United States
Environmental Protection
Agency
Office of Air Quality
Planning and Standards
Research Triangle Park, NC 27711
EPA-453/R-93-045
October 1993
Air
Report to Congress on Hydrogen
EPA Sulfide Air Emissions Associated with
the Extraction of Oil and Natural Gas
''"?<'...!:„. <:,,„..•••;
-------
-------
HYDROGEN SULFIDE REPORT TO CONGRESS - EXECUTIVE SUMMARY
Under section 112(n)(5) of the Clean Air Act (CAA), as amended, Congress required
the Administrator of the United States Environmental Protection Agency (EPA) to carry out a
study to assess the hazards to public health and the environment resulting from the emission
of hydrogen sulfide (H2S) associated with the extraction of oil and natural gas. The
assessment must include a review of existing State and industry control standards, techniques,
and enforcement. This report, developed in fulfillment of section 112 (n)(5), evaluates the
hazards to the public and the environment posed by these emissions.
This study was added to the CAA by the Committee on Environment and Public
Works, chaired by the late Senator Quentin N. Burdick of North Dakota, because of concern
about the health and environmental hazards associated with H2S emissions from oil and gas
wells. Witnesses testified before Congress that these emissions resulted in deterioration of
air quality, death and injury to livestock, and evacuation and hospitalization of residents
located near the release point of such emissions.
Congress considered listing H2S as a hazardous air pollutant (HAP) under
section 1 12(b) of the CAA, which regulates industrial sources of routine emissions of HAPs.
On the basis of information contained in accident records, it was determined that H2S is a
concern from an accidental release standpoint and it would be listed under the accidental
release provisions in section 112(r) of the Act, and not under section 112(b). Substances
regulated under 1 12(r) are known or may be anticipated to cause death, injury, or serious
adverse effects to human health or the environment upon accidental release.
Hydrogen sulfide is produced in nature primarily through the decomposition of
organic material by bacteria. It develops in stagnant water that is low in oxygen content,
such as bogs, swamps, and polluted water. The gas also occurs as a natural constituent of
natural gas, petroleum, sulfur deposits, volcanic gases, and sulfur springs. Natural sources
constitute approximately 90 percent of the atmospheric burden of H2S. Ambient air
concentrations of H2S due to natural sources are estimated to be between 0.11 and 0.33 ppb
(0.15 and 0.46
H2S is a colorless gas with an offensive odor characteristic of rotten, eggs. H2S is
flammable and highly corrosive to metals. It is toxic and care should be exercised in its
presence. There have been several incidences in the United States of deaths of workers
exposed to H2S gases. Other symptoms of exposure include irritation, breathing disorders,
nausea, vomiting, diarrhea, giddiness, headaches, dizziness, confusion, rapid heart rate,
sweating, weakness, and profuse salivation. Levels above 1.5 x 10s ppb are considered life
threatening. Few studies exist measuring effects of natural or accidental exposure of wildlife
to H2S; however, wildlife deaths have been reported in connection with blowouts (a sudden
expulsion of gas or oil well fluids with great velocity).
-------
Natural gas and oil formations may be composed of many gases. The largest volume
and most beneficial gases in this composition are generally the light hydrocarbons (methane,
ethane, propane, and butane). H2S is the most common impurity in hydrocarbon gases. If
an oil and gas formation contains H2S, it is said to be "sour." Although a sour well's oil and
gas can be sweetened by removing the H2S after extraction, the well is always considered
sour once H2S is present.
Certain areas of the United States are especially prone to contain H2S in oil and gas
reservoirs at varying depths underground. Vulnerability zones have been characterized as 14
major H2S prone areas found in 20 States. Texas has four discrete H2S prone areas.
Concentrations as high as 42 percent H2S (by volume) have been found in gas from central
Wyoming.
In the oil and gas industry, H2S may be emitted or released during exploration,
development, extraction, crude treatment and storage, transportation (e.g., pipeline), and
refining. This report focuses on potential hazards of routine emissions and accidental
releases of H2S from the extraction and storage of crude oil and natural gas at well sites.
Potential sources of emissions include flares/vapor incinerators, heater-treaters (an
oil/water/gas separation device), storage tanks, equipment (valves, flanges, etc.), and both
active and abandoned wells.
When H2S is released to the air from an oil or gas well, several factors determine its
possible effects on surrounding residents and the environment. Accidental releases of sour
gas, such as from a well blowout or pipe rupture, are usually at high pressure and will
entrain surrounding air. This causes significant, immediate dilution of the H2S and other
components of the gas, thereby reducing the potential magnitude of the consequences of the
release. Factors such as chemical composition of the expelled gas, release rate, release
orientation, topography and meteorological conditions also determine the effects of such a
release.
Human fatalities from H2S exposure from oil wells in the United States have virtually
all been work-related. Significant public impacts are-rare although evacuations have been
initiated in response to accidental releases and at least one case of loss of consciousness has
been reported as a result of exposure.
Eighteen states have developed ambient air quality guidelines for H2S. Most,
however, do not collect continuous data but rather only monitor for H2S when a complaint is
made. These guidelines range from 160 ppb per 24-hr averaging time to 14 ppb per 24-hr
averaging time. Little data exist to determine actual levels of H2S near oil and gas extraction
sites. North Dakota was the only State found to have a continuous record of H2S
atmospheric levels at several sites. Exceedences of the North Dakota air quality standard
have been minimal in recent years at these monitoring locations. - No specific H2S
environmental (i.e., ecological) protection standards were found to exist. Some States
require notification of the regulatory authority upon accidental release of H2S from oil and
u-
-------
gas wells but few maintain an inventory of such incidences. Reporting of routine emissions
(emissions of small quantities from equipment, pipelines, flares, and storage tanks) was not
required by the States reviewed in this report.
H2S is regulated under a number of United States statutes. It is listed as a hazardous
substance under the Comprehensive Environmental Response, Compensation, and Liability
Act (CERCLA). It is listed under the Emergency Planning and Community Right-to-Know
Act (EPCRA) for emergency planning and preparedness, community right-to-know
reporting, and toxic chemical release reporting. The Occupational Safety and Health
Administration (OSHA) has established General Industry Standards that list worker exposure
concentration, limits, .and Respirator Standards. The National Institute for Occupational
Safety and Health (NIOSH) has produced a criteria document containing recommendations
for safe worker exposure levels and work practices. The United States EPA has the potential
for regulation of new oil and gas well sources through the Prevention of Significant
Deterioration (PSD) program and, as mentioned previously, H2S is listed under the CAA ,
section 112(r) accidental release provisions.
Other standards for worker and public protection from H2S emissions come from the
Bureau of Land Management, Minerals Management Service, and the American Conference
of Governmental Industrial Hygienists.
The oil and gas production industry has guidelines for safe practices regarding H2S.
The American Petroleum Institute, an industry-wide technical organization, has published six
documents regarding H2S in the industry. They pertain to safety practices for drilling,
operation, and equipment.
Findings and Recommendations
As a result of this study, EPA finds that the potential for human and environmental
exposures from routine emissions of H2S from oil and gas wells exists, but insufficient
evidence exists to suggest that these exposures present any significant threat. On the other
hand, an accidental release of H2S from an oil or gas well could have severe consequences
because of its toxicity and its potential to travel significant distances downwind under certain
circumstances. The likelihood (and thus the risk) of an accidental release of H2S or any
other hazardous substance, can be greatly reduced if facility owners/operators exercise the
general duty and responsibility to design and operate safe facilities and if they comply with
existing industry standards and practices, existing regulations, and future guidance and
regulations. Such actions should result in: (1) the safe management of H2S and other
hazardous substances with an emphasis on accident prevention; (2) the preparedness to
properly and quickly respond to chemical emergencies and to provide specialized medical
treatment if necessary; and (3) community understanding of the risks involved. Industry
should ensure that H2S is safely handled and that accidental releases are prevented; that any
releases that do occur are quickly discovered, controlled, and mitigated; and that workers and
the community are informed and prepared to properly respond to a H,S emergency.
111
-------
From the limited data available, there appears to be no evidence that a significant
threat to public health or the environment exists from routine emissions from sour oil and gas
wells. States and industry are encouraged to evaluate existing design, construction, and
operation principles within the framework of process safety management. EPA recommends
no further legislation pertaining to routine H2S emissions or accidental releases from oil and
gas wells at this time. However, the Agency does recommend that the owner/operators of
oil and gas extraction conduct drills and exercises with workers, the community, first
responders, and others to test mitigation, response, and medical treatment for a simulated
H3S accident. Sour oil and gas extraction facilities should be able to rapidly detect, mitigate,
and respond to accidental releases in order to minimize the consequences. The Agency will
continue to investigate the need for additional rulemaking under the accidental release
prevention provisions of the Clean Air Act.
IV
-------
Table of Contents
EXECUTIVE SUMMARY i
Chapter I. INTRODUCTION 1-1
Statutory Requirements 1-1
Scope of Report 1-2
References , 1-5
Chaper H. HYDROGEN SULFIDE FORMATION AND ITS ROLE IN
OIL AND GAS PRODUCTION H-l
Overview n-1
Hydrogen Sulfide in Industry n-3
Potential H2S Emission Sources in the
Oil and Natural Gas Extraction Industry n-3
Oil and Gas Production Operations H-6
Crude Oil n-6
Natural Gas H-10
Stripper Wells H-10
References n-11
Chapter m. HAZARD ASSESSMENT OF OIL AND GAS WELLS ffl-1
Introduction IQ-1
Objective m-1
Focus of Assessment HI-1
Scope and Limitations ffl-l
Hazard Assessment Steps ' IH-2
Hazard Identification ffl-2
Chemical Identity m-2
Location ffl-4
Nature of Hazard EH-4
Exposure Routes, Absorption, Metabolism, and
Elimination DI-4
Acute Human Toxicity ffl-5
Chronic Human Toxicity ffl-7
Ecological Effects DI-8
Flammability, Explosivity, and Corrosivity ffl-8
ACGIH Threshold Limits ffl-10
LC0, m-11
AfflLA Guidelines m-11
NAS/NRC Guidelines . . . -. ffl-13
Exposure and Consequence Analysis ffl-16
Vulnerability Zones ffl-16
• v
-------
Table of Contents (continued)
Exposure Analysis-Routine Emissions m-19
Monitoring Records m-19
Consequence Analysis-Routine Emissions m-32
Exposure Analysis-Accidental Releases m-35
Accidental Release Records ffl-36
Atmospheric Dispersion Analysis m-38
Consequence Analysis-Accidental Releases m-45
Consequence Analysis of Jets from Well Blowouts m-45
Consequence Analysis of Line Ruptures' IH-53
Consequence Analysis of Line Release Seepage m-53
Consequence Analysis of Flare Stack Releases m-56
Consequence Analysis of Releases Collecting at
Ground Level m:56
Accidental Releases-Prevention, Mitigation, and Emergency
Response • m-57
Process Safety Management IH-58
Major Safety Considerations m-59
Abandonment Practices HE-63
Land Use Around Well Sites m-63
Affected Human Populations m-65
Affected Environmental Settings . m-65
Findings m~67
References m-74
Chapter IV. REGULATORY PROGRAMS AND RECOMMENDED
INDUSTRY PROCEDURES IV'1
Introduction *V- *
State Regulations IV"1
Selected Oil and Gas Producing States IV-3
Oklahoma IV'3
Texas IV-5
Michigan ^V-7
California IV"9
A Comparison of H2S Regulatory Programs in Four States . . . IV-19
Other Producing States IV'19
Louisiana IV-21
New Mexico " IV-21
North Dakota IV"22
Pennsylvania IV-24
Wyoming IV-25
VI
-------
Table of Contents (continued)
Federal Regulatory Programs IV-25
OSHA Regulations IV-25
Current Regulations IV-26
Proposed Regulations IV-29
Impact of OSHA Regulations on Occupational and
Human Health IV-31
National Institute for Occupational Safety and Health IV-31
Bureau of Land Management F/-32
Minerals Management Service IV-32
CERCLA and EPCRA IV-33
Clean Air Act, Section 112(r) - Accident Prevention IV-35
Clean Air Act - PSD Program IV-36
Industry-Recommended Safety and Environmental
Protection. Procedures IV-37
API Recommended Practices IV-37
Control Standards IV-37
Control Techniques IV-37
Findings IV-41
References IV-44
Chapter V. RECOMMENDATIONS V-l
Routine Emissions V-l
Accidental Releases V-l
General V-l
Facility and Local Emergency Planning Committee (LEPC) V-2
Preparedness and Response V-2
Research and Further Studies V-3
Glossary G-l
Appendix A BACKGROUND INFORMATION ON THE OIL AND GAS PRODUCTION
INDUSTRY A-l
Exploration and Development A-l
How Oil and Gas are Produced A-l
Downhole Operations A-4
Surface Operations A-6
Overview of the Industry A-8
Principal Production Industry Groups A-13
Diversity of Production A-13
References A-18
Vll
-------
Table of Contents (continued)
Appendix B SUBJECTS OF STATE H2S REGULATIONS AND GUIDELINES . . . B-20
Appendix C ATMOSPHERIC DISPERSION CALCULATIONS FOR H2S RELEASES
FROM OIL AND GAS EXTRACTION FACLUTIES C-l
Introduction C-l
Summary Input and Output Data C-l
Sample SLAB Calculations C-5
SLAB Input C-5
SLAB Output ;:-..- C-7
Sample DEGADIS Calculations C-7
DEGADIS Input C-7
DEGADIS Output C-13
Sample SAPLUME Calculations C-13
SAPLUME Input C-13
SAPLUME Output C-17
References C-20
vui
-------
Figures
Figure n-1. Major H2S prone areas. n-4
Figure n-2. Typical extraction operation showing separation of
oU, gas, and water n-8
Figure IH-1. Components of the hazard assessment exercise ffl-3
Figure DI-2. ERCB H2S probit relations ; DI-12
Figure m-3. The effect of different assumptions on the calculation of
the radius of estimated vulnerable zones HI-17
Figure ffl-4 Example of drilling equipment layout - unconfined location ...... ni-18
Figure DI-5. Class I and DL areas of North Dakota including Lone Butte
and Theodore Roosevelt National Park DI-23
Figure m-6. Well distribution around Theodore Roosevelt National Park,
South Unit m-24
Figure m-7. Well distribution around Theodore Roosevelt National Park,
North Unit IH-25
Figure ni-8 Percentage of times designated H2S concentrations were
measured at the Theodore Roosevelt National Memorial Park -
North Unit monitoring site ffl-27
Figure DI-9. Wells producing between July 1986 and December 1987 surrounding
Lone Butte H2S ambient air monitoring site IQ-28
Figure EQ-IO. Percentage of times designated H2S concentrations were measured
at the Lone Butte monitoring site m-29
Figure IE-11. OU and gas fields m-33
Figure HI-12. Major H2S prone areas shown in relation to number of producing
oil and gas wells in 1990 m-34
Figure HI-13. Distribution of producing sour gas wells in Alberta by
H2S content IH-40
Figure HI-14. Total sulphur generated from producing gas wells in Alberta by
H2S composition of well HI-41
Figure m-15. Simplified representation of a completed sour-gas well ffl-46
Figure IH-16. Possible well flow scenarios HI-47
Figure HI-17. Possible well accidental release geometries EQ-47
Figure HI-18. Predicted H2S and SO2 concentrations for selected well blowout
observations HI-52
Figure IH-19. Possible pipeline rupture scenarios IH-54
Figure'in-20. Possible pipeline release geometries ffl-54
IX
-------
Figures (continued)
Figure ffl-21. Predicted mass release rates - rupture of 6" pipe ni-55
Figure IH-22. Predicted mass release rates - rupture of pipes of
differing diameters HI-55
Figure IH-23. Current land-use pattern by EPA region source HI-64
Figure m-24. Major H2S prone areas shown in relation to 1980 census data m-66
Figure 3H-25. Major H2S prone areas in relation to waterfowl habitats
of major concern DJ-68
Figure m-26. Major H2S prone areas shown in relation to National Forests
and Parks ...,:.. HI-69
Figure IV-1. Parts per million of H2S in some California oil and
geothermal fields IV-13
Figure IV-2. Multi-county districts IV-16
Figure A-l. * Rotary drilling rig A-2
Figure A-2. Cross-section of a well pumping installation A-3
Figure A-3. Main parts of a pumping unit A-5
Figure A-4. Typical extraction operation showing separation of oil, gas,
and water A-7
Figure A-5. 1991 U.S. oil and gas production by state A-10
Figure A-6. States with the most producing gas wells in 1990 . . A-ll
Figure A-7. Gas production in 1990 from the top producing states A-12
Figure A-8. Number of producing oil wells in the U.S. in 1990 A-14
Figure A-9. 1990 U.S. oil production A-17
Figure A-10. States with the largest number of producing oil wells in 1990 A-18
Figure A-ll. Oil production in 1990 from the top producing states A-19
-------
Tables
Table H-l. Physical/Chemical Properties of H2S E-2
Table H-2 Occupations with Potential H2S Exposure n-5
Table n-3 Examples of Potential H2S Routine Emission Sources
and Accidental H2S Release Sources from Oil and Gas Extraction . . . n-7
Table ffl-1. Effects of Exposure in Humans at Various Concentrations in Air .... m-6
Table m-2. Effects of Ecological Exposure to H2S m-9
Table IH-3, . North Dakota H2S Monitoring Studies m-21
Table HI-4. Examples of Accidental Releases of H2S from Oil and Gas
Extraction Operations with Impact on the Public or Environment . . . ffl-37
Table ffl-5. Example Gas Stream Compositions HL-42
Table m-6. Surface Deliverability as a Function of Well CAOF m-48
Table m-7. SLAB and SAPLUME Results — Horizontal Releases
from a Well Blowout ffl-50
Table IV-1. Ambient Air Quality Standards for H2S IV-2
Table IV-2. Highlights of California Laws for Conservation of Petroleum
and Gas Pertaining to H2S Emissions IV-10
Table IV-3. Highlights of Title 14, Chaper 4 of the California Code
of Regulations—Development, Regulation, and Conservation
of Oil and Gas Resources IV-12
Table IV-4. H2S in California Oil, Gas, and Geothermal Fields IV-14
Table IV-5. A Comparison of Four States' H2S Regulatory Programs IV-20
Table IV-6. Summary of Occupational Exposure Standards for H2S IV-27
Table IV-7. Reviewed American Petroleum Institute Documents Pertaining
to H2S in Oil and Gas Production IV-38
Table A-l. 1991 Oil and 1990 Gas Production Estimates A-9
Table A-2. 1990 Oil Production from Stripper Wells by State . / A-15
Table B-l. Subjects of State Hydrogen Sulfide Regulations B-l
Table C-l. Summary of Input and Output Data - Wellhead Blowout Scenarios . . . C-2
Table C-2. Pipe Rupture Scenarios - Inputs and Outputs (Sadenz Model) C-3
Table C-3. SLAB Input - Horizonal Wellhead Release C-9
Table C-4. Partial SLAB Output C-10
Table C-5. Input for DEGADIS Simulation of a Vertical Wellhead Release C-14
Table C-6. Partial DEGADIS Output - Vertical Jet C-14
Table C-7. Input for SAPLUME Runs C-18
Table C-8 Partial SAPLUME Output for Horizontal Plume C-l9
XI
-------
-------
CHAPTER I
INTRODUCTION
STATUTORY REQUIREMENTS
Section 112(n)(5) of the Clean Air Act (CAA or Act), as amended in 1990, requires
the Environmental Protection Agency (EPA) "to assess the hazards to the public and the
environment resulting from the emissions of hydrogen sulfide (H2S) associated with the
extraction of oil and natural gas resources." This assessment must reflect consultation with
the States and shall include a review of State and industry control standards, techniques, and
enforcement. To avoid duplication of work by other EPA offices, the assessment must build
upon a report from the Office of Solid Waste conducted under Section 8002(m) of the Solid
Waste Disposal Act. The Section 8002(m) study is a three-volume report to Congress
entitled Management of Wastes from the Exploration, Development, and Production of Crude
Oil, Natural Gas, and Geothermal Energy (1987).
The EPA Administrator is required by the Act to report to Congress with the findings
of the assessment along with any recommendations. Moreover, under Section 112(n)(5) (or
42 U.S.C.7412(n)5), the Administrator "shall, as appropriate, develop and implement a
control strategy for emissions of hydrogen sulfide to protect human health and the
environment."
This study was added to the Act by the Committee on Environment and Public Works
chaired by the late Senator Quentin N. Burdick of North Dakota. The study was included in
the Act because of concern about the health and environmental hazards associated with H2S
emissions from oil and gas wells. In 1987, Congress received testimony in which witnesses
urged that H2S should be listed as a hazardous air pollutant under the provisions of Section
112 of the Clean Air Act. The witnesses testified that lack of emission controls resulted in
significant deterioration of air quality. There was also testimony that H2S releases from oil
and gas facilities caused death and injury to livestock and required the evacuation and
hospitalization of residents from affected areas.
Congress considered listing H2S as a hazardous air pollutant (HAP) under Section
112(b), which regulates industrial sources for routine emissions of HAPs. On the basis of
information contained in accident records, it was determined that H2S is a concern from an
accidental release standpoint and should be Listed under the accidental release provisions in
Section 112(r) of the Act. The substances regulated under Section 112(r) are known or may
be anticipated to cause death, injury, or serious adverse effects to human health or the
environment from accidental releases. Under the provisions of Section 112(r) of the Act, the
EPA must develop a list of at least 100 substances that pose the greatest risk from accidentaf
releases. The Act listed 16 chemicals, including H2S, which must be included in the Section
112(r)list.
I- 1
-------
A clerical error led to the inadvertent addition of H2S to the Section 112(b) list of
HAPs. However, a Joint Resolution to remove H2S from the Section 112(b) list was passed
by the Senate on August 1, 1991, and the House of Representatives on November 25, 1991.
The Joint Resolution was approved by the President on December 4, 1991. It should be
emphasized that the purpose of this report is not to examine whether or not H2S should be
included in the Section 112(b) list.
SCOPE OF REPORT
The scope of this report is determined by the Congressional directive found in Section
112(n)(5), which is quoted in its entirety in'Exhibit 1. For clarity, the Agency has designed
the report to respond to specific items in the directive within separate chapters or sections of
chapters. It is important to note that although all issues relevant to this study have been
weighed in arriving at the conclusions and recommendations of this report, no single issue
has a determining influence on the conclusions and recommendations.
The directive in Section 112(n)(5) is expanded upon in the paragraphs below.
Detailed methodologies used to analyze and respond to the directive can be found later in this
report and in the supporting documentation and appendices. The principal components of the
Congressional mandate are:
1. Review existing State and industry control standards, techniques, and
enforcement programs.
Currently, there are no Federal ambient air quality standards for H2S. Most oil- and
gas-producing States have their own regulations and enforcement programs. Some States,
such as some hosting major producers, have large H2S programs in place. However, the risk
may exist in States that do not have large programs simply because of the lack of State
regulatory overview. Although Occupational Safety and Health Administration (OSHA)
standards exist that are applicable to oil and gas production, there are no industry-specific
standards. However, the industry has developed recommended practices and technologies to
reduce the potential for H2S emissions.
Current State regulations regarding H2S emissions from the extraction of oil and gas
are summarized in this report, with emphasis on four oil-producing States—California,
Michigan, Oklahoma, and Texas. Industry safety procedures as well as regulations
promulgated and proposed by OSHA and other Federal regulatory programs are reviewed.
2. Assess the hazards to public health and the environment resulting from the
emission of ILS associated with extraction of oil and natural gas resources.
Hydrogen sulfide is a colorless gas almost as toxic as hydrogen cyanide and 5 to 6
times more toxic than carbon monoxide. The principal threat of H2S gas to human life is
poisoning by inhalation (Dosch and Hodgson, 1986). Over the years, there have been
1-2
-------
112(nX5) Hydrogen Sulfide.- The Administrator is
directed to assess the hazards to public health and the
environment resulting from the emission of hydrogen
sulfide associated with the extraction of oil and natural
gas resources. To the extent practicable, the assessment
shall build upon and not duplicate work conducted for an
assessment pursuant to section 8002(m) of the Solid
Waste Disposal Act and shall reflect consultation with
the States. The assessment shall include a review of
existing State and industry control standards, tech-
niques, and enforcement. The Administrator shall re-
port to the Congress within 24 months after the date of
enactment of the Clean Air Act Amendments of 1990
with the findings of such assessment, together with any
recommendations, and shall, as appropriate, develop
and implement a control strategy for emissions ofhydro-
gen sulfide to protect human health and the environ-
ment, based on the findings of such assessment, using
authorities under this Act including sections 111 and
this section.
Exhibit 1. 1990 Clean Air Act Amendments: Mandate
for a Report to Congress on IL,S Emissions
Associated with Oil and Gas Extraction.
I - 3
-------
incidents involving exposure to H2S resulting from accidental releases from oil and gas
extraction facilities that have caused death or injury to humans or animals (Layton, D.W., et
al; Texas Oil and Gas Pipeline Corporation).
Oil and gas extraction, as defined in this study (see Appendix A), includes only the
activities involved in removing oil and/or gas from an established (developed) well. This
report includes not only a review of oil and gas extraction, but also other associated
components of oil and gas extraction such as piping to a separator, separation, and storage.
However, in following the Congressional mandate to address extraction, this report does not
cover activities primarily associated with exploration or well development, nor does it cover
sources such as gas processing plants. It is noteworthy that these plants are potential sources
of H2S releases since one of their functions is to .remove impurities such as produced water,
H2S and/or carbon dioxide. Personnel at these plants are trained in H2S safety. However,
this operation falls outside the definition of extraction.
In addition to assessing the sources of H2S emissions in the extraction industry, this
report discusses related control technologies as well as the health and environmental effects
associated with exposure to accidental H2S releases and routine H2S emissions during
extraction and closely associated production activities. When possible, monitored ambient air
concentrations of H2S and cases of death or injury to humans, wildlife, and/or livestock from
exposure to E^S releases and emissions are documented.
The report culminates with a hazard assessment of H2S routine emissions and
accidental releases from oil and gas extraction activities based on information obtained in the
efforts described in the previous paragraphs. Past and potential hazards from both routine
emissions and accidental releases are identified, the degree of hazard is assessed, and
potentially exposed human and ecological populations are identified.
3. Recommend and, as appropriate, develop and implement a control strategy for
H2S emissions to protect human health and the environment, based on the
findings of such assessment, using authorities under this act including sections
111 and 112.
As stated in a 1987 Senate report on the Clean Air Act Amendments, "Although
many State [H2S regulatory] programs are implemented conscientiously, in some instances
concerns have been raised that some oil- and gas-producing States may not be enforcing their
regulatory programs sufficiently or may have deficient regulatory programs. The purpose of
this subsection is to assess the effectiveness and the level of enforcement of various hydrogen
sulfide control programs. The assessment should assure more uniform application of control
technology, standards and enforcement. The Administrator should examine in particular
means of preventing accidental releases of hydrogen sulfide at remote facilities" (U.S.
Senate, 1987). [EPA identifies and reviews current 'State and Federal regulatory programs
and industry-recommended procedures to reduce routine emissions and accidental releases.
However, the ability to assess the effectiveness of these programs is limited by the lack of
1-4
-------
available emissions-monitoring data and the limited information available on accidental
release incidents.]
In this report, EPA makes recommendations regarding the release of H2S from oil and
gas extraction activities. The recommendations presented in this report do not constitute a
regulatory determination. The Agency is, in several important areas, presenting optional
approaches involving further research and consultation with the States and other affected
parties.
ORGANIZATION OF REPORT
This report addresses two forms of H2S losses to the atmosphere: routine emissions :....-
and accidental releases. (These terms are defined in the Glossary and examples are provided
in Chapter n.) .
Chapter n provides an overview of H2S formation in oil and natural gas deposits and
its presence in numerous industries. Potential sources of routine emissions and accidental
releases from the oil and natural gas extraction industry are identified along with their
causes. Chapter m is a hazard assessment of H2S losses from oil and gas wells. It contains
information on the nature of hydrogen sulfide's hazardous properties; exposure and
consequence analyses for routine emissions and accidental releases; protective guidelines,
prevention, mitigation, and emergency response procedures; and a characterization of land
use around wells and of affected human populations and environmental settings. Chapter IV
reviews and evaluates current State, Federal, and industry-recommended procedures related
to H2S in the oil and natural gas extraction industry. At the end of both Chapters m and IV
are lists of findings to provide the reader with a condensed summary of key information
identified during the development of this report. Chapter V completes the report with EPA
recommendations regarding routine emissions and accidental releases of H2S from oil and gas
extraction operations.
This report contains a glossary of terms commonly used, and three appendices
providing:
• background information on oil and gas production;
• subjects of State H2S regulations and guidelines; and
• atmospheric dispersion calculations for accidental H2S releases.
REFERENCES
Dosch, M.W., and Hodgson, S.F. 1986. Drilling and Operating Oil, Gas, and Geothermal
Wells in an H-S Environment, Publication No. M10. California Department of
Conservation, Division of Oil and Gas, Sacramento, CA.
1-5
-------
Layton, D.W., et al. 1983. Accidental Releases of Sour Gas From Wells and Collection
^Pipelines in the Overthrust Belt: Calculating and Assessing Potential Health and
Environmental Risks. Lawrence Livermore National Laboratory Report UCRL-
53411, Prepared for the Division of Fluid Mineral Operations, Bureau of Land
Management, U.S. Department of the Interior, Washington, DC.
Texas OH and Gas Pipeline Corporation. 1976. 6-inch Natural Gas-Gathering Pipeline
Failure, Meridian, Mississippi, May 21, 1974, PB-250 935, National Transportation
Safety Board, Washington D.C. February 4, 1976.
U.S. Senate. 1987. Clean Air Standards Attainment Act of 1987 "Report of Committee on
The Environment and Public Works. U.S. Senate to accompany S. 1894. Nov. 20,
1987, Report 100-231. U.S. Government Printing Office.
I - 6.
-------
CHAPTER n
HYDROGEN SULFIDE FORMATION AND ITS ROLE IN
OIL AND GAS PRODUCTION
OVERVIEW
Petroleum oil and natural gas originate in organic-rich sedimentary source rocks
composed of decayed marine algae and bacteria and terrestrial plants. In rock formations,
temperature increases with depth. The organic matter (kerogen) in sedimentary rock is
thermally converted to oil and gas at a specific temperature and migrates from the source
rock formation into a reservoir, or trap, formed by less porous cap rock, usually shale.
Once the well has been drilled into the reservoir, the oil and gas flow through the — .....
interconnected pore spaces to the well.
Natural gas may be composed of many gases. Only a few of these gases are typically
found in large concentrations. The largest volume and most beneficial gases in natural gas
are the light hydrocarbons (methane, ethane, propane and butane). Other gases that may
occur in large concentrations are carbon dioxide, nitrogen, and hydrogen sulfide. H2S is the
most common impurity in hydrocarbon gases.
H2S is generated under reducing conditions from high-sulfur kerogens or oils and is
most commonly formed in sedimentary rock formations such as limestone (calcite or calcium
carbonate). H2S can also be generated from hydrocarbon reactions with sulfates in carbonate
rock formations containing anhydrites. Oil and gas formations that do not contain H2S are
called "sweet." Oil and gas formations that contain H2S are described as "sour." Sour gas
is defined by the U.S. EPA as natural gas with an H2S concentration greater than 0.25 grains
per 100 cubic feet (GRI, 1990). Others have defined sour gas as having H2S concentrations
greater than 1.0 grain per 100 cubic feet (Amyx, Bass, and Whiting, 1960) or greater than 2
percent (Curtis and Showalter, 1989). The American Petroleum Institute recommends special
practices (described in Chapter IV) for sour gas when the natural gas's total pressure is
greater than or equal to 65 psia (448 kPa) and the partial pressure of H2S in the gas is
greater than 0.05 psia (0.34 kPa) (API, 1987). It is not known how many sour wells exist in
the United States. Sweet oil wells can become sour due to the introduction of sulfur-
reducing bacteria during enhanced oil recovery injection. Once an oil or gas field becomes
sour, it cannot be made sweet again. However, after extraction from the well, the oil and
gas can be sweetened by processing to remove H2S, and this is a common procedure.
In relatively low concentrations, H2S has a strong rotten-egg odor (Landes, 1953).
However, the sense of smell rapidly becomes fatigued and cannot be relied on to warn of the
continuous presence of H2S. In fact, high concentrations of H2S may cause a loss of smell.
Concentrations of H2S in crude oil vary greatly. In California alone, the Shiells Canyon oil
field measures only 6 x 104 ppb of H2S, while the Santa Maria Valley oil field has reported
H2S concentrations of 2.7 x 107 ppb (27 percent by weight) (Dosch and Hodgson, 1986).
H- 1
-------
Table EU. Physical/Chemical Properties o
Chemical Formula: H2S
Molecular Structure: /Svs
H H
Molecular Weight:" 34.08
Boiling Point: -60.33 °C (-76.59 °F)
Specific Gravity (H2~O=1): 0.916 at -60 °C (-76 °F) (Liquid) 1.54 g/L vapor at 0 °C (32 °F)
Vapor Pressure: 20 atmospheres at 25.5 °C (77.9 °F)
Melting Point -85.49 °C (-121.9 °F)
Vapor Density (AIR=1): 1.19
Solubility in Water: 1 gram dissolves in 242 mL at 20 °C (68 °F)
Flammable Limits: Lower Explosive Limit - (4.3 x 107 ppb)
Upper Explosive Limit - (45.5 x 10 ppb)
Odor Threshold: 20 ppba
Olfactory Fatigue Level: 1 x 10s ppba
Conditions or Materials to Avoid: Avoid physical damage to containers; sources
of ignition; and storage near nitric acid, strong oxidizing materials, and corrosive liquids
or gases (NFPA 1978). Hydrogen sulfide is incompatible with many materials, including
strong oxidizers, metals (NIOSH/OSHA, 1978, p. 112), strong nitric acid, bromine
pentafluoride, chlorine trifluoride, nitrogen triiodide, nitrogen trichloride, oxygen
difluoride, and phenyl diazonium chloride (NFPA 1978).
Hazardous Decomposition or Byproducts: When heated, it emits highly toxic
fumes of oxides of sulfur (Sax, 1984, p. 1552)
Source: U.S. EPA, 1993.
•NIOSH. 1977.
n-2-
-------
Hydrogen sulfide is also called hydrosulfuric acid, sulfurated hydrogen, sulfur
hydride, rotten-egg gas, swamp gas, and stink damp. Table II-1 lists some of the chemical
and physical properties of H2S. It is colorless, has a very low odor threshold, and being
more dense than air, it tends to settle to the ground when released to the atmosphere as a
pure gas (NIOSH, 1977). H2S oxidizes to form sulfur dioxide (SO2).
Exposure to H2S is one potential health and environmental concern associated with
extraction and related operations. H2S is found in Paleozoic carbonates in the Rockies, Mid-
Continent, Permian Basin, and Michigan and Illinois Basins (GRI, 1990). Figure n-1 shows
the areas of naturally occurring H2S. The Gas Research Institute reported in 1990 that H2S
can often occur in association with carbon dioxide (CO2) within the deep portions of a basin
and can comprise more than 30 percent of the composition.
Among the natural gas deposits in the United States, large deposits in central and
north-central Wyoming, in western Texas, in southeastern New Mexico, and in Arkansas
were singled out as rich in H2S. The Health Effects Research Laboratory (HERL) also
reported that H2S concentrations as high as 42 percent may be present in gas from central.
Wyoming. According to the Wyoming State Review (1991), released by the Interstate Oil
and Gas Compact Commission (IOGCC), gas reserves in Wyoming were estimated to be
approximately 11 trillion cubic feet. The IOGCC also reported that the reserves of liquid
hydrocarbons found in western Wyoming are approximately 5 percent H2S. Fifty percent of
the oil produced in Wyoming in 1989 was reported to be sour.
HYDROGEN SULFIDE IN INDUSTRY
Hydrogen sulfide has been cited as a potential hazard for approximately 125,000
employees in 73 industries (U.S. EPA, 1993). Industries with a potential exposure are listed
in Table n-2. The health effects of H2S were recognized in the petroleum industry more
than 50 years ago with the discovery of large deposits of high-sulfur oil in the United States
(Davenport, 1945). In the oil and gas industry, H2S may be. emitted or released during
exploration, development, extraction, crude treatment and storage, transportation (e.g.,
pipeline transmission), and refining. This report focuses on potential hazards of H3S routine
emissions and accidental releases from the extraction and storage of crude oil and natural
gas.
POTENTIAL H2S EMISSION SOURCES IN THE OIL AND NATURAL GAS
EXTRACTION INDUSTRY
Appendix A provides a general overview of the oil and gas extraction industry. Both
the exploration/development and extraction sectors of the industry are described along with
production data for recent years.
Hydrogen sulfide (H2S) complicates oil and gas extraction operations because of its
toxic effects and its corrosive properties. H2S exists as a gas at atmospheric pressure, but it
H-3
-------
Q IS Prone Areas
Source: Gas Research Institute. 1990.
Figure II-l. Major H2S prone areas.
H-4
-------
Table II-2. Occupations with Potential EU
Exposure
Animal fat and oil processors
Animal manure removers
Artificial-flavor makers
Asphalt storage workers
Barium carbonate makers
Barium salt makers
Blast furnace workers
Brewery workers
Bromide-brine workers
Cable splicers
Caisson workers
Carbon disulfide makers
Cellophane makers
Chemical laboratory workers,
teachers, students
Cistern cleaners
Citrus root fumigators
Coal gasification workers
Coke oven workers
Copper-ore sulfidizers
Depilatory makers
Dyemakers
Excavators
Felt makers
Fermentation process
workers -
Fertilizer makers
Fishing and fish-processing
workers
Fur dressers
Geothermal-power drilling
and production workers
Gluemakers
Gold-ore workers
Heavy-metal precipitators
Heavy-water manufacturers
Hydrochloric acid purifiers
Hydrogen sulfide production
and sales workers
Landfill workers
Lead ore sulfidizers
Lead removers
Litho graphers
Lithopone makers
Livestock farmers
Manhole and trench
workers
Metallurgists
Miners
Natural gas production
and processing workers
Painters using polysulfide
caulking compounds
Papermakers
Petroleum production
and refinery workers
Phosphate purifiers
Photoengravers
Pipeline maintenance
workers
Pyrite burners
Rayon makers
Refrigerant makers
Rubber and plastics
processors
Septic tank cleaners
Sewage treatment plant
workers
Sewer workers
Sheepdippers
Silk makers
Slaughterhouse workers
Smelting workers
Soapmakers
Sugar beet and cane
processors
Sulfur spa workers
Sulfur products processors
Synthetic-fiber makers
Tank gagers
Tannery workers
Textiles printers
Thiophene makers
Tunnel workers
Well diggers and cleaners
Wool pullers
Source: NTOSH. 1977.
n-5
-------
is soluble in oil and water. As a result of this solubility, H2S can enter the environment by a
variety of pathways. It can enter the atmosphere as a result of releases of gas containing H2S
or as a result of venting tanks or vessels which contain or have contained oil or water with
significant concentrations of H2S. Waters in the general environment can become
contaminated with H2S by contact with either gaseous plumes or waters that contain H2S.
The potential sources of H2S emissions associated with oil and gas extraction are
summarized in Table n-3.
Routine emission sources may include—
inefficient air emission control devices - ; —
tank venting due to diurnal temperature changes;
volatilization;
generation by sulfur-reducing bacteria in oil deposits; and
migration through poorly plugged wells.
Potential accidental release sources include —
• equipment failures, e.g., valves, flanges;
• piping ruptures due to corrosion, embrittlement, or stress; and
• venting due to unanticipated pressure changes.
Background information on these potential sources is provided in Appendix A.
The crude oil and natural gas industries use a large number of similar yet distinct
industrial processes that together serve a common purpose: to remove hydrocarbons from
subterranean deposits of oil and gas and to produce marketable products for industrial,
commercial, and residential use. Figure H-2 shows the basic components of a typical oil and
gas production operation. From the wellhead, the oil/gas mixture is piped to an oil/gas
separator. Oil/water emulsions and mixtures are then transferred to a heater-treater, which
separates the oil from the water. The treated crude oil is next piped to storage tanks, and the
produced water is piped to a holding tank prior to further treatment and/or disposal. An
emergency pit (a wastewater basin) is also provided. Each of these operations, as well as
other equipment found at a well site, may be a source of H2S in sour oil and gas operations.
Oil and Gas Production Operations
Crude Oil
In the crude oil production process, releases or emissions of H2S to the environment
may occur from a variety of sources, including wellheads, piping, flares, separation devices,
storage vessels, and pumps.
- n-6
-------
Table II-3. Examples of Potential Routine HjS Emission Sources and
Accidental IIS Release Sources from Sour Oil and Gas Extraction
Source
Mechanism
Cause
Flares/vapor incinerators
Heater-treaters ••- -,
Crude oil storage tanks
Water storage vessels
Equipment
(valves, flanges, etc.)
Oil/gas separator
Well plugging
Incomplete combustion
Pressure change,
high pressure
Diurnal temperature
change; filling operations;
volatilization
Volatilization;
sulfur-reducing bacteria
Corrosion and
embrittlement
Migration from well bore
to atmosphere
Design; lack of maintenance
- Pressure above design
specifications
Lack of controls; design
Lack of controls; design
Reaction of water with
metal and HgS; lack of
maintenance; poor
materials
Improper plugging
n-7
-------
Dry Gas
Meter
Oil and Gas
Producuon Well
IS
ie
Produced
Water
Storage
Tan£
Sedim
> a
JP
eat
Enhanced Recovery
or Disposal
Injection Well
barge, or truck
Sediment
ine,
uck
>r
«',' '!•/ V \ '"•"
tfjS&T'&t*
Reservoir
Source: U.S. EPA, 1987.
Figure II-2. Typical extraction operation showing separation of oil, gas, and water
H-8
-------
Flares are connected to points in the system where gas might be directed in case of an
operating problem. Subject to regulatory approval, flares may also burn gases that cannot be
sold. The gases are vented up a tall vertical pipe and then ignited at the top of the pipe,
releasing heat and combustion products. Flares are connected to production vessel
pressure-relief valves, rupture disks, and tank vents, among other places. Few data are
available on the efficiency of flares used in a crude oil production setting; however, the
operating efficiency of a common flare, regardless of industry application, is about 95 to 99
percent (personal communication, Donelson, Texaco, 12/9/92). The combustion product of
H2S is sulfur dioxide (SO^. Incomplete combustion from flares is one possible source of
H2S emissions, and actual pollutant emissions vary depending on the combustion efficiency of
the flare.
Devices, such as heater-treaters, break down water/oil emulsions or mixtures. These
devices operate under pressure and do not normally emit H2S. However, H2S may be
released in accidental situations when the vessel becomes subjected to pressures above design
specifications. The pressure relief valve or a rupture disk will open in a high-pressure
situation, and the gas will be sent through these openings via pipeline to a flare (personal
communication, Donelson, Texaco, 12/9/92).
H2S can potentially be emitted by two processes from vessels used to store water
produced during extraction:
• Dissolved H2S may be contained in the produced water and brought up from
the reservoir. Pressure reductions from subsurface to surface change the
solubility of H2S in water and can release some H2S from solution.
• H2S may be produced by the action of sulfate-reducing bacteria in some
aqueous and oil media. Biocides are used to kill these bacteria and eliminate
H2S formation.
Tanks storing crude oil are another potential source of H2S emissions. H2S can be
discharged to the atmosphere from a storage tank as a result of diurnal temperature change,
filling operations, and volatilization. The process of filling oil-transport vessels is another
potential source of H2S emissions. As the crude oil is loaded, gases containing the pollutant
are displaced to the atmosphere. If the gas amounts do not warrant repressuring into the gas
sales line, a flare may operate to burn the gas given off (personal communication, Donelson,
Texaco, 12/9/92). There have been several accidents involving tanks that have H2S in them.
This is typically a worker safety issue.
Pumps that move the oil during the extraction process can leak oil at the seals
between the moving shaft and the stationary casing, causing a possible release of H2S.
n-9
-------
Natural Gas
Two additional items in natural gas extraction can contribute emissions and releases of
sulfur compounds into the atmosphere: (1) equipment failure (e.g., leaks and ruptured pipes)
due to corrosion or embrittlement, and (2) improperly plugged wells.
Equipment Failure. H2S can attack the crystalline matrix of the steel, leading to
embrittlement and cracking of the steel, which could, in turn, lead to possible leakage of
H2S. This embrittlement is invisible and can occur in a short period of time. Corrosion,
which is caused by chemical reactions of metal with water and H2S, can also cause H2S
leakage. Because of the corrosive nature of H2S in the presence of water, oil and gas
operations take precautions to remove water from gas streams containing H2S. The National,
Association of Corrosion Engineers has a "Standard Material Requirement" entitled
"MR-0175-92, Sulfide Stress Cracking Resistant Metallic Materials for Oil Field Equipment"
which describes corrosion prevention measures. Corrosion resistant materials, coatings, and
chemical corrosion inhibitors may be used to prevent equipment failure and gas releases
where H2S and other corrosives are known to be present (personal communication, Donelson,
Texaco, 12/9/92). This type of accidental release is discussed in greater detail in Chapter
m.
Well Plugging. Improper well plugging may also be a potential source of H2S
emissions. After all of the recoverable natural resources have been removed from a well, it
must be properly plugged to avoid degradation of groundwater and surface water. Plugging
involves placing cement within a wellbore at specific intervals to permanently block the
possible migration of formation fluids containing H2S. Improper plugging may allow BUS (if
present) to migrate out of the wellbore and into the atmosphere. Well plugging is regulated
by the individual states. Plugging bonds are posted and procedures are subject to the
regulatory agency's approval and on-site witness (personal communication, Donelson,
Texaco, 12/9/92). This type of accidental release is also discussed in Chapter m.
Stripper Wells
Stripper wells are defined in Appendix A as producing at most 10 barrels of oil per
day or 100 thousand cubic feet of gas per day. The owners or operators of these wells are
typically smaller producing companies. Although stripper wells are often in remote areas,
many are not completely isolated from the public. The potential exists for livestock,
wildlife, or humans to come into contact with high levels of H2S from stripper wells due to
routine emissions and accidental releases. Although these wells are a potential hazard, no
data were available on the number of sour stripper wells in the United States.
