EPA-600/2-82-040
-dtrmKHay--1982 /yl Qitc /y
PETROLEUM PIPELINE LEAK DETECTION STUDY
by
John R. Mastandrea
Science Applications, Inc.
101 Continental Blvd.
El Segundo, CA 90245
Contract No. 68-03-2532
Project Officer
J. S. Dorrler and R. A. Griffiths
Solid and Hazardous Waste Research Division
Oil and Hazardous Materials Spills Branch
Municipal Environmental Research Laboratory-Cincinnati
Edison, New Jersey 08837
MUNICIPAL ENVIRONMENTAL RESEARCH LABORATORY
OFFICE OF RESEARCH AND DEVELOPMENT
U.S. ENVIRONMENTAL PROTECTION AGENCY
CINCINNATI, OHIO 45268
-------
DISCLAIMER
This report has been reviewed by the Municipal Environmental Research
Laboratory, U.S. Environmental Protection Agency, and approved for
publication. Approval does not signify that the contents necessarily
reflect the views and policies of the U.S. Environmental Protection
Agency, nor does mention of trade names or commercial products constitute
endorsement or recommendation for use.
-------
FOREWORD
The U.S. Environmental Protection Agency was created because of
increasing public and government concern about the dangers of pollution
to the health and welfare of the American people. Noxious air, foul
water, and spoiled land are tragic testimonies to the deterioration of
our natural environment. The complexity of that environment and the
interplay of its components require a concentrated and integrated attack
on the problem.
Research and development is that necessary first step in problem
solution; it involves defining the problem, measuring its impact, and
searching for solutions. The Municipal Environmental Research
Laboratory develops new and improved technology and systems to prevent,
treat, and manage wastewater and solid and hazardous waste pollutant
discharges from supplies, and to minimize the adverse ecomonic, social,
health, and aesthetic effects of pollution. This publication is one of
the products of that research and provides a most vital communications
link between the reseacher and the user community.
This study investigates spill prevention programs for petroleum
pipeline systems. The results of the study can be utilized both in
assessing the need and in the actual development of a spill prevention
and control program for individual pipelines and those nationwide. The
report should be of particular value to research users involved in the
prevention of accidental spills from petroleum pipleline systems. For
further information on the subject contact the author directly or the
Applied Mechanics Division of Science Applications, Inc.
Francis T. Mayo, Director
Municipal Environmental Research
Laboratory
111
-------
ABSTRACT
In this study, oil spills from petroleum pipeline systems are investi-
gated and spill prevention programs presented. The study includes an ex-
tensive review and analysis of the U.S. petroleum pipeline transportation
system, accidental spills nationwide and the available means of effectively
preventing spills.
A current need for improved spill prevention measures for petroleum
pipelines is identified. This situation exists because of the potential of
serious spills. Furthermore, since existing petroleum pipeline systems are
aging, population densities near many new lines are increasing dramatically,
and new lines are expected to be larger and longer, the need for such spill
prevention programs is expected to increase.
Spill prevention programs developed in this report involved the follow-
ing two tasks: (1) proper assessment of the risk; and (2) selection of op-
timum spill prevention measures to adequately minimize the risks. Prior to
this study, both tasks could not be achieved for most petroleum pipeline
systems because of two main problems. First, the contribution of all the
significant factors, i.e., pipeline age, dimension, etc., to the overall
oil spill risk previously could not be assessed primarily because of limita-
tions in reported pipeline leak data and lack of a comprehensive analysis of
the nationwide spill problem. The complexity of the petroleum pipeline sys-
tem in the United States, i.e., system consists of many individual lines of -
various dimensions, age, controls, etc., further complicated this assessment
because there is limited information on the variations between individual
pipeline systems. Secondly, an assessment of all viable spill prevention
measures for reducing the risk (frequency, volume, and the combined frequency
and volume of spills) was not previously available. These tasks and problems
are addressed successfully in this study.
An approach for developing a spill prevention program is presented.
Then recommended spill prevention programs for various types of individual
petroleum pipeline systems and those nationwide are described. Using cer-
tain programs, spill reductions of 75 percent or more are expected. These
programs consist of scheduled inspection and/or leak detection methods used
alone or in various combinations, and are shown to be capable of achieving
significant levels of reduction in the risk of serious spills. Practical
considerations such as the cost of inspections and spillage are also in-
cluded. Means are provided so that individual programs can be readily de-
veloped by operators for their specific individual lines.
IV
-------
CONTENTS
Foreword . i-ji
Abstract iv
Figures vii
Tables xi
Abbreviations and Symbols xvii
Metric Conversion Factors xix
Acknowledgment xx
1. Introduction 1
2. Conclusions 4
3. Recommendations 5
4. Petroleum Transport - Systems, Measurement, Changes and Losses . 9
4.1 Petroleum transportation system description 11
4.2 Petroleum system - losses and specific loss mechanisms ... 39
4.3 Measurement 48
4.4 Changes in petroleum between receipt and delivery 64
4.5 Total measurement system error 65
4.6 Loss statistics 66
4.7 Total petroleum losses—pipeline systems 145
5. Pipeline System Leak Detection and Inspection Methods 146
5.1 General discussion 146
5.2 General description of methods 150
5.3 Compendium of leak detection and inspection methods .... 155
5.4 Status 195
6. Analysis of the Risk of Accidental Oil Spills from Petroleum
Pipeline Systems 223
6.1 Seriousness of oil spills 223
6.2 Problems in assessing the oil spill risk 225
6.3 Analysis and assessment of the risk of accidental spills . . 231
7. Analysis of the Reduction of Oil Spills from Line Pipe by
Utilization of Leak Detection and Inspection Methods 271
7.1 Comparison and evaluation of leak detection and inspection
methods 272
7.2 Oil spillage that potentially can be prevented using the
selected leak detection and inspection methods 297
7.3 Cost analysis 309
7.4 Cost-effectiveness analysis 322
8. Recommendations 327
8.1 Discussion 327
8.2 Benefits 329
8.3 Recommended preventive maintenance program 329
8,4 Recommended developmental and new inspection and leak
detection methods 335
-------
References 337
Appendices
A. Existing U.S. and foreign regulations for transportation
of liquids by pipeline 344
B. Pipeline mileage data 380
C. Summary of liquid pipeline accident reports on DOT form
1000-1 compiled by the Office of Pipeline Safety for
years 1969-1976 398
D. A statistical estimate of pipeline leakage 416
E. Methodology 440
VI
-------
FIGURES
Number Page
1 Petroleum transportation system 12
2 Petroleum transportation model" 13
3 Flowsheet of petroleum production operations 14
4 A pipeline system composed of gathering, trunk and dis-
tribution lines 21
5 Growth of crude oil pipelines in the United States ... 24
6 Mileage of petroleum pipelines in the United States ... 25
7 Illustrative operating conditions: Trans-Alaska Pipeline 32
8 Standard Federal regions (Commerce, EPA and HUD), with
number of liquid pipeline accidents 37
9 Petroleum administration for Defense and Bureau of Mines
refining districts - 38
10 Major system components for crude losses between supply
and refinery input 40
11 Fault tree A]_, oil spilled from petroleum pipeline system 43
12 Fault tree Cj, spill from pipelines 44
13 Curves for estimating volume change when blending volatile
hydrocarbons with crude oil 47
14 Typical system for delivery to pipeline 50
15 Typical PD meter LACT unit flow diagram 51
16 PD meters 52
17 Typical accuracy curve for a PD meter 58
18 Error curves and calculations for inaccurate correction
factors of the coefficient of expansion a for liquids 62
-------
19 Minimum detectable spill loss for a simple pipeline
system 67
20 Spill incidents and volume versus year for petroleum pipe-
line system accidents reported to OPSO between 1968-
1976 77
21 Number of incidents versus year of occurrence for carrier
system components 80
22 Volume of spill versus year of occurrence for carrier
system components 81
23 Number of spills versus year of occurrence by cause ... 84
24 Volume of spills versus year of occurrence by cause ... 85
25 Percentage of number and volume of spills by cause for
line pipe 88
26 Number of incidents versus year of occurrence for type of
fluid transported in pipeline system accidents reported
to OPSO between 1968-1976 91
27 Volume of spills versus year of occurrence for type of
fluid transported through pipeline system accidents
reported to OPSO between 1968-1976 92
28 Number of incidents versus year of occurrence for type of
fluid transported in line pipe, accidents reported to
OPSO between 1971-1975 94
29 Spill incidents and spills per 1000 miles versus year for
pipeline system accidents reported to OPSO between
1968-1976 97
30 Spill volume for mileage and throughput versus year for
pipeline system accidents reported to OPSO between
1968-1976 98
31 Spill incidents per 1000 miles and spill volume per mile
versus year for line pipe transporting liquids in the
U.S. reported to OPSO between 1971 and 1975 101
32 Size distribution of spills from U.S. terrestrial pipe-
lines transporting liquids 104
33 Mean spill size and line pipe diameter for line pipe acci-
dents reported to OPSO, 1971-1975 107
VTM
-------
34 Mean spill versus line pipe diameter for line pipe acci-
dents in the U.S. reported to OPSO between 1971-
1975 108
35 Frequency of spills for pipeline diameters and mileage
for line pipe accidents in the U.S. Reported to OPSO
between 1971-1975 ...... . . 109
36 Spill volume for petroleum pipeline diameters of line
pipe accidents reported in the U.S. to OPSO between
1971-1975 110
37 Diameter of line pipe and causes of spill incidents for
accidents reported to OPSO in 1971 114
38 Spill incidents by year of pipeline system installation
for accidents reported to OPSO between 1971-1975 .... 117
39 Pipeline system spill incidents per 1000 miles of pipe-
line by the year of pipeline system installation for
accidents reported to OPSO between 1971-1975 118
40 Line pipe spill incidents per 1000 miles of pipeline by
the year of pipeline system installation for accidents
reported to OPSO between 1971-1975 121
41 Age of line pipe and causes of spill incidents per 1000
miles of line pipe for accidents reported to OPSO in
1971 and 1974 122
42 Damage from petroleum pipeline system incidents reported
to OPSO between 1968-1978 for crude and product .... 127
43 Number and volume of 10 to 50-barrel spills by year, 1971-
75, Gulf of Mexico outer continental shelf 129
44 Discharge trends for oil only for calendar years 1970-1975 130
45 Pipe and petroleum pipeline system oil spills in the U.S.
for gathering/distribution and production systems in
the United States between 1970-1971 140
46 Cross-country pipeline mean spill size versus pipe diameter 143
47 Cumulative size distribution for oil industry cross-
country pipeline spills, Western Europe and Canada
(assorted pipe diameters) 144
ix
-------
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
Transportation accidents in 1975, National Transporta-
tion Safety Board 1975
Fatalities and injuries versus amount of product loss
in liquid pipeline systems accidents
Symbolism for fault trees
Fault trees, VST anc' ^ST* spill frequency and volume from
typical section of line (reference line)
Relative measure of risk of oil spilled from typical sec-
tion of line (reference line}— leaks, ruptures, mean . .
Fault tree, DE$T» spill damage external to typical section
of line pipe (reference line)
Correction factors for frequency of spills— age, geometry,
length, use, size
Fault tree, CFp, correction factors for frequency of
spills
Fault tree, CF$V» correction factor for the volume of
spills
Correction factors for volume of spills— diameter, length,
use, shutdown time
Fault tree, CFvR or CFyM» correction factors for spill
size — rupture and mean
Fault tree, CF£p,, correction factors for risk of spill
damage external to line pipe
Fault trees, VSTC and F$TC» frequency and mean volume of
spills from line pipe
Fault trees, VNLLC> VNLRC> VMC> nominal largest size of
ruptures, leaks and mean for line pipe spills
Fault trees, RMRR, RMR[., relative measure of risks (barrels/
year) of ruptures and leaks from line pipe
Fault tree, RMRw, relative measure of risk (barrels/year)
of mean spills from line pipe
226
228
235
236
241
243
245
246
252
253
257
259
262
264
266
267
-------
TABLES
Number
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
Percentage of Petroleum Moved by Transportation Modes . . .
Percentage of Crude Petroleum and Petroleum Products
Moved by Method of Transportation, 1938-1974
Volume of Oil Produced and Imported, 1972-1977
Offshore Production of Crude Petroleum
Volume of Petroleum Transported by Pipelines in the
United States
Volume and Type of Oil Transported by Interstate Pipe-
line Systems
Pipeline Transportation System and Associated Subsystems .
Pipeline Mileage in the United States - Total and Inter-
state
Mileage of Petroleum Pipelines by Diameter in the United
States, January 1, 1971 and January 1, 1974
Mileage of Petroleum Pipelines by Diameter in the United
States, January 1, 1971 and January 1, 1968
Relationship between Pipe Diameter, Length, and Volume
Contained Inside
Petroleum Pipeline System Mileage in the U.S., by
Year Installed
Analysis of Crude Products Reported to NPC
Analysis of Clean Products Reported to NPC
Pipel-ine Fill for Petroleum Pipelines in the United States
January 1, 1977 and January 1, 1974
Page
11
16
17
18
19
20
22
26
27
28
29
31
33
34
36
xi
-------
16 Throughput and Losses from ICC Transport Statistics .... 69
17 Petroleum Loss Allowa-nce from Typical U.S. Pipeline Companies
in Percent of Gross Petroleum Shipped 71
18 Spill Statistics for Pipeline Systems Transporting Liquids
in the U.S. for Incidents Reported to OPSO Between
1968-1976 78
19 Number and Volume of Spills for Failures of Petroleum Pipe-
line System Components (In Percent of Total) 79
20 Number and Volume of All Spillage Incidents (In Percent)
for All Pipeline System Components, By Cause Reported
to OPSO Between 1971 and 1975 83
21 Number, Volume and Percentage of All Spillage Incidents
for Line Pipe, By Cause Reported to OPSO Between 1971
and 1975 87
22 Spill Volume and Incidents for Accidents Reported to OPSO
Between 1968 and 1976 for Crude and Produce Pipeline
Systems 90
23 Statistical Estimate of Annual Distribution of Petroleum
Spill Frequencies and Barrels Lost from Line Pipe for
Crude and Product 95
24 Spill Data Summary for Petroleum Pipeline Systems Transport-
ing Liquids in the U.S. for Accidents Reported to OPSO
Between 1968-1976 96
25 Spill Data Summary for Line Pipes Transporting Liquids in
the U.S. for Accidents Reported to OPSO Between 1971-
1975 100
26 Annual Spill Statistics for Line Pipes Transporting
Liquids in the U.S. for All Spill Sizes 100
27 Spill Incidents of Pipeline System for Two Pipeline Diam-
eter Categories for Accidents Reported to OPSO Between
1971-1975 103
28 Summary of Spill Data for Line Pipe Diameters for Liquid
Pipeline Accidents Reported to OPSO Between 1971-1975 . 106
29 Outside Force Incidents on Line Pipe Reported in the U.S.
to OPSO Between 1971-1974 Ill
-------
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
Fault Incidents from Line Pipe Reported in the U.S. to
OPSO Between 1971-1974
Spill Incidents for Pipeline System Age and Estimated
Mileage for Accidents Reported to OPSO Between 1971-1975
Spill Incidents and Volume for Line Pipe Age for Accidents
Reported to OPSO Between 1971 and 1979 .........
Number of Outside Force Incidents by Depth of Cover ....
Fatalities and Injuries for Petroleum System Incidents Re-
ported to OPSO Between 1968 and 1976
Damage from Petroleum Pipeline Systems in the U.S. for In-
cidents Reported to OPSO Between 1968-1976
Number and Volume of Spills Each Year, 1971-1975, Gulf of
Mexico Outer Continental Shelf
General Areas and Locations of Spillage in and About U.S.
Waters, Reported to USCG in 1975
Causes of Spills in and About U.S. Waters, Reported to
USCG in 1975
Sources of Spills in and About U.S. Waters, Reported to
USCG in 1975 ....
