oEPA
United States
Environmental Protection
Agency
Region IX
215 Fremont Street
San Francisco, CA 94105
EPA-909/9-81-003
September 1981
Air
Assessment of VOC
Emissions from Well Vents
Associated With Thermally
Enhanced Oil Recovery
-------
DCN 81-240-016-09-12
EPA 909/9-81-003
ASSESSMENT OF VOC EMISSIONS
FROM WELL VENTS
ASSOCIATED WITH
THERMALLY ENHANCED OIL RECOVERY
. FINAL REPORT
EPA Contract No. 68-02^-3513
Work Assignment No. 9
Prepared by:
G.E. Harris, K.W. Lee, S.M. Dennis, C.D. Anderson, and D.L. Lewis
Radian Corporation
8501 Mo-Pac Blvd.
Austin, Texas 78759
Prepared for:
Tom Rarick
U.S. EPA Region IX
215 Fremont St.
San Francisco, CA. 94806
13 September 1981
-------
CONTENTS
Page
1.0 Introduction and Background 1
2.0 Summary of Results •• 3
2.1 TEOR Well Population Data 3
2.2 Emission Factors • 3
2.3 Well Characteristics Survey 3
2.4 Correlation/Studies 9
3.0 Description of Sources 10
3.1 Enhanced Oil Recovery 10
3.2 Wellhead Design H
3.3 Steam Drive Wells 13
3.4 Cyclic Steaia Wells 14
4.0 Experimental Design 16
5.0 Sampling Methodology 21
5.1 Survey Procedures 21
5.2 Quantitative Sampling Procedures 23
5.2.1 Sampling systems for low, medium and high flow
wells « 23
5.2.2 Sampling procedures 27
6.0 Analytical Methodology 31
6.1 Noncondensible Gas Analysis 31
6.1.1 Fixed gases 34
6.1.2 Hydrocarbon species 35
6.2 Analysis of Collected Liquids 35
6.3 Boiling Point Distribution 36
7.0 Quality Assurance and Quality Control 37
7.1 Systems Audit Results 37
7.2 Performance Audit Results 41
7.2.1 Density 4i
7.2.2 Noncondensible gas analysis 42
-------
CONTENTS (Continued)
7.3 Analytical Precision 45
7.3.1 Volumetric gas flow rate 45
7.3.2 Condensible hydrocarbon emissions 46
7.3.3 Fixed gases 47
7.3.4 Noncondensible hydrocarbon species 49
7.3.5 Density 51
7.4 Equipment Calibration 52
7.5 Data Capture 52
7.6 Data Validation 52
8.0 Detailed Results 56
9.0 Correlation Studies 91
9.1 Correlations Between Survey Parameters 91
9.2 Correlation of VOC Emissions 91
9.3 Regression Analysis on Tested Data 100
10.0 Emission Factor Development 105
10.1 Steam Drive Well Emission Factor 105
10.2 Steam Cycle Well Emission Factor 106
Appendix A A-l
Appendix B B-l
Appendix C C-l
ii
-------
Number
3-1
5-1
5-2
5-3
6-1
8-1
9-1
9-2
9-3
9-4
9-5
9-6
9_-7
FIGURES
Typical production wellhead
Moderate flow sampling train
Diagram of instruments in mobile laboratory
VOC emissions vs. time since last steaming
VOC emissions vs . number of cycles
VOC emissions vs. steam dosage
VOC emissions vs. oil production rate
VOC emissions vs. API gravity of the oil
Page
12
25
26
29
33
90
93
94
95
9J5
9J
9-8
99
111
-------
TABLES
Number Page
2-1 Kern County Producer Survey Summary 4
2-2 Emission Factors 7
2-3 Summary of Cyclic Well Characteristics Data... 8
4-1 Summary of Testing 19
4-2 Sampling Quotas 20
5-1 Selection of Sampling Systems 24
5-2 Major Steps in the Sampling Procedures 28
6-1 Methods for Gas Phase Analysis 32
7-1 Estimated Precision and Accuracy of Test Data 38
7-2 Performance Audit Results for Density Determinations 42
7-3 Performance Audit Results, Noncondensible Gases 43
7-4 Volumetric Gas Flow Rate Variability 46
7-5 Condensible Hydrocarbon Emissions Variability 46
7-6 Summary of Precision for Fixed Gas Analyses 48
7-7 Analytical Variability of Hydrocarbon Samples Analyses 49
7-8 Summary of Precision for Hydrocarbon QC Standard Analyses 50
7-9 Analytical Variability of Density Determination 51
8-1 Survey Results by Field 57
8-2 Breakdown of Non-Blowers by Field 58
8-3 Survey Results by -Producer 59
8-4 Sampling Distribution by Field 60
-------
TABLES (Continued)
Number page
8-5 Sampling Distribution by Producer 61
8-6 Sampling Results by Field 62
8-7 Well Characterization Survey Results 71
8-8 Listing of Emission and Characertization Data for Individual
Cyclic Wells 73
9-1 Correlation Coefficients -for the Survey Data 92
9-2 Correlation Coefficients for Data on Wells Tested 101
9-3 Results of Multiple Regression Analysis on Log of VOC
Emissions , 102
9-4 Results of Multiple Regression Analysis on Log of VOC
Emissions for Western area of Kern County 104
10-1 Summary of Vapor Recovery System Source Tests Used in the
Steam Drive Well Emission Factor 107
10-2 Emission Factors and Variance Data for Steam Cycle Wells ... Ill
v
-------
ACKNOWLEDGMENTS
Radian wishes to acknowledge the assistance provided by members of the
Technical Advisory Committee whose expertise helped guide this study to a
successful completion:
Tom Rarick U.S. EPA Region IX
Harry Metzger California Air Resources Board
Frances Cameron California Air Resources Board
Dean Simeroth California Air Resources Board
Grant Chin California Air Resources Board
Larry Landis Kern County Air Pollution Control District
Stan Bell Tenneco Oil
Sam Duran Getty Oil
Les Clark Independent Oil Producers Association
David Farr Chevron
Alex Nichols Santa Fe Energy
Alan Schuyler ARCO Oil and Gas
Craig Jackson Getty Oil
A special acknowledgment is also due to Getty Oil for allowing the
use of their portable fin fan condenser during the study.
vi
-------
SECTION 1
INTRODUCTION AND BACKGROUND
This document presents the results of a study of VOC (Volatile Organic
Compounds)* emissions from wellhead casing vents associated with thermally
enhanced oil recovery (TEOR) operations. The effort included a survey of
existing source test and well population data, as well as a sampling and
analysis program to measure emissions from uncontrolled cyclic well vents.
These data were used to develop emission factors for both cyclic and steam
drive production wells. This report also includes the results of the surveys
and attempted correlations between well vent emissions and the characteristics
of the well.
The objective of this program is to develop data to refine the estimates
of total VOC emissions attributable to TEOR wellhead casing vents. The state
of California is in the process of reviewing its emission inventories for those
air pollution control districts (APCD's) which have not yet demonstrated attain-
ment of the National Ambient Air Quality Standards CNAAQS) for oxidants. In
several APCD's, the VOC emissions from TEOR operations account for a large
portion of the total VOC emissions in the district. It is necessary, therefore,
to refine the estimates of VOC emissions from TEOR well vents in order to
accurately assess the need for future control.
This study was funded and administered by EPA Region IX. Additional
technical input was received from a Technical Advisory Committee composed of
representatives from the EPA, the California Air Resources Board (GARB), the
Kern County APCD, and the oil industry. The committee met five times during
the course of the program. A project kickoff meeting was held to discuss the
overall objectives and approach to the study. Another meeting was held to
* VOC is defined for this study as total non-methane, non-ethane organic
material.
-------
review the test plan before starting field sampling. A third meeting was
called to discuss a problem encountered in the early testing concerning the
distinction between a steam drive and a cyclic steam well. Another meeting
was held to present the preliminary results shortly after completeing the
field testing phase. The final meeting was held to review the draft final
report.
The results of the testing and surveys are summarized in Section 2.
Section 3 presents a brief discussion of TEOR operations to aid the reader
who is unfamiliar with this type of oil production. Sections 4, 5 and 6 present
the details of the experimental design and the sampling and analytical techniques
used in testing cyclic wells. Section 7 presents a discussion of quality
control for the test program. Section 8 presents the detailed results of
emissions testing and survey data, while that information is used to test for
various correlations in Section 9. Section 10 documents the methodology for
calculating emission factors for both cyclic and drive wells. The appendices
include example data sheets and sample calculations.
-------
SECTION 2
SUMMARY OF RESULTS
The objective of this study is to estimate the VOC emissions from
wellhead casing vents on TEOR projects. This section briefly summarizes all
of the results pertinent to that objective.
2.1 TEOR WELL POPULATION DATA
A survey was made of several sources of population data for both steam
drive and steam cycle wells. The most comprehensive and accurate source of
well population data was found to be a survey made by the Kern County Air
Pollution Control District, the results of which are presented in Table 2-1.
For sources outside Kern County, the population data on file with the
Division of Oil and Gas can be used, but it does have some inconsistencies
in the classification of wells as either drive or cyclic.
2.2 EMISSION FACTORS
Emission factors were calculated for steam drive wells based on compliance
testing of vapor recovery systems. A sampling and analysis program, which
included a survey of 358 wells and quantitative testing of 58 wells, provided
the data to develop an emission factor for cyclic wells. The emission factor
data is summarized in Table 2-2.
2.3 WELL CHARACTERISTICS SURVEY
A questionnaire was completed by producers providing data to characterize
the operations and physical characteristics of each steam cycle well surveyed.
The results of that survey are presented in Table 2-3.
-------
TABLE 2-1. KERN COUNTY PRODUCER SURVEY SUMMARY*
Producer
Arco
Arco
Bell Western
Berry Holding
Berry Holding
Carrec Oil
Chevron USA.
Chevron USA
Chevron USA
Chevron USA
Chevron USA
Chevron USA
Chevron USA
Circle Oil
ELf Oil &Gas
finjsyco
Energy Dev.
Exeter
Exxcn
General Oil
Getty Oil
Getty Oil
Getty Oil
Getty Oil
Getty Oil
Getty Oil
Getty Oil
Gulf
Gulf
Gulf
Gulf
Gulf
Oil Field
Midway-Sunset
Kern Front
Edison
Midway-Sunset
So. Belridge
Kern Front
Cymric
Midway-Sunset
McKittrick
Bel-ridge
Kern River
Poso Creek
Edison/Racetrak
McKittrick
Poso Creek
•pAism
Kern Bluff
Midway-Sunset
Edison
Midway-Sunset
Midway-Sunset
Lost Hills
Cymric
JfcRLttrick
Kern Front
Poso Creek
Kern River
Midway-Sunset
Cymric
Kem Bluff
Lost Hills
Fruitvale
Drive Wells Cyclic Wells
Total-Controlled Total-Controlled
52-0
2-2
0
0
0
0
62-56
90-90
64-64
0
499 - 493
19-19
12-12
22-22
0
0
d
0
0
0
84-84
27 - 27
0
0
0
0
2109 - 2109
0
0
4-4
74 - 74
0
195
42
12
588
21
175
440
140
28
187
78
53
3
10
50
158
36
219
40
70
603
97
76
875
32
16
20
62
26
- 0
- 42
- 0
- 139
0
- 0
- 0
- 0
- 0
- 0
- 38
- 78
- 0
0
0
- 0
- 0
- 0
- 26
- 16
- 0
- 0
- 0
- 0
- 0
- 0
- 638
- 28
- 0
- 18
- 61
- 26
Composite of responses to a questionnaire sent to the producers by a letter
from Leon M. Hebertson, Air Pollution Control Officer, Kern County APCD,
September 12, 1980.
Continued/
-------
TABLE 2-1. (Continued)
Producer
Junniper
Johnson & Broun
Kem Ridge
McCullockOil
McFarland
McFarland
Mobil
Mobil
Mobil
Mobil
Orri dpr1*"-^
Occidental
Petro-Lewis
Petro-Lewis
Petro-Lewis
Santa Fe
Santa Fe
Santa Fe
Santa Fe
Shell
Shell
Shell
Sun Production
Sun Production
Tarmehill Oil
Tenneco Oil
Tenneco Oil
Tenneco Oil
Tenneco Oil
Oil Field
Jasmin
Cymric
So. Belridge
Midway-Sunset
Midway- Sunset
McKittrick
Kern Front
Midway-Sunset
Cymric
Belridge
Midway-Sunset
McKittrick
Poso Creek
Kem Front
Kern River
Midway-Sunset
Kem River
Kern Front
Poso Creek
Midway-Sunset
Mt Poso
Kem River
Kem River
Midway-Sunset
Midway-Sunset
Kem River
Midway-Sunset
Poso Creek
Wheeler Ridge
Drive Wells Cyclic Wells
Total-Controlled Total-Controlled
0
0
2115 - 604
0
0
0
0
0
0
111-39
0
0
0
36-0
27 - 27
163 - 95
41-0
0
0
239-0
257-0
0
3-0
35-0
0
189 - 0
152-0
0
0
13
66
140
44
4
40
330
45
157
52
5
52
82
26
983
136
14
376
608
29
254
147
36
103
0
- 0
- 0
- 0
- 0
- 0
- 0
- 0
- 0
- 0
- 0
- 5
- 37
- 0
- 0
- 84
- 0
- 0
0
- 0
0
- 0
- 16
- 0
- 147
- 0
- 0
0
0
Continued/
-------
TABLE 2-1. (Continued)
Producer
Texaco
Union
Union
Ubion
Victory
Victory
Vfcittier
Vtiittier
TOTALS:
Oil Field
Midway-Sunset
No. Belridge
IfcRLttrick
Midway-Sunset
Cymric
No. Midway
Drive Wells
Cyclic Wells
Total-Controlled Total-Controlled
8-0
12-12
17 - 17
39 - 39
0
0
No. Midway-Sunset 0
Kern Front
Kern Front
Kern River
Boso Creek
Edison
Midway-Sunset
Belridge
Kem Bluff
Lost Hills
Cymric
McKittrick
Fruitvale
Mt Poso
Jasmin
Wheeler Ridge
0
38-2
2868 - 2629
19-19
12-12
862 - 376
2238 - 655
4-4
101 - 101
62 - 56
108 - 39
0
257-0
0
0
6569-3893
59%
38-0
23-23
50-8
150-56
8-0
78-0
122-0
9-0
284-42
1761 - 692
128-115
226 - 26
4377 - 470
274 - 23
30-18
102 - 61
327-0
802-5
26-26
0
0
0
8337-1478
18%
-------
TABLE 2-2. EMISSION FACTORS
Source Type
Overall*
Cyclic
Steam v Western
Wells Kern
County
Central
Kern
County
Steam Drive Wells
VOC Emission 95% Confidence
Factors Interval (Ib/day/well) Emission Factor Basis
(Ib/day/well) Lower
3.6 2.2
4.3 2.3
2.3 0.7
220.3 209.3
Upper
6.2 358 wells surveyed
58 wells tested
271 wells surveyed
7.6 42 wells tested
87 wells surveyed
3.3 16 wells tested
231.3 40 vapor recovery system
tests
963 drive wells represented
* In deriving the overall estimates, average emissions in the cell were weighted by the
proportion between the west and the central areas as determined in the survey. The VOC emissions
of the wells actually tested were averaged within each flow rate group and each area group.
-------
TABLE 2-3. SUMMARY OF CYCLIC WELL CHARACTERISTICS DATA
For All Wells Surveyed
1.
2.
3.
4.
5.
6.
(7.
8.
9.
Parameter
Total Steaming Cycles to Date
Time Since Last Steaming
Steaming Frequency
Soaking Period
Steam Dosage
Oil Production Rate
Cumulative Oil Production
Since Steaming Began
Gravity of the Oil
Water to Oil Ratio
Units
days
mos. /cycle
days
bbl. /cycle
bbl./day
bbl.
"API
___
Mean
Value
8.1
242
9.9
5.8 ,
9731
21.2
49,911
12.9
14.0
Range
Lower
1
1
1
0
640
0.4
302
10.5
0.01
Upper
29
1372
115
23
86,181
1280
320,311
18.9
99
I'umber of
Responses
317
317
228
230
335
302
208
308
292
For Only
Mean
Value
8.0
213
8.8
5.5
10,281
15.6
49,863
12.8
19.8
R
Lower
1
13
1
1
640
2
311
10.5
0.04
Wells Tested
ange
Upper
25
502
24
19
62,089
45
279,938
16.0
97
Numoer of
Responses
56
54
36
36
57
51
34
50
46
00
-------
2.4 CORRELATION STUDIES
An attempt was made to correlate the VOC emissions from cyclic wells to
their operating and physical characteristics. Although some vague trends
could be identified, there was too much scatter in the data to provide
significant correlations. The trends are strong enough to indicate that some
variables do correlate to emissions, but the study population is too small
to quantify the complex inter-relationships of the many variables involved.
-------
SECTION 3
DESCRIPTION OF SOURCES
In order to understand the experimental design and to interpret the results,
it is necessary to understand the fundamentals of oil production using thermally
enhanced oil recovery (TEOR). This section presents a brief discussion of
TEOR technology, especially as it affects the well vent emissions.
3.1 ENHANCED OIL RECOVERY
When an oil producing formation is first drilled, the formation pressure
may be high enough for the oil to flow freely to the surface. As such free
flowing production declines, it is necessary to use some mechanical aid to
induce the flow of oil to the surface. Typically, this is done by pumping
the liquid, but it can also be accomplished by gas lift or by artificially
pressuring the formation with compressed gas. All of these methods are still
considered to be primary production techniques.
As the oil production rate achievable with primary recovery methods drops
off, the producer may consider secondary oil recovery such as water-flooding.
TEOR is a tertiary recovery technique which may include in-situ combustion
Cfire-flooding) and steam stimulation. This report deals solely with the steam
stimulation type of TEOR activities. TEOR is particularly advantageous in the
production of very heavy oils where the high viscosity of the oil retards its
migration through the formation to the well. The injection of steam, on either
a continous or cyclic basis, raises the temperature in the producing zone and
lowers the viscosity of the oil, which increases the achievable production rate.
10
-------
3.2 WELLHEAD DESIGN
Oil production wellheads have essentially identical designs for both steam
cycle and steam drive wells. Figure 3-1 presents a typical design of a
production wellhead.