H- 10
-------
REFERENCES
Amyx, Bass, and Whiting. 1960. Petroleum Reservoir Engineering.
API. 1987. API Recommended Practice 49 (RP 49), Recommended Practices for Safe
Drilling of Wells Containing Hydrogen Sulfide, 2nd ed., Publication No. RP49.
American Petroleum Institute.
Curtis, S., and Showalter, E. 1989. Pumping Units and Prime Movers, SPE Petroleum
Engineers Handbook.
Davenport, S.J. 1945. Hydrogen Sulfide Poisoning as a Hazard in the Production of Oil,
Information Circular 7329. U.S. Department of the Interior, Bureau of Mines.
Donelson, Texaco, Personal communication, 12/9/92.
Dosch, M.W., and Hodgson, S.F. 1986. Drilling and Operating Oil, Gas, and Geothermal
Wells in an H^S Environment, Publication No. M10. California Department of
Conservation, Division of Oil and Gas, Sacramento, CA, 39.
GRI. 1990. Chemical Composition of Discovered and Undiscovered Natural Gas in the
Continental United States. Volume I. Prepared by Energy and Environmental
Analysis, Inc., Arlington, VA, for Gas Research Institute, Chicago, IL.
Landes, K.K. 1953. Petroleum Geology. J. Wiley and Sons, Inc., New York, NY, 106 &
167.
NIOSH. 1977. NIOSH Criteria for a Recommended Standard.... Occupational Exposure to
Hydrogen Sulfide, Publication No. 77-158. U.S. Department of Health, Education,
and Welfare, National Institute for Occupational Safety and Health, Cincinnati, OH,
2, 147.
U.S. EPA. 1987. Management of Wastes from the Exploration, Development, and
Production of Crude Oil, Natural Gas, and Geothermal Energy, Volume /, EPA/530-
SW-88 003. U.S. Environmental Protection Agency, Washington, DC.
U.S. EPA. 1993. Health Assessment Document for Hydrogen Sulfide (HAD). EPA/600/8-
86/026F. U.S. Environmental Protection Agency, Office of Health and
Environmental Assessment, Washington, DC. January 1993.
Wyoming State Review, Interstate Oil and Gas Compact Commission/EPA State Review of
Oil and Gas Exploration and Production Waste Management Regulatory Programs,
October 1991. -
H- 11
-------
-------
CHAPTER HI
HAZARD ASSESSMENT OF OIL AND GAS WELLS
INTRODUCTION
Objective
The objective of this chapter is to evaluate the potential hazards to public health and
the environment resulting from routine emissions and accidental releases of hydrogen sulfide
(H2S) from oil and gas production, i.e., extraction; piping to a .separator; oil, gas, and water
separation; and associated storage.
Focus of Assessment
This hazard assessment was performed in two parts. First, existing H2S ambient air
monitoring data were compared to studies of human health and environmental effects to
determine whether the H2S concentrations measured from routine emissions have potentially
harmful effects. Second, the threat of accidental releases was assessed by identifying past
accidents and their impacts, reviewing atmospheric dispersion analyses (i.e., modeling) of
accidental release scenarios in the literature, and conducting additional analyses. The result
is an assessment of whether routine emissions and accidental releases are at levels that would
require a national control strategy. In addition, this assessment identifies the hazards of H2S,
recommended protective levels, and the areas of the United States potentially vulnerable to
routine emissions and accidental releases of H2S.
Scope and Limitations
This hazard assessment addresses hydrogen sulfide emissions and releases that may
potentially originate from a range of sources beginning with oil and gas wells (after well
development) up through their associated treatment processes, storage units, and piping.
However, it does not include gas processing or oil refining plants. For the potential H2S
emission sources described in Chapter n, non-occupational health impacts are considered
along with environmental impacts (i.e., wildlife, livestock, and vegetation). For wildlife and
livestock, the assessment includes animals that may be exposed to H2S when they wander
onto the well site.
For routine H2S emissions, this hazard assessment is limited by the lack of data
available on ambient air quality around well sites. Only a small amount of ambient
monitoring data collected by States was identified. In addition, no national statistics on the
health and environmental effects of chronic H2S exposure exist. Nor are national statistics on
the frequency and severity of accidental H,S emissions or releases available. Only case
records were located for the assessment of accidental releases. Therefore, the conclusions
drawn from this assessment are based primarily on predictive modeling of accidental releases
m-i
-------
and on a semi-quantitative comparison of ambient monitoring data and non-specific health
effects data.
Hazard Assessment Steps
This hazard assessment was divided into three major parts:
• Hazard Identification
• Exposure Analysis
• Consequence Analysis
Figure TTT-1 displays the various components of this assessment.
The first step in this assessment was hazard identification. It entailed collecting
information on the physical and chemical properties of H2S and its location in the United
States as it occurs (1) naturally in petroleum deposits, and (2) where it has been generated by
sulfur-reducing bacteria that are introduced by enhanced oil recovery processes. The
primary component of hazard identification is determining hydrogen sulfide's hazardous
properties: ignitability, corrosivity, explosivity, and toxicity to human health and the
environment.
The second step, exposure analysis, included identification of the H2S prone areas for
H2S exposure in the United States and the human and ecological populations expected to be
in these zones. The final part of the assessment, consequence analysis, was an examination
of H2S routine emissions and accidental releases occurring at oil and gas wells and the
severity of the consequences.
Since this report examines routine emissions and accidental releases separately, this
chapter first presents hazard identification, which is the same for both routine and accidental
releases. Next, routine exposure and its consequences are discussed. Finally, exposure to
accidental releases and its consequences are presented.
HAZARD roENTUlCATION
Chemical Identity
Hydrogen sulfide is a colorless, flammable gas which, in low concentrations, has a
characteristic odor of rotten eggs. It is a frequent component of crude oil and natural gas.
Hydrogen sulfide gas has the Chemical Abstracts Services (CAS) registry number 7783-06-4;
its physical and chemical properties are summarized in Table n-1.
m-2
-------
]
Hazard Identification
Chemical Identity
Location
Nature of Hazard
Exposure Analysis
Vulnerable Zones
Land Use in Zones
Human Populations
Environment
C
Consequence Analysis
Severity of " ••
Consequences
Adapted from: U.S. EPA, 1987.
Figure III-l. Components of the hazard assessment exercise.
m-3
-------
Location
BUS is found at varying depths in the earth's geological formations. Underground
sources of the gas are often referred to as pockets of H2S. Other natural sources of H2S
include volcanic gases, sulfur deposits, sulfur springs, and swamp gas from anaerobic decay.
Approximately 90 percent of the air emissions of H2S are produced by natural sources (U.S.
EPA, 1993). A portion of this 90 percent results from the routine emissions and accidental
releases resulting from the extraction of oil and gas containing H2S. Figure n-1 shows major
HjS-prone areas of the United States.
Nature of Hazard
Exposure Routes. Absorption. Metabolism, and Elimination
As described in previous chapters, the most rapid route of exposure to H2S is through
the air. Although eye irritation is the basis for the OSHA Permissible Exposure Limit
(PEL), inhalation is the quickest lethal exposure to humans and wildlife. The solubility of
H2S in water decreases as temperature increases; however, drinking groundwater has been
found with noticeable H2S concentrations.
Sullivan and Krieger's Hazardous Materials Toxicology (1992) summarizes the effects
of H2S exposure as follows:
In environmental and occupational exposures, the lung rather than the skin is
the primary route of absorption {Burgess, 1979; Yant, 1930). The dermal
absorption of H2S is minimal (Laug and Draize, 1942). Results from animal
inhalation studies indicate that H2S is distributed in the body to the brain,
liver, kidneys, pancreas, and small intestine (Voigt and Muller, 1955). Within
the body, H2S is metabolized by oxidation, methylation, and reaction with
metallo-or disulfide-containing proteins. Orally, intraperitoneally, and
intravenously administered H2S is primarily oxidized and directly excreted as
either free sulfate or conjugated sulfate in the urine (Curtis et al., 1972). .The
importance of methylation in the detoxification processes of H2S, however, is
unknown (Weisiger and Jakoby, 1980). The reaction of H2S with vital
metalloenzymes such as cytochrome oxidase is the likely toxic mechanism of
H2S (NRC, 1979; Smith and Gosselin, 1979). Reaction with nonessential
proteins may also serve as a detoxification pathway (Smith, Kroszyna, and
Kruszyna, 1976; Smith and Gosselin, 1964). Systemic poisoning occurs when
the amount of H2S absorbed exceeds that which can be detoxified and
eliminated (Yant, 1930; Milby, 1962). Because of its rapid oxidation in the
blood, H2S is not considered a cumulative poison (Yant, 1930; Ahlborg, 1951;
Haggard, 1925)....
- m-4
-------
There are no animal data available regarding the exhalation of H2S after
inhalation exposure. In animals, the excretion of H2S by the lungs is minimal
after peritoneal administration of H2S (Evans, 1967; Gunina, 1957; Susman et
al., 1978). However, because rescue personnel have developed H2S poisoning
shortly after starting mouth-to-mouth resuscitation on victims who had been
poisoned, it is likely that significant H2S is excreted from the lungs (Kleinfeld,
Giel, and Rosso, 1964).
Acute Human Toxicity
The odor perception threshold for H2S is very low. At concentrations between 3 and
20 ppb, the characteristic rotten egg odor is detectable. However, higher concentrations of
H2S in the 1.5 x 10s to 2.5 x 105 ppb range can cause olfactory paralysis. At these
concentrations, the olfactory sense may be lost and exposed persons may be unaware of the
presence of the toxic gas. Thus, odor cannot be relied upon as a warning sign of possible
exposure to H2S. Pulmonary edema, resulting from inhalation of levels between 3 x 105 and
5 x 10s ppb, can be fatal. (See Table m-1.) Inhaling levels between 5 x 105 and 1 x 106
ppb can cause a stimulation of the respiratory system, and rapid breathing (hyperpnea) will
occur followed by cessation of breathing (apnea). The effect of inhaling levels above 1 x 106
ppb is immediate respiratory paralysis followed by death.
Inhalation of levels above 2.5 x 103 ppb can damage organs and the nervous system.
Much of this damage is a result of a lack of oxygen (anoxia) caused by the depression of
cellular metabolism which can occur at 2.5 x 105 ppb. Instances of permanent neurological
damage in humans resulting from acute exposure have been described. Furthermore, animal
data have revealed that changes in the tissues of the brain, lungs and heart can occur from
exposure to the gas.
Irritation of the respiratory tract and eyes is another major effect of H2S exposure.
The gas is readily absorbed through the nasal and lung mucosa. It is very irritating to the
respiratory tract and eyes and can cause serious eye injury above 5 x 104 ppb. The gas can
affect the epithelium of the eye causing inflammation and lacrimation. The Integrated Risk
Information System (IRIS) (U.S. EPA, 1992) lists several signs and symptoms of H2S
exposure including painful conjunctivitis, sensitivity to light, tearing, and clouding of vision.
In addition, permanent scarring of the cornea can occur. At high, and potentially lethal
concentrations, the mucous membranes can be anesthetized so that irritation effects cannot be
relied upon to warn individuals of H2S exposure.
. In addition to irritation, IRIS lists other signs and symptoms of H2S exposure
including labored breathing and shortness of breath, profuse salivation, nausea, vomiting,
diarrhea, giddiness, headache, dizziness, confusion, rapid breathing, rapid heart rate,
sweating, and weakness.
m-5
-------
Table ffl-l. Effects of Exposure in Humans at Various Concentrations in Air
Level of Hydrogen Sulfide
Clinical Effect
Odor perception
threshold
Offensive odor
(rotten eggs)
Offensive odor
(sickening sweet)
Occupational
Exposure Limit
(OEL)
Serious eye injury
Olfactory paralysis
Pulmonary edema,
threat to life
Strong stimulation
of respiration
Respiratory paralysis,
collapse and death
ppb
3-20
<3xl04
>3xl04
IxlO4
5xl04 - IxlO5
1.5xl03 - 2xl03
SxlO5 - 5xl05
SxlO5 - IxlO6
IxlO6 - 2xl06
mg/m3
0.004 - 0.028
<42
>42
14
70 - 140
210 - 350
420 - 700
700 - 1400
1400 - 2800
Reference
Indiana Air
Pollution Control
Board ( 1964)
Ahlborg (1951)
National Research
Council (1977)
National Research
Council (1977)
National Research
Council (1977)
National Research
Council (1977)
National Research
Council (1977)
National Research
Council (1977)
National Research
Council (1977)
Source: U.S. EPA, 1993.
m-6
-------
Hydrogen sulfide may also decrease the body's ability to withstand infection. A
lexicological study exposed rats to 4.5 x 104 ppb of hydrogen sulfide for 2, 4, or 6 hours,
followed by a challenge with an aerosol of staphylococcus epidermis (Rogers and Ferin,
1981). A significant dose-response effect was seen in the number of colonies formed, when
the exsanguinated lungs were harvested from the rats at 30 minutes, 3 hours and 6 hours
post-challenge, and homogenized and grown in a selective growth medium for staphylococci.
Rats exposed to hydrogen sulfide for 4 hours had a 6.5-fold greater percent of colony-
forming units than controls, while those exposed to hydrogen sulfide for 6 hours had a 52-
fold greater percent of colony-forming units. The conclusion reached was that hydrogen
sulfide significantly affected the antibacterial system of the rats by impairing alveolar
macrophages.
However, Higashi et al. (1983), in a cross-sectional study of viscose rayon textile
workers exposed to hydrogen sulfide (average concentration, 3 x 103 ppb) and carbon
disulfide, found no difference between exposed employees and controls in respiratory and
spirometric variables. Similarly, Kangas et al. (1984) found no increased prevalence of
subjective symptoms among cellulose, mill workers exposed to hydrogen sulfide
concentrations of up to 2 x 104 ppb and methyl mercaptan levels as high as 1.5 x 104 ppb,
and much smaller amounts of dimethyl disulfide.
Chronic Human Toxicity
The toxicological data based was reviewed and an inhalation reference concentration
(RfC) was verified by the U.S. EPA Reference Dose (RfD)/RfC Work Group on June 21,
1990. The documentation is available via the Integrated Risk Information System (IRIS)
(U.S. EPA, 1991). The Integrated Risk Information System is an on-line data base
containing EPA risk assessment results and regulatory information. An RfC is defined as an
estimate, with uncertainty spanning perhaps an order of magnitude, of a daily exposure to the
human population (including sensitive subgroups) which is likely to be without adverse
effects during a lifetime (U.S. EPA, 1990). The derivation of the RfC is based on a
complete review of the toxicological literature and encompasses adjustments for exposure
duration and dosimetry. It utilizes uncertainty factors to account for specific extrapolations
between the population in which the effect was observed and the human population. The
critical, usually the most sensitive, effect is the focus of the RfC derivation; for this effect
the no-observed-adverse-effect level (NOAEL), or lowest-observed-adverse-effect level
(LOAEL) if a NAOEL is not available, is identified. Detailed discussion concerning these
issues can be found in U.S. Environmental Protection Agency, 1990.
The RfC for H2S is 9 x 10" mg/m3 (6.7 x 10'1 ppb) and was derived from the
NOAEL for inflammation of the nasal mucosa in. mice (Toxigenics, 1983). The subchronic
study revealed a lowest-observed-adverse-effect level (LOAEL) of 110 mg/m3 (8 x 104 ppb)
and a no-observed-adverse-effect level (NOAEL) of 42.5 mg/m3 (3.05 x 104 ppb). Since the
RfC may change due to evaluation of additional data, the reader is referred to IRIS for the
most current information regarding the RfC for H2S.
ffl-7 -
-------
The extrapolation of the NOAEL to the RfC follows several steps. First, the NOAEL
is adjusted to account for the daily length of exposure in the study; and second, it is
extrapolated to humans, and a human equivalent concentration (HEC) is calculated. Finally
an uncertainty factor is applied. The RfC for hydrogen sulfide is derived using an
uncertainty factor of 1000. The 1000 reflects a factor of 10 to protect sensitive individuals,
10 to adjust from subchronic studies to a chronic study (a subchronic study is carried out
over a shorter period of time and may not accurately reflect cumulative effects), and 10 to
adjust for interspecies conversions and database deficiencies.
Very little data exist on whether H2S can cause carcinogenic, mutagenic, reproductive
or developmental effects in humans or animals. Because of a lack of adequate test data, H2S
is currently placed in Group D, based on the weight-of-evidence criteria in the EPA's
Carcinogen Risk Assessment Guidelines issued in August 1986. A Group D ranking means
that the available data are inadequate to assess a chemical's human carcinogenic potential.
Furthermore, data are inadequate to state that H2S is mutagenic or that it causes reproductive
effects. Limited animal data do suggest that H2S appears to have potential to alter normal
developmental processes. No data on human developmental effects of inhaled H2S have been
located (U.S. EPA, 1993).
Ecological Effects
Data on the ecological effects of H2S are limited (Table ffl-2). McCallan, Hartzell,
and Wilcoxon (1936) and Benedict and Breem (1955) conducted high-exposure fumigation
studies, which noted that young, growing plants were the most susceptible to injury from
exposure to H2S. However, they noted that temperature, soil moisture, and species
differences were important factors affecting the results. Heck, Daines, and Hindawi (1970)
noted that mature leaves were unaffected while damage to the young shoots and leaves
consisted of scorching. Among the plants determined to be sensitive to H2S are clover,
soybean, tomatoes, tobacco, and buckwheat.
According to the EPA Health Assessment Document for H-,8 (U.S. EPA, 1993), few
studies exist that evaluate natural or accidental exposure of wildlife and/or domestic animals
to H2S. However, H2S has been determined to be highly toxic to some fish species. Animal
surveys conducted after a gas well blowout in Lodgepole, Alberta, Canada (Lodgepole
Blowout Inquiry Panel, 1984; Harris, 1986) revealed that large animals were exhibiting signs
of mucous membrane irritation and were avoiding the geographic area. Most cattle in the
exposed area were unaffected. Concentrations of H2S as high as 1.5 x 104 ppb (sampling
time unknown) were measured in the blowout area.
Flammability. Explosivity. and Corrosivity
"Hydrogen sulfide is generally stable when properly stored in cylinders at room
temperature. However, in the air, it is flammable and explosive and may be ignited by static
discharge. It may react with metals, oxidizing agents, and acids such as nitric acid, bromine
m-8
-------
Table HI-2. Effects of Ecological Exposure to H>S
Studies
Aquatic
Mammalian
Species
Bluegill
Rainbow Trout
Fathead Minnow
Mouse
Rat
Level
LCg,, 0.009 - 0.0478 mg/L
LCgQ 0.013 - 0.047 mg/L
LCgj 0.007 - 0.776 mg/L
NOAEL 42.5 mg/m3 (S.OSxlO4 ppb)
LOAEL 100 mg/m3 (8xl04ppb)
NOAEL 42.5 mg/m3 (3.05xl04 ppb)
LOAEL 100 mg/m3 (8xl04ppb)
Source
AQUIRE
AQUIRE
AQUIRE
IRIS
IRIS
IRIS
IRIS
AQUIRE
mis
Aquatic Toxicity Information Retrieval
Integrated Risk Information Service
LC
'so
Lethal Concentration 50
NOAEL No-observed-adverse-effect-level
LOAEL Lowest-observed-adverse-effect-level
ra-9
-------
pentafluoride, chlorine trifluoride, nitrogen triiodide, nitrogen trichloride, oxygen difluoride,
and phenyldiazonium chloride. When heated to decomposition, it emits highly toxic suliur
oxide ftimes" (Sullivan and Krieger, 1992). In pure form, its lower and upper explosive
limits are 4.3 percent (4.3 x 107 ppb) and 45.5 percent (45.5 x 107 ppb), and its auto-ignition
temperature is 260 °C (500 °F) (NIOSH, 1977). The National Fire Protection Association
(NFPA) has classified hydrogen sulfide in the highest flammability class (NFPA, 1974).
In the presence of water, hydrogen sulfide gas is highly corrosive to metals, including
high-tensile steel, which hydrogen sulfide can embrittle. These properties can lead to loss of
containment and accidental releases from ruptures if not controlled. Special precautions must
be taken to prevent spontaneous ignition fires when vessels that previously contained
concentrated hydrogen sulfide are opened. IgnitionJs caused by reaction of iron sulfide with
air to form iron oxide. The conversion of sulfide to oxide produces enough heat to ignite
flammable vapors (Dosch and Hodgson, 1986).
ACGIH Threshold Limits
The American Conference of Governmental Industrial Hygienists (ACGIH) publishes
a book of threshold limit values for chemical substances in the work environment (ACGIH,
1992). The limits are intended for use in the practice of industrial hygiene as guidelines or
recommendations in the control of potential health hazards. When OSHA began setting
standards for employee exposure in the 1970s, they adopted the ACGIH threshold limit
values (TLV's) as their permissible exposure limits. The ACGIH standards are
recommendations rather than regulations; they are updated annually and respond to current
research more quickly than OSHA's regulations.
The current limits for H2S were adopted by ACGIH in 1976. The Threrhold Limit
Value-Time Weighted Average (TLV-TWA) is 1 x 104 ppb or 14 mg/m3, and-the TLV short-
term exposure limit (TLV-STEL) is 1.5 x 104 ppb or 21 mg/m3. The TLV-TWA is defined
as the time-weighted average concentration for a normal 8-hour workday and 40-hour
workweek, to which nearly all workers may be repeatedly exposed, day after day, without
adverse effect. The TLV-STEL is defined as the concentration to which workers can be
exposed continuously for a short period of time without suffering from 1) irritation, 2)
chronic or irreversible tissue damage, or 3) narcosis of sufficient degree to increase the
likelihood of accidental injury, impair self-rescue or reduce work efficiency, also provided
that the daily TLV-TWA is not exceeded. A STEL is further defined as a 15-minute TWA
exposure which should not be exceeded at any time during a workday even if the TWA is
within the TLV-TWA. Exposures above the TLV-TWA up to the TLV-STEL should not be
longer than 15-minutes and should not occur more than 4 times a day, and should be
separated by 60 minutes each.
m-io
-------
One measure of the airborne concentrations of toxic materials that might cause fatality
is the LC0i, which is the concentration that could prove fatal to one percent of those exposed
to it. The LC0, is related to the exposure time, t, by a relationship of the form
LC0, = (k/t)1/n, where k and n are constants that depend on the material in question. This
relationship is a manifestation of the probit equation, which is a well-established way of
presenting the relationship between concentration, exposure time, and probability of fatality.
For H2S, the Center for Chemical Process Safety of the American Institute of
Chemical Engineers (AIChE) has a probit equation which gives k = 83,500 and n = 1.43,
with C in ppb and t in minutes (AIChE, 1989). Thus, for a five minute exposure, LC01=-
8.95 x 105 ppb and, for a one hour exposure, LC0j = 1.6 x 105 ppb.
The Energy Resources Conservation Board (ERCB) of Alberta, Canada (Alp et al.,
1990) has developed an alternative probit equation (shown in Figure m-2) which, for the
LC0i, gives k = l-364xl08 and n = 2.5. For a five minute exposure, this gives LCOI =
.3.75 x 105 ppb and for a one hour exposure gives LC0, = 1.4 x 105 ppb. The ERCB values
are thus more conservative.
AIHA Guidelines
The American Industrial Hygiene Association (AIHA) sets Emergency Response
Planning Guidelines (ERPGs) to protect the general public in the event of an emergency
release. The three ERPGs for H2S, which are time-dependent levels for varying degrees of
potential harm, are defined as follows:
ERPG-3 1 x 1Q5 ppb. the maximum airborne concentration below which it is
believed that nearly all individuals could be exposed for up to one hour
without experiencing or developing life-threatening health effects;
ERPG-2 3 x 104 ppb. the maximum airborne concentration below which it is
believed that nearly all individuals could be exposed for up to one hour
without experiencing or developing irreversible or other serious health
effects or symptoms which could impair an individual's ability to take
protective action.
ERPG-1 100 ppb. the maximum airborne concentrations below which it is
believed that nearly all individuals could be exposed for up to one hour
without experiencing other than mild, transient adverse health effects or
without perceiving a clearly defined objectionable odor.
m-11 -
-------
100
,0
80-1
70-
1 6°"
i
>»
r 40
0
30-
20-
10-
s.
100
1000
Concentration
10000
Note: Concentrations intentionally left in ppm.
Source: Alp et al., 1990.
Figure DI-2. ERCB HjS probit relations.
m-12
-------
For hydrogen sulfide, the ERPG-3 is based on human experience, while the ERPG-2
is based on animal studies and the ERPG-1 is based on the fact that the objectionable odor of
hydrogen sulfide is distinct at 300 ppb (AIHA, 1991). For the purposes of accidental release
dispersion analysis, the ERPG-2 was considered conservative and used as a threshold for
emergency countermeasures.
As stated above, these ERPG values are for an exposure time of one hour. At the
time of writing, there is no definitive guidance on how to extrapolate to shorter durations of
exposure. However, Gephart and Moses (1989) suggest that a constant dosage extrapolation
might be reasonable; that is, (ERPG in ppb)x(exposure time, t, in minutes). = constant, k.
Discussions with one of the_AIHA authors have suggested that, for t < 15 mm, k should be
divided by two. Thus, for H2S, the ERPG-2 is as follows:
• 3 x 104 ppb for an exposure time of one hour
• 1.8 x 105 ppb for an exposure time of five minutes.
The reader should recognize that these extrapolations are tentative and included for
purposes of illustration. They represent one of the greater sources of uncertainty in the
calculations.
NAS/NRC Guidelines
For the last forty years, the NRC's Committee on Toxicology has submitted
emergency exposure guidelines for chemicals of concern to the Department of Defense
(DOD) (NRC, 1986). These guidelines are used in planning for sudden contamination of air
during military and space operations; specifically, they are used to choose protective
equipment and reponse plans after non-routine but predictable occurrences such as line
breaks, spills, and fires. These guidelines are for peak levels of exposure considered
acceptable for rare situations, but are not to be applied in instances of repeated exposure.
An Emergency Exposure Guidance Level (EEGL) is defined as a concentration of a
substance in air (gas, vapor, or aerosol) judged by DOD to be acceptable for the
performance of specific tasks by military personnel during emergency conditions lasting 1 to
24 hours. Exposure to an EEGL is not considered safe, but acceptable during tasks which
are necessary to prevent greater risks, such as fire or explosion. Exposures at the EEGLs
may produce transient central nervous system effects and eye or respiratory irritation, but
nothing serious enough to prevent proper responses to emergency conditions.
Since the 1940's, the NRC has developed EEGLs for 41 chemicals, 15 of which are
listed in Section 302 of the Emergency Planning and Community Right-to-Know Act of 1986
(EPCRA) as extremely hazardous substances (EHSs). Although acute toxicity is the primary
basis for selecting EEGLs, long-term effects from a single acute exposure are also evaluated
for developmental, reproductive (in both sexes), carcinogenic, neurotoxic, respiratory and
other organ-related effects. The effect determined to be the most seriously debilitating,
nr-13
-------
work-limiting, or sensitive is selected as the basis for deriving the EEGL. This
concentration is intended to be sufficiently low to protect against other toxic effects that may
occur at higher concentrations. Factors such as age of the exposed population, length of
exposure, and susceptibility or sensitivity of the exposed population are also considered in
determining EEGLs.
Safety factors are used in developing EEGLs to reflect the nature and quality of the
data. Safety factors for single exposures may differ from those used in chronic studies. In
the absence of better information, a safety factor of 10 is suggested for EEGLs (i.e., the
reported toxicity value should be divided by 10) if only animal data are available and
extrapolation from animals to humans is necessary for acute, short-term effects (NRC, 1986).
The safety factor of 10 takes into account the possibility that some individuals might be more
sensitive than the animal species tested. A factor of 10 is also suggested if the likely route
of human exposure differs from the route reported experimentally (NRC, 1986), for
example, if oral data are reported and inhalation is the most likely exposure route for
humans.
As noted by NRC (1986, p. 7), development of an EEGL for different durations of
exposure usually begins with the shortest exposure anticipated (i.e., 10-15 minutes) and
works up to the longest, such as 24 hours. For H2S, 10-minute emergency exposure
guideline level (EEGL) is 5 x 104 ppb; 1 x 104 ppb is the 24-hour EEGL. The 24-hour/day,
90-day continuous exposure guide level (CEGL) for H2S has been recommended at 1 x 103
ppb (NCCT, 1985). Under the simplest framework, Haber's law is assumed to operate, with
the product of concentration (C) and time (t) as a constant (k) for all the short periods used
(Ct=k) (Casarett and Doull, 1986). If Ct is 30 and t is 10, then C is 3; if Ct is 30 and t is
30, then C is 1. If detoxification or recovery occurs and data are available on 24-hour
exposures, this is taken into account in modifying Ct. In some instances, the Ct concept will
be inappropriate, as for materials such as ammonia that can be more toxic with high
concentrations over short periods. Each material is considered in relation to the applicability
of Haber's law.
Generally, EEGLs have been developed for exposure to single substances, although
emergency exposures often involve complex mixtures of substances and, thus, present the
possibility of toxic effects resulting from several substances. In the absence of other
information, guidance levels for complex mixtures can be developed from EEGLs by
assuming as a first approximation that the toxic effects are additive. When the chemical
under evaluation for development of an EEGL is an animal or human carcinogen, a separate
qualitative risk assessment is undertaken in recognition of the fact that even limited exposure
to such an agent can theoretically increase the risk of cancer. The risk assessment is
performed with the aim of providing an estimate of the acute exposure that would not lead to
an excess risk of cancer greater than 1 in 10,000 exposed persons. The following
mathematical approach, taken directly from NRC (1986, pp. 26-27), is applicable for EEGL
computations for carcinogens:
- m-14
-------
1. If there has been computed an exposure level d (usually in ppm in air), which after
a lifetime of exposure is estimated to produce some "acceptable" level of excess
risk of cancer — say, 1x10^ — this has been called a "virtually safe dose"
(VSD). Computation of the dose d, if not already done by a regulatory agency,
will be computed by the Committee on Toxicology in accordance with generally
accepted procedures used by the major regulatory agencies, i.e., using the
multistage no-threshold model for carcinogenesis and the appropriate body
weight/surface area adjustments when extrapolating from an animal species to
humans.
2. If carcinogenic effect is assumed to be a linear function of the total (cumulative)
dose, then for a single 1-day human exposure an acceptable dose (to yield the
same total lifetime exposure) would be d times 25,600 (there being approximately
25,600 days in an average lifetime); the allowable 1-day (24-h) dose rate would be
dx 25,600
4
3. Because of uncertainties about which of several stages in the carcinogenic process
a material may operate in, and because of the likely low age of military persons, it
can be shown from data of Crump and Howe (1984) that the maximal additional
risk that these considerations contribute is a factor of 2.8. As a conservative
approach, the acceptable dose is divided by 2.8, i.e.,
d x 25.6QQ
2.8
If a lifetime excess risk, R, is established by DOD (for example, at 1 x 10", as
has been suggested by the International Council on Radiation Protection for
nuclear power plant workers), then the appropriate extent of risk at the EEGL
would be
d x 25.600
2.8
level of risk at d
(In the example given here, the level of risk at d was no more than 1 x 1Q-6.) If R
is 1 x 10", then R/risk at d = lO^/lO"6 = 100 (NRC, 1986).
4. If a further element of conservatism is required (for example, where animal data
need to be extrapolated to estimate human risk), an additional safety factor can be
used as divisor.
The NRC's Committee on Toxicology has also developed special public exposure
guidelines upon request from Department of Defense. The Short-term Public Exposure
Guidance Level (SPEGL) is defined as an acceptable ceiling concentration for a single,
-------
unpredicted short-term exposure to the public. The exposure period is usually calculated to
be one hour or less and never more than 24 hours. SPEGLs are generally set at 0.1 to 0.5
times the EEGL. A safety factor of 2 is often used to take into account effects on sensitive
subpopulations, such as children, the aged, and people with debilitating diseases. A safety
factor of 10 may be used to take into account the effects of an exposure on fetuses and
newboms. Effects on the reproductive capacity of both men and women are also considered.
Five SPEGLs (for hydrazine, dimethylhydrazine, monomethyl hydrazine, nitrogen dioxide,
and hydrogen chloride) have been developed by the NRC; all five chemicals are on the list of
EHSs. (U.S. EPA? 1987).
EXPOSURE AND CONSEQUENCE ANALYSES
In this section, potential exposures to and consequences of exposure to H2S from oil
and gas wells are analyzed. The zones of the United States most likely to contain H2S are
identified and the potentially exposed human and ecological populations are discussed.
Routine emissions and accidental releases of H2S are characterized using monitoring records
and dispersion modeling and the consequences are discussed. For accidental releases,
prevention, mitigation and emergency response policies-and procedures are also identified.
Vulnerability Zones
Vulnerability zones are estimated geographical areas that may be subject to
concentrations of H2S at levels that could cause irreversible acute health effects or death to
human populations within the area following an accidental release. For detailed hazard
analyses recommended under the Emergency Planning and Community Right-to-Know Act of
1986 (EPCRA), see Chapter IV; vulnerability zones are based on estimates of the quantity of
hazardous substances released to air, the rate of release to air, airborne dispersion, and the
airborne concentration that could cause irreversible health effects or death. This concept of
vulnerability is used to assess regions most likely to encounter routine emissions or accidental
H-iS releases from oil and gas production. This report does not use the EPCRA
methodology. Rather, the basic tools of a hazard analysis are used to alert the reader to
areas with potential H2S hazards.
Estimated vulnerability zones are shown in Figure DI-3 as circles with different radii
to illustrate how changing conditions or assumptions can influence the vulnerablity zone
estimate. With most atmospheric releases, the actual concentration of the airborne chemical
tends to decrease as it moves further downwind from the release site because of continual
mixing and dilution (i.e., dispersion).
" The American Petroleum Institute (API), an industry-wide technical organization, has
published several recommended practices (RP) pertaining to hydrogen sulfide in the oil and
gas production industry. Figure IH-4 shows API's RP 49 recommended equipment layout to
minimize vulnerability zones for an unconfined area, taking the potential for H2S releases
into consideration. Confinement refers to offshore sites and some land locations confined by
m-16
-------
/ /*
* t
/ / /
* /
I i I •
i' ' i
\ (2) | (4)
! Smaller J Selection of
I amount and . higher level
• rate of release of concern
\
X
X \ \
NN \ \
\
\
\
Release Site
\
\
\ \
\ (3) j (1)
i Use of greater | Radius for
I wind speed j initial
' and less / screening
, atmospheric / zone
/ stability / ,
\
V *
*
/
/ I
t •
Source: U.S. EPA. 1987.
Figure III-3. The effect of different assumptions on the calculation of the
radius of estimated vulnerable zones.
•ffl-17
-------
Wind
Streamer
Contractor's Ibol Puiher
and Drilling Foreman'! Office
Riz Power
Plant
Barricade with
Caution Sign
S«mcid« ith
Cjuiwn Sna
Bncfinf Area and
Protection Center
B^mou Blowout Prevenur
Station & Accumulator Equipment
Source; API. 1987.
Figure III-4. Example of drilling equipment layout - unconfined location.
HI-18
-------
the restriction of area, method of access, terrain, surrounding population distribution, etc. In
an H2S environment, well plot areas should be larger than usual, (i.e.. larger reserve pits.
turnaround room. etc.). The extra space allows for a greater margin of safety in well site
activities and, in turn, a smaller vulnerability zone.
The California Division of Oil and Gas provides guidance on H2S exposure
prevention. In their report, Drilling and Operating Oil, Gas, and Geothermal Wells in an
H^S Environment, the State recommends calculating the well area's potential toxicity from
H2S emissions, if the volume of oil or gas produced and the concentration of the H2S in the
oil or gas are known (Dosch and Hodgson, 1986). From these data, the radius from the
source to the 3 x 105 ppb and 1 x 105 ppb H2S concentration area can be determined on
dispersion-based scales. Potential sources of toxic gas emissions considered in calculating
the toxicity of the well area include wells and associated production, treatment, processing,
and storage facilities.
Calculating vulnerability zones for H2S on a nationwide basis, as in EPCRA hazard
analyses, is difficult because vulnerability zones are designed for site-specific studies.
Therefore, this assessment will take a broader approach to identifying vulnerability zones,
which will be referred to as H2S prone areas. These areas are considered the major areas of
the United States prone to natural occurrences of hydrogen sulfide. Figure H-l identified 14
major H2S prone areas in the United States. The 20 states having H2S prone areas are
Alabama, Arizona, Arkansas, California, Colorado, Florida, Idaho, Indiana, Illinois,
Kentucky, Louisiana, Michigan, Mississippi, Missouri, Nebraska, North Dakota, Oklahoma,
Texas, Utah, and Wyoming. Texas has four discrete areas prone to H2S. However, some
States, such as Louisiana, do not drill to depths of known H2S deposits; in Louisiana, oil and
gas wells appear to be located in more shallow depths.
Exposure Analysis — Routine Emissions
Monitoring Records
Ambient air monitoring programs measure the concentration of pollutants after they
have dispersed from one or more sources. These levels are recorded and tracked
continuously so that the level of exposure and air quality can be assessed over the long term
and under varying meteorological and emission scenarios. Ambient air monitoring is also
used to determine compliance with air quality standards by measuring pollutant
concentrations. With a dispersed, relatively unreactive primary pollutant such as hydrogen
sulfide, often the emissions can be traced back to the specific source.
Many States require ambient air monitoring for hydrogen sulfide at gas plants and
refineries; however, monitoring is not frequently required at oil and gas extraction facilities.
In the preparation of this report, six States (California, Michigan, North Dakota, Oklahoma,
Texas, and Wyoming) were contacted and questioned about the availability of monitoring
-------
data. California. Michigan, Oklahoma, Texas and Wyoming had not conducted pertinent
ambient air monitoring.
The North Dakota State Department of Health and Consolidated Laboratories
(NDSDH&CL) performs ambient monitoring for routine emissions of H2S and has collected
the data since 1980. The following discussion summarizes North Dakota's program to
provide an indication of historical, routine emissions of H2S from wells. Since no other
States have such monitoring data available, this report relies on North Dakota's data to assess
hazards and draw conclusions.
The North Dakota database contains site name, year/month/day monitored, and H2S
value measured. .The database.reflects three background and six special purpose monitors
(i.e., monitors set up as a result of a complaint)". Monitoring periods vary in length from
months to over a decade for a total of 393 months (32.75 years) of data (personal
communication, D, Harman, NDSDH&CL, 8/11/92). Table m-3 shows the North Dakota
data. The data were in half-hour average concentrations up to January 1, 1988, when the
averages recorded were changed to hourly, to correspond with the change in the North
Dakota Ambient Air Quality Standards (NDAAQS). Some monitoring lasted less than a
year; however, monitoring in the Theodore Roosevelt National Park-north unit was begun in
1980 and continues today.
North Dakota's Hydrogen Sulfide Standards - An Historical Review. At the time of
the early monitoring activities, there were two NDAAQS for hydrogen sulfide, both based on
half-hour averages and on odor thresholds but over different time spans. Adopted in 1970,
they were based upon guidelines established in the Interstate Air Pollution Study conducted in
St. Louis in the late 1960s. Those standards were 54 ppb (75 jtg/m3), 1/2-hour maximum
concentration not to be exceeded more than twice per year; and 32 ppb (45 jig/m3), 1/2-hour
maximum concentration not to be exceeded more than twice in any five consecutive days.
The 1/2-hour hydrogen sulfide standards were inconvenient because all of the other pollutants
were being tracked on an hourly basis. To correct the situation, North Dakota developed a
1-hour standard that would afford the same degree of protection as the old 1/2-hour standards
did, while still based on an odor threshold value. Statistically, they narrowed the proposed
standard down to a range of concentrations between 48 ppb and 52 ppb. Montana had an
existing hydrogen sulfide standard of 50 ppb for a 1-hour period, not to be exceeded more
than once per year, and North Dakota decided to adopt the same standard to provide
consistency on both sides of the North Dakota-Montana State border. The 50 ppb (70 /tg/m3)
1-hour hydrogen sulfide standard became effective October 1, 1987.
At the same time that the new standard became effective, a new chapter (Chapter 20)
was added to North Dakota's Air Pollution Control Rules entitled "Control of Emissions
from Oil and Gas Well Production Facilities." The oil companies expressed concern that
the hydrogen sulfide standard was included in North Dakota's table of ambient air quality
standards (NDAAQS) and, by law, exceptions could not be granted. Their position was that
they could not guarantee compliance with the standard at all times, and that the standard was
m-20.
-------
Fable EQ-3. North Dakota BUS Monitoring Studies
Studv
Location
Dates
Year
Ambient
Std. (ppb)
Violation'
(hours)
(ppb)
loffler
lieodore Roosevelt
National Memorial
'ark - North Unit
srgenson
aormas
leodore Roosevelt
ational Memorial
irk - South Unit
>ne Butte
stwood
iza
Farmyard within
1/2 mile of well and
tank battery
Little Missouri River
Valley, near the north
unit park headquarters
Valley with several oil
wells within 1 mile
Farmyard within
1 mile of several wells
Painted Canyon
Rest Area
Little Missouri River
Valley near an oil tank
battery in Little Knife
Oil Field
Lostwood National
Wildlife Refuge
headquarters
Farmyard within 1.5
miles of several wells
Town of Plaza, within
2 miles of several wells
and tank batteries
5/11/80-9/29/80
4/24/80 - 8/2/92
(1990 missing)
10/2/80 - 5/13/82
6/30/82 - 10/31/83
10/17/85 - 6/30/90
1/17/84,7/11/89
12/26/85 - 1/14/91
7/20/89 - 9/18/90
9/4/90 - 8/3/92
1980
1980
1981
1982
1983
1984
1985
1986
1987
1988
1989
1990
1991
1992
1980 •
1981
1982
1982
1983
1985
1986
1987
1988
1989
1990
1984
1985
1986
1987
1988
1989
1985
1986
1987
1988
1989
1990
1991
1989
1990
1990
1991
1992
32
32
32
32
32
32
32
32
32/50
50
50
50/200
200
200
32
32
32
32
32
32
32
32/50
50
50
50/200
32
32
32
32/50
50
50
32
32
32/50
50
50
50/200
200
50
50/200
50/200
200
200
0
0
1
34
31
27
35
12
0
0
0
0
0
0
8
19
13
9
7
0
0
0
0
0
0
1808
1859
1653
1130
320
25 .
0
0
0.
0
0
0
0
2
0
0
0
1
13
4
220
500
158
415
•137
87
73**
39
10
10
32
6
160
230
250
541
353
12
16
18.
9
10
0*
1630
2734
2182
2420
1515
122
0
18
45
46
47
88
0
88
73
152
353
269
urce: Personal correspondence, D. Harman, NDSDH & CL, 8/11/92.
alyais of data prior to 10/1/87 based upon 32 ppb, 1/2-hour average standard, not to be exceeded more than twice in anv
onsecutive aays.
alysis of data between 10/1/87 and 6/1/90 based upon 50 ppfa 1-hour average standard, not to be exceeded more than once
1 year.
alvsis of data after 6/1/90 based upon 200 ppb 1-hour average standard, not to be exceeded more than 1 tune per month.
violation occurs the second time the standard is exceeded.
Monitor out of service much of the time period. "~ " '"" -••- .
Exceedance defined as 2 times the standard.' • • • • -
m-2i
-------
not based on health-related concerns but on odor recognition levels. As a result, a joint
Health Department/Industry task force was established and four new health-based standards
were developed (effective June 1, 1990). These included raising the 50 ppb, 1-hr standard to
a 200 ppb, 1-hr standard - a decrease in H2S protection by a factor of four. These standards,
which remain in effect today, are as follows:
• 1 x 104 ppb or 14 mg/m3) maximum instantaneous concentration not to be
exceeded;
• 200 ppb or 280 ng/m3) maximum 1-hour average concentration not to be exceeded
more than once per month;
• 100 ppb or 140 /ig/m3) maximum 24-hour average concentration not to be
exceeded more than once per year;. -
• 20 ppb or 28 ^g/m3) maximum arithmetic mean concentration averaged over three
consecutive months (personal communication, D. Harman, NDSDH&CL,
8/11/92).
Methodology for Analysis nf Monitoring Data. For the analysis of the monitoring
data, only one of the standards was evaluated for each time period. Prior to October 1,
1987, the data were compared to the 32 ppb 1/2-hour average standard, not to be exceeded
more'than twice in any five consecutive days. After October 1, 1987, and prior to June 1,
1990 50 ppb was the only standard in effect, not to be exceeded more than once per year.
The data collected after June 1, 1990, were compared to the 200 ppb standard which was not
to be exceeded more than once per month. The results of the analysis are tabulated in Table
IH-3.
•
PSD Class I Areas. Several of the North Dakota monitoring programs were
conducted to monitor air quality changes resulting from the oil and gas production industry at
national parks and wildlife refuges. The Federal government established the Prevention of
Significant Deterioration permit program (PSD) to protect areas with good air quality. In
North Dakota, the most important, or Class I, areas include the Lostwood National Wildlife
Refuge and the northern, southern and Elkhorn Ranch portions of the Theodore Roosevelt
National Park (see Figure m-5). Monitoring sites for hydrogen sulfide were set up at all of
these locations except the Elkhorn Ranch locations.
At the lx>stwood Wildlife Refuge, data were obtained for the period from December
26, 1985, until January 14, 1991. Throughout the time period the maximum average
concentration was 88 ppb, recorded as a 1-hour average in 1990. Overall, this was a site
with acceptable air quality with respect to hydrogen sulfide because there were no NDAAQS
violations.
In the Theodore Roosevelt National Park system (see Figures m-6 and m-7 for well
distribution around the park), data were received by NDSDH&CL for the south unit
(obtained at the Painted Canyon Rest Area) from October 17, 1985, to June 30, 1990. The
air quality was very good, with no NDAAQS violations, and a maximum half-hour average
m-22
-------
• Watford City
TRNP-NU
North Dakota
South Dakota
Source: Bilderbeck, 1988
Figure III-o. Class I and H areas of North Dakota including Lone Butte and
Theodore Roosevelt National Park (TRNP). Bold outlined"areas
are Class I; remaining area is Class II.
m-23
-------
A A
N
3km
Source: Bilderbeck, 1988.
