r States
mental Protection
Office of
Reseach and
Development
Office of Energy. Minerals and
Industry
Washington. D.C. 20460
EPA-600/7-77-016
February 1977
ACCIDENTS AND UNSCHEDULED
EVENTS ASSOCIATED WITH
NON-NUCLEAR ENERGY
RESOURCES AND TECHNOLOGY
Interagency
Energy-Environment
Research and Development
Program Report
-------
RESEARCH REPORTING SERIES
Research reports of the Office of Research and Development, U.S.
Environmental Protection Agency, have been grouped into seven series.
These seven broad categories were established to facilitate further
development and application of environmental technology. Elimination
of traditional grouping was consciously planned to foster technology
transfer and a maximum interface in related fields. The seven series
are:
1. Environmental Health Effects Research
2. Environmental Protection Technology
3. Ecological Research
4. Environmental Monitoring
5. Socioeconomic Environmental Studies
6. Scientific and Technical Assessment Reports (STAR)
7. Interagency Energy-Environment Research and Development
This report has been assigned to the INTERAGENCY ENERGY-ENVIRONMENT
RESEARCH AND DEVELOPMENT series. Reports in this series result from
the effort funded under the 17-agency Federal Energy/Environment
Research and Development Program. These studies relate to EPA's
mission to protect the public health and welfare from adverse effects
of pollutants associated with energy systems. The goal of the Program
is to assure the rapid development of domestic energy supplies in an
environmentally—compatible manner by providing the necessary
environmental data and control technology. Investigations include
analyses of the transport of energy-related pollutants and their health
and ecological effects; assessments of, and development of, control
technologies for energy systems; and integrated assessments of a wide
range of energy-related environmental issues.
This document is available to the public through the National Technical
Information Service, Springfield, Virginia 22161.
-------
Accidents and Unscheduled Events
Associated with
Non-Nuclear Energy Resources and Technology
C. Bliss
P. Clifford
G. Goldgraben
E. Graf-Webster
K. Krickenberger
H. Mahar
N. Zimmerman
EPA Contract Number 68-01-3188
\
£-
Project Officer
Steven J. Gage
Deputy Assistant Administrator
Office of Energy, Minerals, and industry
February 1977
Office of Energy, Minerals, and Industry
Office of Research and Development
U. S. Environmental Protection Agency
Washington, D.C. 20460
For sale by the Superintendent of Documents, U.S. Government Printing Office, Washington, D.C. 20402
-------
NOTICE
This report has been reviewed by the Office of Energy, Minerals,
and Industry, U.S.E.P.A., and approved for publication. Approval does
not signify that the contents necessarily reflect the views and policies
of the Environmental Protection Agency, nor does mention of trade names
or commercial products constitute endorsement or recommendation for use.
ii
-------
ACKNOWLEDGMENTS
The authors of this report are grateful to the following persons
for their expert review of part or all of this report:
Mr. Gregory D'Alessio
Mr. Morris Altschuler
Mr. Kenneth E. Biglane
Mr. Alfred Blackman
Mr. Kenneth Bridbord
Mr. James A. Fay
Ms. Barbara Heller
Mr. John Moran
Mr. Lloyd L. Philipson
Mr. Ken Solomon
Mr. L. Donald Williams
United States Environmental Protection Agency
United States Environmental Protection Agency
United States Environmental Protection Agency
National Institute for Occupational Safety
and Health
National Institute for Occupational Safety
and Health
Massachusetts Institute of Technology
Environmental Policy Center
National Institute for Occupational Safety
and Health
University of Southern California
The Rand Corporation
Battelle Pacific Northwest Laboratories
The opinions expressed in this report are those of the authors
and do not necessarily reflect those of the reviewers.
The authors also wish to acknowledge the suggestions made by
Mr. David Okrent, University of California, Los Angeles concerning
the content of the report. Mr. George F. Divine provided much use-
ful information for Section 12.
iii
-------
TABLE OF CONTENTS
Page
ill
ACKNOWLEDGMENTS
TABLES ix
FIGURES xlv
1.0 INTRODUCTION 1
2.0 SUMMARY 29
3.0 COAL 36
3.1 Resource System Overview 36
3.1.1 Exploration 36
3.1.2 Extraction 40
3.1.3 Transportation and Distribution ^
3.1.4 Processing/Beneficiation ^
3.1.5 Reclamation ^
3.2 Accident Overview 46
3.2.1 Exploration 46
3.2.2 Extraction 46
3.2.3 Processing/Beneficiation 58
3.2.4 Transportation and Distribution 64
3.2.5 Land Reclamation 65
4.0 CRUDE OIL 67
4.1 Resource System Overview 67
4.1.1 Exploration 67
4.1.2 Extraction 69
4.1.3 Processing 70
4.1.4 Transportation and Storage 71
4.2 Accident Overview 71
4.2.1 Oil Spills 73
4.2.2 Tanker Accidents 102
4.2.3 Offshore Rig Accidents 102
4.2.4 Pipeline Accidents 105
4.2.5 Transportation (Truck and Railroad) 111
4.2.6 Storage Tank Accidents 116
4.2.7 Refinery Accidents 117
v
-------
TABLE OF CONTENTS
(Continued)
Page
5.0 NATURAL GAS 124
5.1 Resource System Overview 124
5.1.1 Exploration 124
5.1.2 Production 126
5.1.3 Processing 126
5.1.4 Transportation 127
5.1.5 Storage 127
5.1.6 Distribution 127
5.2. Accident Overview 127
5.2.1 Exploration Accidents 129
5.2.2 Production Accidents 136
5.2.3 Processing Accidents 137
5.2.4 Transportation Accidents 137
5.2.5 Storage Accidents 146
6.0 LNG 150
6.1 Resource System Overview 150
6.1.1 Exploration 154
6.1.2 Production 154
6.1.3 Processing 154
6.1.4 Transportation to Liquefaction Plant 154
6.1.5 Liquefaction 154
6.1.6 Storage 157
6.1.7 LNG Tankers 157
6.1.8 Receiving and Regasification 157
6.1.9 Storage 161
6.1.10 Distribution 161
6.2 Accident Overview 161
6.3 Risk Analysis 165
7.0 HYDROELECTRIC 175
7.1 Resource System Overview 175
7.2 Accident Overview 178
vi
-------
8.0 GEOTHERMAL 181
8.1 Resource System Overview 181
8.1.1 Resources 183
8.1.2 Technology 187
8.1.3 Exploration 189
8.1.4 Resource Production 189
8.1.5 Conversion 190
8.2 Accident Overview 193
9.0 OIL SHALE 197
9.1 Resource System Overview 197
9.1.1 Exploration 198
9.1.2 Mining 200
9.1.3 Shale Preparation 200
9.1.4 Processing (Retorting) 200
9.1.5 Refining 201
9.1.6 Product Transportation 201
9.1.7 Land Reclamation/Spent Shale Disposal 201
9.1.8 In-Situ Processing 202
9.2 Accident Overview 203
10.0 SOLAR ENERGY 206
10.1 Resource System Overview 206.
10.1.1 Extent of the Resource 206
10.1.2 Technology 210
10.2 Accident Overview 220
11.0 END USE 225
11.1 Electricity Generation 225
11.1.1 Technology Overview 225
11.1.2 Accident Overview 228
11.2 Other End Uses 235
vii
-------
TABLE OF CONTENTS
(Continued)
12.0 ADVERSE NATURAL AND MAN-CAUSED INCIDENTS 243
12.1 Natural Disasters 243
12.1.1 Hurricanes 244
12.1.2 Tornadoes 245
12.1.3 Floods 247
12.1.4 Tsunamis 248
12.1.5 Snow and Ice Storms 248
12.1.6 Earthquakes 248
12.1.7 Land Subsidence 251
12.1.8 Avalanches and Landslides 251
12.1.9 Volcanoes 255
12.1.10 Meteorites 255
12.2 Adverse Man-Cuased Incidents 255
12.2.1 Airplane and Missle Crashes 255
12.2.2 Sabotage and Terrorism 256
12.2.3 War 257
12.3 Precautionary Measures 259
APPENDIX I - AGENCIES AND ORGANIZATIONS CONTACTED 261
REFERENCES CITED 265
viii
-------
Tables
Number Page
1 Representative Accidents and Unscheduled Events
Associated with Non-Nuclear Energy Resources and
Technology 3
2 Annual Deaths and Injuries Associated with Coal, Crude
Oil, and Natural Gas Fired Electricity Systems for a
1000 Megawatt Power Plant with a Load Factor of 0.75 29
3 Consumption of Bituminous and Lignite Coal, by Consumer
Class, 1974 38
4 Estimated Coal Resources of the World 39
5 Bituminous Coal and Lignite Shipment Methods, 1974 42
6 Bituminous and Lignite Coal Preparation, 1974 43
7 Selected Features of Gasification Processes 44
8 Selected Features of Liquefaction Processes 45
9 Comparison of Coal Resource Technology Options: Human
Health and Safety from Accidents 47
10 Annual Deaths, Injuries and Worker Days Lost for
Uncontrolled Coal-Fired Electricity Systems Associated
with a 10,000 Megawatt Power plant with a Load Factor
of 0.75 51
11 Injury Rates in Selected Industries, 1973 52
12 Fatalities in All Coal Mines, by Principal Cause,
1906-1970 53
13 Number, Average, and Severity of Injuries at Coal
Mines in the United States, by Principal Causes of
Injury and General Work Location, 1970 55
14 Statistics on Injuries and Injury Rates at Coal Mines
in the United States, 1966-1970 59
15 Primary Deficiencies Related to Roof and Rib Fall
Fatalities; Underground Coal Mines, 1967-1970 60
16 Major Disasters at Coal Mines in the United States,
1961-1970 61
IX
-------
Table (cont'd)
Number Page
— . - - - - - *-*- -
17 Fatal and Nonfatal Injuries Sustained During Surface
Strip Mine Activities, 1966-1972. 63
18 Percentage of Crude Petroleum and Petroleum Products
Moved by Method of Transportation, 1938-1974 72
19 Report of Occupational Injuries and Illnesses for the
Year 1975 Covering Operations Subject to OSHA Record-
keeping Requirements Only 74
20 Causes of Fatalities in the Petroleum Industry 1974
and 1975 76
21 Occupational Accidents for Crude Oil and Product
Transport and Refining 77
22 Annual Deaths, Injuries and Workdays Lost for
Uncontrolled Oil-Fired Electricity Systems Associated
with a 1,000-Megawatt Powerplant with a Load Factor
of 0.75 79
23 Contributions of Various Sources to Oil in the Oceans 85
24 Number of Oil Spills by Volume and Source, 1971-1972 86
25 Tankship Accidents, 1969-1973 87
26 Accident Sequence for 47 Tankship Losses for Tankships
over 10,000 Deadweight Tons, 1969-1973 88
27 Major Oil-Spill Incidents - Significant Characteristics QQ
(1956-1969)
28 Summary of Toxicity Data 93
29 Variations in Biologic Effects of Oil with Type of
Oil t 94
30 Biologic Recovery Following Oil Spills gg
31 Effects of Oil on Selected Species 99
32 Extent of Tankship Loss or Damage 1969-1973
33 Description of Loss of Structural Integrity for 47
Tankship Losses for Tankships over 10,000 Deadweight
Tons, 1969-1973 104
-------
Table (cont'd)
Number Page
34 Liquid Pipeline Accident Summary - January 1, 1975
through December 31, 1975 106
35 Pipeline Accident Summary - by Commodity Involved 109
36 Rail and Highway Accidents 112
37 Summary of Rail Accidents Involving Hazardous
Materials, 1971-1975 114
38 Accident Probability 115
39 Accident Loss Summary for Various Processes, 1965-1969 118
40 Losses Over $10,000 Occurring in Gasoline Plants During
the Years from 1959 through 1971 121
41 Refinery Process Unit Losses Originating in Fired
Heaters 123
42 Accident Rates for Extracting, Gathering, Processing,
Transmitting, Distributing, and Storing Natural Gas 130
43 Reports of Occupational Injuries and Illnesses for
the Year 1975 Covering Operations Subject to OSHA
Recordkeeping Requirements Only 131
44 Annual Accident Impact of an Uncontrolled Gas-Fired
Electricity System Associated with a 1,000 Megawatt
Powerplant with a Load Factor of 0.75 132
45 Fatalities by Function and Cause for the Natural Gas
Industry Years 1975 and 1974 133
46 Employee Disabling Injury Frequency and Severity
Rates, by Type of Gas, 1950-1974 134
47 Employee Accident Frequency and Severity Rates of
Selected Industries, 1974 135
48 Accidents and Casualities Reported by Gas System
Operators during 1975 140
49 Summary of Gas Pipeline Accidents and Casualties
Reported during Years 1970-1975 141
50 Number of Incidents by Year of Pipe Installation 142
xi
-------
Tables (cont'd)
Number Page
51 Number of Incidents by Cause Identified
52 Consequences of Incidents 145
53 Number of Corrosion Incidents by Corrosion
Description 146
54 Number of Outside Force Incidents by Primary Cause 147
55 Statistics of Interstate Natural Gas Pipeline
Companies - 1974 148
56 Physical Properties of LNG and Its Constituents 151
57 Projected LNG Import /Receiving Terminals in the
U.S. " 152
58 Projected LNG Imports 156
59 Cryogenic Materials and Associated Temperature
Limitations 160
60 Predicted Frequency of Fatal LNG Tankship Accidents
in U.S. Ports Based on 1,000 Tankship Trips to the
U.S. in 1980 169
61 Probabilities and Consequences of Marine Phase
Accidents 171
62 Land Spill Probabilities and Total Fatality Rates 172
63 Estimated Expected Public Fatalities Per Exposed
Person Per Year 174
64 Resources and Reserves - Gulf Coast Conduction-
Dominated Geothermal Area 188
65 Type of Accident Occurence by Geothermal Resource 194
66 Metal Mining Accidents in Selected Western States 205
67 Site-Selection Criteria for Tidal-Powered Development 208
68 Electric Power Generation by Energy Resource £26
69 Major Power Interruptions 1967 to 1970 as Reported
to Federal Power Commission 231
xii
-------
Number Tables (cont'd) Page
70. Ranking of Accidents in 1975 in Electric 232
Utilities as Reported to the Federal
Power Commission
71. Occupational Injuries due to Accidents in
Generating Plants 233
72. Accident Rates Associated with a 1,000 MWe
Steam Electric Plant 234
73. Motor Transportation Accident Rates 1970-1972 236
74. Annual Industrial Energy Consumption for Various
Fuel Types 237
75. Annual Energy Consumption for Commercial and
Residential Uses 239
76. Transportation Accidents 240
77. Building Fire Losses Caused by Electrical,
Heating, and Cooking Equipment 242
78. Estimate of Effect of Storms Exceeding Pre-
dicted 100 and 200 Year Storm Levels 246
79. Summary of the Effect of Natural Phenomena on
Various Elements of the Oil Production System 249
80. Estimates of Offshore Drilling Facilities
Damage Due to Earthquakes 252
81. Prediction of Earthquake in California by
Fault Theory Using a Computer 253
82. Estimated Efforts of Total and Instantaneous
Failure of Dam Filled to Capacity 254
xiii
-------
Figures
Number Page
1 Coal Resource System Technology Options 37
2 The Crude Oil Cycle 68
3 Spill Volume per Year 81
4 Source of Spill Data from 36 Incidents, 1959-1969 83
5 Composition of Material Spilled 84
6 Observed Persistence of Petroleum Substances
in Various Marine Habitats Following Actual
Oil Spills 97
7 Loss Trends in Hydrocrackers-1962 to 1971 120
8 Natural Gas Energy System 125
9 Typical Production Facility with Safety
Equipment 128
10 Number of Incidents and Cause Versus Year of
Occurrence 139
11 LNG Operations 155
12 Simplified Diagram of Cascade Cycle for
Liquefaction of Natural Gas 158
13 Liquefaction of Expander Cycle 159
14 Event Chain for LNG Spills 167
15 LNG Tankship Accidents Assuming 1000 Tankships
per Year; Predicted Frequency of Classes of
Accidents with a Statistical Number of Fatalities. 153
16 Typical Hydroelectric Power Plant 177
17 Regions of Intense Geothermal Manifestations 132
18 Near-term Scenario for Geothermal Energy Develop-
ment 184
19 Estimated National Growth of Geothermal
Electric Power 135
xiv
-------
Figures Cont'd
Number Page
20 Potential Resource Recovery System Hot-Dry-
Rock Hot Igneous System 191
21 Vapor-Turbine Cycle Diagram 192
22 Shale-Oil Production Complex Schematic
Material Flow 199
23 A Solar Powered Total Energy System for
Commercial Installations 212
24 Central Station Electricity Generation 213
25 A Modular Solar-Sea Power Plant 215
26 The Wind Energy Conversion System 217
27 Approaches to the Conversion of Solar
Energy into Electricity 221
xv
-------
SECTION 1.0
INTRODUCTION
Since the Arab oil embargo, methods of achieving energy independence have
received much attention in the United States. Most of the discussion has con-
cerned availability of various energy resources, availability and cost of tech-
nology, and environmental effects.
Nuclear energy could supply a significant part of our electrical energy
needs. Safety considerations, however, have proven to be a major impediment
to development of nuclear energy.
In order to choose intelligently among the alternatives available to us,
the safety of using our non-nuclear energy resources must be considered. Al-
though some hazards such as oil spills and coal mine explosions have received
significant attention from the public and the regulatory agencies, no compre-
hensive assessment of the safety of developing our non-nuclear resources has
been prepared.
The first step in the assessment of the hazards of non-nuclear systems is
a compilation of existing accident data. The purpose of this report is to
summarize the available information on non-nuclear power systems. All acci-
dents or unscheduled events, whether natural or man-made, are considered.
However, emphasis is placed on major accidents or minor accidents which have
a cumulative major effect.
The energy systems covered by this report are coal, crude oil, natural
gas, liquefied natural gas (LNG), hydroelectric, oil shale, geothermal, and
solar. For the purposes of this report, an accident or unscheduled event is
an unexpected incident which poses a threat to human health or safety, plant,
property, and equipment, or the environment. These incidents can occur during
the construction, operation, or maintenance of the system. Accidents are con-
sidered to be incidents which occur because of factors inherent in the indivi-
dual energy system (e.g., roof collapse of coal mines). Unscheduled events
are incidents which may affect more than one energy system and are caused by
factors external to these systems (e.g., war, earthquakes). Excluded from
consideration are environmental effects or threats to human health, safety,
or property resulting from normal operationss whether continuous or batch
(e.g., SOX and NOX emissions from power plants or water pollution from waste-
water treatment systems at oil refineries). Also excluded are uncontrolled
discharges to the environment such as surface runoff, acid mine drainage, and
fugitive emissions. Because only non-nuclear energy systems are being consid-
ered, accidents associated with the development of geothermal resources using
-------
nuclear explosives (Operation Gas Buggy) are not included. Supply interrup-
tions also are not considered to be within the scope of this study.
In order to identify accidents which might occur, each resource system
was divided into energy cycle elements of exploration, extraction, processing,
transportation (including transmission and distribution), and end use. Tech-
nologies used in each element were examined for their accident potential and
a preliminary list of types of accidents was prepared. This list was then
modified based on telephone contact with governmental agencies, trade asso-
ciations, unions, insurance companies, and other groups with expertise in the
field. A list of the organizations contacted is contained in Appendix I.
These organizations also were asked to supply data concerning the frequency
and severity of each accident.
Table 1 identifies representative accidents or unscheduled events. It is
believed that this table contains all major accidents likely to occur, all
minor accidents which cumulatively have a major effect, and a sampling of other
accidents which might ojcur. This table is not to be considered as exhaustive
because it is impossible to envision or to list all accidents which might occur.
The table also contains an estimate of the frequency and severity of each
accident listed. The following definitions were used in preparing the table;
Major - Having the potential for causing death or severe injury to more
than 10 people, causing more than $500,000 damage to plant, property, or equip-
ment, or having significant ecological impact (i.e., causing severe ecosystem
disruption or killing a significant portion of a segment of the local biolog-
ical community).
Minor or Unknown - All other known accidental or unscheduled occurrences
and unscheduled occurrences whose magnitude cannot be estimated.
Frequent - An accident or unscheduled event which occurs at least 5 times
in the United States during a single year.
Infrequent - An accident or unscheduled event which occurs 4 times or
less in the United States during a single year.
Hypothetical - An accident or unscheduled event which may be expected to
occur but whose frequency cannot be estimated because there is no sound basis
for objective estimation.
The definition of major accidents should not be interpreted as meaning
that accidents with the potential for killing of severely injuring fewer than
10 persons are inconsequential. The death of even a single individual repre-
sents an immeasurable and irreplaceable loss. Rather, the purpose of differ-
entiating between major and minor accidents is to distinguish between poten-
tially catastrophic occurrences such as hydroelectric dam failure and less
severe individual accidents such as slips, falls, or most automobile accidents.
In most instances, estimates of frequency and severity were based on data
presented in the report. For developing technologies, however, data were not
-------
TABLE 1. REPRESENTATIVE ACCIDENTS AND UNSCHEDULED EVENTS ASSOCIATED WITH NON-NUCLEAR ENERGY RESOURCES AND TECHNOLOGY
RESOURCE
ENERGY CYCLE
ELEMENT
TECHNOLOGY AREA
ACCIDENT
SEVERITY FREQUENCY OF OCCURRENCE
Major/Minor Frequent Infrequent Hypothetical REMARKS
COAL
Exploration
Detailed site in-
spections , in-
cluding collec-
tion of surface
and sub-surface
strata for anal-
ysis and evalu-
tion
Extraction/
Reclamation
Underground
Aircraft accident
during aerial
survey
Personnel injury
and/or property
damage from im-
proper use of
explosives (e.g.,
detonations,
flying debris)
Improper use of
boring/drilling
equipment
Explosion, release
of toxic gases
caused by drilling
operation inter-
secting existing
natural gas reservoir
Heavy equipment
accidents during
initial excavation,
site preparation
Explosion and/or fire
from fracturing
activities
Generic hazard
Generic hazard
Generic hazard
(continued)
-------
TABLE 1 (continued)
RESOURCE
ENERGY CYCLE
ELEMENT
TECHNOLOGY AREA
ACCIDENT
SEVERITY FREQUENCY OF OCCURRENCE
Major/Minor Frequent Infrequent Hypothetical REMARKS
Extraction/ Underground
Reclamation (continued)
(continued)
Roof collapse
Face and rib falls
Face, roof, and rib falls,
although affecting a
small number of indi-
viduals per occurrence,
happen so frequently
that the cumulative im-
pact is the greatest
underground mining
hazard
Surface
Bursts (pressure)
Mishandling material,
tools, machinery
Electrocution,
electric arc burns
Gas or dust explosions
Haulage accidents
Intersection with
existing oil or
gas well or
reservoir
Highwall collapse
Gas or dust explosions,
although often severe,
occur infrequently.
Haulage or heavy equip-
ment related accidents
are second most severe
in underground mining
activities
Fall of highwalls or
slides produces sub-
stantial numbers of
non-fatal injuries
(continued)
-------
TABLE 1 (continued)
ENERGY CYCLE
RESOURCE ELEMENT TECHNOLOGY AREA
Extraction/ Surface
Reclamation (continued)
(continued)
Auger
Spoils disposal
SEVERITY FREQUENCY OF OCCURRENCE
ACCIDENT Major/Minor Frequent Infrequent Hypothetical REMARKS
Heavy equipment, x x
haulage accidents
Improper use of x x
explosives
Electrocution x x
Mishandling of hand x x
tools or machinery
Fires x x
People falling x x
Heavy equipment x x
accidents
Tailing dam collapse x x
Fires or explosions x x
Landslides, highwall x x
Heavy equipment operation
has significant poten-
tial for accidents
Generic hazard
Tailing dam collapse
can be catastrophic
Landslides, wall col-
Processing
Reclamation
Beneficiation/
Preparation
(crushing,
sizing, clean-
ing, drying,
etc.)
collapse
Accidents involving
heavy equipment
Fire or explosion from
mechanical friction
or impact
lapse are frequent
but often minor
Human fatality hazard
small, but non-fatal
injuries can be
significant
(continued)
-------
TABLE 1 (continued)
RESOURCE
ENERGY CYCLE
ELEMENT
TECHNOLOGY AREA
ACCIDENT
SEVERITY FREQUENCY OF OCCURRENCE
Major/Minor Frequent Infrequent Hypothetical REMARKS
Processing
(continued)
Beneficiation/
Preparation
(crushing, siz-
ing, cleaning,
drying, etc.)
(cont inued)
Liquefaction
Ignition of coal dust
fires due to local
temperature buildup
Accidental release
of accumulated
dust or fires to
the environment
Haulage accidents
Mishandling of hand
tools or equipment
Gas or dust explosions
People falling
Fires, gas or dust
explosion from
feedstock handling
Pressurized reaction
vessel rupture
Explosion due to
slurry devolitiza-
tion, depressuri-
zation.
Release of reactants,
catalysts, by-prod-
ucts during explosion
Haulage, conveyor acci-
dents are common
Gas or dust explosions
appear to be the two
most significant
hazards resulting in
the greatest number
man-days lost per
occurrence
Few operating data
exist for liquefaction,
gasification, or in
situ gasification.
Therefore, the fre-
quency of most acci-
dents is categorized
as hypothetical.
(continued)
-------
TABLE 1 (continued)
RESOURCE
ENERGY CYCLE
ELEMENT
Processing
(continued)
TECHNOLOGY
AREA
Gasification
ACCIDENT
Fires , gas or
explosions
dust
from
SEVERITY
Major /Minor
X
FREQUENCY OF
Frequent Infrequent
X
OCCURRENCE
Hypothetical
REMARKS
In-situ gasi-
fication
feedstock handling
Pressurized reaction
vessel rupture
Explosion or ignition
of hot aromatic by-
product oils
Release of reactants
to environment
Uncontrolled combus-
tion of coal seam
Land subsidence
Induced seismicity
Leakage of reaction
gases to surround-
ing strata
Uncontrolled fractur-
ing of coal seam
Accidental disrup-
tion of ground
water, aquifers
Available information
indicates that methods
to mitigate or remove
known problems appear
insufficient at this
time.
(continued)
-------
TABLE 1 (continued)
RESOURCE
ENERGY CYCLE
ELEMENT
TECHNOLOGY AREA
Coking
ACCIDENT
Normal industrial
SEVERITY
Maior/Minor
X
FREQUENCY OF OCCURRENCE
Frequent Infrequent Hypothetical
X
REMARKS
Explosions have occurred,
Transportation
and distri-
tion
In-mine trans-
portation
Surface distri-
tribution
hazards associated
with heavy equip-
ment, or high
temperature environ-
ment
Explosions during
heat-up or battery
oven reversals
Haulage accidents
Conveyor fires,
.explosion
Fires, explosions
during transfer
Vehicular accidents
Freight transfer
accidents
Barge, vessel over-
turning , sinking
Railway, truck
collisions
but infrequently
In-mine haulage acci-
dent causes substantial
loss of lives and
production
Generic hazard; coal
specific data are not
available
(continued)
-------
TABLE 1 (continued)
RESOURCE
ENERGY CYCLE
ELEMENT
TECHNOLOGY AREA
ACCIDENT
SEVERITY FREQUENCY OF OCCURRENCE
Ma.1 or/Minor Frequent Infrequent Hypothetical REMARKS
Crude
OH
Exploration
Surveys
Exploratory
Drilling
Extraction
(Production)
Well
Completion
Accidents resulting
from the use of
explosives
Accidents on land
or ship resulting
from handling of
equipment
Blowouts causing gas
leaks, oil spills,
fires and explosions
Collision of a vessel
with a drilling rig
Rig accidents
Blowouts causing gas
leaks, oil spills,
fires and explosions
Oil spills from causes
other than blowouts
Rig accidents
Subsidence
Most explosions are
contained
Small spills are fre-
quent; large spills
are infrequent
(continued)
-------
TABLE 1 (continued)
RESOURCE
Crude
Oil
ENERGY CYCLE
ELEMENT
Extraction
(Production)
TECHNOLOGY AREA
Field Processing
ACCIDENT
Fires and explosions
SEVERITY
Major /Minor
X
FREQUENCY OF OCCURRENCE
Frequent Infrequent Hypothetical
X
REMARKS
Improved
Recovery
Processing Refining
Transportation Tankers and
and Storage Barges
of Crude and
Refined Prod-
uct
Spills of brines or
emulsified oil
Processing platform
accidents involving
explosion, fires,
capsizing, etc.
Subsidence
Steam injection
accidents
Fires and explosions
Oil spills
Failure of waste-
water treatment
equipment
Oils spills due to
collisions, capsiz-
ing, running aground
and other causes
Spills and leaks dur-
ing lightering
bilge wash, etc.
x
X
Diversion to holding ponds
generally prevents re-
lease of untreated mate-
rials to the environment.
Most occurrences are minor
(continued)
-------
TABLE 1 (continued)
RESOURCE
Crude
Oil
ENERGY CYCLE
ELEMENT TECHNOLOGY AREA
Transportation Pipelines
and Storage
of Crude and
Refined Prod-
uct
ACCIDENT
Oil spills due to
pipeline failures,
valve malfunction,
transfer hose rup-
ture and other
causes
SEVERITY
Major /Minor
X
FREQUENCY OF OCCURRENCE
Frequent Infrequent Hypothetical
X
REMARKS
Trucks
Railroad Tank
Cars
Natural Gas Exploration
Tank Farms
Surveys
Fires and explosions x
Truck accidents re-
sulting in fires
and explosions
Derailment or colli-
sions resulting in
fire, explosions, or
. environmental damage
Spills and leaks dur-
ing transfer operations
Fires and explosions x
Tank rupture or
overflow
Accidents involving mis-
use of explosives
Accidents involving radio-
active sensors
Although major accidents
can occur most are
minor
Although major accidents
can occur most are
minor
Not directly associated
with natural gas
properties
Not directly associated
with natural gas
properties
(continued)
-------
TABLE 1 (continued)
ENERGY CYCLE
RESOURCE ELEMENT TECHNOLOGY AREA
Natural Gas Exploration Exploratory
Drilling
Extraction Well completion
Production
Processing Processing
Transportation Pipelines
and Storage
Underground
Storage
Storage Tanks
Peak Shaving
Plants
Local Distri-
bution
SEVERITY
ACCIDENT Mai or /Minor
Blowouts x
Blowouts x
Fires and explosions x
Oil spills x
Release of sulfur x
compounds I^S, S02
Fires and explosions x
Rupture of processing x
equipment
Leaks, ruptures, explo- x
sions, material fail-
ures corrosions 2nd
party equipment
Fires, explosions, con- x
tamination of water
supplies, leaks
Leaks, ruptures, fires, x
explosions
Ruptures, fires, explo- x
sions
Ruptures, leaks result- x
ing in fires and
explosions
FREQUENCY OF OCCURRENCE
Frequent Infrequent Hypothetical REMARKS
x
x
x Results from uncontrolled
release of gas; possible
x failure of downhole valve,
maintenance error.
x
X
X
X
x
X
X
X
(continued)
-------
TABLE 1 (continued)
RESOURCE
LNG
ENERGY CYCLE
ELEMENT
Exploration
Extraction
Processing
TECHNOLOGY AREA
See Natural Gas
See Natural Gas
Pre treatment
Liquefaction
ACCIDENT
See Natural Gas
Fires and explosions
Solids blocking
SEVERITY FREQUENCY OF OCCURRENCE
Major /Minor Frequent Infrequent Hypothetical REMARKS
X X
X X
Transportation Tankers
and storage
Transfer
Storage and
regasification
pipeline
Release of refrigerants
Onboard malfunctions
causing spills and
fires
Impact with 2nd body
Leaks and rupture of
pipeline
Structural failures or
rupture of holding
facilities, "roll-
over"
This includes vessel
failures not related
to LNG but which could
affect LNG tanks
This includes rammings,
groundings, and colli-
sions
Only accidents concern-
ing LNG to date involve
storage facilities
Accidents during
maintenance
Leakage in regasifi-
cation lines
(continued)
-------
TABLE 1 (aontinued)
RESOURCE
Hydroelectric
ENERGY CYCLE
ELEMENT
Resource
Power Con-
version
TECHNOLOGY AREA
Dams and
Reservoirs
Conversion
(Turbines)
ACCIDENT
Dam rupture
Loss of water
Overflow
Reservoir leaks
Boating accidents
Flooding of the
power house
SEVERITY
Mai or /Minor
X
X
X
X
X
X
FREQUENCY OF OCCURRENCE
Frequent Infrequent Hypothetical
X
X
X
X
X
X
REMARKS
Plant damage due to
survey oversights
Personnel in the
course of duty
Geothermal Exploration
Surveys
Conduit failure
Turbine failure
Electrical fires
Power interruption
Accidents involving
equipment
Airplane crashes
Auto accidents
Personnel injuries
and illnesses
x
x
As a result of exposure
to the elements and
operations in remote
areas.
(continued)
-------
TABLE 1 (continued)
RESOURCE
ENERGY CYCLE
ELEMENT
TECHNOLOGY AREA
ACCIDENT
SEVERITY FREQUENCY OF OCCURRENCE
Major/Minor Frequent Infrequent Hypothetical
REMARKS
Geothermal Exploration Drilling
(continued) (continued)
Extraction
Drilling and
Well operation
Blowouts
Subsidence
Rig accidents
Blowouts
Induced seimicity &
seismic induced
failures
Leaks and ruptures
Landslides and
erosion
Well failures
Subsidence
Rig failure
Explosion
Results in emissions of
hot fluids/gases to
environment or to
aquifer, etc.
Including personnel
operating rig
Same as exploration
Wells and pipelines
Caused by poor drilling
and construction
practices
Probably not in vapor-
dominated fields
Usually as a result of
blowout or seismic
activity
Gas ignition
(organic fluids)
(continued)
-------
TABLE 1 (continued!
ENERGY CYCLE
RESOURCE ELEMENT TECHNOLOGY AREA
Geothermal Extraction Pipeline gather-
(continued) (continued) ing system
Conversion
Oil Shale Exploration Mapping and
drilling
Resource Surface mining
Extraction Blasting to
rubbleize
structure
Forming the pit
Excavating the
mineral
SEVERITY FREQUENCY OF OCCURRENCE
ACCIDENT Major /Minor Frequent Infrequent Hypothetical
Pipeline ruptures x x
Pipeline leaks x x
Burns x x
Mechanical accidents x x
Fire x x
Automotive equipment x x
accidents, unfore-
seeable exposure to
elements of nature
Explosion before per- x x
sonnel and equipment
clear the area
Landslides entrapping x x
personnel and/or
equipment
Slips, falls and equip- x x
ment related accidents
REMARKS
A result of an earth-
quake or pressure
buildup
Isobutane mixtures in
the isobutane cycle.
Common to any activity
associated with re-
source evaluation in
remote areas .
(continued)
-------
TABLE 1 (continued)
RESOURCE
ENERGY CYCLE
ELEMENT
TECHNOLOGY AREA
ACCIDENT
SEVERITY FREQUENCY OF OCCURRENCE
Major/Minor Frequent Infrequent Hypothetical
REMARKS
Oil Shale Resource
(continued) Extraction
(continued)
Underground mining
Forming rooms Roof collapse
and pillars
Blasting accidents
Dust explosions
Mine ventilation Failure of equipment
Removal of Automotive equipment
mineral accidents
Vehicle/vehicle-
pillar collisions
Falls from tall equip-
ment
Landslides affecting
haulage road
Land subsidence
Shale Crushing, convey- Accidents involving
Preparation ing, and screen- machine operation
ing
x
X
X
X
Proper roof bolting can
avoid this
(continued)
-------
TABLE 1 (continued)
RESOURCE
ENERGY CYCLE
ELEMENT
TECHNOLOGY AREA
ACCIDENT
SEVERITY FREQUENCY OF OCCURRENCE
Major/Minor Frequent Infrequent Hypothetical REMARKS
Oil Shale
Processing
Retorting (Oil
shale)
Oil-Shale
charging
Fuel supply
system
Shale-Oil
recovery
Spent-Shale
discharge
Refining (Shale
Oil)
Storage (Tank
farms)
Hydrogenation
Coking
Conveyor belt
accidents
Explosions
Oil spills
Conveyor belt
accidents
Fires and burns
generated by hot
spots in spent shale
Oil spills and fires
Explosions and equip-
ment rupture
Conveyor belt
accidents
High-pressure operation
(continued)
-------
TABLE 1 (continued)
RESOURCE
Oil Shale
(continued)
ENERGY CYCLE
ELEMENT
In-Situ
Recovery
TECHNOLOGY AREA
Drilling
Blasting
Igniting
Retorting
Gas Recycling
Oil Recovery
ACCIDENT
Surface accidents
involving blast-
ing and machinery
SEVERITY FREQUENCY OF OCCURRENCE
Mai or /Minor Frequent Infrequent Hypothetical
X X
REMARKS
Underground site probably
will be too deep for
surface effects. No
personnel present un-
derground during re-
torting operation
Preparation of
Underground Site
Product Trans- Pipeline Rupture
portation
Solar Energy
Land/Reclama-
tion/Oil-
Shale Dis-
posal
Direct Conver-
sion Photo-
voltaic
Pump Failures
Groundwater
Control
Materials
Handling
Electrical Energy
Production
Same as for under-
ground mining
Oil spills under or
above ground
Localized fires
Oil spills in pump house
Failure of retaining
dams or embankments
Surface accidents
involving machinery
Solar cell breakage
Wire damage or breakage
Solar cell failure
DC/AC converter failure
Electrical fire
Injuries (burns, falls)
x
x
X
X
X
Mineralized water can
contaminate public
water supplies
(continued)
-------
TABLE 1 (continued)
ENERGY CYCLE
RESOURCE ELEMENT TECHNOLOGY AREA ACCIDENT
Solar Energy Solar-Thermal Solar to heat- Reflector damage
energy conver- Water pipe rupture
sion Hot water storage
failure
Mirror breakage
Heliostat failure
Boiler rupture/leak
Heat exchange fluid
spill or leak
Injuries (burns, falls)
Electrical Energy Steam turbine accidents
Prod.
Electrical fires
Miscellaneous Structural failure
Solar-sea Resource recovery Pump failure
Corrosion (pipeline &
other equipment)
Working fluid leak
(other than water)
Boiler failure
Condenser failure
Turbine failure
SEVERITY
Mai or /Minor
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
FREQUENCY OF OCCURRENCE
Frequent Infrequent Hypothetical REMARKS
X
X
X
X
X
X
x Proposed liquid metal
(Na or Nak) system
X
x Same as for Geothermal,
or Fossil Fuel-fired
steam electric gener-
ating plant.
x
x
x
x
X
X
X
X
Fire and explosion
(energy storage)
(continued)
-------
TABLE 1 Ccontinued)
ENERGY CYCLE
RESOURCE ELEMENT
Solar-Sea
(continued)
TECHNOLOGY AREA
Energy Trans-
portation
Miscellaneous
SEVERITY FREQUENCY OF OCCURRENCE
ACCIDENT Maior/Minor Frequent Infrequent Hypothetical REMARKS
Fire and explosion x x
Ship collision x x
Pipeline rupture x x
Storm damage x x
Wind Power
Resource
recovery
Electrical Energy
Prod.
Miscellaneous
Tidal and Wave
Propeller blade
breakage
Hub damage
Construction & erec-
tion accidents
Injuries during
maintenance
Tower service eleva-
tor failure
Generator failure
Electrical fires
Structural failure
(support tower)
Propeller striking
surrounding objects
Barrage failure or
breaching
Gate failure
Pump & other machin-
ery failure
Storm damage
Earthquake damage
Turbine failure
x
x
x
x
For other than commercial
units.
x
x
Same as for the Hydro-
electric hydraulic
turbine
(continued)
-------
TABLE 1 (continued)
ENERGY CYCLE SEVERITY FREQUENCY OF OCCURRENCE
RESOURCE ELEMENT TECHNOLOGY AREA ACCIDENT Maior/Minor Freauent Infreouent Hypothetical REMARKS
Tidal and Wave
(continued)
Biomass Con-
version
Generator accidents x
Flooding x
Navigation accidents x
Gas explosion x
Fires x
x
x
X
X
X
Same as for the hydro-
electric generators
Same as for any combus-
tion process.
-------
TABLE 1 (continued)
ENERGY CYCLE
RESOURCE ELEMENT TECHNOLOGY AREA
All End Use Generation of
Electricity
Boiler Fired
Plant
Gas Turbine
Plant
Fuel Cell
MHD
SEVERITY FREQUENCY OF OCCURRENCE
ACCIDENT Major /Minor Frequent Infrequent Hypothetical REMARKS
Explosion x
Implosion
Tube rupture
Scalding
Explosion (mechanical x
or pressure)
Asphyxiation
Rupture
Gas leak
Fire
Magnetic explosion x
Burn-through press
explosion
Asphyxiation
Failure of vent or x
vapor handling system
Leak in alkali metal
x
X X
x x
x x
X
X X
X X
X
X
X
X
X
x
x
X
X
X
X
X
General
vapor system
Alkali metal/water
reaction
Fire and explosion
Power failure
Flooding
Electrical fires
Occupation injuries
Fuel leaks (liquid
and vapor)
Oil leaks (containing
PCB)
x
X
X
X
X
X
X
X
May affect residents of
area
(continued)
-------
TABLE 1 (continued)
ENERGY CYCLE
RESOURCE ELEMENT TECHNOLOGY AREA
All End Use Transmission and
Distribution of
Electricity
SEVERITY FREQUENCY OF OCCURRENCE
ACCIDENT Mai or /Minor Frequent Infrequent Hypothetical REMARKS
Ground fault x x
Insulation failure x x
Lightening strikes x x
Aircraft explosion x x
Ship collision x x
Causing possible
power failure
Causing possible
power failure
fire or
fire or
Industrial
Commercial
Residential
(incl. tanker)
Farm equipment
accidents
Aircraft collision
(commercial)
Plant fire/explosion
Plant equipment failure
Plant flooding
Occupational injuries
Toxic substance efflu-
ent or emission
Pipe ruptures (water,
gas, steam)
Electrical fires
Occupational injuries
Building collapse
Pipe ruptures
Fires
Gas explosions
Falls in the home
Gas leaks
Storm damage
x
x
x
x
X
X
X
X
X
X
X
X
X
X
(continued)
-------
TABLE 1 (continued)
RESOURCE
ENERGY CYCLE
ELEMENT
TECHNOLOGY AREA
ACCIDENT
SEVERITY FREQUENCY OF OCCURRENCE
Major/Minor Frequent Infrequent Hypothetical REMARKS
Residential
(continued)
All
Appliance accidents
Heating and air con-
ditioning equipment
accidents
Flooding (storm or
pipe break)
Home accidents
Electrocution
Asphyxiation
Trees falling on lines
Accidents knocking over
power lines
Line breaks due to ice
Power failure due to
wildlife short circuit-
ing
Line overload
Storm damage
Corona discharge effects
Occupational injuries
Electrical equipment
failure
Fire
Electrocution
Hurricane
x
x
x
x
x
x
x
x
x
x
x
X
X
X
X
X
X
X
X
X
X
Accidents affect human
health, personal prop-
erty, and the environ-
ment
Major threat to energy
facilities; 38% of oil
processing and 74% of
natural gas production
is in vulnerable region.
(continued)
-------
TABLE 1 (continued)
RESOURCE
ENERGY CYCLE
ELEMENT
TECHNOLOGY AREA
ACCIDENT
SEVERITY FREQUENCY OF OCCURRENCE
Ma.1 or/Minor Frequent Infrequent Hypothetical
REMARKS
Tornadoes
Floods
Tsunamis
Snow and Ice Storms
Earthquakes
Land Subsidence
Avalanches and
Landslides
Volcanoes
Meteorites
x
x
18% of crude processing
capacity and 12% of
natural gas production
in high risk region.
Minor threat to national
total oil, gas and coal
production.
8% of crude refining
capacity, 3% of gas
production and 15% of
coal production located
in areas of major seismic
risk.
Insignificant threat to
United States energy
facilities.
(continued)
-------
TABLE 1 (continued)
RESOURCE
ENERGY CYCLE
ELEMENT
TECHNOLOGY AREA
ACCIDENT
SEVERITY FREQUENCY OF OCCURRENCE
Ma.i or/Minor Frequent Infrequent Hypothetical REMARKS
Airplane and
Missile Crashes
Sabotage
War
Crashes of airplanes occur
frequently for general
aviation but infrequent-
ly for commercial
aviation.
aFrequency refers to the number of times an event occurs in the United States rather than the number of times the event effects energy
related facilities.
-------
available. In these instances, accidents, frequencies, and severities were
based on subjective assessment of the technologies involved.
The emphasis of each chapter of the report is on major accidents or minor
accidents which cumulatively have a major effect. Because of the limited time
available for completion of the report (ca. 6 weeks), the report was restricted
to summarizing available data. Detailed risk analyses were not performed. No
attempt was made to collect a new and consistent set of data which would allow
comparison of one energy resource area to another. Thus, only limited compar-
isons are made of accident risks among resource areas.
A separate chapter or part of a chapter is devoted to each resource area.
Because most end uses are common to more than one energy system, a separate
chapter is devoted to this subject. Similarly, unscheduled events such as
hurricanes affect more than one energy system and are treated in a separate
chapter.
28
-------
SECTION 2.0
SUMMARY
Accidents can occur at all steps in the energy cycle for all enery systems,
nuclear and non-nuclear. They can be caused by human error, structural or me-
chanical failure, sabotage and natural phenomena such as hurricanes and earth-
quakes. These accidents can cause environmental damage, economic losses, fatal-
ities, injuries, and health impairment. A summary of the frequency and severity
of representative accidents for the coal, crude oil, natural gas, LNG, hydro-
electric, geothermal, oil shale and solar energy systems appears in Table 1.
Although data are available from many sources, it is difficult to assess
the relative risk of accident occurrence among the principal non-nuclear energy
systems because of the lack of consistent data recording and reporting. How-
ever, based on the best available estimate, there probably are a significantly
greater number of deaths and injuries associated with the coal resource system
per megawatt delivered than with the crude oil or natural gas systems (Table 2).
TABLE 2
ANNUAL DEATHS AND INJURIES ASSOCIATED WITH COAL, CRUDE OIL, AND NATURAL
GAS FIRED ELECTRICITY SYSTEMS FOR A 1000 MEGAWATT POWER PLANT WITH A
LOAD FACTOR OF 0.75a
COAL CRUDE OIL
DEEP
Fatalities 4.00
Injuries 112.3
SURFACE ONSHORE OFFSHORE IMPORT
2.64 0.35 0.35 0.06
41.2 32.3 32.3 5.7
NATURAL GAS
TOTAL
0.20
18.3
a Reference 1
Accidents associated with coal-related activities can be generic hazards
(arising from participation in a particular activity in which the accident
rate is generally independent of the coal fuel cycle, e.g., heavy equipment
operation) or specifically coal-related hazards (a generic activity where ac-
cident rates for the coal fuel cycle are different from the industry-wide
average, e.g., underground coal mining).
Underground coal mining is a significantly more hazardous occupation than
underground mining of other substances and both are more hazardous than an
29
-------
all-industry average. The frequency of injuries in underground coal mines is
more than three times the average for selected industries and about one and a
half times the average for underground mining of other materials (Table 11).
The severity of underground coal mining injuries was almost eight times the
all-industry average and about 25% higher than underground mining per se
(excluding coal).
Mine roof and rib falls have accounted for 40 to 50% of annual coal mine
fatalities and haulage related accidents for about 15 to 20%. Gas or dust ex-
plosions can cause many deaths during a single incident but account for about
10% or less of the annual underground coal mining fatalities. Surface activ-
ities associated with underground coal mining and surface coal mining have
accounted for approximately 20% of annual coal mining fatalities during the
past 10 years. Significant health or environmental impacts also may result
from mine tailing dam collapse or from land subsidence associated with coal
extraction.3
Coal processing facilities (e.g., cleaning, sizing, drying operations)
are less hazardous than other elements in the coal fuel cycle. The accident
potential of coal conversion technologies which are being developed is not
known. However, since most liquefaction and gasification processes operate
at high temperature and pressure, accidents involving pressure vessel rupture
may be expected. These accidents may release heavy metals and carcinogens.-^
Other hazards associated with the coal fuel cycle are generic in nature.
Accidents occur during transportation and distribution of coal via rail, truck,
or ship.
Minor occupational accidents in the crude oil system are the most fre-
quent and cumulatively are the most costly in terms of numbers of injuries,
work days lost, and damage to equipment. Major accidents involve spillage,
fire, and explosion. These accidents occur at each stage in the energy cycle.
Oil spills occur on both land and water with the latter predominating in vol-
ume, frequency, and severity. Sources of spills include well blowout, tanker
barge accidents, pipeline ruptures and leaks, other transportation accidents,
and storage tank accidents. The greatest source of spilled oil is the tanker.
Most spills are small, although a few major incidents account for much of the
oil spilled. Coast Guard data for 1972 indicated that 19 spills accounted for
16.25 x 106 gallons whereas 10,126 spills accounted for 4.25 x 106 gallons.^
The major effect of an oil spill is damage to the environment including
flora, fauna, and land. Studies indicate that the toxicity of the oil spill
on marine species will vary in intensity depending upon the hydrocarbon frac-
tions present and the type and age of the organisms affected. Effects of
spilled oil range from minor disruptions of ecosystems with relatively rapid
recovery to major disruptions with slow recovery. Marine birds which encoun-
ter oil spills generally perish from loss of insulation, and ingestion of oil
from contaminated food, and preening. The severity of a spill depends on fac-
tors such as the volume spilled, the composition of the oil, wind, wave and
current conditions, timeliness and effectiveness of cleanup procedures, and
location of the spill.
30
-------
Fire and explosion may result from spills of oil and refined products,
both on and offshore. Major losses of life and property from such an event
occur in refineries and tank farms. The largest and most severe refinery
accident is the loss of hydrocrackers. Equipment losses from hydrocracker
accidents have averaged more than $1 million a year since 1970. Potential
economic losses as a result of storage tank fires and explosions are increas-
ing due to increased crude oil and petroleum product prices and use of larger
storage tanks and tank farms. Destruction of a single 500,000 barrel tank
(replacement cost ^ $2 million) filled with $10 a barrel oil can cause a $7
million loss.5
In the natural gas system, blowouts during drilling of exploratory and
production wells, release of sulfur compounds during processing, and failures
of pipelines account for the largest number of accidents. Sudden uncontrolled
release of natural gas may result in explosions and fires causing damage to
equipment and loss of life or injury to persons in the vicinity. The general
public is exposed to pipeline hazards because they traverse populated residen-
tial and commercial areas.
Pipeline distribution accounts for the largest number of injuries and man-
days lost in the natural gas system at 0.0138 injuries/1012 BTU and 0.324 man-
days lost/1012 BTU, respectively. Most pipeline failures can be attributed to
corrosion and damage by outside forces. Onshore extraction accounted for the
largest number of casualties at 8.1 x 10~5/1012 BTU.6 Other possible sources
of natural gas accidents are malfunction of explosive charges or radioactive
sensors during preliminary stages of exploration, failure of aboveground stor-
age tanks and underground storage cavities, and malfunction of peak shaving
plants. Accidents occurring during exploration are infrequent and not severe.
Failure of an aboveground storage tank entails the loss of large amounts of
gas and the potential for an explosion and large fire.
LNG is a hazardous material because of its cryogenic properties and
flammability. Contact with LNG will cause damage to human tissue. On vapori-
zation and exposure to source of ignition it will burn causing injury and loss
of a life to persons in the vicinity as well as extensive equipment damage.
It may also form a vapor cloud shich may travel great distances before igniting,
thus increasing the damage area. The worst accident to date involving LNG
occurred at a storage facility in Cleveland in 1944. The incident resulted in
130 deaths and $10 million damage. The probability of an accident occurring
due to the same cause is low because of development of materials resistant to
brittle failure at cryogenic temperatures.
Those LNG accidents which are considered to have the greatest potential
for damage are failure of a storage tank, an aircraft crash into a docked ship,
and a collision at sea involving an LNG tanker in which the contents of one or
two tanks are released. Minor accidents include release of refrigerants,
solids blocking the pipeline during or after the liquefaction process, on-
board malfunctions of tankers, leaks in transfer systems from tanker to shore,
and onshore maintenance.
31
-------
The safety of shipping and storing LNG is a matter of significant contro-
versy. Risk analyses have been performed for tanker collisions and failure of
storage tanks. Because no good statistical data base exists, several methods
of risk analysis are used. One approach is to extrapolate data from shipping
accident records. Another approach involves theoretical modelling using
event trees and probability estimates. Analytical and simulation approaches
model the performance of a system in terms of appropriate performance param-
eters such as ship velocities and structural characteristics. One study indi-
cated that the predicted fatalities would be 0.4 per year for tankers calling
at ports in the United States between 1980 and 1985.7
Risk analyses for three California terminal sites performed by the
Federal Power Commission (FPC), Science Applications, Inc. (SAI), and the El
Paso Alaska Company have been compared. All studies indicated that the risk
of a severe accident is low.8 The likelihood of an accident occurring was
estimated to be in the range of one accident per 10,000 years to one accident
per 100,000 years.
The estimates of risk were not identical. The variation in results may
be attributed to different approaches and initial assumptions. Examples of
this include whether rammings and groundings are as potentially harmful as
collisions, the number of tanks which will rupture, the probability of a plume
forming versus immediate ignition, assumptions concerning site and equipment
characteristics, and differences in ship casualty frequency modelling and
vapor cloud characteristics modelling.
Few data exist concerning accidents in the hydroelectric power system;
however, occupational injuries at hydroelectric power plants probably are less
frequent and less severe than at coal, oil, and nuclear fired plants. A sur-
vey for 1972 indicated that occupational injuries at hydroelectric power plants
occurred at about one half the frequency and with one tenth the severity of
the average for all electric generating plants. These results cannot be veri-
fied, however, because some of the original data are not available. Personnel
accidents are the most frequently occurring incidents. These accidents occur
during construction, maintenance, and normal operations.