Component Versus Quantity Spill Category for Gathering/
Distribution System, 1970-1971
Component Versus Quantity Spill Category for Drilling
System, 1970-1971
Component Versus Quantity Spill Category for Production
System and All Systems, 1970-1971
Source Data Summary
Cross-Country Spill Data for Western Europe
Summary of Type and Operational Mode of Leak Detection
and Inspection Methods
Visual and Aided Visual Line Observations
Oil Spill Detectors on or Near the Water
112
116
119
123
124
126
128
132
133
134
135
136
137
138
141
159
161
163
XTM
-------
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
Internal Fluid Variations During Transfer
Detection and Location of Leaked Oil on or at a Short Dis-
tance from Pipeline
Periodic Pressure Tests
Corrosion Inspection
Standard Nondestructive Inspection
Pipeline Inspection Pigs
Survey-Pipeline System Charting and Depth of Burial ....
Miscellaneous
Assumptions for Typical Section of Line Pipe (Reference
Line)
Relative Risk of Oil Spills from a Typical Section of
Line Pipe
Correction Factors for Risk of Spill Damage External to
Line Pipe
Calculation of the Frequency FSTC and Volume V$TC of
Spills for Line Pipe
Calculations of the Nominal Largest Size of Ruptures,
VNLRC» LeakSNLLC and Mean VMC f°r Ll'ne piPe Spills . . .
Calculations of the Relative Measure of Oil Spill Risks
for Rupture RMRR, Leaks RMR|_ and Mean Size RMRw Spills
for Line Pipe
Calculations of the Correction Factor CFED for Risks
External to Line Pipe
Calculations of Risks External to Line Pipe
Inspection and Leak Detection Methods Selected for Further
Evaluation of Line Pipe
Comparison of the Selected Methods for Detection and Loca-
tion of Reference Line Leaks and Ruptures
Spill Reduction Analysis , for Volume of Ruptures for Typi-
cal Section of Line Pipe
165
168
171
174
175
177
180
182
234
240
258
261
263
265
268
269
274
277
_
280
xiv
-------
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
Spill Reduction Analysis for the Volume of Leaks for a
Typical Section of Line Pipe
Reduction Analysis for the Frequency of Oil Spills by
Cause for a Typical Section of Line Pipe (Reference
Line)
Spill Reduction Analysis for the Frequency of Leaks for
A Typical Section of Line Pipe . .
Spill Reduction Analysis for the Frequency of Ruptures
for a Typical Section of Line Pipe
Summary of Risk Reduction Factors for Selected Inspec-
tion and Leak Methods
Analysis of Oil Spilled from the Reference Line that Can
be Prevented or Go Undetected
Analysis of Oil Spillage that Can be Prevented or Go Un-
detected for All U.S. Lines
Assumptions for Reference Line
Equipment and Inspection Service Costs for Selected Leak
Detection and Inspection Methods
Cost of Inspection and Leak Detection Methods
Offshore Pipeline Construction Costs
Spill Cleanup Cost Factors as a Function of the Location
of Spill
Cost-Effectiveness Analysis for All U.S. Lines
Preventive Maintenance Schedules for the Reduction of the
Frequency of Accidental Spills from Line Pipe
Preventive Maintenance Schedules for the Reduction of the
Volume of Accidental Spills from Line Pipe
Preventive Maintenance Schedules for the Reduction of the
Frequency and Volume of Accidental Spills from Line
Pipe
284
290
294
296
298
304
307
311
312
313
316
321
323
331
332
333
XV
-------
ABBREVIATIONS AND SYMBOLS
atm — atmosphere
bbl — barrel
bbld — barrels per day
bblh — barrels per hour
bblm — barrels per minute
°C — degrees Centigrade
cm — centimeter
dwt -- dead weight ton
°F — degrees Fahrenheit
ft — foot
g — gram
gal — gallon
gpd — gallons per day
gph — gallons per hour
gpm — gallons per minute
hr — hour
in ~ inch
IR — infrared
kg — kilogram
km — kilometers
1 — liter
Ib -- pound
LN6 — liquified natural gas
LPG — liquified petroleum gas
m — meter
mg — milligram
mi — miles
min ~ minute
ml — mi Hi liter
mm — millimeter
02 — ounce
psi -- pounds per square inch
UV — ultraviolet
vol — volume
wt — weight
yr — year
SYMBOLS
percent
per
xvi
-------
AGA — American Gas Association
API ~ American Petroleum Institute
BLM -- Bureau of Land Management
BOM -- Bureau of Mines
BS&W — Bottom Sediment and Water Content
CEQ -- Council on Environmental Quality
DnV' — Det Norske Veritas
DOA — Department of the Army
DOE — Department of Energy
DOT -- Department of Transportation
DWP — Deepwater Port
EMAT — Electromagnetic Non-Contact Transducer
EPA — Environmental Protection Agency
ERDA — Energy Research and Development Administration
FOA -- Food and Agriculture Organization
FEA — Federal Energy Administration
GAO — General Accounting Office
HUD -- Housing and Urban Development
ICC -- Interstate Commerce Council
IMCO -- International Maritime Consultative Organization
IP — Institute of Petroleum, Great Britain
LACT — Lease Automatic Custody Transfer Systems
LPG — Liquid Petroleum Gas
NBS — National Bureau of Standards
NDI — Nondestructive Inspection
NOAA -- National Oceanic and Atmospheric Administration
NPC — National Petroleum Council
NTIS — National Technical Information Service
OCS — Outer Continental Shelf
OPSO ~ Office of Pipeline Safety and Operations
PAD — Petroleum Administration for Defense
PD — Positive Displacement Meters
PIRS — Pollution Incident Reporting System
PVTT -- Pressure-Volume-Temperature-Time
SAI -- Science Applications, Inc.
SPCC -- Spill Prevention, Control and Counter-measures
TM -- Turbine Meters
US — United States
USCG — United States Coast Guard
USDA -- United States Department of Agriculture
USDI -- United States Department of Interior
USGS — United States Geological Survey
USN — United States Navy
xvn
-------
METRIC CONVERSION FACTORS
Approximate Conversions to Metric Measures
Symbol
in
ft
yd
mi
in2
ft2
yd2
n.12
oz
Ib
fl oz
qt
gal
ft3
°F
When You Know
inches
feet
yards
miles
square inches
square feet
square yards
square miles
ounces
pounds
short tons
(2000 Ib)
fluid ounces
quarts
gallons
cubic feet
Fahrenheit
temperature
Multiply by
Length
2.5
30
0.9
1.6
Area
6.5
0.09
0.8
2.6
Mass (weight)
28
0.45
0.9
Volume
30
0.95
3.8
0.03
Temperature (exact)
5/9 (after
subtracting
32) -
To Find
centimeters
centimeters
meters
kilometers
square centimeters
square meters
square meters
square kilometers
grams
kilograms
tonnes
milliliters
liters
1 i ters
cubic meters
Celsius
temperature
Symbol
cm
cm
m
km
cm2
m2
m2
km2
g
kg
t
ml
1
1
m3
°C
xvm
-------
ACKNOWLEDGEMENTS
The objective of this project could not have been accomplished without
the assistance of numerous individuals in the oil and gas industries, Fed-
eral and State Government, trade associations, engineering companies, uni-
versities and others. These individuals, dedicated to doing their best to
insure both safety and pollution prevention in petroleum pipeline systems,
graciously responded to inquiries, personnel interviews, provided facility
visits and supplied information requested. Science Applications, Inc. wishes
to convey its sincere appreciation for their efforts.
In particular, Science Applications, Inc. wishes to acknowledge the
support and contributions of the following individuals and organizations:
American Petroleum Institutes
American Gas Association
Atlantic Richfield Oil Co.
Chevron Oil Company
Det norske Veritas
Exxon Research and Engineering Co.
Gulf Oil
Getty Oil Company
International Research and Technology Corp.
Mobil Oil Company
National Transportation Safety Board
Oil Companies' International Study Group (CONCAWE)
Pipeline Digest
Shell Pipe Line Corporation
U.S. Department of Energy - Bureau of Mines
U.S. Department of Interior - Bureau of Mines
U.S. Department of the Interior - Geological Survey
U.S. Department of Transportation - Office of Pipeline Safety
and Operations
U.S. Department of Transportation - Office of Deepwater Ports
U.S. Department of Transportation - Transportation Systems Center.
xix
-------
Science Applications, Inc. would like to give special thanks to Mr.
J. S. Dorrler, Project Officer of the U.S. Environemntal Protection Agency,
for his guidance and valuable discussions and comments on the entire pro-
ject and to Mr. R. A. Griffiths, who succeeded Mr. Dorrler in the latter
phases of this project, for his help and support.
xx
-------
SECTION 1
INTRODUCTION
Transportation of petroleum through pipeline systems in the United
States is relatively safe. However, with the many individual systems of
diverse dimension, age, material, etc., a real and continuous concern
exists over accidents and leakage1. Petroleum accidents and leakage
cause many pollution incidents that are detrimental to both national in-
terest and the oil industry. These accidents result in the loss of large
quantities of petroleum, cause significant environmental problems, and
subject segments of the population to potential hazards. Even small losses
from a pipeline system cannot be tolerated, and it is the responsibility of
government and operating companies to protect the community from possible
damage resulting from a leak.
OBJECTIVES
This study had the overall objective of providing a viable, cost-
effective means of reducing oil spill incidents and volume and thereby
minimizing environmental impacts. Specific objectives included: (1)
estimate total petroleum losses and specific loss mechanisms; (2) accu-
rately appraise oil spill incidents including causes, system component
failures, frequency and volume of spills, causes and key factors relating
to spills; (3) define capabilities of leak detection and/or inspection
techniques both to detect leaks and determine the condition of the line;
(4) estimate oil spillage that potentially can go undetected; and (5) if
feasible and practical, develop or recommend advanced techniques for re-
ducing the problem.
The envisioned result of this study is the development of an effec-
tive spill prevention program for individual pipeline systems of diverse
age, dimensions, controls, etc. The program would be based on scheduled
inspections and/or leak detection to test the condition of pipelines and/
or to check for the presence of leaks. The envisioned application is to
aid operating companies both in evaluating the need and in the actual de-
velopment of a spill prevention and control program for individual lines.
In addition, such information might aid various local and Federal agencies
in evaluating problem areas and available means of minimizing the problem.
-------
SUMMARY
To obtain an overall perspective of the problem of accidental spills
from petroleum transportation systems, the extent and nature of petroleum
losses from the petroleum transportation systems were studied (Section 4).
Estimates were made of the total quantity of petroleum lost and the amount
attributable to each major loss mechanism. Emphasis was directed to the
problem of spills from the transmission line (line pipe). This section
also provides basic information that was used in the detailed analysis of
the oil spill risks (see Section 6) and the capabilities or methods to re-
duce the risk (see Section 7).
Leak detection and inspection methods for petroleum pipeline systems
that might reduce the risk of oil spillage (i.e., reduce frequency of
spills or limit the volume of oil lost) are presented in Section 5. Be-
cause of the high incidence of pipeline spills and the potential for sig-
nificant reductions of the oil spill risks, major emphasis was given to
detection and prevention of leaks from line pipe failures. However,
methods were also included for other pipeline system components. Methods
for prevention of leakage by detection of impending failures and minimizing
of leakage volume through early detection of leakage were emphasized.
Methods included in existing domestic and foreign regulations and current
recommended practices for inspection and leak detection of liquid and gas
pipeline systems were reviewed.
An analysis of the risk of accidental oil spills from petroleum pipe-
line systems is presented in Section 6. The seriousness of oil spills and
difficulties in assessing the risk of oil spills were reviewed. Approaches
accounting for these difficulties are presented and an analysis and
assessment of the risk of oil spills is given. Both the risk of oil spill
from a system and environmental problems created by spillage were examined.
An in-depth analysis of the risk of oil spills from line pipe was earned
out. An analysis was performed on a reference line (typical line), and
correction factors for variations from the reference line were developed
so that the spill risk (potential) could be estimated for most lines. Sim-
plified tables and figures are provided so that an operator can estimate
the spill potential of his own line.
The potential of selected leak detection and inspection methods9 to
reduce the risk of oil spills (frequency and volume of spills and risks
external to the line) from petroleum pipeline systems are estimated in
Section 7. Means are provided for an operator to estimate the risk of oil
spills and the risk reduction capabilities (effectiveness) of inspection
and leak detection methods for his own line. Estimates were made of the
capabilities of leak detection and inspection equipment to detect a leak
of a certain magnitude and/or impending failures. The methods were eval-
uated based on their capability to reduce the frequency and volume of
leak-type and rupture-type spills for the reference line. Factors were
a Selected methods were those judged to be capable of achieving a signifi-
cant reduction in the spill risk.
-------
established that provide a simple means of comparing the capability of
methods to reduce the frequency and volume of spills. These factors aid
a potential user in selection of the most suitable method that will satis-
fy the spill prevention program needs of a particular line. Since costs
are a major consideration in a practical maintenance program for line pipe,
costs of the various options and spillage are estimated and a cost-effec-
tiveness analysis is presented.
Preventive maintenance programs for line pipe are presented and par-
ticular programs are recommended in Section 8. Benefits of these programs
are discussed. Preventive maintenance schedules for achieving specified
reductions in the frequency of spills, the volume of spills and the com-
bination of frequency and volume of spills are identified. Recommendations
were based primarily on qualitative estimates of both the oil spill risks
that exist and the capabilities of various methods to reduce the risks.
However, the recommendations were also based on other practical considera-
tions such as cost of inspections and spillage.
-------
SECTION 2
CONCLUSIONS
This study indicates that the potential exists for serious spills
from the line pipe of petroleum pipeline systems and that various options,
inspection and leak detection methods, are available to significantly re-
duce the oil spill risk. Based on the results of this investigation, pre-
ventive maintenance programs for line pipe are recommended.
The component of a pipeline system most responsible for spill inci-
dents for the largest loss of petroleum, and for the greatest risk to the
environment is the operational transmission line. As pipeline systems age,
an even higher incidence of pipeline failures is anticipated. Once a line
becomes operational, however, no guidelines and/or schedules for effective
inspection exist for U.S. pipeline systems. Furthermore, state-of-the-art
leak detection methods are not commonly used. Even if used, most methods
are limited in sensitivity, and significant quantities of fluid may be lost
before shutdown. The problem becomes particularly critical for certain
pipelines such as those of large diameter (36- to 54-in), underwater lines,
old lines, etc. In the case of large-diameter lines, high volumes of oil
can be spilled before conventional leak detection systems can detect the
leak and initiate shutdown.
Although no methods or procedures exist that can guarantee complete
protection against a pipeline leak, it is possible that the frequency and
severity of spill incidents can be minimized by effective and frequent in-
spection of the operational line. Certain identified inspection and/or
leak detection methods used alone or in various combinations may be effec-
tive in minimizing the risk of a serious spill from individual lines. In
many instances, the various available options may be practical, beneficial,
and cost-effective. Furthermore, preventive maintenance programs can be
implemented (at significant costs) to reduce the possibility of serious
spills and improve the safety record nationwide (see Section 8.3.3.1).
Means are available (see Section 8) for estimating the reduction of
the frequency and volume of spills using preventive maintenance programs.
Such information allows one to estimate future oil spill trends for a par-
ticular pipeline or nationwide.
-------
SECTION 3
RECOMMENDATIONS
Preventive maintenance programs for preventing spills from the line
pipe of operational lines are recommended. These programs, along with new
and-developmental leak detection and inspection methods, should continue to
be investigated. Recommended preventive maintenance programs presented in
Section 8 are primarily intended to aid operating companies in evaluating
the need and in the actual development of a spill prevention and control
program for their own operational line. In addition, this information can
aid various local and Federal agencies in evaluating problem areas and
available means of minimizing the problem.
