator, and the pressure of the crude gathering tank was
assumed to be 14.8 psia (-0.1 psig) for all cases.
Table F-l presents the results from the ASPEN Plus process simulation for oil field
gathering tanks. The methane emission rates from the gathering tank range from 7.3 to 27.1
scf/bbl of oil throughput. The annual US crude production of 2.339 billion barrels of crude
yields an annual methane emission rate of 17.1 to 63.4 Bscf CH4/yr. For comparison the ASPEN
model was used to calculate the annual methane emissions from natural gas field condensate. As
shown in Table F-2, annual emissions are (5.1 to 18.3 scf/bbl) x 788 million barrels of
condensate/yr =4.02 to 14.4 Bscf CH4/yr.
F-2
-------
Sales Gas
Emissions
Methane
SEPARATOR
TANK
Sales
Oil
Crude Oil
Figure F-l. Process Flow Diagram for Simulations to Determine
Methane Emissions from Oil/Condensate Production Tanks
-------
TABLE F-l. METHANE EMISSIONS FROM CRUDE OIL TANKS
WITH VARIOUS PROCESS CONDITIONS
Case
1
2
3
4
5
6
7
8
9
TABLE
Case
1
2
3
4
5
6
7
Separator T
60
60
60
85
85
85
100
100
100
(°F) Separator P (psig)
20
40
60
20
40
60
20
40
60
Methane Emissions
(scf/bbl)
8.8
17,9
27.1
7.8
15.9
24.1
7.3
14,9
22.6
F-2. METHANE EMISSIONS FROM CRUDE OIL TANKS
WITH VARIOUS PROCESS CONDITIONS
Separator T
60
60
85
85
85
100
100
(°F) Separator P (psig)
20
60
20
40
60
20
60
Methane Emissions
(scf/bbl)
6.0
18.3
5.4
10.9
16.5
5.1
15.6
F-4
-------
APPENDIX G
English/Metric Conversions
G-l
-------
APPENDIX G
1 scf methane
1 Bscf methane
1 Bscf methane
IBscf
1 short ton (ton)
lib
1ft3
1ft3
1 gallon
1 barrel (bbl)
1 inch
1ft
1 mile
Ihp
1 hp-hr
IBtu
1 MMBtu
1 Ib/MMBtu
T(°F)
1 psi
Unit Conversion Table
English to Metric Conversions
19.23 g methane
0.01923 Tg methane
19,230 metric tonnes methane
28.32 million standard cubic meters
907.2 kg
0.4536 kg
0.02832 m3
28.32 liters
3.785 liters
158.97 liters
2.540 cm
0.3048 m
1.609km
0.7457 kW
0.7457 kW-hr
1055 Joule
293 kW-hr
430 g/GJ
1.8T(°C) + 32
51.71 mm Hg
Notes
scf
Bscf
MMscf
Mscf
Tg
Giga (G)
Metric tonne
psig
psia
Standard cubic feet. Standard conditions are at 14.73 psia and 60°F.
Billion standard cubic feet (109 scf).
Million standard cubic feet.
Thousand standard cubic feet.
Teragram(1012g).
Same as billion (109).
1000kg.
Gauge pressure.
Absolute pressure (note psia = psig + atmospheric pressure).
G-2
-------
APPENDIX H
Methane Emissions 1986-1992
H-l
-------
Table H-1
1986 Methane Emission Estimate
Petroleum - Production
Emission Source
Annual Production
% Heavy Crude (API<20°)
Total Producing Oil Wells 1993
% Heavy Wells (API<20°)
Fugitives:
Offshore Platforms
Gulf of Mexico
Rest of US
Oil Wellheads (heavy crude)
Oil Wellheads (light crude)
Separators (heavy crude)
Separators (light crude)
Heater/Treaters (light crude)
Headers (heavy crude)
Headers (light crude)
Tanks (light crude)
Compressors (light crude)
Small
Large
Sales Areas
Pipelines
Venting:
Oil Tanks
Pneumatic Devices
CIPs
Vessel Slowdowns
Compressor Starts
Compressor Blowdowns
Completion Flaring
Well Workover
Casinghead Gas
Upsets:
BSD
PRV Lifts
Well Blowout
Combustion Sources:
Gas Engines
Burners
Drilling
Flares
Total
Emission
Factor
2914
1178
0.83
19.58
0.85
51,33
59.74
0.59
202.78
34.4
46.14
16,360
40.55
56.4
12.1
345
248
78
8443
3774
733
96
256,888
34
250,000
0.24
0.526
0.052
Methane
Emissions Units
scfd CH4/platform
scfd CH4/platform
scfd CH4/weli
scfd CH4/well
scfd CH4/sep
scfd CH4/sep
scfd CH4/heater
scfd CH4/header
scfd CH4/header
scfd CH4/tank
scfd CH4/comp
scfd CH4/comp
scfd CH4/area
scfd CH4/mile
scfd CH4/bbl
scfd CH4/device
scfd CH4/pump
scfy CH4/vessel
scfy CH4/comp.
scfy CH4/comp.
scfd CH4/completion
scf CH4/workover
scfy CH4/plat
scfy CH4/PRV
scf CH4/blowout
scf CH4/HPhr
lbCH4/1000gaI
ton CH4/well drilled
Confidence
Interval
27%
36%
30%
30%
30%
30%
30%
30%
30%
30%
100%
100%
30%
97%
88%
40%
83%
266%
157%
147%
200%
200%
200%
252%
200%
5%
10%
100%
Activity
Factor
8,680,000
1 1 .7%
623,000
7.3%
1,092
22
45,230
577,770
10,402
122,627
84,374
16,807
50,345
57,777
706
2,118
4,443
70,000
8,680,000
127,541
133,447
224,729
3,069
3,069
989
46,725
1,114
461,737
2.85
19,337
17,806,000
989
Activity
Units
bbi/d
wells
platforms
platforms
wells
wells
separators
separators
heater treaters
headers
headers
tanks
small g.l. comp.
large g.l. comp.
sales areas
miles
bbl/d
pneumatics
CIPs
sep. and h.t.
gas lift comp.
gas lift comp.
completions
w.o./year
platforms
PRV
blowouts/yr
MMhp-hr
bbl/year
expl. wells
Confidence
Interval
5%
100%
5%
100%
10%
10%
100%
100%
78%
78%
128%
109%
109%
109%
90%
90%
46%
50%
5%
78%
105%
70%
79%
79%
10%
421%
10%
102%
200%
275%
5%
10%
Emissions
(Bscf)
1.161
0.009
0.014
4.129
0.003
2.297
1.840
0.004
3.726
0.725
0.012
12.646
0.066
1.441
38.335
16.061
12.080
0.018
0.026
0.012
0.265
0.004
0.286
0.016
0.001
4.641
0.009
0.002
99.83
Confidence
Interval
29%
38%
1 09%
109%
87%
87%
137%
118%
118%
118%
162%
162%
57%
119%
88%
93%
1 60%
332%
216%
204%
201%
962%
201%
374%
490%
275%
11%
101%
48.8%
to
-------
Table H-2
1986 Methane Emission Estimate
Petroleum - Crude Transportation
Emission Source
Fugitives:
Pump Stations
Pipelines
Metering
Venting:
Tanks
Loading
Truck
Marine
Rail Car
Maintenance:
Pump Stations
Combustion Sources:
Pump engine drivers
Heaters
Total
Emission
Factor
1.06
0,0
4.37E-07
1.02E-05
0.5
1.02E-05
1.56
0.24
Methane
Emissions Units
Ib CH4/yr/mile
Ib CH4/bbl
ton CH4/bbl
ton CH4/bbi
lbCH4/1 000 gal crude
ton CH4/bbl
Ib CH4/y/station
scf CH4/HPhr
Confidence
Interval
100%
10%
100%
100%
100%
100%
100%
5%
Activity
Factor
54,153
6.29E+09
8173000000
7.48E+07
7.52E+10
2.05E+07
542
Activity
Units
miles
bbl/yr
bbl/yr
bbl/yr
gal/yr
bbl/yr
stations
Confidence
Interval
100%
5%
4%
10%
10%
10%
100%
Emissions
(Bscf)
0,0014
0.0000
0.169
0.036
0,892
0.010
0.000
1.108
Confidence
Interval
173%
11%
100%
101%
101%
101%
173%
82.7%
-------
Table H-3
1986 Methane Emission Estimate
Petroleum - Refining
Emission Source
Fugitives:
Fuel Gas System
Pipe Stills
Wastewater Treating
Cooling Towers
Venting:
Tanks
System Slowdowns
Process Vents
Upssts
PRVs
Combustion Sources:
Process Heaters:
Atm. Distillation
Vacuum Distil.