Crude oil production wells are typically completed in a pool or reservoir
with a 6 to 10 inch diameter pipe casing surrounded by cement. The casing and
cement are perforated at the desired depths of production. The crude then
flows into the casing through the perforations and is pumped to the wellhead by
a rod pump connected to the surface pumping unit by a string of rods. The
crude flows through the production tubing into the crude flowline which is
connected to either a main lease flowline or crude storage tank.
During normal production operation, the valve on the crude flowline is
open and the valve on the casing flowline closed. The casing vent may be open
or closed depending on the operational characteristics of the well. If a
negative pressure (relative to atmospheric pressure) develops within the _
casing due to geological properties or pumping practices, the casing vent
valve would be closed to increase the flow of crude through the perforations
into the casing. A high pressure in the well casing would inhibit the flow
of crude into the casing, and the casing vent valve would be opened to relieve
that pressure. With atmospheric pressure in the casing, the casing vent valve
might be open or closed depending on the well operator.
The primary emission point for both steam drive and steam cycle wells is
the casing vent. The occurrence and amount of emissions may vary significantly
between steam cycle and steam drive wells due to differences in steaming
practices. The following two subsections discuss these differences and their
impact on emissions.
11
-------
I J
TO LEASE
FLOWLINE
CRUDE
FLOWLINE
PUMP RODS
PRODUCTION
TUBING
WELL CASING
CASING VENT
70A2210
FIGURE 3-1 TYPICAL PRODUCTION WELLHEAD
-------
3.3 STEAM DRIVE WELLS
Both steam drive wells and steam cycle wells are stimulated by the
injection of steam into the producing formation. In a cyclic operation, the
steam is intermittently injected into the production well itself. In a steam
drive operation, the steam is continuously injected into one well that is
dedicated to that service and oil and connate wa-ter is produced from wells
clustered around the injection well.
It is not always straightforward to distinguish between cyclic and drive
wells. A drive well may occasionally have steam injected directly into the
production tubing, both to clean the tubing and to stimulate production. A
cyclic well may also be indirectly affected by nearby steam injection wells.
The Kern County APCD defines a drive well as a production well which is
completed in the same zone and is within 250 feet of a steam injection well.*
A steam cycle well can then be defined as any well which is intermittently
steamed and produced and does not meet the requirements to be called a drive
well.
Steam drive wells are typically situated in groups or patterns surrounding
a steam injection well. Steam is continuously injected at high pressure into
an injection well which resembles a typical producing well without the pumping
apparatus. During the process of injection, a series of zones develop as the
fluids move from injection well to production well. Nearest the injection
well is a steam zone, followed by a zone of steam condensate, and in front of
the condensate is a region of reduced-viscosity oil moving towards a production
well.
The steam drive, or production well, may also be injected with steam to
reduce the viscosity of the crude nearby. By warming the crude surrounding
a steam drive well completion, the zone of crude moving towards the well may
reach the completion more easily and quickly.
* This distance is based on a 2.5 acre steaming pattern.
13
-------
Several conditions may exist which could result in an emitting steam drive
well casing vent. A typical situation is when steam breakthrough occurs at the
production well. Due to differences in permeability, the steam zone may over-
take the condensate and reduced-viscosity crude zones near the production
well completion. With the casing vent open, this steam rises through the casing
and out the vent. Steam breakthrough usually results in high vent flowrates
for sustained intervals.
Another situation which may result in casing vent emissions is steam
"channeling" or short circuiting. In this case, steam from the injection
well bypasses the crude reservoir via a geological fault in the formation.
This steam would also rise through the casing and exit an open vent.
Emissions from steam drive well vents consist primarily of steam and
entrained water, but may also include carbon dioxide, hydrocarbons, and
hydrogen sulfide. Once steam drive wells begin to emit, they typically continue
to emit.
3.4 CYCLIC STEAM WELLS
As mentioned in the previous section, a cyclic steam well is a production
well that is intermittently steamed and is not affected by any nearby continuous
steam injection wells. The objective of the steaming is to heat the crude oil
in the reservoir surrounding the completion. This reduces the viscosity of the
oil and allows it to flow more freely into the production well. Some major TEOR
operations begin with cyclic steaming and convert to steam drive if the cyclic
steaming project is successful.
When a cyclic well is steamed, the pump rods and pump unit are usually
removed and the production tubing capped off. The crude flowline is then emptied,
casing vent closed, and, depending on individual steaming practices, the casing
flowline valve may be opened. High pressure steam from steam generators is
14
-------
then piped to the well through the crude flowline. The steam is typically injected
through the well tubing and/or casing into the crude reservoir for a period
of 5 to 15 days or until the total amount injected is between 5,000 and
15,000 barrels (as water). At this time the crude and casing flowline valves
are closed and the well is allowed to "soak".
During the soaking period, typically 4 to 10 days, the surrounding crude
becomes less viscous due to heat transfer from the injected steam. After the
reservoir temperature has equilibrated, the pump rod assembly is placed again
into the production tubing and production resumed. At this time, the casing
vent valve is opened. With the vent open, the pressure in the reservoir is
reduced, which causes hot water (condensed from injected high pressure steam)
to flash into steam (and some entrained water), which is emitted from the casing
vent. A crude and water mixture is then pumped to the wellhead. When crude
production has declined significantly the steaming process is repeated. Such
steaming cycles may range from 2 months up to 2 years or more.
Cyclic wells typically exhibit their highest casing vent flowrates immediately
after soaking. The majority of the vent flow is caused by steam condensate
flashing in the crude reservoir and is exemplified by a large steam plume.
Also potentially contained in the casing vent flow are hydrocarbons, carbon
dioxide, and hydrogen sulfide.
Depending on geological characteristics, cyclic wells will have higher
than normal casing vent flowrates for from 1 to 20 days after soaking has
ended. When the flow has decreased, the casing vent may be left open if
positive pressure still exists within the casing, or closed if a negative
pressure is present. If the casing vent is left open, the casing may continue
to emit with little or no plume. It should be emphasized that actual steaming
practices and emission characteristics vary widely depending on the field and
well operator.
15
-------
SECTION 4
EXPERIMENTAL DESIGN
It was determined early in the program that further steam drive well
testing was not warranted, and that the sampling and analysis effort should
concentrate on cyclic wells. The objective was set to quantitatively measure
the emissions of 50 randomly selected uncontrolled steam cycle wells which
were found to be emitting. It was recognized that many more cyclic wells
would need to be surveyed in order to find 50 emitting wells, since cyclic
wells do not always emit on a continous basis.
Data on cyclic well population was available from two sources:
the Division of Oil and Gas (DOG), and
the Kern County Air Pollution Control District (KCAPCD).
There were inconsistencies between these two data bases, largely because of
differing definitions of what constitutes a steam cycle well. The DOG considers
any well which is both steamed and produced in a single year to be a cyclic
well. This results in excluding some cyclic wells which are steamed less
frequently than once per year. It also includes some steam drive wells which
are lightly steamed to clean out the production tubing. The DOG data base
also gave no indication as to whether or not the wells were controlled by a
vapor recovery system.
It was decided that only uncontrolled cyclic wells would be tested. Early
survey efforts indicated that cyclic wells connected to a vapor recovery system
without a check valve could experience a back flow of steam into the well.
Since this might induce artificially high emissions if the vent were opened,
it was decided to omit controlled cyclic wells from testing.
16
-------
The Kern County APCD data offered a more realistic estimate of the
population and distribution of uncontrolled cyclic wells. The definition
used in compiling the KCAPCD survey data was adopted as the definition of a
cyclic well for this program. A cyclic well was defined as one which was
intermittently steamed and produced and was not affected by a nearby steam
injection well. The well operator's judgement was used to determine if a
steam injection well was affecting any given well, but some rough guidelines
were that the well would be considered a drive well if it was completed in the
same zone and within 250 feet of an injection well Chased on 2.5 acre pattern).
Despite the inconsistencies in the DOG data base, it played an important
part in the experiment design. The DOG data base was computerized and
included individual listings for each cyclic well. The KCAPCD survey data,
however, was available only in aggregated form (i.e. broken down only by
field, producer, and controlled/uncontrolled). There was no way to preselect
a random sample of wells based on the KCAPCD data. A hybrid approach was
chosen in which a random list of 1600 candidate wells was generated from the
DOG data. Wells were surveyed from this list, and those which were found to
be drive wells or to be controlled were eliminated from the survey. As many
such candidate wells were examined as necessary to fill survey quotas which
were set to represent the distribution of uncontrolled cyclic wells according
to the KCAPCD survey.
At the outset of the study, it was believed that cyclic wells emitted VOC
primarily during the period of one to two weeks following steaming. The initial
test plan, therefore, called for sampling all wells which were found to be
emitting. The survey demonstrated, however, that while cyclic wells may emit
at somewhat higher rates during the initial depressuring phase following steam-
ing, that about half of them continue to emit throughout the cycle. A
stratified sampling plan was developed to avoid spending too much effort
testing low emitters, the details of which are given below:
17
-------
Survey flow measurement Sampling quota
less than 0.1 liters/minute none
0.1 to 0.99 liters/minute 1 out of 10
1.0 to 5.0 liters/minute 1 out of 4
greater than 5.0 liters/minute all
This plan put the most emphasis on the high emitting sources, especially those
outside the range of an exact reading on the bubble-meter used to determine
the flow rate during the survey.
The experiment design described here resulted in a survey which included
only "true" cyclic wells chosen in a random manner and in proportion to the
population distribution indicated by the KCAPCD survey. Table 4-1 shows the
numbers of wells surveyed and tested compared to the KCAPCD data. The
sampling quotas by survey group are given in Table 4-2. It should be noted
that it was not always possible to sample all sources in the greater than
5.0 liter per minute category. Some sources were inaccessible for the large
van used as a mobile laboratory during sampling. Others were omitted due to
problems with scheduling or a variety of case specific causes. Those sources
not sampled were characterized by the mean emissions of other sources in the
same survey category for emission factor development.
18
-------
TABLE 4-1. SUMMARY OF TESTING
Area / Field
West Side/
Midway-Sunset
Belridge
Cymric
McKittrick
Lost Hills
Subtotal
Central County/
Kern River
Kern Front
Poso Creek
Kern Bluff
Edison
Subtotal
KCAPCD
Survey Population
Uncont. Cyclics
3829
251
327
789
41
5237
1205
263
91
12
200
1771
% of
Population
54.6%
3.6%
4.7%
11.3%
0.6%
74.2%
17.2%
3.8%
1.2%
0.2%
2H°/
,O7,
O C QO/
/_>, O/o
No. in Radian
Survey
189
15
22
44
1
271
42
19
10
0
16
87
% of
Radian Surv.
52.8%
4.2%
6.2%
12.2%
0.3%
75.4%
11.7%
5.3%
2.8%
0%
4.5%
24.6%
No.
Tested
31
3
1
7
0
42
10
4
0
0
2
16
% of
Tests
53.4%
5.2%
1.7%
12.1%
0%
72.4%
17.3%
6.9%
0%
0%
3.4%
27.6%
Grand Total
7008
100.0%
358
100.0%
58
100.0%
-------
TABLE 4-2. SAMPLING QUOTAS
Flow Rate Group
No. in
Survey
% in
Survey
Sampling
Quota
No.
Sampled
% of
Samples
less than 0.1 2,/min. 168 51%
0.1 to 0.99 5,/min. 51 14%
1.0 to 5.0 Jl/min. 93 26%
greater than 5.0 £/min. 46 13%
Totals 358 100%
0
5
24
46
75
0
4
26
Z8
58
0%
7%
45%
48%
100%
20
-------
SECTION 5
SAMPLING METHODOLOGY
The testing of cyclic well vent emissions was done in two stages. A
preliminary survey was conducted to locate the well and to get a rough idea
of its emission status. Selected sources from this survey were then quantita-
tively measured. This section discusses the detailed procedures used in both
surveying and sampling.
5.1 SURVEY PROCEDURES
The objectives of the survey included:
• finding the well,
determining whether or not it was truly an uncontrolled cyclic
well,
measuring the casing vent flow rate,
gathering well characteristics data, and
selection of wells for quantitative sampling.
Each of these functions is discussed in detail in this section.
The first step was to arrange a meeting with a representative of the
company to be tested. The list of random wells to be surveyed was examined,
and the producer's files checked to identify any wells which should be
eliminated from the survey (steam drive wells, fire-flood wells, water-flood
wells, and wells connected to a vapor recovery system). Once this preliminary
survey was completed, the producer's representative and the surveyor began a
field inspection of the remaining candidate wells.
21
-------
At each well site, the surveyor would make a number of observations which
were recorded on the survey data sheet (an example of which is included in
Appendix A) . Each well was carefully checked to insure that it was an un-
controlled steam cycle well by inspecting the area for vapor recovery systems
and steam injection wells. The position of the casing vent valve was noted.
If the casing vent was closed, the well was recorded as a zero emitter. If
the vent valve was open to the atmosphere, the flow rate through the casing
vent was measured using a bubble meter (unless a visual inspection noted a
high flow characterized by a steam plume). A stopwatch was used to measure
the time it took for a bubble to be displaced by 100 ml on a graduated
scale. The elapsed time was measured three times for each source, and the
flow rate corresponding to the average time was recorded on the survey sheet.
An exact flow rate could not be determined for sources emitting greater than
5 liters/minute using the bubble meter, so a static pressure measurement was
also made to aid in characterizing the emissions. The position of the casing
vent valve was left as it was found throughout the survey.
A well characterization data sheet was also filled out for each well
remaining in the survey. This sheet included data on the oil production
rates, life of the steaming project and particulars of steaming practice
(a copy of the sheet is included in Appendix A). These data were taken
for use in trying to correlate the emissions from a well with its physical
characteristics. Since much of this information required a file search, the
well characteristics data sheets were usually left with the producer for
later completion.
The surveyor was also responsible for selecting those sources to be
quantitatively sampled. The sampling quotas given in Section 4 were used
as a guide in this selection. For instance, in the category of wells emitting
between 0.1 and 0.99 liters per minute, only one well in ten was to be sampled.
The surveyor kept a running log of all wells found in this category and select-
ed for sampling the fifth, fifteenth, twenty-fifth, etc. A similar method
was used to select the one in four wells in the 1.0 to 5.0 liter per minute
22
-------
category. An attempt was made to test all wells which surveyed at more than
5.0 liters per minute. Selected wells were typically tested the day following
the survey, or as soon as possible.
5.2 QUANTITATIVE SAMPLING PROCEDURES
The estimate of flow rate and the presence of a steam plume from the survey
were used to select the right sampling procedure for each well. The surveyor
also noted fittings needed and special situations to be encountered by the
sampling crew.
The following parameters were measured at each source test site in
order to meet the objectives of this program:
• volumetric gas flow rate,
• gas phase composition, and
density and volume of condensible organics.
Sampling procedures necessary to obtain volumetric gas flow rate and provide
samples for analysis are described in this section. Only those systems
actually used will be described. Some of the high flow techniques were not
needed but were described in detail in the QA/QC manual.
5.2.1 Sampling Systems for Low, Medium and High Flow Wells
The sampling system varied depending on the noncondensible gas flow
and amount of condensate. Static well casing pressure proved to be of little
use as a third parameter to help in the selection of the sampling procedure.
Table 5-1 contains a list of the two parameters and the systems used for sampl-
ing. The two basic sampling systems used are illustrated in Figures 5-1 and
5-2.
23
-------
TABLE 5-1. SELECTION OF SAMPLING SYSTEMS
Noncondens ible
Gas Flow
Amount of Condensate
(Water plus Hydrocarbons)
Brief Description of System
©
(9)
).! - l.OL/min
.l - l.OL/min
None
Small amount present
-VL.O - 5.0L/min None
^1.0 - 5.0L/min Moderate amount present
>5.0L/min
(up to M.OOOL/min)
None
>5.0L/min
(up to ^1000L/min Small amount present
>5.0L/min Moderate to large amount
(up to VLOOOL/min) present (steam)
>1000L/min
>1000L/min
»1000L/min
None
Large amount present
(steam)
None
Preknockout pot, small condenser, small DGM
(similar to system shown in Figure 5-1)
As in (l) plus second small condenser
(see Figure 5-1)
As in (l)
As in (2)
As in (]p except used large DGM
As in (2) except used large DGM
See Figure 5-2. Preknockout, large knockout,
55 gal. condenser, condenser knockout, small
DGM
As in
except used 2 to 3 DGM's in parallel
As in (?) except used large DGM
As in (l) except used annubar in place of
the DGM (See Figure 5-3)
-------
I -
Ln
Sucker Rod
Production
Oil —
V.H. Valve
Cue lug
T«flon*-llned
PIMP
Oiie-W«y Flow •••trlctar
Figure 5-1. Low Flow Sampling Train
-------
Condenser
100' of Copper
Tuba
1" Bulkhead*
Pressure
Safety
Volva
Production
Oil
Dry Gas Meter ^Rotometer
Figure 5-2. Moderate Flow Sampling Train
-------
5.2.2 Sampling Procedures
Once a well was identified to be sampled, the surveyor and sampling
crew looked over the survey sheet and decided on the best sampling system
to use (see Table 5-1). Even though- ten different systems ware used for
sampling, the steps in setting up and taking the samples were very similar.
Table 5-2 contains a list of steps taken during sampling. A typical sampling
run lasted from one to two hours.
Preparations were made for sampling high flow wells with very large
condensate content (beyond the large condenser capacity). The apparatus
used to measure the flow rates from this type of well is illustrated in
Figure 5-3. Only one of the wells tested required the high flow measuring
devices, and it had very little condensate. The following three paragraphs
briefly describe the procedures which were to be used with each of these high
flow methods. A more detailed discussion of these methods can be found in the
QA/QC manual.
Flow measurement using an S-type pitot tube was based on determining
the cross sectional area of the pipe and the average stream velocity. The
average velocity was calculated from the differential pressure (AP), the
average stream temperature, wet molecular weight, and the absolute static
pressure. Barometric pressure readings were taken twice per day using the
barometer in the mobile laboratory. Static pressure in the pipe was
measured by disconnecting one leg of the S-type pitot and then rotating the
pitot so that it was perpendicular to the gas flow. A liquid trap was
inserted in the gauge line, leading to the upstream pitot tube leg. Static
pressure and AP measurements were measured by connecting a Capsahelic® gauge
to the pitot tube. Temperature of the gas stream was measured using a cali-
brated thermometer.