Since there are numerous variations between lines and wide ranges in
the oil spill risk, it is beyond the scope of this study to recommend pro-
grams for particular lines. Hence, recommendations are generalized and do
not include specific details for implementing methods for particular
lines. The preventive maintenance programs include scheduled inspections
and/or in situ leak detection (some functioning continuously) that could be
effectively developed and implemented. The programs, if implemented, are
expected to improve significantly the prevention and control of accidental
spills of petroleum from operational lines. Nationwide implementation of
such programs is expected to decrease, substantially, the total quantity
of petroleum lost, the number of spill incidents and the number of major
spill incidents. Estimates of these reductions are presented in Section 8.
The recommended approach for development of a spill preventing pro-
gram for an individual or group of lines was based on the following steps:
• Estimate the risk that oil will escape from a line (see Sec-
tion 6.3.4.1).
• Identify locations of a line where a spill may create
special, serious problems (see Sections 6.3.3.4 and 6.3.4.2)
such as the possible damage to the external environment or
cost of spill cleanup.
• Estimate the level of corrective action to reduce the spill
risk to an acceptable level.
• Identify the various options of spill prevention that are
available to achieve the necessary reductions in the spill
risk (see Sections 8.3.2 and 7.1.2).
-------
• Select the optimum methods based on the cost effectiveness
and other considerations of the various suitable options.
A recommended general schedule for the reduction of the frequency and
volume of spills is provided in Section 8.3.2. The general schedule pro-
duces confidence that an individual line or group of lines will have ho
more than (X) barrels spilled and (X) number of incidents per year. The
recommended approach is to first estimate the risk that oil will escape
from the line (see Section 6.3.4.1). Then select one of the following spe-
cific preventive maintenance schedules:
• Reduction of frequency of spills - Table 80
• Reduction of volume of spills - Table 81
• Reduction of frequency and volume of spills - Tabl-e 82.
Using the approach and indicated schedule, selected spillage reductions of
greater than 25 percent, 50 percent and 75 percent may be achieved.
Preventive maintenance programs with specific schedules and applica-
tions were recommended as follows:
(1) Most U.S. Lines:
• Visual inspections of the line by air or ground patrol-
weekly
• One-call system (Reference 12)
• Survey by inspection pig (magnetic flux type or other
comparable device) every 4 years
or
• Hydrostatic tests
- yearly or
- after indication of spill.
(2) Underwater lines:
• Visual inspections of the line by air or ground patrol —
weekly
• Depth of cover inspection by sonar or other comparable
device—every 2 years
• Hydrocarbon probe (towfish or similar device) inspection-
yearly
-------
• Survey by inspection pig (magnetic flux type or other
comparable device)--every 2 years
or
• Hydrostatic test—yearly.
(See also recommended developmental and new methods.)
(3) Large-diameter lines:
• Visual inspection by air or ground patrol--weekly
• One-call system (reference 12)
• Survey by inspection pig (magnetic flux type or other
comparable device)--every 4 years
• Volume comparison, flow rate comparison, and pressure
deviation
or
• Mathematical modeling.
(See also recommended developmental and new methods, Section 8.4.)
(4) Old lines:
• Visual inspections by air or ground patrol—weekly
• One-call system (reference 12)
• Survey by inspection pig (magnetic flux type or comparable
device)—every 3 years
or
• Hydrostatic tests
- yearly or
- after indication of spill .
(5) Lines with two or more reportable spill incidents within 1 year:3
• Visual inspection by air or ground patrol—weekly
• One-call system (reference 12)
aRecommended to be in effect for approximately 3 years after spill inci-
dents are reduced to a normal rate.
-------
• Survey by inspection pig (magnetic flux type or comparable
device)—every year
• Hydrostatic tests—yearly.
This schedule should be in effect until approximately three years after
spill incidents are reduced to a normal rate.
Recommended developmental and new inspection and leak detection methods
include:
• Mathematical modeling
• Passive acoustic array
- retrofit (prevent and/or detect damage from outside forces)
- new lines (prevent damage and detect failures).
-------
SECTION 4
PETROLEUM TRANSPORT - SYSTEMS, MEASUREMENT,
CHANGES AND LOSSES
The extent and nature of petroleum losses from the petroleum trans-
portation systems are studied in this section to provide an overall per-
spective of the problem of accidental spills from petroleum transporta-
tion systems. Estimates are made of the total quantity of petroleum lost
and the amount attributable to each major loss mechanism. Emphasis is di-
rected to the problem of spills from the transmission line (line pipe).
This section also provides basic information that is used in the detailed
analysis of the oil spill risks (see Section 6) and the capabilities or
methods to reduce the risk (see Section 7).
The problems and limitations that affect the accuracy of the estimates
of the total petroleum losses or the losses from specific mechanisms are
discussed. These include:
• Variety of complex petroleum transportation systems in use
• Poor and sometimes inadequate information on petroleum sys-
tems and loss statistics
• Numerous sources of errors that are possible in
- measurement of losses
- petroleum accounting and loss reporting to U.S.
Government agencies
- compilation of petroleum and losses by U.S.
Government agencies
- operational and accounting practices of each
individual company
- other.
Important areas necessary in determining losses are presented and used in
determining actual losses. Specific areas include:
• _Petroleum system description
• Loss mechanisms
• Measurement
• Fluid changes
9
-------
• Instrumentation system accuracies
• Loss statistics.
Each area is discussed in separate subsections.
In order to have an overall understanding of the petroleum transpor-
tation system, the various means of petroleum transport are reviewed in
Section 4.1. Distribution and storage are discussed and the transporta-
tion modes compared. Petroleum pipeline systems are described and de-
tailed information is presented on mileage, diameter, age, type of pipe-
line, and type and volume of oil transported. A pipeline grouping system
is also selected.
Commonly used means of measuring and estimating the losses are exam-
ined in detail for the petroleum pipeline system in Section 4.2. Signifi-
cant loss mechanisms, i.e., evaporation, leaks, and admixture shrinkage,
and false indications of losses that create either a bias or uncertainty
in the loss measurement are identified and briefly discussed.
The ability to accurately measure petroleum (including losses) is
discussed in detail in Section 4.3. Meter errors, calibration errors,
variable conditions that affect meter performance, inaccurate correction
factors, accounting errors, human errors and bias and false indication of
errors are considered.
Changes in fluid between receipt and delivery are discussed in Sec-
tion 4.4. Fluid temperatures, water settling, admixture shrinkage and
other factors are considered.
Total system measurement errors are discussed in Section 4.5. Com-
bined errors of typical leak detection equipment and typical line fluctua-
tions (variations of pressure, temperature, etc.) are considered. System
errors are converted to barrels of oil that can go undetected for a typi-
cal instrumentation system and typical line variations.
An analysis of petroleum loss statistics is carried out in Section 4.6.
This was done primarily to estimate total losses for the petroleum pipe-
line systems and for each loss mechanism. The data analyzed were obtained
primarily from U.S. Government published statistics. However, statistics
from foreign countries, reports from government agencies and programs,
statistics from major oil companies, and loss analyses of reported losses
by Science Applications, Inc. (SAI) and others are used also. An in-depth
analysis of the oil spill statistics is carried out. Particular emphasis
is placed upon spill statistics of pipeline systems. This information is
examined in detail for both the overall pipeline system and the most criti-
cal component, i.e., line pipe. Information such as the number, frequency,
volume and severity of spills is examined in depth to determine key factors,
e.g., line pipe age and geometry, relating to the oil spill risk of pipe-
line systems.
10
-------
Estimates of total petroleum losses and total losses of specific loss
mechanisms for U.S. pipeline transportation systems are given in Section
4.7. This evaluation is made based on the data and information presented
in the previous subsections.
4.1 PETROLEUM TRANSPORTATION SYSTEM DESCRIPTION
The petroleum transportation system in the U.S. distributes and stores
petroleum by various means between the supply and retail outlet to the re-
fineries and markets. Transfer is carried out primarily by pipelines,
tankers, barges, railroad tank cars and tank trucks. A simplified version
of the petroleum transportation system with the main distribution and
storage components is shown in Figure 1. A transportation model is given
in Figure 2. Petroleum supply obtained from production operations is given
in Figure 3.
Quantitative information on the major modes of petroleum transfer, in-
cluding the type and volume of oil transported, and a pipeline grouping sys-
tem is presented in the subsections that follow. Detailed information is
presented for the petroleum pipeline system because it is the major means
of transporting petroleum and responsible for most of the losses in the
petroleum transportation system.
4.1.1 Percentage of Petroleum Moved by Transportation Modes
Pipeline systems currently account for more than half of the petroleum
transported in the U.S. and this percentage has been increasing over the
past forty years. Data on the percentage of crude and gasoline moved by
various modes in 1975 are given in Table 1.
TABLE 1. PERCENTAGE OF CRUDE AND GASOLINE MOVED IN 1975a, BY
METHOD OF TRANSPORTATION
Method Crude Gasoline
Pipeline
Coastal tanker
Barge
Truck
Railroad
77.4
10.75
10.75
0.55
0.55
42
18
18
21
0.35
Reference 2
Trip length varies for the particular transportation mode. Typical trip
lengths (4) for these modes are the following: pipeline, 300 miles;
tankers, 50 miles; barge, 3000 miles (round trip); truck, 500 miles,
trains, 500 miles.
11
-------
ro
IMPORTED
CKIM
OIL
CRUDE
OIL
STORAGE
DOMESTIC
CRUDE
OIL
PIPELINE
COASTAL
TANKER
UARGE
' RAILROAD
TANKCAR
TANK
TRUCK
REFINERY
GASOLINE
STORAGE
PIPELINE
IMPORTED
GASOLINE
COASTAL
TANKER
BARGE
RAILROAD
TANK CAR
MARKETING
TERMINAL
GASOLINE
IANK TRUCK
SERVICE
STATION
Figure 1. Petroleum transportation system. (Source: Reference 2.)
-------
PUMP STATION
TANK FARM
ilL_
-»- Q1 (INPUT AMOUNT OF OIL)
PIPELINE
(A) PIPELINE TRIP STAGE
TANK FARM
VESSEL
LOADING
*- VESSEL
VESSEL I
IN-TRANSIT UNLOADING
(U) TANKER/BARGE TRIP STAGE
Q1
oo oo
LOADING
J - CHU
OO OO
IN-TRANSIT
OO OO
(C) TANKCAR TRIP STAGE
TANK FARM
oooo
TANK FARM
OR TERMINAL
Figure 2. Petroleum transportation model. (Source: Reference 2.)
-------
t
OIL WELLS
CRUDE
JRINE
[MULSION
T"
(
PRODUCTION
GATHERING
SYSTEM
CRUDE
OIL
Figure-3. Flowsheet of petroleum production operations.
14
-------
Currently, over 77 percent of the crude and 42 percent of the gasoline are
transported by pipeline systems. Table 2 shows similar results for crude
and refined products and reveals an increasing trend in the percentage
transported for the years between 1938 and 1974.
4.1.2 Type and Volume of Petroleum Supplied and Transported
Annual statistics on the type and volume of petroleum supplied and
transported in the U.S. are compiled by the Bureau of Mines (BOM)3. Table
3 shows these statistics for the volume of crude and gasoline produced and
imported for the years 1972 through 1977. During this 6-year period, the
total volume increased at an annual rate of about 4 percent from 6.6 bil-
lion to 8.0 billion barrels per year. United States production of crude
both onland and offshore (Table 4) decreased an average of about 2 and 4
percent a year, respectively. Data on foreign and domestic crude and
natural gas liquids transported by pipelines to refinery input and petrol-
eum products turned into pipelines for the 5-year period between 1972 and
1976 are given in Table 5. During this period, total liquids transported
increased an average of about 2 percent a year to 8.9 billion barrels.
Annual statistics on the volume and type of petroleum transported in-
terstate are compiled by the Interstate Commerce Commission (ICC)6. Total
volume received into the interstate system (Table 6) increased from 6.2
billion barrels in 1966 to 9.8 billion barrels in 1976. This represents
an average volume increase of about 5 percent a year over this 11-year
period. The number of barrel-miles in trunk line movement also increased
at a similar rate.
4.1.3 Pipeline Transportation System
Petroleum pipeline transportation systems in the United States consist
of many individual systems of diverse design, dimensions, construction, ma-
terials, age, operation and maintenance. However, most systems transport-
ing liquids through pipelines fall under the jurisdiction of Federal regu-
lation (Title 49- Part 195)7 and must meet the same standards in design,
construction, testing, operation and maintenance, and accident reporting.
These regulations are given in Appendix A.
Typical pipeline systems are shown in the simplified diagram in Fig-
ure 2(A) and the more detailed diagram in Figure 4. In this study, pipe-
line systems (Table 7) are separated into two major areas (onshore and
offshore) with four major components (production, pumping, pipeline and
terminals).
Petroleum pipeline systems are comprised of three main components:
pipelines, storage tanks and pumping stations. Crude gathering lines are
the means through which crude oil is first gathered from production wells.
Crude is operated on, temporarily stored and then transported to refineries
by trunk!ines. Product trunk!ines then distribute products from the re-
fineries to markets and marine terminals. Storage tanks provide short-term
15
-------
TABLE 2. PERCENTAGE OF CRUDE PETROLEUM AND PETROLEUM
PRODUCTS MOVED BY METHOD OF TRANSPORTATION, 1938-1974a
Crude
Year
1938
1948
1958
1968
1970
1972
1973
1974
Pipeline
71.01
68.48
76.35
74.08
74.30
75.75
76.89
74.81
Mater
Carriers
25.58
23.26
16.90
18.62
18.90
16.10
14.13
13.47
Motor
Carriers
1.17
3.86
6.45
7.11
6.65
7.92
8.68
11.29
Railroad
2.24
4.40
0.30
0.19
0.15
0.23
0.30
0.43
Pipelines
6.35
11.36
20.48
30.41
31.12
32.39
32.74
33.54
Product
Mater
Carriers
52.65
44.70
37.51
25.69
26.75
26.92
25.78
25.84
Motor
Carriers
10.59
29.85
36.76
41.35
39.72
38.55
39.31
38.48
Railroad
30.41
14.09
5.25
2.55
2.41
2.14
2.17
2.17
a.
Source: Reference 5.
-------
TABLE 3. VOLUME OF OIL PRODUCED AND IMPORTED, 1972-1977
(THOUSANDS OF BARRELS)
Year
19/2
1973
1974
1975
1976
1977
Produced
3,455.368
3,360,903
3.202,585
3,056.779
2,976,180
2.985,360
Crude
Imported
811,135
1,183.996
1,269,155
1,498.181
1.935.012
2,397.468
Total
4,266,503
4,544,899
4,471,740
4,554.960
4,911,192
5,382,828
Produced
2,319,950
2.401.860
2,337.467
2,379,919
2,516.989
2.581.971
.
Gasoline
Imported
24.787
48,759
74,402
67,249
47,774
78.434
1
1
Crude and Gasoline
Total
2.344.737
2.450.619
2.411,869
2,447.168
2,564,763
2.660.405
Produced
5.775.318
5.762.763
5.540.052
5.436.698
5.493.169
5.567.331
Imported
835.922
1,232,755
1,343.557
1.565,430
1.982.786
2.475.902
Total
6.611.240
6.995.518
6,883.609
7.002,128
7,475,955
8.043.233
-------
TABLE 4. OFFSHORE PRODUCTION OF CRUDE PETROLEUM
(THOUSANDS OF BARRELS)
State
Alaska
California
Total
Federal
State
Louisiana
Total
Federal
State
Texas
Total
Federal
State
U.S. Total
1973
62,132
89,261
18,781
70,480
436,875
381 ,607
55,268
1,419
738
681
589,687
1974
60,308
83,918
16,779
67,134
398,329
351 ,825
46,825
1,081
504
577
543,636
1975
60,358
79,671
15,435
64,236
356,458
316,480
39,978
779
426
353
497,266
1976
54,512
70,587
13,975
56,612
335,849
298,378
37,471
1,522
812
710
462,470
1977
50,043
60,471
12,282
48,189
316,366
282,552
33,814
2,891
382
2,509
429,771
18
-------
TABLE 5. VOLUME OF PETROLEUM TRANSPORTED BY PIPELINES IN THE UNITED STATES
(THOUSANDS OF BARRELS)
I).