Thermal Operations
Cat. Cracking
Cat. Reforming
Cat. Hydrocraking
Cat. Hydrorefining
Cat. Hydrotreating
Alky! & Polymer.
Aromatics/lsomeration
Lube Processing
Asphalt
Hydrogen
Coke
Engines and Flares
Engines
Flares
Total
Emission
Factor
1.02
0.00798
0.01
4.37E-07
580
0.30
0.30
0.50
0.43
0.60
0.60
0.18
0.54
1.05
0.15
0.0
60
0.0
0
0.24
0.0008
Methane
Emissions Units
MMscf CH4/heater/yr
Ib VOC/bbl
Ib VOC/bbl
ton CH4/bbl
# HC/1000 bbc capacity
IbTHC/1000bbl
!bTHC/1000bbl
Ib THC/1 000 bbl
IbTHC/IOOObbl
Ib THC/1 000 bbl
IbTHC/IOOObbl
Ib THC/1 000 bbi
IbTHC/IOOObbl
IbTHC/IOOObbl
IbTHC/IOOObbl
# HC/ton
Included in Thermal Ops
scf CH4/hp-hr
Ib VOC/bb!
Confidence
Interval
100%
100%
100%
100%
100%
100%
100%
100%
100%
100%
100%
100%
100%
100%
1 00%
100%
100%
100%
100%
5%
100%
% Methane
in THC*
1.0%
1.0%
1 .0%
51.0%
51.0%
51.0%
51 .0%
51.0%
51.0%
51 .0%
51.0%
51.0%
51.0%
51.0%
1 .0%
Activity
Factor
3,200
12,715,257
12,715,257
12,715,257
12,715,257
12,715,257
5,662,334
1,566,299
4,855,860
3,147,793
966,926
1,935,990
5,634,335
805,044
532,582
203,018
581 ,578
20,334
12,715,257
Activity
Units
heaters
b/d
b/d
b/d
b/d
b/d
b/d
b/d
b/d
b/d
b/d
b/d
b/d
b/d
b/d
b/d
b/d
MMhp-hr
b/d
Confidence
Interval
50%
5%
5%
5%
5%
5%
5%
5%
5%
5%
5%
5%
5%
5%
5%
5%
5%
100%
5%
Emissions
(Bscf)
3.26
0.009
0.011
0.096
0.638
0.017
0.008
0.003
0.009
0.008
0.003
0.002
0.014
0.004
0.000
0.000
0.154
0.000
4.880
0.001
9.117
Confidence
Interval
122%
100%
100%
100%
100%
100%
100%
100%
100%
100%
100%
100%
1 00%
100%
100%
100%
100%
100%
100%
69.70%
4-
* % Methane in VOC (volatile organic compounds) taken from AP-42 (Reference 28)
% Methane in HC (hydrocarbons) for system blowdowns is taken from AP-42
% Methane in THC (total hydrocarbons) calculated based on data from AP-42
% Methane in HC for asphalt calculated based on data from AP-42
-------
Table H-4
1987 Methane Emission Estimate
Petroleum - Production
Emission Source
Annual Production
% Heavy Crude (APIc20°)
Total Producing Oii Wells
% Heavy Wells (API<20C)
Fugitives:
Offshore Platforms
Gulf of Mexico
Rest of US
Oil Wellheads (heavy crude)
Oil Wellheads (light crude)
Separators (heavy crude)
Separators {light crude)
Heater/Treaters (light crude)
Headers (heavy crude)
Headers (light crude)
Tanks (light crude)
Compressors (light crude)
Small
Large
Sales Areas
Pipelines
Venting:
Oil Tanks
Pneumatic Devices
CIPs
Vessel Slowdowns
Compressor Starts
Compressor Slowdowns
Completion Flaring
Weil Workover
Casinghead Gas
Upsets:
ESD
PRV Lifts
Well Blowout
Combustion Sources:
Gas Engines
Burners
Drilling
Flares
Total
Emission
Factor
2914
1178
0.83
19.58
0.85
51.33
59.74
0.59
202.78
34.4
46.14
16,360
40.55
56.4
12.1
345
243
78
8443
3774
733
96
256,688
34
250,000
0.24
0.526
0.052
Methane
Emissions Units
scfd CH4/platform
scfd CH4/p!atform
scfd CH4/well
scfd CH4/well
scfd CH4/sep
scfd CH4/sep
scfd CH4/heater
scfd CH4/header
scfd CH4/header
scfd CH4/tank
scfd CH4/comp
scfd CH4/comp
scfd CH4/area
scfd CH4/mile
set CH4/bbl
scfd CH4/device
scfd CH4/pump
scfy CH4/vessel
scfy CH4/comp,
scfy CH4/comp.
scfd CH4/completion
set CH4/workover
scfy CH4/piat
scfy CH4/PRV
scf CH4/blowout
scf CH4/HPhr
lbCH4/1000gal
ton CH4/well drilled
Confidence
interval
27%
36%
30%
30%
30%
30%
30%
30%
30%
30%
100%
100%
30%
97%
88%
40%
83%
266%
157%
147%
200%
200%
200%
252%
200%
5%
10%
100%
Activity
Factor
8,349,000
11.0%
620,000
7.4%
1,092
22
45,942
574,058
10,353
121,474
83,504
17,071
50,022
57,406
699
2,096
4,443
70,000
8,349,000
126,361
132,804
222,562
3,037
3,037
857
46,500
1,114
457,313
2.85
19,135
12,497,000
857
Activity
Units
bbl/d
wells
platforms
platforms
wells
wells
separators
separators
heater treaters
headers
headers
tanks
small g.l. comp.
large g.l. comp.
sales areas
miles
bbl/d
pneumatics
CIPs
sep. and h.t.
gas lift comp.
gas lift comp.
completions
w.oVyear
platforms
PRV
blowouts/yr
MMhp-hr
bbl/year
expl. wells
Confidence
Interval
5%
100%
5%
100%
10%
10%
100%
100%
78%
78%
128%
109%
109%
109%
90%
90%
46%
50%
5%
78%
105%
70%
80%
80%
10%
421%
10%
102%
200%
275%
5%
10%
Emissions
(Bscf)
1.161
0.009
0.014
4.103
0.003
2.276
1.821
0.004
3.702
0.721
0.012
12.514
0.066
1.441
36.873
15.912
12.021
0.017
0.026
0.011
0.229
0.004
0.286
0.016
0.001
4.592
0.007
0.002
97.84
Confidence
Interval
29%
38%
109%
109%
87%
87%
137%
118%
118%
118%
162%
162%
57%
119%
88%
93%
160%
332%
216%
204%
201%
962%
201%
374%
490%
275%
11%
101%
48.7%
-------
Table H-5
1987 Methane Emission Estimate
Petroleum - Crude Transportation
Emission Source
Fugitives:
Pump Stations
Pipelines
Metering
Venting:
Tanks
Loading
Truck
Marine
Rail Car
Maintenance:
Pump Stations
Combustion Sources:
Pump engine drivers
] Heaters
(Total
Emission
Factor
1.06
0.0
4.37E-07
1.02E-05
0.5
1.02E-05
1.56
0.24
Methane
Emissions Units
Ib CH4/yr/mi!e
Ib CH4/bb!
ton CH4/bbl
ton CH4/bbl
Ib CH4/1 000 gal crude
ton CH4/bbl
Ib CH4/y/station
scf CH4/HPhr
Confidence
Interval
100%
10%
100%
100%
100%
100%
100%
5%
Activity
Factor
54,886
6.28E+09
8293000000
6.82E+07
8.09E+10
2.08E+07
549
Activity
Units
miles
bbl/yr
bbl/yr
bb!/yr
gal/yr
bbl/yr
stations
Confidence
Interval
100%
5%
4%
10%
10%
10%
100%
Emissions
(Bscf)
0.0014
0.0000
0.172
0.033
0.960
0.010
0.000
1.176
Confidence
Interval
173%
11%
100%
101%
101%
101%
173%
83.8%
-------
Table H-6
1987 Methane Emission Estimate
Petroleum - Refining
Emission Source
Fugitives:
Fuel Gas System
Pipe Stills
Wastewater Treating
Cooling Towers
Venting:
Tanks
System Slowdowns
Process Vents
Upsets
PRVs
Combustion Sources:
Process Heaters:
Atm. Distillation
Vacuum Distil.