Figure IJI-6. Well distribution around Theodore Roosevelt National Park,
South Unit.
m-24
-------
TRNP-North Unit
N Lone Butte
WeU
urce: Bilderbeck, 1988.
III-7. Well distribution around Theodore Roosevelt National Park, North Unit.
m-25
-------
M
1983 Td 70 in 1985). The maximum 1/2 hr time-weighted average concentration recorded
SgThis period was 500 ppb in 1982. Air quality did improve during the second half of
Sdy period, with several years of no NDAAQS violations. This was a result of
NDSDH&CL mandated implementation of rigorous operations and maintenance programs by
well operators involved in the field and tank vapor collection. Also, expansion of a gas-
Serinl feline network contributed to the 'decrease in H2S concentrations because gases
were previously released to the atmosphere.
From 1988 to 1990, the Williston Basin Regional Air Quality Study (BLM, 1990) was
undertaken as a joint project between North Dakota and the Bureau of Land Management
m^o fo^t^omp iance with Federal standards for sulfur dioxide, the resulting product
Smgen Se combustion. Figure m-8 shows the range of concentrations measured at
the site Although over the entire period, 0 ppb was the concentration most frequently
recorded, a decrease in air quality is charted, from 1982 through 1987.
Lone Butte. Lone Butte, is located approximately 11 km from the north unit of
Theodore Roosevelt National Memorial Park (see Figure m-5). Lone Butte had
concentrations of hydrogen sulfide an order of magnitude higher than the other sites. The
monitor at Lone Butte (see Figure ffl-9), in the Little Missouri River Valley near - oU tank
battery in the Lone Butte Oil Field, recorded more than 3000 vio ations of the 1/2-hour
average 32 ppb NDAAQS per year from 1984 to 1986. Air quality did improve at the end
of themonigring period, though not to levels continuously below the NDAAQS of 50 ppb
which was the standard at that time.
Figures DI-10 depicts the range of concentrations measured at the Lone Butte site.
Zero ppb is recorded more than 50 percent of the time through the early years, with an
improvement towards 80 percent of the time by 1989. (The detection limit of the momtonng
equipment was 1 ppb.) The improving trend toward the hydrogen sulfide standard occurred
when the NDSDH&CL correlated the sources of the hydrogen sulfide with the ambient
monitor levels through the use of the prevailing wind direction. The possibility of
NDSDH&CL requiring individual monitoring at each well site convinced the producers to
reduce their emissions (personal communication, D. Harman, NDSDH&CL, S/ll/VZ).
nthP.r Monitoring Sites. Data from thirteen months of monitoring during 1989-1990
were recorded at the Olson farmyard, 1.5 miles from several wells in North Dakota. A
maximum 1-hour average concentration of 88 ppb was recorded. Data were also obtained
from September 4, 1990, to August 3, 1992, from a monitor in the town of Plaza, North
Dakota within 2 miles of several wells and tank batteries. The maximum concentration
recorded on this monitor was 358 ppb, in 1991, with one violation of the NDAAQS
recorded.
m-26
-------
1980 1981 1982 1983 1984 1985 1986 1987 1988 1989 1990 1991 1992
Year
• Oppb
Q 1-10 ppb
£3 11-50 ppb
n >50 ppb
monitor detection limit = Ippb
Source: Personal correspondence, D. Hannan, NDSDH & CL, 8/11792.
Figure IH-S. Percentage of times designated HgS concentrations were
measured at the Theodore Roosevelt National Park -
North Unit monitoring site.
m-27
-------
1
W]
^
J
V,
' NNE
NNW NORTH g- jg7
^<** — 1 r->n
NW S
/\
' \
V
i /_^NE
J Field
A
/ X8839
/ \ VqiSP/i/ ./ \V
mm/
"r*-
\jy
x>s
) Mogi|
^
/ to
wsw\^9i60/Y°
sw\
.4,
•^C.862
>s^ 5ni
6 ~
/ 8486
rvf^^'^ I
Kj 8544^-
\BatSy CC
j^S^SE
3 J J "T
THI SSE
8915^-
9551 o
O
AENE
•N
-4 EAST
/8838
T/ O
mnreasor
o
9062
9063
9509 8g4
O 951
(.
File No.
o
O
o
8545
0
9162J
N
}
N
!
1
r\
LJ
9016
O
O
o
!
MM) _j-
5
z
1
Av
W
b b
00 «J
'i
Operator
Well Name Well No.
8245 Chevron USA. Inc. Bob Creek Federal Unit 1-13-3B
8414 Chevron USA, Inc. Carus Federal 1-30-1C
8486 Apache Corp. Federal 18 1
8543 Chevron USA, Inc. Bob Creek Federal Unit 4-19-1D
8544 Chevron USA. Inc. Bob Creek Federal (Comm) 2-18-48
8545 Chevron USA. Inc. Carus Federal 2-30-4B
8623 Chevron USA, Inc. Bob Creek Federal 3-24-2A
8838 Apache Corp. Carus Amoco Unit "A" 1
8839 Apache Corp. Carus Amoco Unit "C" (SWD) 1
8915 Apache Corp. Lone Butte Federal Amoco "A" 1
9016 Apache Corp. Carus Amoco "B" 1
9062 Apache Corp. Lone Butte Federal Amoco Ti 1
9063 Apache Corp. Federal 19-1-
9159 Texaco, Inc. Bob Creek ' 13-8
9160 Texaco. Inc.
9162 Texaco, Inc.
9167 Texaco. Inc.
Bob Creek 13-11
Bob Creek 36-1
Bob Creek "-13
9509 Chevron USA, Inc. Mormon Butte 5-25-2B
9510 Chevron USA, Inc. Mormon Butte Federal 3-25-3B
9551 Chevron USA. Inc. Foley-Stewart Federal 4-25-3C
N/A North Dakota State
Dept. of Health Air Monitor
N/A Koch Hvdrocarbon Compressor Station
N/A Chevron USA. Inc. Tank Battery
Source: Personal correspondence, D. Harman. NDSDH & CL, 8/11/92.
Figure III-9. Wells producing between July 1986 and December 1987
surrounding Lone Butte H>S ambient air monitoring site.
HL-28
-------
1984 1985 1986 1987
Year
1988
1989
• Oppb
H 1-10 ppb
D 11-50 ppb
D 51-100 ppb
53 101-200 ppb
il >200 ppb
monitor detection limit = Ippb
Source: Personal correspondence, D. Hannan, NDSDH & CL, 8/11/92.
Figure 111-10. Percentage of times designated H^S concentrations were
measured at the Lone Butte monitoring site.
m-29
-------
Only four months of monitoring data from the Roffler site were received by
NDSDH&CL. dating from April 11, 1980, to September 29, 1980. Located in a farmyard
within 1/2 mile of a well and tank battery, the monitor measured very low concentrations
(usually 0 ppb) with a maximum, time-weighted average of 13 ppb recorded. In contrast, at
the Jorgenson monitor, the recorded concentration was as high as 250 ppb. The Jorgenson
monitor was located in a valley within one mile of several wells, and the data received dated
from October 2, 1980, to May 13, 1982. Data from sixteen months of monitoring, from
June 30, 1982, to October 31, 1983, were received for the Kadrmas site. Located in a
farmyard within a mile of several wells, the maximum half-hour averages recorded were 541
ppb, in 1982, and 353 ppb, in 1983. • From these three studies, an analysis was performed
on the monitoring data in comparison to the 32 ppb half-hour standard. The results showed
that the concentration of hydrogen sulfide never exceeded the NDAAQS during the four
months of the Roffler study. Conversely, at the Jorgenson site, the 32 ppb standard was
violated 16 times in 1980, 38 times in 1981, and 26 times in 1982. At the Kadrmas site, the
violation count was 18 times in 1982 and 14 times in 1983.
Williston Basin Study. The Williston Basin Regional Air Quality Study was
undertaken in the late 1980s to assess the air quality impact of oil and gas production in ^
western North Dakota (ELM, 1990). Emissions inventories were prepared and air quality
models were applied to project the impact of sulfur dioxide and hydrogen sulfide emissions
in these 12 selected oil fields with respect to applicable ambient air quality standards and
PSD increments. Study results suggested that exceedances of both sulfur dioxide and
hydrogen sulfide ambient air quality standards could be expected for some fields.
Exceedances of Class I PSD increments for sulfur dioxide were expected for three of the four
Class I areas studied. Further development of the oil and gas fields, where the emissions of
sulfur dioxide and hydrogen sulfide would be possible, would not be permitted unless these
exceedances were addressed.
To arrive at estimated hydrogen sulfide concentrations for the study, two types of
hydrogen sulfide emissions were considered. First a hydrogen sulfide concentration was
obtained through back calculation of the output sulfur dioxide concentrations from the
Industrial Source Complex Model. The predicted sulfur dioxide concentrations were the
result of modeled dispersion of the point source emissions from heater-treaters firing on H2S
contaminated wellhead gas and from flares which burn H2S contaminated wellhead gas when
a gas gathering pipeline is not available. To provide conservative results, combustion
efficiency of 75 percent was used in these calculations, meaning that 25 percent of the
hydrogen sulfide remained unchanged. [Note: As stated in Chapter H, flares, in most
applications, operate at 95 to 99 percent efficiency.] The second emission source used
represented fugitive emissions from leaky valves, tank hatches or pipe connections. These
fugitive sources were estimated as contributing a background concentration of 7 /xg/m3 (50
ppb), derived from the 99th percentile of the 1-hr average monitored ambient air
concentrations at three remote monitor locations (the Theodore Roosevelt National Park's
two sites and the Lostwood site) during portions of 1987 and 1988.
m-30
-------
At the time of the study, the NDAAQS for H2S was 50 ppb 1-hour average
concentration not to be exceeded more than once a year. NDAAQS exceedances were
predicted for 6 of the 12 fields studied using current emissions estimates, with exceedances
predicted for 7 of the 12 fields using future emissions estimates. Of the sites where
modeling suggested NDAAQS exceedances, the yearly second highest (the first occurrence of
ambient hydrogen sulfide concentrations above 50 ppb would be allowed by the law)
expected concentrations exceeded 700 ppb for the Lost Bridge Field and 900 ppb for the
Rough Rider Field.
Modeling results are only an estimate and are often considered accurate when they are
within a factor of two of the actual ambient concentrations. Except for the Lone Butte Field,
ambient monitoring data were not available for the other fields to verify or contradict the —
modeled estimates.
Conclusions. At several locations, for example, Lostwood and the Theodore
Roosevelt-south unit, the monitoring program served as a verification that the air quality was
within the levels allowed by the law. In two cases, the monitoring programs were of too
short of a duration to support any conclusions. When an area is monitored for a short period
of time, as at the Roffler and Olson sites, the full range of meteorological conditions and
emissions scenarios are not represented in the ambient air measurements. Monitoring was
discontinued at Jorgenson and Kadrmas (both monitored in the early 1980s) and at Lone
Butte (the site with the worst air quality) even though numerous NDAAQS violations were
experienced during their last monitored year. This occurred because rigorous inspection and
maintenance scheduling was established and/or the data indicated no air quality problems
existed (personal communication, D. Harman, NDSDH&CL, 11/9/92).
Ambient concentrations of hydrogen sulfide varied for the sites, with maximum yearly
concentrations ranging from half-hour averages, below the 1 ppb detection limit, to 2734 ppb
(2.734 x 103 ppb). Two common factors were the median and mode values. For all of the
monitoring data received from North Dakota, the median and mode values were 0 ppb. In
other words, for each site more than half of all observations recorded below the 1 ppb
monitor detection limit.
Severity of Consequences. No epidemiological studies have been carried out to assess
the effects of hydrogen sulfide exposure resulting from the production of oil and gas. Many
States have enacted ambient air quality standards based upon odor for hydrogen sulfide, since
its odor recognition threshold is so low (i.e., 3 to 20 ppb).
Annual average H2S concentrations, which can more appropriately be compared to a
long-term concentration benchmark such as the RfC, were also calculated from the Lone
Butte site. These values exceeded the RfC by about an order of magnitude from 1984-1987,
dropping to about the RfC level in 1988 and 1989. Since these values indicate the combined
impacts of 9 separate wells, it is reasonable to conclude that: 1) the long-term impact of
routine releases from any individual well is probably not significantly greater than the RfC;
m-31
-------
and 2) the use of a gas-collection system with manifolded flares and rigorous operation and
maintenance programs can significantly reduce long-term H2S impacts.
At low concentrations, odor nuisance and eye and respiratory tract irritation are the
consequences of exposure rather than the toxic properties of the gas. An explanation for an
increased perception of ill health could be related to low level exposure to hydrogen sulfide
and pulmonary infections. A study by Rogers and Ferin (1981) concluded that hydrogen
sulfide significantly affected the antibacterial system of rats by impairing pulmonary
macrophage. However, additional research would be required before any definitive
judgements could be made in human exposure scenarios.
Elevated ambient concentrations in two episodes (one in-the Great Kanaw.ha,River
Valley, WV, in 1950, and one in Terre Haute, IN, in 1964) were reported as 0.41 mg/m
(293 ppb) and 0.46 mg/m3 (329 ppb), respectively (West Virginia Department of Health,
1952; U.S. Public Health Service, 1964). These incidents did not result from oil and gas
production; however, the ambient concentrations recorded were comparable to some
measurements in North Dakota. General symptoms of malaise, irritability, headache,
insomnia, and nausea were'reported by exposed populations. In the Terre Haute incident,
levels measured at a nearby lagoon ranged from 2 x 103 to 8 x 103 ppb). The most common
symptoms reported were offensive odor, foul-tasting water, nausea, vomiting, diarrhea,
throat irritation, shortness of breath, burning eyes and asthma. Milder symptoms included
cough, headache, anorexia, acute asthma attacks, nervousness, weight loss, fever, gagging
and heaviness of chest. The symptoms ceased when the odor disappeared. In an episode in
Alton, IL in 1973 similar symptoms were reported (Illinois Institute for Environmental
Quality, 1974; NRG, 1979). Ambient hydrogen sulfide levels ranged from 25 ppb to higher
than 1 x 103 ppb. Other contaminants, such as ozone and nitrogen oxides were also detected
during this episode (Hoyle, 1973).
A study of the levels of sulfur compounds in vegetation near the Lone Butte oilfield
and Theodore Roosevelt National Park, was conducted during the summer of 1987
(Bilderback, 1988). The study's conclusions confirmed what ambient monitoring had
suggested: the South Unit of the national park may have been impacted by moderately high
levels of atmospheric sulfur pollution, and the Lone Butte oil field was impacted hy high
levels of reactive atmospheric sulfur. Visible signs of vegetation damage were also detected
at the Lone Butte oilfield. Furthermore, Bilderback attributes the elevated levels of hydrogen
sulfide at the North Unit of the Park to the Lone Butte oilfield.
Consequence Analysis — Routine Emissions
As described in Chapter H, several potential sources of routine H2S emissions can be
found at oil and gas production facilities. Figures m-11 and m-12 indicate that 8 States have
a significant overlaps of well fields and H2S prone areas. Using the estimated number of
producing wells in these States (Figure DI-12) as a conservative measure, it appears that as
many as 280.000 oil wells and 54,000 gas wells have the potential for location in an H2S
- , m-32 .
-------
0
Source: IOGCC, 1990.
Figure III-ll. Oil and gas fields.
m-33
-------
# Gas wells
_._ _ ,
US Total
601.520
268J367
N/A = Data not available
0 3 No wells registered in State
Source: Gas Research Institute, 1990.
Figure 111-12. Major B^S prone areas shown in relation to number of
producing oil and gas wells in 1990.
m-34
-------
prone area. Although only a fraction of these wells would actually be sour, these figures
imply that the potential for routine H2S emissions is significant. However, no national
statistics are available to predict the probability of such emissions. The only record of
routine emissions identified is ambient air quality monitoring data from the State of North
Dakota. Nine monitoring studies in 12 years resulted in more than 3,300 violations of the
NDAAQS. The majority of these violations occurred when the standard was developed
based on the more conservative odor threshold rather than on health considerations. Only
one violation was recorded after the health-based (higher concentration limits) standards were
implemented.
A routine emission scenario would be the incomplete combustion of the 'wellhead
gases, allowing some percentage of the hydrogen sulfide to be emitted. In-the. oilfields .of
North Dakota, the concentration of hydrogen sulfide in waste gas stream to flares can reach
30 percent, with the conversion efficiencies of the flaring operations varying from 30 to 100
percent (NDSDH&CL, 1983). (Note, however, that in Chapter n, the common efficiency of
a flare, regardless of industrial application is 95-99 percent.) This scenerio would result in
releases of 0 to 70 percent of the hydrogen sulfide contained in the wellhead gas. In western
North Dakota, the amount of natural gas flared exceeded 1 million cubic feet per month in
mid-1982, dropping to less than half of that amount by mid-1985, as more wells were tied
into a central gas collection system (Liebsch, 1985). As a worst case scenario, if the gas
content were 30 percent hydrogen sulfide, and the combustion efficiency were 30 percent (70
percent of the hydrogen sulfide was emitted unconverted), 210,000 cubic feet of hydrogen
sulfide per month could have been routinely emitted in the mid-1982 time period.
No H2S health or ecological effects studies have been conducted which specifically
target oil and gas production. The most common consequences of exposure to routine
emissions of H2S are the odor nuisance and eye and respiratory tract irritation.
Exposure Analysis—Accidental Releases
The discussion of accidental releases begins with a description of examples of
accidental releases of sour oil and gas in the United States that have impacted the public and
wildlife. These examples are then supplemented by calculations of the consequences of a
series of hypothetical accident scenarios using atmospheric dispersion models. The risk to
the public from an accidental release of H2S is a function of both the potential consequences
and the likelihood of occurrence of an accidental release. Risks from a major accidental H2S
release will vary from facility to facility depending on site-specific factors such as the
population density and distribution of nearby populations and the quality of process safety
management and risk management practiced at the facility. Since risk is a product of both
consequences and likelihood, risk.reduction must take both into account. The accidental
release discussion concludes with an assessment of accident prevention, mitigation, and
emergency procedure measures that, if systematically implemented, could help to prevent or
reduce the likelihood of accidental releases of H2S from sour oil and gas, and mitigate the
m-35
-------
consequences in the event that a release occurs. Supporting details for the atmospheric
dispersion calculations may be found in Appendix C.
Accidental Release Records .
A variety of sources were investigated to locate documentation of accidental sour gas
releases. These sources include: Congressional testimony; literature searches; database
searches; state regulatory authorities; emergency response organizations; and industry
officials. No national statistics regarding sour oil or gas releases were identified. Data base
sources were the Accidental Release Information Program (ARIP) database which is
maintained by the EPA, the Acute Hazardous Events (AHE) database which was developed
by EPA, and the Emergency Response Notification .System. (ERNS) database. ARIP has
records of chemical accidental releases that have occurred since October 1986 with some
detailed information on accident cause. AHE has incident records covering the time period
1982 to 1986 and was developed from various sources including press reports, spill reports
to the National Response Center, and some state and EPA regional office records. ERNS
contains records of releases reported to the National Response Center.
A review of available sources revealed several documented examples of incidents in
oil and gas extraction operations in the United States where accidental releases of H2S have
impacted the public and/or the environment since 1974. There was also a very large sour
gas release that caused some environmental damage in Alberta, Canada during this time
period. Examples of some of these accidents are summarized in Table IQ-4. It should be
noted that these incidents include two accidents related to carbon dioxide injection to improve
recovery rather than from the accidental releases of sour natural gas. One of these accidents
resulted in eight fatalities, and another accident resulted in two injuries. The other incident
resulting in fatalities was the result of fire associated with a natural gas release. However,
effects on the public that are directly related to oil and gas extraction activities have most
often been limited to evacuation. Isolated incidents resulting in hospital treatment have also
occurred. Evacuation may occur as a conservative measure whether or not a life-threatening
situation exists. There have been several documented incidents involving livestock and
wildlife fatalities. In addition to toxicity, the flammability of accidental releases of sour oil
and gas may also present a significant hazard.
Information from the State of Texas shows that there were 145 incidents of sour oil
and gas release during the years 1985 through 1992 (Hall, 1992). These accidents were
generally related to sour oil and gas rather than specifically from extraction activities. In
these incidents, there were 10 deaths (all occupational), and 109 injuries (100 occupational
and 9 public). The Texas incidents may be illustrative of the relative hazard to operating
personnel, the general public, and the environment. These statistics indicate that the major
hazard from oil and gas operations involving H2S would be to workers rather than the public
or wildlife. Workers are more often in close proximity to the wells and associated
equipment.
ffl-36
-------
Table in-4. Examples of Accidental Releases of HjS from Oil and Gas Extraction Operations
with Impact on the Public or Environment
Date and
Location
Effects on Public
Effects on
Environment
Comments
Source
i/21/74
Meridian, MS
!/2/75
)enver City, TX
i/21/81
Jig Piney, WY
0/7/82
Jalgary,
'anada
/88
.ea County, NM
'20/89
curry County,
X
'20/90
aplata County,
0
'29/90
eidelberg, MS
16/91
unbert, MT
19/91
jakum County,
£
1/17/91
lines County,
5 deaths due to
associated fire
8 occupants of
house 200 ft from
well were overcome
by the gas and died.
No impact.on public
No impact on public
1 person physically
incapacitated
Evacuation of nearby
residents, 2 treated
at hospital
2 mile radius
evacuated
No deaths but
2,000 local
residents were
evacuated
12 people were
evacuated
None identified
None identified
40 acres burned
None identified
Deaths of some
jackrabbits and
blackbirds
A number of
moose and other
large animals died
1 horse died
None identified
None identified
None identified
None identified
7 cows, 1 coyote,
and rabbits died
Unspecified
number of wildlife
died
Sour gas gathering pipeline
rupture and subsequent
fire
Gas escaped from gas
injection well. Gas was
93 v/o CO2 and 5 v/o H,,S.
Well blowout lasting 8 days.
Nearest residence was
2 miles awav
Release of 10 million ft3
!L,S per day of accident
An individual changing a
tire was overcome with HjS
CO2 injection line rupture
Well leak
Well blowout and
consequent fire
Incident was caused by
corrosion of gathering line.
Evacuation due to smell.
Sour gas gathering line
rupture, 1.2% I^S
Sour gas gathering line
rupture, approximately
6%IL,S
Texas Oil and Gas
Pipeline Corpora-
tion, 1976
Layton et al., 1983
Layton et al., 1983
Oil Daily, 1982
Corresoondence
NM Oil Conserva-
tion Division, 1992
Texas Railroad
Commision
Hall, 93
ERNS, National
Response Center
Report #01425
Platt's Oilgram
News, 1990
National Response
Center
Texas Railroad
Commision
Hall, 93
Texas Railroad
Commision
Hall, 93
m-37
-------
Atmospheric Dispersion Analysis
Atmospheric dispersion analyses of sour oil and gas releases by computer model were
both reviewed in the literature and conducted. The following issues are discussed prior to
analyses of the consequences of sour gas release scenarios:
• Choice of scenarios;
• Sour gas composition and density;
• Behavior of sour gas upon release; and
• Choice of atmospheric dispersion models.
Choice of Scenarios. The objective in choosing scenarios was Jo., investigate, a.,._ _ ,
representative range of potential accidental release situations including hypothetical worst
case scenarios. Scenarios for atmospheric dispersion analysis were chosen from documented
accidental releases, expressions of public concern, and literature analyses in which dispersion
models were applied to sour gas release scenarios.
The accidental sour gas releases documented in the previous section show some
common causes. Well blowouts and line releases are examples of accidents that have
occurred and resulted in offsite impact. Therefore, these accident scenarios were included in
the atmospheric dispersion analyses. Investigation of some public complaints resulted in
concerns regarding sour gas releases from extinguished flares, collection of sour gas in low-
lying areas, leakage from temporarily abandoned or idle wells, and line leakage
(NDSDH&CL, 1989; U.S. EPA, 1992). These concerns were also investigated as
accidental release scenarios.
Several literature sources provided descriptions of hazards associated with the
operation of sour oil and gas wells in addition to sour gas dispersion analysis to support
scenario development. Hazard/risk analyses and data on the composition of sour gas of wells
in Alberta, Canada (Alp et al., 1990), southwest Wyoming and northern Utah (Quest, 1992),
and western Wyoming and adjoining areas of Utah and Idaho (Layton et al., 1983) were
considered in the choice of scenarios. Assessments of levels of concern (LOG),
concentrations at which H2S is of concern, for acute exposure to H2S were also provided in
these sources. Although H2S alone is more dense than air, in general, the literature pertains
to sour gas mixtures that are typically less dense than air and concludes that sour gas releases
from well blowouts and line ruptures are of most concern as potential causes for levels of
concern to extend significant distances from the point of release.
Sour Gas Composition and Density. The density of sour gas mixtures is of
importance because it is one determinant of whether an accidental release will result in a
plume that travels downwind at ground level or will result in a buoyant plume that rises and
disperses. A dense plume may have a greater impact on humans and wildlife because it
remains at ground level for a period of time. The density of sour gas mixtures at
atmospheric pressure (to which accidental releases of sour gas are discharged) is dependent
m-38
-------
on the temperature and composition of the mixture. The density of a given gas mixture
increases as temperature decreases. Expansion of natural gas released from a pressurized
system results in cooling of the gas. The colder a gas, the higher its density.
There is a wide variety of potential compositions of sour gas mixtures, depending on
the reservoir. The density of these mixtures depends on their composition. In addition to
hydrogen sulfide, natural gas can also contain some or all of the following: hydrogen,
helium, carbon dioxide, nitrogen, methane, ethane, propane, isobutane, n-butane, isopentane,
n-pentane, hexanes, heptanes, and higher molecular weight hydrocarbons. The largest
component is typically methane, with hydrogen sulfide, ethane and possibly carbon dioxide
(CO2) likely to be present in significant proportions. Natural gas must contain some
proportion of hydrogen siilfide in order to be considered'sour. —- •-•-•-...-.
Figure HI-13 illustrates the variability of sour gas composition by showing the
distribution of H2S composition by number of sour gas wells in Alberta, Canada (Alp et al.,
1990). Figure IH-14 presents the same information as a function of the total number of tons
of sulfur from natural gas produced each year. The H2S composition can range from a small
fraction of a percent to over 40 percent! A statistical analysis was performed of the sulfur
composition of wells in the Overthrust Belt in western Wyoming and adjoining areas of Idaho
and Utah (Layton et al., 1983). Volume percentages of sulfur were found similar to those in
the Alberta wells. The sulfur composition ranged from less than 1 percent through 35
percent, with a mean of about 10 percent. Data on H2S in California oil and gas fields
shows fields with H2S concentrations varying from less than 1 x 105 ppb (0.01 percent) to
20 - 30 percent (Dosch and Hodgson, 1986).
In addition to increasing the density of a sour gas mixture, carbon dioxide in
sufficiently large concentrations can extinguish sour gas flares, resulting in uncombusted H2S
being released. CO2 concentrations in various parts of the Overthrust Belt were found to
vary from less than 5 percent by volume to more than 50 percent (Layton et al., 1983).
Some example sour gas compositions are presented in Table ffl-5.. Composition D is
the single composition considered representative of all the data on producing gas wells in
Alberta, Canada. Composition C is a representative gas composition produced by wells in a
southwestern Wyoming sour gas field (Quest, 1992). Data were collected for a producing
well in western North Dakota (U.S. EPA correspondence, 26 October 1992), and the
compositions of streams after processing to recover hydrocarbon condensate at that well are
given by compositions A and B of Table m-5. Composition A shows the gas composition
after high pressure separation, and Composition B shows the composition after low pressure
separation. The low pressure stream has a significantly higher H2S concentration than the
high pressure stream although its flowrate is lower.
H2S alone is more dense than air, while methane alone is less dense than air. Natural
gas mixtures of H2S and light hydrocarbons are typically less dense than air to the extent that
methane predominates in the mixture. The approximate molecular weight of air is 29. The
m-39
-------
25,000
, — 20,000
.......... 15,000
,.-. ..•—..: 10,000
5,000
I960
1965
1970
1975
Year
1980
1985
1990
Source: Alp et al.. 1990
Figure IEH3. Distribution of producing sour gas wells in Alberta by Et,S content.
m-40
-------
8,000,000
7,000,000
6,000,000— '
£ 5,000,000—'
4,000,000
3,000,000
2,000,000
1.000,000
1960
1965
1979
1975
Ye«r
1980
1985
1990
Source: Alp et al., 1990.
Figure HI-14. Total sulfur generated from producing gas wells
in Alberta by H^S composition of well.
m-41
-------
,ble III-5. Example Gas Stream Compositions
Mole Fraction
Sample Well
Molecular High Pressure (A)
Component Weight (well flow)
drogen Sulfide (H,S)
rbon Dioxide (COS)
:rogen (Ns)
•thane CCH4)
nane (C,HS)
spane (C3H,)
;ucane (C4HW)
Butane (C4H10)
'entane (CSHU)
Pencane (CSHU)
«xanes (C^)
•ptanes-h fC_Hw)
erage Molecular Weight
34
44
28
16
30
44
58
58
72
72
86 -I
100+ J
0.075. ' .
0.01
0.003
0.83
0.047
0.012
0.0032
0.0038
0.0016
0.0020
0.0034
19.25
Low Pressure (B)
(vapor recovery
systems)
0.277
0.013
-
0.45
0.10
0.064
0.024
0.026
0.011
0.0086
0.019
28.9
Composition Used Composition
in Cave Creek Risk Used by
Assessment (C) ECRB (D)
0.146
0.027
0.017
0.699
0.058
0.018
0.0042
0.0050
0.0022
0.0018
0.0031
0.0176
23.2
0.30
0.123
0.02
0.55
0.005
0.001
0.001
—
—
—
25.2
m-42
-------
two composite compositions and the high pressure stream shown in Table m-5 have
molecular weights less than 29. Thus, these streams are less dense than air at the same
temperature and pressure. CO2 is also more dense than air at similar conditions and may
cause the density of a gas mixture to be higher than that of air if present in large
concentrations. The low pressure stream has a molecular weight very close to that of air and
with some modification in composition, such as more H2S or CO2 and less methane, could be
more dense than air.
Gas mixtures which are denser than air due to high concentrations of CO2 have
caused fatalities as described in the discussion of release histories. A well blowout near Big
Piney, Wyoming, on June 21, 1981, killed small animals up to about 0.8 km from the well
(Alp et al., 1990) The gaseous emissions from the well were composed of 70 percent C02, "
20 percent methane and 3 to 4 percent H2S. It is not clear that H2S caused the animal
fatalities in this case. However, these emissions were clearly denser than air. The literature
generally describes mixtures that are less dense than air; the studies of hazards/risks
associated with sour gas (Alp et al., 1990; Quest, 1992) referred to in this report used gas
compositions that are buoyant.
In conclusion, sour gas as produced is typically buoyant. There can be atypical cases
where natural gas contains high concentrations of H2S and/or CO2 which results in a denser-
than-air mixture. Also, gas processing such as separation for condensate (liquid
hydrocarbon) recovery at the well site may affect the density of a gas stream.
Behavior of Sour Gas Upon Release. High pressure sour gas releases from well
blowouts and line ruptures are initially high momentum jets which can vary directionally
between the extremes of vertical and horizontal. The jet (high velocity) nature of such
releases is-caused by the differential pressure between the contained gas and the atmosphere
and results in entrainment of the surrounding air into the released gas. Entrainment of air
results in dilution of the released gas and causes its density to approach that of air. Thus, as
air is entrained, both positively and negatively buoyant gas mixtures with air will tend to
have densities approaching that of air. A high velocity jet (such as from a high pressure
source) will entrain air more rapidly and to a greater extent than a low velocity jet from a
low pressure source. Depending on the release conditions, it is possible for a gas mixture to
retain its initial positive or negative buoyancy. Negative buoyancy releases are of greatest
concern because of dense gas behavior and their tendency to travel to ground level where
exposure is likely to occur.
As previously discussed, the effective molecular weight (and thus, the density) of sour
gas mixtures as produced is generally less than that of air with isolated exceptions.
Therefore, models for these cases should consider the various mechanisms that describe the
near-field (near the point of release) and far-field (downwind) behavior of the plume of
released gas and its interaction with the surrounding air. In particular, the models should
contain mechanisms for simulation of the following sequence of effects occurring along a
plume of released gas from the point of release: a) near-field momentum jet modeling; b)
~ ffl-43 - - .- •-.
-------
subsequent positively-buoyant rise or negatively-buoyant sinking; c) potential for a nominally
buoyant plume that is initially on the ground to rise or, if negatively-buoyant, to stay at
ground level; and d) far-field transition to a subsequent Gaussian (passive modeling) phase.
The Gaussian or passive phase assumes random mixing in the far-field due to the action of
atmospheric turbulence; whereas, close to the source, entrainment of air is affected or
sometimes dominated by the released material itself.
Choice of Atmospheric Dispersion Models. The models reviewed in the literature for
analysis of the dispersion characteristics of sour gas were GASCON2, FOCUS, and a
Gaussian dispersion model. Confirmatory, independent atmospheric dispersion analyses were
conducted for most of the scenarios with the SAPLUME, SLAB, and DEGADIS models.
The computer model GASCON2 was specifically developed in Canada to model sour
gas releases from well blowouts and line ruptures (Alp et al., 1990). The model incorporates
high pressure gas jet releases, plume rise or sinking (depending on density) and subsequent
passive atmospheric dispersion. GASCON! was validated by comparison with experiment.
The associated literature also contains extensive discussions on uncertainties and the work
was reviewed by a science advisory board.
The proprietary model, FOCUS, contains a treatment of momentum and buoyancy
effects and transition to subsequent passive atmospheric dispersion (Quest, 1992). The model
has been available for several years and has been used in a number of risk assessments of
toxic and flammable vapors.
The Gaussian dispersion model is suitable for passive releases (Layton et al., 1983).
Therefore, jet momentum effects are neglected and the results are not expected to be reliable
close to the emission source. However, at large distances where low concentrations of H2S
would result (e.g., in the low part per million range), all three of the above models should
converge to similar results.
A well-established model developed by Ooms (1974, 1983) for jet releases of vapors
can model the dispersion of both buoyant and heavier-than-air momentum jets. The EPA has
sponsored the incorporation of the Ooms model into the well-known DEGADIS model
(Spicer, 1988), which can only simulate vertical, but not horizontal releases. Another
proprietary model, SAPLUME, is also based on the Ooms model and can simulate jets at any
orientation (SAIC, 1990).
SLAB was developed by Lawrence Livermore National Laboratory (Ermak, 1989).
This computer model also accepts jets of vertical or horizontal orientation. However, it was
specifically developed for heavy vapors and has not been carefully validated for use with
buoyant plumes, so results must be interpreted with care.
m-44
-------
Consequence Analysis — Accidental Releases
In the following sections, the consequences of accidental releases for a variety of
scenarios are presented.
It should be noted that the calculated consequences of some of the modeled scenarios
are based on very conservative assumptions in order to examine the worst case. The
worst-case scenario is designed to generate the maximum impact off-site. It is considered to
be extremely unlikely and does not take into account a variety of factors that can significantly
reduce downwind impacts. However, the worst-case scenario is useful to facilities and
communities surrounding facilities in gaining an understanding of the potential magnitude of
severe situations. The potential for severe consequences should be taken into account along
with more probable scenarios when setting priorities for community emergency planning.
Consequence Analysis of Jets from Well Blowouts
Figure .m-15 shows the layout of a typical completed sour gas well. A well blowout
is an uncontrolled release from a well during drilling, servicing, or production operations.
Such an accident could occur if a blowout preventer failed during drilling or a subsurface
safety valve fails to operate during production. The possible types of flow from a ruptured
well are shown in Figure IQ-16. A useful simplification is that an accidental release into the
casing is possible during drilling or servicing, while flow would likely be restricted to the
production tube if there were a blowout during normal production operations. Potential flow
orientations are shown on Figure DI-17. Examples evaluated for the purposes of this study
included the extremes of a vertical jet and a horizontal downwind jet.
Flow rates for the scenarios identified in Figure El-16 are functions of such items as
rock permeabilities, gas properties, depth, and tubing and casing diameters. Overall, there
are large variabilities in these parameters. One measure of the potential rate of flow from a
well is the Calculated Absolute Open Flow Rate (CAOF), which is the rate of flow of gas
into the well bore when the pressure is atmospheric. This measure represents a maximum
possible flow rate. The actual flow rates out of a ruptured well will be less than the CAOF
because of frictional effects in the pipework. Thus, the use of CAOF for a release rate is
conservative. Table IEI-6 gives some representative examples of how the CAOF is reduced
for a specific set of well parameters. A flow rate of 2xl07 standard cubic feet per day
(scf/d) was chosen for representative calculations, with a flow rate of 108 scf/d being taken
as an example of a very high flow rate. The bases for these assumptions are presented in
Appendix C.
For the scenarios analyzed for this report, it was assumed that the gas emerges as a
vapor. Since typical pressures are very high (e.g., in excess of 1,000 pounds per square
inch gauge (psig)), the flow is choked .(limited) at sonic velocity.
m-45
-------
Surface
" N'N'N'V>'XVN'v'X'
' * S t t f
•\ ^ ^ •
\ x '
f f
S '
Cement grout seals
space between formations
and casings -
4
/
-------
Casing
Flow
Combined
Flow
Annular
Flow
Tubing
Flow
Casing
Source: Alp et al., 1990.
Figure IQ-16. Possible well flow scenarios.
Downwind Jet
Vertical Jet
Wind Direction
Cloud
Source: Aip et al., 1990.
Figure ffl-17. Possible well accidental release geometries;
m-47
-------
Table HI-6. Surface Deliverafaility as a Function of Weil CAOF
CAOF
(101 mVd)"
Casing
Flow
Annulus and
Tubing Flow
Annulus
Flow
Tubing
- Flow
5000
1000
500
100
50
57.4%
95.0
98.0
99.0
99.2
52.2%
82.5
96.0 -
98.0
99.2
39.3%
76.0
92.0
97.0
99.2
8.5%
26.0-
46.0
90.0
98.6
Source: Alp et ah, 1990.
* At 15°C and lOl.SkPa.
The values in Table III-7 were based on the following well conditions:
Well depth (m) 2660
Casing inside diameter (mm) 156.3
Tubing outside diameter (mm) 73.0
Tubing inside diameter (mm) 62.0
Reservoir pressure (kPa) 15,900
Reservoir temperature (°C) 75
m-48
-------
The temperature of the gas in the well prior to expansion to atmospheric pressure
through the rupture depends on the depth of the gas reservoir. The amount of cooling
depends on the initial pressure and the composition. For the purposes of this analysis, an
expanded gas temperature of 0°C (32°F) was assumed. This assumption is further discussed
in Appendix C.
For a well blowout, the release could continue indefinitely. For illustrative purposes,
it was assumed that any nearby individuals could be evacuated within one hour. The
calculations of distances of concern discussed below assume that the duration of release and
possible duration of exposure is one hour.
For vertical releases of-sour gas from well blowouts, the independent, dispersion
modeling (SLAB, DEGADIS, SAPLUME, and the Gaussian model) and results reported in
the literature (Alp et al., 1990; Quest, 1992) indicate that there will be no concentrations
above levels of concern at ground level, either at the emergency counter-measure (ERPG-2)
or potential fatality (LC01) level. The jet is oriented upwards and, for either buoyant or
negatively buoyant sour gas, dilutes rapidly due to its high momentum.
For horizontal releases from well blowouts, results calculated using the SLAB and
SAPLUME models are given in Table DI-7 for low wind speed and stable conditions.
Releases in the direction of the wind were assumed. Depending on composition, release
rate, and the model used, distances to the LC0j range from 700 meters (approximately 0.4
miles) to greater than 10 kilometers (approximately 6 miles). Distances to the ERPG-2 range
from 2.8 kilometers (approximately 1.7 miles) to greater than 10 kilometers (approximately 6
miles). The atmospheric conditions input into the models represent conditions of high
stability and little atmospheric mixing. Thus, these conditions represent the "worst-case"
because levels of concern will be exceeded for predicted distances from the point of release
that will exceed those for other weather conditions. The results were calculated neglecting .
the possibility of slight buoyancy of the plume even after dilution. DEGADIS results are not
quoted because the jet module of that computer model can only handle vertical releases. For
all the models, results in the range greater than 10 km (6 miles) should be regarded as
beyond the limit of validity and probably conservative (see below).
For comparison, the GASCON2 model calculates an estimated distance of 1.6 km (1
mile) to the LCOI for a composition D flow rate of 2.4x105 m3/d (cubic meters per day), or
8.5xl06 scf/d, and an estimated distance of approximately 5 km (3 miles) for a composition
D flow rate of 9.5xl05 m3/d (3.4xl07 scf/d)(Alp et ai., 1990). From Table HI-7, for
composition D with a flowrate of 6xl05 m3/d (2.1xl07 scf/d), SLAB and SAPLUME predict
a distance of 2.9 km and 3 km (both approximately 1.8 miles) to the LC01, respectively.
These distances and release rates are intermediate to those values in the GASCON2 model.
Therefore, the results calculated with GASCON2 are consistent with the results generated by
SLAB and SAPLUME (to within the uncertainties expected in such models).
m-49
-------
Table EI-7. SLAB and SAPLUME Results - Horizontal Releases
from a Well Blowout
Composition (from Table Hi-fi)
and Flow Rates (m3/d)
A, 6 x 10s m3/d (7.5% H,,S)
B, 6 x 10s m3/d (27% IL,S)
C, 6 x 10s m3/d (15% HoS)
D, 6 x 10s m3/d (30% H,,S)
D, 3 x 10€ m3/d (30% IL,S)
(extreme case)
Predicted Distance
1 h Exposure
(SLAB)
LCoi
700m
2.8km
1.5km
2.9km
7km
ERPG-2
2.8km
7km
4.7km
7km
>10km
Predicted Distance
1 h Exposure
(SAPLUME)
LC01
1km
2.7km
1.5km
3km
>10km
EKPG-2
3.1km
10km
'5.7km
10km
>10km
nr-5o
-------
By contrast, the FOCUS model calculates an estimated distance of 0.7 km to the LCOI
for composition C with a flow rate of 6xl05 nWd (2.1xl07 scf/d)(Quest, 1992). This
prediction is about half that given by the SLAB and SAPLUME calculations, which predict a
distance of 1.5 km (0.9 miles) to the LC01 for composition C with a flowrate of 6xl05 mj/d
(by implication, GASCON2 would predict similar distances). This difference in predictions
may lie within the range of uncertainty of vapor dispersion models; the precise reason for the
difference cannot be determined from the information available about the proprietary model
FOCUS.
Figure HE-18 shows the results of the comparison of observations from actual well
blowouts in Alberta, Canada, with GASCON2 predictions. The actual blowouts were at
Lodgepole (October 17 through December 23, 1982), Clovesholm (September 24-28, 1984)
and Rainbow Lake (December 9-14, 1985). The air quality data associated with each
blow-out were collected with public safety interests in mind and not model verification or
validation. As a consequence, most of the observations were poorly documented wifh respect
to magnitude, location,-averaging time and meteorological conditions. Screening of the data
to select only measurements in which there could be reasonable confidence produced a data
set of 50 (45 of which were from the Lodgepole blowout). For the Lodgepole case, seven
stationary and five mobile units collected data within 50 km of the site.
As can be seen, GASCON2 significantly overpredicts, especially when its predicted
concentrations are in the greater than 3 x 104 ppb range, where overproductions are by as
much as a factor of 10. This concentration is the range of interest for ERPG-2 and LC01.
These overpredictions tentatively (because of the poor quality of the data) suggest that the
GASCON2 results are conservative and, by implication, that the results from the SLAB and
SAPLUME calculations are also conservative.
Possible reasons for conservatism include underestimating the effect of the plume
lifting off the ground. For distances in the several km to the greater than 10 km (6 mile)
range, neglect of dry deposition (fallout, transfer from the air to other surfaces) of the highly
reactive H2S may also lead to overestimation of airborne concentrations. However, it is
more likely that the poor quality of the observations is reponsible for the apparent
disagreements.
Standard text-book calculations indicate that flammable mixtures will not propagate
more than 100 m from the point of release (Quest, 1992). If ignition occurs, potentially fatal
thermal radiation loads could be received up to approximately 100 meters from the source.
Although not pertinent to a discussion of hazards from H2S releases, it should be noted that
SO2 will be emitted as a result of igniting a sour gas stream and may present a toxicity
hazard.
m-51
-------
e
5.
o
•o
Mobile !?2S (3 minute)
Stationary H2S (1 hour)
Stationary K^S (3 hour)
10 20 30 40 50 60 70
Observed Concentration (ppm)
80
90
100
Bottom Left Comer
Bottom Left Comer
2 4 6 8 10 12 14 IS 18 20
Observed Concentration (ppmi
s
'= 0.3
0.2
0 0.1 0.2 0.3 0.4
Observed Conceotrauoo (ppm>
o.s
Figure 111-18. Predicted HgS and SO2 concentrations for selected well blowout
observations.
m-52
-------
Consequence Analysis of Line Ruptures
Releases from line ruptures will behave much like well blowouts unless there is a
means to isolate the rupture. Most gathering systems are not equipped with isolation
systems, and aging pipework presents integrity concerns (particularly when not properly
maintained). Advanced gathering line systems may have emergency shutdown valves (ESDs)
that are remotely or locally operated. ESDs may be manually or automatically operated
(e.g., by a signal from an H2S detector). Figures ffl-19 and m-20 show some typical
configurations .for raptures of lines that are equipped with ESDs. For such releases, the total
mass released is limited by the quantity of gas between ESDs. .The valves may be 1 km to 3
km apart (0.6 mile to 1.8 mile) (Alp et al., 1990).
Figure ffl-21 shows typical mass release rates for the rupture cases identified in
Figure ni-20, assuming a 6" diameter pipe at a pressure of approximately 5,000 kPa (725
psi). Rupture Scenario 4 (no ESD) follows Scenario 1 until a steady state of 2.4x10* m3/d
(8.5xl06 scf/d) is reached after about a minute.