The most severe hydroelectric accident which can occur is dam failure.
In the recent Teton Dam catastrophe, 9 people were killed, 8 were missing,
13,000 cattle were destroyed, and more than $500 million damage was caused.
Dam failures can also cause ecological damage and loss of food and water sup-
plies. Among the possible causes of dam failure are improper siting, improper
construction, erosion, forces exceeding design criteria, and sabotage.
Flooding of the area above the dam also can be serious. Flooding is
caused by heavy rainfall or thawing snow and ice and can be promoted by im-
proper construction practices.
Because there is only limited experience with geothermal resources, data
are sparse. The most severe anticipated accident would be a well blowout re-
leasing hot fluids and steam to the surface and adjacent geologic structures.
A blowout could cause injuries and damage to equipment. If natural gas were
32
-------
present, a fire also could result. A less severe accident would be a pipeline
leak or rupture caused by an earthquake, mechanical failure, human error, or
pressure buildup due to silica precipitation.
Geothermal resource development can also cause subsidence resulting in
damage to buildings and equipment. In lowland coastal areas (e.g., the Gulf
Coast), subsidence may cause flooding and salt water intrusion.
There are few data concerning accidents associated with oil shale re-
sources because the technology required to recover the fuel value of shale is
in the development stage. However, as with other energy systems, it is ex-
pected that the least severe accidents generally will occur most frequently.
It is unlikely that mining of oil shale will be as hazardous as mining
of coal. Roof collapse is less likely to occur for an equivalent sized room
because of the hardness of the shale. However, larger rooms are likely to
be used in oil shale mining. Explosions in the mine due to buildup of flam-
mable gases probably will not occur; however, explosive mixtures of dust may
form.
Explosions and fires may occur in the processing of shale. The incidence
of severe accidents is likely to be similar to that observed for other pro-
cesses involving use of hydrogen under high pressure.
In fossil fueled power plants, electricity currently is generated by
boiler fired or gas turbine plants. Major accidents in these systems result
from explosion in a boiler, turbine, or generator. Explosion and fire gener-
ally occur because of fuel mishandling or component failure. Power failure
may result from explosions from electrical malfunctions, breakdowns, circuit
overloads, or human error, and from accidents involving transmission and dis-
tribution of electricity.
There has not been a significant history of accidents in the various
solar energy conversion systems due to a lack of large scale development.
The meager past experience and the suppositions based upon proposed energy
conversion systems indicate that most expected accidents would be minor in
severity and generally infrequent. Economic losses would be substantial only
in a severe accident involving a large-scale central power facility, such as
the loss of a solar-thermal tower costing $15 million. Occupational injuries
for central solar power facilities have been estimated by the Jet Propulsion
Laboratory at between 0.5 and 1.6 man-days lost per MWe year.
Transmission and distribution accidents generally involve downed power
lines resulting from vehicle crashes, storms, falling trees, and other causes.
Occupational injuries frequently occur during the generation and transmission
of electricity. A National Safety Council study in 1972 showed that the fre-
quency rate for the electric utility industry was less than the average for
all reporting industries with 6.42 injuries per million manhours exposure.
However, the severity rate was higher with 1003 total days charged for inju-
ries per million manhours exposure compared to 655 per million manhours ex-
posure for all reporting industries for the period 1970-1972.1;L
33
-------
The accidents occuring in the transportation sector are those involving
motor vehicles, trains, planes, buses, farm vehicles, and marine craft. The
greatest number of fatalities result from motor vehicle accidents, followed
by farm vehicles, recreational boating, and general aviation. Most rail
accidents causing fatalities are attributable to servicing and other non-
train operations.
The greatest number of injuries in the transportation sector result from
motor vehicle accidents. There are approximately 100 times as many vehicular
injuries as there are injuries from total railroad operations and 20 times as
many as there are from farm vehicles.
Major accidents associated with industrial end use are explosion, fire,
and flooding. Industrial, commercial, and residential accidents are generally
minor though frequent. The most common accidents are those which occur in the
home; however, most are not energy related. Those which are energy end use
related involve appliances, heating and air conditioning, cooking, and the use
of electrical equipment.
Both natural and man-caused unscheduled events can cause injuries, deaths,
adverse effects on health, and destruction of equipment in all non-nuclear
energy resource and technology areas. Natural incidents include hurricanes,
tornadoes, floods, tsunamis, snow and ice storms, eqrthquakes, land subsidence,
avalanches and landslides, volcanic eruptions, and meteorite impacts. The
man-caused events include airplane and missile crashes, sabotage, terrorism,
and war activities.
The probability of external unscheduled events occurring and the damage
caused by them is highly site specific. Based on the frequency of the event
and the total national energy capacity threatened, hurricanes pose the most
serious potential threat of all the natural disasters because a large percent-
age of petroleum and natural gas is extracted and processed in hurricane-prone
areas. Tornadoes and earthquakes may be termed as a lesser but significant
threat to energy production systems. The consequences of a large meteorite
impacting at or near a major energy facility could be severe. The probability
of such impacts are low and highly site specific.
The destruction potential of war is great. Given a postulated "typical
nuclear attack", however, considerable energy production capacity (50% - 70%)
should survive. The greatest destruction will occur in the Northeast, along
the California coast, and at large industrial centers.
The threat of sabotage is serious within the energy industry. Saboteurs
can include enemy agents, dissident groups, dissatisfied employees, and the
mentally unstable. Industrial sabotage may also be performed by workers
wishing to prolong a job. Depending upon intent, the actual destructive ef-
fect of sabotage ranges from slight to severe. Data concerning the effect
of past sabotage activities and predictions regarding their future frequency
are not available.
34
-------
In general, the components most vulnerable to natural disasters and
sabotage are centralized control rooms, power lines, power switching centers,
and nodes in the pipeline networks. Many of these vulnerable points are in
remote, unguarded, or easily accessible areas. Thus a small but knowledge-
able group of saboteurs could cause a major interruption in the flow of fuel
or electricity to the ultimate consumer.
35
-------
SECTION 3.0
COAL
The use of coal as a fuel source for energy production has historically
occupied a major portion of the U.S. market. However, the increased use of
alternate fuel sources (e.g., oil, natural gas, uranium), coupled with adverse
environmental impacts related to coal extraction and utilization, has reduced
the national dependence on coal as a fuel source. In 1948, coal provided 48%
of the U.S. energy supply, while in 1971 it supplied only 18%.12 The major
consumers of United States coal are shown in Table 3.13 xhe declining domes-
tic reserves of alternate fuel sources, when combined with the uncertainties
associated with foreign supplies, has renewed interest in coal as the major
fuel source for the next few decades. It is estimated that the United States
possesses the largest percentage of the world's known recoverable coal
resources (Table 4).^
The increased use of coal as a fuel source will depend, in part, on reso-
lution of various environmental impacts resulting from coal extraction, con-
version, and end use. Included in this cost/risk/benefit analysis of various
energy systems is analysis of accidents occuring in the coal resource cycle.
The following sections discuss possible adverse health or environmental im-
pacts resulting from accidents associated with this cycle.
3.1 RESOURCE SYSTEM OVERVIEW
Of the available domestic reserves of coal, approximately 90% is found in
four geographical areas: Rocky Mountains, Northern Great Plains, Interior, and
Eastern regions. Although the characteristics of the coal deposits vary, the
coal resource system (excluding end use) can be divided into exploration,
extraction, processing, transportation and distribution, and land reclamation
energy cycle elements. Technology options within the extraction and process-
ing elements are shown in Figure 1.
3.1.1 Exploration
Thorough mapping of each coal seam is essential for planning the effi-
cient operation of a specific mine. Available mapping techniques include
review of geophysical and geochemical data, drilling and core samples of strata,
and seismic, gravimetric, or magnetic measurements. Despite the availability
of an array of exploratory tools, drilling remains the primary method for lo-
cating and mapping coal deposits.15 Exploration costs for surface mines
36
-------
Exploration
UNDERGROUND
EXTRACTION/RECLAMATION
Room and pillar
Longwall/short-wall
SURFACE EXTRACTION/
RECLAMATION
area strip
contour strip
auger strip
AUGER EXTRACTION/
RECLAMATION
PREPARATION/
BENEFICIATION
fuel
Liquefaction
-^•liquid fuel
(In situ)
GASIFICATION
(low, intermediate
energy content)
I
GASIFICATION UPGRADING
(high energy content)
Caseous fuel
Figure 1. Coal resource systems technology options.
-------
TABLE 3 CONSUMPTION OF BITUMINOUS AND
LIGNITE COAL, BY CONSUMER CLASS, 1974a
PERCENT OF
CONSUMER TOTAL CONSUMPTION
Electric power utilities 70.6
Bunker, lake vessel, and exports <0.01
Coke production
Beehive ovens 0.2
By-product ovens 16.0
Steel and rolling mills 1.1
Other manufacturing and 10.5
mining industries
Retail deliveries to other 1.6
consumers
100.0
a. Reference 13.
38
-------
TABLE 4. ESTIMATED COAL RESOURCES OF THE WORLD3
REGION
United States
Canada
Latin America
Europe
Africa
Oceania
U.S.S.R.
Peoples Republic
Asia, excl. USSR
RECOVERABLE
RESERVESb (%)
30.8
0.9
0.5
21.4
2.6
4.2
23.1
of China 13.5
, China 3.0
100.0
(651xl09 short tons)
TOTAL
RESOURCES0 (%)
27.2
1.0
0.3
5.6
0.6
1.9
53.1
9.3
1.0
100.0
(ll.SxlO12 short tons)
a. Reference 14.
b. That portion of total reserves currently exploitable under present
local economic condition and available technology.
c. Total amount of coal in the earth's crust which can be successfully
exploited in the foreseeable future.
39
-------
consume approximately 10% of the total capital investment,16 whereas explora-
tion and mapping of underground mines account for less than 1% of the total
capital invested.1''
3.1.2 Extraction
Three coal extraction techniques are used today: underground, surface,
and auger mining. Of the 604 million short tons of coal mined in 1974, 45.9%
was mined by underground techniques, 45.5% by strip mining, less than 1% by
auger techniques, and 7.9% by a combination of strip and auger methods.13
The choice of mining method depends on a number of considerations, including
seam depth and thickness, size of deposit, overburden characteristics, and
local geology.
The majority of underground mines in the United States use either room-
and-pillar techniques or longwall mining. The room-and-pillar method follows
the coal seam, leaving intermittent pillars of coal to support the mine roof.
Normal room-and-pillar mines remove about 45 to 50% of the actual coal seam.-*--*
The room-and-pillar technique can involve hand cutting and loading of coal,
machine cutting and loading, or a combination of both. Mechanical cutting and
loading is the preferred method; less than 0.2% of coal mined underground was
hand loaded in 1974.13
In longwall mining, two access tunnels are dug parallel to the coal seam
about 600 feet apart and a shearing machine is installed between them. The
machine then removes the coal in the seam and transports it from the working
face by conveyors installed in access passages. As the machine progresses
along the seam, the roof collapses behind the working face. In longwall min-
ing, roof collapse is anticipated, while in room-and-pillar mining, roof col-
lapse is accidental and often fatal. An ancillary technique to the longwall
technique is shortwall mining, where the working face is approximately 150
feet long, rather than 600 feet. Longwall mining normally recovers about 80%
of the coal in the seam.15
Longwall mining in the United States is increasing slowly; 3.8% of coal
mined underground in 1974 used the longwall method, compared to 3.1% similarly
mined in 1973. Longwall and room-and-pillar mining methods require similar
surface preparation (i.e., surface support facilities, initial access tunnels,
etc.).
In surface mining, coal seams are exposed by removing the overburden, and
excavated using several types of heavy equipment (i.e., mobile tractors or
dozers, shovels, draglines, or wheel excavators). Explosive fracturing of
overburden and coal seam preceeds removal. When mining occurs on uneven ter-
rain, the overburden is removed to create a flat bench, from which the coal is
subsequently removed (contour mining). On level terrain, trenches are cut in
the overburden to expose the coal seam, and as mining progresses, the spoils
are deposited into the previously mined area (area mining). Auger mining
techniques use large drills or augers to pull the coal out from a horizontal
coal seam. Auger techniques may be applied to the exposed coal after over-
burden removal.
40
-------
3.1.3 Transportation and Distribution
After removal from the coal seam, coal must be transported to a process-
ing facility (if required) and eventually to the consumer. Underground mining
operations generally use conveyors or rail shuttle cars to remove the coal
from the working mine. In 1974, about 9% of underground mined coal was moved
by conveyors, while 81% was removed by shuttle or rail cars.13 The remaining
coal was transported by truck or slurry pipeline. Surface mine haulage sys-
tems include large trucks or conveyor systems. The distribution of the coal
to ultimate consumers usually is achieved by rail (unit or mixed train),
truck, slurry pipeline, barge, or conveyor system. Table 5 indicates the
most common methods of shipment of coal in the United States.13
3.1.4 Processing/Beneficiation
Prior to consumption, mined coal is usually subjected to some degree of
cleaning or sizing. In certain instances, coal is converted into solid (coke),
liquid, or gaseous products prior to combustion. Table 6 presents the amount
of coal mined in 1974 which is subjected to cleaning or preparation steps
prior to conversion or combustion.
Approximately 15% of the total 1974 coal production was used to make
coke for metallurgical smelting operations.13 Coke is the residue remaining
after destructive distillation of coal. Of the total coke production in 1974,
approximately 1.5% was produced in beehive ovens (where the volatile consti-
tuents are not recovered) and the remainder in by-product coke ovens (where
the volatile components are recovered).
Coal also can be converted into liquid or gaseous products in order to
improve its quality prior to end-use. Gaseous fuels of low, intermediate, or
high energy content can be produced from coal. Low and intermediate energy
content gases are produced in a two-stage process involving preparation and
gasification; a third stage, upgrading, is required for the production of high
energy value gases.
In the production and upgrading of gaseous fuels from coal, the carbon-
to-hydrogen ratio is changed - the smaller the ratio, the higher the energy
content. Either hydrogen has to be added to the production gases or carbon
removed. Numerous gasification processes are described in Table 7, along
with process requirements and operating conditions.13»18 Several methods are
available for reducing the carbon-to-hydrogen ratio in the production of liq-
uid fuels; hydrogenation, pyrolysis, and catalytic conversion. Coal lique-
faction methods are described in Table 8, along with process operating param-
eters. 13»18 Most processes described in Table 7 and 8 are still in develop-
mental stages, so many of the operating characteristics remain undefined.
3.1.5 Reclamation
Although the large areal disturbances of surface mining are generally
more visible, both underground and surface mining operations require land
reclamation. The disposal of mine tailings can be accomplished by reinjec-
tion into spent mine shafts for underground mine operation, or by placement
41
-------
TABLE 5. BITUMINOUS COAL AND
LIGNITE SHIPMENT METHODS, 1974a
QUANTITY PERCENT
METHOD (THOUSANDS OF SHORT TONS) OF TOTAL
Railroad
Waterway
Truck
Mine-mouth consumption
Otherb
397,161
67,754
66,382
66,635
5,474
603,406
65.8
11.1
11.0
11.1
1.0
100.0
a. Reference 13.
b. Includes coal used at the mine for power and heat, coa.1 used by
mine employees, and coal shipped by slurry pipeline.
42
-------
TABLE 6. BITUMINOUS AND LIGNITE COAL PREPARATION, 1974a
PREPARATION METHOD
PERCENT OF TOTAL PRODUCTION
SUBJECT TO TREATMENT1"
Mechanical cleaning
Underground0
Surfaced
Augers6
Crushing and screening
No processing
43.9
64.9
29.5
7.9
48.7
7.4
100.O1
a. Reference 13.
b. Total 1974 production: 603,406 thousand short tons.
c. Total 1974 underground production: 277,309 thousand short tons.
d. Total 1974 strip mine production: 275,041 thousand short tons.
e. Total 1974 auger mine production: 51,056 thousand short tons
f. Six percent of all coal mined also was thermally dried.
43
-------
TABLE 7. SELECTED FEATURES OF GASIFICATION PROCESSES3
PROCESS
REACTOR TYPE
COAL REQUIREMENTS
BED TYPE
OPERATING PRESSURE
Low and intermediate
BTU gasification
Lurgi
Koppers-Totzek
Bureau of Mines
Westinghouse
Ash agglomerating
Applied Technology
Corp. (ATC)
In Situ gasification
High BTU gasification
Lurgi
HYGAS
(Steam oxygen or
electrothermal
BY-GAS
SYNTHANE
C02 acceptor
Gasifier
Gasifier
Gasifier
Gasifier
Gasifier
Gasifier
Gasifier
Hydrogasifier
Gasifier and
hydrogasifier
Gasifier,
devolatizer
Gasifier,
devolatizer
Noncaking coal;
sizing required (no fines)
Caking or noncaking;
pulverizing required
Caking or noncaking;
pulverizing required
Caking or noncaking;
pulverizing and drying
required
Caking or noncaking
pulverizing required
Pulverizing required
Channelized coal seam; sur-
face support facilities
Noncaking coal; sizing
required (no fines)
Sizing required (8 to 100
mesh); heating and slurry
Pulverized; liquid slurry
(to rank A)
All coals, 200 mesh; sizing
heating, and volatizing
required
Lignite or subbituminous;
sizing required
Modified fixed
Entrained sus-
pension
Modified fixed
Fluidized
Fluidized
Molten iron bath
Modified fixed
Fluidized
Entrained flow
Fluidized
Fluidized
300-450 PSI
Atmospheric
Atmospheric to
300 PSI
200-300 PSI
Pressurized
Unknown
300-500 PSI
1000 PSI
1000 PSI
1000 PSI
150 PSI
a. References 13, 18.
-------
TABLE 8. SELECTED FEATURES OF LIQUEFACTION PROCESSES'
Ul
PROCESS
Synthoil
H-Coal
Solvent Refined
Coal (SRC)
Consol Synthetic
Fuel (CSF)
COED (Coalcon)
TO S COAL
Fischer-Tropsch
Methanol
REACTION TYPE
Hydrogenation
Hydrogenation
Solvation,
Hydrogenation
Hydrogenation
Pyrolysis
Pyrolysis
Catalytic
conversion
Catalytic
conversion
COAL REQUIREMENTS
Pulverized, dried;
caking or noncaking
Pulverized, dried;
caking or noncaking
Pulverized, dried;
caking or noncaking
Pulverized, dried;
caking or noncaking
Pulverized, dried;
caking or noncaking
Pulverized, dried;
caking or noncaking
Depends on process
Depends on process
REACTOR TEMPERATURE (°F)
850
850
800
800
600-1600
970
-
unknown
OPERATING PRESSURE
2,000-4,000
2700 PSI
1000 PSI
1000 PSI
6-10 PSI
Atmospheric
330-360 PSI
unknown
PSI
a. References 13, 18.
-------
in strip mine trenches for surface mines. The surface storage of these tail-
ings in earthen dams prior to ultimate disposal is a common practice. Land
reclamation may also be required where raw or processed coal awaiting shipment
or consumption has been stored. In all instances, the movement or removal of
coal or spoils piles is accomplished with large earth-moving equipment common
in the construction trades. Planned or unplanned land subsidence may occur in
some cases, especially when underground mining or in situ gasification is used.
The actions required to ameliorate impacts of subsidence will depend on local
conditions.
3.2 ACCIDENT OVERVIEW
Accidents causing death, injury, or environmental damage can occur at any
step in the coal fuel cycle. In Table 1, representative accidents which may
occur have been identified. Severity and frequency estimates which appear in
Table 1 are based on the definitions presented in the introduction.
Quantitative estimates of injury and death rates appear in Tables 9 and
10.8,18,19 From these tables, it is apparent that the majority of occupation-
al death and injuries occur in the underground mining and transportation sec-
tors. It is noteworthy that the death and injury rates associated with under-
ground mining are significantly greater than those associated with surface
mining, either from their inherent severity, their frequency of occurrence,
or a combination of both. Potential accident situations are discussed in the
sections following.
3.2.1 Exploration
The human health or safety hazards associated with accidents during coal
exploration activities are few. All hazards identified are generic in nature,
with no specific coal exploration activity having an unusual accident type or
frequency. Although a significant amount of coal seam mapping must occur
prior to actual mining operations using drilling and coring equipment or ex-
plosives, no significant accidental health, safety, or ecological hazards are
present.
3.2.2 Extraction
Historical data indicate that coal mining is a dangerous occupation.
Table 11 compares the risks associated with underground coal mining with other
industrial activities.2 Coal mining related injuries were both the most fre-
quent and most severe. The threat to personal health or safety from mining
accidents depends upon the extraction technique used, the location of the mine,
the activity of the miner, the location of the individual inside the mine
(underground miners), the experience of the mining crew, the equipment used,
and safety precautions and procedures employed in the mine, and other factors.
Table 12 presents an historical perspective of mining fatalities by prin-
cipal cause over the years.3 The number and severity of injuries sustained at
coal mines in 1970 are presented in Table 13 and frequencies of occurrences
46
-------
TABLE 9. COMPARISON OF COAL RESOURCE TECHNOLOGY OPTIONS:
HUMAN HEALTH AND SAFETY FROM ACCIDENTS3
FUEL CYCLE
TECHNOLOGY OPTION
RESOURCE AREA
ACCIDENT RATE PER
1012 BTU EQUIVALENTS
FATALITIES INJURIES MAN-DAYS LOST
Surface coal
mining and
reclamation
Underground coal
mining and
reclamation
Northwest-area strip
Uncontrolled 0.0025 0.057
Controlled13'0 0.0025 0.057
Central-area strip
Uncontrolled 0.003 0.16
Controlledd 0.003 0.16
Northern Appalachian-
area strip
Uncontrolled 0.005 0.12
Controlled 0.005 0.12
Northern Appalachian
Contour
Uncontrolled 0.005 0.12
Controlledd 0.005 0.12
Central Appalachian
Auger
Uncontrolled 0.0001 0.094
Controlledd 0.0001 0.094
Central Appalachian
Contour
Uncontrolled 0.0018 0.164
Controlledd 0.0018 0.164
Southwest-area strip
Uncontrolled 0 0.059
Controlled0 0 0.059
Eastern coal-area strip
Uncontrolled NC NC
Controlledd 0.005 0.25
Western coal-area strip
Controlled 0.0065 0.31
Central
Room and Pillar
Uncontrolled 0.01 1.01
Controlled 0.01 1.01
Northern Appalachia
Room and Pillar
Uncontrolled U U
Controlled U U
1.
1,
3,
3,
.41
,41
,99
,99
2.49
2.49
2.49
2.49
1.9
1.9
3.30
3.30
0.678
0.678
NC
74
96
37.8
37.8
U
U
NC = not considered, U = unknown
a. Reference 18.
b. In controlled situation, land reclamation and water treatment considered part
of mine operation; in uncontrolled situation, they are not considered
c. Five years assumed for land reclamation.
d. Three years assumed for land reclamation
(continued)
47
-------
TABLE 9 (continued)
FUEL CYCLE
TECHNOLOGY OPTION
RESOURCE AREA
ACCIDENT RATE PER
1012 BTU EQUIVALENTS
FATALITIES INJURIES MAN-DAYS LOST
Coal preparation/
beneficiation
Coal gasification
(low, inter-
mediate energy
content)
Longwall
Uncontrolled U U
Controlled U U
Central Appalachia
Room and Pillar
Uncontrolled 0.022 0-955
Controlled 0.022 0.955
Breaking and Sizing
Northwest
Uncontrolled 0 0.003
Controlled 0 0.003
Central
Uncontrolled U U
Controlled U U
Northern Appalachia
Uncontrolled U U
Controlled U U
Central Appalachia
Uncontrolled U U
Controlled U U
Southwest
Uncontrolled 0 0
Controlled 0 0
Cleaning including washing*3
Uncontrolled 0.0026 0.0053
Controlled 0.0026 0.0053
Cleaning including washing0
Uncontrolled NC NC
Central coal
BuMines
Atmospheric U U
Pressurized U U
Koppers-Totzek U U
Northern Appalachian coal
BuMines
Atmospheric U U
Pressurized U U
Koppers-Totzek U U
Northwest coal
BuMines
Atmospheric U U
Pressurized U U
Koppers-Totzek U U
Lurgi U u
U
U
3.42
3.42
0.148
0.148
U
U
U
U
U
U
0
0
22.9
22.9
NC
U
U
U
U
U
U
U
U
u
u
(continued)
48
-------
TABLE 9 (continued)
FUEL CYCLE
TECHNOLOGY OPTION
RESOURCE AREA
ACCIDENT RATE PER
1012 BTU EQUIVALENTS
FATALITIES INJURIES MAN-DAYS LOST
Coal gasifica-
tion (high
energy content)
Transportation,
in-mine
Eastern Coal
Agglomerating
Fluidized Bedb NC
High-BTU gasification
Central Coal
HYGAS-steam-oxygen U
BIGAS U
Synthane U
Lurgi U
Northern Appalachian-coal
HYGAS-steam-oxygenb U
BIGAS U
Synthane U
Northwest coal
HYGAS-steam-oxygen U
BIGAS U
Synthane U
Lurgi U
COo acceptor U
Solid coal
Solvent refined coal
Northern Appalachian
area U
Central U
Eastern coal
Chemical cleaning NC
Liquefaction
Northwest area
CSF process U
SRC process U
Central area
CSF process U
SRC process U
SRC process0 NC
Northern Appalachian area
CSF process U
SRC process U
Northwest coal
Trucking
Uncontrolled 0
Controlled 0
Central coal
Trucking
Uncontrolled U
Controlled U
NC
U
U
NC
U
U
U
U
NC
U
U
0.027
0.027
U
U
NC
U
U
U
U
U
U
U
U
U
U
U
U
U
U
U
U
U
U
U
U
U
U
U
U
U
U
NC
U
U
U
U
NC
U
U
0.674
0.674
U
U
(continued)
49
-------
TABLE 9 (continued)
TECHNOLOGY OPTION
FUEL CYCLE RESOURCE AREA
Conveyor
Uncontrolled
Controlled
Mine rail
Uncontrolled
Controlled
Southwest coal
Trucking
Uncontrolled
Controlled
Transportation, Unit train
surface Northwest coal
distribution Central coal
Northern Appalachian coal
Central Appalachian coal
Southwest coal
Mixed or conventional train
Northwest coal
Central coal
Northern Appalachian coal
Central Appalachian coal
Slurry pipeline river barge
Central coal
Northern Appalachian coal
Central Appalachian coal
Trucking
Northwest coal
Central coal
Northern Appalachian coal
Central Appalachian coal
Conveyor
Central coal
Northern Appalachian coal
Central Appalachian coal
ACCIDENT RATES PER
1Q12 BTU EQUIVALENTS
FATALITIES INJURIES MAN-DAYS LOST
U
U
U
U
0
0
0.075
0.066
0.065
0.062
0.067
0.075
0.066
0.065
0.062
0.0019
0.0019
0.0019
0.032
0.032
0.032
0.032
0
0
0
U
U
U
U
0.015
0.014
0.599
0.876
0.856
0.767
0.0534
0.599
0.876
0.856
0.767
0.0032
0.0032
0.0032
0.692
0.692
0.692
0.692
0
0
0
U
U
U
U
0.171
0.171
55.6
81.3
79.6
71.4
49.6
55.6
81.3
79.6
71.4
0.243
0.243
0.243
45.4
45.4
45.4
45.4
0
0
0
50
-------
TABLE 10. ANNUAL DEATHS, INJURIES, AND WORK DAYS LOST FOR UNCONTROLLED COAL-FIRED ELECTRICITY
SYSTEMS ASSOCIATED WITH A 1,000-MEGAWATT POWERPLANT WITH A LOAD FACTOR OF 0.75a
OCCUPATIONAL EXTRACTION
HEALTH DEEP5 SURFACEb PROCESSING1* TRANSPORT0 'd CONVERSION6
Deaths 1.67 0.308 0.0238 2.30 0.012
Injuries 85 13.9 2.56 23.4 1.38
Workdays 4,678f 499 99.5 2,340 152.9
lost
TOTAL
TRANSMISSION DEEP SURFACE
NA 4.00 2.61
NA 112.3 41.2
NA 15,280 3,091
NA = Not Available
Ui
t-*
a. Reference 1.
b. References 3, 19.
c. Impacts of coal based on coal•transport exclusively by rail (ave. distance 300 mi.); annual coal supply
for a 1000 MWe plant is 0.1% of total national ton-mileage.
d. Assumed that average injury leads to loss of 100 workdays.
e. Assumed one-half the combined deaths and permanent injuries are assumed to be fatal injuries. Permanent
total disabilities are considered to represent 6,000 workdays lost, and other disabilities are estimated
at 100 days lost.
f. Excluding workdays lost attributed to complicated pneumoconiosis.
-------
TABLE 11. INJURY RATES IN SELECTED INDUSTRIES, 1973'
INDUSTRY
Automobile
Chemical
Machinery
Petroleum
Shipbuilding
Non-ferrous metals and products
Mining, surface6
Tobacco
Construction
Railroad equipment
Quarry6
Underground mining, except coal6
Underground coal mining6
All industries
FREQUENCY RATE^
1.60
4.25
5.81
6.73
7-08
9.31
9.75
12.03
13.59
14.23
17.67
25.26
35.44
10.55
SEVERITY RATEC
176 (110)d
397 (93)
331 (57)
690 (103)
653 (92)
712 (76)
1365 (140)
404 (34)
1544 (68)
1361 (96)
1825 (103)
4431 (175)
5154 (145)
654 (62)
a. Reference 2; Reported by member companies of the National Safety
Council. NSC members generally have better safety programs and
lower injury rates than non-member companies. Data are not com-
parable to Bureau of Labor Statistics (BLS) rates for less severe
injuries (requiring medical treatment but not involving days of
disability).
b. Disabling injuries per 10 man-hours.
c. Lost time (hours) per 10 man-hours.
d. Average days charged per case.
e. Based on data for 1972.
f. Rates not fully comparable from year to year due to reporting
inconsistencies.
52
-------
TABLE 12. FATALITIES IN ALL COAL MINES, BY PRINCIPAL CAUSE, 1906-1970a'b (Percent of Total)
On
FIVE-YEAR
PERIOD
1906-1910
1911-1915
1916-1920
1921-1925
1925-1930
1931-1935
1936-1940
FALLS OF
ROOF AND
FACE
44.5
47.0
48.1
47.3
51.1
55.1
52.4
UNDERGROUND WORKINGS
GAS OR
DUST EX- EXPLO-
HAULINGS PLOSIONS SIVES
12.2
15.3
18.4
16.5
17.0
16.7
17.1
18.0
14.5
8.8
15.3
13.5
6.5
9.8
7.6
5.9
6.0
5.0
4.0
3.1
3.2
ELEC-
TRICITY
2.2
3.4
3.3
3.6
3.9
4.3
3.7
ALL
OTHER
UNDER-
GROUND
5.9
4.6
4.7
4.6
4.5
5.8
6.2
TOTAL
UNDER-
GROUND
90.4
90.8
89.2
92.2
94.0
91.5
92.4
SHAFT
AND
SLOPE
2.9
2.4
2.2
1.7
1.1
1.4
1.3
TOTAL
93.3
93.2
91.4
93.9
95.1
92.8
93.7
SURFACE0
6.7
6.8
8.6
6.1
4.9
7.2
6.3
AVERAGE
NUMBER
FATALITIES
PER YEAR
2,657
2,517
2,420
2,215
2,235
1,240
1,265
a. Reference 3.
b. Beginning with 1941, the cause classification in this historical series was changed as follows: (1)
Fatalities in shafts and slopes have been included in "Underground workings" and are distributed accord-
ing to the specified causes; prior to 1941 these deaths were grouped and shown as "Shaft and slope."
(2) Fatalities from underground machinery accidents are grouped and shown separately; prior to 1941 such
deaths were included in "All other underground." (3) Fatalities from roof or face falls caused by haul-
age equipment or machinery knocking out roof-support have been included in the respective haulage or
machinery classifications; prior to 1941 such deaths were included in "Falls of roof and face." (4)
Fatalities from pressure bumps and bursts and from inrush of water or material are included in "All
other underground;" prior to 1941 such deaths were included in "Falls of roof and face."
c. Includes fatalities that occurred at all other work locations, including the associated surface works of
underground mines, strip mines, culm banks, dredges, and mechanical-cleaning plants; beginning with 1955,
includes fatalities in auger mines.
(continued)
-------
TABLE 12 (continued).
FIVE-YEAR
PERIOD
1941-1945
1946-1950
1951-1955
1956-1960
1961-1965
1966-1970
FALLS OF
ROOF AND
FACE
47.6
50.3
48.9
50.5
47.4
39.0
HAULAGE
20.1
20.0
19.0
14.5
15.7
13.8
GAS OR
DUST EX-
PLOSIONS
8.4
6.4
8.2
6.6
10.2
12.2
EXPLO-
SIVES
2.6
2.5
1.5
2.1
1.1
0.8
ELEC-
TRICITY
2.9
2.0
2.7
3.7
2.2
3.7
MACHINERY
3.3
2.8
3.6
3.9
5.5
6.9
ALL
OTHER
UNDER-
GROUND
6.3
4.4
3.6
5.8
4.0
4.1
TOTAL
91.2
88.4
87.5
87.1
86.1
80.5
SURFACE
8.8
11.6
12.5
12.9
13.9
19.5
AVERAGE
NUMBER
FATALITIES
PER YEAR
1,311
870
522
380
274
246
-------
TABLE 13. NUMBER, AVERAGE, AND SEVERITY OF INJURIES AT COAL MINES
IN THE UNITED STATES, BY PRINCIPAL CAUSES OF INJURY
AND GENERAL WORK LOCATION, 1970a
PRINCIPAL CAUSES OF INJURY
Underground mines :
Underground:
Falls of roof
Falls of face or rib
Pressure bumps or bursts
Inrush of water
Other falling objects
Falls of persons
Handling materials
Handtools
Stepping on objects
Striking or bumping
Haulage
Explosions (gas or dust)
Explosives
Electricity
Machinery
Suffocation
Mine fires
All other
Pneumoconiosis
Total
Shaft and slope:
Falls of roof
Falls of face or rib
Other falling objects
Falls of persons
Handling materials
Handtools
Stepping on objects
Striking or bumping
Haulage
Electricity
Machinery
All other
Total
Total, underground
INJURIES ,
FATAL
76
6
1
-
-
-
2
-
-
-
37
41
3
11
26
-
1
-
-
204
2
-
-
-
-
-
-
-
-
-
-
—
2
206
ALL MINES
NONFATAL
1,320
363
12
-
164
456
2,126
413
173
61
1,651
14
70
78
1,471
4
22
147
9
8,754
4
2
6
24
41
10
5
1
44
3
10
2
152
8,906
SEVERITY,13
ALL MINES
396
150
523
-
35
37
39
40
26
26
206
4,505
382
252
159
6
273
24
2,500
191
2,009
59
21
48
28
16
14
39
37
37
43
5
111
190
a. Reference 3.
b. Severity: number of days charged multiplied by 10" divided by
man-hours. All fatalities and permanent total disabilities have
a standard time-loss charge of 6,000 days.
(continued)
55
-------
TABLE 13 (continued)
PRINCIPAL CAUSES OF INJURY
Surface:
Falls or slides of coal or overburden
Other falling objects
Falls of persons
Handling materials
Handtools
Stepping on objects
Haulage
Explosives
Electricity
Machinery
Suffocation
Fires
All other
Total
Total, underground mines
Strip mines :
Falls or slides of coal or overburden
Other falling objects
Falls of persons
Handling materials
Handtools
Stepping on objects
Striking or bumping
Haulage
Explosives
Electricity
Machinery
Fires
All other
Total
Other surface mining :^
Falls or slides of coal or overburden
Other falling objects
Falls of persons
Handling materials
Handtools
Stepping on objects
Striking or bumping
Hauling
INJURIES ,
FATAL
2
—
—
—
—
—
4
—
5
3
—
—
—
14
220
7
1
—
2
—
—
—
4
1
2
11
1
—
29
2
—
—
—
—
—
—
1
ALL MINES
NONFATAL
—
14
95
228
50
13
125
1
19
30
1
1
48
625
9,531
33
46
276
360
93
29
6
207
8
21
158
16
93
1,346
8
2
23
21
11
2
1
12
SEVERITY, o
ALL MINES
6,000
20
43
28
66
6
264
7
1,266
731
32
14
14
180
189
1,069
164
26
68
13
11
9
159
700
539
463
398
15
162
1,231
41
24
^ *T
16
3
18
6
476
(continued)
56
-------
TABLE 13 (continued)
PRINCIPAL CAUSES OF INJURY
Machinery
Fires
All other
Total
Total, mining
Mechanical cleaning plants :
Falling objects
Falling of persons
Handling materials
Handtools
Stepping on objects
Striking or bumping
Haulage
Explosions (gas or dust)
Electricity
Machinery
Fires
All other
Pneumoconiosis
Total
Grand total
INJURIES ,
FATAL
1
—
—
4
253
—
—
1
—
—
—
3
1
—
1
—
—
1
7
260
ALL MINES
NONFATAL
27
2
3
112
10,989
15
118
155
35
5
5
133
4
9
29
6
49
—
563
11,552
SEVERITY, b
ALL MINES
307
49
11
244
186
210
37
82
16
22
26
193
1,222
10
253
13
12
6,000
119
183
57
-------
during the period 1966 through 1970 are shown in Table 14. It is noteworthy
that the average number of fatalities per five year period has declined sig-
nificantly since early in the century. Governmental regulations, mine safety
programs, union activities, and technological advances have all contributed
to this decrease.
Roof, rib, and face falls accounted for about 50% of yearly mine fatali-
ties until the mid 1960s and still are the most important cause of mine fatal-
ities. Roof and rib falls also accounted for a substantial part of the non-
fatal coal mining accidents. Although the number of persons hurt by a single
roof or rib fall is small, the high frequency with which they occur leads to
a high death and injury total. Causes of roof and rib failure are listed in
Table 15.3,20,21,19 Failure to comply with adopted conventional support sys-
tems or lack of a support plan was the leading cause of fatalities.
The second greatest hazard in underground mining involved mining equip-
ment operation. An engineering safety analysis identified the need for stand-
ardization of controls for mining equipment, noting that the positioning and
responsiveness of equipment controls varied with manufacturer.22 TWO causes
of fatalities are often cited by members of the mining industry—lack of
mining experience and age (youth being a surrogate for inexperience). In a
recent study of these two variables as they related to underground coal mine
fatalities, age was not found to be a significant factor, and mining exper-
ience, per se, was a weak factor compared to job task experience.23
Although explosions and fire are often the most newsworthy accidents
associated with underground mining, only 10-12% of the annual mining fatalities
were caused by these disasters from 1960 to 1970. In fact, the number of
miners killed in mine explosions has been steadily declining. The apparent
reversal in this trend from 1960 to 1970 (see Table 12) can be attributed to
several large explosions in 1963, 1968, and 1970. Table 16 provides details
of major underground mine disasters since 1961.3
As noted above, the frequency of injury and death at surface mines is
significantly less than at underground mines. However, the number of accidents
occurring at surface mines has been increasing because of increasing reliance
on surface mining techniques. The relationship between surface mining activi-
ties and fatal and non-fatal injuries is shown in Table 17.^4
3.2.3 Processing/Beneficiation
The processing of coal can be divided into four categories: preparation
(cleaning, sizing, drying, etc.), coking, liquefaction, and gasification.
Operational histories exist for the first two categories, while the last two
generally are in various stages of development.
Mechanical coal cleaning facilities have been in operation for years.
Table 13 provides some injury data for mechanical cleaning plants.3 There are
few fatalities at these plants. A majority of the non-fatal accidents are as-
sociated with haulage, materials handling, and falls. The potential for explo-
sions involving dust or volatile components exists, but is not a significant
58
-------
TABLE 14. STATISTICS ON INJURIES AND INJURY RATES
AT COAL MINES IN THE UNITED STATES, 1966-1970a'b
Ui
vo
ALL COAL MINES
Number of injuries:
Fatal
Nonfatal
Total
Injury rates:
Frequency per million
man-hours :
Fatal
Nonfatal
Frequency per million tons :
Fatal
Nonfatal
Severity per million
man-hours :
Fatal
Nonfatal
Severity per million tons:
Fatal
Nonfatal
Average severity, days lost
per injury:
Permanent partial disability
Temporary total disability
All injuries
1966
233
10,446
10,679
0.96
42.85
0.43
19.15
5,735
1,973
2,563
882
620
32
176
1967
222
10,115
10,337
0.92
41.84
0.40
18.04
5,509
2,190
2,375
944
621
33
180
YEAR
1968
311
9,639
9,950
1.33
41.12
0.56
17.42
7,960
2,553
3,372
1,081
763
37
248
1969
203
9,917
10,120
0.85
41.76
0.36
17.42
5,129
2,230
2,140
931
600
36
173
1970
260
11,552
11,812
1.00
44.40
0.43
18.94
5,996
2,312
2,558
986
756
33
183
a Reference 3.
b Data exclude office workers.
-------
TABLE 15. PRIMARY DEFICIENCIES RELATED TO ROOF AND
RIB FALL FATALITIES; UNDERGROUND COAL MINES, 1967-1970a
PRIMARY DEFICIENCY
Failure to take down or secure loose roof or rib
Failure to examine roof or incorrect analysis of
examination
Failure to comply with adopted conventional support
system or lack of support plan
Failure to use temporary support
Failure to follow rules or instructions, to heed
warnings, or to take ordinary precautions
Failure to abandon workings when known to be
imminently dangerous
Failure to follow or failure of roof-bolting plans
Failure of mining system
Failure to replace dislodged or removed supports
Failure of conventional support system
Other
Total deaths
1967
5
11
34
12
1
3
6
10
-
12
-
94
FATALITIES
1968
12
16
35
17
2
3
-
8
-
5
-
98
PER YEAR
1969
10
18
20
11
6
3
6
2
-
-
1
77
1970
4
18
25
14
6
5
3
7
2
-
-
84
Reference 3, 19, 20, 21.
-------
TABLE 16. MAJOR DISASTERS AT COAL MINES IN THE UNITED STATES. 1961-70a>b
DATE
1961:
1962:
1963:
1964:
1965:
1966:
1967:
1968:
1969:
1970:
1971:
1972:
Mar. 2
Jan. 10
Dec. 6
Apr. 25
Dec. 16
No major disaster
May 24
Oct. 16
Dec. 28
June 1
July 23
No major disaster
Aug. 7
Nov. 20
No major disaster
Dec. 30
No major disaster
July 22
Dec. 16
MINE
Viking
No. 2
Robena No. 3
Compass No. 2
No. 2
No. 2A
Mars. No. 2
Dutch Creek
Dora No. 2
Siltex
River Queen No. 1
Consol. No. 9
Nos. 15 and 16
Blackville No. 1
Itmann No. 3
LOCATION OF MINE
Terre Haute, Ind.
Herrin, 111.
Carmichaels, Pa.
Dola, W. Va.
Helper, Utah
Robbins, Tenn.
Wilsonburg, W. Va.
Redstone, Colo.
Dora, Pa.
Mt. Hope, W. Va.
Greenville , Ky .
Farmington, W. Va.
Hyden , Ky .
Blackville, W. Va.
Itmann, W. Va.
CAUSE
Explosion
Explosion
Explosion
Explosion
Explosion
Explosion
Mine fire
Explosion
Suffocation
Explosion
Explosion
Explosion
Explosion
Mine fire
Explosion
NUMBER
KILLED
22
11
37
22
9
5
7
9
5
7
9
78
38
9
5
a A single accident in which 5 men or more are killed.
b Reference 3.
(continued'*
-------
1976: March 9
March 11
TABLE 16 (continued)
DATE
1973:
1974:
1975:
MINE
No major disaster
No major disaster
No major disaster
NUMBER
LOCATION OF MINE CAUSE KILLED
Scotia mine
Scotia mine
Oven Fork, Ky.
Oven Fork, Ky.
Explosion
Explosion
15
8
-------
TABLE 17. FATAL AND NONFATAL INJURIES
SUSTAINED DURING SURFACE STRIP MINING
ACTIVITIES. 1966-19723
SURFACE MINING ACTIVITY
Fall of highwall
Haulage Truck Operations
Dozer operations
Front-end loader operations
Shovel and dragline operations
Slips and falls
Railroad car dropping operation
Conveyors in preparation plants
Electrical systems
Handling materials
Machinery maintenance
PERCENT OF
FATAL (%)
12.9
12.3
6.4
11.7
9.4
6.4
9.4
8.1
12.3
3.5
7.6
100.0
TOTAL ACCIDENTS, 1966-72
NON-FATAL (%)
1.0
5.8
2.1
1.8
2.7
32.3
1.5
3.0
2.1
23.5
24.1
99.9
a Reference 24.
63
-------
problem. The potential for fires or explosions from tailings resulting from
coal preparation and treatment will be discussed in the reclamation section.
The coking of coal has been carried out since the turn of the century.
Although there have been incidents of coke ovens or batteries exploding, the
frequency of occurrence has been low. An accidental shift in operating con-
ditions may cause an increase in fugitive emissions, some of which are known
or suspected carcinogens.
Advanced processes involving coal liquefaction or gasification generally
are in the developmental stage and no operating data exist to quantify acci-
dent situations. Some hypothetical incidents can be envisioned, but severity
of impact or frequency of occurrence cannot be specified. It is evident from
Tables 7 and 8 that the majority of coal conversion processes involve reaction
vessels operating at elevated temperatures and pressures.13,18 The failure of
any portion of the system may cause fires or explosions. The rapid, explosive
release of coal conversion reactants or by-products may include polycyclic
aromatic hydrocarbons, or combustible gases. Some process steps such as rapid
depressurization or volatilization are inherently dangerous. The use of pul-
verized coal as feedstock may provide an abrasive situation conducive to com-
ponent failure. Coal feedstock systems may present fire hazards. Although
accident data for these processes are not available, accident histories in
other industries, e.g., oil refining, may be similar.
One gasification process is different from the rest: in situ gasification.
The technology for in situ gasification is currently under development, but
some severe problems have been recognized. The inability to control combustion
in the coal seam may lead to explosive conditions. Explosive fracturing of
the seam, required for controlled combustion, is a significant potential
hazard. Should improper fracturing of the coal seam occur, reaction gases
may escape and uncontrolled combustion may occur. Gases fed from the surface
under pressure to support combustion present a potential fire hazard. Induced
seismicity may result from gasification activities. Land subsidence after
gasification is almost certain to occur. Unfortunately, insufficient informa-
tion is currently available to quantify these risks.
3.2.4 Transportation and Distribution
From Tables 12, 13, and 17, it is evident that coal haulage in coal min-
ing operations currently accounts for between 10 and 15% of mining fatal-
ities. 3»24 in underground mining operations, the hauling subfunction is the
most dangerous; that subfunction accounted for about 72% of total haulage acci-
dents, and for 61% of all haulage-related man-days lost (of which 56% were
compensable injuries). Loading subfunction accidents are low in frequency
(about 10% of haulage-related accidents) but severe (27% of total function
man days were compensable).25 A worker at the controls of a vehicle is in a
more dangerous location than a worker performing supporting tasks. Workers
at the controls were involved in 82% of the total haulage related accidents,
and accounted for 69% of the total man-days lost.25
Rail systems transport approximately 66% of coal produced to the consumer,
22% in unit trains, and 44% via mixed trains. Coal shipments represented 27%
64
-------
of total rail freight in 1969 and 1970.18'26 Estimated rail accident rates
involving some form of coal transport are 0.06 fatal injuries and 0..585 non-
fatal accidents (with 59.5 man-days lost per occurrence) for every 1012 BTU
equivalent tons shipped.18 Comparable analyses indicate slurry pipeline
accident rates of 0.0019 fatalities and 0.0032 non-fatal injuries (with 0.24
man-days lost per occurrence) for 1012 BTU transported. Trucking rates are
estimated at 0.032 fatalities and 0.69 non-fatal injuries (with 45.4 man-days
lost) for 1012 BTU energy shipped. These accident projections are based on
national statistics for motor freight carriers where assumed coal shipment
accident rates are represented as the ratio of total coal shipped to total
freight carrier tonnage for the time span in question. This assumes that
the transport of coal presents no additional accident potential than the
generic rate for all motor carrier transports. This assumption may not hold
in those regions (e.g., Appalachia) where steep, unimproved roads may exist.
3.2.5 Land Reclamation
Land reclamation procedures during or after coal mining operations vary
with the type and location of individual mining operations. Surface mining
techniques (area and contour) and underground mining each require solution of
special reclamation problems. Auger-mining represents a small portion of
total coal production, and refuse generated by auger mining can be handled
using the techniques practiced with underground or surface (strip) mining
activities.
Surface or underground mining and associated coal preparation and treat-
ment generate large quantities of tailings or spoils. The movement of these
piles with large earth moving equipment provides accident situations similar
to those in other construction areas (e.g., surface or underground mine site
preparation). Human health and safety would not be expected to be substan-
tially different from other earth moving operations.
In underground mining operations, it has been estimated that about 2.46
tons of solid waste are produced per 1012 BTU equivalent tons extracted, with
about 4.38 acres of land undermined (assuming an average seam height of 5.3
foot).18 For underground auger mining, about 5.5 acres are undermined for
each 1012 BTU extracted. Of the area undermined, approximately 25% will sub-
side.18 Since subsidence affects a surface area greater than the area under-
mined, the estimates of total land subsidence resulting from mine subsidence
in an average 600 foot deep mine is 10.2 acres per 1012 BTU equivalent tons
extracted.* If the average life of a sizeable underground mining operation
(room-and-pillar) is about 25 years, the time average land impact from sub-
sidence is 127 acre-years per 1012 BTU extracted.18 The time averaged land
impact from subsidence in an auger mine is 51.1 acre-year per 10 BTU. For
longwall mining, where the mine roof collapses as mining progresses, it is
assumed that subsidence occurs over the entire area undermined. The time
averaged land impact from subsidence is estimated to be 280 acre-years per
. i 2
*Assuming 25% subsidence rate of the undermined areas of 4.36 acres/10
Btus in a 600 foot deep mine.18
65
-------
1012 BTU, over a 25 year mine life. Although rates of surface subsidence
from mining activities can be estimated, the severity of impact depends upon
the geographic location of the mine and facilities and resources in the af-
fected area. In populated areas (e.g., Scranton, Pa.), subsidence has proven
to be a severe problem, but in western coal tracts, subsidence may have no
long term economic impact. However, disruption of western aquifers may re-
sult from either mining or subsidence.
Refuse piles from coal mining or preparation activities have often been
used as earthen dams. Since the 1972 Buffalo Creek, West Virginia disaster,
where a refuse pile dam collapsed killing 125 people, more emphasis has been
placed on coal refuse piles as potential hazards. A preliminary assessment by
the Bureau of Mines has indicated that more than 100 coal refuse impoundments
in the Eastern regions present potential hazards.^
Not only can the refuse piles impound water, but they also may catch
fire. Numerous smoldering fires have been reported. These fires can emit
dense fumes and are difficult to extinguish. The remains after a refuse
pile fire, or "red dog," can present disposal problems.
66
-------
SECTION 4.0
CRUDE OIL
4.1 RESOURCE SYSTEM OVERVIEW
Oil was discovered in the United States in the middle of the 19th cen-
tury. Until 1948, the United States was an exporter. The United States had
a total proven reserve at the end of 1973 of 35.3 x 10^ barrels (bbl) of crude
oil.28,29 By the end of 1975, proven reserves dropped to 32.7 x 10$ bbl with
an indicated additional reserve from known reservoirs of 5 x 10^ barrels.
These figures include the Gulf of Mexico.29 The 1974 estimate of petroleum
liquid resources in the United States was put at between 2 to 4 x 10H bbl by
the United States Geological Survey (USGS), while oil company estimates were
only 8.8 x 10$ bbl.30
There are several energy cycle elements involved in the discovery and
transformation of crude oil into a usable fuel. The elements in this energy
cycle are shown in Figure 2.
4.1.1 Exploration
Exploration is undertaken to locate geologic formations which may be po-
tential oil reservoirs. This is generally a three phase effort which includes
regional surveys, detailed surveys, and exploratory drilling.
Regional surveys, often conducted by plane or boat using a magnetometer,
are used to identify promising geologic conditions where oil may be located.
Detailed surveys then examine promising areas using seismic techniques and
core drilling. The seismic technique uses an energy source to generate sound
waves, which are detected and recorded on magnetic tape for data processing
into cross-sectional maps of the subsurface geologic formations. Energy
sources used include explosives in onshore exploration, propane/oxygen detona-
tions in offshore exploration, or high-powered frequency oscillators in either
location.
Core drilling for oil is employed where the situation warrants, using a
rotary drill connected to a number of lengths of pipe called a drill string.
The cuttings are analyzed to characterize the formation being penetrated and
determine the presence of oil. The drilling technique uses a "drilling mud,"
a waterbased slurry of chrome lignosulfate of varying density, to remove drill
cuttings and to maintain hydrostatic pressure in the hole to prevent a blow-
out (unconstrained flow of oil or gas). The drilling mud is circulated through
the drill bit and back out the space between the drill pipe and the bore hole.
67
-------
EXPLORATION
EXTRACTION
00
ONSHORE
Well Completion
Field Processing
Improved Recovery
TRANSPORTATION
INTERMEDIATE
STORAGE
PIPELINE
RR
' TANKCAR
• BARGE
-TANKSHIP
OFFSHORE TOLL
OCS
Well Completion
Field Processing
t
PIPELINE
FOREIGN
CRUDE
and
semi-refined
products
TANKSHIP
PROCESSING
STORAGE
PRODUCT
TRANSFER
REFINERY
Products:
Gasoline
Jet fuel
Heating oil
Lubricants
Petro-chem.
feedstock
£ E«- 1
V +-—
| INTERMEDIATE
j STORAGE
w
^
H
I— PIPELINE —1 I
L— BARGE 1
1 TANKER 1
„ _, -RP TANT^AR
JANK
TRUCK
* BARGE
TANKCAR
PIPELINE
END USE
Figure 2. The crude oil cycle.