Thermal Operations
Cat, Cracking
Cat. Reforming
Cat Hydrocraking
Cat, Hydrorefining
Cat. Hydrotreating
Alkyi & Polymer.
Aromatics/isomeration
Lube Processing
Asphalt
Hydrogen
Coke
Engines and Flares
Engines
Flares
Total
Emission
Factor
1.02
0.00798
0.01
4.37E-07
580
0.30
0.30
0.50
0.43
0.60
0.60
0.18
0.54
1.05
0.15
0.0
60
0.0
0
0.24
0.0008
Methane
Emissions Units
MMscf CH4/heater/yr
Ib VOC/bbI
Ib VOC/bbI
ton CH4/bbl
#HC/1 000 bbc capacity
Ib THC/1000 bbl
Ib THC/1000 bbi
Ib THC/1000 bb!
Ib THC/1000 bbl
Ib THC/1000 bbl
lb THC/1000 bbl
Ib THC/1000 bbl
Ib THC/1000 bbl
Ib THC/1000 bbl
Ib THC/1000 bbl
# HC/ton
Included in Thermal Ops
scf CH4/hp-hr
Ib VOC/bbI
Confidence
Interval
100%
100%
1 00%
100%
1 00%
100%
100%
100%
100%
100%
100%
100%
100%
100%
100%
100%
100%
100%
100%
5%
1 00%
% Methane
in THC'
1 .0%
1 .0%
1 .0%
51.0%
51.0%
51.0%
51.0%
51.0%
51.0%
51.0%
51.0%
51.0%
51.0%
51.0%
1 .0%
Activity
Factor
3,200
12,853,848
12,853,848
12,853,848
12,853,848
12,853,848
5,861,768
1,571,033
4,846,366
3,201,246
980,399
1,882,196
5,713,742
832,898
604,089
1 97,448
596,089
20,334
12,853,848
Activity
Units
heaters
b/d
b/d
b/d
b/d
b/d
b/d
b/d
b/d
b/d
b/d
b/d
b/d
b/d
b/d
b/d
b/d
MMhp-hr
b/d
Confidence
Interval
50%
5%
5%
5%
5%
5%
5%
5%
5%
5%
5%
5%
5%
5%
5%
5%
5%
100%
5%
Emissions
(Bscf)
3.26
0.009
0.011
0.097
0.645
0.017
0.008
0.003
0.009
0.008
0.003
0.001
0.014
0.004
0.000
0.000
0.158
0.000
4.880
0.001
9.128
Confidence
Interval
122%
100%
100%
100%
100%
100%
100%
1 00%
100%
100%
100%
100%
100%
1 00%
100%
100%
100%
100%
100%
69.6%
% Methane in VOC (volatile organic compounds) taken from AP-42 (Reference 28)
% Methane in HC (hydrocarbons) for system blowdowns is taken from AP-42
% Methane in THC (total hydrocarbons) calculated based on data from AP-42
% Methane in HC for asphalt calculated based on data from AP-42
-------
Table H-7
1988 Methane Emission Estimate
Petroleum - Production
Emission Source
Annual Production
% Heavy Crude (APl<20°)
Total Producing Oil Wells
% Heavy Wells (API<20°)
Fugitives:
Offshore Platforms
Gulf of Mexico
Rest of US
Oil Wellheads (heavy crude)
Oil Wellheads (light crude)
Separators (heavy crude)
Separators (light crude)
Heater/Treaters (light crude)
Headers (heavy crude)
Headers (light crude)
Tanks (light crude)
Compressors (light crude)
Small
Large
Sales Areas
Pipelines
Venting;
Oil Tanks
Pneumatic Devices
CIPs
Vessel Slowdowns
Compressor Starts
Compressor Slowdowns
Completion Flaring
Well Workover
Casinghead Gas
Upsets:
ESD
PRV Lifts
Well Blowout
Combustion Sources:
Gas Engines
Burners
Drilling
Flares
Total
Emission
Factor
2914
1178
0.83
19.58
0.85
51.33
59.74
0.59
202.78
34.4
46.14
16,360
40.55
56.4
12.1
345
248
78
8443
3774
733
96
256,888
34
250,000
0.24
0.526
0.052
Methane
Emissions Units
scfd CH4/p1atform
scfd CH4/platform
scfd CH4/well
scfd CH4/weli
scfd CH4/sep
scfd CH4/sep
scfd CH4/heater
scfd CH4/header
scfd CH4/header
scfd CH4/tank
scfd CH4/comp
scfd CH4/comp
scfd CH4/area
scfd CH4/miIe
scf CH4/bbl
scfd CH4/device
scfd CH4/pump
scfy CH4/vessel
scfy CH4/comp.
scfy CH4/comp.
scfd CH4/completion
scf CH4/workover
scfy CH4/pfat
scfy CH4/PRV
scf CH4/blowout
scf CH4/HPhr
Ib CH4/1000 gal
ton CH4/welI drilled
Confidence
Interval
27%
36%
30%
30%
30%
30%
30%
30%
30%
30%
100%
100%
30%
97%
88%
40%
83%
266%
157%
147%
200%
200%
200%
252%
200%
5%
10%
100%
Activity
Factor
8,140,000
1 1 .0%
612,000
7.3%
1,092
22
44,921
567,079
10,108
119,621
82,331
16,692
49,414
56,708
689
2,066
4,443
70,000
8,140,000
124,532
131,090
219,337
2,992
2,992
791
45,900
1,114
450,642
2.85
18,849
14,697,000
791
Activity
Units
bbl/d
wells
platforms
platforms
wells
wells
separators
separators
heater treaters
headers
headers
tanks
small g.l. comp.
large g.l. comp.
sales areas
miles
bbl/d
pneumatics
CIPs
sep. and h.t.
gas lift comp.
gas lift comp.
completions
w.oVyear
platforms
PRV
blowouts/yr
MMhp-hr
bbl/year
expl. wells
Confidence
Interval
5%
100%
5%
100%
10%
10%
1 00%
100%
78%
78%
129%
109%
1 09%
1 09%
90%
90%
46%
50%
5%
78%
105%
70%
80%
80%
10%
421%
10%
102%
200%
275%
5%
10%
Emissions
(Bscf)
1.161
0.009
0.014
4.053
0.003
2.245
1,795
0.004
3.657
0.712
0.012
12.338
0.066
1.441
35.950
1 5.682
1 1 .866
0.017
0.025
0.011
0.212
0.004
0.286
0.015
0.001
4.524
0.008
0.002
96.11
Confidence
Interval
29%
38%
109%
109%
87%
87%
138%
118%
118%
118%
162%
162%
57%
119%
88%
93%
160%
332%
216%
205%
201%
962%
201%
375%
490%
276%
11%
101%
48.7%
oo
-------
Table H-8
1988 Methane Emission Estimate
Petroleum - Crude Transportation
Emission Source
Fugitives:
Pump Stations
Pipelines
Metering
Venting:
Tanks
Loading
Truck
Marine
Rail Car
Maintenance;
Pump Stations
Combustion Sources:
Pump engine drivers
Heaters
Total
Emission
Factor
1.06
0.0
4.37E-07
1 .02E-05
0.5
1.02E-05
1.56
0.24
Methane
Emissions Units
Ib CH4/yr/miIe
Ib CH4/bbl
ton CH4/bbI
ton CH4/bbl
lbCH4/1 000 gal crude
ton CH4/bbI
Ib CH4/y/station
scf CH4/HPhr
Confidence
Interval
100%
10%
1 00%
100%
1 00%
1 00%
100%
5%
Activity
Factor
55,900
6.51 E+09
8668000000
6.81 E+07
8.69E+10
2.22E+07
559
Activity
Units
miles
bbl/yr
bbl/yr
bbl/yr
gal/yr
bbl/yr
stations
Confidence
Interval
100%
5%
4%
10%
10%
10%
100%
Emissions
(Bscf)
0.0014
0.0000
0.180
0.033
1.030
0.011
0.000
1.255
Confidence
Interval
173%
11%
100%
101%
101%,
101%
173%
84.2%
-------
Table H-9
1988 Methane Emission Estimate
Petroleum - Refining
Emission Source
Fugitives:
Fuel Gas System
Pipe Stills
Wastewater Treating
Cooling Towers
Venting:
Tanks
System Slowdowns
Process Vents
Upsels
PRVs
Combustion Sources;
Process Heaters:
Atm. Distillation
Vacuum Distil.