Figure ffl-22 shows mass release rates as a function of time for various pipe
diameters and various ESD separations with an assumed line pressure of approximately 50
atmospheres (735 psi). The variable, td, listed on Figure ni-22 is the time in seconds taken
for 99 percent of the line contents to be depleted after closure of the ESD valves. Md is the
total mass released in kg. As can be seen, for many of the cases, a puff release (rather than
a continuous release) is a reasonable approximation because of the short duration.
The predicted distances of concern for lines with ESD valves that close promptly are
smaller than those for wellhead blowouts because the duration of release is shorter, the total
mass released is smaller, and because shorter exposure times allow higher tolerable levels of
concern.
Calculations from SADENZ, a companion model to SAPLUME for puff releases,
predict that distances to the LC0, for compositions A-D in Table 3H-5 and released masses
specified in Figure IH-22 range from 600 m (0.4 miles) to 4.3 km (2.6 miles). Predicted
distances to the ERPG-2 adjusted for shorter exposure time (method described by Gephart
and Moses, 1989) range from 750 m (0.45 miles) to approximately 5.6 km (3.4 miles). This
is consistent with the calculated results from the GASCON2 model (Alp et al., 1990) and, as
before, somewhat higher than those calculated from the FOCUS model (Quest, 1992).
Consequence Analysis of Line Release Seepage
A survey of several gas pipeline incidents .that were investigated by the National
Transportation Safety Board (NTSB) indicated that, for buried gas pipelines operating above
600 psig, a 1" diameter hole will blow away the soil above the line (Quest, 1992). This will
result in the formation of a crater from which the gas will escape as an unobstructed jet.
For smaller holes (e.g., a 1/4" diameter hole caused by corrosion), the soil remains in place
m-53
-------
End Pipe Rupture
Centre Pipe Rupture
( )•
ESO Value ESO Value f/-x~^J/ ESO Value ESO Value ESO Value / J ESO Value
Operates Operates [/ Operates Operates Operates j / Operates
ESD Valve FaUure
ESO Value ESO Value
Operates Operates
Decay to Steady State
ESO Value ESO Value
Operates Operates
Source: Alp et al., 1990.
Figure IH-19. Possible pipeline rupture scenarios.
ESO Value ESO Value
Operates Operates
LongRupture
Short Rupture
^XXXXXXXX^V/.XXXXXXXXX
KXXXXXXXXXXXXXXXXXXXXX>
Leak
Source: Alp et al.. 1990.
Figure H£-20. Possible pipeline release geometries.
IH-54
-------
-5
ec
ft
a
m
a
1. End Rupture (1 x 1000 m)
2. Valve Failure (2 x 1000 m)
3. Centre Rupture <2 x 500 m)
Time After Rupture
Source: Alp et al., 1990.
Figure ni-21. Predicted mass release rates - rupture of 6" pipe.
600
500 -
400 -
300 -
100 -
Surface Pipeline Release
Pipe Size Sensitivity
1. 1000 m. 154.1 mm ID (6") Pipe
2. 1000 m. 102.3 mm ID <4") Pipe
3. 1000 m, 202.7 mm ID (8") Pipe
4. '1000 m. 254.5 nun ID (10") Pipe
5. 3000m. 254.5 mm ID (10") Pipe
6. 3000 m, 330.2 mm ID (14') IPipe
7. 3000 m. 381.0 mm ID (16") Pipe
m
94
16
82
74
383
334
310
Ma
1450
640
2600
4000
1.2 x 104
2.4 x 104
3.1 x 104
0 30 60 90 120 150
Time After Rupture (*)
Source: Alp et al., 1990.
Figure 113-22. Predicted mass release" rates - rupture of pipes of differing diameters.
m-55
-------
and the vapors migrate to the surface where they are released without any momentum
(although the resulting vapor cloud may still be buoyant enough to lift off).
For a 1/4" diameter hole in a line containing gas at a pressure of 1,000 psig, the
calculated release rate (using standard text book formulae for choked flow) is about 1 Ib/sec,
assuming that the surrounding soil does not reduce the emission rate by physically impeding
the flow. If this gas seeps to the surface, the predicted distance to which the ERPG-2 would
be exceeded for a person who inadvertently enters the plume for five minutes is about 400 m
(0.2 miles) and the predicted distance to the LC0, is about 250 m (0.15 miles) when the
atmospheric stability category is F and the windspeed is 1.5.m/sec (4.9 feet per second),
utilizing composition C from Table ITJ-5. These results neglect the possibility that the plume
might lift off the ground or exhibit dense, gas behavior.
Consequence Analysis of Flare Stack Releases
Results calculated using the GASCON2 (Alp et al., 1990) and FOCUS (Quest, 1992)
models and those carried out independently with the SAPLUME model show that, with or
without sour gas ignition, the plume emitted from a flare stack is a momentum jet with
dilution of the discharge and will rise sufficiently high to avoid concentrations above the
ERPG-2 at ground level.
It is possible that a release of very dense gas from an unignited flare could exhibit
dense gas behavior. For example, in 1950 in the town of Poza Rica, Mexico, 22 people died
from exposure to hydrogen sulfide emitted from, a malfunctioning flare at a gas purification
plant (McCabe and Clayton, 1952). However, in this case, the gas from the well contained 3
percent by volume of H2S and 15 percent by volume of CO2. During the startup period for
the desulfurization units to which the gas was sent, partially processed gas containing 81
percent CO2 and 16 percent BUS was sent to a flare. It was this processed, heavy vapor and
not the produced gas that, upon failure of the flare, descended to ground level. However,
despite the limitations in applicability and the unlikelihood of occurrence, this incident is
illustrative of the potential for severe consequences when managing a dense gas stream.
Consequence Analysis of Releases Collecting at Ground Level
The specific cases listed in Table m-5 are all less dense than air. This has been the
case for all the gas streams investigated for this report for which detailed compositions were
documented. Also, note that the most dense composition on Table ffl-5, stream B which has
a density close to that of air, was obtained after some separation and processing for vapor
recovery. It appears that the concern about heavy vapors containing H2S settling or
collecting in low-lying areas may be justified for only a fraction of wells such as the
previously described Big Piney, Wyoming well blowout and Poza Rica, Mexico flare
incident. It is pertinent to address other situations where this concern is justified.
m-56
-------
Nine people were killed in an incident in Denver City, Texas, when they were
exposed to gas escaping from a well injecting gas into an oil reservoir as part of an enhanced
oil recovery project (Layton et al., 1983). The injected gas was composed of 93 percent by
volume CO2 and 5 percent by volume H2S - clearly denser than air, but as before, gas that
was previously processed and not of as-produced composition.
In general, it is possible that releases directly from wells with unusually dense sour
gas compositions or associated lines could settle in low-lying areas at ground level. These
releases would not be of typical composition.. It is also possible that people entering areas of
seepage such as those previously described for line releases could confuse these with settling
on the ground. It is therefore reasonable to speculate that, in some cases, such concerns
could possibly have arisen from seepage events.- :
The modeling described in the foregoing applies to plumes over flat terrain. In
complex, terrain, it is unlikely that released gas of typical composition will flow into lower
elevations such as valleys because, as previously noted, it is generally not denser-than-air.
However, it is very likely that a small or chronic release will follow the flow of the wind.
Thus, fdr example, on cold, still nights there could be flows of air with relatively little *
turbulence from higher elevations into valleys (katabatic flows). This could carry slowly
diluting H2S with it and potentially cause odors within houses in valleys some distance from
the well. This situation would likely not occur during the day when such air flows are
uncommon. However, as previously discussed, it is possible for sour gas of unusually dense
composition to remain at ground level. Therefore, for such releases, it is conceivable that
flow could "channel" through terrain of low elevations such as valleys. This possibility is
highly uncertain. The study of the behavior of dense gas flow around obstacles and through
rough terrain is controversial and is an area where further research is needed.
Accidental Releases—Prevention, Mitigation, and Emergency Response
The design and operation of sour gas systems require special consideration as a result
of the potential hazards presented by a release of H2S. The hazards of exposure to H2S can
be significantly reduced by the implementation of process safety management principles. A
primary emphasis on containment together with design features for the detection and
mitigation of losses in containment are necessary for safe operations. The degree of
sophistication of individual sour gas system designs will vary depending on site-specific
circumstances and age. Older systems may incorporate relatively simple safety designs when
compared with current state of the art. The presence of sour oil and gas operations in
remote locations or near populated areas may both be justification for the use of advanced
designs. Remote areas may be subject to extended releases if accessibility is limited.
Process safety management and major safety considerations are discussed below.
m-57
-------
Process Safety Management
Facilities that handle hazardous materials have a responsibility to understand the
hazards present at their sites and to take steps to ensure that chemical accidents due to these
hazards are prevented. Many organizations, including the American Institute of Chemical
Engineers - Center for Chemical Process Safety (AIChE-CCPS) and the EPA, have found
that major chemical accidents cannot be prevented by hardware or by technology alone.
Prevention requires comprehensive management systems designed to identify and control
hazards (AIChE, 1989; U.S. EPA, 1988). These management systems are known as Process
Safety Management (PSM) and consist of "comprehensive sets of policies, procedures, and
practices designed to ensure that barriers to major incidents are in place, in use, and
effective. The management systems serve to integrate process safety concepts into the
ongoing activities of everyone involved in the process - from the chemical process operator
to the chief executive officer" (AIChE, 1989). The Occupational Safety and Health
Administration (OSHA) has set standards for process safety management, which are"
discussed in Chapter IV.
PSM consists of several essential elements that work together to allow- safe operation
of a facility;
• Management Commitment: Management must adopt a philosophy that makes
safety an integral part of operation from the top down; an attitude that all accidents
can be prevented and that business must always be conducted safely.
• Process Hazards Analysis or Hazard Evaluation: The purpose of the process
hazards analysis is to systematically examine the equipment, systems, and
procedures for handling a hazardous substance; to identify the mishaps that could
occur, analyze the likelihood that mishaps will occur, and evaluate the
consequences of these mishaps; and to analyze the likelihood that safety systems,
mitigation systems, and emergency alarms will function properly to eliminate or
reduce the consequences of the incident. Thorough process hazards analysis is the
foundation for the remaining elements of the PSM system.
• Process Knowledge and Documentation: Facilities document the details of the
technology and design of the process, its standard conditions and consequences of
deviation from these standards, the known hazards of the chemicals and processes
involved and protective systems for protection of workers, the public, and the
environment. •
• Standard Operating Procedures (SOPs): These'are procedures that describe the
tasks to be performed by the operator or maintenance worker to ensure safety
during operation and maintenance.
m-58
-------
• Training: A program to teach those responsible for designing, operating, and
maintaining the unit or plant. Elements in a management training system include
development of training programs, training of instructors, measuring performance
and determining the effectiveness of training. Training is typically carried out by
facility managers and training staff.
• Maintenance (Process and Equipment Integrity): A formal program to ensure that
equipment is constructed according to design, installed properly, and adequately
maintained.
• Prestartup Review: The purpose of this review is to ensure that all elements of
process safety, including hardware, procedures, and control software, are-in'place
prior to startup, and that all prior issues of concern have been resolved.
• Management of Change: Management must instruct personnel to recognize change
and to evaluate change with regard to process safety.
• Safety Audits: The purpose of safety audits is to measure facility performance, to
verify compliance with a sound process safety program, and to determine that risks
are being appropriately managed.
• Accident Investigation: Accident investigation is a management process by which
the underlying causes of an incident are identified and steps are taken to prevent
similar incidents.
• Emergency Planning and Response: Emergencies involving highly hazardous
substances can have catastrophic results if not handled properly. Employees need
to know and be trained in proper emergency procedures, evacuation requirements,
and notification steps.
Major Safety Considerations
Siting. The magnitude of the potential consequences from human exposure to an H2S
release decreases with distance from the sour oil or gas source. Therefore, operations
involving H2S should be situated as far as possible from residential and commercial structures
to minimize potential hazards to the public. Prevailing weather patterns (e.g., wind
direction), terrain features, transportation routes, population centers, the potential for
evacuation, and the potential for access control are some additional factors to be considered
in siting decisions. These are site-specific factors that must be determined for each location.
ffl-59
-------
At a minimum, well sites should be fenced to maintain some obstacle to approaching the
wellhead.
Materials Selection and Corrosion Prevention. Materials must be chosen that are
suitable for the service into which they are placed. Sour oil and gas operations are often
conducted under high pressure and corrosive conditions. Therefore, in addition to
temperature and pressure considerations, system designs for the wellhead, downhole
equipment, and pipelines must incorporate features to minimize the effects of corrosion in
order to prevent a breach of containment and accidental release of H2S. Several national
engineering standards governing the choice of materials are applicable. Standards include
those by the American Petroleum Institute (API), American Society of Mechanical Engineers
(ASME), and the National Association .of Corrosion Engineers (NACE). One such standard
is NACE Standard MR0175, "Material Requirements for Sulfide Stress Cracking Resistant
Metallic Materials for Oilfield Equipment." Also applicable are the API 6A specifications
for equipment in high H2S concentrations in close proximity to occupied structures.
In addition to proper selection of materials, corrosion inhibiting fluids can be used to
prevent internal corrosion and cathodic protection can be used to prevent external corrosion.
Inhibitor applications include the filling of wells with inhibitor during extended periods of
shut-in and injection into pipelines.
Corrosion monitoring programs should be a normal part of the operations and
maintenance for sour oil and gas systems so that corrosion problems can be anticipated and
repairs made before an accidental release occurs. The need for a corrosion control program
and program monitoring was discussed in the first edition of API RP 55, "API
Recommended Practices for Conducting Oil and Gas Production Operations Involving
Hydrogen Sulfide" (API, 1983). This document has been withdrawn pending publication of
an updated, second edition. Additional discussion of RP 55 can be found in Chapter IV.
Corrosion monitoring systems can take a variety of forms including external monitoring
(ultrasonic or X-ray inspection), corrosion coupons and spool pieces (test pieces),
instrumented "pigs", or in-place instrumentation. Pigs are instruments that can be
transmitted through lengths of larger diameter piping to take measurements of internal
surfaces.
Leak Detection and Mitigation. While systems should be designed to meet the
appropriate standards, there is still the potential for releases to occur as a result of human
error or equipment failure (e.g., corrosion, impact, etc.). A possible design feature for oil
and gas operations is the use of detection systems which monitor for evidence of system
leaks and then isolation systems that can be used to shut off leaks. For H2S-containing
systems, detection systems can focus directly on measurement of H2S, on measurement of
pressure changes which could be indicative of a leak, or temperature indicators that can be
indicative of a loss of containment and subsequent fire. Signals from such detection systems
can be used in modem, sophisticated systems to automatically initiate additional containment
measures such as well shut-in or isolation of sections of pipeline. There are national
m-60
-------
standards for performance and use of H2S monitoring equipment such as these set by the
Instrument Society of America, ISA-S12.15 "Part I: Performance Requirements of Hydrogen
Sulfide Detection Instrumentation" and "Part II: Installation, Operation, Maintenance of
Hydrogen Sulfide Detection Instruments." Not all systems have leak detection or signalling
devices associated with them. Such systems may present a greater hazard potential than
those that have devices because detection would have to be by visual means or by smell.
Any release would continue until detected.
Flares may malfunction resulting in extinguishment of the flame. This may occur due
to several causes including flow of noncombustible compounds (e.g., nitrogen or carbon
dioxide) and high winds. Flares can be equipped with automatic ignition devices to reignite
extinguished flames and supplemental fuel systems to maintain ignition of the flare gas in the
presence of inert gas. Flares should also be constructed at a height that provides for
sufficient dispersion of the discharge.
The equipment used to mitigate releases depends on the operations. For well drilling
and workover operations, a blowout preventer is used. This piece of equipment consists of
high-pressure valves that allow the operator to shut in the well. For operating wells, there
can be subsurface shutoff valves which are located in the well as well as above grade valves
located at the wellhead and in the lines around surface equipment such as separators. Shut-in
may be accomplished automatically via a signal (H2S concentration, pressure change,
temperature) that is received indicating a potential leak. For pipelines, there may also be
isolation or shutdown valves located along the pipeline and these may be automatically
activated if there is an indication of a leak in the pipeline or at the well. Not all systems will
have automatic mitigation capability and isolation would have to be manual in these cases.
Inspection and Monitoring Practices. API RP 55 made recommendations for actions
that were intended to monitor performance of the containment system for the sour oil and
gas. API RP 55 specifically called for inspection of equipment and system performance to
look for indications of corrosion that are indicators of degradation of the sour oil and gas
containment equipment. Inspections were specifically recommended for changes in lift
performance; changes in pressures associated with packed off annuli; and for the condition of
valves, flanges, and connections. The document also recommended that any equipment
failures be evaluated to determine the cause of the failure. Particular attention should be
paid to the effectiveness of the corrosion control program at a site and corrective action
should be considered if there is any indication that the program is inadequate.
API RP 55 also called for the monitoring, maintenance and recalibration of
monitoring equipment (temperature, pressure, composition, etc) to make sure it is functioning
as intended.
Emergency Procedures. In the event of loss of containment of the sour oil and gas,
emergency procedures must be implemented to both restore containment and to protect the
public. API RP 55 called for the preparation of a contingency plan for operations involving
m-61
-------
sour oil and gas. The plans are to contain information that would be needed by personnel
responding to the accident at the site. Among the information that should be in the plan
according to the API recommended practices are:
1. Location of wells and details on the equipment including flow lines, isolation
valves, processing facilities, and tank batteries;
2. Location of safety and life support equipment;
3. Location of telephones and other communication equipment;
4. Potential location of roadblocks for excluding unauthorized personnel for the areas
associated with the accidental release;
5. Location of residences, businesses, parks, schools, roads, medical facilities;
6. Areas that could experience elevated H2S concentrations,(e.g. levels greater than
l-x!05ppb);
7. Potential evacuation routes; and
8. Designated safe areas for operations personnel.
In addition to this information, the plan should have a list of emergency telephone
numbers including company supervisors; residences, schools and businesses; nearby
operators and service companies; local law enforcement agencies; officials responsible for
public facilities that could be impacted; medical assistance personnel, facilities and
equipment; and concerned local, state, and Federal agencies.
Beyond the information listed above, the contingency plan should have an immediate
plan of action. Among the elements in an immediate action plan are the determination of the
potential hazard to the public from the discharge and then an identification of actions to
respond to the hazard (e.g. immediate measures to eliminate the discharge, notification of
responsible supervisors, establishment of a restricted access zone, evacuation of personnel).
API RP 55 also recommended consideration of advanced briefing of public and public
officials so they understand the nature of the hazard, the necessity for emergency response
plans, and the general steps that would be taken in the event of an emergency. Finally, API
PJP 55 called for the updating of the plan as necessary to keep the information in the plan
current and conducting periodic drills so that personnel are familiar with the type of
situations to which they may have to respond.
The Department of the Interior has promulgated regulations that are applicable to sour
oil and gas operations on Bureau of Land Management (BLM) property (BLM, 43 CFR
3160). These regulations call for the preparation of public protection plans for drilling and
production operations where (1) the 1 x 105 ppb H2S radius is greater than 50 feet and the
area includes locations where the public could reasonably be expected to be (e.g. occupied
residences, schools, churches, parks); (2) the 5 x 10s ppb H2S radius is greater than 50 feet
and includes any part of Federal, State, or county or municipal road or highway; or (3) the
1 x 10s ppb H2S radius is greater than 3,000 ft. where facilities and roads are principally
maintained for" public use. The requirements for the content of these public protection plans
are very similar to those called for in API RP 55.
m-62
-------
Abandonment Practices
The termination of sour oil and gas production operations normally results in the
plugging of the well with cement prior to abandonment by the operator. As a result, a
potential exists for H2S to be released from sour oil and gas from the well and associated
equipment if proper precautions are not taken. API RP 55 identified actions that should be
taken at the end of operations. The document specifically called for precautions to ensure
that H2S does not present a hazard to the public and the environment. The document called
for either air purging or water flushing of equipment followed by opening to the atmosphere.
Pipelines then were to be purged and capped. API RP 55 also called for the setting of
cement across formations that could produce H2S.
In some cases, wells may be temporarily abandoned. These wells may also be called
"idle" or "inactive." In temporary abandonments, the well will not be plugged with cement
but perforations may require isolation. Typically, application must be made and approval
given by a state authority to temporarily abandon a well. Conditions justifying temporary
abandonment to a State most often include economic conditions and future utility (IOGCC,
1992). Approval is temporary and of limited duration although extensions may be granted at
the discretion of the state authority. Depending on the state, initial approval periods range
from 6 months up to 10 years. Extensions may be granted for up to an unlimited number of
time periods. In many states, but not all, periodic testing is required on idle wells. For
example, mechanical integrity and pressure tests may be required. These practices are
intended to prevent releases of oil and gas.
Of 215,000 oil and gas wells estimated to have been idle in 1992, approximately
68,000 were thought to have been idled without State approval (IOGCC, 1992). 50,000 of
these wells, known as orphan wells, were believed to have been idled by operators who were
unknown or insolvent. Although the fact that a temporarily abandoned well has not been
reported to the State does not mean the well will be the source of an accidental release, the
lack of control and supervision does represent an unsafe situation and may present a greater
risk to the public and the environment. The majority of States have developed some funding
mechanism and implemented programs to plug and abandon orphan and preregulatory wells
although these activities vary widely from state to state (IOGCC, 1992).
Land Use Around Well Sites
Land use can vary enormously around oil and gas wells. The wells may be found in
urban areas or open rangelands. Figure m-23 shows current land-use patterns by EPA
region (Southerland, 1992). In Regions 6, 8, and 9, which contain the majority of wells in
naturally occurring H2S areas, between 50 and 60 percent of the land is used as range. The
three regions represent about 60 percent of the oil and gas producing wells. In the
Midwest's Region 5, which contains 12 percent of the nation's producing oil and gas wells,
over 50 percent of the land is farmed (U.S. EIA, 1990; U.S. EIA, 1991).
nf-63
-------
Legend
Pasture
Range
Forest
Minor
Source: Southerland, 1992.
Figure 111-23. .Current land-use pattern by ERA. region
m-64
-------
Regarding urban areas, in California, for example, the Division of Oil and Gas
reports that "one-third of California's 1.7 billion barrels of oil reserves are in urban areas or
in areas where residential development is increasing. (The H2S content of these reserves was
not available.) The Los Angeles Basin both typifies the situation and is the most complex
example. Here, a large metropolitan area lies over one of California's major petroleum-
producing provinces. Because oil and gas are so fundamental to the U.S. economy, any
recoverable amounts cannot be ignored. Ways have been developed to produce oil and gas
safely in urban areas, with minimum negative effects. Urban planners, administrators, and
California Division of Oil and Gas engineers work together to ensure a safe partnership
between urban life and oil and gas development" (CDC, 1988).
Affected Human Populations
Figure IH-24 overlays 1980 census data on the H2S prone areas to show the proximity
of major populations to H2S deposits. The figure shows that a wide range in population
density can be found in H2S deposit areas. However, a look at the locations of well fields in
the United States (Figure DI-11) and the number of wells per State (Figure ffl-12) clarifies
the potential exposure of large human populations to H2S from oil and gas wells.
Data were not available to arrive at statistics on individuals exposed to H2S emissions.
Because the number of wells in the U.S. is so great and the diversity of population density
around wells so large, it was not possible to arrive at an estimated affected population. The
photographs in this report show that wells may be found in urban, suburban, and rural areas.
Populations that could be exposed include adults in work settings (e.g., fire stations),
children in schools, shoppers in downtown areas, and people in residential areas.
Affected Environmental Settings
A 1991 study in Wyoming found that, in two years, 237 animals had been killed by
H,S gas. In many oil fields this gas was vented through flare stacks. The researcher stated
that when flare stacks are used, it is possible to install devices, which would prevent raptors
and other birds from using flares as perch sites. Also, wildlife mortality caused by H2S
would be reduced by ensuring that igniters were operating efficiently so that the gas would
be properly flared and not accidentally vented directly into the environment (Esmoil, 1991).
Based on other accident history, one impact on environmental settings has been the loss of
livestock attributed to exposure to H2S. Sixty percent of the U.S. wells are located in EPA
Regions that contain more than 50 percent rangeland. However, many other species of
animals and plants are potentially exposed to H2S concentrations that could cause adverse
effects. Testimony for the Clean Air Act Amendments included statements about episodes in
the Great Plains that resulted in livestock dying and humans being hospitalized (Audubon
Society, 1987).
Twelve percent of all wells are located in EPA Region 5, which is more than 50
percent cropland. As noted in a previous section of this report, soybeans have been
m-65
-------
Urbanized Population
SB 250,000 or more
A 100,000-249,999
• 50,000-99,999
0 H
-------
determined to be sensitive to H2S along with other crops. There has been evidence of
scorching to young leaves and shoots but no effect on mature leaves (Heck et ai., 1970).
Waterfowl habitats of major concern are located in some areas of oil deposits with
H2S, as shown in Figure DI-25. Concern has also been expressed about the deterioration of
air quality in Theodore Roosevelt National Park (Sierra Club, 1987). Figure m-26 shows
the location of national parks and national forests in relation to H2S deposits.
FINDINGS
1. Human exposure to H2S may cause death, as well as symptoms including irritation,
breathing disorders, nausea, vomiting, diarrhea, giddiness, headaches, dizziness, -
confusion, rapid heart rate, sweating, weakness, and profuse salivation. Levels
greater than 1.5 x 105 ppb are life threatening.
2. No epidemiological studies were found on the effects of H2S emissions from oil and
gas extraction/production.
3. Human acute and chronic health effects data and ecological effects data are limited.
4. H2S is classified as a Group D carcinogen, meaning not classifiable as a human
carcinogen. The inhalation RfC is 9 x 10"* mg/m3 (0.67 ppb) in chronic exposures
scenarios. This RfC is not appropriate, however, for assessing concentration-response
relationships in short-term or accidental exposure scenarios.
5. Few studies exist measuring natural or accidental exposure of wildlife to H2S;
however,, wildlife deaths have been reported with blowouts.
6. High exposure studies have shown young, growing plants to be the most susceptible
to H2S injury (clover, soybean, tomatoes, tobacco, buckwheat).
7. Aquatic LC50s show bluegill = 0.009-0.0478 mg/1.
NAOEL for mice = 42.5 mg/m3 (3.05 x 104 ppb).
LAOEL for mice = 100 mg/m3 (8 x 104 ppb).
8. Nationwide, vulnerability zones have been characterized as 14 major H2S prone areas
found in 20 states. Texas has 4 discrete H2S prone areas.
9. North Dakota is the only State known to have routinely monitored ambient H2S at
well sites and surrounding areas.
10. Many oil-and gas producing States require ambient air monitoring for H2S at gas
plants and refineries, but monitoring is not frequently required at oil and gas
m-67
-------
12
CD Waterfowl Habitats
§ H2S Prone Areas
Source: Gas Research Institute. 1990.
Copemder, Boyd. and Stuart. 1986.
Figure 111-25. Major HoS prone areas in relation to waterfowl habitats of
major concern (numbers indicate relative priority of concern).
DI-68
-------
National Forest
National Paries
HoS Prone Areas
Sources: Gas Research Institute, 1990. Rand McNaily, 1992.
Figure ITI-26. Major HgS prone areas shown in relation to National Forests
and Parks.
m-69
-------
extraction facilities, unless H2S emission violations are suspected or complaints are
filed.
11. North Dakota has three background and six special-purpose H2S monitors.
Monitoring periods vary in length from months to over a decade (32.75 years total).
12. At several locations, North Dakota monitoring data verified compliance with State
H2S standards. In two cases, data were from monitoring periods too short to support
any conclusions; these were discontinued even though numerous NDAAQS violations
were experienced their last year monitored.
13. North Dakota's database showed short-term H2S concentrations ranging from O.tq.
2734 ppb. The median value of all monitoring data was 0 ppb.
14. One North Dakota site had maximum short-term H2S concentrations an order of
magnitude higher than the other eight sites. At this site, more than 3,000 violations
were recorded from 1984 to 1986. Concentrations improved greatly from 1986 to
1989, and only one violation occurred after the health-based standards went into
effect.
15. Annual average H2S concentrations at two sites in North Dakota approximated the
RfC after introduction of a gas collection system with manifolded flares.
16. North Dakota flare operating efficiencies have been reported to range from 30 to 100
percent. (At 30 percent efficiency, H2S can be routinely released in significant
concentrations.)
17. The risk to the public of an accidental release of H2S from the extraction of oil and
gas is a function of both potential consequences and likelihood of occurrence.
Judgements of risk should not be made solely on the basis of consequence analysis
alone.
a. Risks may vary from facility to facility depending on site-specific factors such
as the density and distribution of nearby populations and the quality of process
safety management and risk management practiced at the facility.
b. Some facilities present greater risk than others.
c. Risk reduction must take both consequence and likelihood of occurrence into
account.
18. In addition to being toxic, H2S is corrosive to metals in the presence of moisture and
is flammable.
a. Sour gas is flammable due to its composition of light hydrocarbons and EUS.
Howevec, ignition of sour gas does not generally represent a thermal radiation
hazard to the off site public beyond a distance of about 100 meters.
ffl-70
-------
b. The corrosivity of H2S in the presence of moisture can cause equipment
leakage and other losses in containment.
19. If accidentally released to the air under certain circumstances, H2S can present a
threat to public health and the environment.
a. Well blowouts, line ruptures, and equipment leakage have caused accidental
releases of sour gas with documented impacts on public health and the
environment.
b. The impacts on the public in the United States from sour natural gas releases
from extraction activities documented in this study were limited to examples of
hospital treatment and evacuation. A number of fatalities have occurred in the
workplace. A single incident of the release of carbon dioxide containing H2ST-.
from injection activities to enhance recovery resulted in the 1975 fatalities of
eight members of the public.
c. In this study, several incidents were documented as examples of both livestock
and wildlife fatalities resulting from exposure to H2S from accidental releases
of sour gas.
d. The concentration of H2S in sour gas may vary from non-lethal levels to lethal
levels above 30 percent. Unless there are high concentrations of carbon
dioxide and/or hydrogen sulfide, an unprocessed sour gas mixture will usually
be less dense than air and will not usually collect at ground level or in low-
lying areas if accidentally released.
e. Releases of sour gas such as from an extinguished flare or from high-pressure
equipment failures (e.g., well blowouts and line ruptures) will entrain
surrounding air which can cause significant dilution of the hydrogen sulfide
and other components in the gas, thereby reducing the potential magnitude of
the consequences of its release.
f. A release of a sour gas mixture that is denser than air and is not significantly
diluted through release phenomena (such as a jet from a high pressure source)
could, under conservative atmospheric conditions, settle in low-lying areas and
present a toxicity hazard. No documented incidents associated directly with oil
and gas extraction were identified to support this scenario. Thus, this finding
is based on theoretical premises.
20. Atmospheric dispersion modeling of worst-case scenarios shows that accidental
releases of sour gas can have a range of impacts from no public impact to doses
equivalent to the LC01 and AIHA ERPG-3 beyond 10 kilometers from the point of
release.
a. Modeling results indicate that, within a broad range of typical conditions for a
vertical well blowout and emission from an extinguished flare, sour gas
releases will not cause fatalities to the offsite public. This result would also
apply to any similar vertical jet release at wellhead conditions resulting from
equipment or line leakage.
m-71-
-------
b. Modeling results estimate that, in the worst-case, a horizontal release of sour
gas from a well blowout (or similar high release rate jet in a horizontal
orientation from equipment or piping) could produce fatalities in one percent of
the human population exposed at distances up to approximately 10 kilometers.
21. Results from modeling exercises are only gross approximations of what might occur
during an actual accidental release. These results are extremely sensitive to factors
such as the assumed release rates and assumed meteorological conditions. Precise
prediction of downwind effects from an actual release is unlikely for reasons such as:
a. An actual release may have a different release rate than that assumed for a
hypothetical scenario.
b. The composition of an actual sour gas release may differ from that .assumed.,m
a modeling scenario.
c. The meteorological conditions existing during an actual release may differ
from those assumed in a modeling scenario.
d. The effects of surface roughness (e.g., terrain and obstacles) are not fully
understood. It is assumed in the models used that complex terrain and
obstacles increase dispersion.
e. The levels used to predict the onset of toxic effects (i.e., LC01 and ERPG-3)
are highly uncertain.
22. While analysis of the worst-case scenario can be useful to help facilities and the
community surrounding facilities to gain an understanding of the potential magnitude
of severe situations, such an analysis does have its limitations. A worst-case scenario
should be taken into account along with more probable scenarios when setting
priorities for community emergency planning. Note, however, that the worst-case is
designed to generate the maximum impact off-site and is considered to be extremely
unlikely. The worst-case does not take into account a variety of factors that can
significantly reduce downwind impacts.
a. The worst-case scenario does not take into account the role of process safety
management in reducing the probability of loss of containment.
b. The worst-case scenario does not take into account mitigation .actions that can
reduce the amount released into the air.
c. The worst-case scenario assumes terrain and topographical conditions that
minimize dispersion of the plume. Actual conditions may result in greater
dispersion.
d. Worst-case meteorological conditions may not exist during an actual release.
e. The dose that is actually received is uncertain and may be reduced or avoided
by sheltering-in-place or evacuation.
23. Technologies have been developed to detect and reduce the amount of sour gas
released as a result of breaches in containment. These technologies, would serve to
protect the public in inhabited areas and to protect wildlife in remote areas with
m-72
-------
limited access by facilitating quicker mitigation. These technologies include:
a. Subsurface safety valves;
b. Remotely operated isolation valves;
c. Automatically operated shutoff and isolation valves;
e. Remotely monitored pressure and flow meters;
f. Local and remote audible and visual warning signals; and
g. Automatic flare igniters and supplemental fuel sources.
In spite of the availability of detection and mitigation measures, all facilities have not
uniformly adopted such measures. In addition, the reliability of such equipment and
site-specific conditions must.be considered before particular.technologies are adopted
or implemented.
24. Wells drilled in H2S prone areas may or may not contact H2S sources.
25. Eight States have a significant overlap of well fields and H2S prone areas. Therefore,
it is roughly estimated that as many as 280,000 oil wells and 54,000 gas wells have
the potential to be located in an H2S prone area. The actual number of sour wells in
each State was not available.
26. Population densities in urban areas within ranges of 100,000-249,999 and 50,000-
99,999 can be found in H2S prone areas in California, Texas, Missouri, Florida,
Illinois, Kentucky, Oklahoma, Arkansas, Ohio, Michigan, and Wyoming.
27. There have been several documented incidents of wildlife fatalities due to sour oil and
gas releases. No incidents have been documented where large-scale wildlife fatalities
have been caused by H2S, and no national statistics on wildlife incidents were found.
However, a Wyoming study found 237 animals killed by H2S in two years.
28. H2S-prone areas overlap 10 waterfowl habitats of major concern, 18 national forests
and 3 national parks.
29. Land use and, therefore, potential human and ecological exposure scenarios can vary
enormously around oil and gas wells:
a. In EPA Regions 6, 8, and 9 which contain the majority of wells in H2S prone
areas (which represent 60 percent of all wells nationwide), 50 to 60 percent of the
land is used as range.
b. In Region 5 (12 percent of U.S. wells), 50 percent of land is farmed.
c. In California, 1.7 billion bbls of oil reserves are in urban or increasingly
developed residential areas.
m-73
-------
30. ACGHTs recommended TLV-TWA for H2S is 1 x 104 ppb (14 mg/m3) and TLV-
STEL is 1.5 x 104ppb (21 mg/m3).
31. AIHA ERPGs for the general public for H2S are --
ERPG 3 - 1 x 10s ppb (1-hr exposure, not life threatening)
ERPG 2 - 3 x 104 ppb (1-hr exposure, no irreversible or serious health effects)
ERPG 1 - 100 ppb (1-hr exposure, no mild, transient adverse effects or clearly
defined odor).
32. NAS/NRC H2S guidelines for protecting the general public from the effects of
accidental releases are -
90-day continuous exposure guide level - 1 x 103 ppb
24-hr emergency exposure guideline level - 1 x 104 ppb
10-min emergency exposure guideline level - 5 x 104 ppb.
REFERENCES
ACGEH. 1992. 1991-1992 Tfireshold Limit Values for Chemical Substances in the Work
Environment. American Conference of Governmental Industrial Hygienists.
Ahlborg, G. 1951. Hydrogen sulfide poisoning in shale oil industry. Arch Ind Hyg Occu
Med 3:247-266.
AIChE, 1989. American Institute of Chemical Engineers, Technical Management of
Chemical Process Safety, Center for Chemical Process Safety, New York, 1989.
AIHA. 1991. Emergency Response Planning Guidelines: Hydrogen Sulfide. American .
Industrial Hygiene Association. Akron, OH.
Alp, E. et al. 1990. An Approach for Estimating Risk to Public Safety from Uncontrolled
Sour Gas Releases. ERCB Report 90-B (10 Volumes), Prepared by Concord
Environmental Corporation for Energy Resources Conservation Board, Calgary,
Alberta.
API, 1983. API Recommended Practice 55 (RP 55), Recommended Practices for Conducting
Oil and Gas Production Operations Involving Hydrogen Sulfide, October 1981,
Reissued March 1983, 1st ed., Publication No. RP55. American Petroleum Institute.
Audubon Society. 1987. Testimony of John Brophy, Co-chairman of Conservation, Fargo-
Moorhead Audubon Society, North Dakota. Presented to the U.S. Senate
Subcommittee on Environmental Protection, June 19, 1987, Honorable George
Mitchell Chairman, in Hearings before the Subcommittee on Environmental Protection
of the Committee on Environment and Public Works, U.S. Senate, 100th Congress,
m-74
-------
First Session on S.300, S.321, S.1351, and S.1384 bills to amend the Clean Air Act-
Part 2. U.S. Government Printing Office, Washington, 1987.
Bell, R.P. 1978. Isopleth Calculations for Ruptures in Sour Gas Pipelines.
Processing/Canada (July-August 1978).
Energy
Benedict H.M., and Breem, W.H. 1955. The Use of Weeds as a Means of Evaluating
Vegetation Damage Caused by Air Pollution. Proceedings of the Third National Air
Symposium. Pasadena, CA. Los Angles, CA National Air Pollution Symposium, 177-
190.
Bilderback, D.E. 1988. -The Effects of Air Contaminants. ,o.n the. Vegetation of Theodore
Roosevelt National Park and Adjoining Areas. Final Report.
' ELM. 1990. Williston Basin Regional Air Quality Study. U.S. Department of the Interior,
Bureau of Land Management, Cooperative Effort with the North Dakota State
Department of Health and Consolidated Labs; Montana State Department of Health
and Environmental Sciences, and the U.S. Department of Agriculture, Forest Service.
BLM, 43CFR3160. Title 43 Code of Federal Regulations, Part 3160 Onshore Oil and Gas
Operations; Federal and Indian Oil and Gas Leases; Onshore Oil and Gas Order No.
6, Hydrogen Sulfide Operations, Bureau of Land Management. 55 Federal Register
48958, November 23, 1990.
Bureau of the Census. 1983. Characteristics of the Population: General Social and
Economic Characteristics. United States Summary, Part 1, PC80-1-C1, issued
December 1983, 1980 Census of Population. U.S. Department of Commerce,
Washington, D.C.
Burgess, W.A. 1979. Potential Exposures in Industry: Their Recognition and Control. In:
Patty's Industrial Hygiene and Toxicology, G.D. Clayton, ed., John Wiley and Sons,
New York, NY, 1:1149-1222.
Casarett and Doull. 1986. Toxicology. C.D. Klassen, M.O. Amdur, and J. Doull, ed.,
Macmillan Publishing, New York.
CDC. 1988. California Oil, Gas, and Geothermal Resources: An Introduction, 4th ed. Based
on the 2nd edition by D.E. Ritzius. California Department of Conservation, Division
of Oil and Gas, Sacramento, CA.
*
Cloe, W.W. 1985. Selected Occupational Fatalities Related to Toxic and Asphxiating
Atmospheres in Confined Work Spaces As Found in Reports of OHSA
Fatality/Catastrophe Investigations, U.S. Department of Labor.
m-75
-------
Cooperider, A.Y., Boyd, R.J., and Stuart, H.R., eds. 1986. Inventory and Monitoring of
Wildlife Habitat. U.S. Department of the Interior, Land Management Service Center,
Denver, CO.
Curtis, C.G., Bartholomew, T.C., Rose, F.A., and Dodgson, K.S. 1972. Detoxication of
sodium 35 S-sulfide"in the rat. Biochem Pharmacol 21:2313-2321.
Crump, K.S. and Howe, R.S. 1984. The multistage model with a time-dependent dose
pattern: applications to carcinogenic risk assessment. Risk Analysis 4:163-176.
Dielwart, J., Knox, H., and Angle, R. 1979. A Report by the Alberta Petroleum Industry,
Government Environmental Committee un Hydrogen Sulfide Isopleth Prediction: Phase
II Pipe Burst Study. APIGEC.
"Dosch, M.W., and Hodgson, S.F. 1986. Drilling and Operating Oil, Gas, and Geothermal
Wells in an H^S Environment, Publication No. M10. California Department of
Conservation, Division of Oil and Gas, Sacramento, CA, 39.
*
Ermak, D.L. 1989. User's Manual for the SLAB Model, An Atmospheric Dispersion Model
for Denser-than-Air Releases. Lawrence Livermore National Laboratory.
Esmoil, BJ. 1991. Wildlife Mortality Associated with Oil Pits in Wyoming. Master's
Thesis, University of Wyoming, Laramie, Wyoming. Also published as a Final
Report by the Wyoming Cooperative Fish and Wildlife Research Unit, February
1991.
Evans, C.K. 1967. The toxicity of hydrogen sulfide and other sulfides. J Exp Physiol
52:231-248.
Gas Research Institute, Chicago, DL. 1990. Chemical Composition of Discovered and
Undiscovered Natural Gas in the Continental United States. Volume 2: Statistics by
Basin and Formation-Non-Associated Gas. Prepared by Energy and Environmental
Analysis, Inc.
Gephart, L., and Moses, S. 1989. An Approach to Evaluate the Acute Impacts from
Simulated Accidental Releases of Chlorine and Ammonia. Plant/Operations Progress
£8-11.
Guidelines for Chemical Process Quantitative Risk Analysis. 1989. Center for Chemical
Process"Safety of the American Institute of Chemical Engineers, New York.
Gunina, A.I. 1957. Transformation of sulfur-35-Iabeled hydrogen sulfide introduced into
blood. Dokl Akad Nauk SSSR 112:902-904.
_. m-76
-------
Haggard, H.W. 1925. The toxicology of hydrogen sulfide. J Ind Hyg 7:113-121.
Harris, B. 1986. Study into long-term effects on livestock following the Lodgepole
blowout. Alberta Agriculture, Edmonton, Alberta, Canada.
Heck, W.W., Daines, R.H., Hindawi, I.J. 1970. Other Phytotoxic Pollutants. In:
Recognition of Air Pollution Injury to Vegetation: a Pictorial Atlas, J.S. Jacobson and
A.C. m, eds.
Higashi, T., Toyama, T., Sakurai, H., Nakaza, M., Omae, K., Nakadate, T., Yamaguchi,
N. 1983. Cross sectional study of respiratory symptoms and pulmonary functions in
rayon textile workers with special reference-to H2S exposure.----ImL-Mealth 21-2S1-
292. "~
Hoyle, W. 1973. Summary of Alton Special Project (Inter-office correspondence), Illinois
Environmental Protection Agency, Springfield, IL.
Illinois Institute for Environmental Quality. 1974. Hydrogen Sulfide Health Effects and
Recommended Air Quality Standard. Document No. 74-24. Chicago: State of
Illinois, Institute for Environmental Quality, p27.
IOGCC. 1992. A Study of Idle Oil and Gas Wells in the United States, Interstate Oil and
Gas Compact Commission (IOGCC) and United States Department of Energy.
Johnson, D.W. 1991. Prediction of Aerosol Formation from the Release of Pressurized,
Superheated Liquids to the Atmosphere. In International Conference and Workshop
on Modeling and Mitigating the Consequences of Accidental Releases of Hazardous
Vapors. American Institute of Chemical Engineers, NY.
Kangas, J., Jappinen, P., Savalainen, H. 1984. Exposure to Hydrogen Sulfide, Mercaptans,
and Sulfur Dioxide in Pulp Industry. American Industrial Hygiene Association
Journal 45: 787-90.
Kleinfeld, M., Giel, C., Rosso, A. 1964. Acute hydrogen sulfide intoxication: an unusual
source of exposure. IndMed Surg 33:656-660.
Laug, E.P., and Draize, J.H. 1942. The percutaneous absorption of ammonium hydrogen
sulfide and hydrogen sulfide. / Pharmacol Exp Ther 76:179-188.
Layton, D.W., et al. 1983. Accidental Releases of Sour Gas From Wells and Collection
Pipelines in the Overthrust Belt: Calculating and Assessing Potential Health and
Environmental Risks. Lawrence Livermore National Laboratory Report UCRL-
53411, Prepared for the Division of Fluid Mineral Operations, Bureau of Land
Management, U.S. Department of the Interior, Washington, DC.
m-77
-------
Liebsch, E. 1985. State Trends in Well Flaring. North Dakota State Department of Health,
Intradepartmental Memorandum.
Lodgepole Blowout Inquiry Panel. 1984. Lodgepole blowout inquiry phase I report to the
lieutenant governor in council with respect to an inquiry held into the blowout of the
well, Arusco Dome Brazean River 13-12-48-12. Calgary, Alberta, Canada: Energy
Resource Conservation Board; report no. D84-9.
McCabe, L.C., and Clayton, G.D. 1952. Air pollution by hydrogen sulflde in Poza Rica,
Mexico: An evaluation of the incident of Nov. 24, 1950. Arch. Ind. Hyg. Occup.
Med. 6_ 199-213.
McCallen, S.E.A., Hartzell, A., Wilcoxon, F. 1936. Hydrogen Sulphide Injury to Plants.
Contrib. Boyce Tompson Inst. 8: 189-197.