C.
D.
He-finery Input Kuceipts 1972
Crude domestic
Interstate
Pipeline
1 ankers and bargus
Interstate
Pipeline
Tankers and barges
Total - domestic crude
Crude foreign
Pi peli nes
Tankers and barges
Total - foreign crude
Total - domestic and foreign crude (A,B)
Natural gas liquids 302,445
Total - all refinery inputs (A.U.C)
Petroleum products turned into pipelines
Gasoline 1.636.213
Jet fuel 228,476
Kerosene 47,499
Distilled fuel oil 656,798
Natural gas liquids 438,000
Total petroleum products 3,006,986
TOTAL A.B.C.D
1973
1.796,892
148.862
1.108,101
249,819
3,303,674
408,740
775.331
1.184,071
4.487,745
297,482
4,785.227
1.759,322
249,621
46,883
727,019
438,000
3,220.845
8.006,072
1974
1 .691 .458
130,410
1,061,441
211.530
3,095.839
370.654
896,142
1.266.7%
4.362.635
272.416
4,635,051
1,773.951
248,315
35,941
701.798
467.286
3.227.285
7.862,336
1975
1.641.005
134.857
1.021.514
170,168
2.967.544
397,518
1.100,973
1.498,491
4,466,035
259,319
4.725.354
1,822,830
259. 57B
30.864
667,058
504.714
3.285.044
8,010.398
1976
1,622,620
123.824
1,102,118
146.825
2,995,387
443,026
1.487,350
1,930,763
4,925.763
265.451
5.191,214
2.003.207
323.233
30.394
745,335
567,070
3,669,244
8,860,458
-------
TABLE 6. VOLUME AND TYPE OF OIL TRANSPORTED BY INTERSTATE PIPELINE SYSTEMS
Oil transported statistics
No. of barrels received
into system:
1976
1975
1974
1970
1969
1968
1967
1966
From connecting carriers:
Crude oil
Products
Total
Originated:
Crude oil
Products
Total
Total received
into system
No. of barrels delivered
out of system:
To connecting carriers:
Crude oil
Products
Total
Terminated:
Crude oil
Products
Total
Total delivered
out of system
No. of barrels having
truck line movement:
Crude oil
Products
No. of barrel-miles
(truck lines):
Crude oil
Products
2
3
3
3
6
9
2
2
3
3
6
9
5
3
I
1
.6)). 306. 000
5)1.521.000
.122.827.000
,434.362.000
,229,841.000
.664.203.000
.787.030,000
.256.131.000
7)9.675.000
,975,806,000
.764.689.000
,0)6.608.000
.781.297.000
.757.103.000
,824.141,000
,8)3.233,000
,609.173,784
,307.449,94)
2.191.757.000
504.782.000
2,699,539.000
3.492,6)7.000
3.166,659.000
6.659,275.000
9.358,8)4.000
2.125,955.000
691,360,000
2,8)7,315.000
3.548,507.000
2.974,788.000
6.523,295.000
9.340,610.000
5.369,293.000
3.645,461,000
1,545,061.092
1,333.874,961
2.185.749.000
526.239.000
2.711.988.000
3,575.464.000
3.072.546,000
6.648.010,000
9.359.998.000
2.055.769.000
692.200.000
2.747.970,000
3.679,549.000
2.905.822,000
6,585,371.000
9,333,340.000
5,390.999.000
3.588,808.000
1.625.436.501
1.239.411.795
1.729.963.878
415.840.204
2.145.804.082
3.568. 346.99B
2.449.380.365
6.017.727.363
8.163.531.445
1.767. 068.) 33
552.544.037
2.3)9.6)2.170
3.520.580.985
2.306.491.599
5.827.072,584
8,146.684,754
4.756.244.120
2.863.735.39)
1.428.362.484
1.021.484.502
1.637.986.854
395,999.695
2.033.986.549
3.404.988.846
2.316.317.430
5.721.306.276
7.755.292.825
1.696.310,931
546,233.948
2.242.544.879
3.335.595.359
2.163,674.983
5.499.270,342
7.741.8)5.221
4.445.921.966
2.718,870.361
1.321.711.176
995.029.558
1
1
3
2
5
7
1
2
3
2
5
7
4
2
1
.530.981,236
35) .802. 367
.882.783.603
,203,481.568
.203.169.70)
.406.651.269
.289,434,872
.528,033.969
519.796,144
.047.830,113
,188,167.552
.032,761.561
.220.929.113
.268.759.226
,273.298.625
.559.522.918
.270,465.366
951.655.430
1.456.006.801
305.236.793
1.761.243.594
3.017.250.927
2.035.245.6)2
5.052,496.539
6.813.740.133
1,423.072.341
467.072.509
1.890.144.850
3.046.870.063
1,863,409.038
4.910.279,101
6.800,423.95)
3.886.370.6)3
2.338.581.680
). 207. 32) .010
926.121.262
1,388
262
1,651
2,826
1,773
4,600
6,251
1,369
400
1,770
2,833
1.634
4.467
6,238
3,661
2,049
1.181
719
,396.020
.979.877
.375,897
.374.622
.954,368
,328,990
.704.887
.843.9)6
.508,66)
,352,577
,258.057
.581.587
.839.644
,)92.221
,172.872
,934,224
,152,599
,728,561
-------
Oilfield
Petrochemical
Plant
Products Line to Petrochemical
Plants
Products Tankage
Products to Marine
Terminal
Products Pipeline
to Markets
Marme and Tank Terminal
Figure 4. A pipeline system composed of gathering, trunk and
distribution lines. (Source: Reference 8.)
21
-------
TABLE 7. PIPELINE TRANSPORTATION SYSTEM AND ASSOCIATED SUBSYSTEMS
ro
ro
Onshore
1. rrodBctlofl 1.
Mill
Mpln,
ml MI
ruwi
Production ottntrlnf lyitn*
Optrittoiu
billion ortiklnj
Uittr knockout
fill-oil itpiritlon
Uittt Mttr itpiritlon
Hut* nittr dltpoiil
felkttt *'
Vllttl
Flinoti
Stonot tmki
soil*
tilkltl
2. looittr Puwln* Stitlon
Fin* ttctton
Mpln*.
Hock «
-------
storage for receiving, measuring, sorting and routing. Pumping stations
normally are installed every 25 to 150 miles along a line and furnish the
energy to transport the petroleum at proper pressures.
A variety of petroleum pipeline systems exist and significant changes
and trends in construction have occurred in recent years. In order to ac-
count for these, pertinent pipeline system information such as pipeline
dimension (diameter, length), age, type, volume and type of oil trans-
ported, storage, etc., must be considered. Also, a suitable pipeline group-
ing system is necessary for effective utilization of pipeline system data.
This information is presented in the subsections that follow.
4.1.3.1 Pipeline - Construction, Mileage, Diameter, Type, Fl-uid, Age—
The petroleum pipeline system has undergone numerous changes and sub-
stantial growth (Figures 5 and 6) over the past forty years. Rapid accel-
eration of pipeline construction and pipeline flow to the East Coast oc-
curred during the war years in the mid-1940's. Shortly after the war, many
crude petroleum pipelines were taken out of service, or converted to gas
or product pipelines. Since the 1950's larger diameter, long distance
trunk!ines have been constructed.
Petroleum pipelines are normally made of steel pipe with butt welded
joints and are as large as 54 inches in diameter. These pipelines are
usually protected from external corrosion by coatings or cathodic protec-
tion and from internal corrosion by injection of corrosion inhibitors and
chemicals. Petroleum pipelines are normally buried underground or under-
water as protection from external damage. Modern pipelines are made with
heavy wall steel pipe and high tensile and bursting strength. Only ten
years ago, however, standard weight steel pipe with yields to only 50,000
psi was normally used.
Data on total pipeline mileage (Table 8) in service by category are
compiled by BOM9'10'11 and by the ICC6 for interstate pipeline mileage.
Currently there are about 227,000 miles of pipelines and 174,000 miles of
these are interstate. In the past nine years, pipeline mileage has in-
creased at a rate of about 1 percent a year.
Mileage by pipeline diameter is given in Tables 9 and 10. Most of
the pipelines are 12 inches or less in diameter. These sizes currently
account for 191,593 miles or 84 percent of the total pipeline mileage;
mileage has increased only slightly, at a rate of about'0.4 percent per
year, from the 183,519 miles that existed in 1968. Mileage of large diam-
eter pipelines, however, has increased substantially in the past few years.
For example, in 1968 there were 3,312 miles of pipeline larger than 24
inches in diameter and in 1977 there were about 8,268 miles. This is an
increase of about 150 percent over that 10-year period. The relationship
between pipe diameter, length and volume contained inside is given in -
Table 11. Note that volume increased by the diameter squared. Thus,
significant increases in larger diameter pipeline mileages result in huge
increases in pipeline capacity and throughput.
23
-------
I»OS
1932
1973
Figure 5. Growth of crude oil pipelines in the United States
(Source: Reference 8.)-
24
-------
130
80
70
60
50
40
30
20
10
Crude Gathering Lines
CruSeTrunklines
Product
Trunk!ine
1941 1944
1950
1956
1962
1968
1974
Figure 6. Mileage of petroleum pipelines in the United States.
25
-------
TABLE 8. PIPELINE MILEAGE IN THE UNITED STATES
TOTAL AND INTERSTATE
Trunk! ines
Year
1926
1931
1936
1941
1950
1953
1956
1959
1962
1965
1968
1971
1974
1977
Crude
44,470
58,020
57,820
65,180
71,373
75,228
78,594
70,317
70,355
72,383
70,825
75,143b
76,250
77,972
Products
Total
a
a
a
9,001
20,881
27,236
36,420
44,483
53,200
61,443
64,529
72,396b
78,038
81,296
Total
in the United States
44,470
58,020
57 ,820
74,181
92,254
102,464
115,014
114,800
123,555
133,826
135,354
147,539
154,288
159,268
•lathering
Lines
45,700
53,640
52,760
53,170
60,560
68,040
73,526
75,182
76,988
77,041
74,124
71,132
69,247b
67,798
Total
90,170
111,660
110,580
127,351
152,814
170,504
188,540
189,982
200,543
210,867
209,478
218,671b
223,535b
227,066
Total Interstate
1976
58,544
67,913
126,457
39,235
174,072
alncluded in crude lines.
Revised.
26
-------
TABLE 9. MILEAGE OF PETROLEUM PIPELINES BY DIAMETER IN THE UNITED STATES,
JANUARY 1. 1971 AND JANUARY 1, 1974a
ro
Size (Inches)
10
12
14
16
18
20
22
24
26
28
30
32
34"
36
40
42
41)
Total
Tankers
Crude
Jan 1, 19/7
84
468
3,760
9,518
19,303
11.358
10,156
589
5,975
2,059
5,251
3,059
2,256
1,069
-
2,006
-
785
134
527
4
111
77,972
Jan 1, 1974
100
479
3,344
9,668
20,424
11,040
10,183
598
5,668
2,091
4,972
2,664
1,483
1,034
-
960
-
789
5
637
3
108
76,250
Products
.Ian 1. 1977
125
630
2,895
17.706b
27,193
11,958
9,394
2,817
2,077
1,202
1,238
9
566
287
597
520
288
233
1,519
103
-
-
81,2%
Jan 1, 1974 Jan
156b
688b
2,870b
16,786b
27,329b
10,612b
8,703b
2.786
1.954b
1,177
1.072b
9
543
287
546
480
288
233
1,519
-
-
-•
78.038b
Gathering
1, 1977
12,575
10,252
27,382
10,425
5,023
1,075
954
39
49
1
15
1
2
5
-
-
-
-
-
-
-
-
67,7911
Lines
Jan 1, 1974
13,547
11,020
27,687
I0,210b
5,139b
940
626
18
42
1
9
1
3
4
-
-
-
-
-
-
-
-
69.2471'
Total
Jan 1, 1977
12,784
11.350
33,537
37,508
51,519
24,391
20,504
3.445
8,101
3,262
6,504
3,069
2,8?4
1,361
597
2,526
288
1,018
1.653
710
4
111
227,060
Jan 1, 1974
13,803
12,187b
33,901b
36,664b
52,892b
22,G92b
19,512b
3,402
7.6641'
3,269
6.0531'
2,674
2,029
1,325
546
1,440
288
1,022
1,524
637
3
108
223, 5351'
"Revised.
'includes a small amount of 5-inch pipe in trunk!incs.
Includes a small amount of 7-inch pipe In trunk!Ines.
eIncludes a small amount of 9-inch pipe in trunk! iites.
Includes a small amount of 11-inch pipe In trunk I IMPS.
'includes a small amount of 35-inch pipo In trunklinos.
-------
TABLE 10. MILEAGE OF PETROLEUM PIPELINES BY DIAMETER IN THE UNITED STATES
JANUARY 1, 1971 AND JANUARY 1, 1968a
ro
oo
Trunkl lues
Crude
Size (Inches)
2
3b
4C
6d
Be
.of
12
14
16
18
20
22
24
26
28
30
32
349
36
10
42
lotal
Jan 1. 1971
_
479
3.090
9.435
19.269
11.867
10.106
571
5.630
2.002
5,040
2.664
1.504
1.039
-
938
-
792
2
634
4
75,066
Jan 1. 1968
.
269
2,075
8,897
19,525
12,198
9,288
548
5.341
1,992
- 4,765
2.659
1.497
830
-
645
-
. 291
4
-
1
70,8?5
Products
Jan 1. 1971
.
702
2,771
16,732
25,634
9,785
8.732
2.441
1.603
919
907
9
131
217
177
300
288
-
1,058
-
-
72,406
Jan 1. 1968
-
641
2.419
16.002
22.822
8,187
7,189
2,129
1.441
937
1.075
9
137
-
-
104
288
-
1.149
-
-
64.529
Gather tny L tries
Jan 1, 1971
14.463
11.166
27.769
11.280
4,716
872
754
27
65
1
15
-
3
1
-
-
-
-
-
-
-
71.132
Jan 1, 1968
15.318
11,980
28.394
12.016
4.842
846
611
27
70
1
15
-
3
1
-
-
-
-
-
-
-
74.124
Total
Jan 1, 1971
14.463
12,347
33,630
37.447
49.619
22,524
19.592
3.039
7. 298
2,922
5,962
2.763
1.638
1,257
177
1.238
208
792
1 ,060
634
4
21H.604
Jan 1, 196U
15,318
12,890
32,888
36,915
47,189
21,231
17.088
2,704
6,852
2,930
5,855
2,668
1,637
831
-
749
208
291
1,153
-
1
209,4711
Source: Reference 11.
Includes a small amount of 2-Inch pipe In trunk lines.
cIncludes a small amount of 5-inch pipe in trunklines.
Includes a small amount of 7-inch pipe in trunklines.
''I no) mles a small amount of 9-inch pipe in trunk) inos.
Includes a small amount of 11-inch pipe irr trunklines.
yI deludes a small amount of 35-inch pipe in trunklines.
-------
TABLE 11. RELATIONSHIP BETWEEN PIPE DIAMETER, LENGTH,
AND VOLUME CONTAINED INSIDE3
Inside
Diameter
(in.)
2.067
4.026
6.026
8.071
10.020
12.090
24.000
28.000
34.750
40.00
46.500
aSource: Reference
b Lfmil _ 194.965
i n,n^ - 313.766
Lengthb
hold
Miles
45.63
12.03
5.37
2.99
1.94
1.33
0.34
0.25
0.16
0.12
0.09
12.
required to
1000 bbl
Kilometers
73.44
19.36
8.64
4.82
3.13
2.15
0.54
0.40
0.26
0.20
0.15
V(/m1) = 5.129
\l( t\,m\ - t 1Q7
Barrels per
mile of line
22
83
186
334
515
750
2,954
4,021
6,194
8,207
11,090
D2
Barrels per
km of line
14
52
116
208
320
466
1,836
2,499
3,849
5,099
6,891
t
("D" is in inches.)