Thermal Operations
; Cat. Cracking
! Cat, Reforming
I Cat. Hydrocraking
Cat. Hydrorefining
Cat. Hydrotreating
Alky! & Polymer.
Aromatics/lsomeration
Lube Processing
Asphalt
Hydrogen
Coke
Engines and Flares
Engines
Flares
Total
Emission
Factor
1.02
0.0079B
0.01
4.37E-07
580
0.30
0.30
0.50
0.43
0.60
0.60
0.18
0.54
1.05
0.15
0.0
60
0.0
0
0.24
0.0008
Methane
Emissions Units
MMscf CH4/heater/yr
Ib VOC/bbi
Ib VOC/bbi
ton CH4/bbi
# HC/1000 bbc capacity
IbTHC/IOOObbl
IbTHC/IOOObbi
IbTHC/IOOObbl
IbTHC/IOOObbl
IbTHC/IOOObbl
IbTHC/IOOObbl
IbTHC/IOOObbl
ibTHC/1000bbl
IbTHC/IOOObbl
IbTHCAIOOObb!
# HG/ton
Included in Thermal Ops
scf CH4/hp-hr
Ib VOC/bbi
Confidence
Interval
100%
100%
100%
1 00%
100%
100%
100%
100%
100%
1 00%
100%
100%
100%
100%
100%
100%
100%
100%
1 00%
5%
100%
% Methane
in THC*
1 .0%
1 .0%
1 .0%
51.0%
51.0%
51.0%
51.0%
51.0%
51,0%
51.0%
51 .0%
51.0%
51.0%
51,0%
1 .0%
Activity
Factor
3,200
13,246,238
13,246,238
1 3,246,238
1 3,246,238
13,246,236
6,084,486
1 ,644,395
4,766,257
3,352,139
1,024,133
2,059,005
6,099,481
953,940
667,009
201,518
654,990
20,334
13,246,238
Activity
Units
heaters
b/d
b/d
b/d
b/d
b/d
b/d
b/d
b/d
b/d
b/d
b/d
b/d
b/d
b/d
b/d
b/d
MMhp-hr
b/d
Confidence
Interval
50%
5%
5%
5%
5%
5%
5%
5%
5%
5%
5%
5%
5%
5%
5%
5%
5%
100%
5%
Emissions
(Bscf)
3.26
0.009
0.011
0.100
0.665
0,018
0.008
0.004
0.009
0,009
0.003
0.002
0.015
0.004
0.000
0.000
0,174
0.000
4.880
0.001
9.172
Confidence
Interval
1 22%
1 00%
100%
1 00%
100%
100%
100%
100%
100%
100%
100%
1 00%
100%
100%
100%
1 00%
100%
100%
100%
69,3%
* % Methane in VOC (volatile organic compounds) taken from AP-42 (Reference 28)
% Methane in HC (hydrocarbons) for system blowdowns is taken from AP-42
% Methane in THC (total hydrocarbons) calculated based on data from AP-42
% Methane in HC for asphalt calculated based on data from AP-42
-------
Table H-10
1989 Methane Emission Estimate
Petroleum - Production
Emission Source
Annual Production
% Heavy Crude (API<20°)
Total Producing Oil Wells
% Heavy Wells (API<20°)
Fugitives:
Offshore Platforms
Gulf of Mexico
Rest of US
Oil Wellheads (heavy crude)
Oil Wellheads (light crude)
Separators (heavy crude)
Separators (light crude)
Heater/Treaters (light crude)
Headers (heavy crude)
Headers (light crude)
Tanks (light crude)
Compressors (light crude)
Small
Large
Sales Areas
Pipelines
Venting:
Oil Tanks
Pneumatic Devices
CIPs
Vessel Slowdowns
Compressor Starts
Compressor Slowdowns
Completion Flaring
Well Workover
Casinghead Gas
Upsets:
ESD
PRV Lifts
Well Blowout
Combustion Sources:
Gas Engines
Burners
Drilling
Flares
Total
Emission
Factor
2914
1178
0.83
19.58
0.85
51.33
59.74
0.59
202.78
34.4
46.14
16,360
40.55
56.4
12.1
345
248
78
8443
3774
733
96
256,888
34
250,000
0.24
0,526
0.052
Methane
Emissions Units
scfd CH4/platform
scfd CH4/platform
scfd CH4/well
scfd CH4/well
scfd CH4/sep
scfd CH4/sep
scfd CH4/heater
scfd CH4/header
scfd CH4/header
scfd CH4/tank
scfd CH4/comp
scfd CH4/comp
scfd CH4/area
scfd CH4/mile
scf CH4/bbl
scfd CH4/device
scfd CH4/pump
scfy CH4/vessel
scfy CH4/comp.
scfy CH4/comp.
scfd CH4/completion
scf CH4/workover
scfy CH4/plat
scfy CH4/PRV
scf CH4/blowout
scf CH4/HPhr
IbCH4/lOGOgal
ton CH4/well drilled
Confidence
Interval
27%
36%
30%
30%
30%
30%
30%
30%
30%
30%
100%
100%
30%
97%
88%
40%
83%
266%
157%
147%
200%
200%
200%
252%
200%
5%
10%
100%
Activity
Factor
7,612,000
10.9%
603,365
7.2%
1,092
22
43,322
560,043
9,694
117,601
80,651
16,098
48,800
56,004
674
2,023
4,443
70,000
7,612,000
121,966
129,241
214,720
2,925
2,925
560
45,252
1,114
441,140
2.85
18,427
10,120,000
580
Activity
Units
bbl/d
wells
platforms
platforms
wells
wells
separators
separators
heater treaters
headers
headers
tanks
small g.l. comp.
large g.l. comp.
sales areas
miles
bbl/d
pneumatics
CiPs
sep. and h.t.
gas lift comp.
gas lift comp.
completions
w.o./year
platforms
PRV
blowouts/yr
MMhp-hr
bbl/year
expl. wells
Confidence
Interval
5%
100%
5%
100%
10%
10%
100%
100%
78%
78%
130%
109%
109%
109%
91%
91%
46%
50%
5%
78%
105%
70%
81%
81%
10%
421%
10%
102%
200%
276%
5%
10%
Emissions
(Bscfl
1.161
0.009
0.013
4.002
0.003
2.203
1.759
0.003
3.612
0.703
0.011
12.083
0.066
1.441
33.618
15.359
11.699
0.017
0.025
0.011
0.155
0.004
0.286
0.015
0.001
4.422
0.005
0.001
92.69
Confidence
Interval
29%
38%
109%
109%
87%
87%
139%
118%
118%
118%
163%
163%
57%
119%
88%
93%
160%
332%
217%
205%
201%
962%
201%
375%
490%
277%
11%
101%
48.5%
-------
Table H-11
1989 Methane Emission Estimate
Petroleum - Crude Transportation
Emission Source
Fugitives:
Pump Stations
Pipelines
Metering
Venting:
Tanks
Loading
Truck
Marine
Rail Car
Maintenance:
Pump Stations
Combustion Sources:
Pump engine drivers
Heaters
Total
Emission
Factor
1.06
0.0
4.37E-07
1.02E-05
0.5
1.02E-05
1.56
0.24
Methane
Emissions Units
Ib CH4/yr/mile
ib CH4/bbl
Ion CH4/bbi
ton CH4/bbI
ibCH4/1 000 gal crude
Ion CH4/bbi
tb CH4/y/station
set CH4/HPhr
Confidence
Interval
100%
10%
100%
100%
100%
100%
100%
5%
Activity
Factor
55,664
6.44E+09
8686000000
6.86E+07
9.09E+10
1 .98E+07
557
Activity
Units
miles
bbl/yr
bbl/yr
bblVr
gal/yr
bbl/yr
stations
Confidence
Interval
100%
5%
4%
10%
10%
10%
100%
Emissions
(Bscf)
0.0014
0.0000
0.180
0.033
1.078
0.010
0.000
1.302
Confidence
Interval
173%
11%
100%
101%
101%
101%
173%
84.8%
-------
TableH-12
1989 Methane Emission Estimate
Petroleum - Refining
Emission Source
Fugitives;
Fuel Gas System
Pipe SJilis
Wastewater Treating
Cooling Towers
Venting:
Tanks
System Slowdowns
Process Vents
Upsets
PRVs
Combustion Sources:
Process Heaters:
Atm. Distillation
Vacuum Distil.