-------
Offshore exploratory drilling requires a platform to support the drill
rig and other equipment. The types of platform used are the permanently
mounted jack-up, the moored barge, the dynamically positioned and moored
drillship, and the dynamically positioned and moored semi-submersible.
Offshore exploratory drilling has been done mostly by mobile drilling
units. A rig survey31 for 1973 indicated 222 active rigs, 13 inactive, and
132 under construction. Of the active rigs, 127 were bottom supported. The
number of floaters has been increasing, with 71% of those under construction
being of this type. Presently 43% of the active rigs are floaters. It has
been estimated that over the next 10 years the size of the mobile rig fleet
will increase to between 250 and 400 units. It was estimated that approxi-
mately 520 mobile and platform mounted rigs would drill 200 wells during
1975.31
4.1.2 Extraction
When exploratory drilling indicates a discovery, the well is tested to
determine the rate of oil flow and the size of the reservoir. If commercial
production is warranted, a number of wells are drilled to drain the reservoir.
Field processing equipment and pipelines are installed and positive controlled
flow is maintained.
Development drilling uses the same techniques as exploratory drilling ex-
cept that it is carried out with greater accuracy and precision. Offshore
wells on the west and Gulf Coasts of the United States are drilled from fixed
platforms. The wells are drilled into the reservoir, with cemented casings
set and pipe installed to carry the oil to the surface. Offshore wells, and
onshore wells in danger of wellhead damage (or in seismically active areas),
have a downhole safety valve placed either near the bottom of the production
pipe string or at least 100 feet below the sea floor. This valve is designed
to close when the flow through it exceeds a certain limit. Wellhead valve
controls, called Christmas Trees, are either pressure activated or operator
controlled and control the flow rate from the well. Some offshore wells re-
cently have been completed on the ocean floor rather than on a platform.
The oil removed from the well is processed to remove natural gas, produc-
tion water (brine), sand, and other impurities. Natural gas can be separated
using a gravity separator. The gas is transported by pipeline and sold, used
as a lift in recovery of oil, or burned in a flare. Approximately 3.4% of all
gas produced offshore is burned in a flare.32
Brine is normally separated by gravity. If it is emulsified in the oil
other processing methods are required using heaters and surfactants. Addi-
tional treatment of brines using these methods and filters is required before
disposal offshore. If possible, the sand produced is kept in the bottom of
the well by means of chemical binders or screens. If some should surface, it
is removed in the separators.
When the natural flow diminishes, additional oil can be retrieved using
improved recovery techniques. One of these techniques is waterflooding or
secondary recovery. This method involved pumping water down selected wells in
69
-------
a field to force up oil in other wells. All other improved recovery tech-
niques are tertiary methods. The five categories of tertiary recovery tech-
niques are polymer flooding, surfactants, miscible recovery processes, immis-
cible gases, and thermal recovery. Each involves adding a large quantity of
a liquid or gas to the reservoir to improve the flow of oil.
4.1.3 Processing
A petroleum refinery converts crude oil into various products by means of
a combination of processes. These processes include distillation, sulfur re-
moval, cracking and reforming. Principal products of United States refineries
are gasoline, jet fuels and kerosene, diesel and other fuels, lubricants, waxes
and solvents, petrochemical feedstocks, and asphalt. The products and their
proportions vary with refinery design, location, and time of the year. Feed-
stocks also can include natural gas liquids or synthetic crude from oil shale
or coal liquefaction.
Crude oil is classified by density, sulfur content, and other impurities
to meet the requirements of a particular refinery for efficient processing.
Blending of various crudes is often necessary to achieve this.
The first step in refining oil is distillation. Distillation involves
heating crude oil and separating components according to their boiling points.
For example, gasoline boils at less than 250°F, while residual fuel oils boil
at more than 900°F. The residual from distillation at atmospheric pressure is
then distilled under vacuum to achieve further separation. The kinds of prod-
ucts produced will depend on the feedstock and the operating conditions of the
distillation column.
Sulfur is removed by hydrodesulfurization. This process involves the
reaction of hydrogen at high pressure (300-1000 psi) with the sulfur contain-
ing liquid at high temperatures (600-800°F) in the presence of a catalyst to
form hydrogen sulfide (I^S). A fractionation column is then used to separate
the cleaned oil from the H2S. The H2S gas in the system ("sour gas") is cir-
culated in a packed column with an amine solvent, which absorbs the H^S. The
solvent is regenerated, and the I^S is processed to recover elemental sulfur.
Distillation alone may not produce products in the proper proportion to
satisfy market demands. Thus further processing by cracking or reforming
usually is necessary.
Cracking breaks large molecules in the feedstock into smaller molecules
with a higher energy content. Catalytic cracking and hydrocracking are the
two processes currently used in refineries. Catalytic cracking using cata-
lysts of synthetic aluminum on solica in a fluidized bed design is the more
widely used process. Hydrocracking involves reactions under high pressure
(2000-2300 psi) and temperature (approximately 800°F) in the presence of hy-
drogen and a catalyst in a fixed-bed reactor design. There are fewer of these
types of units because of high cost.
70
-------
Reforming, alkylation, and isomerization processes rearrange the molecular
structure of oil components to form high octane gasoline. All three methods
use catalysts. Reforming uses platinum or rhenium in a hydrogen atmosphere at
pressures of 100-200 psi and temperatures of 800-900°F. Alkylation uses acid
catalysts, while isomerization uses platinum oxide at a temperature of 320°F
and a pressure of 400 psi.
In addition to these process units, a refinery has a number of support
facilities. These include stack gas cleaning equipment, wastewater treatment
facilities, and facilities for generating hydrogen, steam, and electricity.
4.1.4 Transportation and Storage
Crude oil and petroleum products are transported by various means to re-
fineries and markets. Among these are tank trucks, railroad tank cars, tankers,
barges, and pipelines. The choice of mode of transportation depends on the
distance to be traveled, the product to be transported, the availability of
the transport system, and other factors. Data on the percentages of crude and
refined products moved by the various modes of transportation is shown in
Table 18.33
Tank trucks carry small quantities for distances less than 500 miles.
Railroad tank cars are competitive with tank trucks for distances greater than
several hundred miles and for tank car quantities. Tankers and barges are
used for long-distance, large volume transportation via water routes, but are
limited by available port facilities. Barges and tankers used for intra-
coastal shipping are generally limited to 125,000 ton tankers, being built for
transporting Alaskan crude. Pipelines are competitive with waterborne and
railroad transportation.
Pipelines are made of welded steel and are as large as 48 inches in diam-
eter. Because of pressure drops along the line, pumping stations are installed
every 50 to 150 miles. These stations use centrifugal pumps powered by elec-
tric motors or diesel engines. They are generally unmanned and remote con-
trolled; however, periodic aerial inspections are made to detect leaks. Off-
shore pipelines are installed from pipelaying barges or ships and are placed
in the bottom by air-blast trenching.
Crude and refined petroleum is stored in tanks ranging in capacity from
several thousand barrels to 1 million barrels. Storage tanks often are
grouped together in tank farms. Two types of storage tanks are used; the fixed
cone roof and the internal floating roof. In the latter the roof moves with
the internal level of liquid. Pipelines connect tank farms to loading facili-
ties for transportation.
4.2 ACCIDENT OVERVIEW
Accidents in the crude oil resource system can occur at many points in
the technology chain. These accidents involve equipment failure, human error,
and natural phenomena. The consequences of these accidents include ecological
71
-------
TABLE 18. PERCENTAGE OF CRUDE PETROLEUM AND
PETROLEUM PRODUCTS MOVED BY METHOD OF TRANSPORTATION, 1938-1974.a
PERCENTAGE OF TOTAL TONNAGE SHIPPED
,INES
RODUCTb
6.35
11.36
20.48
30.41
31.12
32.39
32.74
33.54
WATER
CRUDE
25.58
23.26
16.90
18.62
18.90
16.10
14.13
13.47
CARRIERS
PRODUCT
52.65
44.70
37.51
25.69
26.75
26.92
25.78
25.84
MOTOR
CRUDE
1.17
3.86
6.45
7.11
6.65
7.92
8.68
11.29
CARRIERS
PRODUCT
10.59
29.85
36.76
41.35
39.72
38.55
39.31
38.48
RAILROAD
CRUDE
2.24
4.40
0.30
0.19
0.15
0.23
0.30
0.43
PRODUCT
30.41
14.09
5.25
2.55
2.41
2.14
2.17
2.17
1938 71.01
1948 68.48
1958 76.35
1968 74.08
1970 74.30
1972 75.75
1973 76.89
1974 74.81
Reference 33.
Product pipelines move only light petroleum products.
72
-------
damage, property damage, and injury or death. Representative accidents and
their frequency and severity are shown in Table 1.
Although small spills account for most incidents, the cumulative amount
of oil released is large. The most frequent accidents are those involving
drilling rigs and trucks; however, refining accidents account for approxi-
mately one-third of the occupational injuries in the petroleum industry.34
Occupational injuries and illnesses for 1975 for the exploration and produc-
tion, drilling, refining, and pipeline and marine transportation of crude
oil and products is shown in Table 19.34
A summary of fatalities for 1974 and 1975 for 137 oil and gas companies
are shown in Table 20.35 These data apply to on-duty accidents involving
employees and do not cover those accidents affecting non-employees and private
property.
Occupational health data are summarized on an energy equivalent basis in
Table 21 and in terms of supplying a 1000 MW power plant with fuel for one
year in Table 22. i.15
4.2.1 Oil Spills
4.2.1.1 Occurrence and -Risk—
Although oil spills occur both on land and water, only those on water
tend to be major in terms of environmental impact. This has prompted govern-
mental regulations to prevent spills from occurring, and to contain them and
clean them up when they do occur.
The Coast Guard, the Environmental Protection Agency (EPA), and the United
States Geological Survey (USGS) share responsibility in this area. The Coast
Guard has responsibility for all transportation related facilities and for all
coastal, Ohio River, and Great Lakes spill incidents. EPA has responsibility
for all onshore and offshore facilities not related to transportation. The
U.S. Geological Survey is responsible for regulations for safe drilling and
production practices of oil and gas activities on the outer continental shelf
and inland spills.
Detailed reporting systems have been established by all three agencies to
collect data of sufficient quantity, quality, and depth to permit trend and
risk analyses. The Coast Guard established the Pollution Incident Reporting
System (PIRS) in 1971 and the EPA began the Spill Prevention, Control and
Countermeasure file (SPCC) in 1975. Information gathered by these systems can
be retrieved in terms of source of spill, location of spill, causes, type of
oil, and volume of discharge.
The USGS reporting system consists of two parts—inland and offshore. The
inland monitoring system was started in 1974 and covers only those spills which
occur on federal lands. Although there are field inspectors for inland spills,
incidents are reported on the honor system.
73
-------
TABLE 19. REPORT OF OCCUPATIONAL INJURIES AND ILLNESSES FOR THE YEAR 1975
COVERING OPERATIONS SUBJECT TO OSIIA RECORDKEEPING REQUIREMENTS ONLY3
FUNCTION
Exploration & Production
Drilling0
Refining
Pipeline-Crude & Production6
Marine
Total
NO. OF EMPLOYEES
54,598
1,918
81,698
15,083
3,108
156,405
RECORDABLE CASES
INJURIES ILLNESSES
2,466
515
5,594
754
173
9,502
36
0
321
19
0
376
TOTAL
2,502
515
5,915
733
173
9,878
FATALITIES
9
2
3
6
2
22
TOTAL LOST
WORK DAY
CASES
763
196
1,846
201
66
2,872
NON-FATAL
CASES
W/0 LOST
WORK DAYS
1,730
317
4,066
5.66
105
6,784
a. Reference 34.
b. Geophysical, seismographic, and geological operations, including incidental administrative work. Produc-
tion operations, on- or off-shore, including maintenance and servicing of production properties and trans-
portation of crews and materials.
c. Operation of on- and off-shore rigs including transportation, set-up, and dismantling.
d. Refining of crude oil to produce products including petroleum chemicals if the operation is an integral
part of the refining operation.
e. Gathering systems and trunk line transportation of crude oil; pipeline transportation of refined and semi-
refined products.
f. Ocean, coast, and inland tankship and barge operations as well as land based marine department operations,
-------
TABLE 19. (Continued)
Exploration & Production^1
Drilling0
Refiningd
Pipeline-Crude & Production6
Marine*
Total
TOTAL CASES
4.4
25.45
7.20
5.01
2.26
6.04
INCIDENCE RATES
LOST WORK DAY
CASES
1.35
9.69
2.25
1.30
0.86
1.76
NON-FATAL W/0
LOST WORKDAYS
3.07
15.67
4.95
3.67
1.37
4.15
FREQ.
RATE
6.57
48.93
5.57
6.52
4.44
6.49
-------
TABLE 20 CAUSES OF FATALITIES IN THE PETROLEUM INDUSTRY, 1974 AND 1975'
FUNCTION
EXPLORATION AND PRODUCTION15
DRILLING0
REFININGd
PIPELINE-CRUDE & PRODUCTS6
MARINE
TOTAL
CASES
REPORTED
1975
7
2
2
4
2
17
1974
19
1
10
5
1
36
FIRE AND
EXPLOSION
1975
0
0
1
2
0
3
1974
5
0
3
0
0
8
STRUCK BY
EQUIPMENT
1975
1
1
0
0
0
2
1974
0
1
4
1
0
6
MOTOR
VEHICLE
1975
1
0
0
2
0
3
1974
3
0
1
0
0
4
AIRCRAFT
1975
1
0
0
0
0
1
1974
6
0
0
0
0
6
OTHER
CAUSES
1975
4
1
1
0
2
8
1974
5
0
2
4
1
12
a. Reference 35.
b.-f. See notes b.-f., Table 19.
-------
TABLE 21. OCCUPATIONAL ACCIDENTS FOR CRUDE OIL
AND PRODUCT TRANSPORT AND REFININGa
SYSTEM
CRUDE OIL
PIPELINE
PIPELINE
PIPELINE
TANKER OR SUPERTANKER
OIL TANKER
BARGES
OIL BARGE
TANK TRUCK
TANK CARS
PRODUCT DISTRIBUTION
PIPELINE
UNCONTROLLED TANKER
OR SUPERTANKER
CONTROLLED TANKER
OR SUPERTANKER
UNCONTROLLED SOUTH
AMERICAN TANKER
CONTROLLED SOUTH
AMERICAN TANKER
TANKER
BARGES
TANK TRUCKS
TANK CARS
ACCIDENT RATE PER 1012 BTU EQUIVALENTS
FATALITIES INJURIES MAN-DAYS LOST
6.9 x 10~5
0.009
NC
U
0.0009
U
0.0009
U
U
6.9 x 10~5
U
U
U
U
NC
U
U
U
4.2 x 10 3
0.008
NC
U
0.008
U
0.008
U
U
4.2 x 10~3
U
U
U
U
NC
U
U
U
1
15
NC
U
15
U
15
U
U
0.1
U
U
U
U
NC
U
U
U
77
-------
TABLE 21 (continued)
SYSTEM
FATALITIES
INJURIES
MAN-DAYS LOST
-4
-4
CANADIAN CRUDE/IMPORTED
UNCONTROLLED 4.4 x 10
CONTROLLED 4.4 x 10
MIDDLE EAST CRUDE/
IMPORTED
UNCONTROLLED 4.4 x 10~4
CONTROLLED 4.4 x 10~4
CONVENTIONAL REFINERY
ARABIAN CRUDE 0.0014
KUWAIT CRUDE 0.0014
TOPPING REFINERY
KUWAIT CRUDE 0.0014
0.0362
0.0362
0.0362
0.0362
0.11
0.11
0.11
2.05
2.05
2.05
2.05
25
25
25
a. Reference 15.
NC-not considered
U-Unknown
78
-------
TABLE 22. ANNUAL DEATHS, INJURIES AND WORKDAYS LOST FOR UNCONTROLLED OIL-FIRED ELECTRICITY SYSTEMS
ASSOCIATED WITH A 1,000-MEGAWATT POWERPLANT WITH A LOAD FACTOR OF 0.75a
EXTRACTION TRANSPORTd TOTAL
ONSHORE OFFSHORE PROCESSING0 ONSHORE OFFSHORE IMPORT CONVERSION6 TRANSMISSION ONSHORE OFFSHORE IMPORT
DEATHS 0.21 0.21 0.08 0.05 0.05 0.05 0.010
INJURIES 21 21 5.6 4.5 4.5 4.5 1.15
WORKDAYS
LOST 2,135 29135 785 562 562 562 127.2
NA 0.35 0.35 0.06
NA 32.3 32.3 5.7
NA 3,609 3,609 689
a. Reference 1.
b. Based on injury rate and the number of man-hours worked In oil extraction.
c. Based on the total man-hours worked. Ibid., note 18.
d. Based on 1969 occupational health data and on the share of all oil needed to fuel a 1,000 MWe powerplant for 1 year.
e. One-half the combined deaths and permanent injuries are assumed to be fatal injuries. Permanent total disabilities are
considered to represent 6,000 workdays lost, and other disabilities are estimated as 100 days lost.
-------
The USGS offshore reporting system covers outer continental shelf oil and
gas spills associated with platforms and rigs. Since 1971, operators have
been required to report all incidents. Prior to 1971, reporting of incidents
was spotty.
Several studies of risks and trends of oil spills have been conducted in
order to provide a reliable statistical basis for policy formulation to prevent
oil pollution.36,37,38,39,40 GKY and Associates attempted to derive trends,
based on data in both the PIRS and SPCC files.36 A graphic representation of
spill volume data contained in these files for 1971 through 1975 is shown in
Figure 3.36
Analysis of the data for 1975 indicated that the spill rate for vessels
was approximately 1 gallon spilled for every 10,000 gallons transported. Most
spillage was associated with large spills. Both large and small spills oc-
curred most frequently for vessels, onshore transportation (railroads and
trucks), and onshore wells. Small spills also occurred frequently at storage
facilities. The data also indicated that:
1. The largest number of transportation accidents are associated with
vessels and pipelines (inland),
2. The largest number of non-transportation accidents are associated
with onshore facilities (wells) in the vicinity of the Great Lakes
and inland and offshore wells in the Coastal regions,
3. The ranking of sources with the largest number of incidents is:
vessels, wells (on and offshore), and pipelines, and
4. Major spill volumes tend to derive from vessels, pipelines, and on-
shore wells.
Trend analysis was performed by comparing data for 1975 to that for the
period from 1972 to 1975. Results indicated significant decreasing trends in
small spills from vessels and bulk storage facilities except in the Great
Lakes area, but an increase in large spills from vessels especially on the
east and west coasts. However, care should be exercised in drawing conclu-
sions from these results because the data base is small and covers only a
short time span and because the consistency and accuracy with which spills are
reported cannot be established.
A method has been described which attempts to determine the probability
of a spill, the probable distribution of spill sizes, and the recurrence inter-
val (the average period of time between two spills greater than or equal to a
specific size).37 This approach depends on availability of valid data; however
such data generally have not been available. Data from the PIRS and SPCC data
files may make this approach more useful.
A different approach was taken by Paulson et al in a risk analysis of
large volume oil spills.38 Their method is based on statistical analysis of
the available data. With further development the method may lead to predic-
tions of individual or total spillage.
80
-------
22
20
18
i f.
J. U
14
12
10
8
6
4
2
m cn
0 +J
d
1-1 01
100, 000 gal.
^_
—
—
<~o
on
i^
A
00
CSl
1— 1
ON
ON
:::x:x:x:x:
;::X;X':-Xx
:::::::::::::x:::
•X\'X'X'X'
•X'XvX'X*
•"•'''•"'*'!"X'''
vXvXvX:
:-::X;a^:S:
glSlSijiiiij:
•x*x'i"x""'x
X'X'X'X'X
SiWiiiiiijiji
r-.
fT^i
tj *
ON
o"
t-i
*X*X*XvX
:::::j:i:jxi::::|
SJWxwi
'.*.*." fv~l •'•'•'
'*'.'.* ^J »". .
IvX CN vX
•i'i'i-i'i'Xvi'
•:•:::•:•:•:•:•:•:
CO
IT)
>sD
«\
i-H
r^
:|S:;:;g:|:i:
111!
wisiiss
•:•:•;• o1-:-:":
•:•:•:• n ':i:
^D
CN
c— 1
A
O
.H
vS!:x:j:w
W&Sviv
ijiji-xW:?:
:::i:i:S::':|:|:|:
•SJijiSS?
:;Xx:v:vx
X'X'X."-'X'I
v.'.; /^*s t'X'
•:•:•:• w :%:;
¥:>; rt vX;
:|x;: p^i:>S
•x-i-'Sx;:::
iiSJijix'?!:!:!
X;X;X;X;X;
X'I'X'X'X'I
•XxX'X'X:
?i$si¥i?
:j:j:i::i;:;:|H
^o
i~-
ON
n
r^
i— 1
i:;:i:i:;S:;S
x:;:x':x':i::::
X'I'X***X*X
liGll
't'.'i* r ^ i'.'.'
•I'X' t- ' • • •
ivXPHjXv
x-x w!:¥:;
:i:i:i:cat:|:
:$x to !•:•:•:
:::::::rti:::g
|:|:|i|P4v'i|:
^Si^-S
ill
£:jxW:|i£
'''•"•'t'X't'I'I*
IvXvX*X\
•i'S'i'lvi'X'
:SS:?:?S
:::i:::::::::::::::
1971° 1972 1973 1974
1975
Figure 3. Spill volume per year.
a Reference 36
b Data on large spills not available
81
-------
Table 23 provides an estimate of the relative contribution of various
sources to oil in the ocean.41 Much of this oil enters the water as a result
of normal operations rather than because of accidents.
Data on oil spills show that the quantity of oil spilled varies with the
source and type of accident. Most spills are small with large spills occur-
ring only infrequently. Data for 1971 and 1972 show that one spill
greater than one million barrels occurred in each year (Table 24).42 Analysis
of 36 major oil spills indicated that three-fourths involved vessels (Figure
4).^3 The oil spilled may be either crude or a petroleum product. The petro-
leum products can be further subdivided into light and residual oil fractions.
An analysis of 35 incidents showed that 80% of the oil spilled was crude oil
(Figure 5).43
Shipping is the largest single source of oil pollution. A summary of
data collected from Lloyd's Weekly Casualty Reports, the U.S. Coast Guard,
and oil companies for tankships in excess of 3,000 dwt is given in Table 25.4
The information summarizes 3,183 accidents from 1969 through 1973 and includes
type of accident, deaths and injuries, and resulting overflow. An analysis of
47 tankship losses of over 10,000 dwt (Table 26)4 from the table shows that:
1. Collisions and groundings account for about half of the accidents
and approximately 44% of the outflow.
2. Structural failures, which account for 16% of the accidents, account
for more than one-third of the outflow.
3. Less than one-fourth of the accidents occur at sea, and
4. Approximately 11% of the tanker accidents resulting in out outflow
also result in loss of the tanker. These accidents account for
81% of the oil outflow.
Characteristics of major spills which occurred prior to 1970 are contained in
Table 27.43 Generally, the following conclusions can be reached about past
major spills•.
1. The source of the spill was likely to be a tanker,
2. Crude or residual fuel oil cargoes were involved in 90% of the
spills.
3. The size of the spill was greater than 5,000 bbl, 75% of the time,
with a median size of 25,000 bbl,
4. The spill usually occurred within 10 miles of shore,
5. The duration of a spill incident was usually more than five days,
with a median of 17 days duration,
82
-------
w
d
H
PM
H
O
H
O
H
PM
100% i—
90%
80%
70%
60%
50%
30%
20%
10%
Other
Vessels
Various Tankei
Casualties
Tanker Groundings
Other
Refineries
S 3 ¥:
Offshore
Drilling
Figure 4. Source of spill. Data from 36 incidents, 1956 to 1969.E
Reference 43
83
-------
o
w
hJ
Oi
-------
TABLE 23. CONTRIBUTIONS OF VARIOUS
SOURCES TO OIL IN THE OCEANS3
SOURCE ESTIMATED CONTRIBUTION (%)
(TONS/YR)
TRANSPORTATION
TANKERS, DRY DOCKING,
TERMINAL OPERATION,
BILGES, ACCIDENTS 2,350,000 34.9
COASTAL REFINERIES, MUNICIPAL
AND INDUSTRIAL WASTE 875,000 13.0
OFFSHORE OIL PRODUCTIONS 87,500 1.3
RIVER AND URBAN RUNOFF 2,100,000 31.2
ATMOSPHERIC FALLOUT 660,000 9.8
NATURAL SEEPS 660,000 9.8
TOTAL 6,732,500 100.0
aReference 41
85
-------
oo
TABLE 24. NUMBER OF OIL SPILLS BY VOLUME AND SOURCE, 1971-1972a
SPILL VOLUME (BARRELS)
FACILITY
TERMINAL
1971
1972
SHIP (OFFSHORE)
1971
1972
PIPELINE
1971
1972
PLATFORM
1971
1972
TOTAL
1971
1972
0-1
384
351
4
15
222
15
227
431
837
812
1-10
247
347
6
2
403
24
304
784
960
1,157
10-100
458
544
8
10
496
61
395
728
1,357
1,343
100-
1,000
282
298
0
3
257
61
146
244
685
606
1,000-
10,000
77
71
4
0
41
32
13
20
135
123
10,000-
100,000
19
16
0
0
13
7
2
4
34
27
100,000-
1,000,000
7
5
0
1
2
3
0
0
9
9
1,000,000-
10,000,000
1
0
0
1
0
0
0
0
1
1
a. Reference 42.
-------
TABLE 25. TANKSHIP ACCIDENTS, 1969-19733
NO. OF
TYPE OF ACCIDENT ACCIDENTS
Breakdown
Collision
Explosion
Fire
Grounding
Ramming
Structural Failure
Other
Totals
355
744
104
197
790
473
515
5
3,183
NO. OF ACCIDENTS
RESULTING IN
OUTFLOW (%)
11
126
31
17
123
46
94
4
452
(2.4)
(27.9)
(6.9)
(3.7)
(27.2)
(10.2)
(20.8)
(0.9)
(100)
OUTFLOW
(LONG TONS)
29,940
185,088
94,803
2,935
230,806
13,645
339,181
54,911
951,317
NO. OF ACCIDENTS
RESULTING IN OUTFLOW
TOTAL LOSS (%) (LONG TONS)
2
7
11
1
12
0
15
3
51
(3.9)
(13.7)
(21.6)
(2.0)
(23.5)
(0)
(29.4)
(5.9)
(100)
29,350
140,779
88,780
1,250
134,449
0
322,519
54,790
771,917
NO.
4
26
33
14
0
0
6
0
83
ACCIDENTS 1971-1973 LOCATION OF 452 ACCIDENTS
DEATHS INJURIES PIER HARBOR ENTRANCE COASTAL
5
259
46
34
0
0
37
0
381
53
130
47
10
0
0
32
0
178
0
5
5
10
1
18
8
1
48
1
41
4
2
27
15
9
0
99
1
25
0
0
40
5
4
0
75
5
45
6
1
53
4
7
2
123
SEA
3
9
15
4
2
2
64
1
98
Reference 4.
-------
TABLE 26. ACCIDENT SEQUENCE FOR 47 TANKSHIP LOSSES^
FOR TANKSHIPS OVER 10,000 DEADWEIGHT TONS, 1969-1973*
ACCIDENT SEQUENCE NUMBER
Breakdown-Structural Failure-Sink
Breakdown-Grounding Sink
Collision-Sink
Collision-Explosion/Fire-Sink
Explosion/Fire-Sink
Grounding-Explosion/Fire-Sink
Grounding-Sink
Flooding-Sink
Structural Failure-Grounding-Sink
Structure Failure-Sink
Total
Reference 4
Conversion factor: Long tons x 7.58 = Barrels
OIL OUTFLOW
(LONG TONS)b
(BARRELS)
1
1
2
4
12
1
9
2
1
14
47
16,350
13,000
4,138
136,163
90,030
2,500
134,726
54,669
40,000
282,519
774,095
123,933
98,540
31,365
1,032,116
682,427
18,950
1,021,223
414,391
303,200
2,141,494
5,867,640
88
-------
TABLE 27. MAJOR OIL-SPILL INCIDENTS-SIGNIFICANT CHARACTERISTICS (1956-1969)'
oo
•o
NAME
Algol, tanker
Andron, tanker
Anne Mildred Brovig, tanker
Argea Prima, tanker
USS Bach (DDE-470)
Benedicte, tanker
Chryssi P. Goulandris, tanker
Dutch Coast Spill
Esso Essen, tanker
Fawley, refinery
General Colocotronis, tanker
Hamilton Trader, tanker
Hess Hustler, tank barge
Humboldt Bay, refinery
Kenai Peninsula, tanker
Keo, tanker
Martita, tanker
Moron, refinery
New Castle, power station
Ocean Eagle, tanker
Palva, tanker
Pegasos, tanker
P. W. Thirtle, tanker
R. C. Stoner, tanker
Refinery Loading Site
Robert L. Polling, tank barge
Santa Barbara, platform
Seagate, tanker
Seewarren, storage tank
Ship Shoal, drilling rig
CAUSE OF SPILL
Grounding
Sinking
Collision
Grounding
Grounding
Collision
Unknown
Grounding
Pumping
Grounding
Collision
Grounding
Hose Failure
Collision
Hull Failure
Collision
Pumping
Leak
Grounding
Grounding
Hull Failure
Grounding
Grounding
Hose Failure
Collision
Natural Faults
Grounding
Tank Failure
Storm Shifting
MATERIAL
#6 F.O.
Crude
Crude
Crude
#6 F.O.
Crude
Crude
Residual
Crude
Crude
Crude
Residual
#6 F.O.
Diesel
Crude
#4 F.O.
Bunker C
Crude
Residual
Crude
Crude
Bunker C
Bunker C
Mixed
Crude
#2 F.O.
Crude
Residual
Crude
Crude
VOLUME
(BARRELS)
4,000
117,000
125,000
28,000
14,000
2,600
1,000
30,000
30,000
5,000
40
1,400
1,000
210,000
4,300
16,000
40
83,400
143,000
2,000
4,700
100,000
200,000
2,400
DISTANCE EXTENT OF
FROM.. SHORE CONTAMINATION
(MILES) (MILES)
3 None
15
<1 None
13 Minor
<1 Moderate
Moderate
3 15
<1 Moderate
2 Moderate
45
<1
<1 Minor
120 None
<1
<1 6
<1 Moderate
<1 16
5
270 None
2 Extensive
<1 Extensive
<1 25
<1 13
6 40
<1
6 None
DISTANCE
FROM PORT
(MILES)
<10
>50
<10
10-25
<10
10-25
10
25-50
<10
>50
<10
<1Q
<10
<10
>50
10-25
>50
<10
<10
45
<10
DURATION
(DAYS)
21
75
23
13
7
60
90
5
2
5
65
13
10
55
1
100
3
Reference 43 (Vol. II, P. 244).
-------
TABLE 27. Continued
NAME
USS Shangri-La (CVA-38)
Tampico, tanker
Tim, tank barge
Torrey Canyon, tanker
Waikiki Beach
Waterford Beach
Witwater, tanker
World Glory, tanker
CAUSE OF SPILL MATERIAL
Hull Failure Mixed
Hull Failure Crude
VOLUME
(BARRELS)
DISTANCE
FROM SHORE
(MILES)
Pumping
Grounding
Sinking
Grounding
Unknown
Unknown
NSFO
Diesel
#6 F.O.
Crude
Bunker C
#6 F.O.
200
60,000
7,000
700,000
15
15,000
322,000
40
EXTENT OF
CONTAMINATION
(MILES)
Moderate
3
Moderate
242
7
20
Extensive
None
DISTANCE
FROM PORT
(MILES)
>50
25
>50
DURATION
(DAYS)
13
30
4
10
20
-------
6. The shoreline threatened was at least partially recreational for
80% of the spills, with a median of four miles contaminated, and
7. Three-quarters of the spills occurred within 25 miles of a port.
Blowouts and well-casing ruptures are other potential sources of major
oil spills. A blowout is the uncontrolled flow of oil or gas from a well
bore. A National Petroleum Council survey showed that between 1960 and
1970 there were 106 blowouts in the drilling of 273,000 wells in eight major
oil-producing states.
The survey figures of 106 blowouts may not accurately summarize the situ-
ation. A blowout is considered to be any of the conditions ranging from poten-
tial loss of well control to the catastrophic blowout resulting in fire and
loss of the rig. Potential intermittent loss of control (e.g., 1-hour loss)
are occurrences which are generally unreported and probably relatively common.
Oil drilling (production) operations in southern Louisiana, for example, oper-
ate on the borderline of loss of control. In this area there is a fine balance
between circulation rate and loss of well control, with the line being crossed
in both directions during the course of operations. The catastrophic occur-
rence is often recorded because reporting to a responsible regulatory authority
or an insurance company may be required. The figures for blowouts may there-
fore belong to the severe category only. It is possible, however,that the rela-
tive number of blowouts in relation to wells drilled is decreasing with devel-
opment of well control technology.
Blowouts have also been related to fires, such as the recent offshore rig
accident in the Gulf of Mexico. The best known of the major polluting blowouts
occurred in the Santa Barbara (California) Channel, 6.5 miles from shore, in
1969. Uncontrolled flow lasted from January 28 to February 7. After cementing
of the well, seepage continued at a rate of approximately 300 bbl per day until
September 1969. Seepage then was reduced, by grouting, to approximately 10 bbl
per day.
Well casing is a cemented steel sleeve implanted in the well walls. Its
purpose is to support the well walls and to prevent intrusion of oil, brines,
and mud into the subsurface geologic structure. The Santa Barbara accident
showed that the casing was shallow, extending 238 feet below the sea floor for
the 3,313 foot well. When the well was shut off at the top, fluid pressure
built up in the well and blew out just below the bottom of the casing.
Oil spill control regulations, environmental awareness, and technological
advances have combined to reduce the number of blowouts. Blowout preventer
stacks and cementing of steel casing into the bore hole reduce the frequency
of blowouts. The blowout preventer stack is mounted at the top of the casing
and consists of a series of manual or automatic valves which can close off the
well in the event of loss of control.
When oil spills do occur, several methods can be used to contain and re-
move the oil. For example, containment and recovery can be accomplished by
mechanical methods and sorbent materials, the latter presenting logistical
91
-------
problems of disposal and collection. Mechanically, booms may be used to con-
tain the oil by encircling it and sweeping it to a collection point or to
funnel oil in a "V" configuration to skimmers mounted on barges. This is ef-
fective in protected areas such as harbors. Booms are also used to prevent
oil from entering marinas and tributaries and to protect boats and other
property. Air barriers also have been used effectively in harbors or calm in-
land waters. Containment of oceanfloor seeps is accomplished by means of
hoods connected to surface containers with flexible hosing. Another method
of reducing the amount of oil entering the marine environment is controlled
burning while relief wells are drilled to kill the formation feeding the
producing wells.
Three methods of recovering the oil once it is contained are oil adhe-
sion, open skimmer, and floating-surface skimmer. None of these are satisfac-
tory in heavy seas.
Many treatment processes can be used. Among these are combustion, sink-
ing, and dispersants. The combustion technique has been attempted with little
success and may have detrimental environmental consequences such as sediment
contamination by hydrocarbon residues.44 if used immediately after a spill,
there is danger to personnel and property, while if tried after weathering,
ignition and sustained combustion become difficult due to loss of crude-oil
volatiles. The addition of materials which sink the oil is not widely used
because this transfers the problem to the bottom environment rather than
solving the problem.43,44 Dispersants are not widely used because they may
be toxic to marine life especially in enclosed or shallow waters.
4.2.1.2 Effects—
The effects of an oil spill depend on a number of variables associated
with the spill. These include the volume of the spill, the composition of the
spilled oil, the effects of weathering, marine transport (e.g., dispersion,
trajectory), existing ecosystems, toxicity, physiography, and cleanup opera-
tions (duration, type, extent).
Toxicity is related to the composition of the spilled oil. Low boiling,
saturated hydrocarbons can produce anesthesia and narcosis in low concentra-
tions. They also can cause cell damage and death in lower animals at higher
concentrations. Low boiling aromatic hydrocarbons (e.g., benzene, toluene,
and xylene) are abundant in all oils and most oil products and are that most
immediately toxic fraction.45 Low-boiling aromatics are more water soluble
than the saturates and can kill marine organisms either by direct contact or
through contact in dilute solution. Other components of crude oils which are
both soluble in water and toxic are cresols, xylenols, naphthols, and quino-
line. A summary of toxicity data for various organisms is contained in Table
28.46
The toxicity of oils is greatest for the lightest fractions (Table 29).4?
Refined oils containing a greater percentage of low boiling substances are more
toxic than crude oil.
92
-------
TABLE 28. SUMMARY OF TOXICITY DATA3
ESTIMATED TYPICAL TOXICITY RANGES (ppm) FOR VARIOUS SUBSTANCES
#2 FUEL FRESH
CLASS OR ORGANISM
Flora
Finfish
Larvae
Pelagic Crustaceans 1-10
Gastropods
Bivalves
Other Benthic
Inver t eb r a t e s
aReference 46
Soluble aromatic derivatives (aromatics and napthenoaromatics).
SADD
10-100
5-50
0.1-1
1-10
10-100
5-50
: 1-10
1-10
OIL/KEROSENE CRUDE
50-500
25-250
0.5-5
5-50
50-500
25-250
5-50
5-50
10*
10*
102
103
io4
10*
103
IO3
- 105
- IO5
-IO3
-IO4
-IO5
- IO5
- 10*
-10*
WEATHERED CRUDE
Coating more signifi-
cant than toxicity
93
-------
TABLE 29. VARIATION IN BIOLOGIC EFFECTS OF
OIL WITH TYPE OF OILa
TYPE OF OIL IN
ORDER OF INCREASING
BIOLOGIC EFFECT
Crude Oil
No. 6 Fuel Oil
No. 2 Fuel Oil
aReference 47
TYPE OF ORGANISM IN ORDER
OF INCREASING BIOLOGIC
EFFECT
Mammals, Finfish, Plankton,
Benthos, Birds
Mammals, Finfish, Plankton,
Benthos, Birds
Mammals, Birds, Finfish, Plank-
ton, Benthos
94
-------
The higher toxicity of refined oils is illustrated by the spill of No. 2
distillate oil which occurred at West Falmouth, Massachusetts in 1969. This
spill produced an immediate kill of fish, clams, snails, crabs, lobsters,
worms, and other invertebrates. Oil was incorporated into sediments and con-
tinued to inhibit repopulation nine months after the spill.48
The effects of spills in a estuarine or marine environments can be sum-
marized as follows:
1. Direct lethal toxicity - Dissolved aromatic hydrocarbon derivatives
can disrupt membrane activity in crustaceans, burrowing organisms,
and to a lesser extent in fish and mollusks, leading directly to
organism death.46
2. Disruption of physiological and behavioral activity - Chemical com-
munication can be inhibited leasing to abnormal feeding, reproduc-
tion, orientation, and migration of many marine organisms.46 This
could result from contamination by weathered crude or No. 2 fuel oil.
For most spills, this effect would persist for only a few hours if it
occurred at all. However, in the rare instance in which a highly
toxic product such as No. 2 fuel oil is spilled under unusually ad-
verse sea conditions so that vertical mixing occurs and the sediments
become saturated with the oil, this effect could persist' for several
months or even for a year or more.49
3. Direct coating by oil - Oil may smother or physically inhibit activi-
ties such as feeding and locomotion. Sessile organisms are most sus-
ceptible to coating.
4. Incorporation of hydrocarbons in food chains - Polynuclear aromatic
hydrocarbons, some of which are carcinogenic, are generally present
in oils in concentrations of a few parts per million. These com-
pounds are easily absorbed and concentrated by marine animals. How-
ever, marine animals quickly depurate these hydrocarbons once they
have been removed from exposure to the oil.50,51 Finfish and filter
feeding organisms such as clams and oysters concentrate hydrocarbons
in body tissues creating an objectionable taste.
5. Habitat "changes - Alteration of the physical-chemical environment,
especially in the sediment, will cause shifts in species composition,
relative abundance, and areal distribution. The degree to which oil
will contaminate sediments is a function of the absorption capability
of the sediment, depth of water, vertical mixing, and state to which
the slick has weathered and become more dense.
In addition to these effects, a significant number of sea birds can be
killed by an oil spill. Between 40,000 and 100,000 birds were killed as a
result of the Torrey Canyon spill,52 7,000 in the San Francisco Bay spill,-^
and 3,700 in the Santa Barbara blowout.54 The principal effect of an oil
slick on birds is physical coating. Birds become coated because they may not
recognize an oil slick until they cannot avoid contact. It is felt that the
95
-------
swimming species are more endangered by coating than are aerial species, who
rarely come in contact with oil regardless of its location.49 Birds are warm-
blooded with their feathers serving as insulation. In addition, they have to
be waterproof. A swimming bird encountering oil will dive, generally resur-
facing in the oil. The oil mats the insulating feathers presenting the possi-
bility of the bird freezing to death or dying from disease resulting from body
heat loss. A further effect of the matting is to disturb the buoyancy, dis-
rupting the feeding and swimming capability of the bird, both of which can
cause death. -^ Birds also die from ingestion of oil as a result of preening
or eating oil contaminated food.
The length of time it takes for a marine or estuarine ecosystem to re-
cover from an oil spill has not been established. Although some feel that
there are few long term effects,47 others feel that long term ecosystem dis-
ruption can occur.^ However, it is apparent that the time required for re-
covery depends on many factors including the volume of oil spilled, the kind
of oil spilled, meteorological conditions, and cleanup measures. Table 30
summarizes the biologic recovery observed after several oil spills.4? A sum-
mary of the effects of oil on selected species is shown in Table 31.->°
The toxicity of oil and the persistance of its effects depend, in part,
on weathering which is the alteration of oil by evaporation, dissolution,
microbial decomposition, and chemical oxidation. Evaporation and dissolution
lead to rapid depletion (2-4 days) of the low-boiling, highly toxic frac-
tions. Microbial and chemical oxidation of most higher boiling fractions is
a relatively slow process and may take several years.46,57
Persistence times of oil have been estimated based on observations fol-
lowing a number of spills. One estimate for persistance of weathered oil is
shown in Figure 6.^6
Persistence in the marine environment is related to transport of oil
after a spill. The movement of an oil spill in the marine environment is de-
pendent upon wind, wave action, currents, and turbulence. Poor weather condi-
tions, for example, can cause wide dispersal of oil before control activities
can be effected. Turbulence, as well as some chemical clean-up methods, can
transport oil to the sediments where it can persist for a significant time.
The oil reduces cohesion of the sediment which then gives rise to significant
littoral drift, moving the oil to areas not previously affected by'the spill.
Oil spilled near a shore can leave a visible coating on rocks, sand, and
marsh flora. The extent of the effects will depend upon the use of the shore,
i.e., whether it is industrial, residential, recreational land, or salt marsh
or other wetlands. Recreation shoreline used as a beach cannot be used until
the oil is removed.
An examination of 21 oil spill incidents indicated that 80% occurred
within 10 miles of shore with a median duration of 17 days. Based upon an ap-
parent drift rate of 3.3% of the average wind velocity (assumed 15 knot wind)
it was indicated that oil could reach the shore in less than one day in 80%
of the cases. A further analysis of the data indicated that less than 20 miles
96
-------
10
9
rH
•H
0 8
CO
rl
CO
td 7
01
o
^ 6
CO
1 5
H
11 Persistence
UJ -P>
1 2
cd
S
1
— d-i
—
—
«_
—
c
—
b-
•••••••••i
b-
••••••••W
•••••••MM
b-
•••••••••
n
Offshore Rocky Worm Sandy Mussel Salt
bottom shore and shore reef marsh
clam
flat
Figure 6.a Observed persistence of petroleum substances in various marine
habitats following actual oil spills.f
Reference 56
Analytical determination of oil (Scarratt & Zitko, 1972).
Estimate after 2 years; analytical methods; used #2 fuel oil (Blumer & Sass, 1973)
Visual observation and analytical; JP-5 and #2 fuel oil (Shenton, 1973).
Gas-liquid chromatography analysis (Spooner, 1971).
Maximum times shown do not necessarily imply complete removal of oil, but
may represent an estimate of persistence or termination of study.
97
-------
TABLE 30. BIOLOGIC RECOVERY FOLLOWING OIL SPILLSa
STATE OF BIOLOGIC COMMUNITIES, TIME AFTER SPILL
INCIDENT OIL IMMEDIATELY ONE-HALF YEAR ONE YEAR
S/S Tampico No. 2 fuel oil Extensive damage
Maru
West Falmouth No. 2 fuel oil Extensive
S/S Arrow No. 6 fuel oil Slight to considerable
San Francisco No. 6 fuel oil Slight to considerable
Bay
S/S Torrey Kuwait crude Slight (after applying
Canyon detergents)
Santa Barbara California crude Slight
Repopulation beginning
No evidence of recovery
Area appears normal
Population reduced on
Doxbury Reef
Repopulation under
way in some areas
Communities have
normal numbers -
appear healthy
Many healthy species
Slight repopulation,
some localities
Appears normal
Population somewhat
reduced - repopula-
tion proceeding at
normal rate
Repopulation under
way
Appears normal
LONGER
After 3 years,
recovery nearly
complete -
slightly dif-
ferent from
original
After 2 years,
recovery under
way in most
areas
Repopulation
continuing
aReference 47
-------
SPECIES
COMMON
NAME
TABLE 31. EFFECTS OF OIL ON SELECTED SPECIESa'b
LETHAL SUBLETHAL COATING
UPTAKE
AND
TAINTING
HABITAT
CHANGE
Birds
Rissa tridactyta
Fishes
Alosa spp.
Clupea harengus
Fundulus heteroalitus
Gadus movhua
M-iaropogon undulatus
Morona saxat-il-is
Pseudopteuronectes
ameri.ca.nus
Kittiwake
Alewife x
Herring x
Muramichog x
Atlantic cod x
Croaker
Striped bass
Winter flounder x
x
x
x
Crustaceans
Aoart-ia spp.
Ampel-Lsoa vadorum
Balanus balanoides
Calanus spp,
Crangon spp.
Emerita spp.
Homarus ameniaanus
Paqurus longiaca-pus
Pandalus spp.
Mo Husks
Asqu-ipecten spp.
Crassotrea spp.
Donax spp.
Mercenaries, meroenar-ia
Modiolus spp.
Mya arenia
Mytdus edufis
Zooplankter x
Amphipod x
Acorn barnacle x
Zooplankter x
Shrimp . x
Mole crab x
American lobster x
Hermit crab x
Shrimp x
Scallop x
Virginia oyster x
Coquina clam x
Northern quahog x
Horse mussel
Soft shell clam x
Edible mussel x
x
x
x
x
X
X
X
X
X
X
(Continued)
-------
SPECIES
Mollusks (Continued)
Nittorina hittorea
and spp.
Nassarius obsoletus
Thais lapillus
Worms
Aranicola marina
Nereis vixens
— Stroblospio benedicti
o
Other animals
Astevias vulgaris
StTongy looentvotus
droebaohienis
Plants
Juneus gerardi
Spartina
Spaptina patens
Laminopia spp.
COMMON
NAME
Perwinkle
Common mud snail
Dog whelk
Lugworm
Clam worm
Polychaete
.Starfish
Sea urchin
Marsh rushes
Marsh grasses
Cord grass
Kelp
TABLE 31. (Continued)
LETHAL SUBLETHAL
COATING
UPTAKE
AND
TAINTING
HABITAT
CHANGE
x
x
X
X
X
X
X
X
X
X
X
X
X
X
X
X
^Reference 56
bDoes not list all species for which data have been reported. Rather, an x represents reported data for those species which
were selected for special consideration. An x indicates that some data, regardless of number, have been reported.
-------
of shoreline was affected in 80% of the spills with a median of 3-4 miles. An
analysis of 23 spills showed that recreational shoreline was affected in 85%
of the spills, with residential and industrial each affected in 45% of the in-
cidents. Multiple use accounts for the overlap in percentages. Thus, it is
likely that in the event of a spill, recreational shoreline will be affected
•within one day.
A spill of crude oil also presents a fire hazard. The hazard generally
diminishes with time with the loss of the volatile fractions. Spills of light,
refined products (e.g., gasoline) also are a fire hazard; however, heavy frac-
tions (e.g., No. 6 fuel oil) are much less likely to catch fire.
A fresh spill of crude oil at sea can be easily ignited and may flash to
the tanker resulting in a further toll of property loss and injury with subse-
quent involvement of rescue men and equipment. However, the hazard is general-
ly more acute on land as the spill occurs around a well, refinery units, and
storage tanks. Electrical machinery, such as pumps, are located in and around
these areas which increase the likelihood of a spilled oil fire as a result of
an electrical spark. This has occurred in tank farm areas causing extensive
damage.
Inland water spills have accounted for some of the largest expenditures
of public funds for inland clean-up. For example, flooding of a storage lagoon
near Douglasville, Pennsylvania caused by Hurricane Agnes resulted in release
of 6 million gallons of waste oil to the Schuylkill River. Clean-up costs were
nearly $6 million. Similarly, more than $8 million has been spent to clean up
shorefronts, boats, marshes, and property along 80 miles of the St. Lawrence
River because of a spill of No. 6 fuel oil in June 1976. Clean-up operations
are still continuing.
Numerous adverse effects may result from an oil spill on inland waters.
Vegetation may be damaged and benthic organisms may show an increase in heavy
metal and hydrocarbon concentrations. Marshes, wildlife refuges, and water
fowl (e.g., Canadian geese, ducks) may be damaged. The spilled oil may also
threaten sources of potable water, irrigation supplies, and industrial water
supplies.
Oil spills on land are few in number and account for small volumes
when compared with spills on water. The blowout is the major source during
drilling. Most of these, occurred with high-pressure gas wells as opposed to
oil wells.
The major problem associated with a blowout is seepage into and subsequent
contamination of aquifers used for drinking and other consumptive purposes.
Oil companies have controlled blowouts through the use of blowout prevention
devices (BOP's) and through training of their field crews.
Accidental releases from refineries to land are generally in the form of
oil or oil-contaminated water resulting from leaks, spills, and ruptures.
Regulation and good housekeeping practices dictate that containment methods
such as diking be used around equipment. This localizes the spill to the
processing and storage areas.
101
-------
4.2.2 Tanker Accidents
Tanker accidents result not only in oil spills but also in injury, death,
and loss of property. The United States Coast Guard, Office of Merchant Marine
Safety, analyzed data from 3,715 worldwide tankship accidents from 1969 through
1973.4 Table 25 contains a summary relating death and injury to various types
of accidents. The data indicate that 18.4% of the accidents involved death or
injury. A total of 38 deaths occurred during a three year period with 2.14
deaths for every injury. Approximately 31% of the accidents resulted from
collision. Those accidents accounted for 68% of the deaths and 73% of the in-
juries. Explosions were involved in 40% of the accidents involving death or
injury but resulted in only 12% of the deaths and 26% of the injuries.
Data also were presented which indicate the extent of damage to the ships
involved (Table 32).4 Of the 3,183 accidents, 73% caused only minor damage,
while 15% caused no damage. Only 1.6% of the accidents resulted in total loss.
The actual dollar amount of the damage or losses depends on repair or replace-
ment costs, lost business revenues, increased insurance premiums, and other
factors for which no data were available.
The accident sequence for 47 of the 51 total losses is shown in Table 26.
More than 1/3 of the total losses were caused by structural failure and ac-
counted for 44% of the oil outflow. Causes for loss of structural integrity
for these 47 accidents are shown in Table 33.^
4.2.3 Offshore Rig Accidents
There are four basic types of offshore rigs used in drilling for oil and
gas: submersibles, jack-ups, semi-submersibles, and floaters. A study ana-
lyzed offshore rig casualties, on an industry-wide basis, for the period 1955-
1976.58
The current fleet of operating rigs number more than 340 with a total es-
timated value of over $6 billion. Each year more than 2% of the value of off-
shore rigs is lost due to accidental damage. The leading cause of rig loss is
rig moves during stormy weather. Since 1955 there have been 100 accidents,
each with losses in excess of $500,000, resulting in documented losses total-
ling $438,850,000. These 100 accidents accounted for 121 fatalities. Of
these, jack-ups were involved in the most (61 accidents), accounting for al-
most 64% of all damage losses and 33% of the total fatalities. Floaters,
while accounting for only 6% of the damage costs, were involved in 35% of the
fatalities (42), more than from any other type of rig. Since 1955 the casualty
ratio of rigs damaged to rigs operated has been 3.2%. The offshore drilling
industry operated 3,255 rig-years with a total loss of 43. The fatality rate
for offshore drilling has been estimated at 1 death/340,000 man-days, which is
lower than the onshore construction industry rate.
Mobile drilling rig accidents are of two general types. The first is the
type related to the industrial nature of the business. The second is a result
of the marine environment. Among the first type, the most disastrous accident
is the blowout, which can affect the rig and its occupants in several ways.
102
-------
TABLE 32. EXTENT OF TANKSHIP LOSS OR DAMAGE 1969-19733
TYPE OF INVOLVEMENT
Breakdown
Collision
Explosion
Fire
Grounding
Ramming
Structural Failure
Other
TOTAL
aReference 4
TOTAL
LOSS
2
7
11
1
12
0
15
3
51
HEAVY
DAMAGE
16
64
30
26
63
23
39
1
262
LIGHT
DAMAGE
197
570
52
149
487
412
445
1
2,313
NO
DAMAGE
131
78
10
14
206
35
2
0
476
DAMAGE
UNKNOWN
9
25
1
7
22
3
14
0
81
103
-------
TABLE 33. DESCRIPTION OF LOSS OF STRUCTURAL INTEGRITY
FOR 47 TANKSHIP LOSSES FOR TANKSHIPS OVER 10,000
DEADWEIGHT TONS, 1969-19733
DESCRIPTION~NUMBEROIL OUTFLOW
(LONG TONS)
A. Loss of structural integrity of hull
caused primarily by external forces
or where local material conditions
deteriorated. No explosion or fire
was associated with the accident.