29
-------
Pertinent information on the three types of pipelines, i.e., crude
gathering lines, crude trunk!ines and product or distribution trunk!ines,
are given in Tables 8 through 10. Trunk!ines currently account for about
57 percent of the total pipeline mileage and 72 percent of the interstate
mileage. For the 28-year period between 1950 and 1977 trunk!ine mileage
has increased at a rate of about 2.5 percent a year from 92,254 to 156,268
miles while gathering lines increased at a rate of only about 0.4 percent
a year from 60,560 to 67,798 miles. Essentially all gathering lines (over
99 percent) and most trunk!ines (over 84 percent) are 12 inches or less in
diameter.
Statistics on the main types of petroleum fluids transported, crude
and product, are also presented in Tables 8 through 10. The current
mileage of crude and product trunk!ines is about the same or 79,972 and
81,296 respectively. However, product trunklines are somewhat smaller in
diameter than crude trunklines. For example, over 85 percent of product
trunklines and 69 percent of crude trunklines are 12 inches or less in
diameter.
Mileage of petroleum pipelines by age is not available from BOM sta-
tistics. However, a rough estimate was obtained using statistics from the
Office of Pipeline Safety and Operation (OPSO). Mileage figures for vari-
ous periods were available for natural gas transmission lines and gather-
ing lines. It was assumed that petroleum lines were similar in age to gas
lines. Knowing the total petroleum pipeline mileage and assuming the same
percentage of petroleum lines as gas lines were installed during the same
time periods, an estimated mileage was obtained. Results are shown in
Table 12.
Additional information on pipeline mileage, diameter, type, fluid,
and age such as pipeline activity, activity by state, etc., are given in
Appendix B for the years 1968, 1971, 1974 and 1977.
4.1.3.2 Storage—
Essentially all petroleum in the transfer system is contained in the
pipelines and tanks. Operating conditions for containment are illustrated
in Figure 7. Statistical information on both storage capacities and in-
ventories is compiled monthly by BOM and periodically by the National
Petroleum Council (NPC). A summary of inventories and storage capacities
are presented in Tables 13 and 14. Petroleum inventories as reported by
BOM include:
• Stocks in crude oil trunklines and their terminals
• Stocks in product trunklines and their terminals
• Crude and product inventories at refineries
• Product stocks in bulk terminals.
Not included are stocks located at marketing outlets and military instal-
lations.
30
-------
TABLE 12. PETROLEUM PIPELINE SYSTEM MILEAGE IN THE
U.S., BY YEAR INSTALLED
1974
Gas Lines
Miles by Year Installed Mileage
Estimated 1974
Petroleum Line
Mileage
Unknown Year
Prior to 1930
1930 to 1939
1940 to 1949
1950 to 1959
1960 to 1969
1970 to now
10,130
14,101
18,084
31,906
86,533
90,964
30,147
4
5
6
11
31
32
11
8,941
11,176
13,412
24,588
69,295
71,531
24,588
Total
282,135
223,535
31
-------
3
01
19
18
17
16
15
14
13
12
11
10
9
8
7
6
5
4
3
.2
I
0
SPARE CAPACITY RESERVE
EMERGENCY
BUILDUP
TANK TOPS (UNAVAILABLE CAPACITY)
MINIMUM OPERATING RESERVE
i /EMERGENCY
\ / DRAWDOWN
cc
O
ui
c
o
z
TANK BOTTOMS (UNAVAILABLE STOCKS)
PIPELINE FILL
TIME
Figure 7. Illustrative operating conditions: Trans-Alaska Pipeline.
(Source: Reference 13.)
32
-------
TABLE 13. ANALYSIS OF CRUDE PRODUCTS REPORTED TO NPC£
CO
co
CRUDE OIL INVENTORIES
(Thoutandf of Barrelil
Item
Tola) Inventory! H«ld by Reporting
Companies "
Unavailablt Inventoriet
Tank Bottom j & Refinariet
Operating Requirement/-
Pipclint Fill
Other Unavailable
Total Un*»aiUbla Inventoriei
Unavailabl* at Percent of
Total Rtporttd to NPC
Working Stocki and Available Inventoriei
March 31
1948
213.224
34.067
30.579
68.279
132,926
62.3
80.299
June 30
19SO
224.9-18
38.031
36.618
67.790
142.439
63.3
82.509
March 31
19S2
238.413
39.364
.41.028
;o.5i4
160.908
63.3
87.607
March 31
1954
243.692
41.423
44.341
B0.8B4_
166.848
68.4
77.044
March 31
19S7
226.516
41.277
47,036
69.269
157.682
69.9
67.934
Sept. 30
1982
217,626
41.431
51.722
68.284
150.407
73.2
58.219
Sept 30
1969
240.341 •
44.701
60.311
60.97?
165.989
69.1
74,352
Sept. 30
1973
221.859
43.695
57.141
54.086
154.822
69.8
67.037
Source: Reference 13.
'Excluding producers lease stocks and cargoes in transit from foreign countries.
"Contents of tank bottoms, in refinery pipelines and minimum quantity required
to assure continuous processing, handling and blending various grades of
crude oil.
-------
TABLE 14. ANALYSIS OF CLEAN PRODUCTS REPORTED TO NPC*
CLEAN PRODUCTS
(Thousands ot Barrels)
CO
Item
Total Imintorin Held by
Reporting Companies
Unavailable Inventories
Tank Bononu u
Unfinished a( Refineries
Refinery tints & Operating
Equipment
Ont-Half Average Silt Watir
Cargo Ractipt
Other Unavailable Inventoriet
Pipeline Fill
Pipeline Operating Require-
ments
In Transit- Truck. Tank Car.
Barge & Tanker from
Domestic Source
Total Unavailable Inventories
Unavailable as Percent ol Total
Reported to NPC
Working Stocks and Available Inventories
M«ch31
1948
149.903
22.262
8.9S7
1.S24
9,993
14.778
4.813
6.448
6.462
76.233
50.2
74.670
June 30
1960
180.695
24.176
9.403
1,641
10.077
11.877
4.046
6.702
6,030
73.692
41.0
106.643
March 31
1962
220.283
28.128
10.739
1.768
13.083
14.648
8.202
4.1S8
8.265
86.9*1
39.S
133.292
March 31
1984
251.450
28.866
7.234
1.772
13.417
12.S57
12.747
7.963
7,239
91.794
36.5
159.656
March 31
19S7
294.127
27.834
11.520
1,802
14,866
12.836
14,816
11.162
8,173
103.539
35.2
190.588
Sept 30
1962
385.840
32.648
-b
7.190
15.682
8.977
17.022
9.701
7,926
99.146
25^7
286.694
Sept. 30
1969
430.148
33.783
-b
929
14.686
9.988
35.854
15.651
6,150
117.041
27.2
313.107
March 31
1973
358.472
39.589
'-b
889
14.660
8.294
43.176
8.758
4.928
120.294
33.6
238.178
Sept 30
19^73
443.254
39.822
- b
914
14.342
8.797
44.743
10.139
6.435
125,192
28.2
318,062
Source: Reference 13.
'Unfinished at refineries has been omitted because of change in Bureau
of Mines method of reporting effective January 1, 1962.
-------
Tankage is provided at all receiving terminals and delivery points.
Product tankage is much greater than crude because of its widespread dis-
tribution. In contrast, crude pipeline fill (Table 15) is over 50 percent
higher than products.
4.1.3.3 Grouping Systems—
A pipeline grouping system can be used advantageously in the analysis
of loss and spill data and in the development and implementation of an ef-
fective spill prevention program. Various grouping systems exist such as
those based on regional areas and districts or pipeline corridors. Other
groupings might also be made based on individual states, pipeline size or
age, etc. Regional areas and districts for widely used groupings, i.e.,
standard Federal regions, BOM refining districts, and Petroleum Administra-
tion for Defense (PAD) districts are shown in Figures 8 and 9. The number
of reported liquid pipeline accidents reported to OPSO during 1975 are also
shown in Figure 8.
Selection of a grouping system was based on a number of considerations
such as:
• Pipeline location
• Pipeline age
• Number and volume of reported spills for regional and
district locations
• Areas of high risk
• Implementation of spill prevention program.
Crude and product lines (See Appendix B - Pipeline Mileage by State - and
Figure 5) are in widespread use with somewhat uneven distribution through-
out most of the United States. Eleven states, i.e., Texas, Oklahoma,
Kansas, Illinois, California, Louisiana, Wyoming, Missouri, Ohio, Pennsyl-
vania and New Mexico account for over 76 percent of the pipeline mileage.
Only Texas, however, has a significant proportion (over 10 percent) of
total pipeline activity with over one-third of the crude lines, one-sixth
of the product lines (Appendix B) and one-fifth of pipeline fill (Table 15).
Similarly, the number of spills and losses reported to OPSO occurred through-
out most of the United States with the greatest number of spills and largest
amount of losses occurring in Texas.
A grouping system based on standard Federal regions was judged to be
best for this study. The number and location of Federal regions were con-
sidered to be particularly well-suited for analysis and for implementation
of a spill prevention. Spill statistics show that a particular oil pollu-
tion problem is generally confined to only a few Federal regions. This
allows concentrated effort in a particular region for solution of a regional
problem and also minimizes jurisdictional disputes. For example, areas of
high concentrations of pipelines, of most accidents, and of high risk, such
as offshore, are located in only a few Federal regions. Also, areas of
v
35
-------
TABLE 15. PIPELINE FILL FOR PETROLEUM PIPELINES IN THE UNITED STATES
JANUARY 1, 1977 AND JANUARY 1, 1974a
(ThouMnd barrels)
Catharine lines Crude-oil trunklines Total crude line*
Produce line*
State
Jan. 1, Jan. 1,
1977 197ft
Jan. 1.
1977
Jan. 1.
1974
Jan. 1,
1977
Jan. 1,
1974
Jan. 1, Jan. 1,
1977 1974
Alabaaa * 10 169 161 175 171 4.358 4 246
Alaska 8 12 130 131 138 143 47 11
Arizona - 1 450 446 450 447 727 727;
Arkansas 70 76 702 699 772 775 1,002 985
California 499 509 3,048 2,970 3,347 3,479 1,291 1,379
Colorado 82 74 612 604 694 678 184 138
Connecticut - - - - - 52 55
Delaware - - - - - 14 28
Florida 6 5 42 35 48 40 84 62
G««*-l« - - - - - - 2,725 2,623
I«l*ho - - 228 228
Illinois 223 216 5,830 5.955 6,053 6,171 2,845 3,089
Indiana 31 3ft 1,579 1,541 1,610 1,575 1,743 1,687
lova 50 89 50 89 1,963 1,519
Kansas — 818 678 3,824 3,670 4,642 4,348 2,760 r/ 2,484
Kentucky 117 144 1,725 954 1,842 1,098 17 14
Louisiana 751 495 4,386 4,097 3,137 4,592 4,089 2,924
Maine - - 421 405 421 405 2ft 24
Maryland and District of Columbia - . ... 510 511
Massachusetts - - - ...64 76
Michigan 156 66 4,60ft 3,963 4,760 4,027 790 802
Minnesota - - 4,662 ft.762 4,662 4,762 568 r/ 583
Mississippi 56 52 3.111 3,120 3,167 3,172 4,016 ~2,387
Missouri - - 3,899 3,827 3,899 3,827 2,541 2,486
Montana 107 98 751 703 858 SOI 257 228
Nebraska 60 60 1,046 1,047 1,106 1,107 693 112
Hewada - - 106 92
Hew Hampshire . - 280 208 280 208
Hew Jersey r/ - r/ 745 639
Mm Mexico 421 373 1,495 1,029 1,916 1,402 476 291
Hew York 8 13 77 132 85 145 440 449
Berth Carolina - - - - - - 1,383 1,383
North Dakota 52 48 89ft 943 946 991 194 183
Ohio 114 93 1.045 1,175 1,159 1,268 1,657 1.496
Oklahoma 1.042 1,116 4.356 3.981 5,398 5,097 2,932 r/ 2,548
Oregon - - - - - - 134 240
Fennsylvania 78 92 479 254 557 346 2,304 2,220
Rhode Island - - - . . - 3 3
South Carolina - - - - • 593 833
South Dakota - - - - - 147 147
Tennessee - - 1,235 880 1,235 880 208 184
Texas 2,969 3,129 21.995 17.764 24.964 20,893 7.614 r/ 7,162
Utah 23 28 328 309 351 337 103 ~ r/ 106
Tenant - - 297 297 297 297 - ~ -
Virginia - - - ... 1,533 ^535
Washington - - 115 108 115 108 54ft 500
West Virginia 99 120 75 8ft 174 204 91 72
Wisconsin • - 2,43ft 2,460 2,434 2,460 246 247
Wyoning 295 278 2,368 2,356 2,663 2,634 423 383
Tot.l 8.091 r/ 7,818 78,51* 71,159 86,605 78,977 55,468 r/ 50,121
r/ Revised
Reference 11.
36
-------
X Number of liquid pipeline accidents
In state during 1975
A) Number of liquid pipeline accidents
in region during 1975
Figure 8. Standard Federal regions (Commerce, EPA and HUD), with
number of liquid pipeline accidents.
37
-------
PETROLEUM ADMINISTRATION TOR DEFENSE (PAD) PATRIOTS
BUREAU OF MINES REFINING DISTRICTS
OMS
Figure 9. Petroleum administration for Defense and
Bureau of Mines refining districts. (Source: Reference 11.)
38
-------
similar terrain and population densities are generally confined to specific
regions. Most pipelines and spills, for example, are in region VI while
most inland spills are concentrated in low population densities of-regions
VI, VII and VII and most offshore spills in regions VI and IX.
4.2 PETROLEUM SYSTEM - LOSSES AND SPECIFIC LOSS MECHANISMS
A number of significant loss mechanisms and false indications of
petroleum losses exist in the petroleum transportation system in the United
States. These petroleum losses can occur at any of the numerous system
components such as those between the supply and the retail outlet shown in
Figure 1.
The significant loss mechanisms that cause actual petroleum losses are
the following:
• Evaporation (E)
• Leakage or spills (L)
• Admixture shrinkage (A)
These losses and typical measurement points (M) are indicated in Figure 10
for the petroleum system between the supply and refinery input.
Inaccuracies in petroleum measurement, calibration, estimates and ac-
counting procedures, have resulted in an inability to determine how much
crude is truly lost and to which mechanism (including undetected leaks).
Of the specific loss mechanisms, only evaporative losses have been studied
in detail. Estimates of these losses have been reported. However, studies
are still being carried out to improve both evaporative measurements and
equations for calculating the losses. Evaporative losses are difficult to
estimate. For example, these losses vary widely depending on the produc-
tion operations. Crude losses between the oil wells and crude oil storage
(Figure 3) have not been and probably will not be accurately determined in
the near future for a number of reasons. For example, leakage losses
normally can only be detected by visual inspection or pressure testing be-
cause other means of leak detection are not available. The accuracy of
the reported number of incidents and volume of petroleum lost is suspect.
There has been only limited analysis on spill statistics and regulations
on reporting spills to each government agency are different. Furthermore,
the statistic reporting of these spills by each government agency are
quite different. Thus it is difficult to compare and analyze statistical
spill data.
In this section, the scope of this study for the petroleum system and
losses is given. Then main sources of losses (leakage, evaporation and
admixture shrinkage) between the supply and refinery input and the refinery
output and marketing terminal are reviewed. Fault trees for pipeline spills
are presented. False indications and uncertainties of losses are also
identified.
39
-------
•
Imported
Crude
on
(Supply)
E
, t
1 i !
i | Crude
M | M Storage
i
1 1 1 !