National Response Center, Washington, DC, Incident Report 72218
National Response Center, Washington, DC, Incident Report 76451
NCCT. 1985. Hydrogen Sulfide. National Council Committee on Toxicology, National
Research Council. In: Emergency and Continuous Exposure Guidance Levels for
Selected Airborne Contaminants. Vol. 4, Washington, DC, National Academy Press,
pp. 55-68.
NFPA. 1974. National Fire Codes~A Compilation ofNFPA Codes, Standards, Recommended
Practices, and Manuals: Combustible Solids, Dusts and Explosives. Volume III,
National Fire Protection Association, Boston, MA, 49-160 - 49-161.
NIOSH. 1977. NIOSH Criteria for a Recommended Standard.... Occupational Exposure to
Hydrogen Sulfide, Publication No. 77-158. U.S. Department of Health, Education,
and Welfare, National Institute for Occupational Safety and Health, Cincinnati, OH,
79.
North Dakota State Department of Health and Consolidated Laboratories. 1983. Sulfiir
Dioxide Emissions Inventory for Sources near the Theodore Roosevelt National Park.
Final Report.
North Dakota State Department of Health, Oil Field Complaint File, 5 April 1989.
NRC. 1979. Hydrogen Sulfide. National Research Council. USA Subcommittee on hydrogen
sulflde. University Park Press, Baltimore, MD.
NRC. 1986. National Research Council, Committee on Toxicology. Criteria and Methods
for Preparing Emergency Exposure Guidance Level (EEGL), Short-Term Public
m-78
-------
Emergency Guidance Level (SPEGL), and Continuous Exposure Guidance Level
(CEGL) Documents. National Academy Press, Washington, D.C.
Oil Daily, No. 7773, November 18, 1982.
sAmosn °f the Plume Path <* Gase* Emitted by
a Stack Atmospheric Environment 6 899 - 909: See also: Ooms, GAP Mahieu
and F. Zehs, "IT* Plume Path of Vent Gases Heavier than Air" F^st Internal
Symposium on Loss Prevention and Safety Promotion and Safety Promotio^nthe
C^P, ~neS' ^ HagUC (1974): °°ms' G' "« N-H- MJm, "Dispersion of a
of H^vT, T^^'" IDTAM Symp°Sium °n the AtmoSpheP
S >^ly -?a S ^ SmaU Partides' Delft d983) - Springer , Veriag
Berlin/Heidelberg. -
Platt's Oilgram News, Vol. 68, May 31, 1990.
Quest Consultants Inc. 1992. Hazards Analysis/Risk Analysis Study of Union Pacific
^
Rand,McNaUy. 1992. Road Atlas. Rand McNally, Chicago, IL.
SAIC. 1990. .S4/0 Gwi^pmser Programs for Modeling the Atmospheric Dispersion
-
s
Part 2. U.S. Government Pnnting Office, Washington, 1987.
Smith, R.P., and Gosselin, R.E. 1964. The influence of methemoglobinemia on the lethality
of some toxic amons. Toxicol Appl Pharmacol 6:584-592. Y
Smith, R.P., and Gosselin, R.E. 1979. Hydrogen sulfide poisoning. 7^^21:93-97.
m-79
-------
-------
CHAPTER IV
REGULATORY PROGRAMS AND RECOMMENDED
INDUSTRY PROCEDURES
INTRODUCTION
This chapter identifies and reviews the current State and Federal regulatory programs
and industry-recommended, procedures applicable to either reduce the potential for routine
emissions and/or accidental hydrogen sulfide releases from oil and gas production or to
mitigate the consequences of such emissions and releases. .
STATE REGULATIONS
Currently, there are no national ambient air quality standards (NAAQS) for H2S.
Most oil- and gas-producing States have their own regulations pertaining to H2S gas. Table
IV-1 lists States that have set ambient air quality standards for H2S emissions.
The EPA gathered and reviewed several States' regulations and related guidance
documents and later contacted State agencies to obtain additional information on the unique
aspects of the State regulations governing H2S emissions in the oil and gas industry. EPA
staff also met with officials from North Dakota during a trip to North Dakota oil and gas
well sites. In addition, the Interstate Oil and Gas Compact Commission (IOGCC) was
contacted to obtain information pertaining to regulatory programs (IOGCC, 1990).
This chapter contains a review of existing State regulations for nine States (California,
Louisiana, Michigan, New Mexico, North Dakota, Oklahoma, Pennsylvania, Texas, and
Wyoming).
These nine States were chosen for review because of their large production volumes,
the potential presence of H2S in their well fields, and their distribution across the United
States. The nine States contain over 68 percent of the total oil wells (419,989 wells/613,810
total U.S. wells) and 54 percent of the gas wells (147,360 wells/272,541 total U.S. wells)
producing in the United States in 1991 (Petroleum Independent, 1992). For these States,
regulatory agencies are identified, H2S regulations for routine emissions and accidental
releases are described, enforcement programs are discussed, records and programs to track
accidental H2S release are included, and the effectiveness of each State program is assessed
qualitatively. The qualitative evaluation identifies existing control standards and the
populations or ecosystems the-standard is intended to protect.
In addition, these States account for 67 percent of the total U.S. oil production and 87
percent of the total U.S. natural gas production (Petroleum Independent, 1992). State
regulations for H2S emissions from the oil and gas industry in Oklahoma, Texas, Michigan,
IV - 1
-------
Table IV-1. Ambient Air Quality Standards for HjS
State
California
Connecticut
Kentucky
Massachusetts
Minnesota
Missouri
Montana
Nevada
New York
North Dakota
Oklahoma
Pennsylvania
Rhode Island
Texas
Virginia
Hawaii
Delaware
Indiana
Concentration
(ppb)
30
200
.. 10, ,
14
50a
30b
500"
30b
50°
240
10
200d
100C
100
100
10
80
160
40
30
50
Average Time
(hours)
1
8
.1 . •
24
0.5
0.5
0.5
0.5
1
8
1
I
24
0.5
1
1
0.5
24
1
1
1
'Not to be exceeded more
bNot to be exceeded more
days.
cNot to be exceeded more
dNot to be exceeded more
than two times/year.
than two times/five consecutive
than one time/year.
than one time/month.
. -FV--2
-------
and California were reviewed in greatest detail because they are major oil and gas producing
States. These states have extensive regulations dealing with H2S in the oil and gas industry.
California's air quality program is managed by 33 independent air pollution control districts
and its Division of Oil and Gas is divided into 6 districts where District heads have great
flexibility in enforcing rules. Therefore, California's program is discussed in the greatest
detail.
Selected Oil and Gas Producing States
Oklahoma
The H2S regulations for Oklahoma (10.3.16, "Operation of Hydrogen Sulfide Areas")
were listed in Guidelines for Petroleum Emergency Field Situations in the State of Oklahoma,
a guidance manual that expands on the regulations. The guidance manual contains sections
on characteristics and effects of H2S, recommended guidelines for safe drilling and
production operations in an H2S environment.
The following agencies regulate oil and gas activities in Oklahoma:
Oklahoma Corporation Commission (OCC), Oil and Gas Conservation Division
Oklahoma Air Quality Service
Osage Indian Tribe (OIT)
U.S. Bureau of Land Management
U.S. Environmental Protection Agency.
The OCC has jurisdiction over laws and regulations "relating to the conservation of
oil and gas and the prevention of pollution in connection with the exploration, drilling,
producing, transporting, purchasing, processing and storage of oil and gas..." (OCC, 1986).
The OIT has sole jurisdiction regarding oil and gas operations in Osage County. The U.S.
Bureau of Land Management has responsibility for cases where both surface and mineral
rights are owned by the Bureau or by an Indian tribe other than the Osage Tribe.
As noted earlier hi this chapter, Oklahoma has an H2S ambient air quality standard.
This regulatory program (administered by the Air Quality Service) is used to control routine
emissions (through permit) from oil and gas facilities.
The accidental release of H2S from facilities is regulated by the OCC. Rule 165:10-3-
16 of the OCC rules requires operators to assess their facilities for H2S release potentials that
would fause harm to the public. The rule is applicable to all facilities that handle natural gas
containing 1 x 105 ppb H2S or more and have a significant radius of exposure to cause
adverse effects on the public with the exception of storage tanks. The "radius of exposure"
is that distance from-a source where the ground level concentration of hydrogen sulfide
resulting from a release of gas from a facility is 1 x 105 ppb or 5 x 105 ppb whichever is
applicable in the Rule. The Rule applies as follows:
IV- 3
-------
g
• Does the facility (drilling, producing, injection, storage, etc.) handle
hydrocarbon fluids containing 1 x 10* ppb H2S or more? If yes;
• Determine the 1 x 10s ppb radius of exposure using an equation required in the
Rule or other methods approved by the Commission. The H2S escape rate
from the facility must be determined as required by the Rule.
• If the 1 x 105 ppb radius of exposure is In excess of 50 feet, warning, marker
and security provisions must be provided at the facility.
• If the 1 x 10s ppb radius of exposure is in excess of 50 feet and includes a
public area or if the 5 x 105 ppb radius of exposure is in excess 'Of 50 feet and
includes a public road or if the 1 x 105 ppb radius of exposure is in excess of
3000 feet, control and safety equipment and a contingency plan must be
provided for the facility.
• ,, Facility storage tanks near atmospheric pressure containing 5 x 105 ppb or
greater H2S must have warning signs, wind indicators and possible fencing.
Radius of exposure calculations are not applicable to storage tanks.
• H2S training, injection or flaring provisions, accident notification and other
requirements are addressed in the Rule (personal communication, W. Freeman,
Shell Oil, 6/23/93).
The OCC does not keep an emissions inventory of accidental H2S releases, but it does
keep an inventory of wells with actual or potential H2S problems. Furthermore, an inventory
of inspection data is kept by individual inspectors in the State and the local field offices.
Any emissions of H2S exceeding the OCC standard of 2.5 x 104 ppb must be reported to the
OCC by the emitting facility. Rule 3-2032, H2S Operation, is intended to provide for the
protection of the public's safety in areas where H2S concentrations greater than 1 x 105 ppb
may be encountered.
Drilling facilities are not required to submit data periodically to show that they are in
compliance with regulations. Facilities report release of H2S on an "honor system" once
permits are granted. When noncompliance is discovered, the OCC can use administrative
proceedings to shut down or fine the operation. However, in recent years, there has been no
evidence of noncompliance with the H2S regulations.
The OCC lists training requirements for employees who will work in areas of
potential H2S exposure. The training must cover hazards and characteristics of H2S,
operation of safety and life support systems, and emergency response procedures. OCC
safety inspectors-attend annual industry-sponsored training programs in order to stay current
on safety developments and to check the safety of their breathing equipment. Each H2S
"inspector is required to have an H2S monitor, a manual, H2S gas monitoring test tubes, and a
TV - 4
-------
self-contained air breathing apparatus. Specific H2S provisions also exist regarding H2S
detection and alarm equipment, accident notification, injection, and flaring. In 1991, the
OCC and the industry jointly sponsored an H2S safety seminar. A film about H2S safety was
presented to regulatory and industry personnel, and questions about H2S safety were
answered. Safety training has also been provided to local police, fire, sheriff and ambulance
services, and to interested oil and gas operators, as requested.
The enforcement, field monitoring, and inspection departments of the OCC employ 69
people. The State currently has two H2S inspectors and a third is anticipated. In 1991, one
emergency involving the accidental release of H2S was reported to the OCC. However, the
accident, which resulted hi the death of one worker, was not related to the extraction of oil
and gas resources.
Texas
Six agencies regulate oil and gas activities in Texas:
Railroad Commission of Texas
Texas Water Commission
Texas Air Control Board
Texas Parks and Wildlife Department
U.S. Army Corps of Engineers
U.S. Environmental Protection Agency.
The Railroad Commission regulates most of the operations of the oil and gas industry
but has no authority over the Clean Air Act Amendments. The Railroad Commission is
responsible for the well spacing, construction requirements (casing etc.), and most aspects of
environmental protection and works with other State Agencies to ensure that their concerns
are addressed. The Texas Water Commission works with the Railroad Commission on water
quality issues. The Texas Air Control Board has jurisdiction over the regulation of oil field
activities that generate air emissions. The Texas Parks and Wildlife Department investigates
fish kills and water pollution complaints and evaluates the effects of discharged wastes on
fish and wildlife. The Railroad Commission has jurisdiction over all oil and gas activities on
Federal lands in Texas, regardless of who owns the mineral rights. The U.S. Army Corps
of Engineers has permitting responsibility for activities that would affect statutory wetlands.
The Texas Air Control Board (TACB) is responsible for enforcing the Texas ambient
air quality standard for H2S (discussed previously). Certain allowances are made from the
air standard if the hydrogen sulfide affects only property used for other than residential,
recreational, business, or commercial purposes, such as industrial property and vacant tracts
and range lands not normally occupied by people (i.e., the emission limit is raised to 120
ppb/30 min). If an operator violates these ambient air levels, corrective action must be taken
such as flaring, installation of vapor recovery, etc. Consequently, the unauthorized emission
of H2S that exceeds the time weighted averages for the land use discussed above is a
IV-5
-------
violation of regulation and must be addressed by the operator. In addition, the TACB
requires permits for facilities that handle sour gas emissions from crude oil storage which
also address emergency releases from these type facilities.
Texas regulations on H2S for drilling, extraction, and abandonment are listed under
Statewide Rule 36 - Hydrogen Sulfide Safety, Section 3.36 (051.02.02.036, "Oil, Gas, or
Geothermal Resource Operation in Hydrogen Sulfide Areas," as amended September 1,
1976). The Hydrogen Sulfide Safety Rule in Texas—issued to address accidental
releases—applies to facilities that could expose the public to concentrations of H2S in excess
of 1 x 105 ppb as a result of an accidental release. Operators handling hydrocarbon fluids
containing 1 x 105 ppb or more H2S must determine if the Rule applies to their facility. If it
does, they must calculate the radius of exposure; determine if the public will be impacted* •_
and, if so, install warning signs, ensure security measures, address storage tank
requirements, install appropriate safety equipment, develop contingency plans, provide
training and implement other requirements as necessary. In addition, all operators subject to
Rule 36 must submit a Certificate of Compliance to the Railroad Commission to demonstrate
that they have complied with these requirements. This rule requires that employees working
in HjS areas be trained hi the characteristics and effects of the gas. The Railroad
Commission of Texas publishes a training manual containing this information. The Texas
and Oklahoma regulations are virtually identical. Most of the Texas regulations were
discussed in the previous section on Oklahoma regulations. The Hydrogen Sulfide Safety
Rule in Texas does require safety equipment, alarm equipment, monitors, etc., but does not
specify exact types in an attempt to remain flexible and allow for new technology. It was
designed for the protection of the general public rather than industry, since OSHA rules are
designed to protect industry workers (personal communication, W. Freeman, Shell Oil,
6/23/93).
In Texas, the Railroad Commission does keep an emissions inventory on accidental '
H2S releases. Any emissions of H2S that are found to be of sufficient volume to present a
hazard and/or any H2S-related accidents must be reported to the Railroad Commission by the
emitting facility. Operator certificates are required by the Railroad Commission to
demonstrate that prevention and response measures have been taken to address accidental
releases of H2S.
There was one case of noncompliance during 1991, which involved natural gas
leaking from a pipeline. The Railroad Commission canceled the Certificate of Compliance
for the operators of the well, which prevented the facility from producing or selling the
product until the leak was fixed. In 1991, there were emergencies involving the accidental
releases of H2S. Those accidents were discussed in Chapter TTT,
The enforcement, field monitoring, and inspection departments of the Railroad
Commission employ 215 people. Ground testing for traces of H2S is performed near the
wells. Emission data on each well are submitted to the Railroad Commission using the Form
of Compliance. When noncompliance is discovered, the Commission uses administrative
IV-6
-------
proceedings to implement the following enforcement actions: enforcement letter, pipeline
severance, zero allowable emissions, sealing, permit revocation and/or administrative
penalties. The Railroad Commission may also seek civil penalties through the Attorney
General's Office.
Michigan
The Michigan regulatory program is published in Michigan's Oil and Gas
Regulations - Act 61 (P. A. 1939 as amended and promulgated rules - Circular No. 15,
revised in 1987, MDNR). Most of the regulations in the Michigan guidance were covered in
the sections on Texas and Oklahoma regulations.
A review of Michigan's Oil and Gas Regulations reveals that the State has a
comprehensive set of regulations dealing with H2S. The Michigan rules require extensive
training for all employees and contractors involved in drilling, completing, testing,
producing, repair, workover or service operations. Employees must receive training in the
following areas: physical properties and physiological effects of H2S, effects of H2S on
metals and elastomers, emergency escape procedures, location and use of safety equipment,
the location and operation of detection and warning systems and the location of primary and
secondary briefing areas. Briefing areas are defined hi Michigan's Oil and Gas Regulations
as the areas "nearby where personnel can assemble in case of an emergency." Michigan
defines safety equipment as including items such as first aid kits, dry chemical fire
extinguisher, ropes, flare guns, portable H2S detectors and warning signs.
In addition to training requirements, the Michigan oil and gas regulations contain
comprehensive rules for the preparation of a contingency drilling plan in order to provide a
plan for alerting and protecting, personnel and the public in case of an emergency.
Five agencies regulate oil and gas activities in Michigan:
Michigan Department of Natural Resources (MDNR)
Michigan Department of Commerce, Public Service Commission
U.S. Forest Services
U.S. Bureau of Land Management
U.S. Environmental Protection Agency.
The Department of Natural Resources is responsible for the well spacing, construction
requirements (casing, etc.), and most aspects of environmental protection. The Michigan
Public Service Commission regulates the production of gas from dry natural gas reservoirs
and the safety of gas pipeline construction. When dealing with split estate situations, the
U.S. Forest Service will issue a Special Use Permit which allows an operator to drill within
the-forest boundary. When both the forest surface and corresponding mineral rights are
Federally owned, the U.S. Bureau of Land Management (BLM) issues drilling permits and
the U.S. Forest Service issues Surface Use Plans. The BLM issues drilling permits in all
-IV --7-
-------
cases related to onshore Federal mineral estates (personal communication, T. Alexander,
DOE, 2/22/93).
Worker safety issues are the responsibility of the Michigan Department of Labor.
Part 57 of the General Industry Safety Standards Commission Safety Standards deals with oil
and gas drilling operations safety standards. Under Rule 5717(1), the drilling and servicing
of wells containing H2S shall be conducted as prescribed in the American Petroleum
Institute's Recommended Practice No. 49 (API, 1987).
The MDNR's Air Quality Division regulates H2S emissions from all sources in the oil
and gas industry. Rule 336.1403 states: "It is unlawful for a person to cause or allow the
emission of sour gas from an oil or natural gas producing or transporting facility or a natural
gas processing facility without burning or equivalent control of hydrogen sulfide and
mercaptans." The Rule does allow operators with stripper wells to emit small quantities of
H2S unless one complaint is received from the public which would require some type of
abatement technique to be imposed. All facilities handling H2S are subject to these
regulations.
The Geological Survey Division (GSD) of the Department of Natural Resources
regulates accidental releases of H2S in the oil and gas industry. In addition, it overlaps with
the Air Quality Division on emission controls at production facilities. It appears that two
agencies in the MDNR regulate H2S handling facilities. Under Rule 299.1911-1939,
operators handling hydrocarbon fluids containing more than 3 x 105 ppb H2S must define a
Well Class (defined by the radius of exposure in Rule 299.1912) to determine the
applicability of the Rule. The radius of exposure is defined using the same dispersion
equation as Texas Rule 36. The Rule addresses equipment standards, location standards for
drilling and production equipment, contingency planning, training, drilling, testing,
production operations, servicing operations and nuisance odor requirements (personal
communication, W. Freeman, Shell Oil, 6/23/93).
The enforcement, field monitoring, and inspection departments for oil and gas
regulation by the Geological Survey Division (GSD) of the MDNR employ 47 people. Wells
are retested one year after the initial well test was performed, to check for compliance with
laws. Further periodic tests are required only at the request of the MDNR. When a well is
not in compliance, the MDNR can use administrative proceedings to shut down drilling
processes and production, stop issuing permits to drill, stop well ownership transfers, and
issue fines. Fines are also issued for falsifying records required by the GSD enabling
legislation (Act 61, P.A. of 1939, amended). Violation of the Act or a rule or order under
the Act carries a penalty of not more than $1,000.00 per day that the violation continues. In
1991, there was no evidence of noncompliance for the release of H2S.
The MDNR does not keep an emissions inventory of the accidental releases of H2S
from well blowouts and flare gas releases. Emissions of H2S are reported by industry
personnel to MDNR field personnel, who may keep records on the releases. One incident
IV- 8
-------
was reported to the MDNR in 1990, which involved a pumper who was working on a
storage tank. The exact date and nature of the incident were not available.
California
The following agencies regulate oil and gas activity in California:
• California Department of Conservation, Division of Oil and Gas
• California Water Resources Control Board and the nine Regional Water Quality
Control Boards
• California Department of Health Services
• California Department of Fish and Game, Office of Spill-Prevention and Response,
• California/EPA Department of Toxic Substances Control
• California State Fire Marshall's Office
California Public Utilities Commission
California OSHA
California Air Resources Board and the county or multi-county regional Air
Pollution Control Districts
• California Governor's Office of Emergency Services
• State Lands Commission
• California Coastal Commission
• Local government agencies
• U.S. Bureau of Land Management
• U.S. Department of Energy
• U.S. Environmental Protection Agency.
The Division of Oil and Gas of the California Department of Conservation is
responsible for the management and conservation of oil and gas resources. The Division
issues permits for and inspects the drilling, reworking, and abandonment of oil and gas
wells. Under delegated authority from the EPA, the division also issues underground
injection control well permits for Class n injection wells.
Division 3 - Oil and Gas, part of the California Code of Civil Procedure, contains the
California laws for conservation of petroleum and gas (CDC, 1991). Table IV-2 highlights
key sections of the law applicable to H2S releases. Although, there is no quantitative limit to
H2S emissions, the law grants the supervisor of the Oil and Gas Division, discretionary
authority to control H2S releases to ensure protection of human health and the environment.
California's Code of Regulations contains the oil and gas regulatory program enforced
by the Division of Oil and Gas. These regulations are highlighted in Table IV-3. These
rules include the definition of the term "critical well," requirements for contingency plans,
IV - 9
-------
Table IV-2. Highlights of California Laws for Conservation of Petroleum
and Gas Pertaining to HjS Emissions
Ch
Section
Subject
Description
1, 4,3219
1, 4,3224
1, 4,3228
1, 4.3235
1, 4, 3236
1,4.1, 3241
1,4.2,3251
3, ,3600
Blowout
prevention
Order for repair
Abandonment
of wells
Complaint
Penalty
Strategy to
extract gas in
high risk areas
Define
"hazardous well"
Spacing wells
Where high-pressure gas exists, use
adequate casing and safety devices
Authorizes supervisor to order tests
or repairs needed to prevent damage to
life, health, natural resources, etc.
Protects ground and surface water
from gas-bearing strata
Authority to investigate complaints
For obstructing enforcement,
$100 - $1,000 or up to 6 months
imprisonment per offense
Develop strategy to extract hazardous
gases from abandoned wells to protect
public health and safety
Poses danger to life, health, or natural
resources
Well must be at least 100 feet from
parcel's boundary or public road
IV- 10
-------
rules include the definition of the term "critical well," requirements for contingency plans,
and environmental protection.
The Division of Oil and Gas has also published a guidance document on H2S, Drilling
and Operating Oil, Gas, and Geothermal Wells in an H^S Environment (Dosch and Hodgson,
1986). This guidance document reflects the American Petroleum Institute's publication
RP 49, Safe Drilling of Wells Containing Hydrogen Sulftde (API, 1987) and recommends
safety procedures for H2S release scenarios. The California Division of Oil and Gas
(CDOG) is divided into six districts. Figure IV-1 shows the six districts and the distribution
of H2S in California, presenting parts per million of H2S gas in some California oil and
gepthermal fields. Table IV-4 shows the documented concentration by oil field in each
district. Three of the districts are discussed here.
District 1 of the Division of Oil and Gas has three oil and gas inspectors and seven
energy engineers who inspect well drilling and rework operations. The inspectors wear
tri-gas monitors (H2S, oxygen, and combustibles). The well-permitting program does not
specify H2S limits. All wells are inspected at least once a year. Idle wells must be
pressure-tested periodically to minimize casing leaks. Steam flooding, an enhancement
process that often creates H2S, is used frequently in the district. District 1 authorities know
of past H2S incidents leading to human injuries; however, because records are not
computerized, exact data are not available (personal communication, K. Carlson CDOG
8/27/92).
District 3 has 1,929 producing wells and 2,845 shut-in wells (i.e., no production is
made on the well; its pump is turned off, the stuffing box is closed, and it is inspected to
ensure no leakage). Three field inspectors cover District 3 (personal communication
A. Kollar, CDOG, 8/28/92).
District 4, which includes Kern County, has nine field inspectors, each equipped with
an "escape pack" for H2S protection. An environmental inspection is performed for every
lease on every well. The inspection covers the surface area, well condition, tank condition,
and operation. There are more than 40,000 wells in Kern County alone. District 4 had no'
records of H2S incidents. However, inspectors in Kern County/San Joaquin Air District
(described below) have documented incidents of H2S releases (personal communication R
Bowles, CDOG, 8/27/92).
The California Air Resources Board is authorized to enforce a statewide ambient air
quality limit for H2S emissions of 30 ppb over one hour's averaging time. However,
California's air quality program is managed on a smaller scale by the 33 county or multi-
county air pollution control districts (APCDs) shown in Figure IV-2 (CA Air Resources
Board, 1991). Each district acts as an independent regulatory agency, establishing and
IV-41-
-------
Table IV-3. Highlights of Title 14, Chapter 4 of the California Code of
Regulations - Development, Regulation, and Conservation
of Oil and Gas Resources
Article, Section Subject
Description
Subchapter 1
1,1712
2,1720
2.1,1721
3,1722
3,1724.3
3,1724.4
Scope
Critical well
Well spacing
General
Well Safety Devices
Testing/inspecting
Safety Devices
Onshore drilling and production;
grants Oil and Gas Division Supervisor
authority to establish field rules
Addresses distances to public areas
and navigable waters
Objectives include protecting public
health, safety, welfare and the
environment
Good oilfield practices, blowout
prevention and control plan, prompt
reporting of significant gas leaks
Required of certain critical wells
Test at least every 6 months
Subchapter 2
Environmental
Protection
Requires covers on well cellars,
no excessive leakage including
wellheads and pipelines
'TV-12"
-------
The Genera Geothermal Field"*. :4»: */
40 ppm to 450 ppm »- •
Guadalupe .,'i
13S.OOOppm '
•; .-,,* P<
! '.' .-. "1 k * *•-
I .' •' \ • X ^. '
I / '. •"•
i '• •'" '"*" x -
1 I ': f I "T; Cymric V*^' »i -..
; / ( ./;.. 1.600 ppm \ X"** '\
'.' s \ ~~~~\ *• X v '"* •
\ -« \' X x<>-X-v i.
\ *%. \ N£r
. \ ^-:«.» \ * * ••
\ \ %%-. _ ,
t*r. •"*''••?•«• "*"
VK^T
^RailroadX^"-^ * V-
Tj McKittrick
Santa Maria Valley
97n nnn .»._
Zaca
400 ppm
South ElwwjdOflihort
20.000 ppm
. fa- fldda ^y.
AUfielda an oil fielda except for the Geyat
HjSione,
»v«sss;»..,
Rineon "";
100 ppm -S^.
San MigueUto *"•*"•«!._
200 "P» 5f',Sm ,,.^,. -^.^
Shiella Canyon * (i **«,
€0 ppm
n Gnthemal Oeld.
.
^^ - HjSione,
100 ppm • Numbe« indkaiehigheet report.! H2SCon«ntt,tion.wh«, known
* • California Oirinon of OU and Gai diatrin numban
Note: Concantrmtioiu inwnn
inonaily left in ppm.
' •
Source: Dosch and Hodgson, 1986.
Figure rV-1. Parts per million of E^S gas in some California oil
and geotfaermal fields. Data compiled in 1976.
BalmontOOahore *"
South Cuyama
(**•* UOO ppm
vO
IV- 13
-------
Table IV-4. B^S in California Oil, Gas, and Geothermal Fields
Oil and Gas
District
Fields with BLjS
Concentrations
IrlO5 ppb or Above
Fields with H.jS
Concentrations
Under 100 ppm
Fields with BL.S
Odor, But With
Concentrations Unknown
1
2
3
4
5
6
Geothennal
District
G3
—
Rincon, 1 x 105 ppb
San Miguelito, 2 x 10s ppb
Ventura, 3 x 105 ppb • - -
Casmalia, 1.3 x 107 ppb
Cat Canyon, 2.5 x 10s ppb
Cuyama So., 1.3 x 10s ppb
Elwood So., Offshore, 2 x 107 ppb
Guadalupe, 1.35 x 108 ppb ,
Lompoc, 1.31 x 10s ppb
Orcutt, 3.06 x 105 ppb
Russell Ranch, 1.53 x 10s ppb
San Ardo, 4 x 10s ppb
Santa Maria Valley, 2.7 x' 108 ppb
Zaca, 4 x 10s ppb
Midway Sunset, 1 x 105 ppb
Cymric, 1.6 x 106 ppb
—
—
The Geysers, 4x10* - 4.5xl06 ppb
—
Shiells Canyon
60 ppm
—
—
—
Wilmington, Huntington Beach,
Newport, Torrance, Brea Olinda
Aliso Canyon, Bardsdale, Big Mountain,
' Del Valle, Las Llajas, Oak Park,
Oakridge, Ojai, Piru, Santa Paula, Santa-,
Susana, Simi, South Mountain, Tapo
Canyon So., Temescal, Torrey Canyon,
and West Mountain
Capitan Onshore, King City
Four Deer
North Belridge, South Belridge, Blackwells
Comer, Edison, Northeast Edison, Kern River,
Lost Hills, McKittrick, Mount Poso, Poso
Creek, Railroad Gap, and Wheeler Ridge
Coalinga
—
—
in some California oil and geothermal fields. Data compiled in September 1976. (Data in the first two columns are on
Figure IV-1).
Source: Dosch and Hodgson, 1986.
IV-14 -
-------
enforcing air quality rules tailored to the district's needs. Districts with significant oil
production are:
Ventura County
Santa Barbara County
San Joaquin Unified Air District
South Coast Air Quality Management District
Monterey Bay Unified Air District
Bay Area Air Quality Management District.
. This report highlights H2S emissions programs in four districts: Ventura County,
Santa Barbara County,-San-Joaquin-Unified Air District, and the South Coast District.
Ventura County has Rule 54, "Sulfur Compounds," hi place for air emissions
containing sulfur compounds. This rule, adopted in 1968 and most recently revised in 1983,
includes a limit for H2S not to exceed 1 x 10* ppb by volume at the point of discharge. The
point of discharge includes any distinguishable emission point such as valves, flanges, or
process vents. There are no control technology regulations for H2S in Ventura County other
than these equipment standards. Another H2S rule requires that the aboveground average
concentration at or beyond the property boundary shall not be in excess of 60 ppb for over 3
minutes. The Ventura County limits were adopted in 1968 when the APCD was formed.
Natural emissions of H2S are low in the county's oil well fields, and H2S monitoring is only
performed when a problem is suspected (i.e., when the odor is detected). The APCD uses
hand-held monitoring devices to inspect problem areas. No routine monitoring records are
kept on file in Ventura County, but wells are inspected at least once a year, with large wells
inspected more frequently (personal communication, K. Duval, Ventura APCD, 8/29/92).
Ventura County has an enforcement staff of about 20 people, including 8 field inspectors
(personal communication, K. Duval, Ventura APCD, 11/23/92).
Emission standards in Santa Barbara County are basically the same as in Ventura
County. However, tighter emission limits are applied hi parts of the county with SQ, (an
oxidation product of H2S) nonattainment areas. Rule 309, "Specific Contaminants," for
Santa Barbara County states that sulfur recovery units shall not emit more than 5 x 105 ppb
as SO2 or 1 x 104 ppb as H2S. Rule 310 for odorous organic sulfides states that
concentrations of organic sulfides beyond the property boundary shall not exceed 60 ppb/3
minutes or 30 ppb/hr. For gas produced and used as fuel in equipment on a well site, the
sulfur content limit in the county's northern air shed is 7.96 x 105 ppb sulfur; in the southern
county air shed, the limit is 2.5 x 105 ppb. Control technologies are not used on well heads
for H2S emissions. However, controls do exist for volatile organic compound (VOC)
emissions from well fittings, stuffing "boxes, well cellars, sumps and pits. Rules are being
developed to require these controls, primarily in the surface area of the well cellar to control
the release of VOC. This technology will also control H2S emissions indirectly. The
county's 10 field inspectors inspect wells for all types of emission sources at least once a
year. H2S violations via the total sulfur emission limit are not a problem because by the time
IV- 15
-------
Multi-County Districts
1 Bay Area (415) 771-6000
2 Feather River (July 1, 1990)
Sutler (916) 741-7500
Yuba (916) 741;6484
Great Basin (619) 872-8211
Monterey Bay (408) 443-1135
North Coast (707) 443-3093
Northern Sierra (916 265-1398
South Coast (818) 572-6200
Yolo-Solano (916) 669-6700
San Joaquin Valley (209) 222-6111
County AFC Districts
Amador (209) 223-6406
Butte (916) 891-2882
Cmlaveras (209) 754-6400
Colusa (916) 458-5891
El Dorado (916) 621-5897
Glenn (916) 934-6500
Imperial (619) 339-4606
Lake (707) 263-7000
Lassen (916) 357-8311 xllO
Mariposa (209) 966-3689 '
Mendocino (7070 463-4354
Modoc (916) 233-3939 x401
No. Sonoma (707) 433-5911
Placer (916) 889-7130
Sacramento (916) 386-6650
San Bernardino (619) 243-8920
San Diego (619) 694-3307
San Luis Obispo (805) 549-5912
Santa Barbara (805) 961-8800
Shasta (916) 225-5674
Siskiyou (916) 842-8029
Tehama (916) 527-4504
Tuolumne (209) 533-5693
Ventura (805) 654-2806
Source: California Air Resources Board, 1991.
Figure IV-2. Multi-county districts.
IV- 16
-------
the ambient air quality standard is exceeded, the operator has already been alerted to a safety
problem and is responding. The county has seven currently active H2S ambient monitoring
stations; however, these are at oil and gas processing facilities, rather than at well fields
(personal communication, J. Top, St. Barbara APCD, 8/20/92).
The San Joaquin Unified Air District enforces Rule 407, "Sulfur Compounds," which
limits the emission concentration of sulfur compounds at the point of discharge to 0.2 percent
volume calculated as SO2 (or 2 x 106 ppb SOj). This rule, adopted in 1972 and renumbered
hi 1989, applies to any gas line or vapor control line from a well. Rule 220.1, "New and
Modified Stationary Source Review Rules," has a trigger value for H2S or total reduced
sulfur or sulfur compounds other than SOX of 54.79 Ib/day. If this value is exceeded, the
responsible party must use Best Available Control Technology (BACT).on the emission.,._. .,.
source. Rule 220.1 was adopted in September 1991 and revised March of 1992.
The San Joaquin District does not look at or enforce H2S regulations until the 2 x 106
ppb SO2 emission limit is exceeded, because the rule is based on the impact of SO2 on human
health and the environment, not on the health effects of H2S. No ambient monitoring of H2S
is required by the district. However, the oil companies are required to keep their own
records of SO2 monitoring for two years. Companies also have H2S monitoring data, and the
State has the authority to request these data at any time (personal communication,
M. Amundsen, San Joaquin, 8/21/92).
Kern County, part of the San Joaquin Unified Air District, has three of the largest
producing wells in the United States. The county's production volume is exceeded only by
Alaska, Texas, and Louisiana. The wells in Kern County produce a unique heavy crude and
some use steam injection to enhance pumping. H2S is a problem hi well fields in the county,
where numerous stripper wells (defined in Chapter JJ) are operating. The county has a ten-
person enforcement team that performs inspections at least once a year. Steam casing
collection systems, valves, fittings, etc., are inspected by staff wearing H2S monitors.
Inspectors hi Kern County have been exposed to H2S hi the field. In one case, an inspector
was exposed to greater than 1 x 106 ppb. The case involved a report from a fire department
station downwind of a well and complaints of odor and illness. H2S was measured at the
station at 5 x 104 ppb. The source was a leaking underground gas recovery line. Companies
are required to keep records of such incidents and report them to CAL OSHA (personal
communication, M. Amundsen, San Joaquin Unified Air District, 8/21/92).
During conversations with Kern County representatives, it was noted that an
important control technology for H2S at wells is a casing collection system, which can be
, added to collect natural gas containing H2S that has built up in the casing over time. If the
natural gas pressure is not relieved, well production is hindered. Companies tend to release
this gas to the atmosphere, but a casing collection system can treat the gas by vapor
incineration (98 to 99 percent hydrocarbon destruction efficiency). However, the economic
incentive to put casing collection systems on stripper wells is normally low due to the low
TV - 17
-------
volume of oil produced (personal communication, M. Amundsen, A. Phillips, San Joaquin,
8/21/92).
The South Coast Air Quality Management District has no specific regulations
pertaining to H2S or oil production. Rules in place that indirectly control H2S emissions
include Rule 431.1, "Sulfur Content of Gaseous Fuels," which states that, effective May
1994, natural gas cannot be burned or sold for burning if it contains greater than 4 x 104 ppb
total sulfur. This rule also requires organic vapor recovery systems, which would recover
any H2S gas along with the volatile organics. Rule 402 could also apply to H2S, particularly
for stripper wells that are too small for permitting. This rule is a nuisance rule that could be
used to close wells if, for example, neighbors complained about H2S odors or other health
effects (personal communication, C. Bhatti, South Coast AQMD, 8/25/92). The South Coast
District's enforcement program is managed as part of the Stationary Source Compliance
Office, which has a staff of 500 (personal communication, C. Bhatti, South Coast AQMD,
11/23/92).
, California's Occupational Safety and Health Administration is authorized to administer
the Federal OSHA program. There are two OSHA standards that apply to H2S. One focuses
on the maintenance and use of valves. The second is the Permissible Exposure Limit for
H2S. It is difficult to monitor compliance with this limit because operations are outdoors.
CAL OSHA maintains a database of occupational accidents. No accidents were found in the
database related to H2S releases at California oil wells dating back to 1982 (personal
communication, R. Hayes, CAOSHA, 9/11/92). However, H2S incidents were recorded in
some of the Air Pollution Control Districts and Division of Oil and Gas Districts (personal
communication, M. Amundsen, San Joaquin Unified Air District, 8/21/92).
The California Water Resources Control Board is generally responsible for the
protection of the State's waters and for preserving all present and anticipated beneficial uses
of these waters. The California Department of Health Services is responsible for the
regulation of hazardous wastes. It determines which waste streams and constituents are
hazardous under California's laws. The State Land Commission has joint responsibility with
the Division of Oil and Gas for wells on State-owned, onshore lands.
The Office of Emergency Services administers Chapter 6.95 of the California Health
and Safety Code which states that every business handling any hazardous material greater
than 55 gal., 500 Ib. or 200 cubic feet (gaseous material) must register and develop an
emergency response plan and business plan. If the business handles extremely hazardous
substances onsite exceeding threshold planning quantities (500 Ib for H2S), a preliminary
analysis of the facility must be made to determine if a significant risk potential exists for
accidental release of the extremely hazardous substance. If the potential does exist, the
facility must develop and submit a "risk management and prevention program" that addresses
how to reduce or eliminate the potentiaLfor accidental release (personal communication,
Dr. F. Lercari, Office of Emergency Services, 9/13/93).
TV- 18
-------
A Comparison of H,S Regulatory Programs in Four States
Table IV-5 presents a summary of regulatory programs for H2S across California,
Michigan, Oklahoma, and Texas. This summary addresses the area of "state ... control
standards, techniques, and enforcement" designated for evaluation in Section 112(n)(5) of the
Clean Air Act Amendments. Appendix B tabulates components of the States' regulatory
programs in greater detail.
Texas, Oklahoma, and California have H2S ambient air quality standards in place.
The California standard (30 ppb over 1-hr averaging time) is more stringent than the Texas
standard (80 ppb over 0.5-hr averaging time) and the Oklahoma standard (100 ppb over
0.5-hr averaging-time). -Michigan does not have ambient air quality standards for H2S,
The number of agencies in each State regulating oil and gas operations ranges from
two in Oklahoma and Michigan to eleven in California. The enforcement staff, which
includes inspectors and field monitoring staff, numbers 69 in Oklahoma, 215 in Texas, and
47 in Michigan. California's air emissions program is regulated by districts. The Santa
Barbara District, an area with high concentrations of I^S in its oil fields, has 10 field
inspectors who are also responsible for inspecting other commercial operations. Kern
County, California, has a staff of 10 field inspectors who also have other inspection
responsibilities.
Michigan, Oklahoma, and Texas each have H2S-specific regulations related to public
safety. In California, State law grants the Director of the Division of Oil and Gas discretion
to require additional controls (for areas such as H2S emissions) on a case-by-case basis.
However, none of the four States has specific H2S standards in place to protect the
environment, i.e., ecological protection.
Of the four States reviewed, only Texas maintains an inventory of accidental releases
of H2S from drilling and production operations. However, all four states require notification
when threatening accidental releases occur. None of the four States requires reporting of
H2S routine emissions. "Routine" excludes such incidents as vapor recovery unit failures and
other equipment upsets.
Texas, Oklahoma, and Michigan require worker safety training for H2S. California's
Division of Oil and Gas, however, provides guidance on worker safety in the form of a
publication (Dosch and Hodgson, 1986).
Other Large Producing States
The EPA gathered initial information on several State regulations and later contacted
selected State agencies to obtain additional information on the unique aspects of the State
regulations governing H2S emissions in the oil and gas industry. The results of each State
review are summarized in the following sections.
IV - 19
-------
Table IV-5. A Comparison of Four States' HjS Regulatory Programs
HjS Area Oklahoma
Ambient air . 0.10 (0.5 hr)
quality standard?
Number of State 3
agencies regulating oil/gas
Size of enforcement/ 69
inspection staff
Specific HjS regulations for:
Public Safety Yes
Ecological Protection No
(administered by
environmental agency;
Inventory of accidental • No
releases kept by State?
Routine reporting of emissions No
required?
Notification of a threatening Yes
accidental release?
HjS training required? Yes
"Enforcement staff in California (example counties)
Santa Barbara County Air Pollution Control District:
Kern County (in San Joaquin Unified Air District):
California Division of Oil and Gas - District 7:
California Division of Oil and Gas - District 4:
Texas
0.08 (0.5 hr)
4
215
Yes
No
Yes
No
Yes
Yes
10
10
10
9
Michigan
No
2
47
Yes
Not clear
No
No
Yes
Yes
California
0.03 (1 hr)
6
*
No
No
No
No
Yes
Guidance
IV-20
-------
Louisiana
Five agencies regulate oil and gas activity in Louisiana:
Louisiana Department of Natural Resources, Office of Conservation
Louisiana Department of Environmental Quality
U.S. Bureau of Land Management
U.S. Army Corps of Engineers
U.S. Environmental Protection Agency.
The Louisiana Department of Natural Resources, Office of Conservation, regulates all
subsurface and surface disposal of oil- and gas-associated wastes (Statewide Order Governing
the Drilling for the Producing of Oil and Gas in the State of Louisiana). The office has
primary responsibility for all classes of underground injection control wells. The Office of
Conservation coordinates with the Louisiana Department of Environmental Quality, Office of
Water Resources, on any problem dealing with discharges in the oil and gas industry. The
U.S. Bureau of Land Management has jurisdiction over lease arrangements and post-lease
activity on Federal lands where the mineral rights are Federally held. The Office of
Conservation does not keep an emissions inventory for accidental H2S releases. Any
emissions of H2S that exceed the Office of Conservation standard must be reported to the
Office by the emitting facility.
The enforcement, field monitoring, and inspection departments of the Office of
Conservation employ 34 inspectors. Emission data are sent to the Office of Conservation
when an accidental release has occurred at the well site. The Office of Conservation,
through administrative proceedings, can respond with the following enforcement actions
when compliance is not met: compliance letters, compliance orders, civil penalty
assessments, suspension/revocation of permits and pipeline severance.
In recent years, there has been no evidence of noncompliance and no emergencies
involving the release of H2S from oil or gas wells. The drilling process is not a significant
threat because underground sources of H2S are much deeper than the wells being drilled.
New Mexico
Mexico:
Five agencies have responsibilities for regulating oil and gas activities in New
New Mexico Oil Conservation Division of the Energy, Minerals and Natural
Resources Department (OCD)
New Mexico Oil Conservation Commission
New Mexico Water Quality Control Commission
U.S. Environmental Protection Agency
IV--21
-------
• U.S. Bureau of Land Management.
The Oil Conservation Division of the Energy, Minerals and Natural Resources
Department is responsible for regulating oil and gas industry exploration and drilling,
production, and refining. Its duties include regulating "nonhazardous" liquid and solid
wastes from these operations to protect water quality, public health, and the environment.
The Oil Conservation Commission works in conjunction with the Oil Conservation Division.
The Commission initiates rules and orders to be administered by the Division. The Water
Quality Control Commission develops water quality control standards and water pollution
regulations'. The U.S. Bureau of Land Management has jurisdiction over all Federally
owned land, with the exception of Indian lands.
The Oil Conservation Division of Energy Resources (OCD) keeps emissions
inventories at the district level. There are four districts in the State of New Mexico; any
accidental release of H2S must be reported to the district division of the OCD. The
enforcement, field monitoring and inspection departments of the OCD employ 18 people.
Inspections are made by each district OCD office. In recent .years, there has been no
evidence of noncompliance with the H2S regulations set forth by the OCD, and no
emergencies involving H2S have been reported.
New Mexico's Oil Conservation Commission Rule 118 is intended to provide for the
protection of the public safety in areas where H2S concentrations greater than 1 x 105 ppb
may be encountered. This rule adopts the guidance of the American Petroleum Institute
publications RP 49 and RP 55 (discussed later in this chapter) and covers drilling, extraction,
and abandonment.