These may be broken down into:
1. Structural failure of main hull 12 243,619
girder from excess bending or shear
loading
2. Local structural failure of hull
envelope
a. Failure of hull penetration 2 36,750
b. Local hull plating failure 2 39,169
c. Unknown local structure failure 1 34,000
3. Hull damage caused by collision or
grounding
a. Collision 2 4,138
b. Grounding 11 187,726
Subtotal 30 545,402
B. Loss of structural integrity from damage
caused primarily by explosion or fire or
where explosion or fire contributed to
loss of structural integrity. These may
be broken down into:
1. Explosion or fire initiated in own 12 90,030
ship cargo tanks
2. Explosion or fire set off by vessel
collision or grounding
a. Collision 4 136,163
b. Grounding 1 2,500
Subtotal 17 228,693
a Reference 4~
104
-------
Should the gas or product ignite after a blowout, the drilling rig could easily
catch fire resulting in total destruction. A flowout around a subsea wellhead,
for example, could create sufficient aeration of the water to alter water den-'
sity resulting in the vessel sinking. An effect on jack-ups or any design per-
manently anchored to the bottom, would be the scouring around mats, legs, or
other bottom supports, due to escaping gas, which could cause foundation fail-
ure and subsequent rig damage. Of the 100 major accidents mentioned, 32% were
related to industrial operations and accounted for $115.65 million (26%) of
the total documented losses.
The second type of accident, resulting from the marine environment, ac-
counted for 68% of the accidents. These resulted from moving of the rigs and/
or storms. Forty of the major accidents occurred while the rigs were being
moved. Of these, 35 involved jack-ups. Accidents involving moving of rigs
accounted for losses of $123.8 million (28% of the total). The other 28 major
accidents were storm related in which damage losses approximated $199.4 million
(45% of the total). The industry and government agencies are attempting to
reduce these types of accidents through improved design standards and personnel
training programs.
4.2.4 Pipeline Accidents
The hazards associated with the pipeline accident include fire and explo-
sion with resulting injury and/or loss of life, in addition to oil spill damage.
Liquid pipeline accidents caused approximately $3.2 million in property
damage, the loss of 7 lives, 15 injuries, and the loss of over 319,000 barrels
of commodity during 1975. A summary of pipeline accidents for 1975 is shown in
Table 34.59 A breakdown of accidents in terms of the commodity involved is
shown in Table 35.59 it should be noted that of the commodities lost as a re-
sult of accident, 45.1% was crude oil, 28.3% gasoline, and 21.5% liquified
petroleum gas (LPG). Accidents involving LPG accounted for 86% of the deaths
and 87% of the injuries.*
* Injury and fatality statistics are always subject to discrepancies due to non-
uniform or incomplete reporting. Statistics quoted in this section come from
the Department of Transportation (DOT) and are based on DOT Form 7000-1 filed
during the year 1975. The American Petroleum Institute (API) also has pub-
lished data for 1975 based on information collected from 137 oil and gas com-
panies and their subsidiaries. The API data are for the entire industry and
are further divided by function (e.g., exploration and production, drilling,
pipeline-crude and products). Comparison of the appropriate API and DOT data
for pipeline (crude and liquid respectively) shows significant differences
between the two sets of data:
API DOT
Injuries (lost time) 754 15 employee and
non-employee
Fatalities 6 7 employee and
4 non-employee
The API fatalities include tank truck accidents while the DOT figures ap-
parently do not. The two API estimates of fatalities were obtained from
separate sources. (API data are quoted in Section 4.2.)
105
-------
TABLE 34. LIQUID PIPELINE ACCIDENT SUMMARY - JANUARY 1, 1975, THROUGH DECEMBER 31. 1975
" " — "" "• — ------ - . . .. _ . - . __,.. ..,„. _ . .. _ .. — , — , • , . — • — •
CAUSE OF ACCIDENT
OPERATION ACCIDENTS
Equipment Rupturing Line
Corrosion - External
Equipment Failure
Incorrect Operation by
Carrier Personnel
Corrosion - Internal
Defective Pipe Seam"
Defective Girth Weld
Failure of Previously
Damaged Pipe1
Vandalism
Failure of Previous Weld
Repair
Lightningk
Miscellaneous
Total
NO. OF
ACCIDENTS
75
57b
27
22C
16b
19C
5
4
4
2
2
29
255
PERCENT OF
TOTAL
28.6
22.4
10.6
8.6
5.9
5.9
2.0
1.6
1.6
0.7
0.7
11.4
100.0
DEATHS
CARRIER
EMPLOYEES
0
0
3
0
0
0
0
0
0
0
0
0
3
NON
EMPLOYEES
0
0
0
0
0
0
0
4
0
0
0
0
4
INJURIES
CARRIER
EMPLOYEES
0
0
3
0
0
0
0
0
0
0
0
0
3
(continued)
NON
EMPLOYEES
1
0
0
0
0
0
0
9
0
0
0
2
12
-------
TABLE 34 (continued)
CD
--J
PROPERTY DAMAGE ($)
CARRIER OTHER TOTAL
$ 375,110
89,931
534,010
654,705
2,890
2,259
4,150
20,260
3,505
17,450
297,500
380,429
$2,382,199
$266,751
25,850
9,081
128,150
2,550
12,650
2,015
363,000
15
1,500
0
3,500
$814,882
$ 641,681
115,781
543,091
782,855
5,440
14,909
6,165
383,260
3,520
18,950
297,500
383,929
$3,197,081
LOSS OF
COMMODITY
(BARRELS)
83,724
37,073
20,981
26,619
2,853
61,570
30,131
'31,665
884
2,739
11,356
9,828
319,423
BEFORE
1920
3
2
0
0
0
0
0
0
0
0
1
1
7
NUMBER OF ACCIDENTS BY YEAR OF INSTALLATION
1920- 1930- 1940- 1950- 1960- 1970-
1929 1939 1949 1959 1969 1975 NOT REPORTED
9
10
0
1
6
2
0
0
0
0
0
1
29
3
17
2
3
0
3
3
1
1
0
0
2
35
19
14
1
3
2
4
0
0
2
0
1
4
50
14
9
2
1
2
2
1
1
0
1
0
0
33
21
3
10
8
6
7
1
1
1
1
1
10
71
3
1
8
5
0
1
0
1
0
0
0
7
26
1
1
4
1
0
0
0
0
0
0
0
3
10
(continued)
-------
TABLE 34 (Continued)
Reference 59
"One accident had both external and internal corrosion as the cause. Damages and loss of commodity are included in external
corrosion caused group.
cFour accidents had both incorrect operation by carrier personnel and defective pipe seam as the cause. Damages and loss
of commodity have been included in the incorrect operation by carrier personnel caused group.
Occidents with property damages of $151,000; $130,650; $85,000; and $94,500.
eOne accident - two deaths, three injuries, and $100,000 carrier property damages; another accident - one death and $300,000
carrier property damage; and another accident - $125,000 carrier property damage.
fOne accident had total damages of $275,000.
§Two accidents had a commodity loss of 10,048 barrels (4,000 and 6,048).
^Three accidents had a commodity loss of 32,789 barrels (12,802; 10,000; and 9,987).
•j-One accident - four deaths and 25,500 barrels of commodity lost.
JOne accident - nine injuries and $325,000 other property damages.
^One accident had two tanks (1917 and 1964) struck by lightning, destroyed by fire - $295,000 carrier property damage.
-'-One accident - one injury and $250,000 carrier property damage.
-------
TABLE 35. PIPELINE ACCIDENT SUMMARY - BY COMMODITY INVOLVED3
COMMODITY
Crude Oilb
Gasoline0
Fuel Oil
L.P.G.d
Jet Fuel
Natural
Gasoline
Kerosene
Anhydrous
Ammonia
Condensate
Transmix
TOTAL
NO. OF PERCENT
ACCIDENTS OF
TOTAL
135
41
29
26
7
7
4
3
2
1
255
52.9
16.1
11.4
•10.2
2.7
2.7
1.6
1.2
0.8
0.4
100.0
LOSS
IN
BARRELS
105,871
30,843
24,386
105,228
4,249
21,475
25,030
1,311
850
180
319,423
PERCENT
OF
TOTAL
33.14
9.7
7.63
32.94
1.33
6.73
7.84
0.41
0.27
0.01
100.00
PROPERTY DAMAGE ($)
CARRIER OTHER TOTAL
$1,293,923
771,223
38,401
269,870
462
916
6,919
10
475
0
$2,382,199
$147,841
134,281
14,000
416,000
5,000
94,210
2,150
1,000
100
300
$814,882
$1,441,764
905,504
52,401
685,870
5,462
95,126
9,069
1,010
575
300
$3,197,081
PERCENT DEATHS
OF CARRIER NON
TOTAL EMPLOYEES EMPLOYEES
45.1
28.3
1.6
21.5
0.2
2.96
0.28
0.03
0.02
0.01
100.0
1
0
0
2
0
0
0
0
0
0
3
0
0
0
4
0
0
0
0
0
0
4
(continued)
INJURIES
CARRIER NON
EMPLOYEES EMPLOYEES
0
0
0
3
0
0
0
0
0
0
3
1
1
0
10
0
0
0
0
0
0
12
-------
TABLE 35 (Continued)
aReference 59
^Accidents—$300,000 property damage and one carrier employee death; $151,000 property damage; $295,000 property damage
and 11,356 barrels of commodity lost; 15,000 barrels of commodity lost; $250,000 property damage and one nonemployee injury;
$85,000 property damage; and $125,100 property damage.
cAccidents—$59,431 property damage and 4,122 barrels of commodity lost and $775,000 property damage.
^Accidents—-$58,000 property damage, four nonemployee deaths, and 25,500 barrels of commodity lost; $130,650 property
damage and one nonemployee injury; $325,055 property damage and nine nonemployee injuries; and $100,000 property damage,
two carrier employee deaths, and three carrier employee injuries.
-------
Pipeline spills were the second largest source of oil pollution in
coastal waters.60 In 1971 there were 1,436 leaks spilling a total of 897,685
gallons. An evaluation of the spill data showed that approximately 90% of the
offshore spills and 97% of the oil spilled comes from pipelines leading to
wells less than three miles offshore. Most of the pipeline spillage is the
result of failure of old pipelines.
Pipeline accidents are caused by a number of factors including equip-
ment rupturing lines, internal and external corrosion, structural defects
(seams, welds, repairs), human operating errors, vandalism, and natural
causes. According to the Department of Transportation (DOT), Office of Pipe-
line Safety Operations, equipment rupturing lines (28.6%) and external cor-
rosion (22.4%) accounted for about half of the accidents during 1975.59 These
accidents caused an estimated $757,500 in property damage and a loss of 120,797
barrels of oil, or 38% of the total oil lost in pipeline accidents.
Approximately 43% of the 73 accidents caused by equipment rupturing were
the result of bulldozer, grader, and ditching machine operation. Of these 73
accidents, 47% involved the following activities:
1. Pipeline and cable construction (12 accidents, 16.4%),
2. Moving earth for commercial firms or local government (11 accidents,
15.1%), and
3. Clearing or grading of private land (11 accidents, 15.1%).
The DOT statistics also showed that 78% (57 accidents) of the corrosion
caused accidents were due to external corrosion. This would be expected to
occur with old pipelines since newer ones generally have protective coatings
and cathodic protection. Seventy-two percent were in equipment installed be-
tween 1920 and 1949. However, 75.5% involved lines with cathodic protection.
External corrosion accidents accounted for approximately $115,800 in property
damage but resulted in no death or injury. Two-thirds of external corrosion
accidents involved crude oil in the pipeline while the remaining involved
gasoline (15.8%), fuel oil (12.3%), and jet fuel, furnace oil, and transmix
(5.3%).
4.2.5 Transportation (Truck and Railroad)
Any unintentional release of hazardous materials in transportation must
be reported to DOT's Office of Hazardous Materials Operation (OHMO). For pur-
poses of perspective, a summary of hazardous materials rail and highway in-
cident reports involving death or injury for 1974 and 1975 are presented in
Table 36. 61
111
-------
TABLE 36. RAIL AND HIGHWAY ACCIDENTS3
1974 1975
No. of No. of
Reports Deaths Injuries Reports Deaths Injuries
Rail 68 10 593 74 0 95
Highway 170 18 276 193 27 489
a Reference 61
These figures are based upon immediate notifications which are required when
the release of hazardous materials results in death, injury, estimated proper-
ty damage exceeding $50,000, or spillage of radioactive materials or etiolgic
agents.
According to the DOT, 10 accidents accounted for 22 of the fatalities and
239 injuries. Of these 10, one can be classified as transportation by tank
truck for non-end use. This accident occurred at Eagle Pass, Texas, on April
29, 1975, and involved loss of control of a liquified petroleum gas (LPG) tank
truck. Explosion followed causing 15 deaths and 45 injuries. This one acci-
dent caused 56% of the total highway fatalities, and 9-2% of the injuries for
the year. Of the hazardous commodities most frequently involved in accidents
reported for 1975, gasoline accounted for 12.8% and LPG for 1% of the estimated
8,088.
LPG consists primarily of propane maintained as a liquid at elevated
pressures. LPG is usually transported via truck, rail car, or tanker. Sever-
al major accidents involving transport of LPG have occurred, including the
Eagle Pass incident cited above.
One of the most severe types of accidents which can occur is the BLEVE
(Boiling Liquid Expanding Vapor Explosions). This occurs when a leak of LPG
is ignited. The resulting fire heats tanks which have not been damaged. As
long as the flame is heating the bottom portion of tank containing the LPG,
the relief valve at the top will release the pressure from the resulting vapor.
If, however, the flame is heating the part of the tank where the vapor is,
extreme temperatures will cause the tank to melt and tear. When this happens,
the vapor escapes and explodes. The explosion can have sufficient force to
hurl projectiles over half a mile, generate fireballs over 1,000 feet high,
and cause blasts felt over 4 miles away. One of the worst instances occurred
in Kingman, Arizona on July 5, 1973. Ten minutes after a fire started from
an LPG lead, a BLEVE occurred. Half of the tank truck was hurled 1,200 feet.
Twelve firemen were killed and 95 bystanders were injured.
API has summarized motor vehicle accidents in the petroleum industry for
1975. According to their classification, Marketing - Wholesale includes bulk
distribution of petroleum products, including truck and transport deliveries
and related administrative activities. The figures, however, include retail
112
-------
marketing operations such as transportation of batteries and tires to market-
ing outlets. According to API there were 8 deaths, and 3,356 injuries of
which 1,799 were lost time accidents among employees of the industry.
Rail accident data are generally all inclusive and not limited to type
of cargo. DOT statistics for rail accidents for 1971 through 1975 involving
hazardous materials are shown in Table 37.61
An analysis of accident risks pertaining to hazardous materials ship-
ments by rail, truck, and barge has been completed.62 Although the analysis
focused on damage to cargo in the nuclear industry, many of the results apply
to shipment of non-nuclear cargo.
The analysis was based on 1969 data for accidents involving impact, fire,
or both. Arbitrary categories of accident severity were established in order
to determine a relationship between accident rates and severity for the dif-
ferent modes of transportation. Auto accidents and grade-crossing accidents
were not considered. (Rail cargo is rarely affected by grade crossing ac-
cidents. )
The probabilities for rail, truck, and barge accidents are shown in
Table 38."2 The accidents are categorized by degree of severity in terms of
velocity of vehicle impact and incidence and duration of fire. The totals
are a summary of each accident-severity category. Fire duration was believed
to be unlikely to exceed 1/2 hour in all cases other than those involving
ruptured tank cars or flammable liquids. It was estimated that 1.5% of all
rail accidents involved fires of which 85% last less than 1/2 hour and 1%
longer than one hour.
In 1969, for a total of 61 billion rail car miles, there were 4,971 ac-
cidents, other than at grade crossings, involving death, injury, or property
damage in excess of $750. The overall accident rate was 0.15 train accidents
per million car-miles. For other than rail-crossing accidents this dropped
to 0.08 per million car-miles. Each accident involved an average of 10 rail
cars. Most of the reported accidents were derailments (70%) and collisions
(21%). Accidents with the highest probability of producing significant damage
to shipment containers were those involving derailments.
For trucks, 38,813 accidents were reported in 1969 involving death, in-
jury, or property damage in excess of $250. The overall accident rate was 2.5
accidents per million vehicle-miles while for accidents involving hazardous
materials the rate was 1.7 per million vehicle-miles. Package damage was
severe in all types of truck accidents, but most severe in collisions with
rigid, stationary objects and roll-overs. Fire generally occurred in 0.8% of
truck-truck collisions, 0.3% of truck-auto collisions, 0.6% of truck-fixed-
object collisions, 2% of truck-train collisions, and 1% of roll-over accidents.
Major fire of more than 1/2 hour duration generally occurred in the case of
the truck-truck accident where significant amount of fuel was available (e.g.,
a tank loaded with a fuel). Of all vehicle accidents, 50% involved truck-auto
or bus collisions, while 14% involved collisions with fixed objects, 0.6% with
trains, and 9.5% involved roll-overs.
113
-------
TABLE 37. SUMMARY OF RAIL ACCIDENTS INVOLVING
HAZARDOUS MATERIALS, 1971-1975a
YEAR
1971
1972
1973
1974
1975
NUMBER
OF
ACCIDENTS
80
99
123
148
186
PERSONS
KILLED
1
0
2
10
0
PERSONS
INJURED
57
226
384
613
20
NUMBER
OF
EVACUA-
TIONS
16
18
32
28
19
PERSONS
EVACUATED
3,005
5,274
29,647
10,911
4,755
aReference 61
114
-------
TABLE 38. ACCIDENT PROBABILITY3
SEVERITY
CATEGORY
Minor
VEHICLE
SPEED
(mph)
0-30
0-30
30-50
FIRE *
DURATION
(hr)
1 7.0xlO~n
>1 3.9X10"11
1/2-1 5.1xlO~10
1/2-1 1.5xlO"10
1 l.lxlO"11
1/2-1 1.6xlO~12
l.SxlO"11
>1 1.2xlO~13
1.2xlO~13
5X10'12
IxlO"11
IxlO'10
6xlQ-12
IxlO"10
8xlO~9
8xlO~9
6xlO"13
2xlO~13
8xlO"13
2X10'14
2xlO'14
probabilities are based on the duration of
__
9.3X10-J1
1.3x10 9
3.3xlO"10
_
-
1.6xlO~9
2.3X10"11
-
2.3xlO~1:L
-
-
the fire
and actuarial data on cargo damage. The impact velocities of all
barge accidents were considered to be less than 10 mph, but for
the purpose of this table, minor cargo damage is assumed to be
equivalent to land vehicle impact speeds of 0-30 mph; moderate cargo
damage, 30-50 mph; and severe cargo damage, 50-70 mph.
115
-------
Analysis of the data indicates that the probability of an accident
occurring generally was similar for each mode of transportation. Therefore
a single value for each accident-severity category can be used in general
accident risk assessment. The probability that an accident with a particular
severity will occur is:
Minor Moderate Severe Extra-Severe Extreme
1 x 1(T6 3 x 10-7 8 x 10-9 1 x 10-H1 x 10-13 -
on a per vehicle-mile basis.
Accidents involving truck loading racks were not considered in the prob-
ability analysis. According to the Oil Insurance Association, losses from ex-
plosion and fire at loading racks are small considering the large volume of
flammable petroleum liquids which are handled daily in this manner.64 However,
an accident at a loading rack may cause dollar losses greater than $100,000 and
loss of life. A study indicated that static electrical ignition, human error,
and equipment failure accounted for the largest portion of the overall loss
total in leading rack accidents.64 Other causes were electrical ignition,
vehicle damage, and wind and lightning. Contributing factors were:
1. Fast loading rates (between 500 and 1,200 gpm) which generates static
electricity,
2. Extensive use of product filters which are producers of static
electricity,
3. Dryer, cleaner products (especially in the distillate range) which
retain static charges for longer period of time,
4. Increased handling of more hazardous products,
5. Extensive use of aluminum tank trailers of larger capacity which
can fail structurally within a minute when exposed to intensive
ground fires,
6. Larger, more complicated loading racks,
7. Limited availability of land causing congestion and undue exposure
to other facilities and tankage,
8. Key-stop (unattended) operations, which because of reduced super-
vision, reduces the possibility of controlling fires once started,
and
9. Inexperienced staff because of high labor turnover.
4.2.6 Storage Tank Accidents
The storage of large quantities of crude oil or products at various
points in the crude oil chain from production to retailing present a potential
116
-------
for a major accident. This type of accident involves fire and explosion and
generally results in loss of property rather than life,
Tank fires generally are caused by overfilling, poor housekeeping and
maintenance practices, and lightning. The severity of the resulting fire
depends on tank spacing, common diking, piping within dikes, and availability
of water and foam to fight fires.
According to the Oil Insurance Association (OIA) the potential for large
dollar losses has increased due to the increase in crude oil prices and the
resulting increase in the value of finished products.5 in addition, there is
a trend toward jumbo tanks, those larger than 300,000 barrel capacity. Tanks
of from 300,000 to 1,000,000 barrels are being used for storing crude oil as
well as intermediate and finished products, the latter with various ranges of
flammability.
The risk of accident can present a serious loss exposure problem as illus-
trated by the following example involving a 500,000 barrel tank filled with
crude:
Value of contents @ $10 per barrel $ 5,000,000
Cost of rebuild after a fire @ $4 per barrel
of capacity $ 2,000,000
Total loss exposure $ 7,000,000
Since the trend is to install the tanks at least in pairs and often in groups
of four to ten or more, the potential loss exposure can be high.
4.2.7 Refinery Accidents
The refinery, which contains a number of process units which produce
petroleum products (e.g., LPG, gasoline, jet fuel, diesel fuel, distillate
fuel oil, residual fuel oil, etc.) is the source of many accidents. Major
accidents involve fire and explosion wihch result in extensive property damage
and may cause death and injury.
Many injuries and deaths are associated with refinery accidents. API
data for the refining element indicates that in 1975 there were 5,915 injuries
and illnesses associated with the job, of which 1,846 were lost time accidents
(approximately 1/3), and 3 were fatalities. However, data from the National
Petroleum Refiners Association indicates that there were 8,155 injuries, of
which 2,623 were lost time accidents, and 11 were fatalities.65
Important secondary problems of the refinery accident involving fire and
explosion are air pollution and oil spills. They create a potential problem
to both the worker and the bystander in terms of health and property damages.
The air pollutants characteristic of an oil fire are carbon monoxide
(CO), sulfur dioxide (S02), oxides of nitrogen (NOX), hydrocarbons (HC), and
117
-------
particulates. While the cloud formed as a result of fire and explosion gener-
ally dissipates rapidly, prolonged exposure of these pollutants can cause at
least temporary illness. An atmospheric inversion in the area of such an oc-
currence would prevent the normal dispersion of the pollutants, thus increasing
the change of exposure.
Hydrorefining units, such as the hydrocracker and hydrotreater which use
hydrogen, are widely used in petroleum processing. Statistics compiled by the
OIA indicate that hydrorefining units have the greatest potential for accidents,
in terms of frequency and severity, of any current refining process."-3
Losses due to accidents for various processes from 1965-1969 are summarized
in Table 39.
TABLE 39. ACCIDENT LOSS SUMMARY FOR VARIOUS PROCESSES, 1965-1969
Process
Catalytic Cracking
Catalytic Reforming
Hydrocracking
Crude/ Vacuum Units
Loss Summary
No. of
Accidents
32
59
19
57
1965-1969
Total Amount
of Losses
$3,308,000
3,134,000
6,402,000
1,366,000
Approximate
Average Loss
$103,000
53,000
337,000
24,000
There are several factors to be considered with regard to figures in Table 39.
1. At the end of 1969 there were 5 1/2 times as many catalytic crackers
as hydrocrackers. Therefore, hydrocrackers had a loss frequency rate
more than 3 1/2 times that of catalytic crackers, with a severity
rate more than triple.
2. Almost all catalytic crackers on stream at the end of 1969 had oper-
ated throughout the full five years but few hydrocrackers had been in
service that long.
3. The loss frequency rate for hydrocrackers was 3 times that of cataly-
tic reformers, while the severity rate was more than 6 times greater,
there being 6 1/2 times as many catalytic reformers on stream as hy-
drocrackers .
4. Until the time of this study, the catalytic reformer had been the
most hazardous process in terms of frequency and size of losses.
118
-------
Hydrocracker losses since 1970 averaged well over $1,000,000. Loss trends for
hydrocrackers, in terms of frequency and severity from 1962-1971, are shown
in Figure 7-*
The principal cause of hydrocracker accidents was failure of the heater
tube in the hydrotreater section. A much less important cause was charge
pump failure. Accidents due to other causes were negligible.
Alkylation units, which chemically combine two hydrocarbon molecules in-
to a single molecule, have been considered by the Oil Insurance Association
(OIA) to have had a relatively good accident record. However, accidents in-
volving these units can cause significant property damage. Because volatile,
heavier than air chemicals can be involved in the accidents, large areas can
be affected by fire and explosion.
There are two types of alkylation units, the sulfuric acid (H2SO^) and
hydrofluoric acid (HF). In 1970 there were approximately 70 of each type in
the United States and Canada combined. The H2S04 units had an average pro-
duction capacity of approximately 7,500 barrels per stream day compared with
3,500 barrels per stream day for the HF units. Large units of both types are
becoming increasingly common.
During the period from 1961 to 1972 there were 18 known alkylation unit
accidents. Fiye were ^SO^ units and 13 HF units, or greater than a 2:1 loss
of HF units. In addition, there have been no known H^SO^ unit losses since
1967, whereas HF unit losses began in 1965 and increased in frequency and
severity ever since. HF unit accidents may be due to their age, or deficiencies
in operating or maintenance procedures. Nearly 25% of the losses were caused
by drain valves being left open and unattended.
The 18 unit accidents represent damage in excess of $6,800,000. Three of
the units were covered by Business Interruption Insurance, which represented
an additional $1,400,000. Four of the 18 accidents caused property damage in
excess of $1 million. These 4 units were all less than average size.
A study was made of losses exceeding $10,000 occurring in gasoline plants,
which may or may not be part of a complete refinery."' A determination was
made of the equipment which initiated the accident and the dollar amount of
the loss for each type (Table 40). Pumps accounted for 38% of the damage in
dollars but only for 10% of the number of accidents.
Damage from individual accidents can exceed $1,000,000. For example,
in one incident a fire in a lean oil.pump area resulted in extensive damage to
a refrigeration absorption gasoline plant. The fire was extinguished in one
hour but it caused $2.3 million in damage and 6 months downtime.
Statistics for hydrorefining units covered the period from 1965 through
1969.51 Data compiled after 1970 is not comparable to data for the prior
period because only larger losses were reported due to higher deductibles.
However, frequency and severity of hydrocracker accidents probably have
continued to increase.
119
-------
1750
1500
Frequency
Severity (Avg. loss in $000)
1250
1000
750
500
250
14
1962
Reference 65
1963
1964 1965 1966 1967 1968
Figure 7. Loss trends in hydrocrackers — 1962 to 1971a
1969
1970
1971
-------
TABLE 40. LOSSES OVER $10,000 OCCURRING IN GASOLINE
PLANTS DURING THE YEARS FROM 1959 THROUGH 1971a
INITIATING
EQUIPMENT
Heaters
Pumps
Gas Compressors
Tanks
Piping Failures
(Includes Gasket Failures)
Boiler Fire Box Explosions
Service Buildings
Other
Wind
Hail
TOTAL
aReference 67
NUMBER OF
ACCIDENTS
32
13
23
6
11
7
12
10
16
4
134
LOSS (DOLLA1
1,001,700
6,271,700
3,387,700
1,721,900
2,718,500
150,000
262,000
401,100
390,000
252,900
16,558,300
121
-------
Special consideration was given to the fired heaters as a major source
of accidental fire by OIA."^ These heaters find application in refineries,
petrochemical plants, and other oil and gas facilities. Insurance company
figures show that more than. 200 such fires cost the industry more than $9
million in property damage and production losses between 1965 and 1969.
These heaters are found in a number of process units, including those
previously discussed. Table 41 shows these losses in terms of each type
of unit.
122
-------
TABLE 41. REFINERY PROCESS UNIT LOSSES ORIGINATING
IN FIRED HEATERSa
% OF PROCESS UNIT % OF LOSSES IN EACH
LOSSES ATTRIBUTABLE CATEGORY ORIGINATING
PROCESS CATEGORY TO EACH CATEGORY IN FIRED HEATERS
Hydrocracking 7.6 43.7
Catalytic 15.2 18.7
Catalytic Reforming (Including 27.9 44.1
Hydrotreating)
Polymerization 5.2 9.0
Isomerization
Alkylation
Solvent Refining 0.5 100
Crude, Vacuum Distillation 27.0 30.0
Thermal Processes (Coking, 16.6 22.9
Visbreaking)
TOTAL 100.0
Reference 68
123
-------
SECTION 5.0
NATURAL GAS
5..1 RESOURCE SYSTEM OVERVIEW
Natural gas is composed primarily of methane (CH^) with small amounts of
ethane (C2H6), butane (C4H10), propane (CsHg), and heavier hydrocarbons. Nat-
ural gas may also include water, gaseous sulfur containing compounds, nitrogen,
and carbon dioxide. It may occur three ways: non-associated (very little con-
junctive crude oil), dissolved in crude oil, or associated (as in the case of
a gas-cap where the gas accumulates in the area overlying the oil pool).
In 1950, the total annual consumption of natural gas in the United States
totalled 6,150 trillion BTUs. By 1970, consumption has increased to 22,029
trillion BTUs. This amounted to annual increase of 6.5%. Other energy demands
grew only by 3.5% during the same time period. In 1950 natural gas supplied
18% of the total energy demand. In 1970, this had grown to 32%.
In 1974 the Federal Power Commission estimated the domestic supply of the
lower 48 states and Alaska to be from 1,000 to 2,955 tcf. The proven reserves
at the close of 1973 totalled 250.0 tcf for the entire United States. Of this,
218.3 tcf were attributed to the lower 48 states with approximately 36.1 tcf
offshore and 182.2 tcf onshore. Approximately 88% of the natural gas reserves
in the lower 48 states are located in Texas, Oklahoma, Louisiana, New Mexico,
and Kansas. At the current rate of production, the reserves would be exhausted
in 11.1 years.
5.1.1 Exploration
The elements in the natural gas energy cycle are shown in Figure 8. Ex-
ploration and production of natural gas is similar to exploration and extrac-
tion of crude oil (Sections 4.1.1 and 4.1.2). These functions are briefly
reviewed below.
The initial step in locating new oil and gas reserves involves identify-
ing promising geological formations. This is followed by an in-depth study of
each possible site and finally exploratory drilling. The "bright-spot" tech-
nique using seismic and radioactive methods has proven useful in the detection
of non-associated and associated gas reservoirs.
Exploratory drilling is required to determine whether gas and oil exist
in larg e enough quantities to be profitable. Drilling of offshore forma-
tions presents more problems than onshore. Drilling equipment is mounted on
a platform. These platforms are barges, drill ships, semi-submersibles, or
jackups. Barges are used for shallow water drilling. Drill ships are used
124
-------
EXPLORATION
EXTRACTION
PROCESSING TRANSPORTATION
STORAGE
DISTRIBUTION
1. drilling
2. seismic and
radioactive
testing
expansion of
exploratory
wells to
production
wells
removal of
impurities
separation
from oil
fractiona-
tion
local
transmission
lines
liquefaction
(LPG)
1. above ground
tanks
2. underground
tanks
Figure 8. Natural gas energy system.
-------
in depths of 500 to 3,000 feet. A jackup is self-elevating and has buoyant
hulls so that the platform drilling rig can be floated to the site. When it
reaches the site, the legs are jacked down to the ocean floor and the plat-
form elevated to the appropriate height. A semisubmersible is similar to a
floating rig and is supported by displacement hulls or caissons.
5.1.2 Production
After discovery of natural gas, the well is tested to determine the size
of the reservoir and the possible flow. Additional wells are drilled, care-
fully placed so as to control the flow and drain the reservoir as completely
as possible. Completed development wells contain casing and tubing to carry
oil and gas to the surface. In seismically active areas and offshore sites,
a safety valve is installed near the bottom of the production tubing. This
valve is activated to shut off flows when downhole pressures become uncon-
trolled.
The oil and gas is forced to the surface either by natural pressure or
by artificial means. Artificial means include pumping and injection of
another fluid.
In order to develop an offshore field a fixed drilling platform is con-
structed. From this, between twenty and forty production wells are drilled.
After all the production wells are drilled, the drilling equipment is re-
moved and production equipment is installed. Subsea production systems
are becoming increasingly common. These systems involve placing well-heads
on the ocean floor rather than on platforms. Oil and gas is then pumped to
a nearby fixed platform or shore facility.
5.1.3 Processing
The processing of natural gas involves its separation from associated
oil, water, sand, sulfur, carbon dioxide, and other impurities. The actual
processing method chosen depends upon the composition, location of source,
and end use. Usually, the fluids are first treated to remove water and sand
by physical separation, use of dehydrants, and skimming. After removal of
impurities, the fluids are sent through a sequence of separators to separate
the oil from the gas. Hydrogen sulfide (t^S) is removed by extraction with
ethanolamine. Because of the corrosive and toxic nature of hydrogen sulfide,
special care is taken with gases containing H2S. The gas is then compressed,
fed through a gathering system, and into a high pressure transmission line.
If the gas is rich in methane, it may be processed at natural gas plants
rather than in the field. Besides separating the oil from the gas, these
plants separate the liquid into fractions. The plants produce liquified
petroleum gas (LPG), gasolines, ethane, and heavier liquid hydrocarbons. Much
of the gas produced is used to supply the processing and field energy require-
ments.
If oil and gas are produced jointly, separation must occur in offshore
facilities as well as onshore. If the gas contains water, it is dehydrated by
contact with glycol, pressurized, metered, and pumped to shore by pipeline.
126
-------
If there is no pipeline, or it is economically unattractive to produce the
gas, the gas is pressurized and reinjected into the reservoir. A typical
production facility and associated safety devices are shown in Figure 9.
5.1.4 Transportation
Domestic gas is transported by pipeline. Pipeline systems have diam-
eters from 12 to 42 inches and are usually constructed of steel pipe. They
are usually buried from 2-4 feet below the surface. The gas is compressed
to 600-1,000 psig and the resulting pressure drives the gas through the
pipeline. Valves are installed every 10 to 30 miles along the pipeline to
enable isolation of a section if a large drop in pressure occurs. Pipelines
are coated to protect them from chemical corrosion. In urban areas care
must be taken to prevent corrosion from electrical currents. Stray electrical
charges may cause loss of metallic ions, resulting in a weakening of the pipe-
line. One method to guard against this is cathodic protection in which a
direct current is applied to the pipe.
Construction of offshore pipelines presents special problems. In the
past, pipelines were laid directly on the ocean bottom. Now if the water
depth is less than 200 feet, the pipelines are buried to minimize potential
of damage from natural forces and marine equipment. Pipes are lowered by
a barge into trenches dug by high pressure water jets. Because of access dif-
ficulty once laid, pipes are very carefully inspected.
5.1.5 Storage
Natural gas is often stored at its destination for periods of peak
demand. In 1974, there were approximately 367 underground storage fields
with a total capacity of 6,364 billion cubic feet. Often gas is stored under-
ground in depleted oil or gas reservoirs, aquifers, caverns, and sealed mines.
These areas must be gas tight, have a large capacity, and support high produc-
tion and intake rates. Because of small capacity, storage of natural gas
above ground in tanks is used only on a local basis to meet daily peak demands.
During times of extreme demand, peak-shaving plants are operated. These in-
troduce a mixture of air and LPG to the gas stream.
5.1.6 Distribution
The local distribution of gas to commercial residential users involves
a system of mains, valves, regulators, and meters. The gas usually enters
the local distribution system through city gate stations at pressures of 100-
150 psig. Mains operating at 25-35 psig feed individual customer lines at
0.25-0.35 psig.
5.2 ACCIDENT OVERVIEW
Table 1 presents a summary of the accidents associated with the explora-
tion, extraction, processing, transportation, and storage of natural gas and
their related severity and frequency. Blowouts of well-heads during the drill-
ing of exploratory and production wells, release of sulfur compounds during
127
-------
surface•
NJ
oo
•»< *—**-
u
gas
separator
well
*1
I
7.2
compressor
in
goilf:
o11
free H20
knockout
5' 6
gas
ioili
I
gas
customer
6
-*4-
2 glycol contactor
skim
tank
H.O treat
disposal
)(^customer
1. Subsurface safety device
2. High/low pressure sensors
3. High/low level sensors
Pressure relief valves
Flow check valves
Automatic valves
7. Combustible gas sensors
8. Manual emergency shutdowns
4.
6.
Figure 9. Typical production facility with safety equipment.8
Reference 42
-------
processing, and failures of pipelines from corrosion or outside forces com-
prise the bulk of the accidents occurring from this energy source. Sudden,
uncontrolled release of gas may result in fires and explosions causing injury
to nearby persons and damage to equipment. Accidents affecting the general
public may arise from pipeline failures because these transect residential and
commercial areas.
Table 42 presents injuries and man-days lost per billion BTUs in the
natural gas industry.15 Pipeline distribution accounts for the greatest num-
ber of injuries and man-days lost at 0.0138 injuries/1012 BTU and 0.324 man-
days lost/1012 BTU, Onshore extraction accounts for the largest number of
deaths at 8.1 x 10 /1012 BTU.
Table 43 shows the occupational injuries and illnesses for 1975 for na-
tural gas operations subject to OSHA record keeping requirements.3^ Drilling
had the highest frequency at 48.93 disabling injuries per million man hours
worked. This is an order of magnitude higher than for exploration and produc-
tion, processing, and pipeline transportation.
Table 44 shows the annual impact of an uncontrolled gas-fired electricity
generating system associated with a 1,000-megawatt power plant with a load
factor of 0.75.1 Extraction is responsible for the greatest number of deaths
and injuries. The total number of annual deaths was 0.20 with 0.16 attributed
to the extraction of natural gas. The total number of injuries was 18.3 with
16.0 attributed to extraction. Total number of lost work-days was 1,986 with
1,629 attributed to extraction.
The results of the American Petroleum Institute Review of Fatal Injuries
in the Petroleum Industry for 1975 are shown in Table 45.35 A total of 12
fatalities occurred in 1975. Seven were attributed to exploration and produc-
tion. (There is, however, no distinction made between oil and gas.) No fatal-
ities occurred in gas processing, 2 occurred in drilling, and 3 occurred in
pipeline transportation of gas.
Table 46 presents data on employee disabling injury frequency and sever-
ity by type of gas between the years 1970 and 1974.6^ The frequency for all
natural gas in 1974 was 9.08 where frequency is defined as number of disabling
injuries per million man-hours worked for each injury resulting in a minimum
on one day's loss of time. The severity rate was 357 where severity is de-
fined as number of days lost due to disabling injuries per million man-hours
worked.
The severity and frequency rates for gas utilities are compared to other
industries in Table 47.69 The utilities had the lowest severity rate but the
second highest frequency rate.
5.2.1 Exploration Accidents
Exploration accidents occasionally result from improper use of explosives
or radioactive pellets. One to three such incidents are reported each year.
129
-------
TABLE 42. ACCIDENT RATES FOR EXTRACTING, GATHERING, PROCESSING,
TRANSMITTING, DISTRIBUTING, AND STORING NATURAL GASa
Extraction
Offshore
Onshore
Gathering
Pipeline
Processing
Natural Gas Liquid
Hydrogen Sulfide
Transmission and
Distribution
Pipeline
LPG Trucks
Storage
Underground
Gas Holders
ACCIDENT
FATALITIES
6.9 x 10~6
8.1 x 10~5
3.1 x 10~6
4.1 x 1CT5
2.1 x 10~6
4.0 x 10~5
U
ub
U
RATE PER 1012
INJURIES
0.003
0.004
0.001
0.004
0.002
0.0138
U
U
U
BTU EQUIVALENTS
MAN-DAYS LOST
0.01
0.13
0.025
0.097
0.005
0.324
U
U
U
Reference 15
Unknown
130
-------
TABLE 43. REPORTS OF OCCUPATIONAL INJURIES AND ILLNESSES FOR THE YEAR 1975
COVERING OPERATIONS SUBJECT TO OSHA RECORDKEEPING REQUIREMENTS ONLY'
NUMBER OF
FUNCTION EMPLOYEES
Exploration & 54,598
Production"
Drilling0 1,918
Gas Processing11 5,927
Pipeline-Gas6 16,736
Total 79,179
RECORDABLE CASES TOTAL LOST NON-FATAL INCIDENCE RATE*
INJURIES ILLNESSES TOTAL FATALITIES WORKDAY CASES LOST WORKDAY NON-FATAL
CASES WITHOUT LOST TOTAL CASES CASES W/0 LOST
WORKDAYS WORKDAYS
2,466 36 2502 9
515 0 515 2
299 1 300 2
921 28 949 3
4,201 65 4266 16
763 , 1730 4.4 1.35 3.07
196 317 25.45 9.69 15.67
69 229 4.87 1.12 3.72
168 778 5.75 1.02 4.71
1196 3054 5.27 1.47 3.77
FREQUENCY
RATE
6.57
48.93
5.61
4.97
7.23
a Reference 34.
k Same as Table 4.3 note b).
c Same as Table 4,3 note c).
" Processing of gas to produce liquid hydrocarbons (e.g., ethane, LPG, and natural gasoline).
e Gas gathering and truck/line operations of natural gas transmission lines, excluding retail distribution.
f Per 100 employees.
-------
TABLE 44. ANNUAL ACCIDENT IMPACT OF AN UNCONTROLLED GAS-FIRED
ELECTRICITY SYSTEM ASSOCIATED WITH A 1,000-MEGAWATT POWERPLANT
WITH A LOAD FACTOR OF 0.75a
EXTRACTION PROCESSING TRANSPORT CONVERSION TRANSMISSION TOTAL
Deaths 0.16
Injuries 16.0
Workdays Lost 1,629
0.01
0.05
92
0.02
1.2
145
0.009
1.09
120.4
N.A.
N.A.
N.A.
0.20
18.3
1,986
aReference
132
-------
TABLE 45. FATALITIES BY FUNCTION AND CAUSE FOR THE NATURAL GAS INDUSTRY YEARS 1975 AND 1974'
OJ
CASES
REPORTED
1975 1974
Exploration & 7
Production
Gas Processing 0
Drilling 2
Pipeline 3
Total 12
Total (1974 + 1975) =
19
0
1
_3
13
FIRE AND STRUCK BY MOTOR OTHER
EXPLOSION EQUIPMENT' VEHICLE AIRCRAFT CAUSES
1975 1974 1975 1974 1975 1974 1975 1974 1975 1974
05101
o-o o o v o
00110
0. 0 0 1 2_
05223
31645
00000
00010
1 .2 ° 1 1
41666
35 fatalities
a. Reference 35.
-------
TABLE 46. EMPLOYEE DISABLING INJURY FREQUENCY AND SEVERITY RATES. BY TYPE OF GAS, 1950-1974
a,b
ALL REPORTING GAS UTILITIES
REPORTING COMPANIES RATE
YEAR
1970
1971
1972
1973
1974
NUMBER
0
1
2
3
4
AVERAGE
EMPLOYEES
200,493
210,589
205,371
207,423
208,283
FREQUENCY
6.34
7.15
8.38
8.18
9.08
SEVERITY
548
473
.535
547
357
TOTAL NATURAL GAS
RATE
FREQUENCY
6.34
7.17
8.44
8.18
9.08
SEVERITY
554
478
542
547
357
NATURAL GAS DISTRI-
BUTION AND INTEGRATED
COMPANIES
RATE
FREQUENCY
6.66
7.48
8.98
8.77
9.75
SEVERITY
522
444
495
373
362
NATURAL GAS
TRANSMISSION
COMPANIES
RATE
FREQUENCY
5.02
5.70
5.88
5.67
6.12
SEVERITY
686
634
762
1,287
333
MANUFACTURED AND
MIXED GAS
RATE
FREQUENCY
6.09
6.29
5.42
b
b
SEVERITY
236
208
183
b
b
a Reference 69.
k Frequency rate is number of disabling injuries per million man-hours worked for each injury resulting in a minimum of one day's loss of
time. Severity rate is number of days lost due to disabling injuries, per million man-hours worked.
-------
TABLE 47. EMPLOYEE ACCIDENT FREQUENCY AND SEVERITY RATES OF
SELECTED INDUSTRIES, 1974a
Gas utilities
Electric utilities
Mining, surface0
Mining, underground coalc
Petroleum
FREQUENCYb
10.20
7.41
9-75
35.44
6.73
SEVERITY^
614
1
942
1,365
5,154
690
aReference 69
^Frequency rate is number of disabling injuries per million man-hours
worked for each injury resulting in a minimum of one day's loss of
time. Severity rate is number of days lost due to disabling injuries
per million man-hours worked.
C1972 data
d!973 data
13,5
-------
The most disastrous type of accident that can occur during exploratory
drilling is a blowout. A blowout is the result of a sudden release of pres-
sure resulting in unconstrained flow. Oil and/or gas surges up the hole caus-
ing a loss of control over the well. If the impact pressure is sufficiently
great, physical damage to the drilling rig may result. If ignition occurs,
fires will result causing damage to equipment and death and injury to persons
in the area.
Safeguards have been developed to forestall such blowouts. Drilling mud,
a viscous mixture of clay and water is circulated around the drilling bit to
counteract a sudden flow. Drill holes are encased at the top with steel pipe
set in cement to minimize the possibility of a blowout around the outside of
the drill.
Various devices are used during drilling operations and in producing wells
to reduce or eliminate a blowout. The blowout preventer (BOP) is located in a
stack arrangement outside the drill pipe between the surface casing and the
rotary drilling table. It is generally used to control flows during drilling
operations. On fixed offshore platforms they are located on the platform
above the last string of casing, while on a semi-submersible or drill ship
they may be on the sea floor, also above the last string of casing. The bag
and the ram are the two types of blowout preventers. They are operated by
fluid pressure to close the well opening in order to regain control of pres-
sures in the hole by means of drilling muds.
Subsurface safety valves are located inside the tubing of producing wells
and are either pressure activated or hydraulically controlled. Their purpose
is to control pressure in the well. A blowout preventer may be used with
a producing well if it is being drilled deeper and if there is the possibility
of losing control during workover and the downhole safety devices have been
removed.
5.2.2 Production Accidents
Blowouts are possible during the drilling of production wells as well as
exploratory wells. However, the potential is greatly lessened because the
surrounding geology is much better known and successful exploratory wells have
been drilled. Subsurface safety valves are designed to trigger an immediate
shut down of the drill hole in case of sudden change in pressure. Currently,
this is an area of special interest and much research. Several downhole valves
have failed. In 1970, 10 of 42 valves in Shell's Bay Marchand Fire failed.
In 1971, 4 of 10 valves failed at an Amoco facility leading to a fire. Be-
cause of these events USGS has modified its regulations to include as stand-
ard requirements remotely activated valves which are hydraulically operated
and controlled from the surface.
Loss of well control may occur during servicing of _rigs. Service person-
nel may not work for the company operating the rig, and consequently may be
unfamiliar with operating procedures. A new technique has been developed for
use in servicing procedures which allow insertion of downhole tools in a sepa-
rate parallel circuit while the well is sealed off. This greatly reduces the
136
-------
risk of an unscheduled release of gas. Other safety equipment installed on
production rigs and platforms include pressure, level, and combustible gas
sensors, pressure relief valves, and fire detection and fighting equipment.
5.2.3 Processing Accidents
Processing accidents are those occurring during the separation of oil and
gas and the removal of impurities. Special care must be taken in treating
sour gases, i.e., those containing I^S.
Hydrogen sulfide is odorous and corrosive whether as a free gas or dis-
solved in petroleum. Concentrations of approximately 0.005% cause sub-acute
poisoning when breathed for long periods of time. Inhalations of concentra-
tions of 0.06-0.08% have resulted in death in experiments with animals. Com-
busion of hydrogen sulfide will generate sulfur dioxide (802) and sulfur
trioxide (803). These compounds cause respiratory irritation in humans when
present in concentrations about 5 ppm. Sensitive individuals will notice
1 to 2 ppm and experience bronchial spasms at 5-10 ppm.
Table 42 shows that gas processing for hydrogen sulfide removal was re-
lated to 2.0 x 10~6 deaths per 1012BTUs and 0.002 injuries per 1012 BTUs.
Table 43 'shows that gas processing was responsible for 921 injuries and 3
fatalities in 1975.
5.2.4 Transportation Accidents
Almost all pipeline accidents can be attributed to corrosion damage by
outside forces, construction defects, or material failure. Corrosion may be
internal or external. External corrosion may result from the action of sea
water, ground water, soils, and sediments. Internal corrosion is caused by
hydrogen sulfide, water, or acidic conditions.
External corrosion is lessened by use of coatings, cathodic protection,
and use of materials which resist corrosion, such as plastic, aluminum, and
stainless steel. Internal corrosion can be lessened by interior coatings,
injecting corrosion inhibitors, and chemicals. Chemicals include oxygen sca-
vengers to reduce dissolved oxygen concentrations, dehydrants, and bases.
Several examples serve to illustrate the accidents which may occur. In
Bainbridge, Georgia, on August 26, 1975, a two-inch steel pipeline suffered
damage from an outside force. It had an open end and inadvertant opening of
a valve permitted gas to escape. The gas migrated under the pavement and into
the post office. An explosion and fire resulted killing one person in the
post office and injuring two others, °
In Hugo, North Carolina, on September 29, 1975, three fatalities occurred
from a fire and explosion resulting from the escape of natural gas into an air
conditioning control room. The accident was traced to the weight of a sewer
line causing a split in polyethylene service line carrying 45 psig 12 feet
from the building.
137
-------
The National Transportation Safety Board investigated a pipeline failure
near Monroe, Louisiana, on March 2, 1974. A 30-inch gas transmission pipeline
failed inside a 34-inch casing pipe. Gas burst from a ruptured weld in the 30-
inch line. It blasted a trench 100 feet long, 30 feet wide, and 25 feet deep.
The trench cut through a state highway. The gas ignited and the fire burned 10
acres of forest and sterilized the soil for 700 feet along the pipeline. No
deaths or injuries occurred.
The Eighth Annual Report of Pipeline Safety summarized gas pipeline acci-
dents during 1975 by distribution and transmission and gathering categories
(Table 48).'° There were 9 non-employee fatalities; 8 associated with distri-
bution and 1 associated with transmission. There were 5 employee fatalities;
all were associated with transmission and gathering. There were 200 non-
employee injuries; 191 associated with distribution and 9 with transmission
and gathering. There were 37 employee injuries; 29 associated with distribu-
tion and 8 associated with transmission. There were 1,373 failures. This
amounts to approximately 0.01 deaths per failure and 0.11 injuries per failure.
Damage by outside forces accounted for 71% of the failures, corrosion for 10%,
construction defects for 12%, and other causes for 7%. Table 49 shows the num-
ber of failures and resulting injuries and fatalities for 1970 through 1975.70
The American Gas Association has prepared a report on the analysis of
incidents reported to the Office of Pipeline Safety between 1970 and 1973. In
1970, all companies were required to file a report with the Department of
Transportation of all reportable accidents. A reportable accident is defined
as:
1. Causing death or personal injury requiring hospitalization,
2. Requiring removal of a segment of pipeline from service,
3. Resulting in a gas ignition,
4. Causing damage in excess of $5,000,
5. Significant in the judgment of the operator, or
6. Requiring immediate repair and other emergency action to protect the
public.
Table 50 shows incidents by leaks, and leaks and ruptures by age. There
were 3327 total reportable accidents in the time period or one incident/year/
1213 miles of pipeline in use.
Table 51 shows that between 1970 and 1973 there were 2274 (68.3%) acci-
dents resulted from outside forces." Corrosion caused 479 (14.4%) and 352
(10.6%) were caused by construction or natural defects. A plot of numbers of
incidents versus year of occurrence illustrates the importance of outside
forces as a cause of pipeline accidents. (Figure 10.)6
The risks to persons and property were evaluated in terms of fatalities,
injuries, cost to industry, and damage to operator's property. The number of
138
-------
700 i—
600
500
400
a
u
O
rt
W
300
200
100
Outside Forces
Corrosion
Construction Defect
1970 1971 1972
Year of Occurrence
1973
Figure 10. Number of incidents and cause versus year of occurrence.2
Reference 6.
139
-------
TABLE 48. ACCIDENTS AND CASUALTIES REPORTED BY GAS SYSTEM
OPERATORS DURING 1975a
TOTAL NO.
OF FAILURES
Distribution
Total
Subtotal by cause:
Corrosion
Damage by outside
forces
Construction defect
or material
failure
Other causes
Transmission and
gathering: .
Total
Subtotal by cause:
Corrosion
Damage by outside
forces
Construction defect
or material
failure
Other causes
Gas industry totals
979
94
744
78
63
394
44
237
88
25
1,373
FATALITIES
NON
EMPLOYEES EMPLOYEES
0
0
0
0
0
5
3
0
1
1
5
8
1
5
0
2
1
1
0
0
0
9
INJURIES
NON
EMPLOYEES EMPLOYEES
29
6
7
1
15
8
2
0
5
1
37
191
23
119
25
24
9
4
5
0
0
200
Reference 70
140
-------
TABLE 49. SUMMARY OF GAS PIPELINE ACCIDENTS AND CASUALTIES REPORTED
DURING YEARS 1970-1975a
DISTRIBUTION
FATALITIES
CALENDAR
YEAR
1970
1971
1972
1973
1974b
1975b
NO. OF
FAILURES
676
875
884
893
1,017
979
EMPLOYEES
1
6
2
1
1
0
NON
EMPLOYEES
20
36
26
32
19
8
TRANSMISSION AND GATHERING
INJURIES
EMPLOYEES
32
36
32
48
31
29
NON
EMPLOYEES
170
329
262
285
283
191
NO. OF
FAILURES
343
410
409
471
460
394
FATALITIES
EMPLOYEES
1
2
3
1
1
5
NON
EMPLOYEES
0
1
3
1
3
1
INJURIES
EMPLOYEES
8
14
23
3
7
8
NON
EMPLOYEES
8
10
13
16
13
9
a Reference 70.
b Includes data from telephonic reports to the Office of Pipeline Safety Operations (OPSO). Data from telephonic reports were not
included in accident casualty data for years 1970-1973.