Thermal Operations
Cat. Cracking
Cat, Reforming
I Cat. Hydrocraking
Cat. Hydrorefinlng
Cat. Hydrotreating
Alkyl & Polymer.
Aromatics/Isomeration
Lube Processing
Asphalt
Hydrogen
Coke
Engines and Flares
Engines
Flares
Total
Emission
Factor
1.02
0.00798
0.01
4.37E-07
580
0.30
0.30
0.50
0.43
0.60
0.60
0.18
0.54
1.05
0.15
0.0
60
0.0
0
0.24
0.0008
Methane
Emissions Units
MMscf CH4/heater/yr
Ib VOC/bbI
ib VOC/bbi
ton CH4/bbi
#HC/1 000 bbc capacity
IbTHC/IOOObbl
IbTHC/IOOObb!
!bTHC/1000bb!
IbTHC/IOOObbl
IbTHC/IOOObbl
Ib THC/1 000 bbl
IbTHC/IOOObbl
IbTHC/IOOObbl
IbTHC/IOOObbl
tb THC/1 000 bbi
# HC/ton
Included In Thermal Ops
scf CH4/hp-hr
Ib VOC/bbI
Confidence
Interval
100%
100%
1 00%
100%
100%
100%
100%
100%
100%
100%
100%
100%
100%
100%
100%
100%
100%
100%
100%
5%
100%
% Methane
In THC*
1.0%
1.0%
1.0%
51.0%
51.0%
51.0%
51.0%
51.0%
51.0%
51.0%
51.0%
51.0%
51.0%
51.0%
1.0%
Activity
Factor
3,200
13,400,900
13,400,900
13,400,900
13,400,900
13,400,900
6,143,244
1,698,828
4,902,489
3,385,314
1,070,329
2,076,594
6,255,880
977,726
676,909
207,013
655,151
20,334
13,400,900
Activity
Units
heaters
b/d
b/d
b/d
b/d
b/d
b/d
b/d
b/d
bid
b/d
b/d
b/d
b/d
b/d
b/d
b/d
MMhp-hr
b/d
Confidence
Interval
50%
5%
5%
5%
5%
5%
5%
5%
5%
5%
5%
5%
5%
5%
5%
5%
5%
100%
5%
Emissions
(Bscf)
3.26
0.009
0.012
0.101
0.673
0.018
0.008
0.004
0.009
0.009
0.003
0.002
0.015
0.005
0.000
0.000
0.174
0.000
4. 880
0.001
9.183
Confidence
Interval
122%
100%
100%
1 00%
100%
100%
100%
1 00%
100%
1 00%
100%
100%
100%
100%
100%
100%
100%
1 00%
100%
69.2%
' % Methane in VOC (volatile organic compounds) taken from AP-42 (Reference 28)
% Methane in HC (hydrocarbons) for system blowdowns is taken from AP-42
% Methane in THC (total hydrocarbons) calculated based on data from AP-42
% Methane in HC for asphalt calculated based on data from AP-42
-------
Table H-13
1990 Methane Emission Estimate
Petroleum - Production
Emission Source
Annual Production
% Heavy Crude (APl<20°)
Total Producing Oil Wells
% Heavy Weils (AP!<20°)
Fugitives:
Offshore Platforms
Gulf of Mexico
Rest of US
Oil Wellheads (heavy crude)
Oil Wellheads (light crude)
Separators (heavy crude)
Separators (light crude)
Heater/Trealers (light crude)
Headers (heavy crude)
Headers (light crude)
Tanks (light crude)
Compressors (light crude)
Small
Large
Sales Areas
Pipelines
Venting:
Oil Tanks
Pneumatic Devices
CIPs
Vessel Slowdowns
Compressor Starts
Compressor Slowdowns
Completion Flaring
Well Workover
Casinghead Gas
Upsets:
ESD
PRV Lifts
Well Blowout
Combustion Sources:
Gas Engines
Burners
Drilling
Flares
Total
Emission
Factor
2914
1178
0,83
19.58
0.85
51,33
59.74
0.59
202.78
34,4
46.14
16,360
40.55
56.4
12.1
345
248
78
8443
3774
733
96
256,838
34
250,000
0.24
0,526
0,052
Methane
Emissions Units
scfd CH4/platform
scfd CH4/piatform
scfd CH4/well
scfd CH4/well
scfd CH4/sep
scfd CH4/sep
scfd CH4/heater
scfd CH4/header
scfd CH4/header
scfd CH4/tank
scfd CH4/comp
scfd CH4/comp
scfd CH4/area
scfd CH4/mile
scf CH4/bbI
scfd CH4/device
scfd CH4/pump
scfyCH4/vessel
scfy CH4/comp.
scfy CH4/comp,
scfd CH4/completion
scf CH4/workover
scfy CH4/plat
scfy CH4/PRV
scf CH4/blowout
scf CH4/HPhr
lbCH4/1000gal
ton CH4/well drilled
Confidence
Interval
27%
36%
30%
30%
30%
30%
30%
30%
30%
30%
100%
100%
30%
97%
88%
40%
83%
266%
157%
147%
200%
200%
200%
252%
200%
5%
10%
100%
Activity
Factor
7,355,000
1 1 .0%
602,439
7.2%
1,092
22
43,556
558,883
9,687
116,936
80,105
16,185
48,699
55,888
670
2,009
4,443
70,000
7,355,000
121,299
129,042
213,487
2,906
2,906
620
45,183
1,114
438,597
2,85
18,306
8,773,000
620
Activity
Units
bbl/d
wells
platforms
platforms
wells
wells
separators
separators
heater treaters
headers
headers
tanks
small g.l. comp.
large g.l, comp.
sales areas
miles
bbl/d
pneumatics
CIPs
sep. and h.t.
gas lift comp.
gas lift comp.
completions
w.o./year
platforms
PRV
blowouts/yr
MMhp-hr
bbl/year
expl. wells
Confidence
Interval
5%
100%
5%
100%
10%
10%
100%
100%
78%
78%
130%
109%
109%
109%
91%
91%
46%
50%
5%
78%
105%.