North Dakota
Five agencies regulate oil and gas activities in North Dakota:
North Dakota Industrial Commission, Oil and Gas Division
North Dakota State Department of Health and Consolidated Laboratories
U.S. Department of Agriculture, Forest Service
U.S. Bureau of Land Management
U.S. Environmental Protection Agency.
The North Dakota Industrial Commission, Oil and Gas Division, has regulatory
authority over the drilling and production of oil, and is responsible for protecting the
correlative rights-of the mineral owners, preventing waste, and protecting all sources of •
drinking water. The Bureau of Land Management has jurisdiction over drilling and
production on Federal lands, but the operator must obtain a permit from the Division of Oil
and Gas. Drilling on forest land must comply with the rules of the U.S. Forest Service.
IV-22
-------
Any well completed or recompleted on or after July 1, 1987 must be registered with
the State Department of Health and Consolidated Laboratories (NDSDH&CL). The
registration process includes completion and submittal of a form which provides information
about the well operator, well equipment (such as size and number of storage tanks, existence
of a heater treater and type of fuel on which it is fired, flare stack height, etc.), surface
equipment location, and disposition of produced gas. This form, submitted along with an
analysis showing the H2S concentration of any produced gas, constitutes registration.
Information derived from the registration is entered into a shared database, which is used by
the North Dakota Industrial Commission's Oil and Gas Division, for storing production data;
thus, an emissions inventory which represents actual emissions can be generated from the
database for all registered wells. H2S concentrations in wellhead gas are field-pool specific;
for example, within .the Little Knife Oil Field, gas produced from the Madison Pool will
have an H2S concentration of approximately 9.56 percent, gas produced from the Red River
Pool will be approximately 7.91 percent H2S, and gas produced from the Duperow and
Bakken pools is likely to contain only negligible amounts of H2S. H2S data from the
registrations are, therefore, entered into the database as field-pool specific data (personal
communication, D. Harman, NDSDH&CL, 5/19/93).
The enforcement, field monitoring, and inspection departments of the Division of Oil
and Gas employ 14 people. The NDSDH&CL handles most of these complaint-related
inspections. The Division of Oil and Gas can shut down an operation and fine up to $12,500
per day when compliance is not met. The NDSDH&CL can impose a fine and/or
imprisonment.
H2S typically constitutes between 4 and 10 percent of the oil and gas found in North
Dakota. Because of this prevalence, the State has established an ambient air quality standard
(shown in Table JTV-1).
The NDSDH&CL typically becomes more involved in situations where routine
emissions (as opposed to catastrophic/episodical releases) from a production facility result in
excessive ambient concentrations. This scenario typically manifests itself in the form of
citizen complaints. In these situations, it has been the Department's experience that an
equipment problem, such as flare stack ignitor malfunction (i.e., low efficiency flare),
storage tank gasket degradation and leakage, etc., has been the primary cause. Correction of
the immediate problem and implementation of a more rigorous maintenance schedule will
typically resolve these cases (personal communication, D. Harman, NDSDH&CL, 5/19/93).
Acute, unpredictable releases of H2S, such as natural gas pipeline rupture, etc., are typically
handled by the North Dakota Industrial Commission; however, the Industrial Commission
has had no reports of emergencies involving accidental releases of H2S in the past two years.
IV-23
-------
Pennsylvania
Six agencies regulate oil and gas activities in Pennsylvania:
Department of Environmental Resources,
Bureau of Oil and Gas Management (BOGM)
U.S. Environmental Protection Agency, Region in
Pennsylvania Fish Commission
U.S. Forest Service
U.S. Bureau of Land Management.
The Bureau of Oil and Gas Management (BOGM) was created to coordinate-arid
combine all regulatory activities of the oil and gas industry (Oil and Gas Operators'
Manual). The U.S. Environmental Protection Agency issues permits for underground
injection and secondary recovery. The Pennsylvania Fish Commission identifies pollution of
surface waters and takes appropriate action under the Pennsylvania Fish and Boat Code.
The BOGM does keep records of any accidental releases; however, routine emission
rates are not reported. Nearly all of Pennsylvania's H2S problems have occurred in the
northern part of the State, around Lake Erie.
The enforcement, field monitoring, and inspection departments of the BOGM employ
38 people. The Department of Environmental Resources has the following enforcement
options available when compliance is not met: notice of violation, citation for summary
offense, misdemeanor, civil penalty, injunction, administrative order, consent order and
agreement, permit suspension and/or revocation, and bond forfeiture.
Six wells near Lake Erie have significant concentrations of H2S that could be a threat
to the surrounding environment and people. One incident in 1990 involved discharges of
H2S from a well blowout. Local authorities evacuated a neighboring town until the H2S
could be contained and the well plugged. The blowout did not cause any negative health
effects or other types of injury.
In the pastjTennsylvania explored the possibility of establishing a committee that
would include consultants, gubernatorial appointees, and citizens to examine H2S in relation
to the oil and gas industry and determine if a serious problem exists. It is understood that
this project is currently inactive due to budget limitations.
IV-24
-------
Wyoming
There are four agencies that regulate oil and gas activities in Wyoming:
• Wyoming Oil and Gas Conservation Commission
• Wyoming Department of Environmental Quality
• U.S. Bureau of Land Management
• U.S. Environmental Protection Agency.
The Oil and Gas Conservation Commission has the general authority over oil and gas
production in the State. The Department of Environmental Quality is responsible for land
applications of all types of exploration and production wastes. The Bureau of Land
Management is responsible for all drilling and production on Federal lands.
The Wyoming Oil and Gas Conservation Commission does keep emissions inventories
on accidental releases of H2S. Any accidental release of the gas must be reported to the
Commission immediately.
The enforcement, field monitoring, and inspection departments of the Oil and Gas
Conservation Commission employ ten people. The Commission has the following
enforcement options when compliance is not met: civil assessments, permits denial and
revocations, and bond forfeiture.
In 1989, approximately 2,982 stripper wells in Wyoming produced over 5 million
barrels of oil. In recent years, there have been no signs of noncompliance; however, there
have been emergencies involving accidental H2S releases.
FEDERAL REGULATORY PROGRAMS
Current Federal regulations potentially applicable to the oil and gas production
industry's handling of hydrogen sulfide are summarized below. These include regulations of
the Occupational Safety and Health Administration (OSHA), Bureau of Land Management,
(BLM), U.S. Geological Survey, (USGS), Superfund Amendments and Reauthorization Act
(SARA) Title m, the Clean Air Act, and others. Although the OSHA standards are
applicable only to workers, they are analyzed as guidelines for reducing exposure to H2S
from both accidental releases and routine emissions.
OSHA Regulations
Currently, hydrogen sulfide emissions from oil and gas exploration and drilling are
not directly addressed by OSHA regulations. The regulations that are in effect to protect
workers are: OSHA Standards for General Industry (29 CFR Part 1910.1000), and the
respirator standards (29 CFR Part 1910.134) and the OSHA Process Safety Management
Standards (listed in Chapter m). Industries in which hydrogen sulfide occurs in quantities in
excess of 1500 pounds are covered in the Process Safety Management of Highly Hazardous
IV-25-
-------
Chemicals Standard (29 CFR 1910.119), but retail facilities and remote, unmanned
operations are exempted. Oil and gas well drilling or servicing operations are also
exempted. The potential exists that oil and gas operations that are the focus of this Report to
Congress may be exempt from this OSHA standard if the facility is remotely located or if
servicing operations include those studied in this Report. Table IV-6 lists current and
proposed regulations pertaining to hydrogen sulfide.
Current Regulations
General Industry Standards (29 CFR 1910.1000).. Acceptable concentrations for
chemical exposure are listed in Section 1910 under Table Z-l-A., Limits for Air
Contaminants, of the General Industry Standard (1910.1000). Effective December 31? 1992,
the permissible exposure limit (PEL) time weighted average (TWA) for H2S is 1 x 104 ppb)
(14 mg/m3). That is, an 8-hour time weighted average, such .that an employee's exposure to
hydrogen sulfide in any 8-hour workshift of a 40-hour workweek, shall not exceed 1 x 104
ppb. Also for hydrogen sulfide, the short-term exposure limit (STEL) is 1.5 x 104 ppb (21
mg/m3). The 1.5 x 104 ppb STEL is the employee's 15-minute (time weighted average)
exposure, which shall not be exceeded at any time during the workday. The basis for the
STEL is eye irritation.
The transitional OSHA standard, whose levels have been in effect since 1966, are
ceiling limits and are listed in Table Z-2 of the OSHA standard. The acceptable ceiling
concentration for hydrogen sulfide is 2 x 104 ppb, with an acceptable maximum peak above
the ceiling concentration of 5 x 104 ppb lasting no more than 10 minutes, and occurring only
once in an 8-hour shift, if no other measurable exposure occurs. The definition of a ceiling
is the employee's exposure that shall not be exceeded during any part of the workday. If
instantaneous monitoring is not feasible, then the ceiling shall be assessed as a 15-minute
•time weighted average exposure that shall not be exceeded at any time over a working day.
Respirator Standards (29 CFR 1910.134). The OSHA Personal Protective Equipment
Standard (29 CFR 1910.134) outlines the types of personal protective devices (respirators)
that should be worn when the ambient concentration exceeds the standards. Specific rules
pertaining to hydrogen sulfide are not included in the standard. Covered in the standard are
rules requiring written standard operating procedures, and employee training and screening
for ability to use the equipment. Respirator selection, use, inspection and maintenance,
storage, and cleaning are covered in the standard, as is air quality in supplied air respirators.
Process Safety Management of Highly Hazardous Chemicals (29 CFR 1910.119).
The CAAA instructed OSHA (in section 304), hi coordination with EPA, to promulgate a
chemical process safety standard to prevent accidental releases of chemicals that could pose a"
threat to employees. This standard was finalized in February 1992 (57 Federal Register
6356).
The OSHA requirements for employers include standards to:
IV-26
-------
Table IV-6. Summary of Occupational Exposure Standards for H>S
Agency/Association
Background
Standard or Guideline
Occupational Safety and Health
Administration (OSHA)a
General Industry Standards
29 CFR 1910.1000
Current: Lists acceptable concentrations
for chemical exposure-in the work
environment. H^S — listed under
Table Z-l-A
TWA 10 ppm 8-hour Time Weighted
Average (TWA)
STEL 15 ppm 15-minute Short Term 29
Exposure Limit (STEL)
OSHA Respirator Standards
29 CFR 19'l0.134
Current: Covers respirator selection, use,
inspection and maintenance, storage and
cleaning. Requires standard operating
procedures; employee screening and
training.
No specific rules pertaining to'
OSHA Process Safety Management
of Highly Hazardous Chemicals
Standards •
29 CFR 1910.119
Current: Remote unmanned facilities and
drilling and servicing exempted.
Purpose: To prevent or minimize the
consequences of catastrophic releases of
highly hazardous chemicals. Some
elements specified by the 1990 Clean Air
Act Amendments.
Threshold quantity for HjS: 1500
pounds; meaning that the potential
exists for a catastrophic accident at
facilities with more than 1500 pounds
on site.
OSHA Oil and Gas Well Drilling
and Servicing Standards
29 CFR 1910.270
Proposed: 1983 proposal; OSHA still
supports a specific standard for oil and gas
production, thus their exemption from
29 CFR 1910.119 above.
Specifics pertaining to HjS include:
monitoring programs, personal
protective devices, automatic flare
igniters, spark arresters, drilling mud
programs.
National Institute for Occupational
Safety and Health (NIOSH)b
Criteria Document for a
Recommended Standard for
Occupational Safety and Health
American Conference of Governmen-
tal Industrial Hygienists (ACGIH)b
.Threshold Limit Values for
Chemical Substances in the Work
Environment
Recommendations for safe levels of
worker exposure to ILS.
Standards developed for healthy workers,
not for the public at large.
Professional organization of industrial
hygienists which publishes annually
updated Threshold Limit Values (TLVs)
as guidelines in the control of ocupational
health standards.
HgS ceiling cone.: 15 mg/m3 (approx.
; 1 x 10 ppb), 10-minute sampling,
10-hour workday, 40-hour workweek.
Evacuation: 70 mg/m3 (approx
5xl04ppb)
TLV-TWA- 1 x 104 ppb, for an 8-hour
workday, 40-hour workweek.
TLV-STEL: 1.5 x 104 ppb, 15-minute
weighted average, not more than
4 times/dayday.
"Recommended standard.
enf°rceable shards; 25 of the States and territories run their own occupational safety
IV-27
-------
1) Develop and maintain written safety information identifying workplace chemical
and process hazards, equipment, and process technology;
2) Perform a process hazard analysis which shall include an estimate of workplace
effects of a range of releases and their health and safety effects on employees;
3) Consult with employees and their representatives on the conduct and development
of the process safety management program.
4) Develop and implement written operating procedures for the chemical process;
5) Provide training to employees;
6) Evaluate and monitor contractor safety standards and performance;
7) Perform pre-startup safety reviews for new and modified facilities;
8) Establish maintenance systems for critical process related equipment;
9) Establish and implement written procedures to manage changes to the process;
10) Investigate every incident that has resulted or could result in a major accident;
11) Establish and implement a plant emergency action plan.
OSHA issued its final process safety standard on February 24, 1992.
Appendix A to the process safety standard (1910.119), lists the chemicals that present
a potential for a catastrophic event with respective threshold quantities. For H2S, the
threshold quantity is 1500 pounds. This means that facilities with 1500 Ibs or greater of H2S
on-site would be subject to the process safety management standard. OSHA further requires
that the 25 States and territories with their own occupational safety organizations adopt
similar rules withitf6 months.
Although hydrogen sulfide is covered in this standard, oil and gas drilling or servicing
operations are exempted, along with retail facilities and normally unoccupied remote
facilities. OSHA explains the reason for the drilling and servicing exemptions in its
preamble to the final rulemaking (57 FR 6369), stating that "OSHA continues to believe that
oil and gas well drilling and servicing operations should be covered in a standard designed to
address the uniqueness of the industry." This exclusion is retained in the final standard since
OSHA continues to believe that a separate standard dealing with such operation is necessary.
The potential exists that oil and gas operations that are the focus of this Report to Congress
may be exempt form this OSHA standard if the facility is remotely located or if servicing
• • IV-28 -•••-• -™-- -
-------
operations include those studied in this Report.
regulations pertaining to hydrogen sulfide.
Proposed Regulations
Table IV-6 lists current and proposed
In 1983, OSHA proposed an Oil and Gas Well Drilling and Servicing Standard (48
FR 57202). The proposed standard would supplement the general standards already in effect
and address the operation's unique hazards, such as those related to the unusual equipment,
special situations dictated by the locations of operations, and hazards resulting from well
pressures. According to the Bureau of Labor Statistics, the oil and gas well drilling and
servicing industry was ranked among the most hazardous industries in the United States.
OSHA estimated that 95,000 workers at approximately 5-,4QO..rigs.-were..emplQyed in various
occupations relating to oil and gas well drilling and servicing operations. The National
Institute for Occupational Safety and Health (NIOSH) conducted a study of the oil and gas
industry 2nd provided OSHA with recommendations for developing a standard. In addition
to a discussion of the Bureau of Labor Statistics injury data, NIOSH's "Comprehensive
Safety Recommendation - Land Based Oil and Gas Well Drilling" also referenced in an early
draft a study of data NIOSH received on fatalities and injuries occurring between 1973 and
1978 in Texas and California drilling operations. NIOSH applied these statistics for the
entire drilling industry and concluded that the injury incidence and severity rates for the oil
and gas drilling industry were more than six times the rate of general industry. However,
these statistics include hazards other than H2S.
In 1973 OSHA decided to regulate this industry under its Construction Safety
Standards (29 CFR 1926); however, the applicability of this rule was contested by the
industry. As a result of the industry contention, the Occupational Safety and Health Review
Commission (OSHRC) ruled several times that the construction standards were riot
applicable. According to OSHRC, employers engaged in oil and gas well drilling and
servicing should be subject to the general industry standards found in 29 CFR 1910. New
enforcement problems emerged as a result of applying general industry standards. At the
time of the issuance of the proposed standard, OSHA data showed that the oil and gas
industry received a higher percentage of citations than any other industry. These citations
are issued only when a standard does not exist to address the hazard, but the hazard is well
recognized as a potential source of serious injury. OSHA felt that the high number of
citations indicated the need for standards directed to these hazards in order to assist
employers in meeting their obligations under the Occupational Safety and Health Act. They
stated that it was apparent that the general industry standards either did not address or
inadequately addressed hazards unique to oil and gas production, possibly even contributing
to the higher injury and illness rate experienced by this industry. With the help of data from
numerous studies of injury and illness in The oil and gas production industry, and input from
numerous states, trade associations, labor unions and industry representatives, the draft oil
and gas standards were proposed in 1983. No known action on this proposal has occurred
since then. Currently, the proposed oil and gas well drilling and servicing rule has not been
withdrawn, but it is also not on the regulatory agenda for finalising.
IV-29
-------
OSHA proposed specific requirements for drilling, servicing, and special services
operations performed in areas where a potential for exposure to H2S gas exists. The
requirements proposed establishing and implementing a monitoring program in specified
areas of the rig. The monitoring program would be applicable where the potential exists for
BjS exposure, including areas where data are unavailable or inconclusive with respect to the
potential H2S exposure. The program would use automatic environmental monitoring
systems connected to an employee alarm system. Details of the program and its procedures
would be required from the regulated community in written form. Testing and maintenance
of the monitoring system would also be regulated under the proposal, because improperly
maintained or untested systems may lead to a false sense of security for employees who rely
on them for warning.
Specific respiratory protection equipment requirements were also included in the
prpposed regulation. All employees working in an area of potential hydrogen sulfide
exposure would be required to wear or carry an approved escape-type, self-contained
breathing apparatus. An approved positive-pressure respirator would be required for
employees who remain in or return to the danger area.
*
In Appendix A to the proposed rule, OSHA also suggested the following practices to
control or limit hydrogen sulfide exposure:
• automatic igniters on the flare from the degasser, choke manifold, and mud-gas
separator to burn off hydrogen sulfide;
• spark ancestors for all internal combustion engines to lessen the chance of the
engine serving as a source of ignition in the event of a blowout;
• regular checking of drilling mud to assure it has the right constituents and pH to
counteract H2S;
• addition of hydrogen sulfide neutralizer to the drilling mud to prevent the gas
from reaching the surface;
• installation of H2S monitoring systems on all rigs working within 1000 feet of
known or suspected H2S zones.
Although the oil and gas well drilling and servicing rule (1910.270) was proposed in
1983 and has not been enacted, OSHA has continued to express a preference for a specific
regulation pertaining to the oil^and gas drilling and servicing operation in 1992, by
exempting these industries from the Process Safety Management of Highly Hazardous
Chemicals; Explosives and Blasting Agents Final Rule (29 CFR 1910.109 and 1910.119; 57
FR 6356).
TV - 30
-------
Impact of OSHA Regulations on Occupational and Human Health
OSHA regulations are designed to protect the worker rather than the general public or
the environment. In this respect, they set levels that protect the health of workers exposed
for a 40-hour workweek, rather than residents who may be exposed continuously. The
OSHA permissible exposure limit (PEL) for H2S is 10 ppb. Levels set to protect human
health in general are often much more conservative since they are often based on models
which assume exposure scenarios in which the person is exposed 24-hours a day for a
lifetime. Non-occupational health effects levels may also account for possible developmental
effects on young children and the effects of pollutants on those whose health is already
compromised due to age or a chronic condition.
Four OSHA standards have the potential to protect workers exposed to H2S. Two of
these OSHA standards could apply to both workers and the public, while the other two apply
specifically to workers. The OSHA general industry air contaminants and respirator
standards protect the worker from H2S exposures above certain levels. These standards
address the protection of the worker from an exposure in excess of a set level through the
use of personal protective equipment. The public is not protected through these two
standards, since they aim to protect workers from contact with H2S rather than prevent the
release of the H2S into the atmosphere. The process safety management standard and the
proposed oil and gas well drilling and servicing standard have the potential to protect both
the worker and the general public by preventing the release of H2S.
National Institute for Occupational Safety and Health
Recommendations for safe levels of worker environmental exposure to H2S are
presented in the May 1977, National Institute for Occupational Safety and Health (NIOSH)
Criteria Document for a Recommended Standard for Occupational Exposure to Hydrogen
Sulfide (NIOSH, 1977). Hydrogen sulfide was cited as the leading cause of sudden death in
the workplace (Ellenhorn and Barceloux, 1988). It was recognized as a serious hazard to the
health of workers employed in energy production from hydrocarbon or geothermal sources,
in the production of fibers or sheets from viscous syrup, in the production of deuterium
oxide (heavy water), in tanneries, sewers, sewage treatment and animal waste disposal, in
work below ground, fishing boats, and in chemical operations. Table IV-6 presents specific
work practices recommended by NIOSH for the gas and oil industry.
A ceiling concentration was proposed to prevent eye effects and other adverse effects,
including anorexia, nausea, weight loss, insomnia, fatigue, and headache, from prolonged
exposure to hydrogen sulfide at low concentrations. The proposed ceiling concentration
would also prevent acute eye effects, unconsciousness, and death, which can rapidly follow
exposure to hydrogen sulfide at high concentrations. NIOSH suggests no employee be
exposed to hydrogen sulfide at a ceiling concentration greater than 15 mg/m3 (approximately
1 x 104 ppb), as determined with a sampling period of 10 minutes, for up to a 10-hour work
shift in a 40-hour workweek. Evacuation of the area shall be required if the concentration of
IV - 31 _._.
-------
hydrogen sulfide equals or exceeds 70 mg/m3 (approximately 5 x 104 ppb). NIOSH warns
that the standard was not developed for the population-at-large, and any extrapolation beyond
occupational exposures is not warranted.
The document includes monitoring requirements for all areas where there is
occupational exposure to H2S. First, there should be personal monitoring to detect each
employee's ceiling exposure, with source and area monitoring as a supplement. The
monitoring should be done quarterly, or as recommended by an industrial hygienist.
Recording automatic monitors would be permitted to show short-term (less than 1-minute)
peaks of up to 5 x 10* ppb, as long as no more than one occurs in any 30-minute period.
These recording automatic monitors should be set up to signal spark-proof audible or visual
alarms. They should have different alarms to signal concentrations of 1 x-104 ppb as an alert
level to employees and 5 x 104 ppb as the level for employee evacuation.
The Secretary of Labor weighs NIOSH's recommendations, along with other
considerations such as feasibility and means of implementation, in developing regulatory
standards. The criteria document also contains sections on medical screening and followup
of exposed employees, labeling and posting of H2S hazards, personal protective equipment,
hazard information for employees, work practices, sanitation, and monitoring and
recordkeeping.
Bureau of Land Management
If a sour oil and gas well is located on Federal or Indian land, the facility operator or
owner is subject to the requirements imposed by the Onshore Oil and Gas Order No. 6
developed by the Bureau of Land Management. This order requires submittal of a public
protection plan by operators of sour oil and gas facilities upon detection of the potential to
release a hazardous volume of H2S (defined as concentrations of H2S that exceed 1 x 105
parts per billion hi the gas stream). Site-specific conditions are also criteria for determining
whether or not a facility needs to submit a public protection plan. These conditions include
(1) proximity to public buildings, public gathering centers, and roadways used for public use;
and (2)' radius and concentration of exposure. The order also has requirements for danger
signs, fencing and gates, and wind direction indicators. Additional requirements include well
control equipment, corrosion protection, and automatic safety valves or shutdowns for
accidental release prevention.
The Bureau of Land Management does have procedures for enforcing Onshore Oil
and Gas Order No. 6. Penalties for failure to comply with are cited hi
43C.F.R. 3163.1 (1992).
Minerals Management Service
The Minerals Management Service (Department of the Interior) Outer Continental
Shelf Standard, MMS-OCS-1, Safety Requirements for Drilling Operations in a H2S
-------
Environment is the name for the former U.S. Geological Survey Outer Continental Shelf
(OCS) Standard No.l. In February of 1976, the Conservation Division of the U. S.
Geological Survey (USGS) released offshore rules for safety and pollution prevention in
Standard No. 1, Safety Requirements for Drilling Operations in a Hydrogen Sulfide
Environment (USGS, 1976). Required details of a contingency plan for emergency hydrogen
sulfide situations are listed in the standard, and each platform is required to have the plan
developed prior to drilling. The standard also specifies details of the personnel training
program, and type, storage location and use of personnel protective equipment. Finally, the
standard requires state-of-the-art equipment for blowout prevention, and specifies details of
the mud program, well-testing procedures and flare system.
; The standard requires H2S monitoring-equipment at.all -wells, except when drilling in _
areas known to be free of hydrogen sulfide. Upon encountering hydrogen sulfide, the safety
requirements of the rules go into effect, and when concentrations reach 2x10* ppb the
remainder of the rules dealing with hydrogen sulfide's corrosive effects must be observed.
The precautions in the American Petroleum Institute Recommended Practice for Safe Drilling
of Wells Containing Hydrogen Sulfide, (API RP 49) are considered supplemental to the
requirements of the standard (API, 1987).
Two separate operational conditions are outlined with requirements for warning flags
and notification of authorities. Moderate danger, when the threshold limit value of 10 ppm
is reached, requires the display of signs and flags reading "DANGER - HYDROGEN
SULFIDE - H2S." If the concentration reaches 2 x 104 ppb, protective-breathing apparatus is
required to be worn by all working personnel, and non-working personnel are required to
evacuate to safe briefing areas. Extreme danger, when H2S reaches the injurious level (5 x
104 ppb), is the point when all personnel (or all non-working personnel as appropriate) are
required to evacuate. Radio communications are required to alert all known air and water
craft in the immediate vicinity of the danger.
The Minerals Management Service is in the process of reproposing its standards for
hydrogen sulfide.
CERCLA and EPCRA
The Comprehensive Environmental Response, Compensation, and Liability Act
(CERCLA) of 1980 establishes broad Federal authority to deal with releases or threatened
releases of hazardous substances from vessels and facilities. The Act defines a set of
hazardous substances chiefly by reference to other environmental statutes; currently there are
over 700 CERCLA hazardous substances. Commonly known as "Superfund," CERCLA
requires that the person in charge of a vessel or facility notify the National Response Center
as soon as that person has knowledge of a release of a hazardous substance in an amount
equal to or greater than the reportable quantity (RQ) for that substance. Currently, hydrogen
sulfide is listed as a CERCLA hazardous substance with a reportable quantity of 100 pounds.
IV - 33
-------
On October 17, 1986, the President signed into law the Superfund Amendments and
Reauthorization Act of 1986 (SARA), which revises and extends the authorities established
under CERCLA and other laws. The Emergency Planning and Community Right-to-Know
Act (EPCRA), enacted in 1986 as Title m of SARA, establishes new authorities for
emergency planning and preparedness, community right-to-know reporting, and toxic
chemical release reporting. It is intended to encourage and support emergency planning
efforts at the State and local levels and to provide citizens and local governments with
information concerning potential chemical hazards present in their communities. EPCRA is
organized into three subtitles (A-C), each containing a number of subsections.
Subtitle A establishes the framework for State and local emergency planning. Section
301 requires each State to establish an emergency response commission and local emergency
planning committees. Section 303 governs the development of comprehensive emergency
response plans by local emergency planning committees and provision of facility information
to the committee. Section 302 requires EPA to publish a list of extremely hazardous
substances and threshold planning quantities (TPQs) for such substances. This list was
established by EPA to identify chemical substances that could cause serious irreversible
health effects from accidental releases. The list includes hydrogen sulfide, with a threshold
planning quantity of 500 pounds. Any facility where an extremely hazardous substance is
present in an amount in excess of the threshold planning quantity is required to notify the
State commission and be included in local planning efforts. Section 304 establishes
requirements for immediate reporting of certain releases of reportable quantities of extremely
hazardous substances, and CERCLA Hazardous Substances, to the local planning committees
and State emergency response commissions. These requirements are similar to the release
reporting provisions under Section 103 of CERCLA. Section 304 also requires follow-up
reports on each release, its effects, and response actions taken.
Only those sour oil and gas wells and well-site facilities that have 500 pounds or more
of H2S present at the well facility are subject to the planning requirements. The reportable
quantity of H2S is 100 pounds. Therefore, releases into the environment at or above 100
pounds must be reported in accordance with CERCLA 103 and EPCRA 304.
Subtitle B provides the mechanism for community awareness of hazardous chemicals
present in the locality. This information is critical for effective local contingency planning.
If the owner or operator of a facility is required to prepare or have available a Material
Safety Data Sheet (MSDS) for a hazardous chemical under the Occupational Safety and
Health Act of 1970 and regulations promulgated under that Act, Section 311 requires that
owner or operator to submit MSDSs, or a list of the chemicals for which the facility is
required to have an MSDS, to the local emergency planning committees, State emergency
response commissions, and local fire departments. Under Section 312, owners and operators
of facilities that must submit an MSDS under Section 311 are also required to submit
chemical inventory information on the hazardous chemicals present at the facility. The
threshold for reporting for H2S under sections 311 and 312 is 500 pounds. Only facilities
that have more than the threshold quantity need to report under sections 311 and 312, unless
TV-34
-------
MSDS or inventory information is specifically requested by the State Emergency Response
Commission (SERC) or Local Emergency Planning Committee (LEPC). The owner or
operator must submit an inventory form containing an estimate of the maximum amount of
hazardous chemicals present at the facility during the preceding year, an estimate of the
average daily amount of hazardous chemicals at the facility, and the location of these
chemicals at the facility. Section 313 requires that certain facilities with ten or more
employees that manufacture, process, or use a "toxic chemical" in excess of a statutorily-
prescribed quantity submit annual information on the chemical and releases of the chemical
into the environment. This information must be submitted to EPA and to the appropriate
State offices annually. Hydrogen sulfide is not listed as a toxic chemical for which annual
release information is required.
Subtitle C contains general provisions concerning trade secret protection,
enforcement, citizen suits, and public availability of information.
Clean Air Act Section 112(r) - Accident Prevention
The Clean Air Act Amendments of 1990 established programs to prevent accidental
releases of extremely hazardous substances and to assure that mitigation and response
measures are in place in the event that a release does occur. Section 112(r) of the Clean Air
Act establishes the responsibility for prevention of releases of extremely hazardous
substances as the general duty of owners and operators of facilities that produce, process,
handle or store such substances. Section 112(r) also requires that EPA promulgate a list of
•at least 100 substances that could cause death, injury or serious adverse effects to human
health or the environment. Facilities with threshold quantities of the listed substances will be
required to establish risk management programs and to prepare risk management plans. The
statute requires EPA to promulgate regulations concerning risk management plans and other
aspects of accident prevention. H2S is one substance to which these requirements will apply
as mandated in the statute.
The general duty clause is intended to establish as a responsibility of the facility
owner the prevention of accidental releases and minimization of the consequences of
accidental releases which do occur. Responsibilities include the conduct of appropriate
hazard assessments and the design, operation, and maintenance of a safe facility. This means
that facilities must be equipped for release mitigation and community protection should a
release occur. The clause in the Clean Air Act Amendments refers to and is correlated with
the general duty clause contained in the Occupational Safety and Health Act administered by
OSHA. The OSHA clause was designed for situations for which there is no specific OSHA
regulation or standard. Recognition of the hazard by the owner or operator, or within an
industry, of the industry has been one standard under the OSHA general duty clause (U.S.
Senate 1989). Therefore, the general duty" clause places on the owners and operators of
facilities the responsibility to adhere to applicablejndustry codes and standards for safety,
accident prevention, and response.
IV-35
-------
The accidental release prevention list criteria include severity of acute adverse health
effects, likelihood of accidental release, and potential magnitude of human exposure. A
threshold quantity is to be established for each regulated substance to account for toxicity,
dispersibility, reactivity, volatility, combustibility, or flammability of the substance and the
amount anticipated to cause adverse health effects hi an accidental release. The list and
threshold quantities were proposed on January 19, 1993 (58 FR 5102). H2S is listed as a
toxic, and other substances present at oil and gas sites, such as methane, ethane, propane,
and other hydrocarbons, are listed as flammables. Facilities with threshold quantities of the
regulated substances will be required to prepare risk management plans (RMPs) and
implement risk management programs. The RMPs will include a summary of assessments of
offsite consequences for a range of accidental releases (including worst-case accidental
releases) and a history of accidental releases. Facilities must also describe release prevention
and emergency response programs developed under the risk management regulations as part
of the RMP process.
Clean Air Act - PSD Program
There is no NAAQS which addresses hydrogen sulfide; however, emissions of H2S
are regulated under the Prevention of Significant Air Quality Deterioration (PSD) Program.
PSD is designed to allow for industrial growth within specific air quality goals. The basic
goals of the PSD regulations are (1) to ensure that economic growth will occur in harmony
with the preservation of existing clean air resources to prevent any new nonattainment
problems; (2) to protect the public health and welfare from any adverse effect which might
occur even at air pollution levels better than the national ambient air quality standards; and
(3) to preserve, protect and enhance the air quality in areas of special national or regional
natural, recreational, scenic, or historic value, such as national parks and wilderness areas.
PSD permits are required for stationary sources located in areas designated, pursuant
to section 107 of the CAA, as attainment or unclassifiable for a criteria pollutant. Major
sources or modifications are those emitting either at least 100 tons per year or 250 tons per
year of any pollutant regulated under the CAA, depending on the source category of the PSD
listed pollutants. Major sources in nonattainment areas would be regulated under permit
requirements persuant to Part D under title I of the CAA.
The CAA has set significance levels, below which a PSD permit is not required.
Two tables set the significance values, one for defining significant emissions changes, in tons
per year; and the other for defining significant air quality changes, in /*g/m3. For hydrogen
sulfide, the applicable emissions threshold is the significant emission rate of 10 tons per year.
An exemption from the monitoring provision of the permitting regulations for hydrogen
sulfide is set as a 1-hour average concentration of 0.02 /*g/m3. Hydrogen sulfide emissions
are also counted as part of the Total Reduced Sulfur and Reduced Sulfur, both having
significance values set at 10 tons per year. These pollutant classes are regulated primarily to
avoid nuisance (odor) problems.
IV-36
-------
The applicability of the PSD permit program to oil and gas extraction wells would be
dependent on the amount of emissions and the grouping of the wells (i.e., whether several
wells would be combined for calculation of emissions). In general, it appears that most oil
and gas extraction wells would not likely be subject to PSD regulations based on the
applicability criteria.
INDUSTRY-RECOMMENDED SAFETY AND ENVIRONMENTAL PROTECTION
PROCEDURES
This section summarizes selected industry standards and practices for managing H2S
releases to the atmosphere. The American Petroleum Institute (API) has developed and
published "design, construction,-and operating standards.. Certain., aspects, of these standards .;..
pertaining to accidental release prevention were discussed in the previous chapter.
API Recommended Practices
The American Petroleum Institute (API), an industry-wide technical organization, has
published several recommended practices (RP) pertaining to hydrogen sulfide in the oil and
gas production industry. These voluntary guidelines are intended to maintain worker and
public safety and health. Table IV-7 lists API Recommended Practices pertinent to
production and operations In formations containing H2S.
Control Standards
API RP 49, Recommended Practices for Safe Drilling of Wells Containing Hydrogen
Sulfide (April 15, 1987) and API RP 55, Recommended Practices for Conducting Oil and
Gas Production Operations Involving Hydrogen Sulfide (October 1981; reissued March,
1983; and preparation of a second edition began in 1990) are the two main documents '
dealing with H2S in oil and gas production. It is expected that the revised RP 55 will
provide information similar in scope to that in the document currently under revision, but
with additional detail and more current references. These recommended practices do'not set
a control level for H2S emissions; rather they identify situations to which the practices apply
They are applicable in oil and gas operations where the potential exists for atmospheric
concentrations of H2S to reach 2 x 104 ppb. They also apply "where the fluids handled
contain sufficient H2S to produce a partial pressure above 0.05 pounds per square inch
absolute (psia) and the total pressure is 65 psia or greater, or where internal or external
stresses are present which could result in pipe or equipment failure due to sulfide stress
cracking and/or hydrogen embrittlement" (API, 1987). In these cases, materials must meet
National Association of Corrosion Engineers (NACE) standards.
Control Techniques
The control techniques discussed in the API Recommended Practices take two
approaches to worker and public safety. First, when hydrogen sulfide has already been
__. IV.-37
-------
Table IV-7. Reviewed American Petroleum Institute* Documents Pertaining to I^S in Oil and Gas
Production
Document
Date
Title
Topics Covered
Recommended
Practice 49
(RP49)
Recommended
Practice 51
(RPS1)
Recommended
Practice S3
(RP53)
Recommended
Practice 54
(RP54)
Recommended
Practice 55
(RP5S)
Specification 6A
(SPEC 6A)
2nd Edition
April 15, 1987
1st Edition
October 1974
Reissued
May 1982
2nd Edition
May 25, 1984
2nd Edition
May 1, 1992
1st Edition
October 1981
Reissued
March 19S3
(revision
in progress)
16th Edition
October 1, 1989
Recommended Practices
for Safe Drilling of Wells
Containing Hydrogen Sulfide
_ .API Recommended Onshore-
Production Operating
Practices for Protection of the
Environment
Recommended Practice
for Blowout Prevention
Equipment Systems for
• Drilling Wells
Recommended Practices for
Occupational Safety for
Oil and Gas Well Drilling
and Servicing Operations
Recommended Practices for
Conducting Oil and Gas
Production Operations
Involving Hydrogen Sulfide .
Specification for Wellhead
and Christmas Tree
Equipment, Supplement
land 2
Personnel training and protective equipment.
Locations. Rig and well equipment. Rig operations
in HjS environments. Contingency planning and
emergency procedures. Properties and effects of
H^ and SO2. Sour environment definition.
Producing wells. Lease roads, gathering systems
and pipelines. Production and water Handling
facilities. Oil discharge - prevention and cleanup.
Arrangement (surface and subsea) and/or
installation of: blowout preventers, choke and kill
units and lines, closing units, auxilliary equipment,
pipe stripping, marine riser systems. Inspection
and testing. Sealing components. Blowout
modifications for HjS environments.
Injuries and first aid. Protective equipment.
Fire prevention. Drilling and well servicing rig
equipment and electrical systems. Wireline
service. Stripping and snubbing. Drill stem
testing. Operations (including HgS environment).
Personnel training and protective equipment.
Contingency plans and emergency procedures.
Design, construction, and operating procedures.
Surveillance and maintenance. Continuous HgS
monitoring equipment. Supplementary guidance
and reference material for H«,S operations.
Design and performance. Materials. Welding.
Quality control. Equipment marking, shipping,
storing, and specific requirements.
American Petroleum Institute; 1220 L Street, Northwest; Washington, DC 20005.
IV-38
-------
released, worker and public safety is protected through the use of monitoring programs,
personal protective devices and contingency plans for evacuations. Second, the engineering
approach uses design, construction, and operating procedures to prevent the release of
hydrogen sulfide to the atmosphere. The prevention of equipment damage due to corrosion
(sulfide stress cracking) and the techniques for prevention of blowouts hi API RP 53,
Recommended Practice for Blowout Prevention Equipment Systems for Drilling Wells, are
two main considerations in this more site- specific engineering control technique.
API RP 49, which deals with drilling in a hydrogen sulfide environment, contains the
following recommendations for well siting in order to protect workers from the effects of
hydrogen sulfide accumulation at the well site: "Rig components should be arranged on a
location such that prevailing winds blow across-the rig ma direction, that will disperse any
vented gas from the areas of the wellhead, choke manifold, flare stack or line, mud/gas
separator, drilling fluid tanks, reserve pits, shale shaker, and degasser away from any
potential ignition source (i.e., engines, generators, compressors, crew quarters, etc.) and
areas used for personnel assembly. All equipment should be located and spaced to take
advantage of prevailing winds and to provide for good air movement to eliminate as many
sources of potential gas accumulation as possible" (API, 1987).
Other siting recommendations in API RP 49, shown in Figure m-4, are the use of
caution signs at entrance and exit roads to warn of hydrogen sulfide concentrations above
2 x 104 ppb and danger flags to warn of extreme danger when the concentration exceeds
5 x 104 ppb. These signs are required to stay in place when flaring of the hydrogen sulfide
could produce sulfur dioxide concentrations in excess of 5 x 103 ppb. Protection or briefing
centers should be placed upwind or perpendicular to the prevailing wind, with wind direction
indicators easily visible from the briefing location and all work locations. Mechanical
ventilation, large fans or bug blowers, should be available for use during light wind
conditions to prevent the hydrogen sulfide from accumulating in' low lying locations. The
locations of drilling fluid systems, power plants, burn pits, and flare stacks are also discussed
from the vantage point of worker safety after the release of hydrogen sulfide.
Both API RP 49 (pertaining to drilling in a hydrogen sulfide environment) and API
RP 55 (dealing with production operations) contain recommendations for personnel training.
RP 55 training program topics include: the effects upon humans of various concentrations of
hydrogen sulfide; protective equipment, including the use of self contained breathing
apparatus rather than canister type gas masks (a filtering type mask is not appropriate for
protection from hydrogen sulfide); monitoring devices; emergency procedures; material
selection; and the importance of ventilation. Monitoring equipment that would set off a
visual alarm at 1 x 104 ppb and an audible one at 2 x 104 ppb is recommended. Breathing
equipment requirements are also discussed, including selection and storage (where they are
readily available in an emergency).
Contingency plans are outlined in Section 4 of API RP 55 (API, 1983). They are
recommended for each operation that has the potential for an accidental release capable of
IV r.39,.
-------
exposing the public to hazardous concentrations of hydrogen sulfide. Contingency plans
should include the locations of: equipment that contains hydrogen sulfide, residences and
other public facilities, evacuation routes, safety equipment, telephones, and designated
briefing areas for employees. The contingency plan should also include procedures for
calculating the dispersion of releases and lists of emergency telephone numbers. Finally, it
is suggested that public and local officials should be briefed about the potential hazard prior
to an incident, and that periodic tests of the contingency plan should be conducted.
RP 55 also covers protection of workers from the toxic effects of hydrogen sulfide
due to build-up of gas concentration in confined areas. Protective equipment or purging is
recommended for vessels that have previously held hydrogen sulfide. Extreme caution
should be used when-entering buildings containing equipment used to handle fluids containing
hazardous concentrations of hydrogen sulfide. Routine use of personal protective devices is
suggested hi these instances.
API RP 54, Recommended Practices for Occupational Safety for Oil and Gas Well
Drilling and Servicing Operations (May 1, 1992) also addresses some aspects of personal
protection from the toxic effects of hydrogen sulfide (API, 1992). This document was
released after OSHA's implementation of the 1 x 104 ppb time-weighted average standard.
RP 54 does not mention any specific standard or level, rather it refers the reader back to API
RP 49 and API RP 55, which state that they apply to oil and gas operations where the
potential exists for atmospheric concentrations to reach 2 x 104 ppb (the old OSHA ceiling
standard), or where the gas could cause corrosion of the equipment. API does caution
throughout their documents that the latest local, State and Federal regulations should be
consulted.
Engineering controls used to prevent the production of, or the release of, hydrogen
sulfide to the atmosphere are covered in the recommended practices for drilling and
production (RP 49 and RP 55). API RP 55, pertaining to production, warns of the potential
for introducing sulfur-reducing bacteria, which produce hydrogen sulfide, into a formation
during pressure maintenance or water flooding operations (i.e., enhanced oil recovery).
Operators are warned to be aware of the possibility and to act quickly if introduction occurs.
If care is taken to prevent the bacteria from being introduced into formations that do not
contain hydrogen sulfide, the danger of hydrogen sulfide pollution will be prevented.
Other engineering controls such as those used in design, construction, and operating
practices are covered hi Section 5 of RP 55. API recommends that construction materials
meet specifications of the National Association of Corrosion Engineers (NACE) Standard
MR-01-75: Material Requirements for Sulfide Stress Cracking Resistant Metallic Material for
Oil Field Equipment. These materials include all those that are exposed to fluids containing
hydrogen sulfide and critical to its containment. Process factors for consideration are
discussed, including the concentration of hydrogen sulfide, the maximum atmospheric
temperatures expected, pressure, pH, water content of fluids, mechanical stresses,
corrosional or scale effects on the system, and any others unique to each situation. Finally,
rv-40:, ..:..
-------
piping design should eliminate dead or slow-flow areas where fluids containing hydrogen
sulfide gas can collect.
Drilling fluids are important to the control of the drilling environment. According to
API RP 49 (Recommended Practices for Safe Drilling of Wells Containing Hydrogen
Sulfide), the following practices help to maintain environmental control: maintenance of a
pH of 10 or higher to neutralize hydrogen sulfide (failing to maintain proper pH can cause
release of hydrogen sulfide from the drilling fluid system), the use of chemical sulfide
scavengers, and the use of oil-based drilling fluids. When hydrogen sulfide gas is breaking
out of drilling fluids, the fluids should be routed through a mud-gas separator until the level
is reduced to a safe one. Corrosion inhibitors that create a film which protects the equipment
from pitting and eventual sulfide stress cracking are-also. recommended.,.Finally, extreme
caution is urged hi storing fluids that have been exposed to hydrogen sulfide, and in entering
enclosed areas where drilling fluids have been stored.
Drill stem, casing, tubing, and wellhead selection must meet specifications of API,
NACE, the American Society of Mechanical Engineers, and the American National Standards
Institute, detailed in Section 5 of RP 49. Section 5 also covers procedures for limited entry
tests and equipment considerations for blowout preventer units, closing units, remote choke
control lines, and kill lines. Hydrogen sulfide considerations in mud/gas separators,
degassers and flare system are also discussed.
Abandonment procedures are included in API RP 55, with the disclaimer that the
suggested procedures do not supersede local, State or Federal regulations. Section 6.5
discusses spontaneous combustion of iron sulfide, which is produced by the reaction of H2S
with steel. Because spontaneous combustion is possible when iron sulfide is exposed to air,
RP 55 suggests that iron sulfide be kept wet until it can be burned or buried. Iron sulfide '
also poses a hazard during well servicing operations. Acids react with the iron sulfide to
produce H2S. Damage may also occur in pipes exposed alternately to hydrogen sulfide and
air. API stresses the use of monitoring equipment when well servicing operations are
performed on wells where a hydrogen sulfide hazard exists.