-------
TABLE 50. NUMBER OF INCIDENTS BY YEAR OF PIPE INSTALLATION
AGE OF PIPELINE (YEARS)
RUPTURES
YEAR
1970
1971
1972
1973
Total
1970
1971
1972
1973
Total
N.A.b
57
47
54
81
239
135
127
119
147
528
76-105
1
5
2
2
10
6
21
9
10
46
66-75
5
4
4
3
16
14
13
13
9
49
56-65
7
10
11
8
36
23
26
18
16
83
46-55
6
27
23
30
86
LEAKS
44
61
65
66
236
36-45
10
15
28
30
83
26-35
28
33
31
47
139
16-25
45
72
64
59
240
6-15
73
93
89
105
360
LESS THAN 5
14
44
78
104
240
TOTAL
246
350
384
469
1449
AND RUPTURES
45
61
72
53
231
72
96
82
100
350
136
179
181
151
647
164
197
191
175
727
37
96
132
165
430
786
877
882
892
3327
a. Reference 6.
b. NA = Information on age not available.
-------
TABLE 51. NUMBER OF INCIDENTS BY CAUSE IDENTIFIER3
YEAR OF
OCCURRENCE
1970
1971
1972
1973
Totals
NAb
59
58
43
65
225
CORROSION
104
121
121
133
479
OUTSIDE
FORCES
465
576
630
603
2,274
CONSTRUCTION
DEFECT AND
MATERIAL
FAILURE
48
122
90
92
352
aReference 6
t>NA = Information on age not available.
143
-------
fatalities and injuries is summarized in Table 52.6 In 1973 there were 893
incidents of pipeline failure. There was 1 employee and 32 non-employee
fatalities or one fatality per 27 accidents. There were 48 employees and 285
non-employee injuries or one injury per 2.7 accidents.
Ninety-six percent of the corrosion related incidents were externally
caused. In 83% the corrosion was galvanic (associated with an electrical
current). Table 53 shows that twice as many accidents were caused by pitting
as other sources.6 Coated and cathodically protected pipe had a much lower
failure rate than pipe which was bare or was only wrapped, coated, or cathod-
ically protected.
Sixty percent of the accidents resulting from outside influences were
caused by equipment operated by an outside party (Table 54). Other causes
of damage were earth movement (14.1%), vehicle (7.3%), weather (4.8%), pipe-
line company operated equipment (3.2%), willful damage (1.4%), and other (8.8%).
A tabulation of service interruptions and causes appear in Table 55.'1
For 1974 the total number of interruptions was 71. Outside forces were again
responsible for the majority of failures; 42 of the 71 (59%) were attributed
to action of the elements and earth-moving equipment.
The American Petroleum Institute report on occupational injuries and
illnesses for 1975 indicates that out of 949 incidents, 3 fatalities, 921
injuries, and 28 illnesses resulted (Table 43).
5.2.5 Storage Accidents
Long term storage of gas involves risks of leaks and ruptures. Acciden-
tal release of gas may result in fires and explosions causing damage to equip-
ment and injury and loss of life to persons in the vicinity. Harm to the
public is also possible because these storage facilities are often in dense-
ly populated areas. Above-ground tanks are especially susceptible to damage
from outside forces. For this reason, much work has been done in the develop-
ment of underground storage facilities. Most often, these are naturally
formed cavities.
Depleted oil and gas reservoirs are the most developed forms of under-
ground storage available in the United States today. Their ability to hold a
fluid has already been proven. If the gas is to be stored at pressures greater
than those present at discovery, several safety precautions are taken. This
is because additional pressure may cause fracturing of surrounding rock for-
mations and the subsequent escape of gas. Observation wells are drilled and
completed above the storage horizon and around the flanks in the same horizon
as the reservoir so that escaping gas can be detected quickly. Corrosion
protection measures must also be taken to prevent leakage of gas. If the
storage facility is located in a populated area, equipment should be placed
underground in concrete vaults as protection against vehicles and vandals.
Aquifers may also be used for underground storage. Besides the precau-
tions enumerated above, it is necessary to insure that gas will not migrate
144
-------
TABLE 52. CONSEQUENCES OF INCIDENTS3
YEAR
1970 1971 1972 1973 TOTAL
Number of Employee Fatalities 1 6 2 1 10
Number of Employee Injuries 32 37 32 48 149
Number of Non-Employee Fatalities 20 36 26 32 114
Number of Non-Employee Injuries 170 330 262 285 1047
Number Ignited 275 493 501 505 1774
Number of Explosions Occurred 91 164 170 161 586
Number of Secondary Explosions or Fires 55 123 115 98 391
Estimated Damage in $1000 197 297 574 520 1588
Underground Facility Contributed. 70 64 100 86 320
Reference 6
145
-------
TABLE 53. NUMBER OF CORROSION INCIDENTS3
BY CORROSION DESCRIPTION
YEAR
1970
1971
1972
1973
TOTALS
NA
5
1
3
3
12
PITTING
64
72
82
93
311
GENERAL
35
48
36
37
156
aReference 6
146
-------
TABLE 54. NUMBER OF OUTSIDE FORCE INCIDENTS BY PRIMARY CAUSE3
EQUIPMENT OPERATED
YEAR N.A.b BY /FOR OPERATOR
1970
1971 1
1972
1973 1
Total 2
9
18
25
20
72
EQUIPMENT OPERATED EARTH
BY OUTSIDE PARTY MOVEMENT
330
345
349
348
1372
56
87
99
78
320
WEATHER
8
30
37
35
110
WILLFUL
DAMAGE
9
6
10
7
32
VEHICLE
35
35
51
46
167
OTHER
18
54
59
68
199
a. Reference 6.
b. NA = Information on age not available.
-------
TABLE 55. STATISTICS OF INTERSTATE NATURAL
GAS PIPELINE COMPANIES-19743
SERVICE INTERRUPTIONS OCCURRING ON THE PIPELINE SYSTEM
NUMBER OF
CAUSE INTERRUPTIONS
Action of elements — flood, hurricane, etc.
Corrosion
External
Internal
Coupling failures
External explosion
Fatigue failures
Line-pipe failure during test
Pipeline damaged by plow or other earth
moving equipment
Weld failures
Field weld
Longitudinal
Miscellaneous
Unknown
TOTAL
24
0
0
0
7
0
1
18
2
0
16
3
71
DURATION (HRS.)
338
0
0
0
15
0
30
181
6
0
78
792
1441
aReference 71
148
-------
from the storage area and contaminate potable water aquifers. To prevent
this, observation wells are closely monitored. Pilot gas injection is
started slowly, preferably using an inert gas to trace movement of the fluid.
Salt caverns are suitable areas for storing gas because they are formed
by a leaching process which creates a gas tight container. Hazards arise
from extreme pressure changes causing structural failures which may allow gas
to escape from the cavern. To be sure this does not occur, pressure changes
must be monitored constantly. If escape does occur, highly saline leachate
could contaminate potable water supplies and fresh surface waters.^2
149
-------
SECTION 6.0
LNG
6.1 RESOURCE SYSTEM OVERVIEW
Liquified natural gas, LNG, is composed of liquified methane (95%) with
small impurities of propane (1%), butane (3%), nitrogen (5%), other hydro-
carbons (0.2$). Methane becomes a liquid at -260°F. In the liquid phase it
occupies a volume only 1/600 of its gas phase volume. It is this great volume
reduction that makes liquefaction desirable because it enables large quantities
of gas to be transported across the seas. Some physical properties of LNG ap-
pear in Table 56.7-* LNG is classified as a cryogenic (from the Greek kyros
meaning icy cold and genes, born of) material and it is its cryogenic prop-
erties which render it hazardous.74 if contact occurs, the low temperature
will cause damage to human tissues. If LNG suddenly vaporizes, it may ignite
and burn, causing injury and loss of life and damage to property in the vicin-
ity.
LNG was first used during the early 1940's in peak shaving plants. A dis-
astrous accident in Cleveland in 1944 at a storage facility costing 130 lives
and $10,000,000 in damage greatly retarded growth of the industry. There are
now 23 liquefaction plants, 49 operational peak shaving plants, and 40 satel-
lite storage facilities in the United States. The total storage capacity is
12 million barrels.75
Currently, storage facilities without a liquefaction plant are supplied
by tank trucks carrying 10,000 to 11,000 gal (238-262 bbl) LNG. Transport by
barge and rail is being considered by DISTRIGAS via 30,000 barrel barges from
Staten Island to Brooklyn and Queens, N.Y. DISTRIGAS operates the only cur-
rently functioning LNG terminal. In the first nine months of 1976, twelve tank-
ers delivered 1,200,000 bbl of LNG. Four more tankers are expected before the
end of the year.
There now exist major LNG shipping routes between Algeria and -U.S., Alaska
and Japan, Borneo and Japan, Algeria and England, France, Libya and Spain and
Italy. There are proposed routes from the Persian Gulf to India and Japan,
Nigeria to the U.S., Europe and Brazil, Alaska to United States West Coast,
Carribean via Venezuela to the United States, Ecuador to the United States
West Coast, New Guinea to the United States West Coast and Japan, and Indonesia
to the United States West Coast.
Three base load plants are planned for Alaska (one is now operational) and
13 import receiving terminals are proposed with three operational, 3 under con-
struction, and 7 in planning and approval stages. The planned LNG terminals
and their status are listed in Table 57.75
150
-------
TABLE 56 - PHYSICAL PROPERTIES OF LNG AND ITS CONSTITUENTS1
Composition of typical
peak shaving LNG
(mole %)
Composition of typical
Algerial LNG (mol£ %)
Molecular weight
Melting points °C
(1 atm)
Boiling points °C
(1 atm)
Liquid specific gravity
(at boiling point)
Gas specific gravity
(air=l)
Critical temperature C
Critical temperature
(1 atm)
Evaporation latent
heat (Kcal/Kg)
Total calorific value
Natural ignition
temperature °C
Heated gas ignition
temperature C
Combustion range in
air %
Methane
CH,
95.9
87.4
Ethane Propane
C2H6
2.5
8.4
C3H8
0.9
2.4
Iso-
Butane
C4H10
0.12
0.5
Normal
Butane
C4H10
0.15
0.7
16.04 30.07 44.09 58.12 58.12
-182.5 -188.3 -187.7 -159.6 -138.3
-161.5 - 88.6 - 42.2 - 11.7 - 0.6
0.425 0.550 0.580
0.562 0.581
(normal temperature)
0.554 1.038 1.522 2.006 2.006
82.1 32.4 96.8 135.0 152.0
45.8 48.3 42.0 36.0 37.5
121.9 116.0 101.7
87.5
92.1
9,520 16,820 24,320 31,530 32,010
537 466 405
1,325
5.0-15.0
990
2.2-9.5
990
1.9-8.5
Reference
73
151
-------
TABLE 57
PROJECTED LNG IMPORT/RECEIVING TERMINALS IN THE U.S.
a, b, c
Liquefaction-
Vaporization
Company and
Plant Site
DISTRIGAS CORP.
Everett, MA
PUBLIC SERVICE ELEC &
GAS OF NEW JERSEY
Staten Is., N.Y.
COLUMBIA LNG CORP &
CONSOLIDATED SYSTEM
LNG CO, Cove Pt., MD
ALGONQUIN LNG INC.
(EASCOGAS LNG, INC.)
Providence, RI
NATURAL GAS PIPELINE
CO. OF AMERICA
(Peoples Gas)
Ingleside, TX
TRUNKLINE LNG CO.
Lake Charles, LA.
NORTHWEST NATURAL
GAS CO., Newport, OR
WESTERN LNG C0.d
Terminal Is., Los
Angeles Harbor, CA
SOUTHERN ENERGY CO.
Elba Island, GA
TRANSCO TERMINAL CO.
Racoon Is., N.J.
WESTERN LNG C0.d
Oxnard, CA.
o
Reference 75.
Includes three liquefaction plants and storage for export in Alaska.
Offshore terminal locations not included.
Only one site (if any) will be chosen for an LNG terminal. (continued)
Storage Capacity
MMcf MBbl
3250
6000
5000
6000
5500
6000
1200
7700
4000
6000
4000
974
(1x375+1x600)
1800
. (2x900)
1500
(4x375)
1800
(3x600)
1600
(2x800)
1800
(3x600)
348
2200
(4x550)
1200
(3x400)
1800
(3x600)
1100
(2x550)
Capacity
MMcf/day)
135
360
1000
200
300
375
400
100
400
(4000)
540
520
(4000)
Year of
Operation
1971
1973-1975
1976-Base
Load Plant
1973-lst tank
2 add'l. pend-
ing
1977-planned
1977-planned
1976-under
construction
48 mos. after
approval
1976-under
construction
1977-planned
48 mos. after
approval
152
-------
TABLE 57
(concluded)
Company and
Plant Site
WESTERN LNG C0.d
Ft. Conception, CA
PHILLIPS-MARATHONb
Kenai, AK
EL PASO ALASKA
Pt. Gravina, AK
PACIFIC ALASKA LNGb
CO., Cook Inlet, AK
Storage Capacity
MMcf
7700
2300
7700
3000
MBbl
2200
(4x550)
675
(3x225)
2200
(4x550)
1100
(2x550)
Liquefaction-
Vaporization
Capacity
MMcf/day)
3200
90
3375
400
Year of
Operation
48 mos. after
approval
1969
1981-planned
1979-planned
153
-------
1975
0,2
0.2
0.2
1980
2.2
1.3
1.2
1985
3.5
2.2
1.3
1990
5.0
3.5
1.3
A study on projected LNG tanker fleet growth evaluated three possible
growth rates for LNG imports.76 The three cases were:
Annual Volume Imported
(Trillions of Cubic Feet)
Case
1
2
3
The most likely growth pattern is between Cases 2 and 3, The projected
sizes of LNG tanker fleets were:
Case 1975 1980 1985 1990
2 3 29 46 84
3 3 27 29 29
The projected source and quantity of LNG arriving in the United States
for the years 1975, 1980, 1985 and 1990 are shown in Table 58. Half of the
tankers were expected to be under the American flag.
By 1980 LNG trade may account for 6% of the world's natural gas consump-
tion. According to one projection, the U. S., Europe and Japan will be im-
porting 20 bcf/day of natural gas as LNG.75
An LNG system includes a natural gas source, transportation from source
to liquefaction plant, liquefaction plant, storage and loading facilities at
the export site, transportation by tanker, unloading and storage facilities
at the import site, a regasification plant, and transmission to a major pipe-
line. Figure 11 is a diagram of an LNG resource system.
6.1.1 Exploration - See Natural Gas, Section 5.1.1.
6.1.2 Production - See Natural Gas, Section 5.1.2.
6.1.3 Processing - See Natural Gas, Section 5.1.3.
6.1.4 Transportation to Liquefaction Plant - See Natural Gas, Section 5.1.4.
6.1.5 Liquefaction
The two principal methods used for liquefaction are the cascade cycle and
the expander cycle. The cascade cycle uses a series of refrigerants to sequen-
tially lower the temperature. The refrigerants used are usually liquid propane,
ethane, and methane. Propane is used first to decrease the temperature from
86°F to -44 F. The second stage uses ethane to decrease the temperature from
154
-------
^Customer
Figure 11. LNG operations.'
Reference 375
155
-------
TABLE 58 - PROJECTED LNG IMPORTS'
Year Export Country
Case 2
1975 Algeria
1980 Algeria
Alaska
Trinidad
Total
1985 Algeria
Alaska
Trinidad
Nigeria
Venezuela
Total
1990 Algeria
Alaska
Trinidad
Nigeria
Venezuela
Australia
Indonesia
U.S.S.R.
Total
Case 3
1975 Algeria
1980 Algeria
Alaska
Total
1985 Algeria
Alaska
Trinidad
Total
1990 Same as 1985.
Volume
Tri. cubic ft/yr.
0.2
1.0
0.2
0.1
1.3
1.2
0.2
0.1
0.4
0.3
2.2
1.2
0.2
0.1
0.4
0.3
0.2
0.4
0.7
3.5
0.2
1.0
0.2
1.2
1.0
0.2
0.1
1.3
Number of
Tankers
3
25
2
__2
29
28
2
2
10
4
46
28
2
2
10
4
7
14
17
84
3
25
2
27
25
2
2
29
Reference 76.
156
-------
-35°F to -157 F. The final stage uses methane to accomplish the last reduc-
tion from -148 F to -258 F. A simplified diagram of the cascade system ap-
pears in Figure 12.
The expander cycle uses self-refrigeration. Compressed gas is expanded
through an engine to extract work. At the same time the temperature of the
gas is lowered. The amount of heat loss of the gas is controlled by the
amount of the gas and expander inlet and outlet pressures. This method of
liquefaction is used when large quantities of gas must be depressurized when
going from transmission systems to distribution mains. Energy which would be
lost can be used for liquefaction by reducing the gas pressure through a tur-
bine. Unlike the cascade system, this method does not require the use of
multiple refrigerants. Figure 13 is a simplified diagram of the expander cycle.
After liquefaction, additional purification must occur to remove any com-
pounds which solidify at low temperatures. Examples of these substances are
water, carbon dioxide (CO ), hydrogen sulfide (H S), lubricating oils, odor-
ants, and dust. If these solids are not removed, clogging of transmission
units may occur.'^
6.1.6 Storage
Between the time of liquefaction and tanker loading the LNG is stored in
double walled metal tanks above ground. Tanks are of two types.
The first type has capacities up to 65,000 gallons and operates at pres-
sures between 50 and 250 psig. these tanks consist of an inner liner of
cryogenic material and an outer metal shell usually made of carbon steel. A
list of cryogenic materials and their temperature limitations is contained in
Table 59. Between the shell and the wall is an evacuated space filled with
perlite.
A second type has a capacity of up to 1,200,000 gallons. In this case the
LNG is stored at atmospheric pressures. The space between the inner and outer
wall is filled with insulation in a dry inert atmosphere instead of a vacuum.
A variation of this type contains a suspended deck which eliminates the need
for a roof made of cryogenic material. The capacity of these tanks is up to
24,000,000 gallons.
6.1.7 LNG Tankers
Tankers for transporting LNG are of three types: free-standing, spheri-
cal, and membrane. These tanks differ in shape and insulating material. In
all types, the LNG is contained in tanks in the ship's hold which are independ-
ent of the hull. In 1973, the largest tanker in operation was 75,000 cu. m.;
however, several are being produced with a capacity of 125,000 cu. m.
6.1.8 Receiving and Regasification
To unload, LNG tankers connect to liquid unloading arms and the LNG is
moved from the ships to cryogenic stainless steel or aluminum storage tanks.
157
-------
Heat Exchanger
Compressor
\ /
Heat
Exchanger
Compressor)
Heat
Exchanger
Compressor))^
Cooling Water
Valve
86°F
Liquid
Propane
Refrigerant
Valve
-35°F
Liquid
Ethane
Refrigerant
-157°F
Liquid
Methane
Refrigerant
-258°F
Figure 12. Simplified diagram of cascade cycle for
liquefaction of natural gas.a
Q
Reference 74
158
-------
Feed
Gas ^.
To low pressure
distribution
Compressor
Expansion
Engine
Heat
Exchanger
Phase
Separator
Cold Box
Boil off
from storage
To storage
Figure 13. Liquefaction by expander cycle.*
Reference 74
159
-------
TABLE 59
CRYOGENIC MATERIALS AND ASSOCIATED TEMPERATURE LIMITATIONS3
Material
Carbon steel
Aluminum killed steel
2% Ni steel
3% Ni steel
9% Ni steel
T R
Invar
18/8 stainless steel
Austenitic steels
Copper
Aluminum
Copper and aluminum alloys
_F
-22
-58
-76
-148
-328
Durable well
below LNG
temperatures
(_259°F)
Reference 74
160
-------
The LNG is regasified by passing it through heat exchangers at pressures of
approximately 1200 psig. The LNG is then directly introduced into the pipe-
line.
6.1.9 Storage - See Natural Gas Section.
6.1.10 Distribution - See Natural Gas Section.
6.2 ACCIDENT OVERVIEW
Although the technology for reducing gaseous methane to the liquid phase
has been known for several decades, the industry has experienced very slow
growth. This is in part due to a catastrophic accident in 1944 in Cleveland,
Ohio. A storage tank ruptured, and the ensuing explosions and fire killed 130
persons and caused $10,000,000 in damage. Recently, however, there has been
increased interest on the part of the gas industry concerning importation of
LNG.
Table 1 presents a listing of potential accidents associated with LNG and
an estimate of frequency and severity. This table is based solely on an evalu-
ation of likely accidents considering the technology involved. Those accidents
which are considered to have the greatest potential for harm are groundings,
failure of a storage tank, and a collision at sea involving an LNG tanker in
which the contents of one or two tanks is released. Minor accidents include
release of refrigerants and solids blocking the pipeline during or after
liquefaction, onboard malfunctions of tankers, leaks in the system transfer-
ring LNG from tanker to onshore storage, and accidents during maintenance.
An LNG spill may occur on land in the case of a storage facility, on water
in the case of a tanker spill, or underwater in the case of a tanker sinking.
The consequences of each of these events are different.
If LNG is spilled on the ground, it begins to boil violently and vaporize.
If the spill continues, a pool forms, freezes the ground underneath, and the
boiling rate becomes constant. Several studies have been done concerning the
consequences of an LNG spill on water.77,78 jf LNQ spills on water, it is
usually unconfined. It will spread rapidly and continue to boil at a high
rate. If it is confined, an ice layer forms underneath. As this layer thick-
ens the rate of heat transfer is reduced and the boiling rate is reduced.
If LNG spills, it will either ignite immediately or form a low-lying vapor
cloud. This cloud is denser than air up to -40°F and will hover above the
ground or water. The rate of generation of the vapor is proportional to the
pool area. The vapor cloud will be cigar or pancake shaped. It will begin to
rise when warmed sufficiently and its density drops to that of air. It will
either eventually dissipate or be driven by the wind as a plume.
Distances a plume can travel before it falls below the lower flammable
limit (LFL) are determined by meteorological conditions. Estimates range from
1 to 50 miles. Until the LFL is reached, ignition may occur at any time and
involve flashbacks to the original pool. For LNG to ignite, it must form a
161
-------
5 to 15% mixture with air and an energy source (frictional heating, electrical
sparking, static electricity) must be present. The resulting fire will pro-
duce a radiant heat intensity greater than ignition of an equal amount of gaso-
line, but will be shorter lived.°
If LNG is released underwater because of a tanker accident, it is believed
that the vapor will be greatly dissipated due to the heating of the water. The
resulting flammable portion of the cloud rising from the water will be much
smaller than from an equal size surface spill.°
Much speculation exists as to the likelihood of an LNG spill igniting im-
mediately as opposed to formation and drifting of a vapor cloud some distance
from the site of the spill. A 90% probability of immediate ignition is used
in most of the literature reviewed on risk analysis modelling. However, this
figure is arbitrary, and other values have been used.°
Another consequence of an LNG spill is the superheating of LNG in contact
with water resulting in low energy, non-chemical explosions.'9,80 This
phenomenon has been observed during experiments (tests by U. S. Bureau of Mines
Safety Research Center at Pittsburgh during 1969; Continental Oil Company 1965;
Shell Pipeline Company, in Houston, 1965; Esso Research in New Jersey; and
Tokyo Gas Company, 1969) and during transport of LNG (second voyage of the
Pioneer Methane in 1959 when 4900 gallons of LNG were pumped overboard in a
period of 7 minutes, and at a Wisconsin Natural Gas facility when LNG was
drained from storage into a pond).79 These explosions result from sudden
vaporization accompanied by a rapid increase in volume. A study done by the
NAS in 1972 concludes "The nature of the explosion can be shown to have a
relatively low potential for causing physical damage since it is flameless and
not supported by the energy of a chemical reaction, only the energy of the
superheat."79
In spite of the potential for severe accidents there have been only four
major incidents involving LNG facilities. The most catastrophic was the
Cleveland accident of 1944. A large storage tank containing 38,000 barrels of
LNG collapsed. The collapse was later attributed to brittle fracture of the
tank wall caused by exposure to cryogenic temperatures. The alloy of which the
tank was constructed contained only 2% nickel and was not resistent to brittle
fractures at temperatures below -76°F. Also, no dike existed to contain the
spill. The resulting catastrophe involved a spreading pool fire. As it burned,
the pool flowed into the surrounding residential and industrial area. Vapor
accumulated in closed buildings and sewers and later exploded, demolishing many
structures. Fire also spread by thermal radiation from the flame near the col-
lapsed tank igniting combustible materials at great distances. The flame
reached a height of 2800 feet. In all, 130 people were killed, 300 injured,
80 houses and 10 industrial plants destroyed, with a total property damage of
$10,000,000. The likelihood of an accident of this sort recurring has been
greatly reduced by the development of materials better able to withstand cryo-
genic temperatures (alloys now contain 9% nickel which can withstand tempera-
tures up to -328°F), and the containment of spills through diking.
Three other accidents involved storage facilities. In February 1973,
Texas Eastern's LNG storage tank on Staten Island exploded and burned. Forty
162
-------
workers were trapped and died.75 xhe accident occurred during maintenance of
the tank lining. Investigators concluded that although the tank had been
emptied of LNG for one year, gas vapors had been trapped inside. The flame
from a welder's torch ignited them, and the insulation caught fire.75
At the Northwest Natural Gas Facilities in Oregon a tank exploded during
construction before the LNG was introduced. Four workers died. Investigators
attributed the accident to careless work practices and not to LNG.75
In 1972, natural gas leaked through an air line to the control room of an
LNG plant in Montreal, Canada. A large fire followed, creating more damage
than would have occurred in an area of lesser importance.75
LNG accidents may be classified as two types - those occurring during
marine operations, and those occurring during onshore operations. Accidents
associated with marine phase of operations may be classified as malfunctions
aboard the tanker and impact with a second body.
1. Onboard - Onboard malfunctions may involve spills and fires. LNG
tanks are designed so that a fire in one tank will not ignite another. Care is
taken to place tanks as far as possible from potential ignition sources. On-
board spills may cause imbrittlement and fracture of metal surfaces. Insulated
catchments are standard equipment and should contain spills so that the ship and
crew are not seriously threatened. Since there are spaces within which LNG
vapor may accumulate by accident, the threat of fire cannot be ruled out.
2. Impact with a second body - This group of accidents includes col-
lisions, rammings, and groundings. In rammings (ship to object impact) and
groundings (ship to harbor floor impact) the impact point frequently is in the
bow. Since the LNG tanks are located aft of the bow, many authorities believe
they will not be affected in these types of accidents. Collisions (ship to
ship impact) in which the LNG vessel is struck in its side by another ship are
considered to be more serious.
In a collision, probably only one tank will rupture. Rupture of two tanks
will occur only if impact takes place at a bulkhead intersection between two
tanks. Consequently, the largest possible amount of LNG to be released will
be 75,000 m3 or the contents of two tanks, assuming 37,500 m3 per tank.
Several measures may be taken to reduce the possibility of collisions,
rammings, and groundings. These include crew training, installation of on-
board safety devices, close Coast Guard surveillance including escorts, and
deployment of a zone around LNG vessels excluding all other ships, or halting
of all other traffic while the LNG vessel enters and docks.8
Accidents associated with the onshore phase may be classified as malfunc-
tion of the transfer system, damage to storage tanks, malfunction of vaporiza-
tion plants-, and malfunction of liquefaction plants.
1. Malfunction of Transfer System - The transfer arms, the LNG piping to
the storage tanks, and the trestle carrying the piping are sites of possible
accidents. The risks associated with these operations have been reduced to
163
-------
low levels by the installation of safety devices. The potential risk to the
public has been judged to be small relative to those occurring due to tankers
or land storage tanks.
2. Damage to Storage Tanks - Storage tanks may internally rupture due to
inadequate design, "roll-over" turbulence, rupture or external causes. "Roll-
over" is a phenomenon resulting from improper mixiug of LNC from different
sources and of different compositions. If the heavier liquid is loaded from
the bottom or the lighter liquid from the top, the liquids remain stratified.
If the bottom layer heats, the densities equilibrate and rapid mixing ensues.
This could create heating, boiling, and overpressures sufficient to rupture
the tank. Careful loading and mechanical mixing have prevented accidents re-
sulting from "roll over."
If a spill does occur damage will be minimized if the tank is diked and
located far within the facility boundaries. LNG terminal construction is
regulated by strict building codes. Design of a dike is critical to minimiz-
ing potential damage. The dike should be capable of containing more than 100%
of the tanks' contents and its location and dimensions selected to assure dam-
age will not extend beyond a facility's boundaries. The dike should provide
a small surface area for any spill and have an insulated floor to minimize
heating of the spilled liquid and maintain a low rate of vaporization. This
will permit the cloud to dissipate before it reaches its LFL. Some dikes are
broad and shallow. Some extend only a few feet from the edge of the tank so
they are effectively only an outer wall. The former promote higher evaporation
rates but are less vulnerable to a rupture from the same cause as that for the
tank.
Although codes are in operation, controversy exists as to the safety of
current designs. At public hearings in Cumberland, Rhode Island, in 1973,
concerning the construction of an LNG storage tank, Dr. James Fay of MIT testi-
fied on behalf of the citizens that the National Fire Protection Code 59 for
dike construction did not fully consider the formation of a drifting flammable
cloud, and that existing requirements concerning distances between storage tanks
and public buildings are too small. Dr. Elizabeth Drew testified on behalf of
the utility that existing codes were adequate and provide ample protection to
the public. It was decided that the tank would be built at the original site.
Tanks should have double walls to provide good insulation. The inner wall
must be of a cryogenic material. An additional safety factor is gained if the
outer wall is also cryogenic. In this way, the LNG would be adequately contained
to allow draining if a rupture to the inner lining occurred.
3. Malfunction of Vaporization Plant - It is estimated that only small
leaks of LNG or gas can occur in the piping from LNG tanks and in the vaporiza-
tion process. Valve corrosion may occur leading to leaks and icing, and pos-
sible pressure buildups in piping sections. Insulation failures will cause
frost, icing, and possible leaks. Safety practices of leak detection, spill
prevention and containment, and fire warning devices greatly reduce risks to
the public.
164
-------
Insulation failures and escaping liquid or gas present the possibility of
cryogenic burns from contact with cold liquid, vapor, or surfaces. The severi-
ty of the burn depends upon length of exposure, temperature of the source, and
the rate of heat transfer. If the skin has surface moisture it may freeze onto
the cold object. Skin may adhere to the cold surface when affected area is
removed.
Prolonged exposure may also result in hypothermia, a decrease in body
temperature. Physical and mental capabilities are greatly reduced if body
temperature is decreased. Death results if temperature drops below 80.5°F.
Vapors may reduce oxygen content so that asphyxiation occurs. This may occur
sufficiently slowly that victim realizes it too late to save himself.
Protective measures against these physiological effects involve wearing
safety glasses or goggles, non-conductive fire-resistant hard hats, protec-
tive clothing, gloves, and shoes.
Unless vaporization takes place at a constant rate, temperature cycling
and sudden drops in temperature will create thermal stresses and lead to
material weakening and possible failure.
4. Malfunction of Liquefaction Plants - Liquefaction plants have prob-
lems similar to vaporization plants. Insulation failures, pipe leaks, valve
corrosion with pressure build ups, and thermal stresses occurring during
start-up all may occur. An additional problem is the possibility of blockage
by solids. When methane is liquified, impurities such as water and carbon
dioxide may become solids. If these solids are not removed, they may block
pipelines resulting in back-up pressure and possible rupture.
6.3 RISK ANALYSIS
Risk assessment follows three basic methodologies. These are logic tree
modeling, analytical and simulation models, and statistical models. Logic
tree analysis begins by considering a certain initiating event and then traces
out all the possible events and sequences of events which could possibly occur
as a consequence of this initiating event. An intrinsic problem with this kind
of approach is that not all possible accident event sequences are accounted for.
A second problem is the lack of assurance that failures arising from human err-
or or inadequate design have been accounted for. A third problem is getting
good estimates of component failure probabilities.
Analytical and simulation approaches model the performance of a system in
terms of appropriate performance parameters. The performance parameters such
as ship velocities and structural characteristics are formulated for given
"scenarios." Probability distributions for particular conditions in the
scenarios which will produce an accident are then calculated. Simulation
models develop risk assessments as ratios of frequencies of occurrence of par-
ticular end events and the frequencies of occurrence of the system's activities.
Analytical and simulation models are often labeled as arbitrary because of the
necessary simplifying assumptions.
165
-------
Statistical modeling obviates the problems associated with logic tree
analysis and analytical and simulation models. It develops risk parameters,
probabilities and consequences of accidents based upon historical data. Be-
cause it is based upon actual events, arbitrariness should be eliminated.
Unfortunately, however, historical data is not available in the case of LNG
operations and extrapolations must be made using existing data on large ocean
going vessels.
A generalized study was performed based on an event tree model of the
sequence of occurrences following a tanker accident.? The four events were
spill and generation of vapor at a given rate, dispersal and mixing of vapor
with air, ignition and burning of the flammable mixture, and death of all per-
sons within the flammable plume. The possible chain of events and consequences
following an LNG spill appear in Figure 14.
It was assumed that the flash burning of the vapor plume would be the most
important mechanism causing fatalities and was the only mechanism treated. The
number of fatalities was assumed to be a function of the product of the ground
area covered by the flammable portion of the plume at the moment of ignition,
A, and the population density at the site of the spill. Mathematically, this
is expressed:
No. of fatalities = f(A)(P).
The ground area covered by the flammable portion of the plume at the mo-
ment of ignition, A, is a function of the vaporization rate, Q, the dispersion
of the vapor, V, and a parameter relating flammable plume area and ignition, I.
Thus, probability of occurrence of any accident sequence (Fig. 14) is
the joint probability of the parameter values:
Probability = P(Q)oP(V)oP(l)oP(p)
To obtain an upper limit, the frequency of accidental spills was assumed
to be the same as for petroleum tankers. Based on this, it was assumed LNG
tankship losses would be of two types: a leak of 1,000 bbl/hr or a rupture
resulting in an instantaneous loss of the volume of one tank, 25,000 cu.m. The
average annual loss rates for LNG tankers was found to be 0.03 losses/year for
leak situation and 0.02 losses/year for instantaneous loss situation.
The consequent fatalities and their frequency were calculated for all the
combinations of the values of the variables shown in the event tree diagram
(Figure 14).? The frequency of accidents with a number of fatalities within
a three-fold range were summed. The results in the form of a histogram of
accident frequency versus fatalities/accident appear in Figure 15. Table 60
lists the expected fatalities per accident and their frequency of occurrence.
The predicted statistical fatalities are 0.4/year for 1000 tankers annually
calling at U. S. ports during the 1980-1985 time period. The projections
represent an upper bound since they are based on a number of conservative as-
sumptions. The most significant assumption was that there would be no collision
166
-------
vaporization
large
flammable
plume
fc
confined
plume
vapor air
plume forms
immediate
ignition
inadequate
venting,
wall falls
overheating,
wall falls
Figure 14. Event chain for LNG spills.£
plume
detonation,
fire from
liquid
deaths;
total
destruction
flash plume
fire;
fire at
liquid
deaths;
extensive
destruction
fires set
by burning
liquid
several
deaths and
injuries;
extensive
destruction
localized
fires
few deaths
and injuries.
some
destruction
tank
fragment
projectiles
deaths,
injuries
property,
damage by
fragments
flashing
vapor;
fireballs
persons in
fireball
killed,
flammables
destroyed
-------
nj
01
g
T3
•H
O
O
io *•
10~3
io-5
10~6
J
10-7
—
—
en
O
1
.— 1
0
—
o
I— I
1
*
o
—
CO
i
,— i
o
1— 1
1
en
o
en
0
.-H
^•^M
O
0
i-H
1
O
en
0
o
en
i
o
0
1 — 1
—
o
o
o
.— 1
1
o
o
en
m^t^m
O
o
o
1
o
o
o
i-H
i^^^H
o
o
o
o
1— 1
1
o
o
o
CO
1 0
o
0 0
0 «
0 0
•> en
0
i— i
Fatalities per accident
Figure 15. LNG tankship accidents assuming 1000 tankships per year;
predicted frequency of classes of accidents with a
statistical number of fatalities.3
Reference 7
168
-------
TABLE 60 - PREDICTED FREQUENCY OF FATAL LNG TANKSHIP ACCIDENTS
IN U.S. PORTS BASED ON 1,000 TANKSHIP TRIPS TO THE U.S. IN 1980a
Fatalities Per Accident
0.01-0.03
0.03-0.1
0.1 -0.3
0.3 -1.0
1.0 -3.0
3.0 -10
10 -30
30 -100
100 -300
300 -1,000
1,000 -3,000
3,000 -1 x 104
4 4
1 x 10-3 x 10
Frequency (per year)
Leak Tank Rupture
3.5 x 10
4.9 x 10"
-3
1.8 x 10 3 2.0 x 10 3
1.2 x 10~3 1.6 x 10~3
7.3 x 10"
3.3 x 10"
1.2 x 10"
1.3 x 10"
4.3xl04 9.Ox 10~4
1.7 x 10~4 6.2 x 10~4
3.8 x 10"
2.2 x 10"
1.1 x 10
5.6 x 10"
1.8 x 10"
7.4 x 10"
-4
Reference 7
169
-------
protection provided by double hull structure. Consequently, the probabilities
may be twice as large or one one-hundredth as large as shown.
Risk measurement in the marine phase involves the probabilities of a ship
collision and tank rupture and spill, and the effects of an immediate pool
fire or delayed ignition of a plume. Phillipson presented a comparison of
three models for three possible locations of LNG terminals in California.8
The three sites under consideration were Los Angeles, Oxnard, and Point
Conception. The models were developed by the Federal Power Commission (FPC),
Science Applications, Inc. (SAI), and the El Paso Alaska Company (EAC).
The probabilities, conseuqences, and resulting public risks are shown in
Table 61.
Variations in the results of the studies are the consequences of dif-
ferent assumptions and modelling procedures. SAI considered only collisions
as responsible for vessel casualties whereas the FPC and EAC considered
rammings and groundings to have considerable impact. SAI considered sepa-
rately one and two tank spills, FPC treated only the two tank possibility,
and EAC considered only one tank spills and tanks of smaller capacity.
SAI and FPC assumed a 10% chance of non-ignition at the time of a spill,
EAC excluded the change of a plume being formed. Since the population was
beyond the lethal radius of a spill pool fire, no fatalities could occur in
the sequence considered by EAC. However, EAC's assumption of no plume forma-
tion is not credible.
Risk measurement in the terminal phase of LNG operations considers a
land storage tank spill and a fire or formation of a large plume followed by
ignition at a later time. Only SAI performed these estimates. Table 62
shows that the maximum expected public fatalities per year is 1 x 10~^ for all
three sites."
These tables are summarized in Table 63 for both marine and terminal
operations. The fatality rates have been restated as rates per person
exposed at the site. The disparity in the results of Table 63 indicate the
imperfections and subjectivity unavoidable due to lack of data.
As previously stated, risk analysis may be performed by several methods,
e.g., extrapolation from existing accident data for marine operations or
modelling using event trees. The validity of these methods cannot be evalu-
ated because accident data are lacking. Also, each risk analysis is based
on different assumptions as to the probability and severity of individual
events. Examples of this include whether rammings and groundings are as
potentially harmful as collisions, the number of tanks which will rupture,
and the chance of a formation of a plume and later ignition versus immediate
ignition of a pool.
170
-------
TABLE 61 - PROBABILITIES AND CONSEQUENCES OF MARINE PHASE ACCIDENTS
Casualties Per Year
Probability Per Year
of One-Tank Spillc
Probability Per Year
of Two-Tanks Spill
Probability of Pool
Fire, Given Spill
Probability of Plume
and Delayed Ignition,
Given Spill
Total Expected Public
Fatalities Per Yearc
Los
SAI
10~5-10
10~6-10
10~7-10
0.90
0.10
Angeles
FPC
-2 1.1
-3
-4 io-3
0.90
0.10
Oxnard
SAI FPC
10~5 2.5
io-6 -
\o-7 io-4
0.90 0.90
0.10 0.10
Point
SAI
io-5
io-6
io-7
0.90
0.10
Conception
FPC
2.5
-
io-4
0.90
0.10
EAC
_
io-6
-
1.0
0.0
10
-3
10
-3
0.3
10
-6
0.02
Reference 8
SAI considers only collisions and in several areas of port; FPC and EAC also consider rammings and groundings
C37,500 m3 per tank for SAI and FPC; 24,500 m3 for EAC.
SAI values presented estimated by author from SAI data in different form; 1990 population forecasts and 565
LNG vessel trips per year assumed.
-------
TABLE 62 - LAND SPILL PROBABILITIES AND TOTAL FATALITY RATESS
Ship Tank
Sidewall
io-
io-9
io-6
io~7
io-1
lo"7
Land Tank
and Dike
-U
io-8
io"6
io"9
io-1
io-7
Transfer
Components
io-12
io"8
io-3
io-7
_
io"3
Los Angeles
Initiating Event
Internal Failure
Earthquake
Meteorite
Aircraft
Missile
Fill Factorb
Total at Site
Maximum Expected Public Fatalities Per Person Per Year — 10
Oxnard
Initiating Event
Internal Failure Kf11 - 10~
Ship Collision at Dock 10~
-17 -4
Earthquake - 10 10
Reference 8
Fill factor approximately summarizes the effects of SAI's assumptions that
ship or land tank will generally be only partly filled at any given time, and
that a rupture may occur at a position on the tank above the level of its
contents.
c -11
But multiplied by SAI by 10 to reflect capability to isolate the spill.
Considered by SAI for Oxnard only.
(continued)
172
-------
TABLE 62
(continued)
Ship Tank Land Tank
Sidewall and Dike
io-6 io-9
io-6 io-7
Transfer
Components
io-7
io-4
10-3°
Oxnard (continued)
Initiating Event
Meteorite
Aircraft
Missile
Fill Factor
Total at Site
Maximum Expected Public Fatalities Per Person Per Year — ICf
Point Conception
Initiating Event
Internal Failure lO"11 10~6 ICf3
Earthquake - 10 10
Meteorite 10~9 10~8 10~8
Aircraft 10~7 10~6 10~4
Missile 10~7 10~9 10~7
Fill Factor lO""1 lO"1
Q
Total at Site 10~8 10~6 10~3
Maximum Expected Public Fatalities Per Person Per Year — 10
173
-------
TABLE 63 - ESTIMATED EXPECTED PUBLIC FATALITIES PER EXPOSED PERSON PER YEAR
Marine Operations
Terminal Operations
Los Angeles
SAI FPC
10
-7
10
-7
10
-4
Oxnard
SAI
10
-8
10
-7
FPC
10
-6
Point Conception
SAI FPC EAC
10
-9
10
-7
10
-5
10
-8
Total Expected Public
Fatalities Per Person
Per Year
10
-7
10
-4
10
-7
10
-6
10
-7
10
-5
10
-8
Reference 8
Only SAI gives individual or per person fatality rates. For the FPC and EAC, a proportionating of their
total numbers of fatalities per year of 0.02 and 10~-\ respectively, has been carried out by the author
to give the individual fatality rate estimates shown.
-------
SECTION 7.0
HYDROELECTRIC
7.1 RESOURCE SYSTEM OVERVIEW
Since the first water mills were made operational over 3,000 years ago,
man has looked to water as a source of energy. Recognition of this potential
source would be expected as the energy of a rushing body of water is apparent
by visual examination. It was not until the 1880's, however, that water power
was utilized to generate electrical energy. The first plant opened on Septem-
ber 30, 1882 and utilized the power of the Fox River in Appleton, Wisconsin,
producing direct current (DC). A plant was established on the Willamette
River in Oregon using a more efficient means of transmitting electricity by
generating alternating current (AC). By 1940, approximately 30% of the elec-
tricity generated in the United States was by hydroelectric plants. Because
of increased use of fossil fuels and greater demand for energy, this percent-
age fell to about 15% by 1971.
There are many positive aspects to hydroelectric power generation. These
include low cost of operation and maintenance, low downtime, and multiple use
benefits. For example, the downtime has been shown to be approximately one
quarter that of steam electric generating plants, or about 3% per year. The
low downtime for hydroelectric power plants is indicative of the infrequency
of major accidents affecting power production. Some multiple use benefits
include navigation, flood control, irrigation, public water supply, and recre-
ation.
There also are several drawbacks to use of hydroelectric resources. These
include high capital costs, long transmission distances due to remoteness,
irreversible environmental effects, and the variability of river flow. That
this last factor can be overcome was demonstrated at Hoover Dam, with Lake
Mead capacity equal to 2 years of river flow. However, the greatest draw-
backs to hydroelectric power generation are the limited availability of suit-
able damming sites and flooding of habitable lands when dams are built.
In 1972, the total power generating capacity in the United States was
418 x 103 MWe (megawatts of electricity) of which 57 x 103 MWe (14%) was con-
tributed by hydroelectric power.9 Estimates of hydroelectric capacity in the
U.S. are derived by the U.S. Geologic Survey from their stream-flow data.
The data indicates both significant installed and potential hydroelectric
capacity in the western United States. The North Pacific area accounts for
more than one-third of the total U.S. developed and undeveloped capacity. One
source estimates that theoretically the United States can produce 390 x 10J
175
-------
MWe; however, various problems probably reduce this figure to 179 x 103 MWe.
Another source estimates potential hydroelectric power generation at 186.8 x
103 MWe.^ Using this figure, it can be concluded that in 1972, approximately
30% of our estimated hydroelectric energy capacity was installed generating
capacity. This energy was produced in 1,511 plants throughout the country.
Even if our hydroelectric potential was developed to full capacity, it would
still be unable to supply our total energy needs. However, this energy source
of extremely low pollution, could considerably reduce our needs with respect
to other energy alternatives such as fossil fuel and nuclear plants.
Hydroelectric power derives its energy from the potential and kinetic
energy available in the water source. The potential energy is a function of
the elevation or head of water above the turbine, and the kinetic energy is
dependent upon the flow rate. The available energy can be expressed in terms
of an energy head per unit mass, h£ according to the following relationship:
h = z + v^ where z = elevation of the water
zg surface above a reference
v = mean velocity of the water
g = acceleration of gravity
A hydroelectric power plant is therefore configured to readily convert poten-
tial energy to kinetic energy and to thus take maximum advantage of the re-
source. A pictoral representation of a hydroelectric power station is shown
in Figure 16.
The plant consists of four basic components, the reservoir (forebay),
water conduit (penstock), hydraulic turbines, and exit raceway (tailrace).
The electrical system on the output side of the turbine, such as generation,
transformers, transmission lines, etc., is similar to the generation equip-
ment at power plants using any of the resources (fossil, geothermal).
The reservoir serves as turbine feed water storage as well as providing
sufficient head above the turbine so as to achieve the maximum energy head.
Storage is required during reduced plant load so that the water may be used
for peak load periods and to compensate for low and variable stream and river
flows. Reservoirs are created by building a dam across a natural basin,
through which the water source flows, flooding the basin. Hydroelectric dams
generally are permanent structures built from solid or hollow gravity type
concrete. Other types of dams are buttressed concrete, timber and steel
(rarely used in the United States), earth, rock fill (generally used in re-
mote locations), and the arch.82 It is generally difficult to find suitable
sites for the arch type. Among its requirements are a small length to height
ratio and valley walls composed of good rock which can resist end thrust.
A reservoir may have a depth of over 400 ft., and encompass approximately
247 sq. miles. The Hoover Dam structure at the head of Lake Mead is 726 ft.
high (2nd highest in the U.S.) and 1,244 ft. long. The longest dam construct-
ed was the Amistad on the Rio Grande in 1969 at 3, 202 ft., although its height
is only 285 ft. The dam method of managing the water resource is generally
used in medium-to low-head projects. In some cases smaller regulating
176
-------
transmission
lines
reservoir smzzzszz
power generating
station
transformers
I I
hydraulic turbine
with scroll case
Figure 16. Typical hydroelectric power plant.
-------
reservoirs are used to take advantage of smaller heads. In high- to medium-
head projects part or all of the flow is diverted, through tunnels for exam-
ple, to take advantage of steep gradients along a river. Short- or long-term
holding storage reservoirs may be used in such a system."^
The penstock is the conduit between the dam intake and the turbines in
the powerhouse. This conduit may be divided into nonpressure and pressure
types, the former following the hydraulic gradient, the latter being well
below it and leading to the turbine. The nonpressure conduit may be canals,
flumes, and pipe, while the pressure conduit is usually steel pipe. In low
head plants the conduit is the dam structure.8^ Flow in and out of the con-
duit are controlled by gates and valves at either or both ends.
The turbine generator system uses three types of driving machinery -
propeller turbines, inward flow pressure turbines, and Pelton wheels. These
hydraulic turbines are divided into two classes - impulse type, which use the
kinetic energy of a high velocity stream acting on the circumference; and the
reaction type which uses the combined pressure and velocity of the jet, fill-
ing the runners and water passage within. The impulse units are generally
used at heads exceeding 800 ft. and the jet of water freely discharges and
strikes buckets located around the runner, such as with the Pelton wheel.
The propeller and inward flow pressure turbines are the reaction type, with
the water passing through and filling the vanes and passages. The discharge
from an impulse turbine freely enters the tailrace at atmospheric pressure,
while the closed loop system of the reaction turbine has in integrally de-
signed draft tube which connects the turbine outlet to the tailrace without
disturbing the pressure flow regime. The tailrace is the passage by which
the discharge water exits the turbine and reenters the lower portion of the
river or stream.
In the case of anticipated, rapid-fluctuating peaking power needs, a
pumped-storage system is used which consists of a pumping unit and an elevated
reservoir. In non-peak hours the output of the hydroelectric plant is used
to drive pumps to lift water back up to an elevated reservoir for storage.
At peak hours the steep gradient can be used in providing additional hydro-
electric power. These storage basins can also be used to feed lower reser-
voirs if required by low flow or draught conditions.
7.2 ACCIDENT OVERVIEW
This section will consider accidents associated with hydroelectric
power as an energy source, limiting the discussion to those events which
occur in the system from the reservoir (forebay) through the turbine. The
generator is not included in this discussion because it is similar to the
electric generation system for most other energy sources and is covered in
Section 11. Table 1 shows representative major and minor accidents related
to the hydroelectric system.
The most severe accident involves a dam failure which occurs infrequently.
While there have been a number of dam breaks, only one major hydroelectric dam
has failed in recent years. The Teton Dam, located on the Teton River in
178
-------
Idaho, failed catastrophically in June, 1976. Constructed in 1972, at an es-
timated cost of $65 million, it had an intended estimated capacity of 20,000
KW. The breaching of the dam caused an estimated $500 million in damage'in
the upper Snake River valley, spreading water over 300 sq. miles. It caused
the evacuation of an estimated 30,000 people, killed 9 with 8 missing, while
destroying 13,000 cattle. The cause of failure may have been water pene-
tration of a curtain of grouting (cement based filler) on the dam, compounded
by possible soil instability and improper construction practices; however,
a final report on the dam failure has not yet been released by the Department
of the Interior.
Among the causes of failure for a solid gravity concrete dam are sliding
on a horizontal joint above the foundation, on the foundation, and over-
turning on a horizontal joint within the dam, at the base of the dam, or at
a plane below the base. Additional causes of failure can include erosion
below spillways leading to undermining, and forces exceeding anticipated
design for such items as water pressure, ice pressure, earth pressure, weight
of the dam, weight of the foundation, earthquake forces, and reaction of the
foundation.
A dam failure can cause both direct and secondary events which are detri-
mental to the environment. Among these are the inundation and decimation of
biological species, destruction of food supplies, loss of topsoil permitting
erosion and the loss of ground cover, and contamination of food and water
supplies.
A severe loss of water without dam failure can occur because of seepage.
Serious losses can occur if the dam is on a porous foundation or in long,
unlined canals. In an arid region, the absence of rock formation, a water
table at considerable depth, and a previous foundation can cause a loss of as
much as 30%.^ Careful planning and geologic surveying can eliminate this
occurrence.
Overflow of a dam, caused by spillway failure, heavy rains, or a rapid
thaw may lead to a dam burst. Leaks which may occur would cause minor damage.
In the case of an earth dam, overflow (overtopping) generally leads to failure.
Within the hydroelectric plant there can be several types of accidents
most of which are minor and infrequent. However, a major accident which can
occur is inundation of the plant. This can be caused by conduit failure,
overtopping of hydraulic structures, extreme river flow conditions, or through
conventional openings within equipment. The result would be to remove the
turbine from service and cause extensive damage to electrical circuits and
equipment including the generator.
An example of an unusual environmental accident resulting from a dam
burst, other than the expected drowning and inundation, occurred in Chino-
Farm, Maryland in March of 1972. There was a kill of 2000 geese, of which
88% were found to have died of lead poisoning. It was found that the geese
were exposed to lead in mud highly contaminated with buckshot. This mud
flat resulted from the draining of a pond caused by a dam burst. This type
179
-------
of accident is expected to be rare. In addition, an electrical fire may
occur from partial flooding causing extensive plant damage. This type of
accident has occurred but is infrequent.
Conduit failure may be caused by a number of factors among them corro-
sion and cavitation erosion. Technological advances based on analysis of past
accidents have reduced the frequency of such accidents.
Interruption of power can occur through accidents in the generator and
transmission systems (discussed in the Section 11) or through turbine failure.
Previous statistics demonstrated that the downtime for a hydroelectric sta-
tion amounted to an average of 2 days per year plus normal maintenance time,
or approximately one fourth that of steam electric plants. The downtime,
although not documented, is probably due to mechanical malfunction in the
turbine or valves. Bearings and the other parts of such operating equipment
can fail under stress or through poor maintenance.