A second method for determining volumetric flow was the use of in-line
calibrated orifices. Differential pressure across the orifice was measured
27
-------
TABLE 5-2. MAJOR STEPS IN THE SAMPLING PROCEDURES
Step Task
1 Identify well and mark extent of casing valve
opening.
2 Place preknockout pot on well along with pipe
containing P and T gauges. Take static
temperature and pressure readings.
3 Open preknockout pot valve and set up rest of
system.
4 Test system for leaks.
5 Start condenser and make all initial meter
readings.
6 When analyst is ready for gases, start sampling
by closing preknockout valve and starting
pump inside mobile laboratory.
7 Record T, P and DGM readings periodically
(~10 min intervals) during run.
8 Stop run by shutting off mobile laboratory pump
and closing vent casing valve.
9 Record final DGM readings.
10 Test system for leaks.
11 Disassemble system and at the same time collect
hydrocarbon/water mixtures from all collection
devices.
12 Check to be sure casing valve is in the same
position as when first observed.
28
-------
Orifice Flow Measurement Technique
I tr)
Pressure Safety
Valve
I 3
ID
| Calibrated OrU
I 2" Pipe
Annuber Mounted
Horizontally
| Alternate Flow Measurement Technique
later-Filled
l.lnea with DP
Call Below Pip
I
_ llBHtlllgB V SS
tube/S-type pltot
Probe 1" 00
1" ID
Production
Oil
Pre-
Knackuut
V Teflon*
Dry Ga» Meter
Rotaneter
Figure 5-3. High Flow Sampling Train
-------
using a Capsahelic® gauge. The flow rate was calculated from AP, pipe
dimensions, orifice dimensions, and the orifice coefficient. The orifice
coefficient is a function of the Reynolds number through the orifice and
the ratio of the diameters of the orifice to the pipe. Two interchangeable
orifices of different sizes were used to take measurements of the flow.
The third method for flow measurement was an annubar. Its principle of
operation is similar to that of the S-type pitot tube. The major difference
is that the high pressure sensor uses four impact ports facing upstream, where
an S-type pitot has but a single impact port on the upstream face. Based on
Chebychef calculus for observation averaging, the properly located ports sense
the impact pressure caused by the flow velocity in each of the four equal
cross sectional areas of the stream. The high pressure side of the AP gauge
sees a continuous average of the impact pressure detected by the four sensing
ports. The impact pressure is the sum of pressure due to velocity of the
fluid and the line static pressure. The difference between the high and low
pressure, the AP, is proportional to flow rate according to Bernoulli's
Theorem. An Eagle Eye® differential flow meter was used to measure AP for
the well on which the annubar was used.
30
-------
SECTION 6
ANALYTICAL METHODOLOGY
The two on-site analytical procedures included the determination of
the noncondensible gas composition and the measurement of volumes, density
and temperature of the condensed hydrocarbons. The methods used for gas
phase analysis were independent of the sampling approach. Table 6-1
summarizes the methods for gas phase analysis, including instrumentation
and detection limits. Figure 6-1 is a block diagram depicting the mobile
laboratory instrumentation.
Two of the condensates were chosen for boiling point distribution analysis.
This off-site analysis is described in the final subsection of this section.
6.1 NONCONDENSIBLE GAS ANALYSIS
Before the gas stream from the wellhead casing vent was analyzed, it
was passed through a condenser system. After the condenser a slipstream of
the noncondensible gas stream was diverted to the mobile laboratory for
analysis.
Figure 6-1 illustrates in block form how the 1/4" Teflon® sampling
line was initially attached to the analytical instruments in the mobile
laboratory. This procedure gave variations in analyses due either to well
gas variability or line purging problems. The well gas variability was con-
firmed on a day-to-day basis (see Section 7). In order to integrate the
samples over the sampling period, a 100L Tedlar® bag was attached to the dry
gas meter with all other connections to the instruments eliminated. A compari-
son, of .this technique to the original technique gave identical results using
a well which showed no variation.
31
-------
TABLE 6-1. METHODS FOR GAS PHASE ANALYSIS
I'nrnmeter Description of Method Instrument Lower Detection Levels*
Fixed Gases (N2, Dual Column Gas Chromatograpltlc Fisher Model 1200 0.1% (V/V)
02, CO, C02, II, , Separation with Thermal Conduc- Gas Partltioner
tivity Detection
Methane, Ethane, Single Column Gas Chroma tographic Hewlett-Packard Model 0.1Z (V/V)
C3-C6, C6+ Separation Including Uackfluah 5730 with Model 3380A
with Flame lonizatlon Detector Integrator
*Lower Detection Levels were set by calibration range and program needs and not by
the detection limit of the instruments.
OJ
NJ
-------
LO
U)
•4" Toflot<4
Line From
Uondenaor
1
1
1
1
1
1
| By-Toes
*i /r^^
. Teflon®
| 1 Incd
• P""P
1
>•.
*•
FUed
Analyter
1
Hydrocarbon
Analyser
1
I
1
I
1 *_
Dry
Can
Meter
Recreational Vehicle (RV) Laboratory
Figure 6-1. Diagram of Instruments in Mobile Laboratory
-------
Once the sample was obtained, it was analyzed for fixed gases and
hydrocarbons. The following two sections describe these analytical methods.
6.1.1 Fixed Gases
A Fischer Model 1200 Gas Partitioner was used to measure the fixed gases
(COa, CO, Oa, N2, and CHO concentrations. This instrument was set up with a
0.25cc sample loop, dual columns and dual thermal conductivity (TC) detectors.
When the gases were introduced from the sample loop, they were carried into
Column 1 where COa was retained while the other gases passed quickly through
to the first TC detector to produce a composite peak. The COa then eluted
and was detected. The early eluting composite and the COa were subsequently
detected by the second TC detector. The carbon dioxide was permanently
adsorbed upon entering Column 2. The operating parameters for the analysis
are listed below:
Column 1: 1/8" x 6.5' aluminum packed with 80-100 mesh
Porapak PQ.
Column 2: 3/16" x 11' aluminum packed with 60-80 mesh
Molecular Sieve 13x.
Oven Temperature: 50°C.
Carrier Gas: 8.5% H2 in He at 30 cc/min.
The concentration of each of the species present was determined from
calibration curves generated from the analysis of certified standard mixtures.
The dry molecular weight of the gas stream was calculated, if needed, from the
fixed gas concentrations and major hydrocarbon species (other than CR^ which was
determined in the fixed gas analyses).
34
-------
6.1.2 Hydrocarbon Species
A Hewlett-Packard Model 5730 Gas Chromatograph equipped with dual
flame ionization detectors was used to measure the hydrocarbon species in the
noncondensible gas. A two valve arrangement allowed the introduction of
a known volume of sample (and standards) into the chromatograph and provided
for a blackflush to measure the hydrocarbons with retention times greater
than hexane (Ce+). The column in this instrument was a 3 meter, 1/8" OD
stainless steel tube packed with 10 percent SP1000 (Carbowax plus substituted
terephthalic acid) on 100/120 mesh Chromosorb W AW. This column provided the
optimum separation of the hydrocarbons (Ci to n-C6). The signal from the
flame ionization detector was recorded with a Hewlett-Packard Model 3380A
integrator. A comparison of peak areas to standards was used to quantify
the samples. Species identification was achieved using retention times of
species in the standard mixture. The peak with the retention time closest to
the standard component was assigned that standard component's identity.
6.2 ANALYSIS OF COLLECTED LIQUIDS
There were one to four liquid samples collected at any one source test
site. These included the knockout drum catches and the outlet of the
condenser(s). These catches usually contained both a water and an organic
phase. The water was separated from the hydrocarbons in a separatory funnel.
The volume of the water was measured in a calibrated graduated cylinder and
the temperature measured with a calibrated thermometer. The water was then
discarded.
The hydrocarbon liquids were analyzed for density, temperature, and
total volume. The total volume of the liquids was determined in a calibrated
graduated cylinder. In order to determine density on small amounts of hydro-
carbon that were available, the following procedure was used. Previously
calibrated volumetric flasks (0.500 ml through 10.00 ml sizes) were used to
measure an accurate volume of the liquids. The temperature and the weight
35
-------
of the liquid were determined using an NBS traceable thermometer and a cali-
brated analytical balance, respectively. From these measurements the density
was calculated.
6.3 BOILING POINT DISTRIBUTION
Two samples from the organic condensates were selected for boiling point
distribution analysis. The distribution procedure involved the determination
of the chromatographable organics in the normal hydrocarbon range of Cy to
Ci?. The following gas chromatographic conditions were used for this procedure:
Column: 10' x ?mm ID glass column packed with 10 percent
OV101 on 100-120 mesh Supelcoport.
Oven Program: 50°C for 4 min. , 10°C/min to 250°C and hold.
Carrier Gas: 25 ml/min N2.
Detector: Flame lonization
A standard mixture of C? to Ci? normal alkanes was injected into the
chromatograph to determine retention times. The samples were then injected
and an integrator slicing routine was used to assign that part of the sample
chromatographed between two adjacent hydrocarbons. The results of this
procedure are discussed in Section 8.
36
-------
SECTION 7
QUALITY ASSURANCE AND QUALITY CONTROL
Quality Assurance (QA) and Quality Control (QC) procedures were develop-
ed for this program to assess and document the precision, accuracy, and ade-
quacy of the test data collected during the project. Quality Control pro-
cedures included calibrations, systems checks for each sample run, control
sample analyses, and duplicate samples and analyses. Quality Assurance
activities included a systems audit of sampling procedures, a systems audit
of analytical procedures, a performance audit of laboratory analyses using
audit samples, and a check of the field data reduction procedures. Table
7-1 summarizes the precision and accuracy of the test data generated during
this program. The test data are adequate for the purposes of this program.
The QA/QC data and implications are discussed below. Appendix C
provides details of the various QA/QC data generated in support of the pro-
gram, including:
control charts for analytical quality control samples,
• chain-of-custody forms,
• equipment calibration documentation, and
• systems audit checklists.
7.1 SYSTEMS AUDIT RESULTS
As part of the Quality Assurance program for this project, a systems
audit was performed during the period 27 April through 30 April, 1981. The
37
-------
TABLE 7-1. ESTIMATED PRECISION AND ACCURACY OF TEST DATA
Measurement Parameter (Method)
Experimental Precision
Conditions (Std. Dev.) Accuracy
Comments
oo
Volumetric Gas Flow Rates
Noncondensible Gases
Low Flow Steam and Gases
(Total Stream Condensa-
tion)
High Flow Steam and Gases
a) Total Stream Condensa-
tion
Wellhead Gas 20%
Wellhead Gas 20%
Condensible Hydrocarbons
Wellhead Gas
b) Annubar (GARB) Method Wellhead Gas
20%
20%
Condensate from 10%
Wellhead Gas
+10%
+10%
+ 10%
+ 10%
Estimates are based upon
systems audit results,
equipment calibration and
repeat test data agreement
as discussed in Section 7.3.1.
Estimates based on expected
bias of the method; only one
test conducted using annubar.
+_10% 1 Estimates are based upon
J systems audit results.
Fixed Gases
Hydrocarbon Species
Density
Noncondensible 20%
Wellhead Gas
Noncondensible 20%
Wellhead Gas
Condensate from 10%
Wellhead Gas
+20%
+20% )
+5%
Estimates are based upon
performance audit results
and QC data evaluation.
Estimates are based upon
performance and systems
audit results.
-------
audit was designed to provide a comprehensive qualitative review of the
critical elements of the sampling/analytical procedures to assess their
effectiveness. The audit included evaluations of facilities, equipment,
training, procedures, recording keeping, QC, and reporting.
The precision and accuracy of certain measurement parameters are not
easily quantified by means of performance audits or replicate determinations.
The systems audit provides an alternative means of estimating and confirming
the precision and accuracy of these measurements which include volumetric
gas flow rates and condensible hydrocarbon determinations.
Both sampling and analytical activities were observed on 27, 28, and 29
April, 1981. Surveying activities were observed 29 April, 1981. Generally
the surveying, sampling, and analytical activities observed were consistent
with those specified in the Quality Assurance Project Plan (1) prepared for
this project. Deviations other than those discussed below were deemed to be
justifiable field modifications of the prescribed procedures which would
not adversely affect the data quality.
Several procedural changes and/or corrective actions were initiated as
a result of the systems audit. The most significant modification was the
initation of a bag sample technique for collection and analysis of noncon-
densible gases. The QA Plan stated that duplicate analyses of all noncon-
densible gases would be performed, and that ±20% agreement would be required
for acceptability. Due to the temporal variability of emissions from each
well, repeated injections using a sample loop, as prescribed in the Quality
Assurance Project Plan (QAPP) constituted replicate samples rather than
replicate analysis of a given sample. Data obtained in this manner measured
sample-to-sample variability but not analytical variability as desired.
Also, the 20% agreement limit imposed was inappropriate when applied to
variability of emissions rather than to analytical variability as intended.
The sampling/analysis procedures were amended to include the use of
a Tedlar® bag for sample collection. This procedure allowed the noncondensible
39.
-------
emissions to be collected over a period of time and provided a homogeneous
sample amenable to replicate analysis.
Other actions resulting from the QA system audit included the following:
• A modification was made to the sample logging procedures to incor-
porate the use of a bound and paginated master sample logbook
rather than a looseleaf binder.
The multipoint calibration of the Fisher Partitioner and Hewlett-
Packard Gas Chromatograph was redefined as a linearity check. The
daily single-point response factor checks were accepted as provid-
ing the best calibration in terms of day-to-day precision (repeat-
ability) . Response factor agreement on a day-to-day basis was
required to be ±20%.
The practice of recording intermediate dry gas meter volume
readings during sampling was instituted. Previously, only initial
and final readings had been recorded.
A larger capacity dry gas meter was sent to the field for use with
wells exhibiting high (>200 ft3/min) noncondensible gas flow
rates.
A positive pressure leak check procedure for pre- and post-test
systems checks was defined and initiated.
The 55 gallon drum condenser used for wet and/or moderate flow
wells was rebuilt to provide for easier condensate drainage.
Problems had occurred with pockets of condensate forming in low
spots in the condenser coil.
None of the problems above were judged to be serious enough to have
had significant adverse effects on data quality. The changes made represented
40
-------
an effort to maximize the efficiency and adequacy of the overall sampling/
analytical system and the quality of the data output.
7.2 PERFORMANCE AUDIT RESULTS
A performance audit is a quantiative assessment of the quality of the
data output of a sampling and/or analytical system. The performance audit
was conducted concurrently with the systems audit and addressed the analytical
procedures used for noncondensible gas analyses and for condensate density
determination. The results are expressed as relative accuracy (%A) calcul-
ated as
M—T
%A = ^ x 100,
where, %A = relative accuracy
M = measured value of a standard
T = "true" value of the standard
7.2.1 Density
The performance audit or the density determinations were performed
using four liquid hydrocarbon standards:
• 2-propanol,
methylene chloride,
• acetone, and
• 3-methylpentane.
Two determinations of density were performed on each standard. The
average value is reported. No correction has been made for temperature.
The results are summarized in Table ~h 2 below.
41
-------
TABLE 7-2. PERFORMANCE AUDIT RESULTS FOR DENSITY DETERMINATIONS
Compound
2-propanol
methylene chloride
acetone
3-methylpentane
d measured
0.774
1.30
0.774
0.654
d actual-20°C
0.781
1.3266
0.7899
0.6645
%A
-0.9
-2.0
-2.0
-1.6
The average accuracy of the density determination, -1.6%, is well
within the ±10% acceptability criteria. The slight low bias indicated is
most likely due to the elevated temperature (~95°F) at which the determina-
tions were made.
7.2.2 Noncor.densible Gas Analysis
The performance audit of the gas phase analyses was performed by
challenging the Fisher Partitioner and Hewlett Packard gas chromatograph
with bottled standard gases. Four separate gas mixtures were used as
audit standards for the noncondensible analyses:
(1) C02, CHit, N2 and 02, cylinder j?A9541;
(2) H2, CO, and N2, cylinder #A10753;
(3) C2H6, C3H8 and N2 , cylinder //A5401; and,
(4) CHit, C2H6, CaHg, n-CitHjo, i-CijHio, n-CgHiz, i-C5Hi2 and N2 , Scotty
II® cylinder, SSG Project #44915.
Mixtures #1, #2 and #4 are Certified Master Standards (±2% analytical
accuracy) obtained from Scott Specialty Gases, Inc. Mixture #3 is a Certi-
fied Plus Standard (±1% analytical accuracy) obtained from Scientific Gas
Products, Inc. All four mixtures were analyzed for fixed gases (C02, 02 ,
CO and CE^) using the Fisher Gas Partitioner. Hydrocarbon analyses of
mixtures //3 and #4 were performed using the Hewlett Packard gas chromatograph.
The audit results are summarized in Table 7-3.
42
-------
TABLE 7-3. PERFORMANCE AUDIT RESULTS, NONCONDENSIBLE GASES
Standard Species
#1 C02
CH,,
N2
02
#2 H2
CO
N2
#3 C2H6
C3H8
N2
#4 CHij
C2H6
C3H8
Z C3 +
N2
Instrument
FP
FP
FP
FP
FP
FP
FP
GC
GC
FP
GC
GC
GC
GC
FP
Measured
Concentration
(% V/V)
50.9
42.6
11.6
4.5
NA
6.38
84.7
32.9
10.6
51.5
0.215
0.563
0.312
1.08
81.2
Actual
Concentration
(% V/V)
46.00
39.98
9.96
4.04
4.95
5.10
89.95
29.90
9.99
60.11
0.261
0.251
0.314
0.954
98.2
%A
10.7
6.6
16.5
11.4
—
25.1
-5.8
10.0
6.1
-14.3
-17.6
124
-0.6
13.2
-17.3
NA = Not Analyzed
FP = Fisher Gas Partitioner
GC = Hewlett Packard Gas Chromatograph
43
-------
The measured accuracies of the gas phase analyses are generally within
the specified +20% accuracy limits. Exceptions include:
• CO concentration of standard #2, and
• C2H6 concentration of standard #4.
The discrepancy in the CO determination for mixture #2 was found to be
due in part to the way in which the baseline was established for measurement
of the calibration standard peak height. Because of the presence of a
large CHi, peak in the calibration chromatogram which is partially merged
with the CO peak, the baseline for the CO peak is difficult to determine.