70%
81%
81%
10%
421%
10%
103%
200%
277%
5%
10%
Emissions
(Bscf)
1.161
0.009
0.013
3.994
0.003
2.191
1.747
0.003
3.604
0.702
0,011
11.999
0.066
1.441
32.483
15.275
11.681
0.017
0.025
0.011
0.166
0,004
0.286
0.015
0.001
4.393
0.005
0.002
91.31
Confidence
Interval
29%
38%
109%
1 09%
87%
87%
139%
118%
118%
118%
163%
163%
57%
119%
88%
93%
160%
333%
218%
206%
201%
962%
201%
375%
490%
277%
11%
101%
48.4%
-------
Table H-14
1990 Methane Emission Estimate
Petroleum - Crude Transportation
Emission Source
Fugitives:
Pump Stations
Pipelines
Metering
Venting:
Tanks
Loading
Truck
Marine
Rail Car
Maintenance:
Pump Stations
Combustion Sources:
Pump engine drivers
Heaters
Total
Emission
Factor
1.06
0,0
4.37E-07
1.02E-05
0.5
1 .02E-05
1.56
0.24
Methane
Emissions Units
Ib CH4/yr/mile
Ib CH4/bbl
ton CH4/bbl
ton CH4/bbI
IbCH4/1 000 gal crude
ton CH4/bbl
Ib CH4/y/station
scf CH4/HPhr
Confidence
Interval
100%
10%
100%
100%
100%
100%
100%
5%
Activity
Factor
55,504
6.56E+09
8767000000
6.47E+07
8.91 E+10
2.01 E+07
555
Activity
Units
miles
bbl/yr
bbl/yr
bbl/yr
gal/yr
bbl/yr
stations
Confidence
Interval
100%
5%
4%
10%
10%
10%
100%
Emissions
(Bscf)
0.0014
0.0000
0.182
0.031
1.056
0.010
0.000
1.280
Confidence
Interval
173%
11%
100%
101%
101%
101%
173%
84.6%
-------
TableH-15
1990 Methane Emission Estimate
Petroleum - Refining
Emission Source
Fugitives:
Fuel Gas System
Pipe Stills
Wastewater Treating
Cooling Towers
Venting:
Tanks
System Slowdowns
Process Vents
Upsets
PRVs
Combustion Sources:
Process Heaters:
Aim. Distillation
Vacuum Distil,
Thermal Operations
Gal. Cracking
Cat. Reforming
Cat. Hydrocraking
Cat. Hydroretining
Cat. Hydrotreating
Alkyl & Polymer.
Aromatics/Isomeration
Lube Processing
Asphalt
Hydrogen
Coke
Engines and Flares
Engines
Flares
Total
Emission
Factor
1.02
0.00798
0,01
4.37E-07
580
0.30
0.30
0.50
0.43
0,60
0.60
0.18
0.54
1.05
0.15
0.0
60
0,0
0
0.24
0.0008
Methane
Emissions Units
MMscf CH4/heater/yr
Ib VOC/bbi
Ib VOC/bbi
ton CH4/bbI
# HC/1000 bbc capacity
ibTHC/IOOObbl
IbTHCAIOOObbl
IbTHC/1000bb!
IbTHC/IOOObb!
IbTHC/IOOObbi
IbTHCAIOOObbl
IbTHC/IOOObbl
IbTHC/IOOObbl
IbTHC/IOOObbl
IbTHC/IOOObbl
# HC/ton
Included in Thermal Ops
scf CH4/hp-hr
Ib VOC/bbi
Confidence
Interval
100%
1 00%
100%
100%
100%
100%
100%
100%
100%
100%
100%
100%
100%
100%
100%
100%
100%
1 00%
100%
5%
100%
% Methane
in THC*
1 .0%
1,0%
1,0%
51.0%
51.0%
51 .0%
51,0%
51 .0%
51.0%
51.0%
51,0%
51,0%
51.0%
51.0%
1 .0%
Activity
Factor
3,200
13,409,414
13,409,414
13,409,414
13,409,414
13,409,414
6,121,692
1 ,747,240
4,945,330
3,379,887
1,092,222
2,119,749
6,298,044
1,010,171
724,526
194,649
648,607
20,334
13,409,414
Activity
Units
heaters
b/d
b/d
b/d
b/d
b/d
b/d
b/d
b/d
b/d
b/d
b/d
b/d
b/d
b/d
b/d
b/d
MMhp-hr
b/d
Confidence
Interval
50%
5%
5%
5%
5%
5%
5%
5%
5%
5%
5%
5%
5%
5%
5%
5%
5%
100%
5%
Emissions
(Bscf)
3.26
0,009
0.012
0.101
0.673
0,018
0.008
0.004
0.009
0.009
0.003
0.002
0,015
0.005
0.000
0.000
0.172
0.000
4.8BO
0.001
9.181
Confidence
Interval
122%
100%
100%
100%
100%
100%
100%
1 00%
100%
100%
100%
100%
100%
100%
1 00%
1 00%
100%
100%
100%
69.2%
% Methane in VOC (volatile organic compounds) taken from AP-42 (Reference 28)
% Methane in HC (hydrocarbons) for system blowdowns is taken from AP-42
% Methane in THC (total hydrocarbons) calculated based on data from AP-42
% Methane in HC for asphalt calculated based on data from AP-42
-------
TableH-16
1991 Methane Emission Estimate
Petroleum - Production
Emission Source
Annual Production
% Heavy Crude (API<20°)
Total Producing Oil Wells
% Heavy Wells (APi<20°)
Fugitives;
Offshore Platforms
Gulf of Mexico
Rest of US
Oil Wellheads (heavy crude)
OH Wellheads (light crude)
Separators (heavy crude)
Separators (light crude)
Heater/Treaters (light crude)
Headers (heavy crude)
Headers (light crude)
Tanks (light crude)
Compressors (light crude)
Small
Large
Sales Areas
Pipelines
Venting:
Oil Tanks
Pneumatic Devices
CIPs
Vessel Slowdowns
Compressor Starts
Compressor Slowdowns
Completion Flaring
Well Workover
Casinghead Gas
Upsets:
ESD
PRV Lifts
Wei! Blowout
Combustion Sources:
Gas Engines
Burners
Drilling
Flares
Total
Emission
Factor
2914
117B
0.83
19,58
0.85
51.33
59.74
0.59
202.78
34.4
46.14
16,360
40.55
56.4
12.1
345
248
78
8443
3774
733
96
256,888
34
250,000
0.24
0.526
0.052
Methane
Emissions Units
scfd CH4/p!atform
scfd CH4/pIatform
scfd CH4/weII
scfd CH4/well
scfd CH4/sep
scfd CH4/sep
scfd CH4/heater
scfd CH4/header
scfd CH4/header
scfd CH4/tank
scfd CH4/comp
scfd CH4/comp
scfd CH4/area
scfd CH4/mi!e
scf CH4/bbl
scfd CH4/device
scfd CH4/pump
scfy CH4/vessel
scfy CH4/ccmp.
scfy CH4fcomp,
scfd CH4/completion
scf CH4A/vorkover
scfy CH4/plat
scfy CH4/PRV
scf CH4/blowout
scf CH4/HPhr
IbUWIQOOgal
ton CH4Avell drilled
Confidence
Interval
27%
36%
30%
30%
30%
30%
30%
30%
30%
30%
100%
100%
30%
97%
88%
40%
83%
266%
157%
147%
200%
200%
200%
252%
200%
5%
10%
100%
Activity
Factor
7,417,000
1 0.8%
613,810
7.2%
1,092
22
44,072
569,738
9,776
119,088
31,554
16,376
49,645
56,974
682
2,046
4,443
70,000
7,417,000
123,438
131,478
217,234
2,956
2,956
550
46,036
1,114
446,291
2.B5
18,622
6,715,000
550
Activity
Units
bbl/d
wells
platforms
platforms
wells
wells
separators
separators
heater treaters
headers
headers
tanks
small g.l, comp.
large g.l. comp.
sales areas
miles
bbl/d
pneumatics
CIPs
sep. and h.t.
gas lift comp.
gas lift comp.
completions
w.o/year
platforms
PRV
blowouts/yr
MMhp-hr
bbl/year
expl. wells
Confidence
Interval
5%
100%
5%
100%
10%
10%
100%
100%
78%
78%
130%
109%
109%
1 09%
91%
91%
46%
50%
5%
78%
1 05%
70%
81%
81%
10%
421%
10%
103%
200%
277%
5%
10%
Emissions
(Bscf)
1.161
0.009
0.013
4.072
0.003
2.231
1.778
0.004
3.674
0.715
0.011
12.216
0.066
1.441
32.757
15.544
11.901
0.017
0.025
0.011
0.147
0.004
0.286
0.015
0.001
4.469
0.004
0.001
92.58
Confidence
Interval
29%
38%
109%
109%
87%
87%
139%
118%
118%
118%
163%
163%
57%
1 1 9%
88%
93%
160%
333%
218%
206%
201%
962%
201 %
375%
490%
277%
11%
101%
48.4%
ffi
-------
TableH-17
1991 Methane Emission Estimate
Petroleum - Crude Transportation
Emission Source
Fugitives:
Pump Stations
Pipelines
Metering
Venting:
Tanks
Loading
Truck
Marine
Rail Car
Maintenance:
Pump Stations
Combustion Sources:
Pump engine drivers
Heaters
53 jTotal
: 1
00 * -, r- --- -----
Emission
Factor
1,06
0.0
4.37E-07
1.02E-05
0.5
1.02E-05
1.56
0.24
Methane
Emissions Units
Ib CH4/yr/mile
ib CH4/bb!