Personnel accidents are probably the most frequently occurring events
in the industry, involving construction, maintenance, and normal operations.
A survey of four types of power plants (coal, oil, hydroelectric, and nuclear)
indicates that for a hydroelectric plant the occupational injuries occur at
approximately one half the frequency and with one-tenth the severity of the
average for all electric generating plants."" The survey, covering 1969-72,
included 4 companies with an exposure of 1,711,776 manhours. The frequency
reported was 4.1 disabling work injuries per 1 million employee-hours
exposure. One hundred forty nine total days were lost for work injuries
per 1 million employee-hours exposure. The results for 1972 are not verifi-
able because they involved confidential data and data extrapolated from
those of the electric utilities industry in general.
180
-------
SECTION 8.0
GEOTHERMAL
8.1 RESOURCE SYSTEM OVERVIEW
The geothermal resource reflects the thermal energy represented
by the temperature distribution within the earth's interior over the
range between very high temperatures at the molten core and mild tem-
peratures prevailing at the habitable surface. The existence of this
wide temperature distribution manifests itself periodically through
the eruption of volcanoes and the formation of new land masses. The
temperature distribution pattern is not a uniform one. Higher temper-
atures approach the surface closer in locations where the earth's
crust is younger and less stable, making the easiest access to the
high temperature zones in the geographic areas which are subject to
earthquakes and volcanic activities. These areas are shown in Figure
17.85
Accordingly, the areas having the easiest access to the geother-
mal resources are: Italy, New Zealand, United States, Iceland, Japan,
Soviet Union, Hungary, Mexico, El Salvador, Kenya, Ethiopia, Philli-
pines, Indonesia, Chile, Guadaloupe, Taiwan, and Turkey. These are
the countries where significant activity to exploit the resource has
occurred.
The earliest commercial use of geothermal energy for the genera-
tion of electricity occurred in 1913 in Larderello, Italy. Later,
between 1928 and 1933, a municipal district heating service from geo-
thermal hot water aquifers was installed in Reykjavik, Iceland. In
1940, individual hot water wells began to be used for heating purposes
in New Zealand. The first 0.001 MW experimental generating plant in
Japan, built in 1924, grew to a 0.02 MW plant in 1951. Japanese
activity currently is aimed at a goal of 60 MW, and district and green-
house heating-systems exist.
Currently, the Italian installations generate 415 MW total. New
Zealand generation is about 220 MW, with large use of steam and hot
water for heating. The district heating system in Reykjavik now serves
90% of the homes. About 60% of the population of Iceland is served by
some form of geothermal heat including the heating of greenhouses to
raise fresh vegetables.
The first unit for electricity generation in the United States
was installed in 1960 at the Geysers in northern California. This
unit had a capacity of 125 MW. By 1972, the total installed capacity
had grown to 302 MW. The Geysers presently is producing 522 MWe, and
is projected to reach more than 2000 MWe at maximum development, around
1985.8°
181
-------
oo
NJ
Figure 17.
Regions of intense geothermal manifestations.
'Source: P. 16, Kruger, P. and Otte, C. "Geothermal
Energy" Stanford University Press, Stanford
CA. (1973)
-------
Geothermal manifestations in the United States occur as three sys-
tems which have been described by the United States Geological Survey87
as:
1. Hydrothermal convection systems,
2. Hot igneous systems, and
3. Conduction dominated areas (geopressured zones).
Hydrothermal convection systems involve a fluid, usually water, which
is heated by convection from the surrounding rock. If the convection
and water flows are such that the water is largely vaporized when ex-
tracted, the system is vapor dominated and the resource can be re-
covered as dry or wet steam. If the fluid is recovered as a mixture
of steam and water, the resource is called liquid and exists at tem-
perature levels ranging from above 150°C to below 90°C. Hot igneous
systems involve masses of rock which may be molten in part and exist
at temperatures above 650°C. Conduction-dominated areas involve
underground aquifers or sedimentary basins which have been heated by
conduction from the surrounding strata.
The exploitation of vapor-dominated geothermal resources has been
demonstrated since 1960 in the United States with experience with the
privately-owned and expanding electrical generation facilities at the
Geysers in California. Programs to develop the potential of the other
geothermal resource types have been formulated and are being implemented
by the Energy Research and Development Administration (ERDA). A fore-
cast of the growth of geothermal electrical generation capacity in the
United States through 1986 is shown in Figures 18 and 19.88 This fore-
cast is based on one scenario for development of a research, develop-
ment, and demonstration program. The current Federal program is at-
tempting to achieve the power on line goals in this range. Thus, from
the current generation capacity in California of about 500 MW, electri-
cal generation capacity from all geothermal manifestations is forecast
to grow to about 3000-4000 MW total by the mid 1980's.
8.1.1 Resources
The hydrothermal and hot-igneous resource systems predominate in
the western United States and Alaska. The single conduction-dominated
area discovered to date is the coastal areas of Texas and Louisiana.
Measurement of geothermal reserves is based on the following definitions:
Geothermal Resource Base, The base is the total stored
thermal energy in rock strata above 10 km (6.25 miles)
depth, measured above 150°C.
183
-------
oo
ADDITIONAL INCREMENTS OF
CAPACITY AT THE GEYSERS
DEVELOPMENT OF LIQUID-
DOMINATED SYSTEMS
DEVELOPMENT OF
GEOPRESSURED SYSTEMS
DEVELOPMENT OF HOT, DRY
ROCK SYSTEMS
ESTIMATED CAPACITY
BROUGHT ON LINE, IN
MEGAWATTS d
:UMULATIVE CAPACITY ADDED
1975
b
I
270
270
1
290
560
1
II
EC
425
985
1980
l
III
1
415
1400
l
inn
Ec
525
1925
l
Mill
165
2090
l
620
2710
1
Mini
510
3220
1985
l
III
I
1
500
3720
500
4220
1990
fc
fc
fc-
Figure 18. Near-term scenario for geothermal energy development.3
Reference 88
Each tick represents one plant brought on line
E denotes a thermal extraction experiment
Estimates not available beyond 1987
-------
H
I
H
O
-------
Geothermal Reserves. The geothermal reserves are those
portions of the resource base from which useful energy
forms are recoverable at a cost which is competitive with
the cost of similar energy forms now commercially pro-
duced, and available at comparable locations.
Paramarginal Geothermal Resources. These are the portions
of the resource base from which useful energy forms are
recoverable at a cost which is between one and two times
the cost of similar energy forms now commercially produced.
Submarginal Geothermal Resources. These are the portions
of the resource, recoverable at a cost more than two times
current commercial costs.
Megawatt-Century. A unit of measurement of reserves in
terms of electricity production. It is the generation of
1 megawatt of electricity for 1 century, or for 875 thou-
sand hours; or the generation of 1 kilowatt of electricity
for 875 million hours.
The estimated reserves for the high-temperature hydrothermal con-
vection system (temperatures above 150°C) are the following:
Identified
Reserves (measured and inferred) 3500 Megawatt-Centuries
Paramarginal Resources 3500
Submarginal Resources +1000
Undiscovered
Resources 38,000 Megawatt Centuries
The reserves at the Geysers are estimated at 477 megawatt-centuries or
1590 megawatts over a 30 year period. For the intermediate-hydrother-
mal convection system, the U. S. Geological Survey estimated the re-
source base to be 345 x 1018 calories or 137 x lO-^BTUs. There was no
technological or cost basis for distinguishing between paramarginal
and submarginal resources. No estimates were provided for the low-
temperature hydrothermal system (temperatures below 90°C) ,
The U.S.G.S. conservatively estimated geothermal resources of
hydrothermal convection systems at 11,700 MWe of potential generating
capacity from identified U. S. reserves. Their conservative estimate
of the exploitable geothermal resources, those which are discovered
and undiscovered, is 153,400 MWe or approximately one-third of the
186
-------
present 450,000 MWe power requirements. Mueller89 indicated that
geothermal energy from untapped sources in the U. S. are estimated at
8 x 10/D calories of approximately 5,600 trillion barrels of oil.
Reserve estimates are not practical to make for hot igneous sys-
tems because demonstrated extraction technology and the associated cost
basis do not exist. For partially molten igneous systems no techniques
have been developed to penetrate high temperature molten areas (above
650°C) and to emplace devices for extracting thermal energy and bring-
ing it to the surface. Good estimates of reserves of hot dry rock sys-
tems also are not available.
Reserve estimation for conduction-dominated areas, namely the
coastal zone identified for Texas and Louisiana, is complicated by the
fact that the water contained in the sedimentary layers is overpres-
sured (i.e., the water exists at pressures in the order of 11-14,000
psi at depths of 4-6km, pressures far greater than equivalent hydro-
static pressure), and contains dissolved methane, itself an energy
source. The Geological Survey estimated the reserves on the basis of
independent geopressure-energy recovery, geothermal energy-recovery,
and methane recovery. Three plans for reserve estimation were evalu-
ated. The resource and reserves estimated are shown in Table 64.^0
8.1.2 Technology
Geothermal technology is used to produce useful forms of energy
from geothermal resources, i.e., forms that are readily marketable.
Such forms are the following:
Electricity. Dry-steam flow from a high-temperature hydro-
thermal system such as at the Geysers is already used to
produce electricity in the United States. This energy form
would most likely be produced from exploitation of dry-hot-
rock hot igneous systems.
Hot water for district heating, agricultural and industrial
process uses. The geothermal resource is the intermediate
and low-temperature hydrothermal systems. District heating
systems are already installed and operating at Klamath Falls,
Oregon, and in some other locations in the Western United
States.
Multiple energy forms. In conduction-dominated areas, such
as coastal Texas and Louisiana, a mix of electricity and
pipeline gas could be produced.
The function that must be performed by geothermal technology are
exploration, resource production, conversion, and disposal or injection
187
-------
TABLE 64
RESOURCES AND RESERVES
GULF COAST CONDUCTION-DOMINATED GEOTHERMAL AREA'
oo
00
Resource Base
Equivalent Thermal Energy
Plan 1
Plan 2
-I Q
10 Calories
10,920
6,030
50
1016 Btu
435
239
2
MW-cent MW-20 yrs.
24,380 121,900
MW-cent MW-20 yrs.
38,140 190,700
9,970 49,850
34,350 171,750
38,140 19^,700
Plan 3
3,560 17,800
9,250 46,250
Thermal
Methane
Mechanical
TOTAL 17,000" 676
Reference 90.
Revising this total on a labored pessimistic basis, it is still 9500 x 10 calories (378 x 10 Btu).
°Plan 1: 20 year life, during which well-head pressure gradually decreases to hydrostatic at end.
Flow is limited to constant value throughout the life span. Surface subsidence could reach 15-20 feet.
MW-cent MW-20 yrs,
5,690 28,450
d
Plan 2: 20 year life, during which maximum well-head flow obtained with full reservoir pressure,
surface subsidence could be even more significant.
'Plan 3: 20 year life, during which well-head pressure is increased and regulated to limit surface
subsidence to 3 feet.
-------
of the spent fluids. Once the resource is converted into marketable
form, its transport and distribution could be handled by new or exist-
ing conventional installations such as high-voltage transmission lines
in the case of electricity and high-pressure pipelines in the event
natural gas is produced. In district heating applications, pipeline
transport of hot fluid to the nearest population center and distribu-
tion pipelines within the center are required.
8.1.3 Exploration
The result of exploration is the discovery, delineation, and de-
sign of the extraction procedures for a geothermal reserve. Activities
toward such an end can involve any or all of the following:
1. Airborne surveying for geological analysis and mapping of
underground structures by topography and by measurement of magnetic
fields and infrared radiation, if the area is relatively unknown
structurally and stratigraphically.
2. Field exploration on the ground that could include observation
of geological surface features and measurement of thermal manifesta-
tions, measurement of the water table in available wells, sampling and
chemical analysis of water from hot springs and fumaroles, measurement
of electrical conductivity, seismic noise, and heat flow, and active
seismic exploration.
3. Drilling exploratory wells to determine temperature-depth dis-
tribution, pressure depth distribution, permeability-porosity, fluid
composition, and stratigraphy.
8.1.4 Resource Production
The result of resource production usually is the supply of a steam
or hot water stream suited to the intended method of energy conversion.
In conduction dominated areas of the United States a stream of natural
gas would be produced. Production, therefore, includes drilling and
operating wells to bring reserves to the surface, treating effluents
to remove unwanted solids or liquids, gathering individual sources into
a common supply, and drilling and operating wells which may be needed
for injection of waste fluids into the formation.
Well-drilling techniques are similar to those for oil and gas
drilling, except for problems unique to geothermal resources.^ Ex-
tremely slow penetration rates may be encountered even with bits de-
signed for hard rock. Heat and abrasiveness tend to produce high wear
rates on subsurface equipment. Velocities in the annular spaces, when
steam-water is encountered, can be sonic and lead to high erosion rates.
189
-------
Air colled drill bits may need to be substituted for mud cooled. Par-
ticular attention needs to be given to casing and cementing the well.
Surface equipment consists of liquid/gas separators and gas/
particulate separators. Wells are shut in only if their production
is not required for a long period because substantial disturbance of
well flow rates may result in entrainment of large solid particles
which must be removed. Pipelines need to be protected against over-
pressure by relief valves and rupture discs when steam is not required,
but the well is not shut-in.
For the dry-hot-rock hot igneous system resources, water would be
injected into the formation and water would be brought to the surface.
Figure 20 illustrates a potential technology for the recovery of hot
water from dry-hot-rock formations. Appropriate hydraulic fracturing
techniques would be needed in order to produce flow paths through the
rock between the inlet well and the discharge well.
In the conduction-dominated area, well-head pressure would be let
down at each well through a mechanical device that may or may not re-
cover power before any collection of well-head effluents is attempted.
Alternatively, kinetic and thermal energy would be recovered with
"total flow" devices now in the experimental stage.
8.1.5 Conversion
Conversion technologies depend on whether the geothermal resource
is produced as dry steam, hot water, or hot water and methane mixtures.
Technology is well developed in the case of dry steam because of ex-
periences at the Geysers and many years of experience in electricity
generation. A low-pressure condensing turbine receives saturated steam
at 113.7 psia and exhausts it at 2 psia into a direct contact condenser
fed with water from a cooling tower. The non-condensible gases are
ejected through a two stage ejector equipped with inter and after con-
densers. The steam rate may be 17.6 Ibs. per kwh. Sufficiently hot
(7200°C) liquid-dominated systesm can be operated similarly by allowing
the fluid to flash to steam, though the flashing fraction decreases
with decreasing temperature.
For intermediate temperature systems, which are liquid dominated,
and in cases where steam could be unusually corrosive to the turbine
components, (e.g., in the Imperial Valley, California), an indirect
cycle operating at a lower level temperature difference than for a
a direct steam cycle may be used. Figure 21 illustrates such a cycle
based on isobutane. A 10 MW thermal test loop employing this principle
is being operated at Niland, California. Water temperatures may range
from 160°C to 200°C.
190
-------
power
plant
vertically oriented
crack produced by
hydraulic fracturing
thermal
region, - 570°F
Figure 20.
Potential resource recovery system
hot-dry-rock hot igneous system.3
Reference 1.
191
-------
isobutane condenser
power
boiler feedpump
isobutane turbine for feedpump drive
isobutane boiler
liquid heating boiling & superheating
section section
reinjection
pump
well water
waste well
water
reinjection
well
superheated
well water
deep-well
pump
Figure 21.
Vapor-turbine cycle diagram.3
Reference 98.
192
-------
For the conduction-dominated area, the presence of dissolved
methane in the geothermal water may require careful consideration of
the turbine exhaust pressure because of the need to recompress the
methane to pipeline pressures after condensation of the exhausting
steam. It is unlikely that the methane would be burned for power pro-
duction.
8.2 ACCIDENT OVERVIEW
The geothermal reserves discussed above are dispersed throughout
the western part of the United States in terrain of widely varying
character. It ranges from the densely populated western coastline to
thinly populated areas (about 3 persons per square mile) in northern
New Mexico and Arizona. Because at this time the primary product of
most of the reserves is likely to be electricity, reasonable expecta-
tions are that preference would be given to exploration and development
at sites reasonably close to population centers or transmission grids
in order to minimize the costs of electrical transmissions. District
heating systems would also require geothermal resources close to popu-
lation centers.
A list of accidents associated with geothermal energy production
is presented in Table 1. It should be remembered that there is limited
experience with this source of energy, with the longest U. S. experi-
ence being approximately 15 years at the Geysers. The Larderello field
in Italy, and the Wairakei field in New Zealand, in operation for 70
years and 25 years respectively, provided additional background to sub-
stantiate a number of the possible accidents.
The events which may occur would involve loss of property, hazards
to life and severe detrimental impact on the environment. These oc-
currences are applicable to one or more types of geothermal systems,
as shown in Table 65.
One major occurrence associated with the development of geothermal
wells is the blowout. As with oil and gas wells, this occurs during
any operation involving high pressure fluids. Depending upon the sys-
tem, the blowout would result in hot fluids and steam being released
to the surface or discharged through adjacent geologic structures with
subsequent release to the surface being possible. An event of the first
type occurred during the drilling of one of the early production wells
at the Cerro Prieto field in Mexico.91 It took days to control the
well which geysered steam and salt water. It was estimated that as
much as 10 acre-ft/day could flow as a result of such a blowout.
A second type of blowout occurred at the Geysers. The instabili-
ty of the formation through which the well passes (an old landslide)
allowed the steam to escape into the ground, with the potential of
193
-------
TABLE 65 - TYPE OF ACCIDENT OCCURRENCE BY GEOTHERMAL RESOURCE
VAPOR DOMINATED LIQUID DOMINATED
ACCIDENT HYDROTHERMAL HYDROTHERMAL GEOPRESSURED HOT, DRY ROCK ALL
Exploration
Blowouts x
Subsidence x x
Rig Accidents x
Extraction
x
Blowouts
Induced Seismicity and x
Seismic Induced Failures x
Landslides x
x
Erosion
Rig Failure
Production
Fresh Water Contamination x
Organic Fluid Explosion or Spillage x
Subsidence x x
Clogging and Sealing x
-------
eruption, when attempts were made to cap the well. Proper grouting
can minimize this occurrence. With this technique, liquid cement is
pumped through a series of shallow holes several feet below the sur-
face around the well area to create a mantel. The cemented casing is
firmly surrounded, creating a shield against subsurface intrusion or
eruption alongside the well.
A well blowout can cause damage to equipment, personnel, and ad-
jacent property. The entire drill rig can be lost with the drill
string operator suffering possible death or severe injury. The oc-
currence is manifested by the ground opening and the rig collapsing
into the chasm. As a safety precaution the rig has a slide-for-life
wire which allows the operator to evacuate the rig in a matter of
seconds by sliding down a wire, clear of the drilling area, before the
rig collapses. When natural gas is present, the possibility of fire
also exists.
An earthquake could have a severe effect on a geothermal instal-
lation including rupture of pipelines and splitting of a well. Well
splitting could cause contamination of aquifers supplying potable
water.
A seismic event may be naturally occurring or could possible re-
sult from the reinjection of fluids during production (though this is
an unexperienced and unevaluated conjecture). The disposal of water
and condensate water produced in vapor-dominated fields has been han-
dled by means of reinjection. This has been accomplished successfully
at the Geysers and is being tested at Larderello. However, reinjec-
tion has never been tried in a liquid-dominated field.
A leak in a pipeline may be caused by an earthquake or by me-
chanical or human failure. This is generally minor, but if not re-
paired could lead to major failures of the pipeline. To minimize
the effect of a possible seismic event, the pipelines are usually de-
signed with loops and irregular pathways to absorb ground movement
and shock.
Pressure buildup because of scaling could also cause pipelines
to rupture.89 Hydrothermal waters are often saturated with dis-
solved solids. As the fluid rises in the borehole the temperature
decreases and the fluid becomes supersaturated. When steam is ex-
tracted at the wellhead, the volume of the water fraction decreases
and the temperature drops further. At this point the dissolved solids
concentration in the water exceeds the solubility and precipitation
results.
Because of the precipitation of silica and other dissolved con-
stituents, discharge canals must be cleaned periodically with a
195
-------
pneumatic shovel, which is dangerous and expensive. Precipitation
also causes problems during reinjection because pipes clog and the
permeability of the aquifer may be reduced. Treatment with slaked
lime or adjustment of pH, are under study as solutions to the rein-
jection problem.
Geothermal operations may also cause subsidence. As with oil and
gas production, excessive fluid drawoff can cause a vertically down-
ward shift in the ground. This probably will not occur if the
aquifer consists of fractured rock."^ The only apparent ground move-
ment for a geothermal system has been at Wairakei. ^ The affected
area is greater than 65 km. ^ The region of maximum displacement is
away from the fields and power stations but within 500 m of the steam
mains. The movement is estimated at 40 cm/yr. The Wairakei subsidence
has totaled approximately 4 m since 1956."^ Subsidence can cause dam-
age to buildings and equipment. Should subsidence occur in lowland
coastal areas» flooding could occur. This is a particular problem for
the Imperial Valley of California and for geopressured reservoirs
located along the Gulf Coast.
There are additional events affecting topography which can occur
as a result of construction, operation, and decommissioning of geo-
thermal facilities.95 These include the possibilities of:
1. Landslides and general downslope movement,
2. Sheet wash erosion, and
3. Flooding due to alteration of drainage patterns complicated
by well blowout.
However, these effects have not been observed.
196
-------
SECTION 9.0
OIL SHALE
9.1 RESOURCE SYSTEM OVERVIEW
Oil shales are shales which have a high organic content. This organic
matter is called kerogen and is of indefinite composition, uncertain origin
and usually insoluble in petroleum solvents. Kerogen is an extremely fine-
grained waxy solid intimately associated with the mineral constituents of the
shale. Chemical composition of kerogen in weight percent ranges from 69-80%
carbon, 7-11% hydrogen, 1.25-1.5% nitrogen, 1-8% sulfur, and 9-17% oxygen.
It differs chemically from crude oil in its high oxygen and nitrogen content
and a higher carbon to hydrogen ratio. It can usually be recovered by thermal
decomposition which begins at temperatures around 350°C (662°F). Yields as
high as 150 gallons of oil per short ton of shale have occurred, but most
commercial grades are between 25 and 50 gallons per ton.96
Interest in exploiting the U.S. deposits has existed since the first
decades of the century. Interest peaked during periods of wartime crisis (1917-
18, and 1941-45) when adequate petroleum supplies were considered to be in
jeopardy. Oil shale was declared a leasible mineral in 1920. Shale oil was
produced in the United States from eastern deposits before the Drake discovery
of petroleum in 1859.
A substantial body of technology has been developed and tested on a sig-
nificant scale; however, the generally low prevailing world price of petroleum
has precluded commercial development of a shale-oil industry in the United
States. Estonia, Manchuria, Sweden, and Scotland have operated commercial
plants, and the Estonian and Manchurian industries still operate on a signifi-
cant scale.
The current world price of petroleum has renewed interest in the United
States in the commercial potential of oil-shale resources. In 1974, four of
the six tracts offered by the Department of the Interior were leased to private
interests. But even at present petroleum prices it is not yet clear that these
leases will actually lead to commercial shale oil production before 1985. Never-
theless, the development and refinement of technology continues at a deliberate
pace.
Thus, although the technological basis for an industry can be postulated,
an industry per se does not yet exist. Therefore, an assessment of ths poten-
tial for accident in the operations of an oil-shale industry must be based en
conjecture and reasonable judgment.
197
-------
The concentration of kerogen is measured by the Fischer assay, a standard
laboratory method of thermal decomposition which measures the amount of oil
produced per ton of shale. Interest in resource evaluation is focused on shales
yielding 25 gallons or more per ton of oil,
The major and predominant United States deposits of oil shale occur in the
states of Colorado, Wyoming, and Utah. The deposits underlie 25,000 square
miles of the three states, and about 17,000 square miles contain oil shale of
potential value (i,e,, assaying 25 gallons or more per ton). Of this, roughly
80% occur in Colorado, 15% in Utah, and 5% in Wyoming. Eighty percent occurs
on public lands administered by the Department of the Interior. The total re-
serves delineated have been estimated by the United States Geological Survey
at about 418 billion barrels of oil, of which 80 billion barrels are recoverable
by modern mining methods.97 Seven functions must be performed to produce from
the resource a product mix of liquid-hydrocarbon fuels suited to the market now
served by petroleum sources. These functions are:
1. Exploration,
2. Mining,
3, Shale Preparation,
4, Processing (Retorting),
5, Processing (Refining),
6. Product Transportation, and
7. Land Reclamation/Spent-Shale Disposal.
9.1,1 Exploration
Current technological development is based on two alternative approaches.
In the conventional approach all functions after mining, are performed above-
ground. In the in-situ approach, the shale preparation and retorting functions
are combined with the mining function and all three performed underground.
Figure 22"° illustrates the interrelationships among the functions, and pre-
sents a schematic material flow for a postulated commercial operation producing
100,000 barrels per day of a "syncrude," i.e., an upgraded shale oil that can
be accepted by a conventional petroleum refinery and pumped through a long-
distance pipeline.
A system of core holes is drilled on a widely-spaced grid to estimate and
delineate oil shale reserves. The grid is closer spaced if the exploration
objective is formulation of a mining plan. In both cases, the cores are re-
moved, identified as to location, and samples subject to Fischer assay for
determination. The equipment used consists of surveying tools for locating
the drilling positions, truck-mounted engine-powered rotary drill rigs, and a
chemical laboratory for the assays.
198
-------
vo
VO
mining
o
H
O
o
o
Q Q)
o m
CO
H
£> Cd
O -U
^ a
0)
x e
0)
VO rH
• P.
rO P<
CO
shale
preparation
crushed
rock
155,000
T/D
oil-shale
processing
(retorting)
NOTE
Material flow based on 30 gal/ton
shale
Crude shale oil specific gravity
of 0.93
Syncrude specific gravity of 0.80
excess
gas
gas
purification
(H2S removal)
17,040 T/D
crude shale oil
<;i04,700 BBL/D)
«> ft
r~1 **s^
rt H
tC
CO O
0)
CO
00
rH
3
C!
stack gas
shale-oil
processing
(refining)
ammonia
(276 T/D)
syncrude
(14,000 T/D)
(100,000 BBL/D)
coke
(1,710 T/D)
86 T/D
170 T/D
256 T/D
•sulfur.
. en
Figure 22. Shale-oil production complex schematic material flow.3
Adapted from Reference 2
-------
9.1.2 Mining
Several techniques can be used for mining oil shale. These are:
1. Surface, open pit,
2. Underground, shaft entry,
3. Underground, adit entry, and
4. In situ.
No experiences exist with respect to surface mining (open pit) of oil
shale. The hardness of the rock indicates that such a mining operation would
involve loosening the shale by drilling and blasting, and then collecting the
broken shale by power shovels. Conventional drag-line removal of over burden
would not be practical. Drilling and blasting also probably would be required
for over-burden removal.
Extensive experience exists with the underground mining of oil shale.
Since 1940, the Bureau of Mines and various private groups have cumulatively
extracted more than 2 million tons of oil shale in underground mining opera-
tions. Conventional room and pillar mining techniques have been employed,
with the notable feature that the rooms are unusually large in area and height
because of the roof support offered by the hard shale. Room widths of 60 feet
and room heights of 80 feet are practical. The freedom of movement thereby
generated enables large oil-shale removal machinery to be employed. Multi-head
drilling machines for charging explosives, 20 cubic yard front-end loaders,
and 100 ton capacity trucks have been employed. Future development could in-
volve automated rail haulage, or underground portable crushing equipment feed-
ing high capacity conveyor systems.
9.1.3. Shale Preparation
The preparation needed is essentially size reduction and classification,
and depends upon the process to be employed for retorting (thermally decompos-
ing) the oil shale. The equipment used is the conventional assembly of crushers
and screens. Fine grinding is not needed for the retorting processes now on
the horizon. The preparation plant receives the run-of-mine oil-shale rock by
truck, rail, or conveyor, and delivers sized rock to the retorting plant prob-
ably by conveyor.
9.1.4 Processing (Retorting)
Retorting processes differ in the manner in which heat is applied to reach
the kerogen decomposition temperature of 800-900°F and in the manner of flow
of the rock through the retorting equipment. Two processes employ gas-to-solid
exchange and downward or upward flow of the prepared oil shale respectively.
Another process employs solid-to-solid heat exchange in a rotating kiln. In
all cases the kerogen decomposes, the oil fraction is vaporized or entrained
as droplets in the effluent gases, and the oil product collected as crude shale
200
-------
oil. The uncondensed gases are used as fuel. The residual carbon in the spent
shale can be burned for heat recovery. Electrostatic precipitators may be
necessary in order to collect the fine oil droplets (mist) in the effluent gases
Multiple retorts, operating in parallel, are needed to meet capacity require-
ments. The operation depicted in Figure 22 could involve 12 retorting units.
9.1.5 Refining
Crude shale oil, although a hydrocarbon and a liquid, is unacceptable to
a refinery because of its highly viscous waxy nature and its high nitrogen con-
tent. The first property makes it difficult to pump shale oil in pipelines
and the second interferes with catalyst activity in fluid-bed cracking proces-
sing. The carbon/hydrogen ratio in shale oil is higher than for crude petro-
leum, and the basis for producing a crude petroleum equivalent (or syncrude)
is the adjustment of this ratio, either by carbon removal or by hydrogen ad-
dition. The former approach involves an initial coking step, and the latter
an initial hydrogenation step. Hydrogenation has an advantage, since the
operating conditions can be selected to remove the nitrogen content of the
shale oil as ammonia and incidentally remove sulfur as hydrogen sulfide.
Current interest in the refining of shale oil is focusing on the production
of a syncrude that can be accepted by a conventional petroleum refinery. A
specially designed shale-oil refinery is technically feasible, but its economic
feasibility would seem to require large-enough volumes of shale oil and a
localized market for the products.
9.1.6 Product Transportation
Since the first oil-shale operations probably will produce a syncrude,
product transportation will entail pipeline transport of the syncrude to desig-
nated petroleum refining centers. The equipment for transport will comprise
the storage area (tank farm) at the shale oil refinery, the pipeline and pump-
ing stations, and the receiving tank farm at the petroleum refinery. Trans-
port distances could reach 1,500 miles.
9.1.7 Land Reclamation/Spent Shale Disposal
Techniques used for land reclamation and spent shale disposal will depend
on whether mining is open pit or underground. As already noted, no open-pit
oil-shale mining operation of significance has yet been undertaken in the U.S.,
nor is it clear when such an operation may occur. The prospects are that an
open-pit coal mine, with the exception that the equipment used would be suited
to harder rock, and that the backfill operations for the pit would need to
provide for disposal of spent shale before the overburden is replaced. The
volumes of backfill material, whether spent shale or overburden, will be greater
than the original volume disturbed during mining, so that the final ground
contours after backfill will be raised from the original. Some field crushing
and compacting may be needed to prepare the final surface for the backfill of
the top soil, which would be available in limited quantities. In any event,
the structure of the backfull volume may have a high permeability for ground
water flow as compared to the original structure, even though compacting of
201
-------
backfill in layers is foreseen. Potential leaching of the mineral content of
the backfilled material would therefore be intensified.
Less land reclamation operations would be needed in the underground mining
of oil shale than for strip mining or for underground coal mining. The options
for the disposal of spent shale are to fill the mined-out areas, or to estab-
lish an environmentally controlled dumping area where new land surfaces would
be built. The more attractive of these two options would be determined by the
site characteristics and relative locations of mining and processing facilities.
In the dumping option, major efforts are required to identify suitable techniques
for revegetating the new surfaces and minimizing water percolation.
For the shale oil production complex envisioned in Figure 22, production
of 100,000 barrels per day (14 thousand tons per day) of syncrude requires
daily disposal of 134 thousand tons of spent shale. The daily production of
spent shale would cover an area of about 27 acres to a depth of one foot.
9.1.8 In-Situ Processing
In-situ processing of oil shale combines the functions of mining, shale
preparation, retorting, and land reclamation/spent shale disposal as a special-
ized in-situ operation which produces shale oil of similar nitrogen and sulfur
content but of reduced viscosity. Thus, a refining function would still be
needed to produce a syncrude, although this operation could be less intense.
Private activities since 1972 have focused on a large underground trial opera-
tion which currently involves an in-situ prepared volume of oil shale 250 feet
high and 100 feet in square cross-section. Government activities have focused
on fracturing the oil shale formation to increase its permeability to gas flow,
then flowing air from one well to another through the fractured formation,
igniting the shale at the air inlet well, and collecting the retorted products
at the outlet well. Federal work has also involved simulating retorting of
columns of oil shale of known size distribution in above-ground equipment on
the scale of 150 ton batches.
As practiced currently in experimental underground work, a volume is
excavated under the delineated column of oil shale to be retorted, and the ex-
cavated oil shale removed to the surface. Explosives are placed in the de-
lineated column and detonated in a fashion designed to fracture and break down
the oil shale to fill the volumes of the excavated shale plus the fractured
shale with broken pieces and to produce an underground column having a pre-
dictable porosity. Retorting is begun by igniting the top surface with air and
feeding the column from above with air and recycled gases. The shale is re-
torted, the oil collected in a sump at the bottom of the column, from which it
is pumped to the surface. The recycled gases and residual carbon in the spent
shale fuel the retorting operation. The process continues until a column is
completely retorted, at which point is is abandoned.
The key factors affecting the attractiveness of in-situ processing are the
yield of oil achieved in comparison with that for above-ground processing and
the degree of use of the oil shale deposit itself. These two factors signifi-
cantly affect the production costs of the oil. Achieving higher yields with
202
-------
in-situ processing need not be critical because a number of factors inherently
favor in-situ processing. Among these are reductions in the cost of environ-
mental compliance, the ability to retort oil-shale grades too low for above-
ground processing, the potential to eliminate considerable capital cost items,
and the reductions in costs by elimination in the handling of large quantity
of solid materials.
9.2 ACCIDENT OVERVIEW
The estimated frequency and severity of accidents expected to be associated
with oil shale resources is shown in Table 1. Because recovery of oil shale
for use as a fuel source is still in the development stage, this table is based
on analogy to similar technologies and reasoned judgment rather than available
data.
In formulating Table 1 it was assumed that standard training and safety
programs and incentive awards for mine and plant personnel will be used in a
new shale-oil facility, and that the results will tend to minimize occurrence
of incidents such as tripping on stairs, falling from heights, impact by falling
objects too heavy for hard-hat protection, accidents with automotive equipment,
and careless handling of electrical equipment. It also was assumed that under-
ground mining on in-situ retorting operations will not lead to appreciable
seismic effects because of their relatively small size and the general seismic
stability of the geological structure in the three states possessing the re-
serves.
Only relatively minor accidents should occur frequently, that is, more
than four times a year. At any one site, with competent supervision, a fre-
quency of once per year would be unusual for many accidents for the presently
contemplated production of approximately 157,000 tons/day of shale rock (Fig-
ure 22).
Roof collapse is a severe problem in the mining of coal. Roof collapse
is less likely to occur in the mining of oil shale for an equivalent sized room
because of the hardness of the shale. However, the rooms used for oil shale
mining are likely to be larger than the rooms used for coal mining.
The potential for the occurrence of explosions within processing equipment
is particularly significant in cases where high-pressure hydrogenation of crude
shale-oil is the basis for the refining process. Experience with this type
of operation indicates that close attention must be given by operating person-
nel to observing appropriate startup, on-line, and shutdown operating practices
and to assure proper maintenance of the equipment. The likelihood of^"gassy"
atmospheres within the mine is remote. Thus gas explosions are not likely.
However, explosive (combustible) dust mixtures may form.
Cameron has pointed out that health and safety should be improved in oil-
shale plants and mines built to contemporary standards and in any case should
not be compared to coal mining." He suggests that the copper industry is more
akin to oil shale.
203
-------
The rationale for this approach could be in the fact that the character
of the rock in oil shale and copper mining is similar, the sought after min-
erals in the rock are of the same level in value, and the tonnages extracted
may be of a similar order of magnitude. He quotes the Bureau of Mines as
stating an average annual fatality rate in the copper mining, milling, and
smelting industry was 21 between 1967 and 1970. Table 66 presents fatal and
non-fatal injury statistics for the metals mining industries in four western
states from Bureau of Mines sources.100
One may expect that the mining technologies will be similar, but also
that the mix of open-pit and underground mining will be different. If the
mix should turn out to be significantly different, then the comparison may not
be valid.
204
-------
TABLE 66
METAL MINING ACCIDENTS IN SELECTED WESTERN STATES
o
Ui
(1971 and 1972)
State
Arizona
Colorado
Utah
Wyoming
Average Men
Working Daily
1971 1972
12210
4298
5157
1985
12035
3465
4555
1915
Days
Active
1971
320
260
307
284
1972
329
280
311
290
Thousands of
Manhours
Worked
1971
31243
8926
12645
5065
1972
31729
7690
11361
4939
Number of
Fatal
1971
11
3
4
3
1972
6
5
3
_
Injuries
Non-Fatal
1971
807
411
210
112
1972
810
314
110
97
Reference 100
-------
SECTION 10.0
SOLAR ENERGY
10.1 RESOURCE SYSTEM OVERVIEW
Early uses of solar energy depended on the focusing effect of
materials. Solar heating was used by Joseph Priestly in 1774 to
decompose HgO into mercury and oxygen. A solar distillation system,
built in 1872, was used in a remote Chilean desert location to Pr°7n,
vide up to 6,000 gallons per day of fresh water to nitrate miners.
Also in the 1870's, a solar-heated steam boiler was demonstrated at a
Paris exhibition. However, significant technological development of
solar energy use did not occur until the early 1950's.
This section examines the extent of the resource, the associated
technology, and the accidents which occur in the solar energy resource
system. Four general types of solar energy systems are included in
this discussion. These are direct conversion, wind, tidal and wave,
and biomass conversion.
10.1.1 Extent of the Resource
Solar energy has not as yet been used to an appreciable extent
for several reasons, including a lack of technological development,
little experience of potential users with solar-type systems, and a
lack of sufficiently flexible and inexpensive storage capacity. This
section estimates the extent to which solar energy is available in the
United States. For perspective, these estimates are related, where
possible, to worldwide availability and United States energy consump-
tion.
?1
The solar input to the earth is estimated at 5.3 x 10 BTU per
year. "^ The solar energy input to the United States is estimated at
1.5 x 10 BTU per square mile per year. The total energy consumed
in the United States in 1970 was about 6.5 x 1016 BTU, or roughly the
energy received by 4,300 square miles in one year. Assuming a 10%
efficiency for converting solar energy input to usable energy, it
would have required 43,000 square miles, or approximately 1.5% of the
land area of the contiguous 48 states, to provide the energy consumed
in the United States in 1970. This area is about equivalent to one
tenth of the area used for growing farm crops.103 The projected growth
in the consumption of all major energy resources between 1974 and the
year 2000 is 9-03 x 1016 BTU/yr.104 Assuming the same solar input to
the United States as well as comersion efficiency, it will require
approximately 60,206 square miles or 2.1% of the total land area of the
48 contiguous states to provide this energy growth. The growth
206
-------
in electric power consumption for the period 1974-2000 is estimated at
2.315 x 10 BTU/yr., which represents about 252 of the gross energy
consumption requirement (15,433 square miles). ^
The use of walls and rooftops of existing buildings as solar col-
lectors could reduce this land area requirement. A projection has
been made for meeting the energy requirements of Arizona and 11 western
states by 1990 using photovoltaic conversion. A 192 square mile solar
cell array, coupled with pumped storage facilities could provide
14,300 MWe, which is the estimated energy need for Arizona for 1990.
Similarly, to meet 40% of the electrical power needs of the 11 western
states by the same year would require an array covering 2200 square
miles. •* A proposed solar-thermal power plant, covering 13,000 square
miles of desert is estimated to be able to provide 1 x 106 MWe. The
waste heat from this plant would be used to produce 50 billion gallons
of water per day, sufficient to meet the needs of 120 million people.106
The solar-sea resource, based upon the thermal gradients between the
solar heated surface water and the colder temperature of the sea at
great depths, has been estimated.107 Existing thermal gradients in the
Gulf Stream are sufficient to generate 2.08 x 107 MWe per year or
6.2 x 10 ' BTU per year. This may be compared to the estimated total
thermal energy requirement for the United States by 1980 of 3.20 x 10
MWe or 9.6 x 1016 BTU per year.
The generation of electricity through the use of wind power has
been estimated. It would require approximately 22,000 wind energy
conversion systems (WECS) each with a 50 foot rotor radius, operating
continuously in a 20 mile per hour wind, to generate 1% (1.6 x 10' MWhe
or 5.5 x lO1^ BTU) of the electrical energy used in the United States
in 1971. This assumes a wind-to-electrical energy conversion efficiency
of 25%. The amount of power which these same windmills could gener-
ate would represent only about 0.24% of the growth in electric power
consumption between 1974 and the year 2000. It is stated by others
that under certain optimum design parameters it requires 50 windmills
with 50 foot rotor radii operating in a 20 mph average wind to gen-
erate 1 MWe of electrical power.108 Compared to the previous esti-
mate the 50 windmill configuration of optimum design would indicate
a 6% conversion efficiency. A recent design of a 190 ft. rotor WECS,
by the General Electric Space Division under the auspices of ERDA,
is estimated to be capable of generating 1.5 MWe in a 22.5 mph wind-
stream and 0.76 MWe in an 18 mph windstream. Compared to previous^
examples, this unit indicates a conversion efficiency exceeding 30%.
Extrapolated estimates of available wind energy, at an elevation of
50 meters, is given as 2 MWh/M2/year for average wind velocities of
about 15.7 mph., and 3 MWh/M2/year for 18 mph wind velocities.
The tidal power resource is estimated in terms of dissipation
energy and usable energy.102' 109» U1 The total tidal energy dis-
207
-------
TABLE 67. SITE-SELECTION CRITERIA FOR TIDAL-POWERED DEVELOPMENT'
N3
O
CO
Country and
site
France :
Lorient
Brest
Alber-Benoit
Alber-Vrach
Arguenon
and Lancieux
La Fresnaye
Ranee
Rotheneuf
Chausey
Somme
Tidal
range ,
m
4.5
6.4
7.4
7.4
11.4
11.4
11.4
12.0
12.4
9.3
Basin
area,
km2
16.0
92.0
2.9
1.1
28.0
12.0
22.0
1.1
610.0
49.0
Potential annual
electrical-energy
production,
106 kwhp/y
97
1,130
48
18
1,090
470
860
48
28,140
1,270
Barrage
length,
m
750
3,640
200
190
6,275
2,760
725
330
23,500
5,100
L/S
in/km2
47
40
69
173
224
230
33
300
39
104
L/E,
m/(106 kwhe/y)
7.7
3.2
4.2
10.6
5.8
5.9
0.8
6.9
0.8
4.0
United Kingdom:
Severn
United States:
Passamaquoddy
11.5
7.5
44.0
120.0
1,750
2,025
3,500
4,270
80
36
2.0
2.1
Reference 102.
Calculated from:
x 106) R2S kwhe/yr.
-------
sipated on earth is 3 x 106 MWe with the tidal energy dissipated in
bays and estuaries being approximately 1 x 106 MWe. Usable tidal
energy in the world has also been estimated at 6.4 x 10^ MWe.102 The
efficiency of converting tidal energy to electrical energy by means
of current technology, is estimated at 15%. ^
Selection of sites for using tidal power is based upon the con-
figuration of the location, as determined by the ratio of the barrage
length (L) to the area of the basin (S) , L/S, the barrage being a
barrier of varying design which is placed across the mouth of a bay
to form an enclosed basin. Suitability for production is determined
by the ratio of the barrage length (L) to the potential annual
electrical energy production (E) from the tidal basin, L/E. An opti-
mum site would have the smallest ratios. This potential power pro-
duction is expressed by E = (0.3 x 106) R2S KWe/yr., where R is the
tidal range and S is the area of the basin. The potential energy
production from tidal power for several basin sites is shown in Table
67.
Wave energy dissipated on the earth is estimated to be on the or-
der of 2.5 x 10 MWe or about that of tidal energy dissipation. A
proposed power plant design, using the horizontal movement of waves as
the energy source, is estimated as having an 80% efficiency in con-
verting mechanical energy to electrical energy. A single unit about
the size of a supertanker is estimated to be able to generate approxi-
mately 50 MWe per
It is estimated that 0.02% of the total solar energy flux of the
world is available input to photosynthesis. This amounts to 4 x 10
MWe.102 The energy produced is ultimately released through plant de-
cay, used as animal forage and human food, or stored as a fuel. The
electrical energy resource which can be derived through combustion
of plant matter can be estimated. Assuming a best plant growth rate
which may be expected, of 40 tons of dry material per acre compared
with trees and grasses which are between 4 and 26 tons per acre, and
a heating value of 16 x 106 BTU/ton (63% that of coal) , the optimum
heat-energy production for a single crop is 6.4 x 10° BTU/acre/year
or 4.1 x 1011 BTU /mile2 /year. Assuming a 33% efficiency for the con-
version from thermal to electrical energy, the required area to pro-
vide the thermal energy requirements for electrical energy in the
United States for 1970 would be approximately 4.15 x 104 square miles
or 1.4% of the total land area of the contiguous 48 states.11^ The
growth in electrical power between 1974 and the year 2000 is estimated
to be 6.783 x 1012 KWh (2.315 x 10^6 BTU).104 Based on the previous
assumptions, the land area required to accommodate this growth is
5.65 x 10* square miles or 1.9% of the total land area of the con-
tiguous 48 states.
209
-------
Crop rotation may reduce the land requirement by a factor of two
or three. Based upon the average solar energy incident on the earth
of 1,500 BTU/ft2, the overall efficiency of converting solar energy to
electrical energy via photosynthesis is about 2.7%. This efficiency
is reduced further, to about 2% of the original energy due to leaf
shading, plant growth problems, combustion, steam raising, and conver-
sion of mechanical energy to electrical power. It is estimated that
worldwide energy conversion by terrestrial plants to combustible mat-
ter by photosynthesis is about ten times the present rate of energy
consumption by mankind. -"-^
The decomposition or digestion of algae and sewage by anaerobic
bacteria is potentially a useful source of methane. A pound of dry
organic material produces between 4.5 and 6.5 SCF of methane with a
heat value of 103 BTU/SCF or 56-81% of the heating value of the origi-
nal organic material.9 It has been estimated that 100% of the United
States gas needs in the year 2020 can be supplied using 5% of the
United States land area with a 2% overall energy conversion efficiency,
and be economically competitive with other gas producing techniques. -1-"
10.1.2 Technology
The technology of the four basic types of solar energy systems,
direct conversion, wind, tidal and wave, and biomass conversion, will
be reviewed in terms of their operating systems and the equipment in-
volved. The resulting end use of these systems may be electric power,
hot water and space heating, absorption refrigeration, and air con-
ditioning.
There are three methods for the direct conversions of solar
energy: photovoltaic, solar-thermal, and ocean thermal energy con-
version (OTEC). Photovoltaic conversion involves direct production of
electrical energy resulting from the absorption of solar radiation on
solid state devices, commonly referred to as solar cells. The photo-
voltaic system consists of three basic components: the collector,
energy storage, and power conditioning. The collector is a solar cell
which is constructed of semiconductor materials, generally silicon,
gallium-arsenide, or germanium. The power generated by the cell de-
pends upon its having a high light collection efficiency (low re-
flectivity) as well as absorption coefficients which permit the con-
version to electrical energy by the interaction of the desired radia-
tion energy with the configuration of the solid state cell materials.^
A solar cell, 2 inches in diameter, may provide an electrical power
output of 1.5 watts with a solar radiation intensity of 100 mW/cm.
For operational systems, a number of cells are mounted on a common
frame with a linked electrical output to form an array. Depending
upon the application, a number of these modules can be linked to-
gether. An existing system which powers navigational lights on a
210
-------
buoy consists of 80 modules, each having five 2.17 inch diameter sili-
con cells, which charge 28-100 amp-hr, 12 volt storage batteries.106
The energy produced by any of the solar energy conversion methods
may be stored until demand requires its use. Currently, the electrical
energy from solar cells is usually stored by charging batteries. The
electrical energy may also be transformed into other forms of energy
which can be used at a later time. Some of these alternate energy
storage forms include pumped hydroelectric, compressed air, flywheel,
fuel cell, and superconducting magnets. The type of energy storage
will depend upon the use of the electrical energy output as part of a
transmission network as well as economics. The transmission of the
electrical energy ultimately produced, and after power conditioning
COC/AC conversion, is the same as for any of the other energy re-
source systems.
Solar-thermal power conversion involves focusing the radiant
energy on a heat transfer medium, such as water flowing in a pipe,
boiler or collector. Depending upon the system configuration and solar
collector design, such a system can provide space and water heating
and electricity generation. The two classes of systems discussed here
are the small residential/commercial systems and the large central
power and special industrial use facility.
The small systems consist of a collector, a thermal storage unit,
and an end use subsystem. The collector is generally some configura-
tion of a mirrored reflector. These can be planar with a device for
focusing the radiant energy on the water pipe, or parabolic with the
pipe located at its focal length. The radiant energy heats the water
which then may be used to heat a building, via hot water coil heaters
for example, and/or may be stored as a hot water supply. A residen-
tial solar collector would be able to heat water from 25°C to ap-
proximately 65°C. A commercial system, as shown in Figure 23, pro-
duces steam at temperatures near 750°C, this temperature being a
function of the design and efficiency of the solar collector.9 A
suitably designed and sized heat-storage unit can provide a long-term,
high-temperature heat energy reservoir needed to supply sufficient
steam to power a steam turbine for electrical power generation. Ther-
mal storage in small systems uses an insulated tank, with water being
circulated to the collector for reheating during daylight hours.
The large facility, for central power generation or special in-
dustrial application, consists of the collector, receiver, thermal
storage, and power generation subsystems (Figure 24).y The collector
subsystem consists of a 2-axis steerable array of reflectors (mirrors)
called heliostats, which track the incoming radiant energy. The array
of heliostats focuses the reflected energy on the receiver, usually
211
-------
N3
M
S3
solar collector
on building roof
to rooftop
heat radiator
pump
high-temperature
heat-storage unit
(100 days storage)
surplus heat
low-temperature
heat-storage unit
(4 days storage)
alternator
electrical
power
to building heating
and cooling system
Figure 23. A solar-powered total-energy system for commercial installations.8
Reference 9
-------
light-absorb ing system
electric generator
electric power
storage
tank
Ho pipeline
TTTTT
sun-tracking reflector surfaces
sunlight
electrolysis
unit
Figure 24. Central station electricity generation.3
Reference 9
-------
a parabolic reflector, which concentrates the energy on a boiler of
small surface area. The resulting steam produced by the boiler can
drive a turbine which drives a generator for production of electrical
energy. Hot water from the boiler and waste steam from the turbine
may be stored for future use. Storage may be in insulated tanks or
one of several media whose technologies are being studied. Some of
these thermal media include: Caloria HT 43, heat transfer oil; a mix-
ture of Caloria and granite rock; Hitec, a eutectic mixture of potas-
sium nitrate, sodium nitrate and sodium nitrite; a Caloria/Hitec com-
bination; sodium hydroxide; and long steel ingots with longitudinal
channels through whcih steam passes transferring its heat to the
steel; and thermia, a hydrocarbon fluid. In addition, an electroly-
sis unit powered by solar-thermal energy converted to electrical ener-
gy would permit energy to be stored in the form of hydrogen.
Several special purpose industrial solar furnaces have been built
and operated in Japan, France, and Russia. A 70KW furnace built in
Japan in 1963 is used both for research on the properties of materials
at high temperatures and to manufacture fused alumina crucibles. The
furnace produces temperatures in excess of 3400°C using a 10 meter
diameter parabolic concentrator. A 1-MW furnace was built in the
Pyrenees at Odeillo-Fort Romeu in the 1950's and completed in 1970.117
At its altitude of 5900 feet the sun shines approximately 180 days a
year with solar intensities as high at 1000 watts/M^. This solar
energy is incident on a collector of 23,000 square feet, consisting
of 63 heliostats arranged in eight tiers to track the sun, reflecting
the energy to the parabolic reflector. The reflector, 135 feet high
and 175 feet wide consists of 9500 mirrors having a focal length of
59 feet, concentrating the energy into a 2 foot diameter area. Sixty
percent of the total thermal energy, about 600 KW, is further concen-
trated in a one foot diameter area at the focal point. A smaller
furnace in Russia is generating 3500°C temperatures for the produc-
tion of high purity refractories.-^
Ocean thermal/energy (solar sea) conversion systems (OTEC) are
based on the principle of a heat engine which uses the temperature
gradient between the warm, solar-heated upper layer of water (the
heat source), and the cold, deeper water (the heat sink). This type
of energy conversion system consists of the water intakes, boilers,
and a condenser and turbine. A schematic of a proposed modular OTEC
plant is shown in Figure 25.106, 9
Two experimental power plants of this type were built by the
French at Abidjan off the Ivory Coast in 1956, using the thermal
gradient of 36°F to generate 3.5 MW each.106 The intake pipes were
eight feet in diameter to a depth of three miles, approximately
three miles offshore. The boiler, condenser and turbine are the
214
-------
Condenser
module —
Cold water -
intake pipe
Figure 25. A modular solar-sea power plant.8
Reference 106
215
-------
working heart of the system. Propane, as one of the working fluids
which may be used, is pumped through the boiler where the warm water
vaporizes the propane at approximately 150 psi. This vapor passes
through a turbine producing, an energy conversion. The vapor is then
condensed with the cold, deep water as the heat sink, and the cycle
continues. The pumps derive power from the propane turbine. The
proposed plant would be neutrally buoyant at a depth which would
minimize the pressure differentials in the boiler and condenser.11^
Current ERDA estimates for such a system are on the order of $1200-
1500 per KWe. 12°
A portion of the solar energy received by the earth is converted
to surface winds. Windmills have used wind energy for centuries for
pumping water and grinding grain. Using this energy for electric
power generation, however, is a recent development. The wind energy
conversion system (WECS) consists of the rotor, the conversion unit,
and storage units. The conventional WECS design, a horizontal axis
device, is only one configuration for extracting energy from the wind.