If the peak height for CO is measured assuming a flat baseline (rather
than by the crude tangent skim method which was used), and a new CO response
factor calculated, the measured concentration of CO in the audit gas becomes
5.5Z. This new value represents a relativft accuracy of 13.7%. In any event,
the accuracy of the fixed gas analyses does not adversely impact the emission
factors since the fixed gas composition is used only for calculation of mole-
cular weight of the gas.
The high positive bias in the C2Hs analysis may be attributed to the low
concentration in the audit standard 00.251%) as compared to the calibration
standard 05.0%). The purpose of this low range standard was to assess the
validity of precision data generated early In the program using a QC standard
for Cj-C6 hydrocarbons at 0.1%. furthermore, the C2Hs values also do not
adversely impact the calculated emission factors since neither CE* nor C2H6
values are included in the calculations.
Although it is not indicated in the table of results, analysis of audit
standard #3 using the Fisher Partitioner gave a false positive result for
02 (25.7% reported). Ethane (C2H6) apparently has the same retention time
as oxygen on the Fisher. The 02 to N2 ratio of air is used to subtract
this false positive out of the analytical data for samples. Due to the
synthetic nature of the audit standard, this correction could not be
44
-------
applied, so the 02 value was simply not reported. If the C2H6 to N2 ratio
in the audit standard (0.497) is used to correct the analytical result, as
below,
26.7 - (0.497 x 51.5),
where 26.7 = reported concentration of 02, %
0.497 = ratio of C2H6 to N2 in the audit gas mixture
51.5 = audit gas mixture measured concentration of N2 in the audit
gas mixture, %,
the resulting 02 concentration is 1.1%. This method of correcting the data
seems to be a satisfactory solution.
7.3 ANALYTICAL PRECISION
Table 3-1 of the Quality Assurance Project Plan (1) and Table 7-1 present
the original precision estimates for each major measurement parameter. The
estimates represent the maximum expected standard deviation of the measure-
ment, expressed as percent of the mean (relative standard deviation, RSD).
7.3.1 Volumetric Gas Flow Rate
The precision of the volumetric gas flow rate determinations was estimated
to be 20%. All flow measurements for this program with the exception of well
number 3 were made using the total stream condensation methods discussed
in the QAPP and in Section 5. The precision of the method is thus a function
of the precision of the dry gas meter volume measurement and the measurement
of elapsed time. The systems audit indicated acceptable compliance with gas
flow rate measurement procedures and the calibration of the gas meters indicated
that all were within the required ±5% accuracy limit. Three wells were tested
twice each during the program. The results of the repeat measurements of
volumetric flow rates are summarized in Table 7-4.
45
-------
TABLE 7-4. VOLUMETRIC GAS FLOW RATE VARIABILITY
Well
No.
164
173
176
Date of
First Test
4/16/81
4/16/81
4/21/81
Flow Rate
(ACFM)
0.51
0.20
0.102
Date of
Second Test
4/23/81
4/23/81
4/22/81
Flow Rate
(ACFM)
0.54
0.22
0.075
Repeatability
(RSD)
1.0%
6.7%
21.6%
Only the tests of well number 176 showed a flow rate precision (.repeat-
ability) for the two measurements in excess of 20%. It is believed that
this was due primarily to a temporal variation in well emissions rather
than variability in the sampling procedure. The reduction in emissions during
the second test was obvious at the time the well was sampled, and was signifi-
cant enough that the sampling team switched to the "low flow" sampling
apparatus for the second test.
The data above and the systems audit observations support the conclusion
that the overall precision of the flow rate measurement data is within the
estimated 20%.
7.3.2 Condensible Hydrocarbon Emissions
The results of the condensible hydrocarbon emissions for the repeat
tests discussed above are summarized in Table 7-5 below.
TABLE 7-5. CONDENSIBLE HYDROCARBON EMISSIONS VARIABILITY
Well
No.
164
173
176
Date of
First Test
4/16/81
4/16/81
4/21/81
Condensible
HC Emissions
(Ibs/hr)
0.002
1.050
0-454
Date of
Second Test
4/23/81
4/23/81
4/22/81
Condensible
HC Emissions
(Ibs/hr)
0.004
0.899
0.015
Repeatability
(RSD)
47.1%
2.7%
132%
46
-------
As discussed previously, the high variability for well number 3 is
believed to represent the actual temporal variation in the emissions, and as
such, the variability does not reflect measurement variability. The precision
of 47% indicated for well number 1 is attributed to the low condensible
emissions. The systems audit of the condensible hydrocarbon emissions measure-
ment sampling procedures indicated that proper procedures were used for sample
collection. Based upon the above data and the systems audit results, it is
felt that the overall precision of the condensible hydrocarbons measurement
was within 20% as estimated.
7.3.3 Fixed Gases
As discussed in Section 7.1, the systems audit of the analytical system
resulted in a revision of the sampling/analytical procedures for fixed
gases and noncondensible hydrocarbons. Prior to the audit, the method of
gas phase analysis consisted of analysis of replicate samples. After
instituting the bag sampling procedures, replicate analyses were performed
upon each sample. Thus, two different types of variability may be calculated
from these data:
• sample-to-sample variability of well emissions, (sample repeatabil-
ity) , and
• analytical variability with respect to analysis of samples (sample
replicability).
The data from the quality control standard analyses may also be used to
assess analytical variability. The data from replicate analyses of the QC
standard at one site under a given set of instrument conditions and using the
same response factor represent one measure of analytical variability: standard
replicability. Since the QC standard was analyzed at each site with each set
of sample analyses, the site-to-site or day-to-day analytical variability may
also be quantified. This measure of precision is referred to as standard re-
peatability. The data for both standard and sample repeatability and replicabil-
ity for fixed gases are summarized in Table 7-6 below.
47
-------
TABLE 7-6. SUMMARY OF PRECISION FOR FIXED GAS ANALYSES
Species
CO
COa
02
N2
CH,,
Analytical Variability
Standard
Replicability
(PRSD)
1.81%
1.23%
2.18%
5.00%
1.67%
Standard
Repeatability
(RSD)
3.5%
4.6%
5.2%
7.8%
2.8%
Sample
Replicability
(PRSD)
ND
3.44%
2.52%
0.79%
0.34%
Sample Variability
Sample
Repeatability
(PRSD)
ND
4.47%
45.8%
21.2%
4.88%
ND= Not Detected
The values in Table 7-6 above for standard repeatability represent the
relative standard deviation for the measurements. Values indicated for standard
and sample replicability and for sample repeatability represent the pooled
relative standard deviation (PRSD), which is a measure of the variability of
the relative standard deviations for n sets of data calculated as
PRSD
/ n
fz
i=i
n
Z
i=i
Y 2
xi
DF±
~ X
where X-^ = relative standard deviation of data set i
DFi = degrees of freedom for data set i Ck^-i)
n = total number of data sets
k^ - number of data points in data set i
i = data set 1, 2, 3, ... n
The terms for degrees of freedom in the above equation allow the data to be
weighted according to the number of data points in each data set.
It should be noted that the last category, "Sample Repeatability", is not
actually a measure of analytical precision. Rather, it indicates the net
variability arising from two sources:
48
-------
• temporal variability of well emissions, plus
• analytical variability.
Comparing these values to either standard or sample replicability indicates
that the analytical variability was generally less than short-term temporal
variations in the emissions themselves. As indicated in Table 7-6, the pre-
cision of the fixed gas analyses is well within the 20% estimate for all
categories, except Na and QZ sample repeatability.
7.3.4 Noncondensible Hydrocarbon Species
The data for precision of noncondensible hydrocarbon speciation analyses
may be categorized in the same way as those for fixed gases. The only major
difference is that several different hydrocarbon QC standards were used during
the course of the program. The data for analytical repeatability and replica-
bility of samples is summarized in Table 7-7 below.
TABLE 7-7. ANALYTICAL VARIABILITY OF HYDROCARBON SAMPLES ANALYSES
Species
CH^
C2H6
£ C3-C6
I Cfff
Replicability
(PRSD)
3.44%
5.09%
11.6%
8.21%
Repeatability
(PRSD)
7.38%
23.6%
17.5%
11.5%
Repeatability and replicability for each of the various QC standards is
summarized in Table 7-8 below. The validity of these estimates of precision
is limited in some cases due to the small number of applicable data points.
In each case, the number of pairs of analyses upon which the calculated
precision is based is indicated (n = number of pairs).
49
-------
TABLE 7-8. SUMMARY OF PRECISION FOR HYDROCARBON QC STANDARD ANALYSES
Species
Low Standard Mixture*
High Standard Mixture1
0.5Z Propane Standard3
5.0% Propane Standard
Replicability Repeatability Replicability" Repeatability Replicability Repeatability Replicability Repeatability
(RSD. n - 1) (RSD. n - 10) JPRSD, n - 1) (RSD. n - 25) QPRSU. n - 2) (RSD. n - 7) (n - 0) (RSD. n - «)..-
CHu
13.0
13.0
8.06
6.52
C2H6
6.22
65.6
6.78
7.17
C3Ha
5.82
66.1
5.02
7.59
11.1
11.1
8.2
'CH,, " 0.074%; C2H6 " 0.109%; C3H8 - 0.103%; Scotty II® Mix #236.
2CHM " 40.0%; C2H6 " 4.0%; C3H8 - 1.6%; SSG Cylinder 01A5924
'Mixture contained only C3H8 in N2, therefore no values for CHi, or C2B» could be obtained using this standard; SGP Mini-Mix®, Ref. 229987.
° ""No replicate analyses of this standard wera made
-------
The data in Tables 7-7 and 7-8 indicate that the precision was generally
well within the estimated 20%. The exceptions to this generally represent cases
where the analyte concentration approached the detection limit and/or was
more than an order of magnitude lower than the concentration of the calibration
standard. The performance audit results relect the difficulty of obtaining
accurate measurements near the detection limit of the method, and confirm this
conclusion.
7.3.5 Density
The systems audit of analytical procedures revealed that daily control
sample density determinations were not being performed as prescribed in the QAPP.
This procedural deviation was documented in the QA audit report and corrective
action was recommended. However, the use of a control standard for density
was never implemented as a routine procedure. There is therefore no data
available for calculating the precision of the method over the duration of the
project. Two density determinations were performed on each of the audit
standards, however, and the analytical variability may be estimated from these
data. These results are summarized in Table 7-9 below.
TABLE 7-9. ANALYTICAL VARIABILITY OF DENSITY DETERMINATION
Compound
2- Propanol
Methylene Chloride
Acetone
3- Methylpentane
d Measured
(Mean)
.774
1.30
.774
.652
Repeatability
(RSD)
0.09%
0.0%
0.09%
0.11%
Based on these data, the pooled relative standard deviation is less than 0.1%.
51
-------
Despite the limited data available for estimating the precision of the
density analyses, the performance audit results support the conclusion the
overall precision was within the specified 10%.
7.4 EQUIPMENT CALIBRATION
The checkout and calibration of source sampling equipment is essential
to maintaining data quality. Accepted calibration procedures were used to
calibrate the sampling equipment used in this program. These procedures are
detailed in the Quality Assurance Project Plan. The results of the pertinent
calibrations are documented in Appendix C. These data indicate that the test
data were obtained using acceptable equipment.
7.5 DATA CAPTURE
Table 3-3 of the QAPP indicates an expected data capture of 90% for
each applicable measurement parameter. A total of 62 tests were conducted
on 59 wells during the course of the project. Three tests (Table 7-4)
were judged to be questionable in the field and the wells were retested.
The results of the first test on these wells are not included in the emissions
factor data base. One test was rejected as invalid during the data review
and validation process. Thus, a total of 58 valid tests were conducted. The
valid data percentage of the total tests conducted is therefore 93.5%. The
scope of work required 50 tests. The valid data percentage of the total
tests required in the scope of work is therefore 116%.
7.6 DATA VALIDATION
The overall sampling, analytical, and data reduction scheme for this
project was designed to maximize valid data output. A number of different
criteria were used to assess the validity of the test results. The validation
process was an integral part of all phases of the testing. Specific aspects
related to validation included:
52
-------
• the use of preformatted data sheets which served as procedural
checklists,
• the delineation of specific control limits and acceptability
criteria for leak checks, calibrations, analytical precision, etc.,
• on-site review of field data,
review and evaluation of comments and notations concerning
problems and/or special situations related to all sampling and
analyses,
recalculation of all data for 10% of the tests (six wells chosen
at random), and
subjective evaluation of reasonableness of test data and resulting
correlations.
Three wells were retested during the course of the project, as mention-
ed in Section 7.5. The initial tests of these wells were judged to be of
questionable validity due to apparent equipment problems. A fourth test
was rejected during the final review process because of a number of sampling
and analytical problems which were noted on the data sheets.
The calculations check on 10% of the tests identified a number of minor
calculation errors. Of these, however, only one ultimately impacted the
emission data by more than 5%. This error was a failure to add the coridensible
VOC emissions to the noncondensible VOC emissions for total VOC emissions.
Although the manually calculated data indicated a total VOC emissions value
equal to only the noncondensible VOC emissions for that well, the error was
ultimately corrected during the computerized phase of the data reduction process
and did not impact the reported result.
53
-------
Statistical treatment of the audit data and QC data is consistent with
the definitations and procedures outlined in Volume I of the EPA Quality
Assurance Handbook for Air Pollution Measurement Systems (2). Outliers
were identified using the Dixon criteria and rejected at the 5% significance
level.
54
-------
REFERENCES
1. Balfour, W.D. and D.L. Lewis, Assessment of VOC Emissions From Well
Vents Associated with Thermally Enhanced Oil Recovery, Quality Assur-
ance Project Plan, Austin, Texas, Radian Corporation, 1981.
2. "Quality Assurance Handbook for Air Pollution Measurement Systems",
Volume I, "Principles", U.S. Environmental Monitoring and Support
Laboratory, Research Triangle Park, North Carolina, January 1976.
EPA-600/9-76-005.
55
-------
SECTION 8
DETAILED RESULTS
The preliminary field survey was designed to confirm that candidate
wells were true uncontrolled cyclic wells and to make a rough estimate of
the emission range for each well. A total of 829 candidate wells were
examined, out of which 358 were determined to be uncontrolled cyclic wells.
The flow rate from the casing vent valve was measured using a bubble meter, a
dry gas meter, or a visual estimate for each well remaining in the survey.
Table 8-1 presents a summary of the results of the survey categorized by the
producing field and the casing vent flow rate. In addition to the individual
fields, data are presented for all the fields in western Kern County and those
in central Kern County in aggregate form. Table 8-2 provides a breakdown of
the non-blowing wells by field. Table 8-3 presents survey data broken down
by producer.
A total of 58 wells (out of the 358 surveyed) were selected for quantita-
tive emission measurement. The distribution of these sampled wells by field
is given in Table 8-4 and by producer in Table 8-5. The results of each of
the 58 tests is given, organized by field, in Table 8-6.
A well characterization survey form was left with each producer to be
completed for each well retained in the survey. This form was designed to
provide information on steaming practices and production characteristics of
the well, which were to be used in correlating casing vent emissions. Table
8-7 presents a summary of the well characterization data obtained. The mean
value of all responses is presented for each parameter, along with the standard
error and the number of responses. No data is presented here on the oil/water
ratio because it appears that many respondents used widely differing forms of
expressing that ratio; the resulting indicators would have no meaning. A full
56
-------
TABLE 8-1. SURVEY RESULTS BY FIELD
Survey Flow
Rate Range
(liters per minute)
less than 0. 1
0.1 to 0.99
1.0 to 5.0
greater than 5.0
Total Surveyed
Number of Wells Found in Each Survey Range
4-1
cu
CO
§
C/3
1
cd
•5
s
101
28
44
16
189
o
• «-4
M
4-1
4-1
U
a
18
10
12
4
44
o
•H
O
14
3
4
1
22
01
60
T)
t-l
01
m
8
1
3
3
15
co
rH
r-l
•H
w
4->
co
a
0
0
1
0
1
a
o
•H
00
Si
i»_4
cd C
4J M
O 0)
4-1 4-1
43 CO
en ts
141
42
64
24
271
M
0)
^
*H
P5
a
t-i
15
4
15
8
42
a
o
pt(
a
M
H t^
168
51
93
46
358
-------
TABLE 8-2. BREAKDOWN OF NON-BLOWERS BY FIELD
Number of Wells of Each Type Found in Each Field
tn
oo
V.
^•x.
\. Field
^s.
^x^
^X^^
^^^^
^*sx.
^x.
^x.
^s.
^x^
Type of Non-Blower — ,.
^N
No detecable flow
Detectable flow less than O.l£/min.
Negative flow
Normal production-casing closed
Well being steamed
Well soaking
Well being worked over
Casing vent clogged
4J
-3
0
0
0
0
0
0
0
0
M
^.
5
g
&4
8
1
3
0
0
1
0
2
•u
0
O
En
g
M
0
1
1
1
0
0
0
0
»^
0)
-------
TABLE 8-3 SURVEY RESULTS BY PRODUCER
Survey Flow
Rate Range
(liters per minute)
less than 0.1
0.1 to 0.99
1.0 to 5.0
greater than 5.0
Total Surveyed
1
12
2
2
3
19
Number of Wells Found in Etch Survey Rang*
2
0
0
2
0
2
3
4
0
0
0
4
4
2
0
0
1
3
5
1
0
0
0
1
6
2
0
0
0
2
7
33
6
13
3
55
8
1
0
5
4
10
9
28
14
22
10
74
10
1
1
0
0
2
11
7
1
1
0
9
12
15
9
6
9
39
13
3
1
0
0
4
14
2
0
0
0
2
15
5
1
2
4
12
16
17
7
17
2
43
17
12
3
11
8
34
18
12
3
4
1
20
19
1
0
0
0
1
20
3
0
4
0
7
21
3
0
0
0
3
22
3
1
2
1
7
23
1
2
2
0
5
o
168
51
93
46
358
-------
TABLE 8-4. SAMPLING DISTRIBUTION BY FIELD
Survey Flow
Rate Range
(liters per minute)
less than 0. 1
0.1 to 0.99
1.0 to 5.0
greater than 5.0
Total Sampled
Number of Wells Sampled
4J
01
CO
C
en
!>••
CO
*
• H
s
0
3
18
10
31
x
o
•H
VJ
4-1
4-1
• H
1
0
0
3
4
7
o
• H
1
0
0
0
1
1
0)
00
-a
• H
M
,-1
0)
PQ
0
0
1
2
3
to
H
rH
a
4-1
CO
3
0
0
0
0
0
a
o
•H
C>0
0)
{tf
rH
2 e
o
2
e
.s
0
1
3
6
10
4J
C
0
M
PM
l-i
S
0
0
1
3
4
•3
0)
M
o
o
0}
0
PL.