ton CH4/bbl
ton CH4/bbl
lbCH4/1 000 gal crude
ton CH4/bbl
Ib CH4/y/station
scf CH4/HPhr
Confidence
Interval
100%
10%
100%
100%
100%
100%
100%
5%
Activity
Factor
59,034
6.69E+09
8914000000
6.72E+07
9.00E+10
1.91E+07
590
Activity
Units
miles
bb!/yr
bbl/yr
bbl/yr
gal/yr
bbl/yr
stations
Confidence
Interval
100%
5%
4%
10%
10%
10%
100%
Emissions
(Bscf)
0.0015
0.0000
0.185
0.033
1.067
0.009
0.000
1.296
Confidence
Interval
173%
11%
100%
101%
101%
101%
173%
84.4%
-------
TableH-18
1991 Methane Emission Estimate
Petroleum - Refining
ffi
1
SO
Emission Source
Fugitives:
Fuel Gas System
Pipe Slllls
Wastewaler Treating
Cooling Towers
Venting:
Tanks
System Slowdowns
Process Vents
Upsets
PRVs
Combustion Sources:
Process Heaters:
Atm. Distillation
Vacuum Distil.
Thermal Operations
Cat. Cracking
Cat. Reforming
Cat. Hydrocraklng
Cat. Hydrorefining
Cat. Hydrotreating
Alky! & Polymer.
Aromatics/Isomeralian
Lube Processing
Asphalt
Hydrogen
Coke
Engines and Flares
Engines
Flares
Total
Emission
Factor
1.02
0.00798
0.01
4.37E-07
580
0.30
0.30
0.50
0.43
0.60
0.60
0.18
0.54
1.05
0.15
0.0
60
0.0
0
0.24
0.0008
Methane
Emissions Units
MMscf CH4/heater/yr
Ib VOC/bbl
Ib VOC/bbl
ton CH4/bbl
#HC/1 000 bbc capacity
ibTHC/1000bbl
IbTHC/1000bbl
IbTHC/1000bbl
lbTHC/1000bbl
!bTHC/1000bbl
IbTHC/IOOObb!
lbTHC/1000bb!
lbTHC/1000bbl
IbTHC/IOOObbl
IbTHC/IOOObbl
# HC/ton
included in Thermal Ops
scf CH4/hp-hr
Ib VOC/bbl
Confidence
Interval
100%
100%
1 00%
100%
100%
100%
100%
100%
100%
100%
100%
100%
100%
100%
100%
100%
100%
100%
100%
5%
100%
% Methane
in THC*
1 .0%
1 .0%
1 .0%
51.0%
51.0%
51 .0%
51.0%
51.0%
51.0%
51 .0%
51.0%
51 .0%
51 .0%
51.0%
1.0%
Activity
Factor
3,200
12,486,545
12,486,545
12,486,545
12,486,545
12,486,545
5,748,91 4
1,644,146
4,484,353
3,208,753
1 ,071 ,567
1 ,985,098
5,986,416
971,827
690,707
176,456
627,123
20,334
12,486,545
Activity
Units
heaters
b/d
b/d
b/d
b/d
b/d
b/d
b/d
b/d
b/d
b/d
b/d
b/d
b/d
b/d
b/d
b/d
MMhp-hr
b/d
Confidence
Interval
50%
5%
5%
5%
5%
5%
5%
5%
5%
5%
5%
5%
5%
5%
5%
5%
5%
100%
5%
Emissions
(Bscf)
3.26
0.009
0.011
0,094
0.627
0.017
0.008
0.004
0.008
0.008
0.003
0.002
0.014
0.005
0.000
0.000
0.166
0,000
4.880
0.001
9.117
Confidence
Interval
122%
100%
100%
1 00%
100%
100%
100%
1 00%
100%
100%
100%
100%
100%
100%
100%
100%
100%
100%
100%
69.7%
" % Methane in VOC (volatile organic compounds) taken from AP-42 (Reference 28)
% Methane in HC (hydrocarbons) for system blowdowns is taken from AP-42
% Methane in THC (total hydrocarbons) calculated based on data from AP-42
% Methane in HC for asphalt calculated based on data from AP-42
-------
Table H-19
1992 Methane Emission Estimate
Petroleum - Production
Emission Source
Annual Production
% Heavy Crude (APl<20°)
Total Producing Oil Wells
% Heavy Wells {API<20<1)
Fugitives:
Offshore Platforms
Gulf of Mexico
Rest of US
Oil Wellheads (heavy crude)
Oil Wellheads (light crude)
Separators (heavy crude)
Separators (light crude)
Heater/Treaters (light crude)
Headers (heavy crude)
Headers (light crude)
Tanks (light crude)
Compressors (light crude)
Small
Large
Sales Areas
Pipelines
Venting:
Oil Tanks
Pneumatic Devices
CIPs
Vessel Slowdowns
Compressor Starts
Compressor Slowdowns
Completion Flaring
Well Workover
Casinghead Gas
Upsets:
ESD
PRVUftS
Well Blowout
Combustion Sources:
Gas Engines
Burners
Drilling
Flares
Total
Emission
Factor
2914
1178
0,83
19.58
0.85
51.33
59.74
0.59
202.78
34.4
46.14
16,360
40.55
56.4
12.1
345
248
78
8443
3774
733
96
256,888
34
250,000
0.24
0.526
0.052
Methane
Emissions Units
scfd CH4/platform
scfd CH4/platform
scfd CH4/weII
scfd CH4/well
scfd CH4/sep
scfd CH4/sep
scfd CH4/heater
scfd CH4/header
scfd CH4/header
scfd CH4/tank
scfd CH4/comp
scfd CH4/comp
scfd CH4/area
scfd CH4/mfie
scf CH4/bbl
scfd CH4/device
scfd CH4/pump
scfy CH4/vessel
scfy CH4/comp.
scfy CH4/comp.
scfd CH4/completlon
scf CH4/workover
scfy CH4/piat
scfy CH4/PRV
scf CH4/blowout
scf CH4/HPhr
Ib CH4/1000 gal
ton CH4/well drilled
Confidence
Interval
27%
36%
30%
30%
30%
30%
30%
30%
30%
30%
100%
100%
30%
97%
88%
40%
83%
266%
157%
147%
200%
200%
200%
252%
200%
5%
10%
100%
Activity
Factor
7,171,000
1 1 .9%
594,189
7.1%
1,092
22
42,247
551 ,942
9,533
115,195
78,850
15,698
48,095
55,194
659
1,978
4,443
70,000
7,171,000
119,475
127,275
210,257
2,861
2,861
450
44,564
1,114
431,957
2.85
18,023
4,718,000
450
Activity
Units
bbl/d
wells
platforms
platforms
wells
wells
separators
separators
heater treaters
headers
headers
tanks
small g.l. comp.
large g.l. comp.
sales areas
miles
bbl/d
pneumatics
CIPs
sep. and h.t.
gas lift comp.
gas lift comp.