Other designs, which may provide cost competitive operation, include
the cross wind horizontal axis, the vertical axis, and the vortex
rotor types. The rotation of any of these rotors drives the energy
conversion equipment. The energy conversion unit may consist of an
electric generator, a mechanical pump, or a direct heat pump. The
choice of energy storage system will determine which conversion
subsystem will be used. The electric generator energy output can be
stored in batteries, in a pumped hydroelectric system, in flywheels,
as hydrogen generated by an electrolysis unit, or as compressed air
to be used later to drive a turbine. The mechanical pump can be used
with a fluorocarbon to act as a heat pump, with air as a compressor,
or to pump water for a hydroelectric configuration. The wind energy
conversion system is shown in Figure 26. Several wind power plants
of the horizontal axis type have been built. In 1941, the output
from the 1.25 MWe Grandpa's Knob plant, near Rutland, Vermont, was
connected to a utility transmission grid. This unit had two stain-
less steel blades, 175 feet in diameter and weighing 7.5 tons each
and an electric generator mounted on a 110 foot tower, which turned
to face the wind. The plant, costing $1 million, achieved full power
with a wind velocity in excess of 30 mph, achievable about 70% of
the time.-"-21 Based upon the experience of this plant, the Federal
Power Commission conducted a study of wind electric power generation
integrated with utility network grids. The study concluded that wind
energy would be economical for plant capacities of 5-10 MW.-^ The
largest foreign plant was built and operated in France from 1958 to
1960. Using a flexible 105 foot diameter, three-bladed propeller,
the rated 800 KWe was reached with a wind velocity of 49 mph at 47
rpm. The plant had reached a maximum power of 1.2 MWe.122 The
National Aeronautics and Space Administration (NASA) is testing a 100
KWe wind generator at the Lewis Research Center Plum Brook facilty
216
-------
Head-On
Energy Horizontal Crosswind
Extraction Axis Horizontal Axis
I I
Energy
Conversion
Mechanical
Pump
Vortex
1
Other
Electric
Generator
Freon Air
Heat
Energy Pump
Storage
Energy Use
Water
Pumped
Hydro
Direct Heat
Churn
Centrifugal Pump
Friction
Electrolysis
Stored
Hydrogen
Compressed
Air-
Bat te
1
ries
)
•»
*•
!*-
Irrigation
Q_r
Aqueduct
Fuel Water
T
Supplement to
Natural Gas
or Propane
Pre
Heat Steam
Plant
Other
~ Supplement to
Sale 0:"-1 or Gasoline
Electrical
Use or Electric
Vehicles
Space Heating
Figure 26. The wind energy conversion system.
217
-------
at Sandusky, Ohio, in which power is expected to be achieved at 40 rpm
in a wind speed of 18 mph.123 Small residential units of 2-5 KW out-
put which are in use, deliver their power to storage batteries for
later use.
Tidal energy was used as a source of mechanical energy in France,
Germany, Russia, England, and the United States until the late nine-
teenth century; however, the use of tidal energy as a source for elec-
tricity generation has only recently come under examination.
This resource makes use of the tidal range, the height difference
between high and low tide, in conjunction with a low-head turbine to
produce electricity. To place a perspective on the available head, the
open ocean has a tidal range of two feet. The effects caused by water
interacting with the continental shelf (shoaling), water in a bay
interacting with the erection of barriers (resonance), and water in-
teracting with the geometry of the basin (funneling), can amplify the
tidal range thus enhancing the potential for available energy. As an
example, the shoaling effect can produce an amplification factor of
three or four times the range of the open ocean.-*-23 ^he resonance
effect produces an amplification factor of almost 2.7 for the tides at
the mouth of the Bay of Fundy.124 The tidal energy conversion system
consists of a barrage (the barrier transforming a bay into a basin),
the sluiceways and control gates for controlling the filling and
emptying of the basin, and a power house containing the turbines and
generators.
The turbines used are of three major designs, bulb-type, tube-
type, and straight-flow. All are designed to operate at maximum
efficiency under low energy heads typical of tidal power plant con-
ditions. The generator is designed with a rotor on the end of the
turbine blades which turns in a recess in the water channel. The
stator surrounds the rotor recess and is not exposed to the flowing
water. There are a number of designs for barrages which will not be
discussed here other than to mention that the design is determined
by conditions at a particular location. The sluiceways and gates
must be able to operate rapidly and reliably, opening and closing on
each tidal cycle, dissipating the dynamic loads from wave action, and
withstanding the corrosion effects of sea water.
There are four techniques for using tidal energy: one-way,
single-basin generation; two-way, single-basin generation; multiple-
basin; and pumped storage. The one-way, single-basin system is the
simplest and oldest of the techniques. The basin is formed by plac-
ing a barrage, together with sluiceways, gates and power house,
across the mouth of an estuary or bay. As the tide rises, the water
passes through the sluiceways, filling the basin. When the level is
218
-------
hxgher than sea level, the gates are closed and the water is emptied
through the turbines in the powerhouse, generating electricity. Due
to the cycle of tides, the time required to fill the basin, and the
time it takes to empty the basin through the turbines, this method
yields 5 hours of electrical power generation and requires about 7
hours for refilling.9 The two-way, single-basin system differs from
the one-way system in that electric power is generated during water
movement both in and out of the basin. In this scheme the head is
lower than in the first scheme thus requiring larger and more expen-
sive turbines. This is the technique used in the Ranee tidal-power
project in France. The power output from both the one-way and two-
way systems are time dependent. The multiple-basin configuration
consists of two, one-way, single-basin installations which are linked
to provide more of a continuous output. The two basins, one at a
higher level than the other, are both separated from the sea by means
of barrages, sluiceways and gates, and from each other by a barrage
containing a water flow path which passes through the turbine. The
high level basin is filled to a level higher than the sea. Water is
then released to the lower basin, generating electricity. When the
low-level basin level increases to that of the high level one, its
sluice gates open to the sea dropping its level. When the sea level
is that of the lower-level basin, the gates are closed and the low-
level basin begins to fill again, dropping the high-level basins
level, at which time the cycle begins again. The period of this cycle
is approximately 12.5 hours. 9 The fluctuating head in this scheme
causes a power factor fluctuation of approximately 2 during a single
cycle, and a factor of 5 during the lunar month. Although there is
relatively continuous power output, the total energy derived from this
configuration is about half that of the one-way, single-basin scheme
for that basin.
The pumped storage concept is a continuation of the tidal-power
basin configuration, where the electrical output of the tidal power
conversion is used either to pump water to a pumped storage hydro-
electric power plant in a high-head scheme, or to move water to a
higher basin in a low-head scheme.126' 127 The high-head scheme
permits the use of low cost turbines because they can operate at a
high hydraulic efficiency. The low-head pumped storage configuration
is similar to the multiple-basin system previously described.
The use of tidal power as part of an electrical power network
has the major deficiency of being variable. The tidal cycles and the
varying head during power generation in each of the configurations
introduces a variability that is not related to energy demand. There
fore, at best, this source must be used to generate energy to be
stored for use during peak demand times.
219
-------
The action of wind upon the ocean surface creates waves which can
be a source of power. Both the vertical and horizontal action of waves
may be useful sources of energy. Present devices use the vertical
motion in providing power to navigational buoys. The current simple
system consists of a riser pipe, a check valve, and a turbogenerator.
The vertical motion forces the water to a level above the wave height.
With sufficient head, the water is drained to the sea through the
turbogenerator.128 There are proposals which would use the horizontal
motion of waves for generating electricity.
Biomass energy conversion involves the absorption of solar
radiation by plants, and through photosynthesis, storing the energy in
the form of large molecular combustible matter, followed by a conver-
sion of this stored energy to electrical or other energy forms. Bio-
mass production consists of maximum use of efficient farming techniques,
as well as sewage treatment operations. The farming process will not
be discussed here as it is a very diversified field depending upon a
number of variables such as type of crop, location, cultivation, and
heat value. The system also includes the harvesting and transportation
of the resources in order to make use of it efficiently. The major
processes for the conversion of biomass are anaerobic digestion, com-
bustion, acid hydrolysis and pyrolysis. Anaerobic digestion involves
the action of anaerobic bacteria in decomposing algae or sewage in
closed chambers. This process yields a combustible gas containing
approximately 60% methane, which is then used as a fuel for electrical
or heat energy production. Also produced in the digestion process is
a liquid enriched in nitrates and phosphates which has application as
a fertilizer.
Acid hydrolysis involves the conversion of biomass to sugar. The
sugar then undergoes fermentation producing alcohols which may then
be used as a fuel for heating. Pyrolysis is the high temperature des-
tructive distillation (thermal decomposition) of the organic materials,
in the absence of oxygen, yielding liquid and gaseous fuels.
The use of the various solar energy conversion techniques for the
generation of electricity is shown to have variability which does not
follow user demand.H^ ^he combination of each of the sources, their
storage, and conversion to electrical energy provides an approach to
electrical power generation, as shown in Figure 27.^06
10.2 Accident Overview
Table 1 presents representative accidents and unscheduled events
associated with the solar energy resources. While there has not been
any large scale development of solar energy, several of the accidents
listed have occurred at demonstration, pilot, and small scale plants
which have been built and operated.
220,
-------
Electricity
Thermal or Chemical Energy
Mechanical Energy
Technological
Collection
Solar
Radiation
Chem. Fuel
Gen.
to
Natural
Collection
Biolog.
(Heat)
Power While
Sun Shines
Heat
Storage
Power
Plant
Remote Power
on Demand
Hydro
Power
Plant
Figure 27. Approaches to the conversion of solar energy into electricity.3
Reference 106
-------
The accidents associated with photovoltaic conversion are minor
and are expected to be infrequent. Solar cell damage can be minimized
with care in shielding and installation. It is anticipated that roof
mounted units would be less likely to sustain damage caused by the
crashing of objects into or onto the array. Injuries associated with
this technology would most likely be those resulting from falls or
electric shocks and burns while installing or servicing the equipment.
Solar- thermal related accidents can be major because this type of
system has been used in large scale projects. These systems involve
significant costs, such as for the heliostats and reflectors. As an
example, the heliostats for the French solar furnace cost $21 per
square foot with 23,000 square feet used. With systems being proposed
to use one square mile of heliostats, the costs could be quite sub-
stantial. The cost of a tower used in a generating station as shown
in Figure 24 is estimated at $15 million. Accidents involving the
large scale systems may include objects crashing into the heliostats
and reflector, rupture of piping seals or the boiler which carry
pressurized water and steam at temperatures ranging from 150°F to
1000°F, or gross structural failure as may be caused by an earthquake.
The smaller residential size solar-thermal system would not repre
sent significant capital costs. By the mid-19601 s, Israel had over
100,000 solar water heating systems installed. Equipment failures in
these systems have been shown to be so infrequent as to cause the man-
ufacturers to increase the guarantee period from 3 to as long as 8
years.129
Other unscheduled events would involve injuries or unknown health
effects. Glare caused by the numerous reflecting surfaces, may af-
fect the general population in the vicinity of the plant as well as
pose a possible threat to birds or other wildlife in its path. Con-
centrated radiation may also pose a threat to the same segments of the
ecosystem. The effect of the accidental release of some proposed
heat transfer fluids on the population and the surrounding environ-
ment is, at this point, uncertain. Therefore, no hazard potential for
a leak or rupture of such fluids can be stated now. Occupational
hazards could be expected during installation, maintenance, and oper-
ation of a plant or even the smaller residential units. Occupational
injury for solar power is estimated at 0.5 - 1.6 man-days lost/MWe/yr .
Occupational accidents are those occurring in similar areas of other
energy resources and include burns, falls, strains, and injuries due
to falling objects and contact with operating equipment. Accidents
involving the steam turbine or the electric generating components are
those described previously in the accident report under geothermal or
fossil fuel-fired power plants. Storm damage to the various solar
systems may occur. The damage sustained as a result of these storms
may include cell and reflector breakage, bent reflectors (home solar-
222
-------
thermal units), pipe ruptures, roof leaks, heliostat failure, and loss
of the energy resource conversion capabiltiy.
OTEC systems are expected to have relatively minor accidents as-
sociated with them. The plant may be isolated from large populations
and would not have any crew on board, except for maintenance and re-
pair functions. There have been few such systems which have been
operational. The two units off the Ivory Coast in 1956 suffered from
corrosion, a problem expected to be prevalent with this type of sys-
tem. The piping, 8 feet in diameter and extending 3 miles deep, be-
came so difficult to maintain that the plant was finally abandoned.
Another possible accident may involve explosion or fire involving
the working fluid used, which may be propane, butane, or ammonia, or
a stored energy form such as liquid hydrogen and oxygen. A large
scale explosion could conceivably destroy the plant which is a sizable
investment. The working fluid can also be accidentally released with
no resulting fire or explosion. The effect this may have on the sur-
rounding ocean ecosystems are not known at this time. Accidents also
may occur in transporting the stored energy forms from the plant to
the mainland, by ship or pipeline. The accidents associated with this
are similar to those shipping accidents encountered when transporting
crude oil, natural gas, and LNG.
There are several types of accidents which are associated with
wind energy systems. The plants which have been built at Grandpa's
Knob (1939) and in France (1958-1960) both suffered from broken blades.
The blade breakage on the French unit created unbalanced forces which
tore up the hub, destroying the plant. The horizontal and vertical
axis turbine rotors may cause significant E-M wave interference. The
extent of this event is not really known at this time, but may in-
volve communication and navigation interference of aircraft. There
can be a probability of spawning tornadoes with some designs of the
vortex type. The top of the vortex column produced may pose a hazard
to air navigation. There are also those accidents which may be ex-
pected with WECS. These include ice shedding from blades; blades
striking people, vehicles and other objects; and wind turbulence
causing damage and harm to people, structures, flora, and fauna. De-
sign considerations and site planning may eliminate or minimize these
occurrences. A major accident may occur if the tower supporting the
entire unit failed causing a collapse. This might occur as a result
of design error, storm, earthquake, or other external cause.
Accidents associated with tidal energy systems would include
plant damage such as plant flooding, structure collapse, barrage fail-
ure or control gate failure, personnel injuries in the course of
operation, maintenance and installation; and marine ship collisions or
damage. Structural failure may occur as a result of storm damage,
design error, or earthquake. Accidents involving the turbine and
electrical generation are similar to those given for the hydroelectric
plant.
223
-------
Biomass conversion systems may have accidents such as explosion
and fire resulting from the production of methane, in the anaerobic
digestion process, or hydrogen and oxygen when power is used to dis-
sociate water. Depending upon the cost of such facilities, design
of gas handling and storage equipment, and the location of the plant
relative to workers and population centers, such accidents may be
major. Normal combusion techniques for converting biomass to other
energy forms would sustain accidents similar to other combustion
processes discussed in the coal and oil sections. As with other ener-
gy sources, occupational injuries will probably be more prevalent,
although minor, and would include those accidents occurring in agri-
cultural activities.
224
-------
SECTION 11.0
END USE
11.1 ELECTRICITY GENERATION
The commercial production of electricity arose out of the development of
the dynamo, the regulator, a distribution network, circuit switching and pro-
tective devices, measurement equipment (meters), and the electric light bulb.
The first commercial electric generating station was located on Pearl Street
in New York City. It was activated in 1882 with a power output of 792 KWe,
a capacity approximate one-thousandth that of modern plants. Since that time
our dependence on electricity has grown substantially to where approximately
25% of the total energy consumed today is used for electric power genera-
tion.1 y> Projections indicate that electric power generation will account
for about 40% of energy consumption in the year 2000.129,15,9 Electricity
production from all energy sources in 1974 totaled 1.86 x 1012 KWhg while pro-
jections for 1980 and 1990 are on the order of 2.70 x 1012 KWhe and 4.7 x 1012
KWhe, respectively.1™ The energy sources used to produce electricity are
those which were discussed previously in this report. The contribution of
various sources to the production of electricity is shown in Table 68 for 1973
and 1974 with projections for 1980 and 1990.13°
Electric power generation from fossil fuels is based upon the conversion
of chemical energy to electrical energy via mechanical energy transformation.
This latter transformation is based upon the principle of a conductor moving
in a magnetic field inducing a current in the conductor. The basic components
in the electric power generation system discussed in this section are the
energy source, the power plant, the generator, and the transmission and dis-
tribution network.
11.1.1 Technology Overview
The basic types of power plant used in the generation of electricity are
the boiler-fired, gas turbine, combined cycle, fuel cell, and magnetohydrody-
namic. The hydroelectric power plant (including pumped storage plant), using
the hydraulic turbine, was treated in Section 7.1 and will not be discussed
further here. Both the fuel cell and magnetohydrodynamic type are in the
developmental stages and therefore are not commercialized.
In boiler-fired plants, a solid, liquid, or gaseous fuel is burned, the
heat energy being transferred to a working fluid (water and/or steam). The
heat energy in the working fluid is converted to mechanical energy by a tur-
bine and finally to electrical energy by driving a generator.
225
-------
TABLE 68. ELECTRIC POWER GENERATION BY ENERGY RESOURCE3
Ni
RESOURCE
Contribution by Resource
1973
Coal
Oil
Gas
Hydroelectric
Nuclear
Other
Total
846.0
310.7
336.0
271.1
83.3
2.3
1849.3
(45.7)
(16.8)
(18.2)
(14.7)
(4.5)
(0.1)
(100.0)
828.
298.
320.
300.
112.
2.
1861.
1974
4 (44
2 (16
2 (17
3 (16
0 (6
7 (0
9 (99
, KWhe
x 10
9 (% of total)
1980 Proj.
.5)
.0)
.2)
.1)
.0)
.1)
.9)
1310
453
242
292
400
3.5
2700
(48.
(16.
(9.
(10.
(14.
(0.
(100
5)
8)
0)
8)
8)
1)
.0)
1990 Proj .
1720
450
190
319
2000
21
4700
(36.
(3.
(4)
(6.
(42.
(0.
6)
6)
8)
5)
5)
(100.0)
aReference 130
-------
A steam power plant uses a conventional boiler and a steam turbine, and
generally is equipped with cooling systems. This is the most common type of
plant, accounting for approximately 78% of the U.S. generating capacity'.15
These systems transfer heat from conventional fossil fuels, nuclear, geother-
mal, or solar sources to water to produce high-pressure, high-temperature
steam which drives a turbine. The steam is then condensed to water and re-
cycled. To improve the efficiency of these plants various types of boiler and
turbine equipment and techniques are utilized. With fossil fueled systems,
additional gas cleaning devices must be used.
Gas turbine plants often are used to accommodate peak loads due to their
fast start-up time, low initial cost, and short delivery time. They now
represent approximately 8% of the installed generating capacity in the United
States. In this type of system, the gaseous or vaporized fuel is injected
into a combustion chamber together with compressed incoming air, the resulting
high-pressure, high-temperature exhaust being used to drive the turbine which
in turn drives both the generator and the compressor. This is similar to the
type of engine which is used in jet aircraft. The gas turbine, however, re-
quires a clean gas stream, necessitating a clean burning fuel or source of
high-temperature thermal energy such as a nuclear reactor.
The combined cycle power plant is a combination of these systems. In
this type, the hot exhaust of the gas turbine is used to generate steam in an
unfired boiler. The steam then drives a conventional steam turbine. As an
example, a 1000 MWe plant contains four gas turbines and their associated
generators as well as a steam turbine with its generator. These plants are
presently being used to serve intermediate system loads.
The proposed fuel cell system produces electricity from a controlled
electrochemical oxidation. The three types of fuel cells are hydrogen-oxygen,
hydrocarbon-oxygen, and reformer. The latter consists of a first stage re-
former reacting steam with coal or hydrocarbons to produce, a fuel composed of
hydrogen and carbon monoxide. This mixture is then used in the fuel cell.
Since intermediate heat steps are not required, the fuel cell can theo-
retically approach 100% efficiency. The reactants in such a fuel cell are
consumed only when the external circuit is completed, allowing the electrons
to flow and the electrochemical reaction to occur. Under continuous operation,
the removal of heat, water, and inert material entering the cell is necessary.
The power produced is direct current (DC) and therefore must be converted be-
fore use in normal existing power systems.
Magnetohydrodynamic (MHD) power plants may be used to produce electricity
in the future. This type operates at high temperature and bypasses the heat-
to-mechanical energy conversion step. With a conventional generator a magnet
is spun around a stationary conductor. With MHD the conductor is an electri-
cally conductive fluid flowing through a rectangular duct immersed in a mag-
netic field. Three types of MHD systems are currently being investigated:
open-cycle plasma, closed-cycle plasma, and the liquid metal system. The out-
put of an MHD system is direct current (DC) generated from the induced voltage
drop across the duct. Conversion from DC to AC would be required for com-
patibility with existing power systems.
227
-------
The electrical output of the electric generating station is directed to
its ultimate use by transmission and distribution networks. The transmission
system consists of the lines and cables (overhead and underground), trans-
former, converters and switches, and controlling and metering equipment.
Transmission lines have developed over the years to meet demands of usage.
Currently lines carrying up to 765 KV are in service with plans for 1500 KV
lines. Super-conducting power lines are being developed for AC and DC trans-
mission. There are now approximately 40,000 miles of overhead transmission
lines and 2,000 miles of underground transmission cables using about 4 million
acres of land.
Ultrahigh voltage (above 745 KV) transmission is currently of great
interest as a means of increasing the capacity and efficiency of transmitting
power. Hazards accompanying high voltage and cost and environmental problems
associated with rights-of-way for overhead transmission systems have led to
increased use of underground transmission. Two methods are currently being
studied. One method is compressed gas insulation in which a compressed gas
such as sulfur hexafluoride (SF^) fills the pipe in which the wire is suspended
thus improving heat transfer and lowering the dielectric loss. The other
method involves cryogenically supercooling certain metals which causes a re-
duction in their resistivity, enabling smaller wires to be used for carrying
more power.
A distribution system provides power at a usable level to the customer.
Such a system consists of subtransmission lines (69-138kv), primary distribu-
tion lines (2.4-34.5kv), distribution transformers, secondary transmission
lines (120-240v), and service lines to the customer.
11.1.2 Accident Overview
The accidents which may occur in the course of electric power generation,
transmission, and distribution are shown in Table 1. Those technologies which
have unique components are listed separately. Generators, transmission, and
distribution systems are generally common for all energy sources and are there-
fore listed under a general heading.
Data on accidents associated with electrical power generation are sparse.
Present Federal Power Commission (FPC) regulations require accidents to be
reported to the FPC only when loss of power occurs. Thus, it is conceivable
that an explosion in a boiler killing or injuring many workers will not be
reported because an auxiliary boiler or generator took over and prevented a
power outage. Many state regulatory agencies have reporting requirements
which are more strict than those for the federal government (e.g., the New
York State Public Service Commission).131
A serious accident which can occur at a boiler fired plant is explosion
of the boiler, whether it is of conventional or fluidized bed design. Boilers
operate at a combustion temperature of approximately 1500°F and pressures of
10 atmospheres.18 Such pressures can present an explosion hazard. The proba-
bility of explosion can be increased by the high dynamic stresses caused by
high temperature and erosion in localized areas where temperatures may be
228
-------
greater than 1500°F. Additional problems in a fluidized bed combustor are
fires which can lead to explosion and interior coal dust explosions which can
increase the pressure four to thirteen times.132
Fires and explosions may involve the fuel and fuel handling equipment.
One major problem with using crude oil as a fuel is liquid and vapor leak-
age.133 For example, an oil explosion and fire in the Essex generating plant
in Newark killed a worker thirty feet away. In San Antonio in 1971, an oil
line ruptured spraying a hot steam line and causing a fire.I34 However, all
boiler accidents may not necessarily be due directly to the fossil fuel itself.
For example, a boiler exploded in western Pennsylvania because of a servomotor
oil leak which caused the plant to remain down for 43 days.
Several additional but infrequent accidents can occur in the boiler fired
plant. Subatmospheric pressure in the condenser could cause an implosion,
destroying the condenser. A tube rupture in the boiler can occur, leading to
damage to the equipment as well as shutdown. A high pressure, high tempera-
ture steam line could rupture causing injury or death to nearby workers. Sta-
tistics on these accidents were not available.
The most severe accidents which may occur at a gas turbine plant are
explosion, asphyxiation, and ruptured lines. Explosion, with possible sub-
sequent fire, can occur in the turbine, compressor, combustor, and recuperator.
In gas turbine and steam plants, explosions may occur in the fluidized bed
combustor system with ejection and dispersion of the bed contents. The con-
tainment vessel in an atmospheric fluidized bed system could rupture as a
result of accidental pressurization whereas pressurized vessels can suffer
burn-through and explode. Both atmospheric and pressurized fluidized bed
systems, fired with coal, also run the risk of fire from spontaneous combus-
tion. An additional accident which may occur in gas turbine plants is
asphyxiation from toxic working fluids. Two developmental gas turbine systems,
integrated low-BTU gasification and refined coal, could release toxic sub-
stances such as hydrogen sulfide," carbon monoxide, and coal tar volatiles
from leaks, pressure ruptures, process failures, or human error. The fluid-
ized-bed coal combustor could release alkali metal hydroxides through leaks,
pressure ruptures, error, or spilling. In the condenser or boiler an acci-
dent could occur which would result in an alkali metal water reaction causing
fire and explosion.
The experimental fuel cell system may be subject to two accidents. The
first would involve a gas leak which may result in asphyxiation, fire, or ex-
plosion. For example, an oxygen leak near electrical equipment could cause
a fire. The second, would be a fire caused by the failure of the heat removal
system. The frequency of such accidents cannot be estimated at the present
time.
As with other developmental systems, the MHD system has several potential
accidents associated with its operation. These accidents include explosion
and fire, asphyxiation, and magnetic explosion. If the vent or vapor^handling
system fails a pressure explosion may occur involving the cryogenic liquids
used in the system. Any surfaces at liquid helium temperatures exposed to the
229
-------
atmosphere would liquefy or freeze the air. If either the liquid or solid
air is contaminated with the fuel, coal dust, or hydrocarbons, the mixture
can be explosive. A magnetic explosion is unique to an MHD system because
it uses a large superconducting magnet providing very high flux densities.
The stored magnet energy is estimated to be on the order of 570,000 BTU or
approximately 280 pounds of TNT. An accidental rapid release of this energy
in either mechanical or electrical form is a magnetic explosion. It can be
caused through a loss of coolant or by mechanical damage. -" The resulting
explosion could injure or kill workers as well as destroy the equipment. The
probability of these types of accidents occurring has not been established.
Some other accidents which may occur in the generation or electricity
include power failure, flooding, electrical fires, and occupational injuries.
The power failure, usually caused by an accident in the plant which knocks
out the generator, causes widespread, though general damage. Customers are
unable to use machinery, refrigerator and freezers, heating and air condi-
tioning equipment, lights, etc. In addition, failure in a major generating
facility may affect the network into which the power is fed causing imbal-
ances. This in turn could cause a widespread power failure such as occured
in the northeast United States in 1964. The power failure may be due to
mechanical or electrical equipment failure, storms and/or flooding, or short
circuiting caused by animals, birds, or humans. Major power interruption
reported to the FPC due to accidents in generating stations for the years
1967-1970 are shown in Table 69-
Power outages and occupational injuries are associated with the trans-
mission and distribution network. Line failure is a major cause of power
outages. This can occur as a result of lightning strikes, falling trees,
ice accumulation, storms, and vehicles striking power poles. Electrical
occurrences such as short circuiting, insulation failure, and line overload
and resulting fires cause power outages. A list of accidents by customer-
hours lost is shown in Table 70. 136
One of the accident types unique to overhead power transmission is the
effect of corona discharge from the lines. Ozone created as a result of
corona discharge can damage plants and crops. Transmission lines also could
effect the operation of electrical medical devices such as cardiac pacemakers;
however, data on this subject are lacking.
The occupational injuries associated with workers involved with power
generation of all plant types have been tabulated. H» 137, 138 The types of in-
juries include those affecting the back, legs, arms, face, and eyes. Causes
include slipping, lifting, pulling, being struck by objects, steam burns,
electrical burns and shock, and inhalation. The most prevalent injury was
back strain.
A study was conducted of generating plant accidents for different fuel
types for 1969-197 2. 1J/ The frequency and severity of these accidents, which
occurred during 15 million manhours of work at 6 utilities, are tabulated in
Table 71.1J/ Accident rates expressed in terms of a 1000 MWe steam electric
plant with a load factor of 0.75 are shown in Table 72. 137 The published data
230
-------
TABLE 69. MAJOR POWER INTERRUPTIONS 1967 TO 1970
AS REPORTED TO FEDERAL POWER COMMISSION
YEAR TYPE OF ACCIDENT
1968 Lost boiler fire
1968 Generator relayed out bearing vibration
1968 Water pump
1968 Boiler draft fan failed
1969 Lost control on plant gas burner
1970 Flashover on generator excitor
1970 Boiler tube rupture
.231
-------
TABLE 70. RANKING OF ACCIDENTS IN 1975 IN ELECTRIC UTILITIES AS REPORTED TO THE FEDERAL POWER COMMISSION3
Accident Group Customer Hours Lost Power Lost (MWeh)
Accidental shutoff and overload includes
overloading, tripping and failures of
circuit breakers and relays 6,580,883 4,334
Meteorological phenomenon (not mentioned
elsewhere) includes line failures due
to wind, hurricane, ice, tornado, etc. 2,643,281 6,416
Vehicles or other objects striking
transmission lines 2,608,018 5,889
Electrical storms 333,385 1,141
Extraneous fires (destroyed plant) 112,500 3.6b
Inherent faults 73,862 337b
Arcing 30,987 57
Explosions 27,246 b
Total 12,382,916 18,177.6+
aReference 136
bOne or more incidents not reported.
-------
TABLE 71. OCCUPATIONAL INJURIES DUE TO ACCIDENTS IN GENERATING PLANTS3
Accident Rates
Plant Type
Frequency0
Severity6
1969 1970 1971 1972 1969 1970 1971 1972
Coal
Oil
Combination fossil
Total, all fossil
Hydroelectric
Nuclear
Total, all plants
Electric production plants^
Electric utilities industry0
3.6 5.6 11.1 10.8
0 13.45 6.91 13.69
121 1930 2410 1950
0 930 2850 461
3.
3.
3.
1.
3.
8
6
9
3
6
4.
4.
4.
3.
4.
0
7
0
4
5
6
7
0
3
6
6
6
.4
.5
.6
.1
.5
.82
.08
8.2
9.1
4.1
3.0
8.1
360
290
268
136
280
82
605
3750
84
930
275
900
79
18
755
1540
1710
149
43
1430
aReference 137
bEdison Electric Institute data
CNational Safety Council, 1972 Accident Facts
dFrequency rate is the number of disabling work injuries per 1 million
employee-hours exposure
eSeverity rate is the total days charged for work injuries per 1 million
employee-hours exposure
233
-------
TABLE 72. ACCIDENT RATES ASSOCIATED WITH A 1000 MWe STEAM ELECTRIC PLANT3
CLASSIFICATION Annual Occupational Accident Rates
Coal Oil Gas
Fatalities due to accidents
(frequency rates'3)
Power plant operation
Published data 0.03 0.03 0.03
Philadelphia Electric 0.10 0.013 0.007
Company Survey
Non-fatal injuries due to accidents
(frequency rates'3)
Power plant operation
Published data 1.2 1.2 1.2
Philadelphia Electric 3.3 1.2 0.7
Company Survey
Man-days lost to accidents, all types
(severity ratesc)
Power plant operation
Published data 350 350 350
Philadelphia Electric 750 130 70
Company Survey
aReference 137
^Frequency rate is the number of disabling work injuries per million
employee-hours exposure
cSeverity rate is the total days charged for work injuries per million
employee-hours exposure
234
-------
used is that of the National Safety Council (NSC) extrapolated by the authors.
NSC figures for 197211 show that for the sampled electric utilities (218 units)
the frequency rate for fatal and permanent total disability was 0.12 injuries
per million manhours exposure. The frequency of permanent partial disability
was 0.25 per million manhours, temporary total disability 6.05 per million man-
hours, and all disabilities 6.42 per million manhours. The utilities ranked
tenth in frequency of injuries among, all industries. The severity rate was
1,043 total days charged for injuries per 1 million manhours exposure. For
the 3-year period, 1970-1972, the frequency for electric utilities as 6.31
per million manhours with a severity of 1,003 days charged per million man-
hours. These figures indicate that the frequency rate of accidents for all
disabilities in the electric utility industry are below the average, but the
severity is above average. Transportation accidents involving employees were
higher than the overall rate11 for all reporting truck fleets as shown in
Table 73.11 Although the overall number of occupational injuries are large,
these accidents are mostly of a minor nature.
11.2 OTHER END USES
In addition to electric power generation, which represents approximately
25% of total energy consumption,1™ end uses of the various resources are
transportation, industrial, residential, and commercial.
Transportation accounts for approximately 25% of the total energy con-
sumption, which is the same as the amount used in power generation. 2"»15 of
that, the automobile uses more than half (52%) with trucks using about 22%,
domestic commercial aircraft more than 9%, ships and barges 4%, and railroads
less than 4%. Military and government installations, automotive and aviation
lubricants, and miscellaneous-transportation modes account for the remainder
(approximately 8%).15>18
The industrial sector is the largest user of energy accounting for approx-
imately 28% of the total consumption in 1971. It was estimated that this use
represented 22,623 x 1012 BTU's.139 Of this energy, more than half was used
in fuel burning or other thermal operations. Manufacturing used approximately
85% of the industrial energy consumed, with agriculture and mining equally
dividing the remainder. The annual energy consumption for six major indus-
tries is shown in Table 74.15 The chemicals group includes the manufacture
of basic, intermediate, and end use chemicals including Pharmaceuticals. It
was reported that in 1972 more than 30% of the tonnage and 60% of the value
of all organic chemicals were petrochemicals.
Paper and allied products use was basically for heat and mechanical
operations, which has been estimated at 90% and 10%, respectively.15
Stone, clay, glass, and concrete usage is limited to kiln heat and me-
chanical operations for crushing, conveying mixing, etc. The food and
allied products group includes the manufacture of beverages and food as
well as vegetable and animals fats and oils, prepared animal and bird feed,
and manufactured ice.
235
-------
TABLE 73. MOTOR TRANSPORTATION ACCIDENT RATES 1970-1972£
No. of Fleets
Vehicles Vehicle miles
(000)
Accident Rate 1970-1972
Accident Rate
Public Utility -
Electric
64
38,837
463,922
11.47
11.53
Trucks - all
1,866
244,745 4,742,568
9.86
10.86
^Reference 11
-------
TABLE 74. ANNUAL INDUSTRIAL ENERGY CONSUMPTION FOR VARIOUS FUEL TYPESa>b
to
u>
Industrial Group
Chemical
Primary Metals
Stone, Clay, Glass,
and Concrete
Paper and Allied Products
Food and Allied Products
Transportation Equipment
Coal and Coke
696
1,706
702
373
177
78
Natural
2,335
1,095
1,289
420
378
84
Fuel Type (1012
Gas Residual
183
170
433
196
65
14
BTU's/year)
Distillate
59
77
180
71
49
1
Electricity
1,083
1,018
317
304
261
146
Total
Energy
4,356
4,138C
2,921
1,364
930
323
Total
3,732
5,601
1,061
437
3,129
14,032
aReference 15
^Petroleum refining not included. Most data are for 1970. Paper and transportation equipment data
are for 1971.
cIncludes 72 x lO^ BTU's of petroleum coke used in primary aluminum production.
-------
The transportation equipment group includes the manufacture of equipment
for passenger and freight transport on land, sea, and in the air. The major
component in this group is the manufacture of motor vehicles (cars, trucks,
tractors, buses, and trailers).
The residential and commercial sector accounts for approximately 23% of
the total consumption. Both areas include energy use for space heating, water
heating, cooking, clothes drying, refrigeration, air conditioning, lighting,
and various appliances (e.g., television, can openers) and equipment (e.g.,
elevators, computers, office machines). The largest use within residential
and commercial areas is space heating, followed by water heating and refriger-
ation. An example of the fuel consumption for the various uses are shown in
Table 75.18 For residential and commercial use the combined consumption of
natural gas and distillate oil accounted for 50% of the total energy consump-
tion for the sector.
There are many accidents associated with end use of the various energy
sources. Some representative accidents for each sector are shown in Table 1.
The transportation sector has a variety of accidents which occur fre-
quently and can be considered major. The most common accident is the ve-
hicle collision involving one or more cars, trucks, or buses. This event
becomes major when there are multiple vehicle collisions such as occur
periodically on our interstate highways and turnpikes. The passenger bus
accident may kill or injure a number of the passengers and may result from
collision with another vehicle, a fixed object, running off the road, or
collision with a train. While there are a significant number of train ac-
cidents, only about one-third are a result of regular passenger and freight
operation. The remainder are involved with servicing and other operations.
There has been an increase in accidents, injuries, and damage over the past
several years which has been ascribed to deferred maintenance of track and
equipment.140 Data for these accidents are shown in Table 76.
Commercial aircraft accidents can be major or minor. A fire, explosion,
collision, or mechanical failure which occurs during a commercial flight and
which prevents the operation of flight systems can result in the loss of
many lives as well as the airplane. Commercial airline fatalities reached a
low of 146 in 1970 but increased to 467 in 1974. ^
Private plane accidents, while numerous, are generally minor. However,
private planes occasionally crash into occupied buildings such as schools.
Motorized farm equipment accidents involving death or injury are re-
latively common but generally affect only the operator. Farm vehicle ac-
cidents accounted for 3,200 deaths in 1973, for a death rate of 33.8 deaths/
100,000 farm residents. There were also 110,000 disabling injuries in this
group.2
Ship collisions can be either major or minor depending upon the nature of
the accident. Accidents involving tankers are representative of many accidents
238
-------
TABLE 75. ANNUAL ENERGY CONSUMPTION FOR COMMERCIAL AND RESIDENTIAL USESa'b
Fuel Type (1012 BTU's/year)
U)
End Use
Residential
Space heating
Air conditioning
Water heating
Refrigeration
Cooking
Commercial
Space heating
Air conditioning
Water heating
Refrigeration
Cooking
Total
Natural Gas
4,375
14
945
NC
328
1,857
43
419
NC
120
8,099
LPG
476
NA
85
NA
56
83
NA
18
NA
5
723
Distillate
2,294
NA
214
, NA
NA
569
NA
NC
NA
NA
3,076
Residual
NC
NA
NA
NA
NA
1,173
NA
NA
NA
NA
1,173
Electricity
708
570
744
1,126
310
NC
868
263
111
26
5,392
Coal
NC
NA
NA
NA
NA
427
NA
NA
NA
NA
427
Total
7,853
584
1,988
1,126
694
4,109
911
700
111
151
18,893
^Reference 18
bData covers 1970
NA = Not applicable
NC = Not considered
-------
TABLE 76. TRANSPORTATION ACCIDENTS
js
O
Transportation Mode
Railroad3
Collisions
Derailments
Other
Total (train)
Total (all including
service and non-
train)
Airplane
Commercial Operations
General Aviation
Motor Vehicle0
Highway 16
Bus (Commercial
and school)
Boating - Recreational
Farm Vehicles
Total Accidents
1973 1974
1,657 1,551
7,389 8,513
652 630
9,698 10,694
27,620 30,896
43 47
4,255 4,425
,648,100^
202,000
Fatalities
1973 1974
13
40
86
139
1,908
227 467
1,412 1,438
55,800 46,000
410
1,750 l,475h
3,200
Accident Ratee
Total Fatalities
1973
1.99f
8.89f
0.78f
1.08
24.18
0.016
1.14
22.34e
1974 1973
1.86f
10.22f
0.76f
1.28
27.2s
0.019 0.003
1.09 0.193
4.3
33. 81
1974
0.003
0.180
3.6
aReference 140 GReference 2
Reference 141 Reference 142
^Accident rates: Trains - accidents/million train-miles;
Airplanes - accidents/million aircraft flown-miles;
Buses - accidents/million vehicle-miles;
Motor vehicles - accidents/100 million vehicle-miles;
General Aviation - accidents/100,000 aircraft-miles
•^Accidents/Million train miles
^Casualties/Million train miles; casualties
include killed and injured.
^Estimated
^eaths/lOOjOOO farm residents
-------
involving large ships. Tanker collisions were discussed in Section 4.0.
Accidents involving pleasure craft are numerous but minor,
Table 68 indicates representative types of energy related accidents in
the industrial sector. Accidents such as fire, explosion, and flooding
comprise a major proportion of the accidents, Accidental release of hazardous
substances also can occur.
The commercial and residential sector accidents are generally minor
although almost all occur frequently. While a residential gas explosion
usually is minor, major explosions can occur. For example, a gas explosion in-
New York City on January 13, 1967 engulfed several city blocks in fire causing
extensive damage. The cause of such accidents are generally gas leaks which
are ignited. These potentially are common occurrences as evidenced by the
widespread usage of natural gas as shown in Table 75.
Accidents occur most frequently in the home. These accidents include
falls, fires, burns, asphyxiations, electrocution, and poisoning. In 1973
there were 26,000 deaths (a rate of 12.4/100,000 persons), 4.1 million dis-
abling injuries, and 18 million non-disabling injuries.
Most home accidents do not involve use of energy. However, accidents
involving appliances, heating, and other equipment can cause injury, damage
to equipment, or fire resulting in extensive damage. The damage caused by
these accidents usually is small. A summary of building fire losses caused
by electrical, heating; and cooking equipment is shown in Table 77. The
average accident caused less than $2000 in property damage.
241
-------
TABLE 77. BUILDING FIRE LOSSES CAUSED BY ELECTRICAL, HEATING, AND COOKING EQUIPMENTa»b
Cause Number of Fires Loss ($ million)
Electrical 162,600 315.8
Wiring and general equipment 101,600 203.1
Motors and power consuming 61,000 112.7
appliances
Heating and Cooking Equipment 155,200 177.6
Equipment, defective or
misused 89,400 116.7
ho
£j Chimneys, flues overheated
or defective 21,800 16.2
Hot ashes, coals 6,800 3.9
Combustibles near heaters, 37,200 40.8
stoves
aReference 11
^Estimated for 1972
-------
SECTION 12.0
ADVERSE NATURAL AND MAN-CAUSED INCIDENTS
This section deals primarily with events that are of such major propor-
tions that they may have catastrophic effects. In general, these incidents
"arrive with very little warning, have rapid development, and have potential
for substantial destruction.ul^3 xhe natural incidents considered are hurri-
canes, tornadoes, floods, tsunamis, snow and ice storms, earthquakes, land
subsidence, avalanches and landslides*, volcanic eruptions and meteorite im-
pact. Also covered are anthropogenic unscheduled incidents: airplane and
missile crashes, sabotage, terrorism, and war activities.
An overview of the severity and frequency of each of these is given in
Table 1. An in-depth study of each of these is clearly beyond the scope of
this paper. The purpose here is rather one of collection of summary infor-
mation, and the presentation of a broad overview of the vulnerability of the
non-nuclear energy systems. Precise cost figures of damage to energy facili-
ties as a result of these incidents were difficult to determine and are gen-
erally not included.
All industrial installations are potentially vulnerable to disasters.
In general, their vulnerability is dependent upon their geographic location,
their physical characteristics, and their dependence on other industrial en-
tities for supply materials, transportation, and other resources. Often the
magnitude of the vulnerability can be reduced by proper management and proper
contingency planning.
Emphasis is placed on the vulnerability of petroleum and the natural gas
industries since together they supply 75-80% of the total energy requirements
of the nation.
12.1 NATURAL DISASTERS
The effect of a natural disaster on an industrial installation can often
be so severe and damaging that it can be compared to the effect of a nuclear
blast whose point of detonation is a few miles from the plant. In particular,
the high wind velocities and low pressures associated with tornadoes, and the
extreme heat generated by a major fire are quite similar to the condition
following a nuclear explosion. The loss of life and economic damage can be
catastrophic. Up to 26 natural disaster areas are declared in a single year.
Natural disasters cause about 500 to 600 fatalities each year and economic
losses average 10 to 15 billion dollars.
243
-------
The types of damage produced by natural disasters include flood damage to
storage facilities, processing facilities, and transmission lines, physical
stress damage to processing facilities, rupturing of pipelines, and explosions
and fires resulting from escaped gases and highly flammable liquids.
Natural disasters are often concentrated in specific geographic locations.
Unfortunately, these generally coincide with the concentration of energy in-
dustries. These issues and others are elaborated below for each type of na-
tural disaster.
12.1.1 Hurricanes
Hurricanes or tropical cyclones develop over ocean areas in latitude 8°
to 15°N, then generally follow a north-easterly course approximating the East-
ern coastline. The counterclockwise winds associated with hurricanes may ex-
ceed 200 miles per hour. Winds in excess of 100 miles an hour are common in
the storm's path. Hurricanes are accompanied by torrential rainfall, result-
ing nearly always in serious flooding. *
On the average, eight hurricanes develop during a year, of which four
affect some part of the United States. Of these, approximately one each year
is of devastating proportions, with more than 50 lives lost, and economic dam-
age amounting to billions of dollars. Hurricane Betsy in 1965 caused close to
five billion dollars damage. It is estimated that losses from hurricane Agnes
in 1972 exceeded that figure. During the years 1951 through 1965, the total
loss of life due to hurricanes was 5,475, or an average of 110 per year.
Total economic damage (in unadjusted dollars) was over 4 billion dollars, for
an average of 87.6 million per year.144
Hurricanes generally develop in or near the Caribbean Sea or the Gulf of
Mexico. They then move north and east, generally hugging the coastline, until
finally dissipating over land, or heading out into the Atlantic. Hurricanes
have been known to travel as far north as Maine; however, the states most
affected are Texas, Louisiana, Florida, Alabama, Georgia, and North and South
Carolina. This same area contains a large concentration of oil refineries and
natural gas installations. By 1971 figures, 5,032,600 barrels of crude oil
per stream, day, or 38% of the national total, were processed in refineries lo-
cated in this region.144 For natural gas, 16,633,870 million cubic feet, or
74% of the total national natural gas supply came from this
Coal production areas are not greatly threatened by hurricanes. Approxi-
mately 5% of the national production total is mined in the hurricane risk area.
According to sources at the National Hurricane Center, the past summer
(1976) was something of an anomaly with regard to storm paths which affect the
United States. The whole hurricane pattern appears to have shifted towards
the east and south. No storms developed in the Gulf of Mexico or in the Carib-
bean. (This is only the second time in this century that this was the case).
Several of the Atlantic storms which did develop, recurved much further south
than usual. Similarly, there were an unusually large number (four) of major
Eastern Pacific storms which affected the land areas of Mexico and some parts
of California instead of remaining out in the Pacific. This unusual pattern
244
-------
appears to be related to the high atmosphere wind circulation patterns which
affect the Northern Hemisphere, and which seem to have shifted There is
however, no evidence to indicate that this shift is permanent,'and that next
year s pattern will be similar. Estimates of risk to facilities must be based
on the long-term histories of natural events, not on the anomalies of a given
year.1'*"
The damage caused by hurricanes is due to flooding and to the high wind
velocities which may carry a great deal of debris. Low-lying storage facili-
ties and processing facilities are vulnerable to floods. Flooding may also
break^communication and power lines, and disrupt aboveground transportation
facilities. This may result in a serious disruption of necessary supplies.
Offshore drilling platforms are particularly vulnerable. The Council on Envi-
ronmental Quality (CEQ) has studied the development of petroleum reserves lo-
cated in the Outer Continental Shelf (OCS) in the Atlantic and off the coast
of Alaska. CEQ has calculated the probability of one or more storms exceeding
the 100 year and 200 year design storm (the worst storm expected in 100 or 200
years) and the corresponding impact on offshore facilities. These figures are
shown in Table 78. These figures all assume a field life of 20 years. Doub-
ling the assumed field life results in a doubling of all the estimates. The
table clearly shows that though there is potential for damage to the well plat-
form, the storage facilities, and the tanker moorings, the probability of a
well blowout is low.
12.1.2 Tornadoes
Tornadoes are small, intense cyclones in which air spins at high speeds.
Tornadoes appear as funnels, containing condensed moisture, dust, and debris.
Wind speeds in a tornado exceed those in any other kind of storm. They are
estimated to be between 200 and 250 miles per hour. 4'
Tornadoes occur much more frequently than hurricanes, but they are more
short-lived (less than a day) and impact a relatively small area. Because of
their intensity and narrow limits of their paths, they take a heavier toll of
lives per year than hurricanes, but result in less economic damage. Approxi-
mately 10,000 people lost their lives in tornadoes during the years 1916
through 1970, an average of 194 per year. Property loss was estimated to ex-
ceed $40 million per year.144
Most tornadoes occur in the midwest. Oklahoma has the highest rate of
occurrence, approximately 300 per year. At least 200 per year occur in north-
ern Texas and eastern Kansas. The remainder of the midwest, as far north as
Iowa and as far east as western Ohio, has at least 100 per year. The northern
halves of Mississippi and Alabama also report at least 100 per year. The in-
cidence in the remainder of the country is insignificant.147
Out of a total national crude oil processing capability of 13,284,985
barrels per stream day in 1971, 3.6% are processed in plants located in Okla-
homa, 3.0% in Kansas, less than 0.1% in Nebraska, 6.6% in Illinois, 4.7% in
Indiana, and 0.1% in Missouri.144 Thus, approximately 18% of crude processing
capacity is located in areas where the incidence of tornadoes is greater than
100 per year.
245
-------
TABLE 78
ESTIMATE OF EFFECT OF STORMS EXCEEDING PREDICTED
100 AND 200 YEAR STORM LEVELS
PROBABILITY OF DAMAGE OR DESTRUCTION OF
100 Year Storm
design margin of safety =1.5
design margin of safety = 2.0
200 Year Storm
design margin of safety = 1.5
design margin of safety = 2.0
ilLITY OF
K COLLAPSE
9%
4%
5%
2%
PROBABILITY OF
WELL BLOWOUT
0.36%
0.16%
0.2%
0.08%
FLOATING
STORAGE
9.5%
4.4%
4.7%
2.2%
UNDERWATER
STORAGE
9.5%
4.4%
4.7%
2.2%
TANKERS SINGLE
POINT MOOR
4.4%
2.2%
-------
Of a total national production of 22,53,698 million ft3 of natural gas
in 1972, 8.0% came from Oklahoma, 3.9% from Kansas, and negligible amounts
from Nebraska, Illinois, Indiana, and Missouri. Approximately 12% of the
natural gas production is therefore located in areas where the incidence of
tornadoes is greater than 100 per year.145 The production capacity for
petroleum and natural gas located in Texas is not included here since these
wells are generally on the coastline where the incidence of tornadoes is low.
Coal production is not threatened to any significant degree by tornadoes.
Only 2% of the total national production capacity, or 11,660 short tons, is
mined in areas of high tornado incidence.
The extremely powerful wind-forces associated with tornadoes are their
primary danger. Small pieces of debris which are carried up acquire pierc-
ing strengths of projectiles. The effect is similar to the destruction caused
by explosives. Another danger is the extreme low atmospheric pressure in the
center of the tornadoes. Air at normal pressures within a building enclosed
by a tornado can blow out the walls because of the pressure differential. The
most vulnerable aspect of any installation is the actual physical plant and
its fittings. Aboveground communication and power lines are also vulnerable.
12.1.3 Floods
Floods occur as a result of a variety of atmospheric and/or environmental
conditions. Among these are rainstorms, hurricanes, tidal waves, snowmelt,
and dam failure. The geology and vegetation of a region significantly influ-
ence the severity of a flood. If the soil surface is porous, excessive water
can quickly drain out. If it is composed of clay, drainage is poor, and much
debris consisting of mud and silt is produced. An area which is heavily wooded
or containing grasslands will entrap considerable precipitation on leaf sur-
faces, and thereby reduce flood potential. ^
The total damage caused by flooding in 1966 in the continental United
States was approximately 1.7 billion dollars. In spite of various flP°d con-
trol measures, the predicted damage in 1980 is 2.4 billion dollars. The
loss of life associated with floods in the United States is not great, how-
ever, averaging less than 50 per year.
The areas most prone to flooding are the Northern Midwest states, par-
ticularly the Upper Mississippi, Missouri, and Arkansas River valleys. Next,
with about half the incidence rate of the Northern Midwest, are the northern
parts of both coastlines, and the states south of the Great Lakes region, par-
ticularly Indiana and Ohio. Less than 10% of the total national capacity of
oil production and gas production is based ,in these areas.J-44,l« The floo
-------
12.1.4 Tsunamis
Tsunamis or seismic sea waves are generated by sudden displacement of
earth under water, as during an earthquake. They are also, but infrequently,
caused by underground landslides. Though potentially very destructive (the
Krakataw volcanic eruption in 1883 caused a tsunami over 90 feet high, which
rolled over Java and Sumatra, and killed about 30,000 people), they are quite
rare. Since 1900, only seven major tsunamis with wave heights of 15 feet or
above have been reported in the world. However, the most recent, in 1964, hit
Crescent City, California, and resulted in 119 deaths and $104 million of dam-
age. In a risk assessment study for a liquified natural gas terminal (LNG) in
Oxnard, California, it was estimated that a tsunami of over 20 feet has the
probability of occurring once in a hundred years.
The study further states that should a tsunami occur, it would result in
some minor flooding of the LNG facility, but the force of the water would not
be sufficient to cause damage. However, LNG tankers would have to leave the
dock area, and the ship-to-shore transfer system would need to be drained.
Sufficient warning usually is available in the United States for tankers to
leave fixed berths. No large scale damage or LNG spills are anticipated as
a result of tsunamis. The previously cited CEQ study, summarized in Table
79^150 does show, however, that tsunamis could seriously threaten underground
storage facilities and tankers moored at fixed berths.