0
0
0
0
0
g
o
CO
•rl
'O
H
0
0
0
2
2
a
o
• rt
60
S
rH
Cfl r-l
4J Cd
O M
W W
•9 §
w o
0
1
4
11
16
>>
4-1
S
3
o
o
1— 1
cd C
4-1 M
O CU
H ttfi
0
4
26
28
58
-------
TABLE 8-5. SAMPLING DISTRIBUTION BY PRODUCER
Survey Flow
Rate Range
(liters per minute)
less than 0.1
0.1 to 0.99
1.0 to 5.0
greater than 5.0
Total Surveyed
Number of Wells Sampled
. ' . i— I
1
0
l
l
1
3
2
0
0
1
0
1
3
0
0
0
0
0
4
0
0
0
0
0
5
0
0
0
0
0
6
0
0
0
0
0
7
0
2
3
3
8
8
0
0
0
2
2
9
0
1
9
6
16
10
0
0
0
0
0
11
0
0
1
0
1
12
0
0
2
5
7
13
0
0
0
0
0
14
0
0
0
0
0
15
0
0
0
1
1
16
0
0
4
2
6
17
0
0
1
6
7
ia
0
0
i
i
2
19
0
0
0
0
0
20
0
0
2
0
2
21
0
0
0
0
0
22
0
0
0
1
1
23
0
0
1
0
1
to
u
o
H
0
4
26
28
58
-------
TABLE 8-6. SAMPLING RESULTS BY FIELD
FIELD: BELRIOQE
SUMMARY OF MASS EMISSIONS
CYCLIC TEOR WELLHEAD CASINO VENTS
TOTAL
NON-CONOENSIBLES
WELL
NUMBER
1
2
3
VUU 1
EMISSIONS
(LB/DAY)
13.19
3.18
39. 81
HTUKUliAKDUN
EMISSIONS 1
(LB/DAY)
131.88
68.57
8S.68
ViUNUCNSlHl.E
1YDROCARBONS
(LB/DAY)
8. 88
0.00
0.00
CH4
(LB/DAY)
117.88
68.21
39.93
C2HB
(LB/DAY)
0.79
0.18
10.12
C3-C8
(LB/DAY)
1.92
1.21
29.68
C6+
(LB/DAY)
2.69
1.87
B.94
NJ
Continued/
-------
TABLE 8-6. (Continued)
SUMMARY OF MASS EMISSIONS
CYCLIC TEOR WELLHEAD CASINO VENTS
FIELD: CYMRIC
TOTAL
VOC HYDROCARBON CONDENSIBLE
WELL EMISSIONS EMISSIONS HYDROCARBONS
NUMBER (LB/DAY) (LB/DAY) (LB/DAY)
NON-CONOENSIBLES
CH4 C2H6 C3-CB CB+
(LB/DAY) (LB/DAY) (LB/DAY) (LB/DAY)
16
105.87
229.07
101.24
122.87
0.33
0.84
3.99
Continued/
-------
TABLE 8-6. (Continued)
SUMMARY OF MASS EMISSIONS
CYCLIC TEOR WELLHEAD CASINO VENTS
FIELD: EDISON
TOTAL
VOC HYDROCARBON CONDENSIBLE
WELL EMISSIONS EMISSIONS HYDROCARBONS
NUMBER (LB/DAY) (LB/DAY) (LB/DAY)
38 1.43 138.49 0.00
3B 1.80 382.58 0.00
NON-CONDENSIBLES
CH4 C2H6 C3-C8 C8+
(LB/DAY) (LB/DAY) (LB/DAY) (LB/DAY)
133. 8B 1.21 0.19 1.24
188.33 178.48 0.18 1.83
Continued/
-------
TABLE 8-6. (Continued)
Ul
FIELD: KERN FRONT
VOC
WELL EMISSIONS
NUMBER ( LB/DAY )
54 0.22
55 0.39
56 2 . 07
57 3.31
SUMMARY OF MASS EMISSIONS
CYCLIC TEOR WELLHEAD CASINO VENTS
TOTAL
HYDROCARBON
EMISSIONS
(LB/DAY)
27.34
SO. 66
144.82
188.39
CONDENSIBLE
HYDROCARBONS
(LB/DAY)
0.00
' O.OO
O.57
1.48
NON-CONDENSIBLES
CH4
(LB/DAY)
27. OS
90.25
142.39
182.71
C2H6
(LB/DAY)
0.08
O.O1
0.35
0.38
C3-C6
(LB/DAY)
O.04
O.OO
0.38
0.39
C6+
(LB/DAY)
0.18
O.39
1.13
1.46
Continued/
-------
TABLE 8-6. (Continued)
Ov
FIELD: KERN RIVER
£$$&4E4c$££&4t4t$44t4c;|:4[$£^
VOC
WELL EMISSIONS
NUMBER (LB/DAY)
73 0.81
74 1.74
75 O.69
76 O.24
77 2.88
78 O.84
79 4.48
80 2 . 18
81 25.89
82 4O . 05
83 0.54
SUMMARY OF MASS EMISSIONS
CYCLIC TEOR WELLHEAD CASINO VENTS
t**^*****-([***^***«******^**J'*****'*'*'^**^****'*'*«******«^^***^^*^»^********
TOTAL
HYDROCARBON
EMISSIONS
(LB/DAY)
88.70
22.O2
01. 08
32. 6O
8.79
11O.30
8.42
198.37
3O.24
85.24
41.22
r»»*p^»*»»»»»»»»»
CONDENSIBLE
HYDROCARBONS
(LB/DAY)
0.00
O.OO
O.OO
O.OO
0.74
0.00
4.29
0.41
25.68
4O.53
0.00
»»»»»»»»»»»»»»»»»»»»»»»iit'*t»»»»*^»t»<^^»^^'^»
NON-CONDENSIBLES
CH4
(LB/DAY)
84.09
20.24
9O.33
32.25
5.90
109.39
0.98
195.68
4.33
24.18
40.33
C2H8
(LB/DAY)
2.11
O.O4
O.O8
0.11
0.02
0.07
O.OO
O.BB
0.02
O.11
0.35
C3-C6
(LB/DAY)
0.05
0.45
O.03
0.11
0.80
0.00
0.04
O.29
0.12
0.18
0.08
C8+
(LB/DAY)
0.45
1.29
0.88
0.13
1.52
0.84
O.13
1.48
0.09
O.28
0.48
Continued/
-------
TABLE 8-6. (Continued)
FIELD: MCKITTRICK
WELL
NUMBER
116
117
118
119
120
121
122
VOC
EMISSIONS
(LB/DAY)
16.01
2.48
1.73
88.56
0.21
2.37
8.37
SUMMARY OF MASS EMISSIONS
CYCLIC TEOR WELLHEAD CASING VENTS
TOTAL
HYDROCARBON
EMISSIONS
(LB/DAY)
39.55
2.89
37.49
396.14
8.13
4.17
509.51
CONDENSIBLE
HYDROCARBONS
(LB/DAY)
3.12
2.17
0.04
1.54
0.08
0.27
O.OO
NON-CONDENSIBLES
CH4
(LB/DAY)
23.44
0.41
35.73
307 . 25
7.92
1.78
500.83
C2H6
(LB/DAY)
0.10
0.00
0.03
0.32
0.00
0.02
0.31
C3-C8
(LB/DAY)
4.41
0.01
0.14
10.36
0.03
0.70
0.48
C8+
(LB/DAY)
8.48
0.30
1.55
76.87
0.09
1.40
7.91
Continued/
-------
TABLE 8-6. (Continued)
co
SUMMARY OF MASS EMISSIONS
CYCLIC TEOR WELLHEAD CASINO VENTS
FIELD: MIDWAY-SUNSET
^4t^:+^^E^t
-------
TABLE 8-6. (Continued)
SUMMARY OF MASS EM
CYCLIC TEOR WELLHEAD
FIELD: MIDWAY-SUNSET
***********************************************
TOTAL
VOC HYDROCARBON CONDENSIBLE
WELL EMISSIONS EMISSIONS HYDROCARBONS
NUMBER (LB/DAY) (LB/DAY) (LB/DAY)
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
0.79
8.17
4.00
1.26
1.26
12.08
4. 81
4.05
7.31
3.40
3.59
1.21
2.37
1.28
11.92
2.49
13.44
8.94
61.22
15.95
38.50
89.63
164.63
182.46
81.87
25.42
35.86
50.81
25.19
44.31
0.35
0.78
0.00
0.15
0.00
8.54
0.49
0.53
3.62
1.00
0.19
O.OO
0.83
0.41
8.69
ISSIONS
CASINO VENTS
*************
CH4
(LB/DAY)
1.69
4.47
4.65
59.85
14.48
26.07
84.69
158.95
172.62
77.65
21.64
34.62
48.41
23.32
32.24
****************************
NON-CONDENSIBLES
C2H8
(LB/DAY)
0.01
0.80
0.29
0.11
0.24
0.35
0.34
1.63
2.63
0.83
0.19
0.03
0.04
0.58
0.15
C3-C6
(LB/DAY)
0.19
4.85
2.96
0.07
0.59
2.11
0.35
1.13
0.35
1.46
1.29
0.12
0.07
0.34
0.81
C8+
(LB/DAY)
0.24
2.53
1.04
1.04
0.67
1.43
3.78
2.39
3.33
0.93
2.11
1.09
1.48
0.53
2.42
Continued/
-------
TABLE 8-6. (Continued)
SUMMARY OF MASS EMISSIONS
CYCLIC TEOR WELLHEAD CASING VENTS
FIELD: MIDWAY-SUNSET
TOTAL NON-CONDENSIBLES
VOC HYDROCARBON CONDENSIBLE
WELL EMISSIONS EMISSIONS HYDROCARBONS CH4 C2H8 C3-C8 C8+
NUMBER (LB/DAY) (LB/DAY) (LB/DAY) (LB/DAY) (LB/DAY) (LB/DAY) (LB/DAY)
191 4S.74 71.83 41.39 2B.7O 0.39 1.98 2.38
-------
TABLE 8-7. WELL CHARACTERIZATION SURVEY RESULTS
1.
2.
3.
4.
5.
6.
7.
8.
Well Parameter
Total Number of Steam Cycles to Date
Mean Value (X)
Standard Error (SE)
Number of Observations (N)
Frequency of Steaming (Months/Cycle)
X
SE
N
Time Since Last Steaming (Days)
X
SE
N
Steam Dosage (Barrels)
X
SE
N
Soaking Period (Days) _
X
SE
N
Oil Production Rate (Bbl/day)
X
SE
N
Cumulative Oil Production
Since Steaming Began ( 10 3 Bbl)
X
SE
N
API Gravity of the Oil (°API)
X
SE
N
Western
Region
8.5
0.6
240
10.0
0.7
224
236.3
10.6
238
9719
573
250
5.8
0.3
226
25.2
6.0
215
54.9
4.6
181
12.7
0.1
245
Central
Region
6.9
0.6
77
7.5
1.5
4
260.5
22.0
79
9798
531
85
3.0
0
4
11.3
1.0
87
21.1
4.4
27
13.7
0.2
63
Overall
8.1
0.3
317
9.9
0.7
228
242.3
9.7
317
9731
448
335
5.8
0.3
230
21.2
4.3
302
49.9
4.1
208
12.9
0.1
308
71
-------
listing of the site survey data, the testing results, and the well characteriza-
tion data for all 358 wells is presented in Table 8-8.
Two of the organic condensate samples were chosen for further characteriza-
tion by gas chromatography. The two samples selected both came from Kern River
field and the same producer but had widely differing steaming project ages.
The intent was to determine if the composition of the condensible organics
emitted from a well changes after an extended period of steaming. Figure
8-1 presents the results of that analysis in graphical form. Well number
82 is the newer well, having undergone 4 steaming cycles during a steaming
project 40 months old. Well number 77 that has been steamed more appears
to be emitting a lower molecular weight condensate, but it is difficult to
determine whether that might be due to the steaming history or other factors.
72
-------
TABLE 8-8. LISTING OF EMISSION AND CHARACTERIZATION DATA FOR INDIVIDUAL CYCLIC WELLS
NUMBER
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
FLOW
RATE
(L/MIN)
6.0
1.5
6.0
0.0
0.0
0.2
1.2
0.0
0.0
0.0
0.0
6.0
O.O
0.0
1.2
TOTAL
VOC STEAMING
EMISSIONS CYCLES
(LB/DAY) TO DATE
13.19 14
3.18 12
35.61 7
5
r m
18
f m
13
17
, f
13
11
10
2
2
TIME
SINCE
STEAMING
(DAYS)
279
259
190
368
m
124
421
4
.
463
165
410
380
231
FIELD: BELRIDGE
STEAMING SOAKING
FREQUENCY PERIOD
(MOS/CYCLE) (DAYS)
14.0
9.0
9.0
27. 0
t t
8.0
, .
8.0
8.0
f ,
8.0
5.0
8.O
8.0
2.0
STEAM
DOSAGE
(BBL/CYCLE)
8317
10516
11062
9636
f
1577
,
10099
11476
,
2538
7772
7323
7938
9677
OIL
PRODUCTION
RATE
(BBL/DAY)
3.0
13.0
39.0
2.0
t
4.0
,
11.0
20.0
.
17.0
31.0
18.0
12.0
9.0
CUMULATIVE
OIL
PRODUCTION
(BBL)
17910
54000
80730
8940
f
17780
t
39600
77970
.
62220
59520
38400
5760
,
GRAVITY
OF THE
OIL
(DEG API)
12.7
13.2
13.4
18.2
t
12.5
t
12.0
12.9
.
13.8
16.5
12.5
13.8
15.0
OIL TO
WATER
RATIO
0.1
0.0
0.7
0.0
f
0.0
t
0.0
O.2
.
0.2
1.9
0.1
1.0
1.1
Continued/
-------
TABLE 8-8. (Continued)
NUMBER
TOTAL TIME
FLOW VOC STEAMING SINCE
RATE EMISSIONS CYCLES STEAMING
(L/MIN) (LB/DAY) TO DATE (DAYS)
FIELD: CYMRIC
STEAMING SOAKING
FREQUENCY PERIOD
(MOS/CYCLE) (DAYS)
STEAM
DOSAGE
(BBL/CYCLE)
OIL
PRODUCTION
RATE
(BBL/DAY)
CUMULATIVE
OIL
PRODUCTION
(BBL)
GRAVITY
OF THE OIL TO
OIL WATER
(DEO API) RATIO
18
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
6.O
0.0
0.0
0.0
O.O
0.0
0.7
3.6
O.O
0.0
0.6
0.0
1.8
0.0
4.0
O.O
0.0
o.e
0.0
3.4
0.0
0.0
103.87
3
13
7
8
11
4
14
4
9
1
9
7
4
5
3
3
3
1
1
18O
130
327
169
247
153
468
400
385
187
321
43
267
344
406
265
7.O
14.0
24.0
20.0
16.0
29.0
11.0
B.O
lOiO
6.0
12.0
9.0
B.O
8.0
12.0
6.O
B.O
9
9
B
9
9
2
9
9
8
9
7
9
7
12400
11424
11109
8114
12369
12222
11038
9965
12103
2360
12199
11058
12423
13384
10396
9211
8200
7837
7998
3O.O
18.0
2.0
4.0
17.0
24.0
2.0
30.0
30.0
8.0
1B.O
7.0
40.0
4O.O
1B.O
12.0
12.0
3.0
30.0
2430O
102080
10920
20O4O
95370
87840
28250
28448
98300
1500
53100
14070
51800
48800
35100
10440
11180
1360
9000
12.0
12.0
13.5
15.8
11.9
12
12.0
14.7
12.2
13.0
13.3
11
12
12
11
13
12
12
13.0
O.I
0.3
0.0
0.7
0.7
0.1
0.4
0.1
0.8
0.0
2.5
O.I
0.8
0.2
0.0
O.4
2.5
Continued/
-------
TABLE 8-8. (Continued)
NUMBER
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
FLOW
RATE
(L/MIN)
5.0
6.0
2.4
4.0
0.0
0.0
0.0
0.6
0.0
0.0
3.0
4.0
1.2
3.0
6.0
5.0
TOTAL
VOC STEAMING
EMISSIONS CYCLES
(LB/DAY) TO DATE
1.43 1
1.80 1
1
15
11
3
11
8
8
7
2
2
1
1
2
1
FIELD: EDISON
TIME
SINCE STEAMING SOAKING
STEAMING FREQUENCY PERIOD
(DAYS) (MOS/CYCLE) (DAYS)
255
317
335
> • •
• • •
182
323
336
254
264
425
335
396
396
31
212
STEAM
DOSAGE
(BBL/CYCLE)
8600
11000
10000
7000
6500
9500
6500
7000
7300
6500
9800
7900
6300
6500
8500
9900
OIL
PRODUCTION
RATE
(BBL/DAY)
24.0
44.6
10.4
8.0
1.0
11.9
4.0
1.0
3.0
5.0
8.2
15.3
24.7
35. 3
38.4
23.0
CUMULATIVE
OIL
PRODUCTION
(BBL)
2100
8500
4000
72000
17000
23000
21600
12600
17640
19800
8700
16400
7500
12000
15000
4600
GRAVITY
OF THE
OIL
(DEQ API)
16.0
15.6
18.9
15.8
14.4
15.8
14.0
15.3
15.4
14.0
18.6
16.6
15.6
16.1
18.5
18.6
OIL TO
WATER
RATIO
88.3
26.4
32.5
97.8
91.7
83. 1
77.3
85.7
91.4
94.6
80.4
41.5
71.8
54.2
82.5
24.1
Continued/
-------
TABLE 8-8. (Continued)
NUMBER
54
55
58
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
TOTAL
FLOW VOC STEAMING
RATE EMISSIONS CYCLES
(L/MIN) (LB/DAY) TO DATE
2.0 0 22 15
6.O O
6.0 2
6.0 3
0.0
3.0
4.5
4.0
5.0
2.0
5.0
2.4
6.0
0.0
6.0
4.5
6.0
O.O
39
07 3
31 2
17
12
13
15
.
f
t
a
m
5
5
6
4
0.5 2
FIELD: KERN FRONT
TIME
SINCE STEAMING SOAKING STEAM
STEAMING FREQUENCY PERIOD DOSAGE
(DAYS) (MOS/CYCLE) (DAYS) (BBL/CYCLE)
408 13962
52
181
219
147
336
473
391
233
21
79
B
23
26
59
386
267
278
191
5317
12209
15319
1762O
13176
14096
11689
7756
5391
8609
5919
5862
5516
14755
1363B
11852
149BB
116O8
OIL CUMULATIVE GRAVITY
PRODUCTION OIL OF THE
RATE PRODUCTION OIL
(BBL/DAY) (BBL) (DEO API)
9.0 13. B
11.0
13.0
17.0
7.0
19.0
10.0
6.0
3.0
10.0
12.0
16.0
B.O
11.0
41.0
9.0
9.0
1B.O
6.0
13.0
13.8
12. B
13.6
14. 0
13.0
14.0
13. 0
13.0
13.0
13.0
13.0
13.0
12. B
12.8
12.8
12.7
12.6
OIL TO
WATER
RATIO
0.1
oil
0.1
0.1
0.1
0.3
O.1
.