Hydrogen sulfide in oil and gas production is also mentioned in API RP 51, API
Recommended Onshore Production Operating Practices for Protection of the Environment
(October 1974, reissued May 1982). General information on the protection of personnel and
equipment are presented in this document (API, 1982).
FINDINGS
1.
Eighteen States have short-term H2S ambient air quality standards. Four of the nine
major oil and gas producing States reviewed in this report do not have ambient air
standards.
IV - 41:
-------
2.
3.
4.
5.
6.
7.
8.
9.
10.
11.
Ambient air quality standards range from 160 ppb per 24 hr average time to 50 ppb
per 0.5 hr average time.
The number of State agencies involved in controlling oil and gas operations varies
widely.
The size of enforcement staffs at the State level varies greatly, with some staff having
inspection responsibility beyond oil and gas operations.
No specific H2S environmental (i.e., ecological) protection standards were found for
Texas, Michigan, Oklahoma and California.
Not all States maintain notification requirements for accidental releases of H2S from
oil and gas wells. Some do require notification when a threatening accidental release
occur.
Reporting of routine H2S emissions is not required in Texas, Oklahoma, Michigan, or
California. "Routine" excludes such incidents as vapor recovery unit failures and
other equipment upsets.
NIOSH suggests no employee be exposed to H2S at a ceiling concentration greater
than 15 mg/m3 (about 1 x 104 ppb) for up to a 10 hr work shift in a 40 hr work week.
Evacuation is required if the concentration equals or exceeds 70 mg/m3 (SxlO4 ppb).
NIOSH requires monitoring in work areas with alarms sounding at 1 x 104 ppb and
5 x 104 ppb.
The Minerals Management Service requires for offshore rigs drilling in an H2S
environment: contingency plan, personnel training, state-of-the-art blowout prevention
equipment, monitoring equipment and response procedures at 1 x 104, 2 x 104, and
5 x 104 ppb. Special mud programs, well-testing procedures, and flare systems are
also required. This Federal regulatory program does not have an equivalent onshore
program.
The PSD permit program applies to significant emissions of H2S from new sources
emitting greater than 250 tons per year (or 100 tons per year for certain source
categories) of any regulated pollutant, i.e., major PSD sources. It also applies to
modifications of existing facilities if the net emissions increase of H2S from the
modification is significant. In either case, the significant emission rate for H2S is 10
tons per year. Also, permits do not require monitoring if the 1-hr average
concentration is below 0.014 ppb (0.02 Atg/m3). H2S is also regulated under the PSD
program for its nuisance odor as part of a larger group of Total Reduced Sulfur and
Reduced Sulfur (significant > 10 tons/yr).
IV --42--
-------
12. Accidental releases of H2S can be prevented by application of process safety
management principles. The following are among the ways that these principles are
adopted:
a. Under the Clean Air Act, as amended, industry has a responsibility to
identify hazards, take the actions necessary to prevent chemical accidents,
and to take action to mitigate accidents in the event that they do occur.
b. OSHA has promulgated a process safety management standard that requires
facilities to implement process safety management programs for chemicals
including H2S to protect workers from accidents. These same measures can
also prevent chemical accidents that might affect the public. However, the
OSHA requirements do not apply to remote or unstaffed facilities such as
most oil and gas well sites.
c. Under the Clean Air Act, as amended, EPA must promulgate rules that
require facilities handling H2S to implement a risk management plan designed
to prevent chemical accidents that adversely affect the public.
d. The Bureau of Land Management's Onshore Oil and Gas Order No. 6
addresses the prevention of accidental releases of H2S on Federal or Indian
lands.
e. Several State programs address the prevention of accidental releases of H2S.
States with such programs include Oklahoma, Texas, Michigan, California,
and New Mexico.
f. Voluntary industry initiatives (e.g., codes, standards, recommended
practices) such as the API RP 55, Recommended Practices for Conducting
Oil and Gas Operations Involving H2S, which is currently being revised,
have been implemented by many facilities.
13. A number of Federal and State requirements exist for emergency planning in the
event that an accidental release of H2S occurs.
a. Facilities handling quantities of H2S greater than threshold amounts are
subject to the emergency planning requirements of the Emergency Planning
and Community Right-to-Know Act (EPCRA).
b. The accidental release prevention provisions of the Clean Air Act
Amendments will require facilities handling amounts of H2S above threshold
quantities to implement an emergency response program.
c. For Federal and Indian lands, the Bureau of Land Management requires
public protection plans for sour oil and gas production operations that meet
certain criteria.
d. Several States require contingency plans in the event of accidental H2S
releases. State requirements include those of Oklahoma, Texas, Michigan,
California, and New Mexico.
IV-43
-------
e. API RP 55 recommends that contingency plans be developed for oil and gas
extraction facilities where an accidental release of H2S could be immediately
hazardous to life or health.
API. 1982. API Recommended Practice 51 (RP 51), Recommended Onshore Production
Operating Practices for Protection of the Environment, October 1974, Reissued May
41982, 1st ed., Publication No. RP51. American Petroleum Institute.
API. 1983. API Recommended Practice 55 (RP 55), Recommended Practices for
Conducting Oil and Gas Production Operations Involving Hydrogen Sulfide, October
' 1981, Reissued March 1983, 1st ed., Publication No. RP55. American Petroleum
Institute.
API. 1984. API Recommended Practice 53 (RP 53), Recommended Practice for Blowout
Prevention Equipment Systems for Drilling Wells, 2nd ed., Publication No. RP53.
American Petroleum Institute.
API. 1987. API Recommended Practice 49 (RP 49), Recommended Practices for Safe
Drilling of Wells Containing Hydrogen Sulfide, 2nd ed., Publication No. RP49.
American Petroleum Institute.
API. 1989. API Specification 6A (SPEC 6A), Specification for Wellhead and Christmas
Tree Equipment, with Supplement 1 and 2, 16th ed., Publication No. SPEC6A.
American Petroleum Institute.
API. 1992. API Recommended Practice 54 (RP 54), Recommended Practices for
Occupational Safety for Oil and Gas Well Drilling and Servicing Operations, 2nd ed.,
Publication No. RP54. American Petroleum Institute.
California Air Resources Board. 1991. Multicounty Air Pollution Control District Map.
Sacramento, California.
CDC. 1991. California Laws for Conservation of Petroleum and Gas. Publication No,
PRC01. California Department of Conservation.
Dosch, M.W., and Hodgson, S.F. 1986. Drilling and Operating Oil, Gas, and Geothermal
Wells in an H^S Environment, Publication No. M10. California Department of
Conservation, Division of Oil and Gas, Sacramento, CA.
EUenhora, M.J., and Barceloux, D.G. 1988. Medical Toxicology. Elsevier Science
Publishing Co., New York, NY.
IV--44--- -.
-------
Guidelines for Petroleum Emergency Field Situations in the State of Oklahoma. 1983.
Recommended to the Oklahoma Petroleum Industries by the Industry Advisory
Committee and the Oklahoma Corporation Commission - Oil and Gas Corporation
Conservation Division. October 1, 1983.
IOGCC. 1990. The Interstate Oil and Gas Compact Committee Bulletin, Volume IV,
Number 2. The Oil and Gas Compact Commission, Dallas, TX.
Michigan's Oil and Gas Regulations - Act 61 (P.A. 1939 as amended and promulgated
rules - Circular No. 15, revised in 1987). Michigan Department of Natural
Resources.
NIOSH. 1977. NIOSH Criteria for a Recommended Standard.... Occupational Exposure to
Hydrogen Sulftde, Publication No. 77-158. U.S. Department of Health, Education,
and Welfare, National Institute for Occupational Safety and Health, Cincinnati, OH,
2, 147.
OCC. 1986. The Corporation Commission of the State of Oklahoma. General Rules and
Regulations of the Oil and Gas Conservation Division.
Oil and Gas Operators' Manual. Pennsylvania.
Petroleum Independent. 1992. The Oil and Natural Gas Producing Industry in Your State,
1992-1993. Petroleum Independent, Vol. 62. No. 7. Independent Petroleum
Association of America.
Statewide Order Governing the Drilling for and Producing of Oil and Gas in the State of
Louisiana.
U.S. Geologic Survey. 1976. Outer Continental Shelf Standard No. 1, Safety Requirements
for Drilling in a Hydrogen Sulftde Environment, GSS-OCS-1.
U.S. Senate. 1989. Report on The Clean Air Act Amendments of 1989, S-1630. U.S.
Government Printing Office.
IV-45
-------
-------
CHAPTER V
RECOMMENDATIONS
ROUTINE EMISSIONS
From the limited data available, there appears to be no evidence that a significant
threat to public health or the environment exists from routine H2S emissions from sour oil
and gas extraction. States and industry are encouraged to evaluate existing design,
construction, and operation principles within the framework of process safety management.
EPA recommends no further legislation pertaining to 'routine ET2S emissions from oil and gas
extraction at this time.
ACCIDENTAL RELEASES
General
The EPA recommends no further legislative action pertaining to accidental H2S
releases from oil and gas extraction activities at this time. The regulations already
promulgated, and being developed, under the authorities provided to EPA in CERCLA,
EPCRA, and the accidental release prevention provisions of the CAA, provide a good '
framework for the prevention of accidental releases and preparedness in the event that they
occur.
• EPA should track implementation of current and future industry standards and
recommended practices at sour oil and gas extraction facilities. An example of such
industry standards is the American Petroleum Institute Recommended Practices for
Conducting Oil and Gas Production Operations Involving Hydrogen Sulfide (API
RP55). EPA should consider outreach specifically directed at non-participating
sectors.
• The EPA should participate in the investigation of any accidental releases associated
with H2S that cause or have the potential to cause public impacts in order to
determine the root cause of such accidents. Such investigations should be coordinated
with the Occupational Safety and Health Administration (OSHA) in order to
encompass worker safety issues.
• The EPA should continue to investigate the need for additional rulemaking under the
accidental release prevention provisions of the Clean Air Act to require
implementation of certain prevention, detection, monitoring and mitigation efforts at
facilities where extremely hazardous substances (such as H2S) could generate denSe
gas clouds and impact the public.. The level of voluntary industry initiatives and
degree of participation, and accident history should be taken into account.
V-l
-------
Facility and Local Emergency Planning Committee (LEPC)
Facilities that handle hazardous substances that could form dense vapor clouds if
accidentally released, such as H2S, should work closely with their LEPC to prevent accidents
and to be prepared to respond to such accidents.
• Facilities should identify and thoroughly understand the hazards and conditions that
can lead to accidental releases and the potential impacts on the public. These hazards
and potential impacts should be communicated to the LEPC.
• All sour oil and gas extraction facilities and the LEPC for that area should conduct
drills and exercises with workers, the community, first responders and others to .test
mitigation, response, and medical treatment for a simulated major H2S accident. All
such facilities should have training programs in place for H2S emergencies.
Preparedness and Response
All sour oil and gas extraction facilities should actively conduct outreach efforts to
ensure that the community is aware of the hazards of HjS, that protective measures are in
place to prevent public health impacts, and that proper actions will be taken during an
emergency. Such outreach should be conducted through the LEPCs.
• All sour oil and gas extraction facilities should able to rapidly detect, mitigate, and
respond to accidental releases in order to minimize the consequences. Site-specific
risk factors should be taken into account.
• Because a general duty exists to design, operate, and maintain a safe facility, owners
and operators of sour oil and gas facilities should use appropriate equipment for the
facility to provide public safety and should implement a program to remedy the
,, effects of wear and tear and corrosion on equipment.
• In addition to regular inspection of ail equipment, owners and operators should pay
particular attention to corrosion monitoring of existing flow and gathering lines and to
the condition of temporarily abandoned equipment. Remedial action should be taken
before accidental releases occur.
• EPA should foster the development and continued refinement of release detection and
mitigation systems for hazardous substances, such as H2S, in order to improve their
reliability and effectiveness.
•
• All facilities that handle oil and gas with potentially harmful levels of HjS should
have proper medical treatment supplies and trained personnel available and should
ensure that first responders, hospitals, and clinics in the area are prepared to treat H2S
exposure.
V-2
-------
Research and Further Studies
• Further study on the acute exposure levels of H2S that result in irreversible health
effects or lethality hi humans should be continued in order to improve emergency
planning tools such as atmospheric dispersion models.
• Further research on the effects of surface roughness and obstacles on dense-gas
dispersion behavior should be continued to determine their influences on toxic
substance concentrations in a dispersing vapor cloud. The Liquefied Gaseous Fuels
Spill Test Facility could be used for spill tests to assist in this research.
• EPA should continue to study the issues surrounding worst-case releases, their .......
consequences, and the likelihood of worst-case or other significant releases for
extremely hazardous substances and the role and relationship of these issues to
prevention, preparedness, and response.
-V - 3
-------
-------
GLOSSARY
Abandon: To cease producing oil or gas from a well when it becomes unprofitable. A wildcat may
be abandoned after it has been proven nonproductive. Usually, before a well is abandoned, some of
the casing is removed and salvaged and one or more cement plugs are placed in the borehole to
prevent migration of fluids between the various formations. In many States, wells may not be
abandoned unless approved by an official regulatory agency.
Accidental Release: The unanticipated emissions of a regulated substance or other extremely
hazardous substance into the air from a stationary source.
Acid: Any chemical compound, one element of which is hydrogen, that dissociates in solution to
produce free-hydrogen ions. For example, hydrochloric acid, HC1, dissociates in water to produce
hydrogen ions, H+, and chloride ions, Cl'.
Additive: A substance or compound added in small amounts to a larger volume of another substance
to change some characteristic of the latter. In the oil industry, additives are used in lubricating oil,
fuel, drilling mud, and cement for cementing casing.
Air drilling: A method of rotary drilling that uses compressed air as its circulation medium. This
method of removing cuttings from the wellbore is as efficient or more efficient than the traditional
methods using water or drilling mud; in addition, the rate of penetration is increased considerably
when air drilling is used. However, a principal problem in air drilling is the penetration of
formations containing water, since the entry of water into the system reduces its efficiency.
Alkalinity: The combining power of a base, or alkali, as measured by the number of equivalents of
an acid with which it reacts to form a salt.
Annular injection: Long-term disposal of wastes between the outer wall of the drill stem or tubing
and the inner wall of the casing or open hole.
Annulus or annular space: The space around a pipe in a wellbore, the outer wall of which may be
the wall of either the borehole or the casing.
API: The American Petroleum Institute. Founded in 1920, this national oil trade organization is the
leading standardizing organization on oil-field drilling and production equipment. It maintains
departments of transportation, refining, and marketing in Washington, D.C., and a department of
production in Dallas.
Artificial lift: Any method used to raise oil to the surface through a well after reservoir pressure has
declined to the point at which the well no longer produces by means of natural energy. Artificial lift
may also be used during primary recovery if the initial^eservoir pressure is inadequate to bring the
hydrocarbons to the surface. Sucker-rod pumps, hydraulic pumps, submersible pumps, and gas lift
are the most common methods of artificial lift.
Barrel (bbl): A measure of volume for petroleum products. One barrel (1 bbl) is equivalent to 42
U.S. gallons or 158.97 liters. One cubic meter (1 m3) equals 6.2897 bbl.
.. „. .... G-l .......
-------
Basin: A synclinal structure in the subsurface, formerly the bed of an ancient sea. Because it is
composed of sedimentary rock and its contours provide traps for petroleum, a basin is a good
prospect for exploration. For example, the Permian Basin in West Texas is a major oil producing
area.
Bit: The cutting or boring element used in drilling oil and gas wells. Most bits used in rotary
drilling are roller-cone bits. The bit consists of the cutting element and the circulating element. The
circulating element permits the passage of drilling fluid and uses the hydraulic force of the fluid
stream to improve drilling rates. In rotary drilling, several drill collars are joined to the bottom end
of the drill-pipe column for added weight. The bit is attached to the end of the drill collar.
Slowdown: The emptying or depressurizing of a material from a vessel. The material thus
discarded.
Blowexut preventer (BOP): Equipment installed at the wellhead, at surface level on land rigs and on
the seafloor of floating offshore rigs, to prevent the escape of pressure either in the annular space
between the casing and drill pipe or in an open hole during drilling and completion operations.
Blow out: To suddenly expel oil-well fluids from the borehole with great velocity. To expel a
portion of water and steam from a boiler to limit its concentration of minerals.
Borehole: The wellbore; the hole made by drilling or boring.
Casing: Steel pipe placed in an oil or gas well as drilling progresses to prevent the wall of the well
from caving in during drilling and to provide a means of extracting petroleum if the well is
productive.
Casing string: Casing is manufactured in lengths of about 30 ft, each length or joint being joined to
another as casing is run in a well. The entire length of all the joints of casing is called the casing
string.
Cement: A powder consisting of alumina, silica, lime, and other substances which hardens when
mixed with water. Extensively used in the oil industry to bond casing to the walls of the wellbore.
Cement plug: A portion of cement placed at some point in the wellbore to seal it.
Christmas tree: Assembly of fittings and valves at the tip of the casing of an oil well that controls
the flow of oil from the well.
Close-in: A well capable of producing oil or gas, but temporarily not producing.
Collar: A coupling device used to join two lengths of pipe. A combination collar has left-hand
threads in one end and right-hand threads in the other. A drill collar.
Commercial production: Oil and gas output of sufficient quantity to justify keeping a well in
production.
---• - = ••--G-2 "
-------
Completion fluid: A special drilling mud used when a well is being completed.. It is selected not
only for its ability to control formation pressure, but also for its properties that minimize formation
damage.
Completion operations: Work performed in an oil or gas well after the well has been drilled to the
point at which the production string of casing is to be set. This work includes setting the casing,
perforating, artificial stimulation, production testing, and equipping the well for production. It is
done prior to the commencement of the actual production of oil or gas in paying quantities, or in the
case of an injection or service well, prior to when the well is plugged and abandoned.
Corrosion: A complex chemical or electrochemical process by which metal is destroyed through
reaction with its environment. Rust is an example of corrosion.
Crude oil: Unrefined liquid petroleum. It ranges in gravity from 9° to 55° API and in color from
yellow to black, and it may have a paraffin, asphalt, or mixed base. If a crude oil, or crude, contains
a sizable amount of sulfur or sulfur compounds, it is called a sour crude; if it has little or no sulfur, it
is called a sweet crude. In addition, crude oils may be referred to as heavy or light according to API
gravity, the lighter oils having the higher gravities.
A
Cuttings: The fragments of rock dislodged by the bit and brought to the surface in the drilling mud.
Washed and dried samples of the cuttings are analyzed by geologists to obtain information about the
formations drilled.
Demulsify: To resolve an emulsion, especially of water and oil, into its components.
Desander: A centrifugal device used to remove fine particles of sand from drilling fluid to prevent
abrasion of the pumps. A desander usually operates on the principle of a fast-moving stream of fluid
being put into a whirling motion inside a cone-shaped vessel.
Desilter: A centrifugal device, similar to a desander, used to remove very fine particles, or silt, from
drilling fluid to keep the amount of solids in the fluid to the lowest possible level. The lower the
solids content of the mud, the faster the rate of penetration.
Disposal well: A well into which salt water is pumped; usually part of a saltwater-disposal system.
Drill: To bore a hole in the earth, usually to find and remove subsurface formation fluids such as oil
and gas.
Drill collar: A heavy, thick-walled tube, usually steel, used between the drill pipe and the bit in the
drill stem to weight the bit in order to improve its performance.
Drill cutting: The formation rock fragments that are created by the drill bit during the drilling
process. . *
Drilling fluid: The circulating fluid (mud) used in the rotary drilling of wells to clean and condition
the hole and to counterbalance formation pressure. A water-based drilling fluid is the conventional
drilling mud in which water is the continuous phase and the suspended medium for solids, whether or
not oil is present. An oil-based drilling fluid has diesel, crude, or some other oil as its continuous
-------
phase with water as the dispersed phase. Drilling fluids are circulated down the drill pipe and back
up the hole between the drill pipe and the walls of the hole, usually to a surface pit. Drilling fluids
are used to lubricate the drill bit, to lift cuttings, to seal off porous zones, and to prevent blowouts.
There are two basic drilling media: muds (liquid) and gases. Each medium comprises a number of
general types. The type of drilling fluid may be further broken down into numerous specific
formulations.
Drill pipe: The heavy seamless tubing used to rotate the bit and circulate the drilling fluid. Joints of
pipe 30 ft long are coupled together by means of tool joints.
Drill site: The location of a drilling rig.
Drill stem: The entire "length of tubular pipes, composed of the kelly, the drill pipe, and drill collars,
that make up the drilling assembly from the surface to the bottom of the hole.
Drill string: The column, or string, of drill pipe, not including the drill collars or kelly. Often,
however, the term is loosely applied to include both the drill pipe and drill collars.
Emulsion: A mixture in which one liquid, termed the dispersed phase, is uniformly distributed
(usually as minute globules) in another liquid, called the continuous phase or dispersion medium. In
an oil-water emulsion, the oil is the dispersed phase and the water the dispersion medium; in a water-
oil emulsion the reverse holds. A typical product of oil wells, water-oil emulsion also is used as a
drilling fluid.
Embrittlement: Through chemical reactions with H2S, steel and other materials become more brittle
and more likely to break.
Emulsion breaker: A system, device, or process used for breaking down an emulsion and rendering
it into two or more easily separated compounds (like water and oil). Emulsion breakers may be (1)
devices to heat the emulsion, thus achieving separation by lowering the viscosity of the emulsion and
allowing the water to settle out; (2) chemical compounds, which destroy or weaken the film around
each globule of water, thus uniting all the drops; (3) mechanical devices such as settling tanks and
wash tanks; or (4) electrostatic treaters, which use an electric field to cause coalescence of the water
globules. This is also called electric dehydration.
Enhanced oil recovery (EOR): A method or methods applied to depleted reservoirs to make them
productive once again. After an oil well has reached depletion, a certain amount of oil remains in the
reservoir, which enhanced recovery is targeted to produce. EOR can encompass secondary and
tertiary production.
EPA: United States Environmental Protection Agency.
• ^^
Exploration: The search for reservoirs of oil and gas, including aerial and geophysical surveys,
geological studies, core testing, and the drilling of wildcats.
Extraction: The physical removal of oil and gas from a well.
G-4
-------
Field: A geographical area in which a number of oil or gas weils produce from a continuous
reservoir. A field may refer to surface area only or to underground productive formations as well.
In a single field, there may be several separate reservoirs at varying depths.
Flare: Combustion of wastegases, such as H2S or natural gas, which are not able to be profitably
brought to market.
Flowing well: A well that produces oil or gas without any means of artificial lift.
Formation: A bed or deposit composed throughout of substantially the same kinds of rock; a
lithologic unit. Each different formation is given a name, frequently as a result of the study of the
formation outcrop at the surface and sometimes based on fossils found in the formation.
Gas plant: A plant for the processing of natural gas, by other than solely mechanical means, for the
extraction of natural gas liquids, and/or the fractionation of the liquids into natural gas liquid products
such as ethane, butane, propane, and natural gasoline.
Heater-treater: A vessel that heats an emulsion and removes water and gas from the oil to raise it to
a quality acceptable for pipeline transmission. A heater-treater is a combination of a heater, free-
water knockout, and oil and gas separator.
Hydrocarbons: Organic compounds of hydrogen and carbon, whose densities, boiling points, and
freezing points increase as their molecular weights increase. Although composed of only two
elements, hydrocarbons exist in a variety of compounds because of the strong affinity of the carbon
atom for other atoms and for itself. The smallest molecules of hydrocarbons are gaseous; the largest
are solid.
Ignitability (RCRA): The hazardous characteristic of ignitability for purposes of RCRA is defined in
40 CFR 261.21 and is generally a liquid with a flash point less than 140 degrees F., a non-liquid that
causes fire under a friction condition, an ignitable compressed gas, or is an oxidizer.
Inhibitor: An additive used to retard undesirable chemical action in a product. It is added in small
quantities to gasolines to prevent oxidation and gum formation, to lubricating oils to stop color
change, and to corrosive environments to decrease corrosive action.
Injection well: A well in which fluids have been injected into an underground stratum to increase
reservoir pressure.
Kelly: A pipe attached to the top of a drill string and turned during drilling. It transmits twisting
torque from the rotary machinery to the drill string and ultimately to the bit.
(median lethal concentration): The concentration of a chemical required to cause death in
50% of the exposed population when exposed for a specified time period, and observed for a
specified period of time after exposure. Refers to inhalation exposure concentration in the context of
ajr toxics (may refer to water concentration for tests of aquatic organisms or systems).
G-5
-------
Lease: A legal document executed between a landowner (or a lessor) and a company or individual,
as lessee, that grants the right to exploit the premises for minerals or other products. The area where
production wells, stock tanks, separators, and production equipment are located.
Lowest-observed-adverse-effect level (LOAEL): The lowest dose or exposure level of a chemical in
a study at which there is a statistically or biologically significant increase in the frequency or severity
of an adverse effect in the exposed population as compared with an appropriate, unexposed control
group.
Mud: The liquid circulated through the wellbore during rotary drilling and workover operations. In
addition to its function of bringing cuttings to the surface, drilling mud cools and lubricates the bit
and drill stem, protects against blowouts by holding back subsurface pressures, and deposits a mud
cake on the wall of the borehole to prevent loss of fluids to the formation. Although it originally was
a suspension of earth solids (especially clays) in water, the mud used in modern drilling operations is
a more complex, three-phase mixture of liquids, reactive solids, and inert solids. The liquid phase
may be fresh water, diesel oil, or crude oil and may contain one or more conditioners.
Natural gas: Naturally occurring mixture of hydrocarbon and non-hydrocarbon gases found in
geologic formations beneath the earth's surface. The principal hydrocarbon constituent is methane.
No-observed-adverse-effect level (NOAEL). The highest experimental dose at which there is no
statistically or biologically significant increases in frequency or severity of adverse health effects, as
seen in the exposed population compared with an appropriate, unexposed population. Effects may be
produced at this level, but they are not considered to be adverse.
Odor perception threshold: The lowest concentration at which a substance is first able to be
smelled.
OH- base muds: A drilling fluid that is a water-oil emulsion with oil as the continuous phase. The
oil content ranges from 50-98% oil. Oil muds are used to reduce drilling torque and to stabilize
reactive shales that impede the drilling process.
Oil and gas separator: An item of production equipment used to separate the liquid components of
the well stream from the gaseous elements. Separators are vertical or horizontal and are cylindrical
or spherical in shape. Separation is accomplished principally by gravity, the heavier liquids falling to
the bottom and the gas rising to the top. A float valve or other liquid-level control regulates the level
of oil in the bottom of the separator.
Oil field: The surface area overlying an oil reservoir or reservoirs. Commonly, the term includes
not only the surface area but also the reservoir, wells, and production equipment.
Operator: The person or company, either proprietor or lessee, actually operating an oil well or
lease.
Packer: A piece of downhole equipment, consisting of a sealing device, a holding or setting device,
and an inside passage for fluids. It is used to block the flow of fluids through the annular space
between the tubing and the wall of the wellbore by sealing off the space. The packer is usually made
up in the tubing string some distance above the producing zone. A sealing element expands to
: . . . ,, . ....G-6 - .1., , . ,.,..„.- - „
-------
prevent fluid flow except through the inside bore of the packer and into the tubing. Packers are
classified according to configuration, use, and method of setting and whether or not they are
retrievable (i.e., whether they can be removed when necessary, or whether they must be milled or
drilled out and thus destroyed).
Perforate: To pierce the casing wall and cement to provide holes through which formation fluids
may enter, or to provide holes in the casing so that materials may be introduced into the annulus
between the casing and the wall of the borehole. Perforating is accomplished by lowering into the
well a perforating gun, or perforator, that fires electrically detonated bullets or shaped charges from
the surface.
Permeability: A measure of the ease with which fluids can flow through a porous rock.
pH: A measure of the acidity or alkalinity of a solution, numerically equal to 7 for neutral solutions,
increasing with increasing alkalinity and decreasing with increasing acidity.
Primary recovery: Oil production in which only existing natural energy sources in the reservoir
provide for movement of the well fluids to the wellbore.
Produced water: The water (brine) brought up from the hydrocarbon-bearing strata during the
extraction of oil and gas. It can include formation water, injection water, and any chemicals added
downhole or during the oil/water separation process.
Producing zone: The zone or formation from which oil or gas is produced.
Production: The phase of the petroleum industry that deals with bringing the well fluids to the
surface and separating them. Production also includes storing, gauging, and otherwise preparing the
product for the pipeline.
Production casing: The last string, of casing or liner that is set in a well, inside of which is usually
suspended the tubing string.
RCRA (Resource Conservation and Recovery Act): The Federal statute enacted in 1976 (and
subsequent amendments) which amended the Solid Waste Disposal Act. Among other things, RCRA
and its amendments established and/or augmented three significant programs: the hazardous waste
management program, the solid waste program, and the underground storage tank program.
Reference concentration (RfC): An estimate (with uncertainty spanning perhaps an order of
magnitude) of a daily inhalation exposure of the human population (including sensitive subgroups) that
is likely to be without an appreciable risk of deleterious effects during a lifetime.
Reservoir: A subsurface, porous, permeable rock body in which oil or gas or both are stored. Most
reservoir rocks are limestones, dolomites, sandstones, or a combination of these. The three basic
types of hydrocarbon reservoirs are oil, gas, and condensate. An oil reservoir generally contains
three fluids-gas, oil, and water-with oil the dominant product. In the typical oil reservoir, these
fluids occur in different phases because of the variance in their gravities. Gas, the lightest,'occupies
the upper part of the reservoir rocks; water occupies, the lower part; and oil occupies, the
intermediate section. In addition to occurring as a cap or in solution, gas may accumulate
. - - . - - - - G-7
-------
independently of the oil; if so, the reservoir is called a gas reservoir; Associated with the gas, in
most instances, are salt water and some oil. In a condensate reservoir, the hydrocarbons may exist as
a gas, but when brought to the surface, some of the heavier ones condense to a liquid or condensate.
At the surface the hydrocarbons from a condensate reservoir consist of gas and a high gravity crude
(i.e., the condensate). Condensate wells are sometimes called gas-condensate reservoirs).
Rig: The derrick, drawworks, and attendant surface equipment of a drilling or workover unit.
Routine emissions: The anticipated emissions of a regulated substance or other extremely hazardous
substance into the air from a stationary source during its normal operation.
Secondary recovery: Any method by which an essentially depleted reservoir is restored to producing
status by the injection of liquids or gases (from extraneous sources) into the wellbore. This injection
effects a restoration of reservoir energy, which moves the formerly unrecoverable secondary reserves
through the reservoir to the wellbore.
Shale shaker: A series of trays with sieves that vibrate to remove cuttings from the circulating fluid
in rotary drilling operations. The size of the openings in the sieve is carefully selected to match the
size of the solids in the drilling fluid and the anticipated size of cuttings. It is also called a shaker.
Short-term exposure limit (STEL): A time-weighted average that the American Conference of
Government and Industrial Hygienists (ACGIH) indicates should not be exceeded any time during the
work day. Exposures at the STEL should not be longer than 15 minutes and should not be repeated
more than 4 times per day. There should be at least 60 minutes between successive exposure at the
STEL.
Shut-in well: A non-producing well with its pump turned off, and the stuffing box closed, which has
been inspected to ensure there is no leakage.
Sour: Containing hydrogen sulfide or caused by hydrogen sulfide or another sulfur compound.
Stripper well: A well nearing depletion that produces a very small amount of oil or gas.
Tail gas: gas that leaves a sulfur recovery process after most of the H2S has been converted to SO2.
Tank battery: A group of production tanks located in the field, used-for storage of crude oil.
Tertiary recovery: A recovery method used to remove additional hydrocarbons after secondary
recovery methods have been applied to a reservoir. Sometimes more hydrocarbons can be removed
by injecting liquids or gases (usually different from those used in secondary recovery and applied with
different techniques) into the reservoir.
Threshold limit value (TLV): The concentration of a substance below which no adverse health
effects are expected to occur for workers, assuming exposure for 8 hours per day, 40 hours per week.
TLVs are published by the American Conference of Governmental Hygienists (ACGIH). This listing
may be useful in identifying substances used in the workplace and having the potential to be emitted
into the ambient air.
G-8
-------
Time-weighted average (TWA): An approach to calculating the average exposure over a specified
time period.
Tubing: Small-diameter pipe that is run into a well to serve as a conduit for the passage of oil and
gas to the surface.
Uncertainty factor (UF): One of several, generally 10-fold factors, applied to a NOAEL or a
LOAEL to derive a reference dose (RfD) from experimental data. UFs are intended to account for
(a) the variation in the sensitivity among the members of the human population; (b) the uncertainty in-
extrapolating animal data to humans; (c) the uncertainty in extrapolating from data obtained in a less-
than-Hfetime exposure study to chronic exposure; and (d) the uncertainty in using a LOAEL rather
than a NOAEL for estimating the threshold region.
Volatile: Readily vaporized.
Waterflood: A method of secondary recovery in which water is injected into a reservoir to remove
additional quantities of oil that have been left behind after primary recovery. Usually, a waterflood
involves the injection of water through wells specially set up for water injection and the removal of
the water and oil from the wells drilled-adjacent to the injection wells.
Wellbore: A borehole; the hole drilled by the bit. A wellbore may have casing in it or may be open
(i.e., uncased); or a portion of it may be cased and a portion of it may be open.
'&
Well completion: The activities and methods necessary to prepare a well for the production of oil
and gas; the method by which a flow line for hydrocarbons is established between the reservoir and
the surface. The method of well completion used by the operator depends on the individual
characteristics of the producing formation or formations. These techniques include open-hole
completions, conventional perforated completions, sand-exclusion completions, tubingless
completions, multiple completions, and miniaturized completions.
Wellhead: The equipment used to maintain surface control of a well, including the casinghead,
tubing head, and Christmas tree.
Workover: One of more of a variety of remedial operations performed on a producing oil well to try
to increase production. Some examples of workover operations are deepening, plugging back, pulling
and resetting the liner, and squeeze-cementing.
Workover fluids: A special drilling mud used to keep a well under control when it is being worked
over. A workover fluid is compounded carefully so it will not cause formation damage.
G-9
-------
-------
APPENDIX A
BACKGROUND INFORMATION ON THE OIL AND
GAS PRODUCTION INDUSTRY
-------
-------
APPENDIX A
BACKGROUND INFORMATION ON THE OIL AND GAS PRODUCTION
INDUSTRY
EXPLORATION AND DEVELOPMENT
Although geological and geophysical studies provide information about potential
accumulations of petroleum, only exploratory drilling can confirm the presence of petroleum.
Rotary drilling, the primary .drilling method in the United States, provides a safe way
to control high-pressure oil/gas/water flows and .allow for. the. simultaneous drilling of the
well and removal of cuttings. This makes it possible to drill wells over 30,000 feet deep
Figure A-l illustrates the process. Most rotary drilling operations employ a circulation
system using a water- or oil-based fluid, called "mud" because of its appearance. The mud
is pumped down the hollow drill pipe and across the face of the bit to provide lubrication and
remove cuttings. Cuttings are removed at the surface by shale shakers, desanders, and
desilters; they are then deposited in the reserve pit excavated or constructed next to the* rig.
Air drilling, which is considerably faster and less expensive than drilling with water- or oil-
based fluids, is used in areas where high pressure or water-bearing formations are not
anticipated.
Potential producing zones are normally measured and analyzed during exploratory
drilling. If evidence, of hydrocarbons is found, a drill stem test can show whether
commercial quantities of oil and gas are present. If so, the well is prepared for production.
This is called "completion." The most common method is the "cased hole" completion.
Production casing is run into the hole and cemented permanently in place. Then one or more
strings of production tubing are set in the hole, productive intervals are isolated with
packers, and surface equipment is installed. The well is not actually completed until a gun
or explosive charge perforates the production casing and begins the flow of petroleum into
the well (U.S. EPA, 1987). Figure. A-2 shows a cross section of a common well.
While a well is being drilled, heavy fittings have to be installed at the surface where
the casing is attached, as each string of casing is inserted into the hole. Each part of the
casing head is supported by a part of the casing head which was installed at the top of the
next larger string of casing when it was run (U.S. EPA, 1987).
HOW OIL AND GAS ARE PRODUCED
Production operations generally include all activities associated with the recovery of
oil and gas from geologic formations. They can be divided into activities associated with
downhole operations and activities associated with surface operations. Downhole operations
include primary secondary, and tertiary recovery methods, well workovers, and well
stimulation activities. Activities associated with surface operations include oil/gas/water
..A-l
-------
A Five-Sheave Crown Block
B Four-Sheave Traveling Block
C Hook
D Swivel
E Kelly
F Standpipe and Rotary Hose
G Derrick Floor
H Rotary Machine
J Rotary Machine Drive from
. Main Transmission
K Vibrating Mudscreen
L Outlet for Drilling Fluid
M Drawworka
N Main Engines
O Main Transmission (engines to
drawworks and to pump)
P Suction Tank
Q Pump
Source: Royal Dutch/Shell, 1983.
Figure A-l. Rotary drilling rig.
-A-2
-------
Pumping Unit
Oil and Gas
is - Gas
Surface Casing ».
Tubing
Plunger
Pump
Traveling Valve
Standing Valve
Gas
Oil and Some Gas
Perforated Top
Suction Pipe
Packer
Liner
,'7>i>7
«%•*•«•.
Producing Formation
Source: Royal Dutch/Shell, 1983.
Figure A-2. Cross section of a well pumping installation.
-------
separation, fluid treatment, and disposal of produced water. The term "extraction" is
commonly used to refer to activities associated with getting oil or gas to the surface;
production includes both extraction and the surface operations involved in processing the
materials extracted from the well. Production, as discussed in this report, is limited to the
processing and storage that occurs at the well site. Transportation and further processing is
not included in the scope of this report.
Downhole Operations
The initial production of oil or gas from the reservoir is called primary recovery.
Natural pressure or artificial lift methods (surface or subsurface pumps and gas lifts) are used
to bring the gas or oil out of the formation and to the surface (see Figure A-3). High-
pressure gas can also be injected to lift the oil from the reservoir.
During the primary recovery stage, natural pressure in the reservoir may decline and
artificial lift may be needed. One of three general types of pumps may be used: (1) pumps
at the bottom of the hole run by a string of rods; (2) pumps at the bottom of the hole run by
high-pressure liquids; and (3) bottom-hole centrifugal pumps (API, 1976).
The pumping unit includes a complete set of surface equipment that imparts an up-
and-down motion to the sucker-rod string, which is connected the bottom-hole pump. Figure
A-2 shows the parts of such as unit. Deep wells often require the long-stroke pumping
provided by hydraulic units.
A stuffing box is used in a pumping well to pack or seal off the pressure inside the
tubing so that liquid and gas cannot leak outside the polished rod. A stuffing box consists of
flexible material or packing housed in a box which provides a method of compressing the
packing. The packing material gradually wears out and must be replaced before it loses its
effectiveness as a seal (API, 1976).
Primary recovery methods alone can produce oil and gas from most reservoirs, but
over the life of the well production gradually decreases. Some form of secondary recovery
will eventually be needed in nearly all wells. Secondary recovery methods inject gas or
liquid into the reservoir to maintain pressure. The most frequent method is waterflooding,
which involves injecting treated water (seawater, fresh water or produced water) into the '
formation through a separate well.
When secondary recovery methods are no longer adequate, the last portion of the oil
that can be economically produced is recovered by tertiary methods. These include
chemical, physical, and thermal methods or some combination. Chemical methods involve
injection of fluids containing substances such as surfactants and polymers. Miscible oil
recovery methods inject gases such as carbon dioxide and natural gas that combine with the
oil. Thermal recovery methods include steam injection and in situ combustion (or "fire
flooding"). The injected gases or fluids from secondary and tertiary recovery operations are
"A-4.
-------
1 Prime Mover or Power Plant
2 Gear Reducer
3 Crank and Counter Weight
4 Pitman
5 Walking Beam
9 Horse Head
7 Counter Weight
8 Sampson Post
9 Bridle
10 Carrier Bar
11 Polished Rod Clamp
12 Polished Rod
13 Stuffing Box
14 Tee
IS Tubing Ring
16 Casing Head
17 Casing Strings
18 Tubing String
19 Sucker Rod
20 Fluid Level
21 Rod Pump
Source: API, 1976.
FignrerA-3. Main parts of a pumping unit.
A.-5-
-------
dissolved or mixed with the oil produced by the well and must be removed during surface
production operations (U.S. EPA, 1987).
Workovers are another type of downhole production operation. Workovers are used
to restore or increase production when downhole mechanical failures or blockages, such as
sand or paraffin deposits have inhibited the flow of a well. Fluids circulated into the well for
a workover must be compatible with the formation and must not adversely affect
permeability. The workover fluid may be reclaimed or disposed of when the well is put back
into production. Workover fluids are similar to completion fluids, which are special fluids
used when the well is completed (ready for the production phase), to minimize formation
damage and control potential problems such as H2S corrosion.
Other chemicals are used periodically or continuously to inhibit corrosion, reduce
friction, or simply keep the well flowing (U.S. EPA, 1987).
Surface Operations
As fluids are pumped to the surface, they are collected and treated to separate the
various components (oil, gas, gas liquids, and water). Figure A-4 shows the separation
process. These surface operations become more complex as secondary and tertiary recovery
methods are employed. The ratio of water and other fluids to oil tends to increase as
producing reservoirs are depleted. In new wells little or no water may be produced. The
volume of water produced by stripper wells varies greatly. Stripper wells may produce more
than 100 barrels of water for every barrel of oil, especially if waterflooding is used as a
secondary recovery (U.S. EPA, 1987).
Separation involves the use of equipment to separate the gas, oil and water from each
other. The actual separation may be accomplished in a single step or several steps depending
on the relative amounts and the physical characteristics of the material which is delivered to
the surface. Complete separation may require several stages involving different pressures,
temperatures, and possibly additives if the material is delivered to the surface at a high
pressure and the oil and gas are present in an emulsion.
After separation, the gas is transported by pipeline to a gas processing facility if the
quantities from a specific well are adequate. If the quantities are inadequate, the gas is
flared (burned). Gas processing facilities remove inerts (N2, CO2), hydrogen sulfide (H2S),
and liquids (oil and water) to produce pipeline quality gas which has a nominal heating value
of 1000 BTU per cubic foot. Gas can also be re-injected into the well if necessary to help
manage the reservoir or the production from the well.
Oil that is recovered from the separators at the well is placed in tanks and transported
to a refinery for processing. This transportation is by pipeline if the quantities are adequate
to justify installation of a pipeline or by truck if the production is small.
A-6
-------
Dry Gas
Oil and Gas
Production Well
Gas
Defavdrator
Oil and Gas
Separator
•==^-l IP
- te
J, —
ras
• t
Oil
Water
] Meter
i=Toj
pipel
^
Water
I^J^
Produced
as Water
ine Storage
Tank -
j: — — - — u^
1 Sediment
Emergency Pit
_l
> OS
~~*" t
^R
=A:
T
b
leter
3 oil pipeline,
arge. or truck
I
>
Enhanced Recovery
or Disposal
Sediment
Reservoir
Source: U.S. EPA, 1987.
Figure A^. Topical extraction operation showing separation of oil, gas, and water
A-7
-------
Water recovered from the separators at the well is placed in tanks or pools. This
water will ultimately be reinjected into the producing formation, injected into a disposal well,
or discharged. Reinjection into the producing formation and injection into a disposal well are
the most common methods for water disposal; discharge is rarely used. Permits are usually
required for these water disposition options.
The equipment used at the surface to control the well is called the well head. If high
production or significant gas pressure is expected, the well head is usually built of cast or
forged steel, and machined to a close fit. These sealed fittings prevent well fluids from
blowing or leaking at the surface. Parts of the well head may be designed to hold pressures
up to 20,000 Ib per sq in (psi). Some well heads are just simple assemblies to support the
weight of the tubing in the well, and may not be built to hold pressure: .For stripper, wells, .;
or other low-production, low-pressure wells, a simple well head can be used as long as only
small amounts of gas are produced with the oil (API, 1976).
High pressures or corrosive gases such as H2S require well heads with special valves
and control equipment to control the flow of oil and gas from the well. These are
constructed of heavy metal and installed above the casing head or tubing head before the well
is completed. This collection of valves is called a Christmas tree because of its shape and the
large number of fittings branching out above the well head. The tree diverts fluids through
alternative chokes (API, 1976).
Safety measures should be adequate to prevent high pressure wells from going out of
control. Equipment is available that automatically shuts off production if there is damage to
the wellhead or to automatic surface safety valves at the wellhead.
Simpler types of Christmas trees can be used on low pressure or pumping wells.
Pressure gauges on the well head and Christmas tree measure the pressure in the casing and
tubing. If the pressures under various operating conditions are known, better control can be
maintained (API, 1976).
OVERVIEW OF THE INDUSTRY
The U.S. petroleum industry drilled its first oil well in 1859. Since that first well,
the oil and gas industry has grown to be extremely complex and diverse. In 1990,
approximately 869,887 wells in over 33 States were producing oil and gas in the United
States. The oil and gas obtained from these wells is found at depths ranging from 30 feet to
30,000 feet below the earth's surface. The major U.S. areas of onshore production include
the southwest (including California), the midwest, and Alaska, with lesser contributions from
the Appalachians. Table A-l lists production estimates for the oil and gas producing States.
In 1990-1991, Texas led all States in oil and natural gas production, turning out 705 million
barrels of oil and 6.3 trillion cubic feet of natural gas (Petroleum Independent, 1992).
Figure A-5 shows U.S. oil and gas production by State. The bar graph in Figure A-6 shows
distribution of States containing more than 70 percent of gas wells in the U.S. Some of these
A-8
-------
September 16,1993
DRAFT - DO NOT CITE OR QUOTE
Table A-l. 1991 CHI and 1990 Gas Production Estimates
State
Alabama
Alaska
Arizona
Arkansas
California
Colorado
Florida
Illinois
Indiana
Kansas
Kentucky
Louisiana
Maryland
Michigan
Mississippi
Missouri
Montana
Nebraska
Nevada
New Mexico
New York
North Dakota
Ohio
Oklahoma
Oregon
Pennsylvania
South Dakota
Tennessee
Texas
Utah
Virginia
West Virginia
Wyoming
Federal Waters
Other
U.S.