12.1.5 Snow and Ice Storms
Snow and ice storms are familiar and frequent phenomena in the northern
parts of this country. The loss of life resulting from winter storms is often
high. A United Nations study stated that 1,620 lives were lost in the United
States during the period 1947-1970. This compares with 1,890 due to tornadoes,
1,730 due to hurricanes, and 680 due to floods for the same period.
Though economic damage can be large, the major economic loss is generally
loss of livestock and loss of farm produce.-*-^ Substantial industrial instal-
lations are not seriously threatened by winter storms, except in the disrup-
tion of transportation facilities and overhead electric power and communica-
tion lines. Overhead lines are particularly susceptible. It has been reported
that during a winter storm in 1960, ice deposits of over eight inches in diam-
eter built up on some electric power lines, with a loading of twelve pounds
per foot causing many lines to fail. 2 In general, however, damages resulting
from winter storms are repaired in a matter of days. Winter storms are not
considered a major hazard to energy production facilities.
12.1.6 Earthquakes
Earthquakes are movements of ground surface. The initial motion which
occurs to relieve a point of stress produces shock waves which radiate out
from the center (epicenter). This motion is generally related to a fault,
which is a line along which earth moves in different directions; e.g., up on
one side and down on the other, or in opposite horizontal directions. Earth-
quakes are measured on an intensity scale, the Richter Scale, or the modified
Mercalli scale. The Richter Scale is a logarithmic scale, i.e., a value of
248
-------
TABLE 79 SUMMARY OF THE EFFECT OF NATURAL PHENOMENA ON VARIOUS ELEMENTS OF THE OIL PRODUCTION SYSTEMa
\£>
ELEMENT
Platform
Pipeline
(U Ashore
M
2 Afloat
o
w Underwater
Tankers Underway
Tankers Moored at
single point moors
Tankers at Fixed
Berth
SEVERE
STORM
Slightb
None
Slightf
Moderate
Moderate
Slightk
Slight1
Slight
EARTHQUAKE
VIBRATION
Slight0
None
Slight^
None
Serious
None
None
None
SOIL STABILITY
Slightd
Serious6
Slightd
Slight1
Serious
None
Slight1
None
TSUNAMI
None
None
None
None-'
Serious
None
None
Serious
VOLUME OF OIL AT RISK PER
500 to 1500 bbls/well/day
10,000 bbls or more
EVENT
Up to 1,000,000 bbls or greater11
200,000 to 1,000,000 bbls
100,000 bbls or greater
500,000 to 2,000,000 bbls
Reference 150
Storm forces in both areas are comparable to those in the North Sea.
Provided earthquake resistant design features are used.
Provided careful soil analysis program is followed.
It may be possible to reduce threat by line routing over less susceptible areas
Provided tanks are sited away from flood prone areas.
Provided free surface effect is reduced.
Dikes give protection against damaging oil spill.
Assumes control can be regained before floating storage grounds or capsizes.
Provided floating storage is moored in deep water.
Assumes regular inspections and prudent seamanship.
Assumes ship control is regained before grounding .
-------
one increase in the number (for example, from 4 to 5) means that the intensity
of the quake is higher by a factor of 10. Quakes of intensity 4 or less cause
minor damage. At an intensity of 4.5, local damage is caused. Intensity 7
indicates a major earthquake which can be recorded over the entire earth. On
the average, 14 earthquakes of intensity 7 or greater occur during the year
throughout the world. The maximum intensity reported was 8.6. A quake of
this magnitude releases tremendous amounts of energy (for comparison, the
Hiroshima atom bomb would have rated 5-7 on the Richter Scale), and results
in serious damage or destruction of practically all constructed edifices.
Earthquakes also produce secondary effects such as tsumanis, avalanches, and
landslides.^*°
Major earthquakes are potentially highly destructive. Throughout the
world, on the average, 10,000 die each year from earthquakes, with damage
amounting to $400 million dollars. Two major recent earthquakes in the United
States illustrate their potential for causing damage. The Alaskan Good Friday
Earthquake of 1964 measured between 8.0 and 8.6 on the Richter Scale, making
it one of the most severe recorded. Three hundred people (out of a total
Alaskan population of 250,000) were killed. Property damage was estimated at
$210 million dollars. Industry and civil life were completely disrupted with
•the destruction of harbors, docks, railroad tracks, bridges, highways, power
facilities, and structures of many kinds. Had the earthquake occurred in a
more populated part of the country, the loss of lives and damage would have
been many times greater. The San Fernando (California) earthquake in 1971
measured 6.6 on the Richter Scale, which is considered "moderate" rather than
"severe". Nevertheless, 65 lives were lost, and damage has been estimated at
$500 million dollars.1^° It has been estimated that if an earthquake as severe
as the 1906 San Francisco quake (estimated intensity of 8.25) were to hit the
area, the damage would be on the order of $20 billion.^^
The entire western part of the United States exhibits great seismic
activity. Areas of major seismic risk (frequent quakes with some damage)
include essentially all of California, and most of Nevada, Utah, Idaho, and
Montana. There are also pockets of major seismic activity centered around
Puget Sound, southwest Kentucky, southern Illinois, southeastern Missouri,
the South Carolina coastline, along the St. Lawrence Seaway, and in Alaska.
Moderate seismic activity may be found in the remainder of the West, around
Kentucky, Tennessee, and Indiana, around the Appalachian mountains, and in
most of New England and northern New York. The southern coastline, includ-
ing Texas, Louisiana, and Florida, is the only section of the country which
seems to be relatively free of seismic risk.
Approximately 18% of our total crude oil refining capability is located
in regions of major seismic risk, with another 14% located in moderate risk
regions. ^ The total natural gas supply is little affected. Only 3% is
located in the area of major risk, and an additional 5% in the area of moder-
ate risk. ^5
The most vulnerable part of the oil and gas industry are the pipelines.
Several major pipelines are located in California. Some cross major faults
more than 20 times.' Similarly, some of the major pipelines from the oil
and gas fields in Louisiana and Texas cross the significant risk region of
250
-------
the Kentucky, Missouri, Illinois area.145 The newly constructed Alaska pipe-
line crosses over four major active faults.148 The coal mining industry is
also vulnerable to earthquakes. Approximately 15% of the total national coal
production is located in earthquake-prone areas. The primary hazard from
earthquakes in coal mining is roof and sidewall collapse.
All parts of energy industry installations are potentially vulnerable
to earthquakes. The sudden collapse of part of a structure in which oil or
gas is housed or processed will often result in serious fires. Transmission
pipelines can be sheared by seismic movement. Landslides, caused by earth-
quakes, can severely damage storage tanks and processing machinery. Espe-
cially vulnerable are storage tanks which are built on inclines which can be
carried along by powerful landslides.
The CEQ Outer Continental Shelf oil and gas study can again be used as
an example of potential impact. The CEQ estimates for damage to offshore
facilities due to earthquakes are given in Table 80. The figures given
assume a field age of 20 years. Doubling the field age causes a doubling
of the estimated failure probabilities.^3
Table 79, giving the summary of the effects of natural phenomena on
the oil production system, shows that earthquakes are considered a serious
threat to underwater storage tanks.
Although probability estimates for failure of hydroelectric dams per se
are not available, a study of the probability of failure of 13 dams in Cali-
fornia has been performed..3 Probability estimates are presented in Table 81.
Estimated fatalities and damages are present in Table 82. These fatality and
damage estimates are based on the assumption of total and instantaneous fail-
ure of dams filled to capacity.
12.1.7 Land Subsidence
Subsidence can be defined as any displacement of a generally level
ground arising from surface or subsurface causes.-^" Earthquakes often cause
land subsidence. Other causes are the uncontrolled pumping of water or oil
from underground. This is a known problem in the development of oil fields
and is discussed elsewhere. Land subsidence (other than that associated with
earthquakes) is not generally considered to be a major hazard to the energy
industries.
12.1.8 Avalanches and Landslides
With the possible exception of a few oil and gas pipelines, no large
industrial installations are located in the path of snow avalanches. These
pose no significant danger to the oil and gas industry. Some coal mines may
be threatened by avalanche activity. Data on economic losses due to avalanches
are not readily available.
Landslides occur on slopes consisting of a variety of geological mate-
rials and develop through a variety of mechanisms and causes. Lanslides are
often initiated by earthquakes. Landslides can also be brought about by
251
-------
TABLE 80 ESTIMATES OF OFFSHORE DRILLING FACILITIES DAMAGE DUE TO EARTHQUAKES3
(FIGURES ASSUME A DESIGN SAFETY FACTOR OF 2.0)
MAGNITUDE OF QUAKE
(RICHTER SCALE)
STRUCTURE IS DESIGNED
TO WITHSTAND
LOCATION
OF STRUCTURE
AVERAGE NUMBER OF
TIMES QUAKE EXCEEDING
GIVEN MAGNITUDE
WILL CAUSE PLATFORM
COLLAPSE
AVERAGE NUMBER OF
TIMES QUAKE EXCEEDING
GIVEN MAGNITUDE
WILL CAUSE WELL
BLOW OUT
AVERAGE NUMBER OF
TIMES AN UNDERWATER
STORAGE TANK WILL
BE DESTROYED
6.6
7.2
Ln
8.6
(most severe
earthquake recorded)
Atlantic
(seismic hazard is
low to moderate)
Atlantic
Gulf of Alaska
(seismic hazard
is serious)
Gulf of Alaska
0.23
0.09
2.8
0.28
0.009
0.0036
0.11
0.011
0.23
0.09
2.8
(no estimate given)
a Reference 150
-------
TABLE 81 PREDICTION OF EARTHQUAKE IN CALIFRONIA BY
USING A COMPUTER3
FAULT THEORY
Name
of Dam
St. Francis
Van Norman
San Andreas
Lower Crystal Springs
Stone Canyon
Encino
San Pablo
Folsom
Shasta
Chatsworth
Mulholland
Upper San Leandro
Lake Chabot
Estimated
Probability
of MM VIII
per Yearb
NC
0.014
0.044
NC
0.012
0.012
0.076
NC
NC
0.013
NC
0.12
0.12
Estimated
Probability
of MM IX
per yearc
NC
0.0049
0.014
NC
0.0014
0.0013
0.032
NC
NC
0.0028
NC
0.063
0.057
Estimated
Probability
of MM X
per year^
NC
0.00003
0.003
NC
0.0003
0.0002
0.011
NC
NC
0.00003
NC
0.023
0.021
a Reference 150
b Significant Probability of Failure
c Substantial Probability of Failure
d High Probability of Failure
NC = Not Calculated
253
-------
TABLE 82 ESTIMATED EFFECTS OF TOTAL AND INSTANTANEOUS
FAILURE OF DAM FILLED TO CAPACITY3
NAME OF DAM
FATALITIES
DAY NIGHT
DAMAGE ASSESSED
IN U.S. DOLLARS
St. Francis
Van Norman
San Andreas
Lower Crystal Springs
Stone Canyon
Encino
San Pablo
Folsom
Chatsworth
Mulholland
Upper San Leandro
Lake Chabot
Shasta
a Reference 150.
b No allowance for evacuation.
- Not calculated -
72,000 123,000
21,000 33,000
(
125,000 207,000
11,000 18,000
24,000 36,000
260,000 260,000
14,000 22,000
180,000 180,000
36,000 55,000
34,000 34,000
8
3 x 10
1.1 x 108
5.3 x 108
5 x 10y
7.7 x 10y
6.7 x 108
6 x 10'
7.2 x 10C
1.5 x 10C
1.4 x 10*
254
-------
improper disposal of mining debris. The debris (called "tailings") is
i^.^i^]^
" ^ "« UUW11' une nunared torty-four people lost their lives. In gen-
la?ioni?n are n0t C°nsidered a maJ°r hazard to energy industry instal-
12.1.9 Volcanoes
Thfre are no currently active volcanoes in the continental United States.
Although mountains making up the Cascade Range of Washington, Oregon, and
California are volcanic, and there are active volcanoes in both Alaska (Mt.
Katmai) and Hawaii, (Mauna Loa, Kilauea), none appear to pose a significant
threat to energy systems.
12.1.10 Meteorites
Approximately 3500 meteorites strike the earth each year. Most of these
are very small, weighing under 10 pounds. On the average, one meteor weigh-
ing as much as a ton impacts the U.S. in a given year. The probability of a
meteor hitting a major industrial installation (in this case an LNG facility)
has been calculated to be 6.07 x 10~9, of which 5% would penetrate storage
tanks. It may be concluded that meteorites present no significant danger
to energy installations.
12.2 Adverse Man-Caused Incidents
The destruction potential of man-caused adverse incidents is very great.
For example, a well-planned act of sabotage or a nuclear weapon can complete-
ly demolish all structures within an area. Other man-caused events may pro-
duce minor immediate damage, but their cumulative effect may be significant
(e.g., industrial sabotage). Two types of man-caused destructive incidents
are discussed: those which are unintentional (airplane and missile crashes),
and those which are planned and executed with the intent to harm and destroy
(sabotage, terrorism, and war).
12.2.1 Airplane and Missile Crashes
The commercial air carrier fleet consists of approximately 2500 aircraft.
These generally have an excellent safety record. The number of major acci-
dents in which more than 100 people are killed are less than 1 per year. The
total number of accidents of all kinds average 30 per year. The figures for
general aviation aircraft is approximately 180,000. These average approxi-
mately 3800 accidents per year of which approximately 50% occur within 5 miles
of an airport. These accidents are generally minor in terms of loss of lives
and economic loss. If a falling aircraft or missile were to penetrate a stor-
age tank or pipeline and cause a fire, the damage could be extensive. How-
ever, the probability of an airplane crash into a plant has been estimated to
be approximately 2.5 x 10~4. 7 The probability of a mis-fired missile or
debris from an exploding missile hitting a plant or a pipeline is less than
one in 10 billion.7 Although these estimates are highly site specific, these
events do not appear to present significant hazards to energy industry install-
ations.
255
-------
12.2.2 Sabotage and Terrorism
No industrial installation is immune to sabotage. Sabotage may be per-
formed by enemy agents, by dissident groups, by disgruntled employees, or by
individuals who are mentally unstable.1^8 Figures on the prevalence of sabo-
tage are not available. Much of the data is classified. It is also possible
that some cases of sabotage are not recognized but are treated as accidents,
and that other cases are recognized but not reported.
It is impossible to determine the total amount of damage due to sabo-
tage. Until the 1960's, this nation was relatively free of civil disorder.
Violence as an expression of group dissatisfaction was not the norm. Since
the mid 1960's, however, sabotage and terrorist activities have increased.
The use of bombs and incendiary devices has multiplied. It is now estimated
that there are approximately 2000 bombing incidents per year, resulting in
20 to 25 deaths, and 170 to 180 injuries. Generally the bombings are poli-
tically motivated, with political targets such as government installations.
Often the bombings have been directed at "Big Business", especially banks.
The energy industries have not been a frequent target. Out of approximately
ninety violent acts which occurred during 1965-1975, only two had energy in-
stallations as a target.151 There is, however, no way of predicting what
future damage will be the result of sabotage and terrorism.
Industry experts feel that industrial sabotage and arson is not an
infrequent occurrence. Generally, the individual acts result in only minor
economic damage and are regarded more as an irritant than a danger. Most
industrial sabotage is done by disgruntled employees. Acts of industrial
sabotage frequently occur during strikes. For example, delivery vehicles
to and from an installation are waylayed and damaged or destroyed. Com-
munication lines and power lines may be cut. Some industrial sabotage is
done only to prolong the duration of an activity, so that the workers are
able to stay on the job and get paid for a longer period of time. The im-
proper welding found on sections of the Alaska pipeline, currently under
construction, may fall into this category.
The petroleum, gas, and power industries are important, visible, and
easily accessed targets of potential sabotage. The following is a list of
particularly vulnerable aspects of these industries:
1. Many oil and gas wells are clustered on relatively small
offshore platforms.
2. Often oil and gas storage tanks are next to producing equipment.
3. The production of many wells is funneled into a few pipelines.
4. Pipeline pumping stations are above ground, unmanned, unguarded,
and often located in remote areas. Pipelines are a tempting tar-
get for the saboteur, since he can cuase not only the loss of oil
or gas but also an extensive fire. It has been stated that a
trained group of a few hundred persons knowledgeable as to the
location of major pipelines and control stations and with destruc-
tion in mind, could starve our refineries from crude or cut off
256
-------
most of the natural gas.10 In the gas distribution system the
major point of vulnerability is the "city gate", where the regu-
lators, meters, and points of odorization are located.1^5
5. Many of the control of operations at the wells and at the refiner-
ies is done by computer. Often control of several operations is
located in one control room. These control rooms often have
little or no security. Computers are easily tampered with using
magnets or other means of destroying their data banks stored on
magnetic disks or tapes. Knowledgeable personnel may incorporate
changes into the computer programs which could cause millions of
dollars of damage.
6. The petroleum and gas industries are heavily dependent upon the
availability of electric power. Approximately 50% of total crude
production is dependent upon electric power. A serious power
outage could bring a major refinery to a halt.
The power industry is most vulnerable at major switching stations.
These are above ground, unguarded, and easily accessible. The transmission
lines themselves are also vulnerable. The power industry is dependent upon
oil, gas, and coal for the generation of electricity. Thus, the inter-
dependence of these energy industries is one of their most vulnerable
aspects.
12.2.3 War
It is unlikely that a conventional war will be fought in this country.
Therefore, vulnerability of energy installations to war means vulnerability
to nuclear war.
The effects of a nuclear weapon are as follows:
1. The nuclear explosion is an extremely powerful detonation. Shock
waves and tremors will be felt for many miles. Secondary earth-
quakes may be initiated.
2. A large portion of the energy of a nuclear explosion is emitted
in the form of light and heat. This is capable of causing skin
burns and starting fires.
3. An electromagnetic pulse, sometimes called radioflash, is pro-
duced which can damage electrical or electronic equipment in the
area.
4. Radioactive earth and other materials are drawn up into the
atmosphere. These radioactive particles then fall back to earth
over a period of time and cover a wide area. This is called
"fallout". The full range of effects from fallout are often
not apparent till months or years later, when the radioactive
materials have entered the natural food cycle and show up in
vegetable and animal products, and, ultimately, in the human body.
257
-------
There are no data available on the likelihood of a nuclear attack.
However, the Office of Emergency Preparedness has developed the following
profile of a "typical" attack:
1. The attack will last 48 hours and consist of 450 weapons repre-
senting a total yield of 2,300 megatons.
2. Most of the weapons will fall during the first hours.
3. Explosions will be evenly divided between air and ground bursts.
4. Nuclear devices will be directed at military, industrial and
population centers.
5. Ports and transportation centers will be especially hard hit.
6. Fallout will be a serious problem, but most parts of the. country
will be accessible within two weeks. No major section of the
country will be completely isolated due to fallout. Fallout in-
tensities at one hour past the beginning of the attack will ex-
ceed 10,000 roentgens per hour in a few cases. However, many
areas will receive less than 100 roentgens per hour.
Such an attack would reduce the population of the United States from 200
million to 145 million. During the first month, 31 million would require
hospitalization. Deaths during the first year would further reduce the
total population to 125 million, giving a total decrease of 38%. (The
population figures given here are all based on 1970 statistics).
The damage created by a 20 megaton blast would be as follows.
1. At impact site:
A 0.66 mile diameter crater will be produced. The fireball will
have a diameter of 2.9 miles.
2. Within 5 miles of impact:
The destruction will be essentially total. Ninety-eight percent
of the population will die.
3. 5 to 7 miles from impact:
Damage will be heavy, 40% of population will die, another 25%
will be injured.
4. 7 to 11 miles from impact:
Damage will be moderate. Five percent of population will die,
20% will be injured.
258
-------
5. 11 to 15 miles from impact:
Damage will be light, 10% of population will be injured.
6. More than 15 miles from impact:
There will be no direct damage or injury. Fallout will be the
only danger.
The primary targets for nuclear explosions will be major government,
military, and industrial installations. This implies that the North East
Corridor, the California Coast line, and the large industrial complexes
clustered around Chicago will be attacked. Due to their wide geographic
distribution, oil and gas wells are less vulnerable than refineries and
processing plants which tend to be clustered in a few major areas. The
power system is particularly vulnerable. The elimination of a few of the
major nodes in the power transmission network could cause loss of power
over a large area, such as the entire Northeast.
12.3 Precautionary Measures
The destructive aspects of natural disasters may roughly be grouped
into the categories of excessive wind, flooding, and earth movement. In
addition, due to the flammability of natural gas and many petroleum prod-
ucts, explosions and fires are a constant danger.
The best defenses against disasters are to design and build facilities
which stand a chance of withstanding the destructive forces, and to be aware
of whatever advanced warnings are available, and to take appropriate action.
The effect of the man-caused disasters of sabotage and war are similar
to those caused by natural disasters. Their severity varies, however, from
slight to the point of total annihilation.
1. Structural Safety Aspects - Careful planning goes into the design
of any major industrial installation. Among the many design criteria,
safety factors have great importance. It is customary, for example, to
'design a structure to withstand the worst 100 year or 200 year storms. In
hurricane or tornado-prone areas, this means designing for wind forces in
excess of 200 miles per hour. For example, storage tanks must be able to
withstand impacts from debris swept up by such winds. If flooding is a dan-
ger, low-lying processing and storage facilities are either avoided, or are
constructed appropriately. The siting of a major installation is generally
studied with great care. Clearly, construction on top of a known seismic
fault is to be avoided. In certain parts of the country, faults are quite
prevalent, and some structures such as pipelines must cross over them. Pipe-
line construction now includes several techniques that ensure that the lines
can "give" (at least to some degree) when subjected to pressure due to earth
movement.
259
-------
Careful consideration is also given to duplication of function or
backup facilities. For example, some degree of manual control for the
operation of the facilities must be provided, should the controlling com-
puter be destroyed or become disabled (all modern petroleum processing
plants are operating with computer control). Alternate sources of elec-
tric power are generally included in the plans. Power lines are, unfortun-
ately, among the most vulnerable components of the total energy systems.
Finally, specific "safety" areas are included in the design. These
areas may serve as a refuge for employees and as a repository of valuable
company documents. The design safety factors of these areas are several
times those used in the rest of the installation. Presumably they are able
to withstand moderate explosions and fires and provide protection from radio-
active fallout.
2. Immediate Protective Measures - Most natural disasters, as well as
nuclear attack, are preceded by some advance warning. The Federal Govern-
ment has instituted a Natural Disaster Warning System. Notice of impending
hurricanes, tornadoes and floods is given generally hours before the ex-
pected event. Some research is now being done to also develop earthquake
warning systems.
Regardless of the amount of advance warning, the best protective mea-
sure for any installation is a well-conceived and exercised Emergency Plan.
Such a plan should be put into effect as soon as a major emergency becomes
apparent. The Petroleum and Gas Industries have spent considerable effort
on the development of emergency plans. The following specific points are
taken from the emergency planning guidelines relating to nuclear attack, as
promulgated by the Office of Civil Defense. They are applicable to any
major emergency situation. Actions to be taken are:
1. Establish areas of relative safety for all personnel.
2. Ensure continuity of management.
3. Coordinate mutual aid with federal, local and state agencies, fire
and police departments, other installations, local utilities, etc.
4. Establish methods of communication.
5. Arrange emergency staffing of vital jobs such as repair crews.
6. Establish evacuation and shut-down procedures.
7. Arrange for emergency power supply.
8. Establish plant security procedures.
9. Establish fire control procedures.
10. Protect vital records.
11. Obtain emergency supplies.-
260
-------
APPENDIX I
AGENCIES AND ORGANIZATIONS CONTACTED
GOVERNMENT
Bureau of Labor Statistics
Bureau of Mines
Bureau of Reclamation
California Assembly - Office of Research
California State Energy Commission
City of Oxnard Planning Commission
Defense Civil Preparedness Agency
Energy Research and Development Administration
Environmental Protection Agency
Federal Energy Administration
Federal Power Commission
Federal Highway Administration
Federal Railroad Administration
Mining Enforcement and Safety Administration
National Highway Traffic Safety Administration
National Transportation Safety Board
New York State Public Service Commission
Occupational Safety and Health Administration
Office of Pipeline Safety Operations
Office of Safety Programs - Department of Transportation
261
-------
GOVERNMENT (continued)
Senator Barry Goldwater's Office
United Statss Coast Guard
United States Senate Committee on Interior and Insular Affairs
United States Senate Internal Security Committee
INDUSTRIAL AND COMMERCIAL FIRMS
Advanced Systems Laboratory, AMF, Inc.
Booz, Allen and Hamilton
Oil and Gas Consultants International
Petty-Ray Geophysical, Inc.
Scientific Applications, Incorporated
Seismograph Service Corporation
Standard Oil of California
INSURANCE COMPANIES
American Insurance Association
Industrial Research Insurance Company
Insurance Information Institute
Insurance Institute of America
Oil Insurance Association
MISCELLANEOUS
Air Pollution Control Association
American Academy of Environmental Engineers
American Institute of Mining Engineers
American Water Works Association
Chemical and Engineering News
262
-------
MISCELLANEOUS (continued)
Denver Research Institute
Institute for Disaster Research - Texas Institute of Technology
National Academy of Sciences
Society of Exploration Geophysicists
Society of Petroleum Engineers of AIME
Texas Petroleum Research Committee
Water Pollution Control Association
TRADE ASSOCIATIONS
American Boiler Manufacturers Association
American Gas Association
American Institute of Mining, Metallurgical, and Petroleum Engineers
American Petroleum Institute
American Petroleum Refiners Associations
Association of Oil Pipelines
Bituminous Coal Association
Edison Electric Institute
Electric Power Research Institute
Independent Petroleum Association of America
Independent Refiners Association of America
Independent Refiners Association of California
Independent Terminal Operators Association
Institute of Gas Technology
Mid Continental Oil and Gas Association
National Coal Association
National Fire Protection Association
263
-------
TRADE ASSOCIATIONS (continued)
National Gas Supply Committee
National Oil Jobbers Council
National Petroleum Council
National Petroleum Refiners Association
National Safety Council
Petroleum Equipment Institute
UNIONS
International Oil, Chemical, and Atomic Workers Union
International Union of Petroleum Workers
United Mine Workers of America
264
-------
REFERENCES CITED
1. Council on Environmental Quality, Energy and the Environment, Electric
Power, Washington, B.C., 1973. pp. 58.
2. National Safety Council. Accident Facts, 1974. Chicago, Illinois. 1974.
pp. 26.
3. Bureau of Mines, Department of the Interior. Injury Experience in Coal
Mining, 1970. Information Circular 8613. 1973.
4. Card, Lt. Cdr. James C., Ponce, Paul V., and Snider, Lt. Cdr. Warren D.
Tankship Accidents and Resulting Oil Outflows, 1963-1973. In: Proceedings
of the Conference on Prevention and Control of Oil Pollution. American
Petroleum Institute, EPA, and the U.S. Coast Guard, San Francisco, Cali-
fornia, 1975, pp. 205-213.
5. Oil Insurance Association. Loss Information Bulletin. No 2 (November
1972), No. 400-1 (July 1974), No. 400-2 (March 1975). OIA. Chicago,
Illinois.
6. American Gas Association, Analysis of the Office of Pipeline Safety
1970-1973, Reportable Incident Data for the Natural Gas Distribution
Companies, 1975, pp.33.
7. Science Applications, Inc. Risk Assessment of Storage and Transport
,tof Liquefied Natural Gas and LP-Gas, 1974.
8. Philipson, Lloyd L., "The Systems Approach to the Safety of LNG Import
Terminals." Prepared for the State of California Energy Resources Con-
servation and Development Commission, Draft Report, pp. VI-1 to VII-3.
9. Penner, S.S. and L. Icerman. Energy: Non-Nuclear Energy Technologies,
Vol. 2. Addison-Wesley, Reading, Massachusetts, 1975.
10. Comparative Assessment of Orbital and Terrestrial Central Power Systems,
Jet Propulsion Laboratory (Internal Document) Pasadena, California,
March 1976.
11. National Safety Council, Accident Facts, 1973 Edition, National Safety
Council, Chicago, Illinois, pp.96.
265
-------
12. Department of the Interior. United States Energy through the Year 2000.
Washington, B.C., 1972.
13. Bureau of Mines, Department of the Interior. Coal-Bituminous and Lignite
in 1974 (preliminary). 1976.
14. World Energy Conference. Survey of Energy Resources. New York, N.Y.
1974. pp. 61.
15. University of Oklahoma, 1975. Energy Alternatives, a Comparative Analysis.
Science and Public Policy Program. Norman, Oklahoma. 1975. pp. 1.1 -
1.121.
16. Bureau of Mines, Department of the Interior. Cost Analyses of Model
Mines for Strip Mining Coal in the United States. Information Circular
8535. 1972.
17. Katell, S. and E.L. Hemingway. Basic Estimated Capital Investment and
Operating Costs for Underground Bituminous Coal Mines. U.S. Bureau of
Mines (DOI) Information Circulars 8632, 8641. 1974.
18. Hittman Associates, Inc. Environmental Impacts, Efficiency, and Cost
of Energy Supply and End Use. Columbia, Maryland, 1975. Vol. I, Cht
3; Vol. II, Cht 2-8.
19. Bureau of Mines, Department of the Interior. Injury Experience in Coal
Mining. Information Circular 8599. 1969.
20. Bureau of Mines, Department of the Interior. Injury Experience in Coal
Mining. Information Circular 8555. 1967.
21. Bureau of Mines, Department of the Interior. Injury Experience in Coal
Mining. Information Circular 8556. 1968.
22. Theodore Barry and Associates. Equipment Operation Safety Study. Los
Angeles, California. 1972. pp. 66-70.
23. Theodore Barry and Associates. Industrial Engineering Hazards Associated
with Underground Coal Production, Vol. I. Los Angeles, California. 1971.
Vol. I, pp. 66-81.
24. Theodore Barry and Associates. Industrial Engineering Study of Hazards
Associated with Surface Coal Mines. Los Angeles, California. 1974.
25. Trauke, R.J. and J.E. Hopper. Accident Analysis by Functional Classi-
fication for Bituminous Coal Mines in 4-Foot to 8-Foot Seam Heights.
FMC Corporation. Santa Clara, California. 1972. pp. 27-36.
26. National Coal Association. Coal Facts, 1974-1975. Washington, D.C.
1975.
266
-------
27. Mining Enforcement and Safety Administration, Department of the
Interior. Mine Refuse Impoundments in the United States. Information
Report 1028. 1976. pp. 10-21.
28. Department of the Interior. Minerals Yearbook, Volume 1 - Metals
Minerals, and Fuels. 1973. pp. 903-1018.
29- American Petroleum Institute, American Gas Association, Canadian
Petroleum Association. Reserves of Crude Oil, Natural Gas Liquids,
and Natural Gas in the United States and Canada and United States
Productive Capacity as of December 31, 1974. Volumes 29 and 30.
30. Gillette, Robert. Oil and Gas Resources: Did U.S.G.S. Gush Too High.
Science, 185: 127-130, July 12, 1974, in University of Oklahoma, Energy
Alternatives: A Comparative Analysis. Prepared for CEQ, ERDA, EPA, FEA,
FPC, DOI and NSF, University of Oklahoma, Norman, Oklahoma, 1975 (p. 3-3)
623 pp.
31. Yearbook of Science and Technology. McGraw-Hill, 1975.
32. Department of the Interior. Statement, Questions and Policy Issues
Related to Oversight Hearings on the Administration of the Outer Con-
tinental Shelf Lands Act, Held by the Senate Committee on Interior and
Insular Affairs, Pursuant to S. Res. 45, March 23, 1972. Department
of the Interior, Washington, D.C. in University of Oklahoma, Energy
Alternatives: A Comparative Analysis. Norman, Oklahoma, 1975 (p. 3-19).
33. Association of Oil Pipelines, Press release, June 7, 1975; May 26, 1976.
34. American Petroleum Institute. Summary of Occupational Injuries and Ill-
nesses in the Petroleum Industry. API, Washington, D.C., April 1976,
24 pp.
35. American Petroleum Institute. Review of Fatal Injuries in the Petroleum
Industry for 1975. API, Washington, D.C., April 1976, 6 pp.
36. Young, G.K., D. Evans, and R.U. Jettmar. Analysis of Oil Spill Trends.
A Report to the Council of Environmental Quality. GKY Associates.
Alexandria, Virginia. August 1976.
37. Bayaert, Bruce. Analysis of Oil Spill Accidents for Environmental
Impact Statements. In: Proceedings of the Conference on Prevention and
Control of Oil Pollution. American Petroleum Institute, EPA, and the
U.S. Coast Guard, San Francisco, California, 1975. pp. 39-45.
38. Paulson, A.S., A.D. Schumaker, and W.A. Wallace. A Risk-Analytic
Approach to Control of Large-Volume Oil Spills. In: Proceedings of the
Conference on Prevention and Control of Oil Pollution. American Petroleum
Institute, EPA, and the U.S. Coast Guard, San Francisco, California, 1975.
pp. 301-306.
267
-------
39. United States Coast Guard. Marine Environmental Protection Program:
An Analysis of Mission Performance. Department of Transportation,
Washington, D.C., August. 1975.
40. Operations Research, Inc., Spill-Risk Analysis Program: Interim
Report E-2: Facilities Analysis. Operations Research, Inc., May 20,
1975.
41. National Academy of Sciences. Petroleum in the Marine Environment,
Washington, B.C., 1975, pp. 104-107.
42. Council on Environmental Quality. OCS Oil and Gas - An Environmental
Assessment. U.S. Government Printing Office, Washington, B.C., 1974.
43. National Petroleum Council. Environmental Conservation - The Oil and
Gas Industries (2 Volumes). Washington, D.C., 1971 and 1972.
44. Blumer, M., et al. A Small Oil Spill. Environment, Vol. 13, No. 2,
March 1971.
45. Blumer M. Oil Contamination and the Living Resources of the Sea. In:
FAO Technical Conference on Marine Pollution and Its Effects on Living
Resources and Fishing. Rome, Italy, December 1970.
46. Moore, Stephen, Robert Dwyer and Arthur Katz. A Preliminary Assessment
of the Environmental Vulnerability of Machias Bay, Marine to Oil Super-
tankers. Report No. 162. Massachusetts Institute of Technology,
January 1973.
47. Society of Petroleum Engineers, Perrine, Richard L., UCLA, editor.
Petroleum and the Environment. Society of Petroleum Engineers, Dallas,
Texas, 1975.
48. Blumer, M. and J. Sass. The West Falmouth Oil Spill, Report No. WHOI-
72-19. Woods Hale Oceanographic Institution, Woods Hale, Massachusetts,
1972.
49. Stongham (1972) quoted in: Brown, Richard D., Considerations in the
Establishment of Discharge Standards for Petroleum Refinery Effluents.
Presented at the Conference on Management of Petroleum Refinery Waste-
waters. January, 1976.
50. Battelle (1973) quoted in: Brown, Richard D., Consideration in the
Establishment of Discharge Standards for Petroleum Refinery Effluents.
Presented at the Conference on Management of Petroleum Refinery Waste-
waters. January, 1976.
51. Anderson (1973) quoted in: Brown, Richard D., Consideration in the
Establishment of Discharge Standards for Petroleum Refinery Effluents.
Presented at the Conference on Management of Petroleum Refinery Waste-
waters. January, 1976.
268
-------
, and Jerome H.
Ca^bri P and the Marlne Environment. Ballinger Publishing
., Cambridge, Massachusetts, 1974.
. » D°nald F" Carl H" Hershner, and Jerome
Co r Spills and the Marine Environment. Ballinger Publishing
Co., Cambridge, Massachusetts, 1974.
54. Stronghan, D. , quoted in: Boesch, Donald F., Carl H. Hershner, and Jerome
H. Milgram Oil Spills and the Marine Environment. Ballinger Publishing
Co., Cambridge, Massachusetts, 1974.
55. Nelson-Smith, A. Oil Pollution and Marine Ecology. Planum Press, New
iork, M.Y., 1973.
56. See Reference 42. Volume I.
57. Brown, Richard D. Considerations in the Establishment of Discharge
Standards for Petroleum Refinery Effluents. Presented at the Conference
on Management of Petroleum Refinery Wastewaters, January 1976.
58. Huff, John R. Study Analyses Offshore Rig Casualties. In: Oil and Gas
Journal, 74(48): pp. 71-86, 1976.
59. Department of Transportation, Office of Pipeline Safety Operations. Sum-
mary of Liquid Pipeline Accidents Reported on DOT Form 7000-1 From January
1, 1975, through December 31, 1975. Department of Transportation, Washing-
ton, D.C. , May 1976.
60. Boesch, Donald F., Carl H. Hershner, Jerome H. Milgram. Oil Spills and
the Marine Environment. Ballinger Publishing Co., Cambridge, Mass., 1974.
61. Department of Transportation. Sixth Annual Report of the Secretary of
Transportation on Hazardous Materials Control (CY 1975) . Washington,
D.C., 1976.
62. Brobet, William A., Transportation Accidents: How Probable? Nuclear
News, May 1973.
63. National Fire Protection Association. BLEVE - Boiling Liquid Expanding
Vapor Explosions (film) .
64. Oil Insurance Association. Truck Loading Rack Safety, Publication No.
71. Oil Insurance Association, Chicago, Illinois.
65. National Petroleum Refiners Association. Annual Summary of OSHA, Injury
Statistics, 1975. National Petroleum Refiners Association, 1976.
66. Oil Insurance Association. Hydroref ining Process Units - Loss Causes and
Guidelines for Loss Prevention, Publication No. 101. Oil Insurance
Association, Chicago, Illinois, 1974.
-------
67. Oil Insurance Association. Our Plant Is On Fire - Case Histories of
Gasoline Plant Losses, Publication No. 302. Oil Insurance Association,
Chicago, Illinois, 1972.
68. Oil Insurance Association. Fired Heaters - Loss Causes and Guidelines
for Safe Design and Operation, Publication No. 501. Oil Insurance
Association, Chicago, Illinois, 1971.
69. American Gas Association, Gas Facts, 1974, 209 pp.
70. Department of Transportation, Eighth Annual Report on the Administration
of the Natural Gas Pipeline Safety Act of 1968, 1975, pp. 1-30.
71. Federal Power Commission, Statistics of Interstate Natural Gas Pipeline
Companies, 1974.
72. Federal Power Commission, National Gas Survey, Volume 5, 1973, pp. 157-179.
73. National Academy of Science, Proceedings of Conference on LNG Importation
and Terminal Safety, Boston, June 1972, p. 262 (Washington, D.C., The
Academy, 1972).
74. American Gas Association, Introduction to LNG Personnel Safety, 1973,
53 pp.
75. Van Horn, Andrew J., Liquefied Natural Gas: Public Concerns, Safety
Issues, and Decision Making. Harvard University, Cambridge, Massachusetts,
unpublished manuscript (draft 1).
76. Wing, John F., Trends in World Shipping. Conference Proceedings on Bulk
Transportation of Hazardous Materials by Water in the Future. July 9-10,
1973. National Academy of Sciences, pp. 11-31.
77. Fay, James A. Unusual Fire Hazard of LNG Tanker Spills. Combustion
Science and Technology, 1973, Vol. 7, pp. 47-49.
78. May, W.G., W. McQueen, and R.H. Whipp. Spills of LNG on Water. Esso
Research and Engineering Company. Paper presented at the American Gas
Association Distribution Conference, May 1973.
79. National Academy of Sciences, LNG-Water Explosions, 1972, 51 pp.
80. Murray, F.W., D.L. Jaquette, and W.S. King. Hazards Associated with the
Importation of Liquified Natural Gas. The RAND Corp., June 1976. 63 pp.
81. Meyer, R.F. World Energy Conference, U.S. Geological Survey, OER,
Detroit, Michigan, 1974.
82. Creager, William P. and Joel T. Justin. Hydroelectric Handbook, 2nd
Edition. John Wiley and Sons, New York, 1950.
270
-------
83' o?rtLEwerS £0nfere"ce/UrVe? °f Energy Resources. National Committee
of the World Energy Conference, New York, 1974. Chapter VI.
84. Jertalett, AD and R j. Fox. Accident-Rate Sample Favors Nuclear.
Electrical World. July 15, 1974.
85. Kruger P and C. Otte. Geothermal Energy. Stanford University Press,
Stanford, California, 1973.
86. Statement of William M. Dolan, President, Geothermal Resources Council,
xn support of S.2608 prepared for the Committee on Finance of the U.S.
Senate. Reprinted in: The Geothermal Resources Council Newsletter,
March/April/May 1976.
87. White, D.E., and D.C. Williams. Introduction in Assessment of Geothermal
Resources in the United States - 1975. Circular 726, U.S. Geological
Survey, Department of the Interior, Washington, D.C., 1975.
88. Swetnam, G.F., Geothermal Industry Position Paper: EPA Regulatory Options
and Research and Development Information Needs. The MITRE Corp.,
November 1976, 97p.
89. Mueller, Diana L. TRQ. See Promise in Geothermal Energy - A Look at
East Mesa._ Geothermal Energy, Vol. 4, No. 4, April 1976.
90. Papadopulos, S.S., R.H. Wallace, Jr., J,B. Wesselman, and R.E. Taylor.
Assessment of On-Shore Geopressures-Geothermal Resources in the Northern
Gulf-of-Mexico Basin, in Reference 3.
91. U.S. Geological Survey. Assessment of Geothermal Resources in the United
States - 1975. Circular 726. U.S.G.S., Department of the Interior, Wash-
ington, D.C., 1975.
92. Goldsmith, M. Geothermal Resources in California - Potential and Prob-
lems. California Institute of Technology. 1971.
93. Axtmann, Robert C. Impact of a Geothermal Power Plans. Science, Vol. 187,
No. 2179, March 7, 1975. Also, Reference 9.
94. See Reference 10.
95- Duncan, L.J., S.A. Hans, E.P. Krajeski and H. Williams. Recommended
Procedural Guidelines for Environmental Analysis of ERDA Geothermal
Programs/Projects. MITRE Corporation Document MTR-7150. McLean, Virginia.
1976. 241 pp.
96. Levarsen, A.I. Geology of Petroleum. W.H. Freeman and Company, San
Francisco, California, 1967. pp. 28-29.
97. United States Department of the. Interior. Final Environmental Statement
for the Prototype Oil Shale Leasing Program. Washington, D.L.,
Six Volumes (variously paged).
271
-------
98. Katell, Stone and Weilmon. Oil Shale - A Clean Energy Source.
In: Proceedings of the Seventh Oil Shale Symposium, Quarterly
of the Colorado School of Mines. Vol. 69, No. 2, April 1974.
99. Cameron, Russel J. Attachment of Letter No. 57, p. 7. In:
Vol. V of the Final Environmental Statement for the Prototype
Oil Shale Leasing Program, Department of the Interior, November
1972.
100. United States Bureau of Mines, Minerals Yearbook, Volume II.
Washington, D. C., 1973.
101. Daniels, Farrington. Direct Use of the Sun's Energy. Yale
University Press, New Haven, Connecticut. 1964. Chapter 2.
102. Gustavson, M. R. Dimensions of World Energy. MITRE Document
M75-4. The MITRE Corporation, McLean, Virginia. January 1975.
37 pp.
103. Cherry, William R., Harnessing Solar Energy: The Potential.
Astronautics and Aeronautics, pp. 30-36. August 1973. Cited
in reference 106.
104. Dupree, Walter G., Jr., and John S. Corsentino. United States
Energy Through the Year 2000 (Revised). Bureau of Mines, U. S.
Department of the Interior, Washington, D. C., December 1975.
65 pp.
105. Rich, J. E. and J. G. Hewitt, Jr. Large Terrestial Solar
Arrays, in the Proceedings of the Intersociety Energy Conver-
sion Engineering Conference, Boston, Mass., August 1971.
Cited in reference 106.
106. Williams, J. Richard. Solar Energy, Technology and Applications.
Ann Arbor Science Publishers, Inc., Ann Arbor, Michigan. 1974.
120 pp.
107. Anderson, J. H. and J. H. Anderson, Jr. Power from the Sun,
Via the Sea, presented at the Susquehanna Section of the ASME,
November 1964. Cited in reference 9.
108. Putnam, P. C., Power from the Wind. D. Van Nostrand Company,
New York, N.Y. 1948. Cited in reference 9.
109. Jeffreys, H, Tidal Friction, Nature, 246, pp. 346, 1973. Cited
in reference 9.
272
-------
110. G. E. Space Division. Design Study of Wind Turbines SOLW to
to 3000 kW for Electric Utility Applications, Analysis and
Design, February 1976, King of Prussia, Pennsylvania.
111. Munk, W. H. and G. J. F. MacDonald. The Rotation of the Earth,
A Geophysical Discussion. Cambridge Monographs on Mechanics
and Applied Mathematics, Cambridge University Press, Cambridge,
England, I960. Cited in reference 9.
112. Isaacs, J. D. and D. J. Seymour. The Ocean as a Power Source,
International Journal of Environmental Studies, 4, pp. 201-205.
1973. Cited in reference 9.
113. Salta, S. H. Wave Power. In: Nature, 249, pp. 720-724, and
Chemical and Engineering News, 52, p. 16, October 14, 1974.
Cited in reference 9.
114. Donovan, Paul and William Woodward. An Assessment of Solar
Energy as a National Energy Resource, NSF/NASA Solar Energy
Panel, University of Maryland, December 1972. Cited in refer-
ence 106.
115. Brinkworth, B. J., Solar Energy for Man. John Wiley & Sons,
New York, N.Y. 1972.
116. McElroy, W. D. UCSD/NSF (RANN) Conference on Enzyme Engineering
as Related to Food and Energy Production, La Jolla, California,
July 22-25, 1974. Cited in reference 9-
117. Trombe, F. Solar Furnaces and Their Applications. In: Solar
Energy, Vol. 1, No. 2, pp. 9-15. 1957. Cited in reference 106.
118. Solar Energy. Moscow News, No. 2, January 1974. Cited in
reference 6.
119- Zerier, Clarence. Solar Sea Power. In: Physics Today, pp. 48-
53. January 1973. Cited in reference 106.
120. ERDA, Program Plan, Office of Solar Electric Applications,
Internal Working Document, August 6, 1976.
121 Wilcox, Carl. Motion Picture History of the Erection and Opera-
tion of the Smith Putnam Wind Generator. Wind Energy Conversion
Systems Workshop, NSF/RA/W - 73-006, Ed. by J. H. Savino, pp. 8-
10, December 1973. Cited in reference 106.
122. Noel, John M., French Wind Generator Systems. Wind Energy Con-
version System Workshop Proceedings, NSF/RA/W - 73-006, pp. 186-
196, December 1973. Cited in reference 106.
273
-------
123. Edmonson, William B. A 100 KW Windmill Generator. Solar Energy
Digest, Vol. 2, No. 4, p. 4, April 1974. Cited in reference 106.
124. Wilson, E. M. Tidal-Energy Development in the Handbook of Applied
Hydraulics, Section 42, C.V. Davis and K.W. Sorenson, ed.,
McGraw-Hill, New York, N.Y., 1969- Cited in reference 9.
125. Pelletier, B. P. and R. M. McMullen. Sedimentation Patterns in
the Bay of Fundy and Minas Basin. In: Tidal Power, T. J. Gray
and 0. K. Gashus, eds., Plenum Press, New York, N.Y., 1972. Cited
in reference 9-
126. Sorensen, K. E. Pumped-Storage and Tidal Power. In: Tidal
Power, T. J. Gray and 0. K. Gashus, eds., Plenum Press, New York,
N.Y., 1972. Cited in reference 9.
127. Haswell, C. K., et. al., Pumped Storage and Tidal Power in Energy
Systems. In: Journal of the Power Division, Proceedings of the
American Society of Civil Engineers, 98, pp. 201-220, 1972.
Cited in reference 9.
128. Isaacs, J. D. and D. Castel. Wave-Powered Generator System.
University of California Sea Grant Annual Report. Scripps In-
stitute of Oceanography, La Jolla, California, 1971-72. Cited
in reference 9.
129. Sobotka, R. Economic Aspects of Commercially Produced Solar
Water Heaters. Solar Energy, Jol. 10, No. 1, pp. 9-14, 1966.
Cited in reference 106.
130. Comparative Assessment of Orbital and Terrestial Centrol Power
Systems, Jet Propulsion Laboratory (Internal Document), Pasadena,
California, March 1976.
131. National Science Foundation. 1973, Energy, Environment, Pro-
ductivity-Proceedings of the First Symposium on RANN:- Research
Applied to National Needs, Washington, D. C.
132. Chew, Bernard B. The Impact of Recent Developments on the Future
of Electric Power. Proceedings of the American Power Conference,
Volume 37, pp. 978-985. 1975.
133. Personal communication from J. Hoffman, Power Division, New York
Public Service Commission.
134. Rao, C. S. L., Safety Aspects of Fluidized-Bed Combustion Sys-
tems, MITRE Corporation, April 1976.
274
-------
135. American Petroleum Institute. Crude Oil as a Burner Fuel:
Petroleum Safety Data, Washington, D. C., July 1974.
136. Fort Worth Regional Office (FWRO) Report of Principal Activities
to the Federal Power Commission, November 1971.
137. Weller, A. E. Report on Safety Aspects of Advanced Energy Con-
version Systems. Battelle Columbus Laboratories, Columbus, Ohio,
March 1, 1976. 14 pp.
138. Federal Power Commission Reports, Washington, D. C., 1975
(4 quarters).
139. Bertolett, A. D. , and R. J. Fox. Accident-Rate Sample Favors
Nuclear Electric World. July 15, 1974. pp. 40-41.
140. New York State Public Service Commission Records.
141. Dupree, Walter G., Jr., and James A. West. United States Energy
Through the Year 2000. Department of the Interior, Government
Printing Office, Wahsington, D. C., 1972. pp. 30, in Ref. 2,
pp. i3-22.
142. Federal Railroad Administration. Summary and Analysis of Ac-
cidents on Railroads in the United States. Accident Bulletin
No. 143. U. S. Department of Transportation, Federal Railroad
Administration, Office of Safety, 1974. 91 pp.
143. National Transportation Safety Board. Annual Report to Congress,
1975. National Transportation Safety Board. Washington, D.C.,
1975. 50 pp.
144. National Transportation Safety Board. Annual Report to Congress,
1974. National Transportation Safety Board. Washington, D.C.,
1974. 91 pp.
145. Defense Civil Preparedness Agency. Disaster Planning Guide for
Business and Industry. May 1974. 54 pp.
146. Stephens, Maynard M. Minimizing Damage to Refineries from
Nuclear Attack, Natural and Other Disasters. Prepared for the
Office of Civil Defense, February 1970. 241 pp.
147. Stephens, Maynard M. Vulnerability of Natural Gas Systems. Pre-
pared for the Defense Civil Preparedness Agency, June 1974. 108 pp.
148. Personal communication with Dr. Miles Lawrence, the National Hur-
ricane Center, National Weather Service, Coral Gables, Florida.
275
-------
149. Strahler, Arthur N., and Straler, Alan H. Environmental Geo-
Science: Interaction Between Natural Systems and Man. Hamilton
Publishing Company, California, 1973. 411 pp. plus Appendices.
150. Bolt, B. A., W. L. Horn, G. A. Macdonald, and R. F. Scott.
Geological Hazards. Springer-Verlag, New York, Heidelberg,
Berlin, 1975. 328 pp.
151. Citron, Robert. Smithsonian Institution. Notes on the Develop-
ment of a United Nations Natural Disaster Program. Prepared for
the Office of Science and Technology of the United Nations, March
1971, 26 pp.
152. Ayyasenamy, P., B. Hauss, T. Hsiih, A. Mascata, T. E. Hicks, and
D. Ohrint. Estimates of the Risks Associated with Dam Failure.
Prepared for the United States Atomic Energy Commission. Univer-
sity of California at Los Angeles, 1974, 375 pp.
153. Brennan, C. D., etc. al. The Threat to Licensed Nuclear Facili-
ties. The MITRE Corporation. September 1975. 171 pp.
276
-------
TECHNICAL REPORT DATA
(Please read Instructions on the reverse before completing)
EPA-600/7-77-016
3. RECIPIENT'S ACCESSION>NO.
Accidents and Unscheduled Events
Associated with Non-Nuclear Energy Resources
and Technology
5. REPORT DATE
February 1977
6. PERFORMING ORGANIZATION CODE
AUTHORISIC. Bliss, P. Clifford, G. Goldgraben, E. Graf-
Webster, K. Krickenberger, H. Mahar, N. Zimmerman
8. PERFORMING ORGANIZATION REPORT NO.
PERFORMING ORGANIZATION NAME AND ADDRESS
The MITRE Corporation, METREK Division
1820 Dolley Madison Boulevard
McLean, Virginia 22101
10. PROGRAM ELEMENT NO.
1NE 626
11. CONTRACT/GRANT NO.
68-01-3188
2. SPONSORING AGENCY NAME AND ADDRESS
Office of Energy, Minerals and Industry
Office of Research and Development
U.S. Environmental Protection Agency
Washington, D.C. 20460
13. TYPE OF REPORT AND PERIOD COVERED
"Final Report
yx
14. SPONSORING AGENCY CODE
EPA/600/17
15. SUPPLEMENTARY NOTES
16. ABSTRACT
Accidents and unscheduled events associated with non-nuclear energy
resources and technology are identified for each step in the energy cycle.
Both natural and anthropogenic causes of accidents or unscheduled events
are considered. Data concerning these accidents are summarized. Estimates
of frequency and severity are presented for all accidents. The energy
systems discussed are coal, oil, natural gas, LNG, hydroelectric, geothermal,
and oil shale.
KEY WORDS AND DOCUMENT ANALYSIS
=SCRIPTORS
b.lDENTIFIERS/OPEN ENDED TERMS
Accidents, Non-Nuclear energy,
Hazards
18. DISTRIBUTION STATEMENT
Normal distribution
Coal, Natural gas, Oil,
Geothermal. Solar, Oil
shale, Sabotage, Earth-
quake, Tsunamis,
Accidents.
19. SECURITY CLASS (ThisReport)
Unclassified
20. SECURITY CLASS (Thispage)
Unclassified
F-10
22. PRICE
EPA Form 2220-1 (9-73)
------- |