.
O.B
0.2
0.1
0.3
0.0
Continued/
-------
TABLE 8-8. (Continued)
NUMBER
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
FLOW
RATE
(L/MIN)
4.0
1.5
3.0
0.6
6.0
6.0
6.0
6.0
6.0
6.0
6.0
0.0
0.0
0.6
0.0
0.0
O.O
0.6
4.0
1.5
6.0
1.5
0.0
1.2
0.0
4.0
2.4
2.0
0.0
0.0
1.5
VOC
EMISSIONS
(LB/DAY)
0.51
1.74
0.69
0.24
2.86
0.84
4.46
2.16
25.89
40.95
0.54
.
t
.
p
.
,
p
.
.
.
.
TOTAL
STEAMING
CYCLES
TO DATE
15
19
4
17
12
4
8
10
3
4
5
11
15
11
20
4
2
10
7
15
13
12
12
12
24
6
9
4
8
6
5
FIELD: KE
TIME
SINCE STEAMING
STEAMING FREQUENCY
(DAYS) (MOS/CYCLE)
443
411
112
171
269
206
30
370
38
83
141
727
135
280 12.0
276 6.0
265
86
250
105
236
1353
886
294
147
318 6.0
284
280
301
208
378
292
RN RIVER
SOAKING STEAM
PERIOD DOSAGE
(DAYS) (BBL/CYCLE)
659
640
8200
7000
11032
11310
14817
6500
12628
16023
13236
13159
6000
3 4000
3 3000
3000
6250
7000
7200
8354
10595
1 1 100
10811
11268
3 4000
11061
10055
10308
10353
18890
15078
OIL
PRODUCTION
RATE
(BBL/DAY)
8.0
17.0
20.0
10.0
9.0
2.0
9.0
10.0
7.0
9.0
13.0
7.0
5.0
2.0
2.0
4.0
1.0
8.0
20.0
2.0
9.0
12.0
8.0
18.0
2.0
1.0
13.0
5.0
11.0
19.0
22.0
CUMULATIVE
OIL
PRODUCTION
(BBL)
B
25200
72000
,
.
f
17280
.
f
f
f
28800
t
t
10000
350
72000
72000
GRAVITY
OF THE
OIL
(DEG API)
13.6
13. B
13.8
13.1
,
.
.
12.2
,
,
t
.
12.0
14.0
13.0
12.0
13.0
12.7
13.2
,
.
.
.
t
14.0
,
.
,
.
OIL TO
WATER
RATIO
0.0
10.3
94.1
97.0
44.0
96.0
73.0
87.3
92.0
97.0
90.0
62.0
96.0
64.0
98.0
95.0
91.2
96.8
90.9
89.0
52.0
62.0
63.0
44.0
70.0
99.0
58.0
97.0
53.0
66.0
48.0
Continued/
-------
TABLE 8-8. (Continued)
NUMBER
104
1O5
108
107
108
109
110
111
112
113
114
TOTAL
FLOW VOC STEAMING
RATE EMISSIONS CYCLES
(L/MIN) (LB/DAY) TO DATE
2.4 B
1.2
4.O
O.O
0.0
0.0
0.8
0.0
0.0
O.O
2.0
4
3
3
.
t
f
1
1
1
2
FIELD: KERN RIVER
TIME
SINCE STEAMING SOAKING STEAM
STEAMING FREQUENCY PERIOD DOSAGE
(DAYS) (MOS/CYCLE) (DAYS) (BBL/CYCLE)
230 1 1039
414
181
BBS 8
m
(
f
277
272
291
8B
o :
16828
18903
» 3OOO
,
1B92
28112
27B1B
12891
6400
OIL CUMULATIVE GRAVITY
PRODUCTION OIL OF THE
RATE PRODUCTION OIL
(BBL/DAY) (BBL) (DEO API)
13. 0
6.0
16.0
1.0
4.0
1.B
3.0
14.0
12.0
19.0
t
is!o
13.0
13.0
13.0
_
.
t
2.0 2400 12. B
OIL TO
WATER
RATIO
68. 0
81. 0
66.0
99. 0
m
t
f
BI!O
36.0
BB.O
98.8
--.I
CO
Continued/
-------
TABLE 8-8. (Continued)
FIELD: LOST HILLS
TOTAL TIME OIL CUMULATIVE GRAVITY
FLOW VOC STEAMING SINCE STEAMING SOAKING STEAM PRODUCTION OIL OF THE OIL TO
RATE EMISSIONS CYCLES STEAMING FREQUENCY PERIOD DOSAGE RATE PRODUCTION OIL WATER
NUMBER (L/MIN) (LB/DAY) TO DATE (DAYS) (MOS/CYCLE) (DAYS) (BBL/CYCLE) (BBL/DAY) (BBL) (DEG API) RATIO
115 4.8 . 1 243 . . 3981 10.0 3000 1B.8 0.8
^ Continued/
so
-------
TABLE 8-8. (Continued)
00
o
NUMBER
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
148
FLOW
RATE
(L/MIN)
e.o
B.O
2.8
6.0
1.2
1.8
6.0
1.2
O.3
0.5
0.0
0.0
0.0
O.O
2.4
O.O
0.0
O.8
O.O
0.5
2.4
0.0
0.6
O.O
0.0
2.8
0.5
0.0
0.0
O.7
1.8
TOTAL TIME
VOC STEAMING SINCE
EMISSIONS CYCLES STEAMING
(LB/DAY) TO DATE (DAYS)
16.01
2.48 IB
1.73
88.58 4
0.21 3 170
2 . 37 2
8 . 37 1
8 342
• • •
15
9
10
8
14
16
14
18
14
14
13
7
14
12
10
13
9
11
6 407
4 183
8
5 277
FIELD: MC
STEAMING
FREQUENCY
(MOS/CYCLE)
io!o
.
e.o
B.O
e.o
e.o
19. 0
t
e.o
17.0
e.o
e.o
12. 0
10.0
11. 0
10.0
12.0
e.o
e.o
6.0
10B
115
13. 0
6.0
e.o
11.0
18.0
28.0
12. 0
14.0
KITTRICK
SOAKING
PERIOD
(DAYS)
2
t
2
<
2
2
9
,
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
9
9
2
9
STEAM
DOSAGE
(BBL/CYCLE)
11200
,
1OOOO
10314
1OOOO
10420
7890
.
10000
8630
1OOOO
10000
9278
10690
11 ISO
10900
10864
10OOO
12000
10OOO
12389
12600
11B30
10000
10000
12500
7O8B
8300
11BOO
11 446
OIL
PRODUCTION
RATE
(BBL/DAY)
ia!o
.
8.0
18.0
33.0
4.0
11.0
.
4.0
1.0
,
9.0
2.0
17.0
10. 0
22.0
8.0
6.0
24.0
1.0
24.0
10.0
7.0
B.O
8.0
7.0
3.0
3.0
16.0
7.0
CUMULATIVE
OIL
PRODUCTION
(BBL)
81360
t
10853
10800
9302
1064
41910
,
181731
B1O42
6O07O
BB273
104231
121904
115257
134702
185153
117405
10B728
17206
101552
90896
53281
70802
13B28
79201
10890
10800
B7718
J7010
GRAVITY
OF THE
OIL
(DEO API)
13.0
f
13. 0
13.0
13. 0
13. 0
IB. 8
,
13.0
13. 0
13.0
13.0
13.0
13.0
13.0
13.0
13.0
13.0
13.0
13.0
13.0
13.0
13.0
13.0
13. 0
13.0
11.7
12.9
13.0
11.8
OIL TO
WATER
RATIO
o'.e
<
O.B
1.1
0.2
O.B
2.8
,
1.6
0.7
1.0
O.B
1.0
0.8
1.0
1.0
0.9
O.9
0.6
O.B
o.e
0.7
O.B
0.8
0.3
0.3
0.4
0.2
0.6
1.O
Continued/
-------
TABLE 8-8. (Continued)
FIELD: MCKITTRICK
NUMBER
147
148
149
150
151
152
153
154
155
158
157
158
159
160
FLOW
RATE
(L/MIN)
0.0
0.0
0.6
0.4
1.6
0.0
1.8
O.O
0.0
1.2
0.9
0.0
2.5
0.0
TOTAL
VOC STEAMING
EMISSIONS CYCLES
(LB/DAY) TO DATE
7
4
5
6
5
2
4
4
3
2
2
2
1
2
TIME
SINCE
STEAMING
(DAYS)
790
.
,
t
.
,
.
344
303
STEAMING
FREQUENCY
(MOS/CYCLE)
9.0
15.0
B.O
B.O
11.0
8.0
B.O
13.0
B.O
5.0
6.0
6.0
8.0
6.0
SOAKING
PERIOD
(DAYS)
9
2
2
2
2
2
2
2
9
t
2
2
2
2
STEAM
DOSAGE
(BBL/CYCLE)
9610
8808
10000
10000
12172
10000
10000
10896
7982
10680
10000
10OOO
9440
10000
OIL
PRODUCTION
RATE
(BBL/DAY)
2.0
25.0
32.0
15.0
26.0
8.0
7.0
16.0
1B.O
11.0
B.O
4.0
12.0
10.0
CUMULATIVE
OIL
PRODUCTION
(BBL)
6100
13788
36085
26083
36085
B502
6920
7022
13500
8930
2B47
1375
1763
302
GRAVITY
OF THE
OIL
(DEG API)
12.9
13.0
13.0
13.0
13.0
13.0
13.0
13.0
13.8
13.0
13.0
13. 0
13.0
13.0
OIL TO
WATER
RATIO
0.2
0.5
0.8
0.8
0.5
0.9
0.4
0.4
0.0
0.4
0.4
0.0
0.6
0.9
Continued/
-------
TABLE 8-8. (Continued)
CO
FIELD: MIDWAY-SUNSET
NUMBER
161
182
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
FLOW
RATE
(L/MIN)
"B.O
0.8
0.9
0.9
2.4
1.2
1.2
6.0
1.5
1.0
6.0
1.2
6.0
1.2
6.0
6.0
1.2
1.2
1.8
1.2
2.2
6.0
6.0
4.0
6.0
1.2
2.8
2.5
1 .8
1.2
6.0
VOC
EMISSIONS
(LB/DAY)
33.79
1.21
3.23
1.84
0.71
1.64
O.01
26.91
5.88
4.34
141.49
6.78
21.91
2.07
10.07
0.79
8.17
4.0O
1.28
1.28
12.08
4.61
4.05
7.31
3.40
3.59
1.21
2.37
1.28
11.92
45.74
TOTAL
STEAMING
CYCLES
TO DATE
10
1
24
9
25
5
20
8
11
IB
7
2
19
15
8
14
11
6
7
8
7
6
4
8
4
4
2
4
1
2
1
TIME
SINCE
STEAMING
(DAYS)
502
330
125
295
405
36O
,
15
97
.
t
311
31
165
224
158
76
157
188
178
13
338
82
274
28
464
381
144
251
115
218
STEAMING
FREQUENCY
(MOS/CYCLE)
15. 0
(
11.0
10.0
7.0
24.0
8.4
3.O
10. 0
9.0
17. 0
<
B.O
8.9
11.0
12.0
6.0
10. 0
8.0
11.0
9.0
6.0
9.0
,
8.5
4.0
12.0
B.O
0.0
8.0
.
SOAKING
PERIOD
(DAYS)
9
10
0
B
3
9
3
5
9
7
8
19
9
3
7
7
B
8
B
4
0
7
3
8
3
B
0
0
3
9
9
STEAM
DOSAGE
(BBL/CYCLE)
11642
12183
11200
1848
8209
1BOOO
4878
2654
12454
8080
9OOO
82089
11077
5018
7000
11228
BOOO
800O
11928
129OO
1250O
9000
5497
9340
4773
2847
12200
10500
8454
1500O
8527
OIL
PRODUCTION
RATE
(BBL/DAY)
29.0
9.0
21.0
4.0
t
3O.O
f
18.0
22. 0
29.0
1O. O
13.0
22.0
.
20. 0
38. 0
2B.O
4.0
17.0
6.0
18.0
9.0
.
23.0
.
8.0
18.0
B.O
,
2.O
3.O
CUMULATIVE
OIL
PRODUCTION
(BBL)
134444
€
279938
10800
f
1BOOOO
.
1O8OO
738 1O
111380
,
3388
88890
.
,
1764OO
38930
,
30090
151852
237B8
.
.
7333
.
2880
443B
4078
.
460
311
GRAVITY
OF THE
OIL
(DEG API)
13.6
12.3
13.7
12.9
12.0
11.4
12. 0
13. B
14.1
t
13.4
12.8
11.9
12.0
10. B
11.0
12.0
11.2 '
11.0
13.7
13.2
12. B
12.0
12.8
12.0
14.8
13.7
13.7
12. 0
11.4
12.0
OIL TO
WATER
RATIO
1.8
1.B
0.1
0.1
t
0.7
t
2.B
0.4
1.8
0.3
1.3
O.8
.
0.1
3.B
2.1
4.0
2.1
o.a
0.7
0.8
.
0.8
.
0.0
3.8
0.1
.
0.2
0.9
Continued/
-------
TABLE 8-8. (Continued)
FLOW VOC
RATE EMISSIONS
NUMBER
192
193
194
195
193
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
2t7
218
219
220
221
222
(L/MIN) (LB/DAY)
0.
0.
0.
0.
0.
0.
0.
0.
0.
1.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
o.
0.
1.
0.
0.
0.
0.
0.
2.
0
0
6
0
0
0
0
0
0
2
0
0
0
0
0
0
6
4
0
0
0
0
0
0
0
0
6
0
0
0
5
TOTAL
STEAMING
CYCLES
TO DATE
6
29
6
7
4
10
3
S
2
18
21
17
5
8
11
22
20
29
m
IB
14
11
16
17
16
,
.
20
TIME
SINCE
STEAMING
(DAYS)
42
.
141
468
427
158
319
1372
330
402
196
137
85
265
42
242
,
286
366
89
.
151
295
436
317
194
420
338
365
197
FIELD: MIDWAY-SUNSET
STEAMING SOAKING STEAM
FREQUENCY PERIOD DOSAGE
CMOS/CYCLE)
6.0
8^4
12. 0
8.0
8.0
11.0
19.0
17.0
12.0
14.0
9.0
3.0
6.0
4.0
12.0
%
B.3
9.4
6.5
1l!o
6.O
10.0
6.0
10.0
11.0
f
.
5.0
(DAYS)
S
_
3
9
9
11
10
10
10
5
0
0
5
5
5
7
3
3
3
6
16
12
7
7
6
4
4
4
(BBL /CYCLE)
1217
4880
10370
10588
12200
1066S
15366
12497
11000
11200
11900
11642
2889
2660
6000
7073
7597
5814
_
9113
7998
12118
8054
5250
8060
10987
10646
3931
6209
OIL
PRODUCTION
RATE
(BBL/ DAY)
6.0
.
0.0
16.0
21.0
50.0
7.0
63.0
1.0
58.0
9.0
37.0
18.0
4.0
15. 0
t
t
_
42 .'o
15.0
30.0
41.0
23.0
12.0
22.0
1.0
3.0
23.0
CUMULATIVE
OIL
PRODUCTION
(BBL)
8480
.
1921B
28792
24979
178425
23912
163327
1129
235433
182572
S66 10
16200
3840
_
.
t
211170
92031
•4237
78272
120060
60480
.
.
f
70000
GRAVITY
OF THE
OIL
(DEG API]
13.6
12.0
13.8
12. S
13.3
14.5
15.1
14.8
17.1
13.7
13.7
11.0
13.6
13.5
13.6
12^0
12.0
12.0
12.' 8
12.0
12.0
12.0
m
11.3
11.3
12.6
11.8
OIL TO
WATER
1 RATIO
0.1
0.1
0.1
0.1
0.8
0.0
0.4
0.0
0.4
0. 1
3.7
0.5
0.1
1.7
.
.
o!s
0.3
0.5
0.7
2.1
4.0
1.6
0.0
2.0
3.B
Continued/
-------
TABLE 8-8. (Continued)
00
FIELD: MIDWAY-SUNSET
NUMBER
223
224
225
228
227
228
229
230
231
232
233
234
235
238
237
238
239
240
241
242
243
244
245
248
247
248
249
250
251
252
253
FLOW
RATE
(L/MIN)
O.O
2.5
O.8
0.0
O.B
0.0
O.O
O.5
0.0
0.0
0.0
0.0
1.2
0.0
1.2
0.0
O.8
0.5
0.5
0.0
0.0
0.8
4.5
0.0
0.0
8.0
0.0
3.8
1.8
0.0
1.2
TOTAL
VOC STEAMING
EMISSIONS CYCLES
(LB/DAY) TO DATE
7
15
9
9
9
9
18
11
24
19
14
25
,
20
_
'. 12
11
5
20
18
20
19
18
8
7
8
3
17
t
7
TIME
SINCE
STEAMING
(DAYS)
473
323
284
298
179
172
288
221
22
257
253
197
4
85
.
21
273
1169
214
205
148
271
11
42
425
12
73
288
4
t
337
STEAMING
FREQUENCY
(MOS/CYCLE)
14. 0
8.0
12. 0
12.0
14.0
9.0
8.O
15.0
8.9
12.0
12.0
8.4
_
8.0
.