T T T '
1 A L j
Oil yells
Crude Oil
Production
vSupply)
Refinerv
Figure 10. Major system components for crude losses
between supply and refinery input.
40
-------
4.2.1 Scope
It is beyond the scope of this study to determine all losses and
losses from each individual petroleum transfer system and each system com-
ponent. Hence, U.S. losses will be determined for the petroleum transpor-
tation system with the major losses, i.e., the pipeline transportation
system.9 The following losses and loss mechanisms are investigated in
this study:
• Total losses of crude between the supply (crude oil
storage, coastal tankers and barges) and refinery
input via pipeline system (see Figures 1 and 10).
• Total losses of product between refinery output and
marketing terminal via pipeline systems (see Figures
1 and 4).
• Individual loss mechanisms of the pipeline system be-
tween the crude supply (crude oil storage, coastal
tankers and barges) and refinery input via pipeline
system (see Figures 4 and 10).
• Individual loss mechanisms of the pipeline system be-
tween the refinery output and the marketing terminals
via pipeline system (see Figure 4).
• Spill losses from pipeline systems of petroleum produc-
tion operation (see Figure 3).
4.2.2 Losses
4.2.2.1 Spills-
Spills from pipeline systems are normally detected by visual monitor-
ing or metering. These spills can occur at almost any location and for a
variety of causes between the supply and refinery input and the refinery
output and marketing terminal.
Main components and subcomponents that are the major sources of leaks
from the pipeline system are given below.
aThis pipeline transportation system would be expected to have the major
losses because most petroleum is transported by crude and product pipe-
line systems. About 77 percent of the crude and 42 percent of the gaso-
line, and most of the crude and product are contained in pipelines and
storage tanks.
41
-------
1) Pipeline
Line pipe
Piping
Valves
Flanges
Fittings
Heaters
Miscellaneous
control, metering and monitoring equipment
2) Storage tanks
Tanks
Piping
Valves
Flanges
Fittings
Seals
Pumps
Miscellaneous control, metering and monitoring equipment
Spills from other pipeline system components include:
3) Pumps
Flanges
Gaskets
Seals
Fittings
Piping
Miscellaneous control, metering and monitoring equipment
4) Production
• Gas-oil separators.
Fault trees are used here for an overview of the sources of leaks
(failures) and in Section 6.3.2 for the detailed analysis of the risk of
oil spills from petroleum pipeline systems. Fault trees11* are logic
diagrams that describe the ways in which component failures and other
events can result, either alone or in combinations, in the top of the
event trees. The top event is the undesired event. The probability of
occurrence of the top event is desired. It can be calculated when all
the failure probabilities of the basic events of the fault are known.
These basic events on the tree are identified in Figure 6 and are com-
bined through "AND" and "OR" gates.
A summary fault tree of petroleum spills from the petroleum pipeline
system for the main areas and major systems (see Section 4.1.3) is given
in Figure 11. A fault tree for the main source of the spills, the pipe-
line, is given in Figure 12. Specific values of the variables (a,b,c...)
42
-------
-p.
CO
A-
_£..."""
ONSHORE
|
SPILL SPILL SPILL
FROM FROM FROM
PROOUCTIOU PIPELINES STORAGE
OPERATIONS lERMIilALS
A A A
OIL SPILLED FROM
PIPELINE TRANSPORTATION SYSTEM
~§
-~L- r C-,
SPILL FROM SPILL
BOOSTER FROfl
PIJMP PRODUCTION
STATIONS OPERATIONS
JT T
1 OFFSHORE j
1 1
SPILL SPILL SPIIL
FROM FROM FROfl
UNDERWATER MAR 1! IE PI IMP 1 'SO
PIPFLI.1ES TFRMltlALS PLATFORMS
_. .. f ^
A A /A
() Spills per year
Figure 11. Fault tree A,, oil spilled from petroleum pipeline system.
-------
a^ft-yr
Spill from
Onshore
Pipelines
Spill from
Pipelines
aop/ft-yr
Other Failures
Including Rupture
and Damage
a2/ft-yr
a3/ft-yr
as_p/ft-yr
a4/ft-yr
1
Spill from
" iderwater
.pe lines
to
a5/ft-yr
Other Failures
Including Rupture
and Damage
a6/ft-yr
Figure 12. Fault tree C, spill from pipelines. (Source: Reference 14.)
-------
of spill frequency, volume, etc., for particular pipelines are given in
Section 6.3.2.
4.2.2.2 Evaporation-
Evaporative losses for each loss mechanism and system component can
be determined either by metering of fluids or calculations based on evapo-
rative loss equations. Main sources of evaporative losses include:
1) Storage tanks
Breathing
Wetting
Filling
Emptyi ng
Boiling
2) Production
• Gas-oil separation
- Flaring
- Venting
- Incomplete extraction of butanes and pentanes from gas
- Storage
• Emulsion
- Accelerated evaporation by heat
- Storage
• Lease tank
- Accelerated evaporation because of splashing
- Storage
3) Transportation-pipeline
Filling tanks
Emptying of tanks
Propulsion (pumping)
Air eliminators
Valves
Fittings
Connections
4.2.2.3 Admixture Shrinkage-
Admixture shrinkage15 occurs when light hydrocarbons are mixed with
heavy ones and result in a combined volume that is less than the volume of
each individual hydrocarbon added separately. The resultant volume loss,
however, is only an apparent volume loss and there is no loss of weight.
Also, limited data indicates an increase in shrinkage with temperature
increase and a decrease in shrinkage with a pressure increase. Admixture
45
-------
shrinkage occurs quite often, typically when light hydrocarbon products
are blended into the crude stream. Figure 13 shows curves for estimating
volume change when blending volatile hydrocarbon with crude oils.
4.2.2.4 False Indications-
False indications of lossa occur when a loss appears to occur but
actually does not. Also, false indications may balance or cover up (by
showing a gain) an actual loss. These false indications create either
a constant bias or random uncertainty in the measurement of petroleum
and in particular, losses.
4.2.2.4.1 Bias in Loss Measurement
A number of conditions may exist which indicate either a loss or
gain, but in fact, are merely a bias in a specific direction in the
petroleum measurement. These include:
• Use of incorrect coefficient of expansion. (Coefficient of
expansion losses occur because the petroleum heats up and
expands as it is shipped. Non-linear expansion of the
petroleum occurs, but linear tables are used to correct
this error. This results in an apparent loss at the de-
livery end of the pipeline.)
• Entrained free air in the line
• Incorrect back pressure on meters
• Higher prover temperature than the average daily temperature
• Human bias in reporting shortages rather than overages
• Human bias in reporting lower volume and incidents of spills.
4.2.2.4.2 Random Uncertainty or Inaccuracy in Loss Measurement
Uncertainties in petroleum loss measurements exist primarily because
of metering limitations such as meter accuracies and calibration errors.
However, other factors also contribute to this uncertainty. These include:
aThese loss indications are described in more detail for the appropriate
factor that contributes to measurement error in Section 4.3.
46
-------
o
c
o
J3
s-
(O
o
o
s-
TJ
O)
(O
o
en •
c in
C O)
O) U
i — C
-Q CU
C (U
r—
O) O
•i- CU
4J -O
(O 3
10 .c
O) 4J
»r-
S- X
o
to
cu
3
O
cu
3
D1
(IdV 030) 33N3«3dJIQ
47
-------
• Variable conditions that affect meter performance
• Accounting errors
• Human errors
• Gravitation between tanks
• Inaccuracies in estimating the volume in the supply lines
• Physical changes in the volume of the fluids
• Changes in fluid between receipt and delivery.
4.3 MEASUREMENT
Accurate measurement of petroleum including quantities lost in trans-
fer and storage is extremely important for both cost and environmental
considerations. However, accurate measurements are difficult to obtain.
The petroleum industry is currently in the process of improving measure-
ments. Studies are being carried out in areas such as metering accuracies
and techniques, expansion and loss tables and equations for computing
evaporation losses. In spite of the current inaccuracies and errors in
petroleum measurement, a significant amount of information on petroleum
losses can be obtained from commonly used metering and accounting techni-
ques. Furthermore, this information can be used to estimate losses
nationwide.
Petroleum must be accurately measured to quantitatively determine
either total losses or the total loss for a particular loss mechanism.
Under ideal conditions, i.e., 100 percent pure petroleum; nominal varia-
tions in flow, and identical metering at each end of the pipeline system,
receipt and delivery meters should provide identical readings. This would
allow simple accounting procedures to determine overage and shortage re-
sulting from either the total loss or a specific loss mechanism. This is
not the case, however, and various factors such as:
• Meter errors
• Meter proving errors
• Variable conditions that affect meter performance
• Inaccurate correction factors
• Accounting errors
• Human errors and bias
contribute to an overall error in the measurement of petroleum. Addition-
ally, changes in the fluid between receipt and delivery that cannot or are
not measured, also contribute to the overall error.
On a short-term basis (monthly) or for a particular pipeline, these
factors all contribute to some degree to measurement error, and an accurate
determination of the true losses is extremely difficult using existing
48
-------
measurement and accounting techniques. On a long-term basis (yearly) or
for a large number of barrels of throughput, i.e., yearly throughput of
petroleum transported interstate, most of these factors would.follow ran-
dom normal distribution and the error would average out to almost zero.
However, a number of factors that-contribute to measurement errors are
still present such as:
• Coefficient of expansion
• Admixture shrinkage
• Entrained free air in the lines
• Back pressure errors in meters
• Prover temperatures that are higher than the daily
average temperature
• Evaporation
• Bias in reporting shortages rather than overages
• Measurement uncertainty.
Measurement techniques and errors will be reviewed to ascertain the
accuracy and limitations of measurements for quantifying petroleum losses.
Techniques used in industry for both metering and accounting of petroleum
will be reviewed and capabilities of these techniques evaluated. Then
metering, meter proving, variable conditions that affect meter performance,
correction factors, accounting, personnel reports (human errors and bias)
and false indications are discussed in separate subsections. Changes in
the fluid between receipt and delivery are discussed in Section 4.4.
4.3.1 Measurement Techniques—Metering and Accounting
Petroleum losses, for the vast majority of U.S. pipelines, are measured
with standard custody transfer systems at receipt and delivery. Figures 14
through 16 show the general arrangement of the commonly used Lease Automatic
Custody Transfer System (LACT)16. Positive displacement meters (PD)a or
turbine meters (TM) are normally used to measure the volume of petroleum
that is received and delivered. These meters provide a display of gross
barrels. In most systems, temperature changes of the petroleum due to
changes in temperature barrels are also displayed on the meter. These
values are either automatically or manually recorded on meter run tickets.
Measurements of BS&W (bottom sediment and water content) are made separately
with a sampler system, and net barrels are further corrected by this amount
aFor brevity, metering errors for only PD meters (most commonly used meter)
will be discussed in detail in this section to demonstrate the error
problem. See also Section 4.5.
49
-------
(A) PRODUCTION SUMCC WITH TWO MCASUKINC TANKS
.VCNT
'Flu. VACVt VALVt
ON BOTH
(•I PRODUCTION SUftCC WITH UCA5IMING TANK 1 MPEUNC SUM*
CC}TWO MCA9UHINC TANKS WITH
(A) Continuous Delivery to Pipeline.
OCUVCHT VM.VC
CAI OOtJHf MCASUHIN6 TANK-NO MMGC
OCUVtKT VACVt
(•I SIMOU SJKCC WITH VMCU MCASUnMG TANK
(B) Intermittent Delivery to Pipeline.
Figure 14. Typical system for delivery to pipeline.
(Source: Reference 16.)
50
-------
PD
en
SURGE
TANK
h-O
MONITOR
PROBE
-0
°
3-WAY
DIVERTER
VALVE
O
TRANSFER
PUMP
STRAINER
HXI-
RECIRCULATING
PUMP
SAMPLER
BACK-PRESSURE CONTROL
(ALTERNATE LOCATIONS)
TO PROVER
TO DEHYDRATION FACILITY OR WET-OIL STORAGE
Figure 15. Typical PD meter LACT unit flow diagram. (Source: Reference 16.)
-------
LIQUID INTAKE
LIQUID TRANSITION
LIQUID OUTLET
(A) Brooks-Birotor Meter Principle of Operational Diagram
OUTER
HOUSING
CAM
STATIC
LIQUID
BUOES
PATH
OF BUOES
ROWING
LIQUID
MEASURING
CHAMBER
INNER
UNIT
HOUSING
BLADE
BEARING
ROTOR
(B) Smith Meter-Principle of Operational Diagram
Figure 16. PD meters. (Source: References 17 and 18.)
52
-------
on the meter run ticket. Meter prover systems are used to periodically
prove (calibrate) the meters. Measuring-tank LACT (Figure 14) systems
are used for a small percentage of the petroleum transported. Tanks or
special vessels are the measuring devices in the automatic transfer of
oil from leased to transporting carriers. Tank levels are manually read
and then recorded on meter run tickets.
Petroleum losses can be determined from periodic monitoring of over-
ages and shortages using standard petroleum accounting procedures. Over-
ages and shortages are obtained by taking the difference between meter
tickets for net barrels (corrected for temperature and BS&W) received into
and delivered out of the system and correcting for changes in system in-
ventory. For short-term monitoring (daily, weekly, monthly), this method
may be used to determine if large losses exist. For long-term monitoring
(yearly), small losses and in some instances the specific loss mechanism
can be determined. Generally, most losses (shortages) are considered to
be accounting errors but are normally investigated if the shortage exceeds
0.1 percent of throughput.
4.3.2 Metering Errors
4.3.2.1 Positive Displacement Meters-
Positive displacement meters are the most commonly used means of measur-
ing flow in custody transfer systems. Two slightly different types of PD
meters that are widely used in the petroleum industry will be briefly de-
scribed.
A Brooks Instrument Division device, Figure 16(A), consists of two
aluminum spiral fluted rotors within a measuring chamber dynamically
balanced, but hydraulically unbalanced. As the product enters the intake
of the measuring unit chamber, the two rotors divide the product into pre-
cise segments of volume momentarily and then return these segments to the
outlet of the measuring unit chamber. During this "liquid transition,"
the rotation of the two rotors is directly proportional to the flow rate
of liquid throughput. A gear train located outside the measuring unit
chamber conveys mechanical rotation of the rotors to a mechanical or
electronic register for gross totalization of liquid throughput.
A device by Smith Meter Systems, Figure 16(B), consists of a rotor
with four evenly spaced slots that revolves on stainless steel bearings.
The slots control the position of two blades that are at right angles to
each other. As liquid flows through the meter, the rotor and blades re-
volve around a fixed cam. Ball bearings fixed to the blades roll around
the cam, causing the blades to move radially. The successive movement
of the blades, outward toward the case wall, forms a measuring chamber of
precise volume between the blades, the rotor, the case wall, the cover,
and the bottom of the case. A continuous series of these closed chambers
is produced, four for each revolution of the rotor. Neither the blades
nor the rotor contact the stationary walls of the measuring chamber.
53
-------
In addition to the gross totalization function, PD meters normally
include an accessory for automatic temperature compensation to provide
net registration.3 Net volume or net registration differs from the gross
metered volume depending on the average liquid temperature and base tem-
perature and the coefficient of expansion of the liquid. One typical
temperature compensator system consists of the following: (1) bulb (that
converts temperature change to linear motion) installed in the motor hous-
ing and exposed to the metered liquid; (2) a gravity selector (coefficient
of expansion selector); (3) a main gearing system; (4) a secondary gearing
system that includes an integrator; and (5) a median/leal register for net
totalization of liquid throughput. Net registration operates continuously
regardless of flow rate and temperature variations.
Assuming manufacturer recommendations on operating conditions, ranges,
etc., are followed, typical performance specification of these meters are:
• Linearity ±0.1 percent over rated flow range
• Accuracy ± 0.25 percent over flow range
±0.2 percent over 5 to 1 flow range
• Repeatability ±0.02 percent over flow range
• Temperature ± 1°F over temperature range.