Number of
Producing
Oil Wells"
872
1,466
22
7,265
43,375
6,596
..:'.'.•• 83
31,874
7,506
45,470
22,741
23,812
0
4,570
2,168
854
3,854
1,440
46
18,546
4,043
3,546
30,089
95,468
0
22,338
149
736
188,829
1,972
25
15,950
11,397
4,468
25
601,520
Oil Production
(thousands of
barrels)b
18,538
647,310
122
10,387
350,900
30,857
5,674 -
19,954
3,000
55,427
5,411
390,406
0
19,675
27,033
146
19,810
5,890
4,012
67,247
416
36,716
10,008
112,274
0
2,643
1,648
508
705,460
27,604
15
2,143
103,855 •
NA
NA
2,684,687
Number of
Producing
Gas Wells"
2,038
109
NA
3,460
1,169
5,097
NA •
356
1,311
14,043
11,713
13,530
NA
1,438
629
NA
2,428
NA
NA
19,537
5,406
103
34,697
27,919
NA
30,000
52
527
48,075
742
819
37,000
2,431
3,591
147
268,367
Gas Production
(million cubic
feet)b
135,276
402,907
2 IOC
174,956
362,748
242 897
6 483 • - -
677
399
573,603
75,333
5,241,989
22
172,151
94616
7
50 429
7QQ
NA
965,104
25,023
52 169
154,619
2,258,471
2am
177,609
881
2,067
6,343 146
145,875
14,774
178,000
735,728
NA
NA
18,561,596
Combined Source: Petroleum Independent, September 1992, attributes the individual column sources to-
World Oil.
b Energy Information Administration.
-NA Not available.
-------
2.643
10.0081 177,609
18.95413.0001 154.619
"677~ "39T
Oil (thousands of barrels)
Gas (million cubic feet) '
UATbtal
2.648.687
18,561.596
Combined Source: Petroleum Independent, September 1992, attributes the oil and gas
sources to: World Oil Energy Information Administration.
Figure A-S. 1991 U.S. oil and gas production by State.
A-10
-------
LA
KY
Source: World Oil (in Petroleum Independent, September 1992)
Figure A-6. States with the most producing gas wells in 1990.
A-11
-------
LA
OK
NM WY
State
KS
AK
CA
Source: Energy Information Administration (in Petroleum Independent, September 1992).
Figure A-7. Gas production in 199O from the top producing states.
A-T2.
-------
States, however, are not the largest gas producers. Figure A-7 shows that Texas, Louisiana,
Oklahoma, New Mexico, Wyoming, Kansas, Alaska, and California account for 92 percent
of domestic gas production. Alaska, California, Louisiana, and Texas account for 78 percent
of domestic oil production.
Principal Production Industry Groups
The industry can be divided into four groups. The first group consists of the major oil
companies. These companies are highly vertically integrated, which means that they perform
both "upstream" activities (oil exploration, development and production) and "downstream"
activities (transportation, refining and marketing).
The second group is the large independents. These companies primarily explore,
develop, and produce oil and gas, but do not perform downstream activities. Some large
independents produce oil and gas only, while others provide such additional services as
contract drilling and pipeline operations.
The third group is the small independents. Little information is available that would
characterize this group quantitatively. However, small independents are known to have
fewer wells and/or lower production wells. The lower operating expenses of small
independents makes it more affordable to continue producing small quantities from low
volume wells.
The fourth group consists of companies that provide a variety of specialized services
to the oil and gas drilling rigs and platforms, such as designing, manufacturing and
installing specialized hardware. They also provide geophysical support, drilling mud and
logging services.
Diversity of Production
Production from individual wells varies greatly from a high of 11,500 barrels per day
to less than 10 barrels per day. As shown in Figure A-8, over 70 percent of U.S. oil wells
are stopper" wells. The definition of a stripper well varies from State to State However
these wells are generally defined as wells that produce 10 barrels of oil per day or less or '
100 thousand cubic feet (mcf) of gas per day or less. In 1990, 463,854 stripper wells existed
and produced a total of 383,197,000 barrels of oil (NSWA, 1991). Stripper well production
oS Jo?^o-7mnTn?e A"2' Figure A"9 shows that striPPer wells produced 14 percent of the
Z,o84,687,000 barrels of oil produced in the United States in 1990 (U.S. EIA, 1991- U S
FJA, 1987). Figure A-10 shows the proportion of stripper wells in the io States with the
largest numbers of wells overall. In all 10 States, stripper wells comprised more 50 percent
of producing wells. However, Figure A-11 demonstrates that in the 10 top oil producing
States, oil from stopper wells is relatively low in volume. These wells typically are near
depletion of recoverable natural resources and produce only a small quantity of oil or gas
A-13
-------
Strippers3
76%; 463,854 Wells
Large Producers
24%; 149,956 Wells
a Strippers are defined as those producing 10 barrels
a day or less.
Source: Interstate Oil and Gas Compact Commission and
National Stripper Well Association.
Figure A-8. Number of producing oil wells in
the U.S. in 1990.
A-14-
-------
Table A-2. 1990 Oil Production from Stripper Wells by State
Location
Alabama
Alaska
Arizona
Arkansas "
California
Colorado
Florida
Illinois
Indiana
"Kansas
Kentucky
Louisiana
Michigan
Mississippi
Missouri
Montana
Nebraska
Nevada
New Mexico
New York
North Dakota
Ohio
Oklahoma
Pennsylvania
South Dakota
Tennessee
Texas
Utah
Virginia
West Virginia
Wyoming
U.S.
Number of
Producing
Wells"
872
1,466
22
' 7,265 .-••••
43,375
6,596
83
31,874
7,506
45,470
22,741
23,812
4,570
2,168
854
3,854
1,440
46
18,546
4,043
3,546
30,089
95,468
22,338
149
736
188,829
1,972
25
15,950
11,397
601,520
Number of
Producing
Stripper0
Wells
514
0
12
7,290 —
26,128
5,234
0
33,700
5,764
45,227
19,330
17,695
3,967
615
375
3084
1,269
0
15,261
3,748
1,205
29,576
73,345
21,800
26
923
127,790
1,026
22
15,975
2,953
463,854
Percentage of
Producing Wells
Which Are
Stripper Wells*
58%
0%
55%
•-..'•NA* • • - •-•
60%
79%
0%
NA*
77%
99%
85%
74%
87%
28%
44%
80%
88%
0%
82%
93%
34%
98%
77%
98%
17%
NA*
68%
52%
88%
NA
26%
77%
Amount of
Crude Oil
Produced
(thousands
of barrels)13
18.538
647,310
122
10,387
350,900
30,857
5,674
19,954
3,000
55,427
5,411
390,406
19,675
27,033
146
19,810
5,890
4,012
67,247
416
36,716
10,008
112,274
2,643
1,648
508
705,460
27,604
15
2,143
103,855
2.684,687
Combined Source: Petroleum Independent, September 1992, attributes the individual column sources to-
a World Oil.
b Energy Information Administration.
c Interstate Oil and Gas Compact Commission and National Stripper Well Association.
*Petroleum Independent warns "[number of producing stripper wellsHata cannot be compared to "Producing Oil Wells"
table due to different sources and technology."
NA Unable to calculate.
A-15
-------
9
Table A-2. 1990 CHI Production from Stripper Wells by State (continued)
Amount of Crude Oil
Produced from
. Stripper Wells
(thousands of
barrels)0
Percentage of
Crude Oil
Produced from
Stripper Wells
1,486
0
26
5,693
36,405
5,698
0
18.520
3,002
40,873
4,338
7,154
4,599
802
120
2,449
2,011
0
14,296
383
2,053
7,271
78,599
2,622
64
419
135,850
1,035
12
2,122
5,297
389.197
-3%
0%
21%
55%
10%
19%
0%
93%
NA*
74%
80%
2%
23%
3%
82%
12%
34%
0%
21%
92%
6%
73%
70%
99%
4%
83%
19%
4%
80%
99%
5%
14%
A-16
-------
Strippers*
14%; 383,197,000 Barrels
Large Producers
86%; 2,684,687,000 Barrels
a Strippers are defined as those producing 10 barrels
a day or less.
Source: Interstate Oil and Gas Compact Commission and
National Stripper Well Association.
Figure A-9. 1990 U.S. oil production.
A-17
-------
IX OK KS CA IL OH IA KY PA NM
Large Wells
Stripper Wells (wells producing 10 barrels a day or less)
Source: World Oil (in Petroleum Independent, September 1992).
Figure A-10. States with the largest number of producing oil wells in 1990.
'A-18
-------
800
ND
CO
• Large Wells
H Stripper Wells (wells producing 10 barrels a day or less)
Source: World Oil (in Petroleum Independent. September 1992).
Figure A-ll. OH production in 1990 from the top producing state.
A-19
-------
REFERENCES
API. 1976. Primer on Oil and Gas Production. American Petroleum Institute, Committee
• on Vocational- Training and Executive Committee on Training and Development.
American Petroleum Institute, Dallas, TX.
API. 1983. Introduction to Oil and Gas Production. 4th ed. American Petroleum
Institute, Committee on Vocational Training and Executive Committee on Training
and Development. American Petroleum Institute, Dallas, TX.
IOGCC. 1990. The Interstate Oil and Gas Compact Committee Bulletin, Volume IV,
Number 2. The Oil and Gas Compact Commission, Dallas, TX.
NSWA. 1991. National Stripper Well Survey. National Stripper Well Association.
Petroleum Independent. 1992. The Oil and Natural Gas Producing Industry in Your State,
1992-1993. Petroleum Independent, Vol. 62. No. 7. Independent Petroleum
Association of America.
Royal Dutch/Shell Group of Companies. 1983. The Petroleum Handbook, 6th ed. Elsevier
Science Publishers B.V., Amsterdam, The Netherlands, 38, 52.
U.S. ELA. 1990. Natural Gas Annual. U.S. Energy Information Administration.
U.S. EIA. 1991. Petroleum Supply Annual. U.S. Energy Information Administration.
U.S. EPA. 1987. Management of Wastes from the Exploration, Development, and
Production of Crude Oil, Natural Gas, and Geothermal Energy, Volume I, EPA/530-
SW-88 003. U.S. Environmental Protection Agency,. Research Triangle Park, NC.
World Oil Magazine, Forecast Review Issue. February 1992.
A.20
-------
APPENDIX B
SUBJECTS OF STATE H2S REGULATIONS AND GUIDELINES
-------
-------
Table B-l. Subjects of State HjS Regulations and Guidelines
Regulations and Guidelines Oklahoma
Texas
Michigan
Characteristics and Effects of H,S
(including emergency rescue, resuscitators,
effects on metal and artificial respiration)
Initial Testing
Periodic Gas Analyses
Nuisance Odors
Guidelines for Safe Drilling Operations
A. Location Requirements "
B. Drilling Equipment
(Including blowout preventer, controls,
piping and accessories, etc.)
C. Monitoring Equipment
(including alarm systems and gas
detection equipment)
D. Personal Protective Equipment
(including all personnel, breathing
apparatus, equipment specs., etc.)
E. Employee Physical Requirements
F. Training Requirements
G. Drills and Orientations
H. Maintenance of Equipment
I. Warning Systems
J. Evacuation •
Guidelines for Safe Production Operations
A Applicability
B. General Provisions
1. Concentration Determination
Radius of Evacuation (ROE)
Escape Rate Volume Determinators
Storage Tank Provisions
. . . ppm ROE in excess of . . . feet
Implementation
Control and Safety Equipment
Contingency Plan
Training
Injection Provision
Certificate of Compliance Provision
Accident Notification
NA
NA
NA
NA
NA
NJ
NA
NA
NA
NA
•
•
•
NA
NA
2.
3.
4.
5.
6.
7.
8.
9.
10.
11.
12.-
NA
NA
NA
•
NA
NA
«
NA
•»
*
NA
NA
NA
NA
Not available in reviewed literature.
• The subject was identified under the States H-S regulations or guidelines
-A Rule 36 references APIRP19. - •' . - .
NJ Not under Congressional jurisdiction. • -
MMIOSH Required by Michigan OSHA.
•CA grants the supervisor of the Oil and Gas Division discretionary authofity'to control H^S releases.
-------
-------
APPENDIX C
ATMOSPHERIC DISPERSION CALCULATIONS FOR H,S RELEASES
FROM OIL AND GAS EXTRACTION FACILITIES
-------
-------
APPENDIX C
ATMOSPHERIC DISPERSION CALCULATIONS
FOR H2S RELEASES
FROM OIL AND GAS EXTRACTION FACILITIES
INTRODUCTION
The purpose of this appendix is to provide supporting details for the analyses of
atmospheric dispersion of H2S conducted for this report. In Chapter HI, computer models
were used, together with information on published studies of sour gas releases, to examine
the range of predictions of the distances of concern ;for scenarios of H2S releases from
wellheads or pipelines. The inputs to the analyses are reviewed, and the outputs of three
sample calculations for two of the scenarios are described. Outputs for a horizontal wellhead
release are described for calculations using the SLAB and SAPLUME models. The output
for a vertical wellhead release using the DEGADIS model is also described.
SUMMARY INPUT AND OUTPUT DATA
Summary data for the wellhead blowout and pipe rupture scenarios are presented in
Tables C-l and C-2, respectively. As described hi Chapter m, analyses for wellhead
blowouts were conducted using the SLAB, DEGADIS, and SAPLUME models. Analyses of
the pipe rupture scenarios were conducted using the SACRUNCH and SAPLUME models.
The wellhead blowout scenarios in Chapter m result from various assumed flow rates
as presented again in Table C-l. The following discussion presents some justification for the
choice of these flow rates. Flow rates are functions of such factors as rock permeabilities,
gas properties, depth, and tube and casing diameters. In practice, there are large variabilities
in these parameters. One measure of the potential rate of flow from a well is the Calculated
Absolute Open Flow Rate (CAOF), which is the rate of flow of gas into the well bore when
the pressure is atmospheric. A sample of 15 wells in western Wyoming had CAOFs with a
geometrical mean of 4.7xl06 standard cubic feet per day (scf/d) or l.SxlO5 cubic meters per
day (m3/d) (Layton et al., 1983). The 95 percent confidence interval spanned the range from
2.1xl05 scf/d (5.9X103 m3/d) to 108 scf/d (2.8xl06 m3/d). Alp et al. (1990) considered
CAOFs of between 5x10* and 5xl06 m3/d as representative of wells in Alberta and chose 106
m3/d as representative for the purposes of risk analysis. The Quest report (1992) considered
CAOFs in the range 2.2xl05 to 7.3xl06 m3/d for a system of wells in southwestern
Wyoming. The actual flow rates out of a ruptured well will be less than the CAOF because
of frictional effects in the pipework. By contrast, the Quest report and Layton et al. use the
CAOF as a conservative estimate of flow rate. Based on the above discussions, a flow rate
of 2x107 scf/d was chosen" for representative calculations, with a flow rate of 10s scf/d being
taken as an example of a very high flow rate.
C-l
-------
TABLE C-l
SUMMARY OF INPUT AND OUTPUT DATA
WELLHEAD BLOWOUT SCENARIOS
SCENARIO-
INPUTS
Flow rate (m'/d)
Vol. % HiS
Density* ® 0*C (kg/m3)
Release temperature (*C)
Total release rate (kg/s) • • •
Release rate of H.S (kg/s)
Ambient temperature f C)
Relative humidity (%)
Atmospheric stability category
Windspeed (m/s)
Surface roughness length (m)
Effective area of release (m,)
OUTPUTS: HORIZONTAL RELEASE
SLAB:
Distance to:
LC,,(m)
ERPO-2(m)
SAPLUME:
Distance to:
LC,, (m)
ERPO-2(m)
OUTPUTS: VERTICAL RELEASE
SLAB:
Distance to:
LC,,(m)
ERPO-2(m)
DEOADIS:
Distance to:
LC,,(m)
ERPG-2(m)
SAPLUMEi
Distance to:
LC0,(m)
ERPG-2(m)
A
6x10*
7.5
0.862
0
5.99
0.79
5
75
F
1.5
0.1
0.02
700
2,800
1,000
3,100
0
0
0
0
0
0
B C
6x10* 6xl05
27 15
1.293 1.038
0 0
8.98 . 7.21
2.85 1.58
5 5
75 75
F F
1.5 1.5
0.1 0.1
0.02 0.02
2,800 1,500
7,000 4,700
2,700 1,500
10,000 5,700
0 0
0 0
0 0
0 0
0 0
0 0
D
6xl03
30
1.128
0
7.83
3.17
5
75
F
1.5
0.1
0.02
2,900
7,000
3,000
10,000
0
0
0
0
0
0
D(E)*
SxlO6
30
1.128
0
39.2
15.8
5
75
F
1.5
0.1
0.1
7,000
> 10,000
> 10,000
> 10,000
0
0
0
0
0
0
* Scenarios from Table III-7.
* For comparison, density of air® 0°C= 1.293 kg/m3.
" E"< Extreme Case.
C-2
-------
TABLE C-2
PIPE RUPTURE SCENARIOS
INPUTS AND OUTPUTS (SADENZ MODEL)
Parameters/Scenario
Composition A1, rupture
of 4" diameter pipeline6
1 Composition from Table HI-5.
b Spacing between emergency shutdown valves is 1,000 m.
• Spacing between emergency shutdown valves is 3,000 m.
d From Figure IH-22.
Composition D*, rupture
of 16" diameter pipeline5
INPUTS
Total mass released (kg)d
Total mass of H,S (kg)
Duration of release (s)d
Density @ 0°C (kg/m3)
Release temperature <°C)
Ambient temperature (°C)
Relative humidity (%)
Atmospheric stability category
Windspeed (m/s)
Surface roughness length(m)
OUTPUTS
Distance to:
LQ, (m)
ERPG-2 (m)
640 31,000
84 12,500
16 310
0.862 1.128
0 (32 °F) 0 (32 °F)
5 (41 °F) 5 (41 °F)
75 75
F F
1.5 1.5
0.1 0.1
600 ' 4,300
750 5,600
C-3
-------
Table C-l also presents values for the effective area of release. These values are
derived by dividing the volumetric release rate by the velocity of release and were not the
bases for the release scenarios. As stated in Chapter IH, the velocity of release was assumed
to be "choked," or limited, to sonic velocity (approximately 330 m/s) as a result of the high
initial gas pressure.
The temperature of the gas in a well prior to expansion to atmospheric pressure
through a rupture depends on the depth of the gas reservoir. The amount of cooling that
results from expansion to atmospheric pressure as a result of release depends on the initial
pressure and the composition. Alp et al. (1990) assume a representative release temperature
of 15°C (288 K) at atmospheric pressure. In the Quest report, the authors assume a reservoir
temperature of 60°C and calculate expansion temperatures of between -9°C and 3°C. The .
calculated results of wellhead blowout and pipeline rupture scenarios in this study are based
on a representative release temperature of 0°C. This temperature is below the assumed
ambient temperature of 5°C.
Atmospheric conditions characterized by low turbulence and low wind speed provide
for decreased dilution of a released chemical with the surrounding air. Thus, these
conditions are directionally conservative in terms of potential exposure to accidental releases.
Atmospheric thermal stability, impacted by the difference between surface and air
temperatures, is often described by PasquUl atmospheric stability categories. These categories
range from high turbulence (A) through low turbulence (F). The "F" category is typical of
still, nighttime conditions (AIChE, 1989). This category was chosen for the calculations
conducted to conservatively evaluate the wellhead blowout and pipeline rupture scenarios.
* Wind speeds of less than 2 m/s are considered low and create little turbulence. The
calculations used in this study's analyses assume a wind speed of 1.5 m/s to conservatively
simulate nonturbulent conditions. Actual conditions of A - D stability and higher wind
speeds will cause more rapid dilution of an accidental release and will result in a decreased
affected distance. The assumption that conditions of low wind speed and stable atmospheric
conditions exist uniformly for extended distances also provides conservatism to the analyses.
Terrain is another factor that may influence atmospheric dispersion of a release. The
surface roughness length is a measure of the "roughness" of the terrain. Roughness is a
function of the type of terrain and the presence of such features as trees and buildings. The
models in this study assume that the study of the behavior of dense gas flow around obstacles
and through rough terrain is controversial and is an area where further research is needed.
Rough terrain will cause more turbulence to atmospheric flows above it than smooth terrain.
The value of surface roughness length, 0.1 m, used in the calculated dispersion predictions,
is considered to be an intermediate roughness length and typical of highly vegetated rural
terrain. It should be noted that lower, more conservative values would be more appropriate
in flat, barren terrain.
.G-4
-------
SAMPLE SLAB CALCULATIONS
SLAB Input
The following illustrates how the input is prepared for SLAB, using composition D
from Table ffl-5 as an example. The SLAB input is displayed on Table C-3. The SLAB
users' manual provides further guidance (Ermak, 1989).
Line 1: IDSPL is the spill source type. For an evaporating pool, IDSPL =1. For a
horizontal jet release IDSPL=2. For a vertical jet release IDSPL=3. For a puff,
IDSPL = 4. For the present example, the release is assumed to be horizontal, IDSPL=2.
Line 2: NCALC is a numerical substep parameter. The code developer recommends using
NCALC = 1. However, NCALC can be increased if numerical stability problems are
encountered.
Line 3: WMS is the molecular weight of the wellhead gas in kg/gmole. From Table ni-6,
it is 0.0252 kg/gmol (from 25.2 g/gmol). Note, however, that the value given in Table C-3
is 0.0289 kg/gmol, for the following reason. Initially, the dilution of the plume is dominated
by entrainment caused by its high momentum (its initial velocity equals that of sound).
There is considerable dilution in this early phase and, by the time it is over, the density of
the plume is only slightly less than that of the surrounding atmosphere. Work on marginally
buoyant plumes shows that they are not likely to lift off the ground (Briggs, 1973).
However, SLAB runs with WMS = 0.0252 kg/gmol show predicted plume rise that
continues to a height of over 100 m. This is regarded as physically unrealistic and the
computer model is "fooled" into ignoring plume rise by setting WMS equal to the effective
molecular weight of air which is 28.9 g/gmol (0.0 289 kg/gmol). As noted above, this is
thought to be physically realistic. The results -predicted in this way will be conservative if
plume rise does in fact take place.
Line 4: CPS is the vapor heat capacity at constant pressure. Similar to the above molecular
weight calculation, the gas mixture vapor heat capacity is calculated by summing the product
of the constituents' mole percent and vapor heat capacity. For composition D it is
aroximatel 1500 J/k/K.
approximately 1,500 J/kg/K.
Lme 5: TBP is the boiling point of the released material. For a pure vapor release, SLAB
does not in fact use this quantity, which has been arbitrarily set equal to the boiling point of
methane, 111.5K.
Lme 6: CMEDO is the liquid mass fraction in the initial release and is set to zero because
the release is pure vapor.
Lines 7, 8: DHE=509,880 (J/kg) and CPSL=3,349 (J/kg/K). are the heat of vaporization
and the liquid specific heat for methane. Their values are taken from Table 2 of the SLAB
C-5
-------
Users' Guide. When the released material is pure vapor, as it is in the present case, and the
temperature of the cloud does not drop below the boiling point, these values are adequate
because the liquid properties will not be used in the SLAB calculation. However, a value for
all SLAB input properties must be specified whether they are used or not.
9: RHOSL is the liquid density of the released material. This is another quantity that
is not used in the calculations. It has been set equal to the density of water (1,000 kg/m3).
Lines 10.11: SPB and SPC are parameters that go into the saturated vapor pressure formula:
Ps = Pa*exp[SPA - SPB/(T + SPC)],
where P, is the saturated vapor pressure, Pa is the ambient pressure (l.OlxlO5 N/m2), SPA is
defined in the code and T is the local cloud temperature. Table 2 of the SLAB Users' Guide
contains some values of SPB and SPC, but not for the mixture modeled here. When these
values are unknown, the Users' Guide recommends default values of SPB = -1 and
SPC = 0. The code then uses the Clapeyron equation to define the value of SPB. When the
released material is pure vapor, as it is in the present case, and the temperature of the cloud
does not drop below the boiling point, this default is adequate because the saturation pressure
will not be used in the SLAB calculation. However, a value for all SLAB input properties
must be specified whether they are used or not.
Lines 12-17: These lines specify the spill parameters. TS is the temperature of the released
material, taken to be 273K. QS is the rate of release, estimated at 20 million scfd (7.69
kg/s). AS is the effective area of the release, 1.93xlO'2 m2, obtained by dividing the
volumetric flow rate by the speed of sound (340 m/s). TSD is the duration of the release,
3,600 .s, the assumed duration of release for a wellhead blowout. QTIS is zero except when
modeling an instantaneous puff release. Finally, HS is the height of the release, arbitrarily
taken to be 5 m (close to the ground).
Line 18: TAV is the concentration averaging time. This is set equal to 3,600 to be
consistent with the exposure time of concern.
Line 19: XEFM is the maximum downwind extent of the calculation. A value of 10 km is
used in order to obtain cloud concentration results at large distances away from the release.
It is set to 2xl04 m, which should be enough to ensure that any results of interest lie within
this distance.
Lines 20-23: ZP(I) allows the user to specify up to four heights at which the concentration is
calculated as a function of downwind distance. ZP(1) is set to 1.6 m (approximate head
elevation above grade). The remaining ZP(I)s are zero, which means that SLAB only
considers the first height.
Lines 24-29: These lines specify the meteorological conditions. ZO is the surface roughness
length, which is set to 0. 1 m, depicting a relatively smooth surface. ZA is the height at
"•"-• ..... ";"!; ...... ' ................ • ........ • - •-•• -C-6 -•-• • •-- - ....... •• .....
-------
which the windspeed is measured (10 m). UA is the windspeed at height ZA (1.5 m/s). TA
is the ambient temperature (273K). RH is the relative humidity (75%, chosen as being
typical of Category F weather conditions). Finally, STAB is the stability class (F=stable).
The weather conditions (Category F with a low windspeed of 1.5 m/s) have been chosen to
simulate unfavorable (close to worst case) conditions.
Line 30: TER is the end of file designator. TER < 0 terminates the run.
SLAB Output
A partial SLAB output corresponding to the inputs of Table C-3 is given in Table
C-4. The interpretation is as follows. The first column gives the downwind distance, x.
The second column gives the time at which the maximum concentration arrives at x and the
third gives the duration of cloud passage. As can be seen, the duration of passage remains
equal to the duration of release until the cloud has traveled several kilometers downwind.
The fourth column gives the approximate half-width of the plume, bbc. The remaining six
columns give the average concentration (volume fraction) at a height of 1.6 m (as chosen in
the SLAB input) for six off-axis distances that are multiples of bbc, 0.5, 1.0, 1.5, etc. The
predicted concentrations are zero close in because the plume was arbitrarily released at a
height of 5 m. As the plume broadens, the concentrations at height 5 m rise above zero to a
maximum at about 25 m to 30 m downwind and then begin to decline as the plume dilutes
further.
The effective ERPG-2 is 100 ppm and the effective LCOI is about 4.7 x 105 ppb.
These number values are derived as follows: the ERPG-2 for pure H2S for an exposure time
of 1 hour is 3 x 104 ppb. The volume concentration of H2S in composition D is 30 percent
(see Table ffl-5). Therefore, the overall concentration of the total released material when the
H2S in it is at 3 x 104 ppb is 30/0.3 = 1 x 10s ppb. Similarly, the LC01 for pure H2S is
1.4 x 10s ppb for an exposure time of 1 hour (see Chapter HT). Therefore, the effective LC01
for the plume is 140/0.3 = 4.7 x 105 ppb. As explained in Chapter m, the ERPG-2 is
regarded as a threshold at which emergency response might be necessary and the LC01 is an
approximate threshold for the occurrence of fatalities among the affected population.
Reading down the column headed "y/bbc=0," the concentrations first fall below 4.7 x 105
ppb (= a volume fraction of 4.7x10^) at a distance of about 3 km and below 1 x 105 ppb (=
a volume fraction of 1.0x10"*) at a distance of about 7 km.
SAMPLE DEGADIS CALCULATIONS
DEGADIS Input
Table C-5 displays the DEGADIS input for the same case as was prepared for SLAB
in Table C-3 except that DEGADIS can only simulate a vertical jet release. The chosen
values for most of the parameters have already been explained in the section on SLAB.
C-7
-------
Lines 1-4 allow the user to input up to four lines of title.
LineS requires the windspeed (1.5 m/s) and the height at which the windspeed is measured
(10 m).
Line 6 gives the surface roughness length (0.1 m).
Line 7 requires the parameter INDVEL, the atmospheric stability category (F=6) and the
Monin-Obukhov length RML. For INDVEL= 1 (the present case) the model calculates RML
from the stability category and the surface roughness length, so the user does'not need to
specify a value for RML.
Line 8 requires the ambient temperature (273K), the ambient pressure (1 atmosphere) and the
relative humidity (75 %).
Line 9 gives the surface temperature, which is here set equal to the ambient temperature
(273K).
Line 10 is a name for the released gas, in this case CPD for Composition D.
Line 11 is the molecular weight, 25.2.
Line 12 is the averaging time, taken to be equal to the duration of release, 3,600 s. It is
used to calculate the increase in the effective width of the plume as a function of exposure
time.
Line 13 is the temperature of the released gas, 273K.
Line 14 contains the upper level of concern (470 ppm, expressed as a volume fraction), the
lower level of concern (100 ppm) and the height at which the concentrations are calculated
(1.6 m).
Line 15 contains first a variable INDHT=0, meaning that heat transfer from the ground is
not included, which does not matter here because the plume, air, and ground all have the
same temperature. The second entry is the specific heat of the released gas at constant
pressure (1,500 J/kg/K). The third entry, CPP=0, indicates that an approximation was made
in which the specific heat does not vary with temperature.
Line 16 is a parameter "NDEN" that is used to specify the density profile of the released
material. For NDEN=0, the release is assumed to be an ideal gas with specific heat at
constant pressure 1500J/kg/K. Water condensation effects are taken into account.
Line 17 is the mass rate of release, 7.69 kg/s.
C-8
-------
Table C-3. SLAB Input - Horizonai Wellhead Release
Value
2 (horizontal), 3 (vertical)
1
0.0289
1500.
111.50
0.0
509,880.
3,349.
1,000.
-1.0
0.0
273.
7.69
1.93xlO'2
3,600.
0.
5.
3,600.
20,000.
1.6
0.
0.
0.
0.1
10.
1.5
273.
75.
F .
-1.
Parameter
IDSPL
NCALC
WMS
CPS
TBP
CMEDO
DHE
CPSL
RHOSL
SPB
SPC
TS
QS
AS
TSD
QTIS
HS
TAV
XFFM
ZP(1)
ZP(2)
ZP(3)
ZP(4)
ZO
ZA
UA
TA
RH
STAB
TER
Line No.
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
C-9
-------
Plane
J
1 1
g. 3
* S
° I
ea u
CO
ion
I J
&•
u
I ^
ed fTAV = 3,600 s) Voume Concen
Ave
F™'«»
Concentratlon
"
2gjS22£!2o»
2222O'='222222O'-'000
xxxxxxxxxxxxxxxxx
b b b b b b b b b b b b •) od <-4 — --' p~ vi od •/•> O -< CS t^ -^ TT (N -.' §
^-^
Bd « -a
•oj
J 23
U 3
•— a
o o
a^ U ^"^
_g x 42-
Downwi
.
ooooooooooo
XXXXXXXXXXXXXXXXK
pt^j^^cncST^pnp^oovovoomo—'escM
o«r>odes—"--vd^rr^Tr^c^vdod^—1-^—4
obbbboooobbbbbbbb
xxxxxxxxxxxxxxxxx
PPPPPPPPPpppf<^~"-Oioqfnfopo\rrmoocs^rvooooS
OOOOOOOOOOOOCO — ro — — ON vo — vo'cs'iXoo'—'—'—'—;—.'
vo
oo p en vo p
vq
p
oo ON —'
X
r-.
o
c-io
nu
co
-------
I
4)
SB
II
N
C3
CJ
2
s
"o
o
S
3
O)
BO
2
V
11111222222222222221111111°^°°^
xxxxxxxxxxxxx "--" — — — — — — — — — — — — — — — —
xxxxxxxxxxxxxx
X X X X X
F22l2l2F2l2ll222222l2ll22222°"2"2"°
TT7TTTTTTTTTT7v?''>>n'n'n'n'n0=>
KXKXXX-
vo
XXX X XXX X. XX K..K..K. X X X
r*"> «-* r**i [•"••• vi vo ON NO C
ri rn es CN •—' — — —' «
S S 8 s S S
NO TT tn rn M -.•
._. - C O m o w-i in
O P-; — 0<3 t~- O u-J
«"*t^^f*^-^ootOrn
S
•
•n
o
TT^TTTT'T'r'r'nmvi
.U.Ul^l^Vl^l-TjT.^.- ._ ._
xxxxxxxxxxxxx
-• — —' O\ vo'
2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2m"0"0^^^'o o
xx x x x x x x x x x x x x x x'x'x'x""'"" — — —
en i""»«»— OXP^NO
xxxxxxxxxxxx
^rpnescs — o\vdrn~~r~-
63 n JS
13 .0
§ «
2 2 2 2 ° 2 2 2 2 2 2 2 2 2 2 22112111111111 o'isb
-------
Table C-S. Input for DEGADIS Simulation of a Vertical Wellhead Release
Value
Release from a
Well Head:
Verticle Jet Simulation
1.5 10.
0.1
1 6
273. 1.
273.
CPD
25.2
3,600.
273.
4.7X10-4 l.OxlO-4
0 1,500.
0
7.69
5.0 0.0192
3,600.
50.
Line Number
f
1
2
. . < 3
4
k,
5
6
0. 7
75. 8
9
10
11
12
13
1.6 14
0.0 15
16
17
18
19
20
C-12
-------
LjneJS contains the height of release (5 m) and the effective diameter (0.0192 m)
Line 19 is the duration of release, 3,600 s.
Line_2Q is the distance between points at which DEGADIS calculates the output.
DEGADIS Output
A partial DEGADIS output is given in Table C-6. The first column gives the
distance downwind and the second gives the elevation. As can be seen, the plume rises
substantially because of its initial momentum. The third column gives the concentration of
the released gas as a mole fraction, the fourth column gives the concentration in kg/m3 and
the fifth column gives the density in kg/m3. As can be seen, the density rapidly approaches
that of the surrounding air, 1.29 kg/m3. The fifth column gives the temperature of the
plume, which remains constant at 273K because the released plume and the air both have that
temperature. The sixth column gives the plume horizontal standard deviation a y and the
seventh column gives the vertical standard deviation, a z (the concentration across'the plume
is approximated by a Gaussian distribution in DEGADIS). As can be seen, at a height of-
1.60 m, the predicted width of the plume (the distance across the wind to the upper or lower
Levels of Concern, LC01 and ERPG-2) is zero so that LOCs are not predicted to be seen at
ground level. This is a typical result for vertical jets of sour gas in stable weather
conditions, whether DEGADIS, SLAB, or SAPLUME is used.
SAMPLE SAPLUME CALCULATIONS
SAPLUME Tnpnt
Table C-7 contains the input for the model SAPLUME corresponding to Table C-3
which contains the SLAB input for a horizontal release with composition D The first few
lines of input begin with four asterisks and are title cards, followed by a blank which tells
the code that the titles have ended. Each subsequent line or group of lines begins with a
keyword, followed by numbers in exponential notation to three significant figures.
"SITE" tells SAPLUME that there is a site with one radius and one sector (this is the
default when the model is not considering a real site). The following line gives the one
radius, arbitrarily set at 10,000 m, with one person arbitrarily set at that point (in the mode
of operation chosen for the current problem, the model ignores these numbers).
"WEATHER" specifies that one weather condition only, category F is being
considered (because the 1.000 that follows -WEATHER begins at space 151.' For E the space
would be 51, for D, 41 and so forth. The model can consider all six weather categories at
once with up to four velocity subdivisions in each.) In this case, there is one velocity
subdivision, specified as 1.5 m/s (first line after weather), and the probability that the wind
blows into the one sector is unity (second line after weather).
.,„.,.... ._.... „ ,. -,, .. , _ C-13 . . , ... .. .
-------
&>
QJ
l
3
"3
O
£2
S
w
Q
£
vo
6
ID
1
.S
"3
1
c
53
a)
"3
u
'u
eu
m
"35
Q.
TJ
C3
£
A
1
OJ
tn
es
at
•33
OS
fj
•S M ^
a?-l~
1C
• •
'1
- e •§"
B --
E *7
2 !«i
H S
M M .
II
d:
SI'
*"§
1^
r
ll
a e.
"5 —
r
a
J? -|
.
1 § ~
Ii~
2
22 222 S § S
00 OOO OOO
Sll III Hi
-
-
<=>
K
O\
•MW M c« r^ nr*oo
VI
1
,.._,
r**O S^ON * T oo *n
*
ooo ooo oob bob
tn «o 3 Wi vo «t> Mln)S *^ 5t r*
r* m me oi cs o •- — rj r4«*if*{
S5S S S 5 S S S SS'5
ooo ooo ooo ooo
«n vo r* •evim v% i.n •— ^t ao
-------
I*
11
s
III
Ill-
e •?•?•? *?•?•? «?•>*> •?•?•? •?*>•* •*•»»» ««
n „ S SSS SSS SSS SSS SSS SSS S b o bob bob bob ob'o
I 2 lj * * * * K « KMK **-' ~~ ~
is5 ss§ S§3 I§S ^§s ssa SSB ssa ss?s ss;a ~ss ?.
_ ^ ^000 000 000 000 000 000 000 000 000 000 O
f 5
00000 000 000 000000 000000 000 00 O- -0-0 O-
SI
^i1^0; "^«^"* t^°i™ ^T**! *^p
n«
-------
U
ju
J3
ill-
is-
11
il
*&
H O
O J,
•N "^
D J.
i IN d tN m
222 22
K K K K K
8 5 S 15
c* ci c* cl ci
000 00
000 O
Q £
-2
-a
K K K K
ci ri fi
000 00
J
a E
III
C-16
-------
The entries on the "PROPERTIES" line are as follows: the ambient temperature is
273K at which temperature the density of air is 1.29 kg/m3 and its specific heat at constant
pressure is 990 J/kg/K. At a temperature of 273K, the density of the released gas is 1.141
kg/m3 and the specific heat at constant pressure is 1,500 J/kg/K.
The entries on the "SOURCE" line are as follows: the rate of release is 7.69 kg/m3 at
a temperature of 273K. The amount of air initially entrained with the source is zero. The
angle of release is zero radians (horizontal). The height of release is 5 m. The initial
momentum flux is 2,540 kg m s'2 and is the product of the rate of release and the exit
velocity (the speed of sound is approximately 330 m/s).
The "INTERVAL" line specifies that SAPLUME calculations start at a,downwind
distance x of 0.1 m and that calculations are performed at a uniform spacing of 0.15 in
log,0(x).
On the "ROUG" line, the surface roughness length is 0.1 m and the windspeed is
measured at a height of 10 m.
The "HAZARD" line specifies two levels of concern. As explained above in the
discussion of the SLAB results, these are the LCm of 470 ppm (approximately 5.32X104
kg/m3) and the ERPG-2 of 100 ppm (l.HlxlO4 kg/m3).
"VGRAD" informs SAPLUME that it should consider the velocity gradient and the
temperature gradient in the atmosphere. SAPLUME uses standard textbook formulae for
these gradients. If the first entry after VGRAD were zero, velocity would be constant as a
function of height. Similarly, if the second entry after VGRAD were zero, the temperature
of the atmosphere would remain constant as height increases.
A value of 3 after "NEUT" specifies one of three parametrizations for the standard
deviations in the Gaussian model once the released material has evolved out of the jet phase
NEUT =3 corresponds to a parameterization that is appropriate for a rural site
DUR" specifies that the duration of release is one hour.
Finally, the repetition of "END" terminates the run of SAPLUME.
SAPLUME Output
A partial SAPLUME output corresponding to the input in Table C-7 is given in Table
C-8. This table indicates that, for hazard level 1 (i.e., the LCm of 470 ppm discussed above)
the plume touches down at a downwind distance of approximately 63 m and extends to about
3 km, covering an area of about 105 m2 (one tenth of a square kilometer). The table of pairs
of values of downwind distance, x, and width can be coupled to a plotting routine to give
contours of constant concentration. Similarly, hazard level 2 (the ERPG-2 of 100 ppm)
C-17
-------
Table C-7. Input for SAPLUME Runs
•••*
*•••
EPA Hydrogen Suffide Runs
January 1993
No Protective Measures
Composition D - 30% HjS at Wellhead
2xl07 SCFD:
Horizontal Release
HjS Release Rate - 3.073 kg/s
Total Mixture Release Rate - 7.69 kg/s
Hazard Level - ERPG-2 (100 ppm) and
LC(,, (470 ppm) Both Adjusted for Stream Composition
Category F Weather, Windspeed 1.5 m/s
SITE
1.000K10*
1.000
WEATHER
1.500
PROP
SOURCE
INTERVAL
ROUG
HAZARD
5320x10-*
VGRAD
NEUT
DUR
END
END
1.000 1.000
1.000
2.730X102 1.290 9.900X102 2.73QX102 1.141 LSOOxlO3
7.690 2.730X102 0.000 0.000 5.000 2^40xl03
LOOOxlO1 l.SOOOxUT1
LOOOOxlO-1 1.000x10'
2.000
1.141x10*
1.000 1.000
3.000
1.000
C-18
-------
OI
3
'C
o
j=
I
O
es
3
ju
.Q
u
'a
1
I
£
i
j
>
u
a
N
S
1
a
a
a
z
o
a
VI
OS
w
\e
o «
8
£
|
o
w
o
u
3
o
a
ee
Ed
S
i
s
<
a
z
a IT 222
•o o- c! ci o- ci eS I S S f; § J3 3
— 222 o'oo'o'c.'o'c,"^
X1sl|ssa5li3555i
2 S •§•• 2^2- 3-^2--3 -g -|--B~|
H —~ "O *O "o
tfcsciiiiiii§^2§l
^00*45 'o'o'ot ' 2 "2 "2
cieicJcio-lliislG^g
^ « 222 oooo'e.'c.'Ss'e,
?s c n >j M |^ TJ *3 ^3 ^3 *"* **" *"—
^^ *^ ^^ ^™ *^ ^" ^5 c4 ^S S\ ^« 3« mS S^
<=> o o- o o ci ci o". ci
o o o
2 5 0 oo
— — £> m
^ o b b 2 _ JS m os
s s oo a a 2
^^ ^^ ^N "• f^ i
1
_
a
1
u
O —
^^- 222<2
-------
extends from about 60 m to about 12 km downwind, covering an area of approximately
106 m2.
The above results are close to those predicted by SLAB. The higher result is about
50% larger than that predicted by SLAB. However, the difference is within the range of
uncertainties expected for these dispersion models. As noted above, the neglect of dry
deposition means that the predictions are likely to be conservative.
REFERENCES
AIChE, 1989. Guidelines for Chemical Process Quantitative Risk Analysis. American
Institute of Chemical Engineers,, NY._
Alp, E., et al., 1990. An Approach for Estimating Risk to Public Safety from Uncontrolled
Sour Gas Releases. ERCB Report 90-B (10 Volumes), Prepared by Concord
Environmental Corporation for Energy Resources Conservation Board, Calgary,
Alberta.
Briggs, G.A., 1973. Lift-Off of Buoyant Gas Initially on the Ground. National Oceanic and
Atmospheric Administration, Atmospheric Turbulence and Diffusion Laboratory File
ATDL 83, Oak Ridge, TN.
Ermak, D.L., 1989. User's Manual for the SLAB Model, An Atmospheric Dispersion Model
for Denser-than-Air Releases, Lawrence Livermore National Laboratory.
Layton, D.W., et al. 1983. Accidental Releases of Sour Gas From Wells and Collection
Pipelines in the Overthrust Belt: Calculating and Assessing Potential Health and
Environmental Risks. Lawrence Livermore National Laboratory Report UCRL-
53411, Prepared for the Division of Fluid Mineral Operations, Bureau of Land
Management, U.S. Department of the Interior, Washington, DC.
Quest, 1992. Hazards Analysis/Risk Analysis Study of Union Pacific Resources Company's
Wahsatch Gas Gathering Pipeline System. Prepared by Quest Consultants, Inc., for
Union Pacific Resources Company, Fort Worth, TX.
C-20
-------
HTNO.
EPA-453/R-93-045
- TITLE. AMO SUaTrrue'
Report to Congress on Hydrogen Sulfide Air Emissions
Associated with the Extraction of Oil and Natural Gas
_
3. RECIPIENT'S ACCESSION NO.
5 RFPORT DATE
' October 1993
. PERFORMING ORGANIZATION CODE
OAQPS - project lead - Laurel Driver
OSWER - project lead - Ed Freedman
9. PERFORMING ORGANIZATION NAMS AND ADDRESS
3. PERFORMING ORGANIZATION REPORT NO.
1O. PROGRAM EUEMENT NO.
II. CONTRACT/GRANT NO.
68-D2-0065
. SPONSORING AGENCY NAME AND ADDRESS
Office of Air ..Quality Planning and Standards (MD-13)
U.S. Environmental Protection Agency
Research Triangle Park, NC 27711
13. TYPE OF REPORT AND PERIOD COVERED
Final
14. SPONSORING AGENCY CODE
redetoSsubmit this"^ ^ °f ^ Clean A±* ACt Amendments of
the EPA
7.
KEY WORDS AND DOCUMENT ANALYSIS
DESCRIPTORS
b.lOENTIFIEHS/OPEN ENDED TERMS
c. COSATI Field/Group
Hydrogen Sulfide
Air Emission Sources
Hydrogen Sulfide Report to Congress
Clean Air Act Amendment of 1990
Oil and Natural Gas Extraction
8. DISTRIBUTION STATEMENT
Unlimited
19. SECURITY CLASS (This Report)
unclassified
21. NO. OF PAGES
20. SECURITY CLASS (Tkiipa$f
unclassified
22. PRICE
EPA Form 2220— 1
v. 4— 77) PREVIOUS EDITION is oasouere
-------
------- |