18.0
13.0
24.0
7.3
8.0
7.0
7.0
8.0
4.O
4!o
3.0
24.0
8.1
t
12. 0
SOAKING
PERIOD
(DAYS)
8
7
7
7
B
11
12
B
B
B
3
t
B
.
10
8
7
3
3
B
7
3
B
4
B
B
B
B
,
7
STEAM
DOSAGE
(BBL/CYCLE)
10OOO
B430
7OOO
7000
7OOO
BOOO
8S8B
2907
4209
8748
10290
6078
f
11418
t
70000
8000
7OOO
4718
6824
10234
4600
9849
240O
8173
1818
2846
2500
6087
,
7500
OIL
PRODUCTION
RATE
(BBL/DAY)
18.0
12.0
1B.O
2.O
41.0
0.4
49. 0
33.0
|
10.0
60.0
.
,
BO.O
.
34.0
30.0
78. 0
.
,
28. 0
1B.O
,
1O. O
18.0
2.0
7.0
B.O
.
.
11.0
CUMULATIVE
OIL
PRODUCTION
(BBL)
8BB69
3O399
,
,
,
20734
162351
141118
.
103234
289800
.
.
243000
,
,
,
*
,
.
10B7BO
B9860
.
980O
,
1880
3780
40800
•
.
86800
GRAVITY
OF THE
OIL
(DEG API)
11.9
11.9
13.2
10.7
12.8
12.0
12.8
12.8
12.0
12.8
11.0
12.0
.
11.0
.
12.1
12. B
12.4
12.0
12.0
11.0
.
12.0
13.4
12.8
13.4
13. B
13.0
12.0
.
13.0
OIL TO
WATER
RATIO
0.8
0.3
0.8
0.7
B.8
0.0
O.B
0.8
.
O.7
1.8
,
,
2.0
.
2.8
3.3
2.8
.
.
2.B
7.6
.
1.O
0.9
0.3
0.2
0.3
•
.
O.B
Continued/
-------
TABLE 8-8. (Continued)
CO
Ui
NUMBER
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
FLOW VOC
RATE EMISSIONS
(L/MIN) (LB/DAY)
0
5
1
0
4
2
6
0
0
2
0
0
0
1
0
0
1
0
0
0
0
0
0
0
0
0
• 1
0
0
0
2
.6
.0
.5
.0
.O
.4
.0
.0
.0
.4
.0
.0
.0
.5
.7
.0 i
.5
.0
.0
.0
.6
.0
.0
.O
.0
. 1
.8
.0
.0
.0
.4
TOTAL
STEAMING
CYCLES
TO DATE
5
9
12
B
11
8
16
7
12
9
8
13
13
12
6
9
9
3
10
. 7
1O
8
17
f
13
8
9
2
5
2
TIME
SINCE
STEAMING
(DAYS)
223
251
287
.
129
63
55
415
158
418
263
191
64
358
274
344
320
257
277
176
135
t
141
168
188
310
149
392
FIELD: MIDWAY-SUNSET
STEAMING SOAKING STEAM
FREQUENCY PERIOD DOSAGE
(MOS/CYCLE) (DAYS) (BBL/CYCLE)
15.0
13.0
10.0
13.0
9.0
11.0
8.0
11.0
8.0
9.0
11.0
7.0
8.7
7.0
8 . 7
9.0
sio
11.0
8.0
4.0
5.0
8.0
9.1
8.0
,
8.0
B
10
9
7
3
7
5
5
7
10
13
5
3
9
m
3
8
9
5
3
22
5
a
5
3
3
9
B
13
2000
9000
11932
7000
6059
9000
5000
3089
9278
10058
4394
5000
8461
10234
82223
8538
7780
13267
5000
8132
9299
2402
5000
14698
8375
6892
9965
2424
51232
OIL
PRODUCTION
RATE
(BBL/DAY)
1 .0
is!o
89.0
29.0
13*0
54 .'O
3S.O
80.0
9.0
14.0
32.0
is!o
78 !o
7!o
45.0
128o!o
9.0
35.0
10. 'o
14*0
uio
43. 0
CUMULATIVE
OIL
PRODUCTION
(BBL)
'
320311
175387
80850
94228
80451
9238O
112971
378OO
215574
2057
105749
75115
8640
60881
23400
3049
12800
13083
GRAVITY
OF THE
OIL
(DEQ API!
13.0
13.2
13.0
11.0
12.0
14.8
12.0
12.7
12.0
12.0
12.0
12.0
12.0
11.8
12.5
12.0
12.0
13.0
12.0
12.0
12.8
15.1
12.0
11.0
12.0
12.0
12.8
14. S
12.8
OIL TO
WATER
) RATIO
1 .0
0.4
3.9
3.8
0.8
0.0
0.7
0.3
0.3
0.8
0.0
O.4
4.1
0.4
0.0
o.a
1.3
O.O
I'D
i'o
0.6
1.3
Continued/
-------
TABLE 8-8. (Continued)
oo
NUMBER
285
288
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
308
307
308
309
310
311
312
313
314
315
FLOW VOC
RATE EMISSIONS
(L/MIN) (LB/DAY)
0.8
2.4
0.0
O.O
0.0
0.6
0.0
0.0
0.8
O.O
0.0
O.O
8.0
0.8
0.0
0.0
3.0
0.0
1.8
0.0
O.O
0.0
0.8
0.0
0.0
O.O
0.9
0.6
0.6
0.0
0.0
TOTAL
STEAMING
CYCLES
TO DATE
17
10
10
16
9
6
6
3
9
7
t
6
8
.
t
7
B
.
7
_
7
8
4
3
3
4
6
3
2
1O
TIME
SINCE
STEAMING
(DAYS)
113
50
41
9O
49
225
120
271
174
410
.
372
408
412
,
1
258
198
.
157
_
279
254
257
245
177
419
315
347
42
FIELD: MIDWAY-SUNSET
STEAMING SOAKING STEAM
FREQUENCY PERIOD DOSAGE
(MOS/CYCLE) (DAYS) (BBL/CYCLE)
4.0
7.0
8.0
4.0
8.0
10.0
11.0
p
7 ,o
9.0
.
9.0
10.0
t
.
^
7.0
9.0
t
B.O
8.0
B.O
12.0
11.0
14.0
B.O
8.0
12.0
18.0
4.0
B
,
18
B
B
8
1O
11
B
0
6
7
7
t
,
3
11
t
B
(
9
11
0
7
10
7
12
10
B
B
5000
3420
12100
500O
10217
8000
10000
38891
11928
11700
4000
7OOO
7000
1O202
.
.
5933
11932
.
2420
,
5970
6003
9074
10000
10OOO
7000
9599
9000
7124
2285
OIL
PRODUCTION
RATE
(BBL/DAY)
28.0
44.0
11.0
20.0
33.0
33.0
1B.O
29.0
18. 0
18.0
3.0
B.O
2.0
7.0
.
.
,
28. 0
.
23.0
.
10.0
2B.O
8.0
10.0
38. 0
10.0
4.0
8.0
,
48. 0
CUMULATIVE
OIL
PRODUCTION
(BBL)
38937
98380
74582
49344
55440
t
.
B882
28SBO
30123
,
,
.
.
.
.
,
4B2B2
.
241BO
.
48422
33804
23058
.
.
.
11430
.
.
B4OOO
GRAVITY
OF THE
OIL
(DEQ API)
12.0
• (
13.4
12.0
11.0
11.2
11.2
13.2
11. 0
13.4
13. B
11. B
11. B
11.3
.
.
12.0
14.4
.
13. B
.
12.0
12.0
13.7
11.4
10.9
11.4
11.6
12. B
12.0
14.1
OIL TO
WATER
RATIO
0.0
1.3
0.8
0.0
2.3
11.0
O.B
0.7
2.1
1.B
0.3
0.9
2.0
0.3
.
.
.
0.1
.
1 .8
.
0.3
0.2
0.3
B.O
1.4
B.O
0.9
0.4
.
O.B
Continued/
-------
TABLE 8-8. (Continued)
FLOW VOC
RATE EMISSIONS
NUMBER
318
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
(L/MIN) (LB/DAY)
1
1
1
0
0
0
0
0
o
0
o
0
2
0
0
0
0
0
1
0
0
0
0
6
O
0
6
0
0
0
0
.2
.2
.8
.0
.6
.0
.0
.0
.0
.0
.0
.8
.0
.0
.0
.0
.0
.0
.2
.0
.0
.0
.0
.0
.0
.0
.0
.0
.0
.0
.0
TOTAL
STEAMING
CYCLES
TO DATE
5
5
2
4
3
3
3
3
4
6
4
6
6
2
4
1
4
3
1
2
1
3
2
f
1
1
1
4
4
TIME
SINCE
STEAMING
(DAYS)
119
24
348
347
IS
110
366
303
447
.
100
144
114
77
299
91
271
203
176
101
106
251
S3
231
241
270
290
87
74
219
FIELD: MIDWAY -SUNSET
STEAMING SOAKING STEAM
FREQUENCY PERIOD DOSAGE
(MOS/CYCLE)
7.4
6.8
13.0
s!o
11.0
4.0
6.0
10.0
5.0
4.0
9.O
6.0
5.0
5.0
3.5
3.0
6.0
5.0
7.5
s'.o
3.O
,
t
6.0
3.0
2.O
(DAYS)
3
4
3
5
3
11
0
0
0
7
f
5
5
5
5
3
3
23
20
0
3
3
11
5
5
11
12
11
3
7
5
(BBL/CYCLE)
71S3
7892
6032
15377
5902
7400
11700
12000
11800
10000
12000
1795
5083
3964
5000
6532
5797
86181
6978
11900
10944
7560
7531
3879
2648
13623
11528
12894
7197
7830
8045
OIL
PRODUCTION
RATE
(BBL/DAY)
t
.
17.0
f
2.0
59. 0
10.0
17.0
33.0
5.0
15.0
37. 0
18.0
24.0
f
f
20.0
17.0
7.0
.
10.0
25.0
122.0
18. 0
20.0
18.0
f
83.0
8.0
CUMULATIVE
OIL
PRODUCTION
(BBL)
.
,
978
1009
5087
7623
.
_
10800
39980
11520
12198
,
.
3883
8768
3607
_
2084
11250
21960
3279
4469
3571
t
332
1820
GRAVITY
OF THE
OIL
(DEC API)
12. 0
12.0
12.0
11.9
12.0
13.3
13.7
13.8
13.7
10.8
10. S
14.7
18. 0
15.0
12.0
12.0
12.0
11.5
12.0
13.7
12. 0
12.0
12.9
14.2
14.6
11.8
12.9
11.5
12.0
11.8
11.8
OIL TO
WATER
I RATIO
€
o!e
.
0.3
0.5
0.5
2.5
1.6
0.5
0.1
3. 1
2.0
0.0
,
.
0.3
0.7
0.3
O!B
0.3
3.6
1.2
2.0
0.8
.
0.5
0.8
Continued/
-------
TABLE 8-8. (Continued)
FIELD: MIDWAY-SUNSET
TOTAL TIME OIL CUMULATIVE QRAVITY
FLOW VOC STEAMING SINCE STEAMING SOAKING STEAM PRODUCTION OIL OF THE OIL TO
RATE EMISSIONS CYCLES STEAMING FREQUENCY PERIOD DOSAGE RATE PRODUCTION OIL WATER
1°.?*!!. (B.L/DAY)
-------
TABLE 8-8. (Continued)
FIELD: POSO CREEK
NUMBER
350
351
352
353
354
355
356
357
358
359
FLOW
RATE
(L/MIN)
0.3
0.0
0.0
6.0
0.0
2.0
0.5
0.4
0.0
5.0
TOTAL
VOC STEAMING
EMISSIONS CYCLES
(LB/DAY) TO DATE
4
4
4
4
3
2
3
3
2
1
TIME
SINCE STEAMING
STEAMING FREQUENCY
(DAYS) (MOS/CYCLE)
141
f t
t t
t t
214
184
201
82
113
169
SOAKING STEAM
PERIOD DOSAGE
(DAYS) (BBL/CYCLE)
9002
3000
3000
8816
10216
10843
13951
10937
9212
1O204
OIL
PRODUCTION
RATE
(BBL/DAY)
8.0
6.0
10.0
11.0
9.0
14.0
8.0
5.0
4.0
45.0
CUMULATIVE GRAVITY
OIL OF THE
PRODUCTION OIL
(BBL) (DEG API)
12.5
2500 12.7
B100 13.0
12.9
12.9
12.7
12.5
12. 5
12.9
12. 5
OIL TO
WATER
RATIO
O.O
0.0
0.0
0.1
0.0
0.0
0.0
0.0
0.0
0.2
CO
Footnote:
Well //146 was not included in the study.
-------
Well No. 82
Well No. 77
O
O
0
C7 C9 C11 C13 C15 C17 +
CARBON NUMBER
I
u.
O
C7 C9 C11 C13 C15 C17 +
CARBON NUMBER
98° 151° 196° 234° 270° 302°+ 98° 151V 196" 234° 270° 302° +
BOILING POINT (9C) BOILING POINT («C)
Figure 8-1. Condensate characterization.
-------
SECTION 9
CORRELATION STUDIES
It was desired to determine if any strong correlations existed between
the rate of VOC emissions from a well and any of its physical characteristics
or operating practices. This was attempted by plotting VOC emissions versus
each well parameter, performing correlation analysis, and finally, multiple
regression analysis. It was evident from the results of these efforts that
VOC emissions are affected by so many variables in such a complex manner that
no clear-cut correlations could be developed.
The remainder of this section presents the results of the correlation
studies. The graphical presentations indicate some logical trends, but as
the numerical analyses indicate, there is too much scatter (caused by variable
interdependency) to quantify the relationships.
9.1 CORRELATIONS BETWEEN SURVEY PARAMETERS
One of the first steps in this effort was to check for correlations
between the variables which characterize the well. Such interdependency of
variables could mask potential correlations to the VOC emissions. Table
9-1 presents the results of this check in the form of paired variables which
have a significant correlation coefficient.
9.2 CORRELATION OF VOC EMISSIONS
The objective of this portion of the study is to relate VOC emissions
to other characteristics of the well. Figures 9-1 through 9-7 present plots
of VOC emissions against well characteristic data available. The plotting
9.1
-------
TABLE 9-1. CORRELATION COEFFICIENTS* FOR THE SURVEY DATA
Variable Pair
1. Age of Steaming
2. No. of Cycles
3. No. of Cycles
4. No. of Cycles
5. Frequency
6. Frequency
7. Steam Dosage
8. Soaking Period
9. Age of Steaming
10. Flow Rate
11. Time Since Last
Steaming
12. Time Since Last
Steaming
Frequency
Steam Dosage
Soaking Period
Cumulative Production
Steam Dosage
Soaking Period
Soaking Period
Production Rate
Cumulative Production
No. of Cycles
Frequency
Soaking Period
Pearson
Correlation
Coefficient
0.41
-0.15
-0.18
0.65
0.17
0.22
0.38
0.29
0.65
-0.13
0.23
0.16
Sample
Size
204
317
222
207
228
194
230
190
177
317
219
219
*0nly coefficients significant at 95% or higher
92
-------
i ono
lino
cd
jz
0)
C
o
0]
-H
I '°
r «• -0.15
p ~ 0.31
n - 50
t 3
3 3<*
? "»
144
33
3
30 tCO 150 200 ZSO SOO 350
Time since last steaming (days)
tso soo
-•+-------4.
550 600
Figure 9-1. VOC emissions vs. time since last steaming.
-------
1000
100
OJ
^3
CO
C
O
W 10
CO
•H
W
u
o
14
l"»
12
4
r •» -0.14
p - 0.32
n " 51
"3*32 3 3
, a « - 3 a
3 323
-4 — 4— 4 — 4 — - + -~ + - .. + — 4...4— .4 — + - —+ —. +-..+--_+. — 4.__ + ._. + 4...4... + ...+ ...*... + .
1 ? 3 "» S fc 7 B 9 10 11 12 13 1« IS 16 17 18 i» 20 2» 22 23 2H 25
Number of Cycles Since Steaming Began
Figure 9-2. VOC emissions vs. number of cycles.
-------
I
1000 +
100
SO
Ul
a
ft
CO
en
•H
r = 0.04
p = 0.78
n = 52
10
u
o
3
3 3
3
3 H 2
3 2
3 «•
<4 3<4 3
(*•» 2 <«
3 3
?000
6000 8000 10000 12000 1UOUO 16000
Steam Dosage (Bbl. /Cycle)
laouu
Figure 9-3. VOC emissions vs. steam dosage.
-------
1000 t
tfl
T3
O
•H
M
I
U
O
100
10
23
0 4
_*_-„---.--+.
0 5
« (.
3 2
5 ,
3 t
3
-------
vo
1000
100
w
C
o
•H
u
w
•H
s
It
g 10 3 43
3
3 t
OH 2
3 3
3
jnonn
• + -.•
31
r = -0.10
p = 0.61
n = 30
guoon voooo J?OODO isoouo isoono piooou JIUQOO 270000 300000
Cumulative Oil Production (Bbl.) f
Figure 9-5. VOC emissions vs. cumulative oil production.
-------
I
1II (10 »
nl
-a 100
s
CO
c
£ .3
tn
CO
-H
§
u 10
o
0 <
1(
r - -0.02
p - 0.89
n - 46
„
M
If
H 4
4 4
<* 3 43
3
3
3
3 4
4432 3
333 243 44
<4 32 3^3
^ _ 3 423
).P 10.5 11.0 11.5 12.0 13. 5 13. 0 13. t) 14, U !•»,& 15.0 15.5 1*>.0
Figure 9-6. VOC emissions vs API gravity of the oil.
-------
1000
V 180-
O
c
E
M
I
S
S
I
o
N
S
L
B
10-
0
XXX
10 23-30
40 S3 60
FUOW RATE L/MIN
r = -0.15
p = 0.31
n = 50
70
•rt | i 't f-i *'ri 'i-i | i fi f • n ft ry
80 93 ISO
Figure 9-7. VOC emissions vs. survey flow rate.
-------
symbols 2, 3, and 4 represent the flow rate category from the preliminary survey.
Although some vague trends are recognizable, the scatter effectively prohibits
any strong conclusions from this size of data set.