Errors in PD meters can result from a variety of sources. Some of these
are:
• Slippage - passage of unmetered fluid through the internal
tolerances (clearance between rotating piece and stationary
piece so no metal contact) or the meter
• Rotor drag - changes of rotor speed caused by internal or
external loading effects
• Measurement volume changes
• Temperature compensator nonlinearity and hysteresis
• Component damage or malfunctions
• Meter wear.
The effect on the performance of the meters depends on these sources and on
a variety of conditions such as operating ranges, fluids and environment.
If the meters are regularly proved (calibrated) with line fluids at typical
line operating conditions, the performance can be improved greatly and
aNet registration is defined in the U.S. as liquid volume at a referenced
base temperature of 60°F.
54
-------
effects of error sources can be significantly reduced. Proving is normally
done at periodic intervals, i.e., weekly or bi-weekly, for custody transfer
systems and accuracies of about ±0.02 percent to ± 0.05 percent are
routinely achieved. In many cases, however, operating conditions may be
changed and meters not proved for these conditions, i.e., line fluids may
be changed or significant variations of line operating conditions may
occur. Even in the case of a meter without defects this may result in
accuracies of ± 0.25 percent or worse.
Discussions with petroleum industry personnel indicated that one of
the main complaints with PD meters was their poor reliability. This is a
problem even though most meters have an automatic shudown that occurs if
the meters are defective and do not register a certain amount of through-
put. Malfunctions can result in a variety of errors. For example, the
main components that malfunction are the stacks which contain the tempera-
ture compensator, printing head and other mechanical parts. The tempera-
ture compensator unit is considered to be the weakest link in the stack
and in the PO meter. This component requires high maintenance and malfunc-
tions cause a substantial error in the measurement. For example, a mea-
surement error of only 1°F is equivalent to a volume error of about 0.04
percent for a typical crude.
4.3.2.2 Tank Gauging—
Tank gauging is less accurate than PD or turbine meters. It is a
standard method for metering in custody transfer, but it is used to mea-
sure less than 10 percent of the petroleum transferred. Thus, tank gauging
errors, although quite high, do not significantly affect the ability to
accurately determine nationwide petroleum losses.
A variety of errors affect the overall measurement and limit the ac-
curacy to about ± 0.5 percent. These errors include:
• Measurement errors - inability to measure accurately such
things as height of liquid, container size and average
temperature.
• Gravitation between tanks - leakage in a valve between two
tanks can cause a loss of product in one tank and a gain in
another that may not be observed.
4.3.2.3 BS&W Metering--
Bottom sediment and water content metering is very inaccurate and this
has not been improved upon for many years. This is due to two main prob-
lems. First, the sample is collected and put into a centrifuge which can
be read out only at increments of 0.05, 0.10, 0.15, 0.2, 0.3, 0.4, 0.5 and
so on. Since one must typically read in increments of 0.1, a roundoff
error of 0.1 percent is possible for each measurement. Secondly, there
is a large sampling error in BS&W primarily because three sampling steps
are required for a sampling. The steps are as follows: (1) take sample
55
-------
out of line; (2) take sample from the original sample; and (3) take a
third sample for the centrifuge. Each sampling step is subject to being
non-representative. Also, for low transport velocities19 the original
sample taken out of the^line is affected because of non-uniform distributed
dispersion. In theory, BS&W sampling errors are purely random and should
not contribute to an error over a large period of time (1 year). However,
on a daily or monthly basis, the error can be substantial. On an individual
line, there can be a significant error problem because the BS&W on one end
of the line might not agree with the BS&W on the other end.
BS&W measurement, in general, is very inaccurate and is considered
the weakest link in the measurement of petroleum at receipt and delivery.
BS&W content in delivered petroleum ranges between 1 and 2 percent of the
total volume. Since BS&W readings are required in the correction for net
barrels, there can be significant errors in the net barrels correction.
This error occurs regardless of the other metering accuracies achieved.
BS&W errors can be evaluated on individual lines by comparing the BS&W
reading at each end of the line for separate batches.
Because of BS&W inaccuracies, leak detection systems (see Section 5)
that monitor gross barrels can be much more accurate than comparable sys-
tems that measure net barrels.
4.3.3 Meter Proving Errors
Performance in this report primarily pertains to the accuracy or re-
peatability of a measurement for a specific meter sensitivity or meter
factor. This is determined originally by manufacturer calibrations and
after installation by user provings (calibrations) at one or more flow
conditions for the fluid transported. Normally, meter proving is done
near the operating conditions for the fluid transported during the day on
a weekly or bi-weekly basis and a single meter factor is used.
Meter proving over the range of variable conditions for the fluid
transported and then correcting the data (ideally by computer or by manual
conversion of a graph or meter factor curve) is often done. This accounts
for some of the variable effects on the meters and significantly reduces
errors. Even if this is done, some errors may not be corrected. For ex-
ample, external temperatures may vary greatly over a 24-hour or weekly
time interval and it would be extremely difficult to account for these
variations.
4.3.4 Variable Conditions that Affect Meter Performance
There are many variable conditions that could affect the performance
of PD meters. These include:
• Meter wear, damage, or internal deposits
• Flow rate
• Viscosity
56
-------
• Temperature
- Liquid
- External
- - Liquid expansion
- Vaporization
- Temperature and viscosity
• - Lubricating qualities of the liquid
• Liquid pressure
• Liquid density
• Entrained free air
• Torque
• Back pressure
• Other.
Effects of many of these variable conditions cannot or are not normally
accounted for and a significant error band is generally present.
Each of the variables identified are discussed briefly and an esti-
mation of the errors given in instances where the errors are significant
and/or the qualitative effects are known. Also, error sources are iden-
tified that do not follow a random normal distribution.
4.3.4.1 Meter Wear, Damage, Internal Deposits--
Meter wear, internal damage and internal deposits can change the
volume of the PD measurement chamber and affect slippage. Thus, incorrect
volume measurements and unknown amounts of liquid may pass through the
meter unmeasured. Foreign matter from sources such as solids in the bot-
tom of the tanks or in the liquid itself is particularly harmful to the
meter. Normal change from meter wear between proving may amount to 0.1
percent.
4.3.4.2 Flow Rate-
Variations in the flow rate have a significant effect on the error
over the flow range of the meter. This can be seen, for example, in
Figure 17 where a typical accuracy curve is given for a PD meter.
This error is mainly due to slippage. It is less repeatable and inac-
curate at the flow rates because of a capillary seal effect. This effect
is usually minimized by operating the meter at the medium and high ranges
and re-proving the meter as conditions change. Meter factor may be used
from a curve that has been developed rather than reproving.
57
-------
t +0.2.
S +o.r.
£ 0
°- -0.1
1 "°-2
a:
ti j
f-
.
_ — •
— ,
i i i i i i i i i
U.S.GPM 50 100 150 200 250 300 350 4Qn ^e SO
FLOW RATE
Figure 17. Typical accuracy curve for a PD meter.
4.3.4.3 Viscosity-
Variations in viscosity affect the meter performance by changing the
slippage of the liquid through the meter. Viscosity changes cause either
thickening or thinning of the liquid and this affects the volume that can
slip through the tolerances of the meter unmeasured. For example, as the
temperature of a liquid increases, liquid thinning occurs and more fluid
can slip through the meter. Viscosity variations can occur from changes
in the liquid, in the temperature, or the liquid and the temperature.
Re-proving the meter when a fluid change occurs is the best way to correct
for this error although plots of meter factors as a function of viscosity
are often used. Viscosity effects are almost negligible for light products.
With a heavy crude, such as a number 6, viscosity variations can result in
a small error.
4.3.4.4 Temperature--
Variations in temperature have the most significant effect on meter
performance and cause various errors. These variations occur primarily
from the following:
• Liquid temperatures
• External temperatures
• Liquid expansion
• Liquid vaporization
• Viscosity.
The combinations of the effects of many of these variables in many instances
are inseparable. This can cause significant errors that cannot be detected
or corrected even by frequent proving.
Liquid Temperature—With liquid temperature variations, thermal expan-
sions or contractions occur in the meters that are not necessarily linear
or precisely known. This is caused by dissimilar or different materials
and results in mechanical clearances that vary. Both the volume of the
measuring chamber and slippage can be affected. Proving the meter at or
near the operating temperature of liquid helps to minimize this error.
58
-------
Typically, this error would be in the same direction for similar custody
transfer PD meters and receipt and deli very-and may not significantly af-
fect overages and shortages. However, the magnitude of this erro'r in-
creases and may not be in the same direction (affecting overages and
shortages) for very dissimilar or different internal meter materials.
This occurs, for example, when different types or sizes of meters are
used at receipt and delivery.
External Temperature—External temperature is conveyed through the
metal to the meter measuring elements. These temperatures cause thermal
expansion or contraction of the meter by an external temperature gradient.
At the same time, a different internal temperature gradient from the liquid
also causes thermal expansion or contraction. (See previous subsection.)
For example, a higher external temperature (120°F) causes thermal expan-
sion of the meter while a lower internal fluid temperature (70°F) causes
contraction. Both external and fluid temperature measurements are re-
quired for accurate meter factors to determine the volume of the fluid.
This is not normally done. Furthermore, meter proving is usually done
only during the day when the external temperature is always higher than
the average temperature for a 24-hour period. Thus, the effects of ex-
ternal temperature typically are not known and not corrected for in meter
proving.
Liquid Expansion—Temperature variations of the liquid cause varia-
tions in volume that depend upon the characteristic thermal coefficient
of the liquid. This variation has the most important effect on meter per-
formance of any of the variable conditions. This is because an error of
only 1°F is equivalent to a volume error of about 0.04 percent for a
typical crude. Proving is usually done close to the temperature of the
fluid transported and this tends to reduce measurement errors to small
errors that are of a random normal distribution.
The temperature bulb of the temperature compensator unit is not cali-
brated separately because it is not a significant source of error as long
as the meter is proved at the temperature at which you are operating.
Periodically, it is removed and installed in a temperature bath and cali-
brated at true temperature. If large temperature variations (> 10°F) oc-
cur between meter provings, use of a single meter factor may result in a
substantial error, i.e., about ± 1.125°F or a ± 0.05 percent volume error
for each PD meter. If, instead of the temperature compensator unit, a
manual reading of temperature is carried out, an arithmetic correction is
applied and an even larger error can be expected.
Temperature and Viscosity—Variations of external temperature, liquid
temperature and viscosity of fluids result in overall effects on the meter
performance that are difficult to determine and in many cases inseparable.
The following example will demonstrate the problem. If the fluid tempera-
ture is increased, the viscosity of the fluid will decrease. The fluid
will thin down and will increase squirm through the meter (slippage) and
not be measured at the same time; the clearances in the meter will de-
crease. Thus the fluid attempts to pass through the meter, i.e., because
59
-------
of the lighter fluid and increased slippage, and at the same time the blade
clearances expand and become tighter and tend to reduce slippage. These
effects compensate one another so as to reduce slippage. The overall ef-
fects of these variations are dependent on the particular meter because one
meter may expand at a different rate than another primarily because of dif-
ferent materials.
4.3.4.5 Liquid Pressure--
Meter performance is affected only by changes in operating pressures
that are high enough to affect the dimensions of the meter by elastic de-
formation. A 100 psi pressure change causes an error of about 0.04 per-
cent. Assuming typical operating conditions, the pressure has to be off
a great deal before this variation becomes significant. Since crudes have
considerably higher viscosity than products, the accuracy of meters in
crude service normally varies less with flow rate than metered products.
If pressure changes are significant, the measurement ticket value must be
corrected by the multiplication factor Cpl.
Pressure pulsations and excessive surges as well as excessive pres-
sures caused by thermal expansion can cause errors in the meter reading.
Normally, surge tank, expansion chambers, relief valves or other protec-
tive devices are used to minimize these problems.
4.3.4.6 Entrained Free Air-
Entrained free air in the liquid occurs mainly at unloading facilities
and can affect the drag of the meter rotor. As the air strikes the measur-
ing elements, the air will be metered and the metered air will momentarily
speed up the meter. This occurs because there is less resistance to air
moving through the meter than the liquid and results in higher readings be-
cause of the speed-up of the meter and the measurement of air.-
4.3.4.7 Back Pressure—
If back pressure is off, the meter should read low. However, the
value of this error is difficult to determine and is, in fact, debated as
an error source. However, on turbine meters back pressure is a signifi-
cant source of error. A formula for minimum back pressure is recommended
by API to correct for back pressure variations.
4.3.5 Inaccurate Correction Factors
Inaccurate correction factors can cause errors in a number of measure-
ments that can affect petroleum loss values. The most significant one is
the common use of an inaccurate coefficient of expansion a for the liquid.
4.3.5.1 Coefficient of Expansion a—
Field measured volumes are corrected for expansion from ambient tem-
perature at the measurement point to the volume at 60°F and 14.7 psia using
ASTIM-IP20 Petroleum Measurement Tables. These tables give densities linear
60
-------
with temperature. This is based on the value of the expansion coefficient
a, which is assumed to have a linear relationship a = l/V(dV/dT)p, that is
a function of standard density at 60°F and 14.7 psia. This simple rela-
tionship of a to temperature and standard density was shown to be incor-
rect by Kurtz21. Downer and Inkley22 made coefficient of expansion measure-
ments for crude that showed a values of the 1952 ASTIM-IP tables to be about
8 percent low. However, products were fairly representative of the curve
given from the ASTIM-IP tables. Figure 18(A) shows the expansion coeffi-
cient of the curve from ASTIM-IP tables and those measured for crude oil
and products. Figure 18(B) shows the expected calculation error for a 1°F
temperature range.
Use of inaccurate coefficient of expansion tables causes a very sub-
stantial hypothetical loss to occur for petroleum transfer. This is be-
cause the average temperature before transfer (receipt) is higher than the
average temperature after transfer (delivery). Both volume measurements
are currently corrected using the incorrect coefficient in the ASTIM-IP
tables and a "hypothetical" or "apparent" loss results.
4.3.6 Accounting Errors
Petroleum accounting provides records of receipts, deliverables, in-
ventories, overages and shortages, etc., that are quite useful in estimat-
ing overall system losses and establishing the possible existence of leaks.
Accurate records of losses or leaks appear to be simple to obtain, but in
actual practice various error sources make this quite difficult to achieve.
A few of the accounting procedures that are normally carried out include:
• Receipt and Delivery - Periodic totalizing meter tickets for
net barrels received into and delivered out of the system.
• Inventory - Periodic totalizing of tank gauging records and
line capacity.
• Overages and Shortages - Periodic computing of the difference
between receipt and delivery tickets and then correcting for
inventory changes.
In practice, these procedures often cannot be carried out without the pos-
sibility of numerous errors (usually small) that can make daily, weekly or
monthly accounting quite inaccurate. Over a long period of time, however,
most accounting errors would be expected to follow a random normal distri-
bution and average out close to zero. Yearly accounting would be expected
to be an order of magnitude more accurate than monthly accounting. Sources
of accounting errors are discussed in the subsections that follow.
4.3.6.1 Inability to Make all Measurements at the Same Time-
All meters (receipt, delivery, tank gauges) in a system are normally
read every ten days and/or the last day of each month. Since it is almost
impossible to read all meters at the exact scheduled time, significant
61
-------
Mean coefficient of Expansion
"5 LL.
O
+-> ~~»
CO
U) O
T- <— 1
(J
£s
>+- c
O) O
O •!-
O l/l
c
C (Q
(O O.