completions
w.o./year
platforms
PRV
blowouts/yr
MMhp-hr
bbl/year
expl. wells
Confidence
Interval
5%
100%
5%
100%
10%
10%
100%
100%
78%
78%
130%
109%
109%
109%
91%
91%
46%
50%
5%
78%
105%
70%
81%
81%
10%
421%
10%
103%
200%
277%
5%
10%
Emissions
(Bscf)
1.161
0.009
0.013
3.945
0.003
2.158
1.719
0.003
3.560
0.693
0.011
11.810
0.066
1.441
31.671
15.045
11.521
0.016
0.024
0.011
0.120
0.004
0.286
0.015
0.001
4.326
0.002
0.001
89.64
Confidence
Interval
29%
38%
1 09%
1 09%
87%
87%
139%
1 1 8%
1 1 8%
1 1 8%
1 63%
163%
57%
119%
88%
93%
160%
333%
21 8%
206%
201%
962%
201%
376%
490%
277%
11%
101%
48.4%
ffi
to
o
-------
Table H-20
1992 Methane Emission Estimate
Petroleum - Crude Transportation
Emission Source
Fugitives:
Pump Stations
Pipelines
Metering
Venting;
Tanks
Loading
Truck
Marine
Rail Car
Maintenance:
Pump Stations
Combustion Sources:
Pump engine drivers
Heaters
Total
Emission
Factor
1.06
0.0
4.37E-07
1.02E-05
0,5
1.02E-05
1.56
0.24
Methane
Emissions Units
Ib CH4/yr/mile
Ib CH4/bfal
ton CH4/bb!
ton CH4/bbl
IbCH4/1 000 gal crude
ton CH4/bbl
Ib CH4/y/station
scf CH4/HPhr
Confidence
Interval
100%
10%
100%
100%
100%
100%
100%
5%
Activity
Factor
54,675
6.54E+09
8825000000
7.19E+07
9.26E+10
7.52E+06
547
Activity
Units
miles
bbl/yr
bbl/yr
bbl/yr
gal/yr
bbl/yr
stations
Confidence
Interval
100%
5%
4%
10%
10%
10%
100%
Emissions
(Bscf)
0,0014
0.0000
0.183
0.035
1.098
0.004
0.000
1.321
Confidence
Interval
173%
11%
100%
101%
101%
101%
173%
85.1%
-------
Table H-21
1992 Methane Emission Estimate
Petroleum - Refining
Emission Source
Fugitives:
Fuel Gas System
Pipe Stills
Wastewater Treating
Cooling Towers
Venting:
Tanks
System Slowdowns
Process Vents
Upsets
PRVs
Combustion Sources:
Process Heaters:
Atm. Distillation
Vacuum Distil.
Thermal Operations
Cat, Cracking
Cat. Reforming
Cat. Hydrocraking
Cat. Hydrorefining
Cat. Hydrotreating
Alkyl & Polymer.
Aromatics/lsomeration
Lube Processing
Asphalt
Hydrogen
Coke
Engines and Flares
Engines
Flares
Total
Emission
Factor
1.02
0.00798
0.01
4.37E-07
580
0.30
0.30
0.50
0.43
0.60
0.60
0.18
0.54
1.05
0.15
0.0
60
0.0
0
0.24
0.0008
Methane
Emissions Units
MMscf CH4/neater/yr
Ib VOC/bbl
Ib VOC/bbl
ton CH4/bbl
# HC/1000 bbc capacity
IbTHC/IOOObbl
IbTHC/IOOObbl
lbTHC/1000bfal
!b THC/1 000 bbl
Ib THC/1 000 bbl
Ib THC/1 000 bbi
IbTHC/IOOObbl
IbTHC/IOOObbl
IbTHC/IOOObbl
IbTHC/IOOObbl
# HC/ton
Included in Thermal Ops
scf CH4/hp-hr
Ib VOC/bbl
Confidence
Interval
100%
100%
100%
100%
100%
100%
100%
100%
100%
100%
100%
1 00%
100%
100%
100%
100%
1 00%
100%
100%
5%
100%
% Methane
in THC*
1.0%
1.0%
1.0%
51.0%
51.0%
51.0%
51.0%
51.0%
51.0%
51.0%
51.0%
51.0%
51.0%
51.0%
1 .0%
Activity
Factor
3,200
13,410,527
13,410,527
13,410,527
13,410,527
13,410,527
5,849,509
1,582,073
4,585,047
3,171,605
1,083,645
1,880,137
6,208,001
982,173
701,466
170,852
662,633
20,334
13,410,527
Activity
Units
heaters
b/d
fa/d
b/d
b/d
b/d
b/d
b/d
b/d
b/d
b/d
b/d
b/d
b/d
b/d
b/d
b/d
MMhp-hr
b/d
Confidence
Interval
50%
5%
5%
5%
5%
5%
5%
5%
5%
5%
5%
5%
5%
5%
5%
5%
5%
100%
5%
Emissions
(Bscf)
3.26
0.009
0.012
0.101
0.673
0.018
0.008
0.004
0.009
0.008
0.003
0.001
0.015
0.005
0.000
0.000
0.176
0.000
4.880
0.001
9.183
Confidence
Interval
122%
100%
100%
1 00%
100%
100%
100%
100%
1 00%
100%
100%
100%
100%
100%,
100%
100%
100%
100%
100%
69.2%
ffi
to
% Methane in VOC (volatile organic compounds) taken from AP-42 (Reference 28)
% Methane in HC (hydrocarbons) for system blowdowns is taken from AP-42
% Methane in THC (total hydrocarbons) calculated based on data from AP-42
% Methane in HC for asphalt calculated based on data from AP-42
-------
NIML-RTP-142
TECHNICAL REPORT DATA
(Please readIn tint ctions on the reverse before comp
I. REPORT NO,
EPA-600/R-99-010
7.
4, TITLE AND SUBTITLE
Methane Emissions from the U. S,
Industry
Petroleum
5. REPORT DATE
February 1999
6. PERFORMING ORGANIZATION CODE
7. AUTHOR(S)
Matthew R. Harrison, Theresa M. Shires, Richard
A. Baker, and Christopher J. Loughran
8. PERFORMING ORGANIZATION REPORT NO.
9. PERFORMING ORGANIZATION NAME AND ADDRESS
Radian International LLC
P. O. Box 201088
Austin, Texas 78720-1088
10. PROGRAM ELEMENT NO.
11, CONTRACT/GRANT NO.
68-D2-0160
12.SPONSORING AGENCY NAME AND ADDRESS
EPA, Office of Research and Development
Air Pollution Prevention and Control Division
Research Triangle Park, NC 277U
13. TYPE OF REPORT AND PERIOD COVERED
Final; 4/95 - 4/96
14. SPONSORING AGENCY CODE
EPA/600/13
is. SUPPLEMENTARY NOTES ^PPCD project officer is David A. Kirchgessner, Mail Drop 63,
919/541-4021.
A The report quantifies methane (CH4) emissions from the U. S. petroleum in-
dustry by identifying sources of CH4 from the production, transportation, and refi-
ning of oil. Emissions are reported for the base year 1993 and for the years 1986
through 1992, based on adjustments to the base year calculations. An extensive liter-
ature search identified 54 reports as having some potential applicability for estima-
ting CH4 emissions from the petroleum industry. Each report was reviewed and
subjectively ranked based on data quality. Only seven reports were useful for this
study. Methods for estimating emissions were developed when data gaps were iden-
tified. For the base year 1993, approximately 98 billion (10 to the 9th power) stan-
dard cubic feet (Bscf) +/- 44% of CH4 emissions are attributed to the petroleum in-
dustry. Standard error propagation techniques were used to determine the precision
of the estimate to a 90% confidence bound. (NOTE: As concentrations of greenhouse
gases increase in the atmosphere, their potential impact on global climate has be-
come important. Although greenhouse gases, such as carbon dioxide (CC2), CH4,
and nitrogen oxides (NOx), occur naturally in the atmosphere, recent attention has
focused on the increased emissions resulting from human activities. CH4 is the sec-
ond largest source (after C02) of anthropogenic greenhouse gas emissions.)
17.
KEY WORDS AND DOCUMENT ANALYSIS
DESCRIPTORS
b.lDENTIFIERS/OPEN ENDED TERMS
c. COSATI Field/Group
Pollution
Methane
Emission
Petroleum Industry
Greenhouse Effect
Pollution Control
Stationary Sources
13B
07C
14G
05C
04A
18. DISTRIBUTION STATEMENT
Release to Public
19. SECURITY CLASS (ThisReportJ
Unclassified
21. NO. OF PAGES
158
2O. SECURITY CLASS (Thispage}
Unclassified
22. PRICE
EPA Form 2220-1 (9-73)
------- |