The variable "VOC emissions" was added to the list of variables tested
for correlation in the previous subsection, and the results were presented
in Table 9-2. Most of the same variable pairs repeated their significant
correlation. VOC emissions was found to correlate significantly with only
the survey flow rate (which, was not a true well characteristic, but only a
rougher measurement of the emission rate).
9.3 REGRESSION ANALYSIS ON TESTED DATA
Multiple regression analysis was used to evaluate the combined effects
of all the possible independent variables upon the logarithm of the VOC
emissions as the dependent variables. A logarithmic transformation of the
VOC emissions was used due to the lack of normality of this variable. Many
responses to the questions about well characteristics were missing, so the
analysis involved a trade-off between sample size for the analyses and
inclusion of some of the variables.
A "dummy variable" was created to denote the area of the well. It was
coded as a "1" if the well was in the western portion of the county and a
"0" otherwise. After a series of models were evaluated, the most important
two variables appear to be the flow rate and the variable distinguishing west
and central wells. Table 9-3 shows the regression model using the log (VOC
emissions) as the dependent variable. The R2 value or multiple correlation
coefficient is 0.37. This says that 37% of the variability in the data is
accounted for with this model. The addition of other variables made only a
negligible improvement in this. The remaining variability must be explained
by one or more factors that were not measured in this study.
An additional analysis was carried out for Western wells only since a
few of the well characteristics were only available for this group. The model
100
-------
TABLE 9-2. CORRELATION COEFFICIENTS* FOR DATA ON WELLS TESTED
Variable Pair
1. Age of Steaming
2. Age of Steaming
3. No. of Cycles
4. No. of Cycles
5. Frequency
6. Frequency
7. Production Rate
8. VOC emissions
Frequency
Cumulative Production
Steam Dosage
Cumulative Production
Steam Dosage
Cumulative Production
Cumulative Production
Flow Rate
Spearmans
Correlation
Coefficient
0.55
0.79
-0.29
0.76
0.34
0.62
0.46
0.39
Sample
Size
30
28
56
34
36
26
34
58
*0nly coefficients significant at 95% level or higher
101
-------
TABLE 9-3. RESULTS OF MULTIPLE REGRESSION ANALYSIS ON LOG (VOC EMISSIONS)
Independent
Variables .
Intercept
Survey Flow Rate
Area (West/Central)
Regression Significance
Coefficients
-1.63
0.43 P< 0.01
1.70 P< 0.01
R = 0.37
Significance of Regression equation
degrees of
freedom F Significance
Regression
Error 55
Total 57
**
2 16.19 P< 0.01
*p is the probability that the coefficients are significantly different
than 0.
**p is the probability that this model accounts for a significant portion
of the variability.
102
-------
selected for this is given in Table 9-4. In addition to flow rate, soaking
period and possibly API gravity of the oil had a significant effect. 44%
of the variability of the Western wells is accounted for with this model.
103
-------
TABLE 9-4. RESULTS OF MULTIPLE REGRESSION ANALYSIS ON
LOG (VOC EMISSIONS) FOR WESTERN AREA OF KERN COUNTY
Independent
Variables
Intercept
Survey Flow Rate
Soaking period
API gravity
Regression
Coefficients
-7.21
0.43
0.16
0.51
Significance
*
p< 0.01
p=0.03
p=0.06
R = 0.44
Significance of Regression equation
Regression
Error
Total
degrees of
freedom F Significance
**
3 7.06 p< 0.01
27
30
*p is the probability that the coefficients are significantly different
than 0.
**p is the probability that this model accounts for a significant portion
of the variability.
104
-------
SECTION 10
EMISSION FACTOR DEVELOPMENT
The objective of this study was to provide refined data for calculating
the VOC emissions from wellhead casing vents associated with TEOR operations.
To that end, the test data on steam cycle wells and the test data on vapor
recovery systems serving steam drive wells was used to develop emission factors.
The emission factors, their confidence intervals, and a brief explanation
of the methods of development are contained in the next two subsections. A
more rigorous explanation of the development of the steam cycle emission factor
follows in Appendix B.
10.1 STEAM DRIVE WELL EMISSION FACTOR
Although very little testing has been done on individual steam drive wells,
a large body of data is available on vapor recovery systems which serve steam
drive wells. The vapor recovery system compliance tests measure the VOC recovered
through condensation and lost out the stack. The sum of the recovered and lost
VOC represents the total emissions of the steam drive wells connected to the
system and, therefore, can be used to calculate the average uncontrolled emissions.
Table 10-1 presents a summary of the vapor recovery system data used in calcula-
ting the steam drive well emission factor.
The use of vapor recovery system tests to calculate uncontrolled steam
drive well emission factors represents an approximate model. The actual
emissions from the wells may be affected by the recovery system back-pressure
or by back-flow of vapors into wells with a casing pressure lower than the
recovery system header.
105
-------
The data listed in Table 10-1 represents only about half of the existing
vapor recovery system source test data. Other tests were omitted from the
emission factor calculation for a variety of reasons. In some cases, there
were anomalies noted in the test results or procedures. For some tests, the
number of wells connected to the system was not known. Many tests were done
on systems with a mixture of steam drive and steam cycle wells. It was noted
in the early testing on this program that when cyclic wells were attached to
a vapor recovery system without a check valve, that it was possible to induce
back-flow from the vapor recovery system into the casing. Since this factor
could not be quantified easily, tests on hybrid systems without check valves
were not used in calculating the emission factor for drive wells. The remaining
data base is still quite large with 963 observations.
The emission factor is based on a weighted average of the individual
system test results. This results in an emission factor estimate of 220.3
pounds per day per well. An analysis of the variation between individual
system tests was used to calculate the confidence intervals surrounding the
emission factor estimate, which were found to range from 209.3 to 231.3 lb/day/
well.
10.2 STEAM CYCLE WELL EMISSION FACTOR
The results of the field testing done in this study were used to develop
the cyclic well emission factor. The data consisted of a survey of 358 randomly
selected wells which classified each well into one of four casing vent flow rate
strata. The lowest flow rate strata (less than 0.1 liter/minute) was assigned
a zero emission rate. A subsample of each of the other three strata were tested
to determine the mass emission rate of VOC from the casing vent. A total of
58 wells were tested, with most emphasis being placed on the highest flow strata.
Calculation of the emission factor was done in two steps. The first step
was to obtain an emission factor and confidence interval for wells emitting at
a rate greater than or equal to 0.1 liters/minute, referred to here as blowing
106
-------
TABLE 10-1. SUMMARY OF VAPOR RECOVERY SYSTEM SOURCE TESTS USED IN THE
STEAM DRIVE WELL EMISSION FACTOR
Chemecology
Test Report
Number
A- 647
A-661
A- 685
A- 8 24
A-979
Vapor Recovery
Producer System No.
Chevron CC-1-31
CC-2-31
CC-1-32
CC-2-32
CC-3-32
Chevron CC-1-5
CC-4-32
CT-3-5
CC-1-9
CT-2-5
CT-3-31
CT-1-4
CT-2-4
CT-5-3
Chevron CC-3-3
CT-4-3
Belridge
—
Chevron CT-4-3
3-CC-l
CT-1-3
CC-3-3
CT-2-4
CC-2-9
CC-4-32
Date
Tested
10/19/78
10/19/78
10/20/78
10/20/78
10/20/78
11/13/78
11/14/78
11/15/78
11/15/78
11/16/78
11/20/78
11/20/78
11/21/78
11/22/78
1/16/79
1/16/79
10/9/79
10/9/79
7/29/80
7/29/80
7/30/80
7/30/80
7/31/80
7/31/80
8/1/80
Total VOC
in Feed
(Ib/day)
791
1773
764
1910
1375
19704
4066
756
12859
7212
24432
2352
4356
7212
4298
5998
1462
4090
3730
806
2297
3955
9322
9451
2540
Number
of Wells
3
5
3
5
6
55
8
12
44
28
29
40
31
28
15
33
41
21
33
13
17
15
31
25
8
Emission
Factor
(Ib/day/well)
263.7
354.6
254.6
382.0
229.2
358.3
508.3
63.0
292.3
257.6
842.5
58.8
140.5
257.6
286.5
181.8
35.6
194.7
113.0
62.0
135.1
263.7
300.7
378.0
317.5
Continued/
-------
TABLE 10-1. (Continued)
o
00
Chemecology
Test Report
Number Producer
A-992 Chevron
A- 1002 Chevron
Vapor Recovery
System No.
CC-1-9
CC-1-5
CT-2-5
CT-3-5
CC-1-31
CC-3-31
CC-3-32
CC-2-32
CC-1-32
CC-2-31
CC-1-27
CT-16Z
CC-36W-1
CC-31X
CC-26C
Date
Tested
8/4/80
8/5/80
8/5/80
8/6/80
8/6/80
8/7/80
8/7/80
8/8/80
8/8/80
8/11/80
8/11/80
8/12/80
8/12/80
8/13/80
8/14/80
Total VOC
in Feed
(Ib/day)
4596
8681
10841
2013
1308
11963
1075
1918
2659
3850
1650
3037
7970
6305
6794
Number
of Wells
44
55
28
12
3
29
6
5
3
5
31
37
62
41
53
Emission
Factor
(Ib/day/well)
104.5
157.8
387.2
167.8
436.0
412.5
179.2
383.5
886.3
770.0
53.2
82.1
128.5
153.8
128.2
Totals = 212,171
963
Weighted Average = 220.3 Ib/day/well
95% Confidence Interval = 209.3 to 231.3 Ib/day/well
-------
wells. The second step was to combine this with an estimate and a confidence
interval for the proportion of blowing wells. This estimate was obtained from
the survey data. A brief description of these steps is included here and more
detail is included in Appendix B.
The emission factor for the blowing wells was calculated using the assumption
that this data had a lognormal distribution. The wells selected for testing
were stratified by flow rate and area. The mean emission rate for these blowing
wells was a weighted average using estimates of the proportions within each
strata that were obtained from the survey data. The variance was calculated
as a variation of the variance for a stratified sample. This was necessary since
only estimates of these proportions were available. A 97.5% confidence interval
was calculated for the mean emissions. A scale bias correction factor was
calculated to convert the log scale values to data scale.
The second step involved calculating a 97.5% confidence interval for the
proportion of wells that were blowing. This information was combined with
the information from step one to produce an emission factor and 95% confidence
interval for both blowing and non-blowing wells combined. This emission factor
was calculated as follows:
proportion of „ average emissions
Emission factor = wells emitting from emitting wells
The confidence intervals were combined in a similar manner.
This analysis resulted in an emission factor of 3.6 pounds per day per
well. The 95 percent confidence interval surrounding that emission factor
is estimated to be 2.2 to 6.2 pounds per day per well. This emission
factor estimate compares favorably with the simple arithmetic model which
results in a mean emissions estimate of 3.75 pounds per day per well. The
lognormal model was chosen because it allows the computation of more meaning-
ful confidence intervals.
The emission factor presented above represents all wells in Kern County.
Emission factors were also calculated on a more dissociated basis, by field
109
-------
and by areas. The calculation of emission factors by field was not productive,
since most fields had too few tests to make a firm estimate. The grouping of
fields in western Kern County separate from those in central Kern County,
however, provided some interesting results. Table 10-2 presents a comparison
of the overall cyclic well emission factor to those for the western and central
county areas.
110
-------
TABLE 10-2. EMISSION FACTORS AND VARIANCE DATA
FOR STEAM CYCLE WELLS
Area
Kern County -
Overall*
Western Kern
County
Central Kern
County
Emission
Factor
(Ib/day/well)
3.60
4.31
2.26
95%
Interval
Lower
2.21
2.32
0.70
Confidence
(Ib/day/well)
Upper
6.24
7.61
3.34
Scale Bias
Correction
Factor
3.34
3.02
2.51
Variance
of Data
in Logs
2.41
2.21
1.84
Standard
Error
in Logs
0.182
0.210
0.272
Arithmetic
Model Estimate
(Ib/day/well)
3.75
4.19
2.10
* In deriving the overall estimates, average emissions in the cell were weighted by the
proportion between the west and the central areas as determined in the survey. The VOC emissions
of the wells actually tested were averaged within each flow rate group and each area.
-------
50272-101
REPORT DOCUMENTATION
PAGE
l._REPORT NO.
EPA 909/9-81-003
3. Recipient's Accession No.
4. Title and Subtitle
Assesment of VOC Emissions from Well Vents
Associated with Thermally Enhanced Oil Recovery
5. Report Date Issued
September 1981
7. Authors) 8. Performing Organization Rept. No
G.E. Harris, K.W. Lee, S.M. Dennis, C.D. Anderson, D.L. Lewis DCN 81-240-016-09-12
9. Performing Organization Name and Address
Radian Corporation
8501 Mo-Pac Blvd.
P.O. Box 9948
Austin, Texas 78766
10. Project/Task/Work Unit No.
9
11. Contract(C) or Grant(G) No.
EPA #68-02-3513
(G)
12. Sponsoring Organization Name and Address
U.S. EPA Region IX
215 Fremont St.
San Francisco, CA 94806
13. Type of Report & Period Covered
Final
14.
15. Supplementary Notes
16. Abstract (Limit: 200 words)T
s)t~The objective of this document is to provide improved data for
determining the inventory of VOC emissions from wellhead casing vents associated with
thermally enhanced oil recovery (TEOR) in California. Both steam drive and cyclic steam
wells are examined in terms of emissions and population. The study concentrates on Kern
County.""]
"The details of a testing program conducted to determine the emissions from cyclic
steam wells are presented, along with the results of a survey of the characteristics of the
well, the producing field, and the steaming operation. Vrhe results of correlation studies
are also presentedTj An emission factor for cyclic wells is developed. (The data base
presented consists of a survey of 358 wells of which 58 were quantitatively tested."""
Emission data for steam drive wells is presented in the form of compliance tests for
vapor recovery systems associated with steam drive operations. This report presents a
summary of the applicable test data which was found and an emission factor developed from
that data.
17. Document Analysis a. Descriptors
b. Identifiers/Open-Ended Terms
c. COSATI Field/Group
Air Pollution
Crude Oil
Hydrocarbons
Thermally Enhanced Oil Recovery
Emission Factors
Volatile Organic Compounds
13H
18. Availability Statement
Release Unlimited
19. Security Class (This Report)
Unclassified
20. Security Class (This Page)
Unclassified
21. No. of Pages
22. Price
(See ANSI-Z39.18)
See Instructions on Reverse
OPTIONAL FORM 272 (4-77)
(Formerly NTIS-35)
Department of Commerce
-------
DO NOT PRINT THESE INSTRUCTIONS AS A PAGE IN A REPORT
INSTRUCTIONS
Optional Form 272, Report Documentation Page is based on Guidelines for Format and Production of Scientific and Teenm-si Reports .
ANSI Z39.1S-1974 available from American National Standards Institute, 1430 Broadway, New York, New York 10018. Each separately
bound report—for example, each volume in a multivoliime set—shall have its unique Report Documentation Page.
1. Report Number. Each individually bound report snail carry a unique alphanumeric designation assigned by the ^Mfuntiing orga-
nization or provided by the sponsoring organization in accordance with American National Standard ANSI Z39.23-1S74 Technical
Report Number (STRN). For registration of report code, contact NTIS Report Number Clearinghouse. Springfield. VA <£?J,61. Use
uppercase letters. Arabic numerals, slashes, and hyphens only, as in the following examples: FASEB/NS-75/«7 and FAA/
RD-75/09.
2. Leave blank.
3. Recipient's Accession Number. Reserved for use by each report recipient.
4. Title and Subtitle. Title should indicate dearly and briefly the subject coverage of the report, subordinate subtitle to the main
title. When a report is prepared in more than one volume, repeat the primary title, add volume number and include subtitle for
the specific volume.
5. Report Date. Each report shall carry a date indicating at least month and year. Indicate the basis on which it was selected (e.g..
date of issue, date of approval, date of preparation, date published).
6. Sponsoring Agency Code. Leave blank.
7. AuthorfrMSive name(s) in conventional order (e.g.. John R. Doe. or J. nuuert Doe). List author's affiliation if it differs from
the peifoililing organization.
8. Performing Organization Report Number. Insert if performing organization wishes to assign this number.
9. Performing Organization Name and Mailing Address. Give name, street, city, state, and ZIP code. List no more than two levels of
an organizational hierarchy. Display the name of the organization exactly as it should appear in Government indexes such as
Government Reports Announcements & Index (GRA & I).
10. Project/Task/Work Unit Number. Use the project task and work unit numbers under which the report was prepared.
11. Contract/Grant Number. Insert contract or grant number under which report was prepared.
12. Sponsoring Agency Name and Mailing Address. Include ZIP code, ate main sponsors.
13. Type of Report and Period Covered. State interim, final, etc.. and. if applicable, inclusive dates.
14% Performing Organization Code. Leave blank.
15. Supplementary Notes. Enter information not included elsewhere but useful, such as: Prepared in cooperation wrth . .
of ... Presented at conference of... To be published in ... When a report is revised, include a statement wheth** t .
report supersedes or supplements the older report.
16. Abstract Include a brief (200 words or less) factual summary of the most significant information contained in the report. If the
report contains a significant bibliography or literature survey, mention it here.
17. Document Analysis, (a). Descriptors. Select from the Thesaurus of Engineering and Scientific Terms the proper authorised terms
that identify the major concept of the research and are sufficiently specific and precise to be used as index entries for cataloging.
(b). Identifiers and Open-Ended Terms. Use identifiers for project names, code names, equipment designators, etc, Use open-
ended terms written in descriptor form for those subjects for which no descriptor exists.
(c). COSATI Reid/Group. Field and Group assignments are to be taken from the 1964 COSATI Subject Category List Since the
majority of documents are multidisciplinary in nature, the primary Field/Group assignment(s) will be the specific discipline,
area of human endeavor, or type of physical object The application(s) will be cross-referenced with secondary Reid/Group
assignments that will follow the primary posting(s).
M. Distribution Statement Denote public reusability, for example "Release unlimited", or limitation for reasons other than
security. Cite any availability to the public, with address, order number and price, if known.
19.&20. Security Classification. Enter U.S. Security Classification in accordance with U.S. Security Regulations (i.e.. UNCLASSIFIED).
21. Number of pages. Insert the total number of pages, including introductory pages, but excluding distribution list, if any.
22. Price. Enter price in paper copy (PC) and/or microfiche (MF) if known.
OPTIONAL FORM 272 BACK (4-77)
------- |