O) X
ZI uu
70S
608
508
408
\ o Crude oils
\ J Fuel oils
-•\ Other products
^~ ASTM at 60°-100°F
•
Vox" Pr°P°sed 60°-100°F
— ^\£ curve for crude oils
•OQ
• QJ|p
iCL ^-ASTM at 140°-170°F
~. , X,
0.7 0.8 0.9
Sp Gr 60°/60°F
(a) Expansion Coefficients—Table ASTM-IP and Measured
0.006
-a
O) 0)
•£ » 0.005
fo C.
r— fC
|^ 0.004
-------
errors result. For example, if a custody transfer meter that is measuring
1000 bblh is read 15 minutes late an error of 250 bbl would occur,
4.3.6.2 Damaged Meter—
If a damaged meter is found, the meter is assumed to have gone bad be-
tween the last time the meter was read or proved. This can present a
large error for short-term accounting, particularly for meters that monitor
large throughput.
4.3.6.3 Pipeline System Complexity-
Petroleum pipeline systems in the U.S. vary from simple single line
systems to very complex multiple systems. The latter systems often have
numerous inputs and metering locations, use different lines and transport
different fluids. It is extremely difficult to implement error-free ac-
counting on these complex systems, particularly over a short time, i.e.,
weekly, monthly.
4.3.6.4 Tank and Line Inventory--
Measurement of the inventory in pipelines and tanks is typically quite
good and the relative error quite small because the line is usually full
and the volume does not change appreciably. Tank gauges are inaccurate
(about ± 1/2 percent) but tanks usually contain only a small portion of
the petroleum of throughput (< 3 percent). Thus if a worst case reading er-
ror of 1 percent on all tanks in the U.S. were five billion barrels, inven-
tory would be about 150 million barrels. Thus a 1 percent reading error
of the tank gauges would result in an inventory error of about 1,500,000
barrels or a system error of 0.03 percent. Over a long period of time,
however, most gauging errors would be expected to follow random normal
distribution and average out close to zero.
4.3.6.5 Personnel Errors—
(See Section 4.3.7.)
4.3.7 Human Errors and Bias
Human errors and bias can contribute to both the overall accounting
errors and actual errors in spill reporting. These errors cannot be es-
timated quantitatively and range from mistakes in reading meters to in-
correct inputs to the computer.
Gravitation between storage tanks, for example, can result in loss
of product from one tank and a gain in another if valves between any two
tanks leak. Loss or gain in a tank may not be detected (see also Section
4.3.6.4).
63
-------
Some of the most important errors are bias errors in reporting petro-
leum losses. Two of these are particularly significant. First, there
appears to be a tendency to report shortages rather than overages because
overages are difficult to explain. Second, there is a tendency to under-
estimate the size of small spills, particularly those close to the minimum
reportable size. This latter error exists primarily because of the extreme
difficulty in accurately estimating the actual spill size. This error can
result in reports of the incidents and volume of spills that are too low.
4.4 CHANGES IN PETROLEUM BETWEEN RECEIPT AND DELIVERY
Changes in petroleum can occur between receipt and delivery that may
not or cannot be correctly measured by normal custody transfer metering.
This creates problems in determining what is being measured, particularly
for long lines that are of large diameter. These changes can result in
significant errors in volumetric measurement particularly over short periods
of time, i.e., days, weeks.
4.4.1 Fluid Heating
Fluids at receipt in storage tanks are usually at a higher average
temperature than at the delivery point. This temperature difference is
estimated to be about 10°F. The major effect of this temperature difference
is in the error that results from use of the inaccurate coefficient of ex-
pansion a. A 10°F difference for crude, for example, would result in a
0.04 percent error in the measured volume. Also, petroleum entering the
pipelines and cooling before being taken out can shrink. For example,
cooling of 5°F can cause shrinkage of approximately 0.25 percent in volume
and result in substantial errors if temperature measurements at either end
of the line are in error. The temperature difference between fluids at
each end of the line may be quite different for an individual line or lines
in certain areas.
Petroleum heats up and expands as it is transported because the energy
put into the pipeline is transferred to the petroleum. This heating effect
depends upon a number of factors such as flow rate, pumping and pipeline
dimensions. If the pipeline is operated at a high rate, friction builds up.
Crowding of the pipeline (working at capacity or above) also tends to gen-
erate a large amount of heating. These effects become more pronounced as
the size of the line increases because the large pipeline mass transfers
off less of the heat that is generated.
Effects of external temperatures and internal fluid heating during
transport are difficult to estimate and measure for individual lines.
Thus, differences in fluid temperatures at each end of the line present
difficulties in accurately measuring the true volume of the fluid.
64
-------
4.4.2 Mater Settling
If BS&W is high, i.e., 1 to 2 percent or more, water tends to settle
at the bottom of the line. This can result in a significant difference in
the BS&W at each end of the line and result in a larger error in net bar-
rels. As noted previously, at low transport velocities there is a non-
uniform distributed dispersion of water, solid particles and hydrocarbons
which affect BS&W sampling accuracy.
4.4.3 Admixture Shrinkage
(See Section 4.2.2.3.)
4.4.4 Miscellaneous
Other less significant changes that can occur include:
• Cavitation
• Viscosity changes due to temperature
• Vaporization if line pressure goes too low.
4.5 TOTAL MEASUREMENT SYSTEM ERROR
The measurement accuracy for individual pipeline systems must include
the combined effect of the various measurement errors (see Section 4.3)
and fluid changes between receipt and delivery (see Section 6.4). This
information is necessary in determining the minimal spill loss that can
be detected. Means of obtaining a value for the measurement accuracy are
presented in Section 4.5.1. An estimate of the minimal spill loss that
can be detected is presented in Section 4.5.2 for a simple line.
4.5.1 Means of Computing Measurement System Error
In general, the total system error will be greater than any of its
individual components. Hence, it is important to account for significant
errors of individual components. This can be done as follows: First, es-
timate the error (in standard deviations) of each individual component
(Ec), i.e., flow meters, pressure transducers, etc. This is done by com-
puting the square root of the sum of the squares of the individual com-
ponents
where 0]_ ... an = errors of individual components. Then computing the
total systems error (Et) by taking the square root of the sum of the
square of individual error component,
65
-------
4.5.2 Computation of Minimal Detectable Spill Loss for a Simple
Pipeline System
Estimates of the overall measurement error, which consists of the com-
bined errors of typical metering (leak detection equipment, etc.), line
fluctuations and other items, are necessary in order to determine the mini-
mal spill loss (in barrels) that can be detected. Assuming the simple
line and instrumentation identified in Figure 19(A), the total system er-
ror is computed in Figure 19(B).
4.6 LOSS STATISTICS
Results on measurement and fluid changes in Sections 4.3 and 4.4 in-
dicate that total losses and individual loss mechanisms potentially can be
determined for the U.S. petroleum industry from existing loss statistics.
However, system inaccuracies caused by both limitation and errors in mea-
surements and unaccounted for fluid changes are inherently present and
must be considered. In cases where there is a large throughput and many
measurements are taken over a large period of time, most random normal
errors tend to be negligible and reasonable loss estimates can be obtained.
In this section, total petroleum losses and specific loss mechanisms
are studied based on nationwide statistics. Information is obtained pri-
marily from U.S. Government agencies and programs, foreign government
agencies,2 major oil companies, and studies of statistics of spills and
other losses.
Statistics on overall total losses are reviewed in Section 4.6.1.
Data in this section indicate that substantial losses do exist. Loss
statistics on the specific loss mechanisms of oil spills and factors af- .
fecting these losses are discussed in detail in Section 4..6.2. Causes
and factors relating to failures of various pipeline systems components
are studied.
4.6.1 Total Losses
No published reports, that are available, directly estimate total
petroleum losses in barrels. Petroleum information reported by oil com-
panies and published in statistics by the U.S. Government is in barrels
received into and delivered out of the petroleum transportation systems.
This information does not specifically identify shortages or losses in
barrels. However, indirect estimates of these losses can be made from
available U.S. Government statistics.
aForeign statistics will also be reviewed. In some countries, pipeline
regulations are much more stringent than in the U.S. (See Appendix A.2).
Also pipeline age, installation, etc., are not typical of those in the
U.S. Thus, spill statistics may differ from those in the U.S.
66
-------
35
30
25
20
CO
CD
15
C/5
-------
Also, data on petroleum loss allowances or tariff deductions (in per-
cent of throughput) published by U.S. companies to cover actual pipeline
system transportation losses can be used to directly estimate total losses.
4.6.1.1 ICC Statistics-
Total petroleum losses can be estimated from ICC transport statistics. 6
Table 16 shows ICC statistics for the total barrels of crude and product
received into and delivered out of U.S. pipeline systems (pipeline, stor-
age tanks, etc.) engaged in the interstate transportation of petroleum for
the 12-year period between 1964 and 1975. Data are included here for ap-
proximately 170,000 miles of pipeline and an average trip length of about
300 miles per barrel. Petroleum losses are obtained by taking the differ-
ence between the received and delivered barrels. Average losses were rela-
tively consistent and, except for 1965, ranged from 0.17 to 0.30 percent
of barrels received into the system between 1964 and 1976. An average
loss of 0.21 percent (standard deviation 0.07 percent) occurred over this
period. For the year 1975, this average loss would amount to about 20
million barrels of oil. These losses in Table 16 can only be used as es-
timates, particularly on a yearly basis, primarily because the inventory
is not included in these ICC statistics. Assuming inventory changes gen-
erally vary in a purely random way (see Section 4.1.3.2), it would be ex-
pected to average close to zero over a long time period.
Unpublished ICC information23 on overages and shortages in the dollar
cost of oil is reported by individual companies. These statistics are in-
dicative of substantial losses of petroleum. Unfortunately, these over-
ages and shortages are given in dollars and the barrel cost is not provided.
Furthermore, for some companies overages and shortages reported may not
show a loss even if an actual loss exists. This is due to different re-
porting companies. For example, some companies assume a certain percentage
loss allowance between receipt and delivery and this is included as part of
the normal operating expense. If the actual loss does not exceed the set
loss allowance, no shortages are reported. Other companies report the ac-
tual overages and shortages. Regardless of various shortcomings, the exist-
ing information provides a rough estimate of the cost of lost oil. A sub-
stantial effort would be required, however, to obtain an accurate estimate
of the cost of lost oil. Inquiries would have to be made to the individual
companies for reporting procedures and barrel costs and the data compiled.
Even then, it is uncertain that dollar costs could be converted to an ac-
curate estimate of barrels lost.
4.6.1.2 Petroleum Loss Allowance-Tariffs Published by U.S. Companies-
Tariffs by pipeline transportation companies are of sufficient accu-
racy that they can be used to determine total losses in pipeline transpor-
tation systems. Tariffs or petroleum loss allowances are charged by ship-
pers to cover all actual losses from evaporation, shrinkage, and all other
unavoidable losses during transportation, so that the shipper always has
enough oil on hand to deliver a certain amount of oil. The tariff is
68
-------
TABLE 16. THROUGHPUT AND LOSSES FROM ICC
TRANSPORT STATISTICS
Year
1964
1965
1966
1967
1968
1969
1970
1974
1975
Average
Standard
1964-1970
1974-1976
Number of Barrels
Received into System
5,575,851,696
5,867,582,598
6,251,704,887
6,813,740,133
7,289,434,872
7,755,292,825
8,163,531,445
9,359,988,000
9,787,030,000
Loss
Deviation
47,717,134,000
28,505,842,000
Number of Barrels
Delivered
Out of System
5,565,112,394
5,863,651,079
6,238,192,221
6,800,423,951
7,268,759,226
7,741,815,221
8,146,684,754
9,333,340,000
9,757,103,000
47,624,636,000
28,431,053,000
Loss
Barrels
10,739,302
3,931,519
13,512,666
13,316,182
20,675,646
13,477,604
16,846,691
26,648,000
29,927,000
92,498,000
74,800,000
Loss
%
0.19
0.07
0.22
0,20
0.28
0.17
0.21
0.28
0.30
0.21
0.07
0.19
0.03-
69
-------
quantity of oil deducted from the amount received for shipment and the
balance is the net quantity deliverable.
Loss allowance values are determined in a number of different ways
by pipeline companies. Some companies' charges are based on a previous
year's experience and their tariff may change yearly. Other companies
base their tariffs on long-term loss figures and other considerations.
Tariff information is provided by individual companies in their Rules
and Regulations of Transport of Liquids. The following example is a typi-
cal deduction charge by Laurel Pipeline Company in their rules and regula-
tion #8.
"Gauging, Metering, Testing and Deductions
Quantities of petroleum products shall be determined at the
Carrier's option by tank gauges and computations made from tank
tables showing 100 percent of the full capacity of the tanks or
by the Carrier's positive displacement meters. Such quantities
shall be corrected to a temperature of 60°F.
Shipper and consignee shall have the privilege of being
present or represented during gauging, metering reading, meter
calibration and testing of shipments by Carrier.
One-half percent of all petroleum products received for
transportation at point of origin shall be deducted and retained
by Carrier to cover losses due to shrinkage and evaporation in-
cident to pipeline transportation, and the net balance will be
the quantity deliverable to consignee."
Tariff deduction information is furnished to the ICC and is summarized for
individual companies in a private publication.21*
The tariff charge varies from one company to another. Typical rates
range from about 0.05 to 0.50 percent. Published tariffs for a typical
sampling of U.S. companies is given in Table 17. Shell pipeline, for
example, has a tariff charge of 0.15 percent for crude and Teromo's new
30-inch pipeline has a tariff of 0.1 percent. Currently, the average in-
dustry tariff rate for a sampling of interstate lines (Table 17) is about
0.22 percent3 (about 0.16 percent for crude and 0.33 percent for product)
of the petroleum received into the system. This amount is in reasonable
agreement with the 0.21 percent average loss from ICC statistics. In the
1960's, the loss allowance used was about 0.1 percent, but has been re-
vised upward based on recent shipper experience.
Excludes tariff rates (about 6 percent of companies) that are in excess
of 0.50 percent.
70
-------
TABLE 17. PETROLEUM LOSS ALLOWANCE FROM TYPICAL U.S. PIPELINE
COMPANIES IN PERCENT OF GROSS PETROLEUM SHIPPED
Crude
Company
Product
Loss Allowance
(*)
Company
Loss Allowance
Arapahue
ARCO
Belle Fourche
Buckeye
Butte
Chevron
Explorer
Pipeline Limited
Jawhawk
Keanai
Lakhead
Phillips
Osage
Ohio Oil Gathering
Sohio
Tomahawk
Texoma
Shell Pipeline
Pure Transportation
Williams
Wesco
0.25
0.20
0.20
0.25
0.15
0.25
0.25
0.10
0.05
0.10
0.05
0.10
0.25
0.05
0.05
0.10
0.15
0.10
0.50
0.10
0.20
0.10
0.15
0.25
0.10
0.20
Airforce Pipeline Inc.
American Petrofina
Amoco
ARCO
Buckeye
Chase Transportation
Gulf
Laurel
Pinto
Plantation
Phillips
Southern Pacific
Texas Pipeline Co.
Williams Pipeline
Gulf
Average Tariff
0.50
0.20
0.50 .
0.50
0.10
0.50
0.50
0.25
0.125
0.25
0.25
0.25
0.25
0.50
0.33
Average Tariff
0.16
71
-------
4.6.1.3 BOM Statistics-
Bureau of Mines statistics3 on production, imports, exports, storage,
refinery and consumption in the U.S. includes a loss estimate of 0.1 per-
cent of the total crude supply for losses between supply and disposition.
The percentage, which agrees with the early ICC loss tariff, was chosen
based on an undocumented study in the early 1960's by BOM, API and others.
Individual loss mechanisms that make up the 0.1 percent value cannot be
identified.
BOM annual reports also provide statistics on total losses. After
careful and extensive review, it was determined that these statistics do
not lend themselves to analysis and were not considered further.-
4.6.2 Oil Spills
Statistics of reported oil spills on land and water in the United
States normally include spill location, source, cause, type of oil, vol-
umes discharged, company, etc., and are collected by the following four
Federal government agencies: (1) Office of Pipeline Safety and Opera-
tions; (2) Environmental Protection Agency (EPA); (3) United States Geo-
logical Survey (USGS); and (4) United States Coast Guard (USCG). Each
has a specific jurisdiction that in total encompasses essentially all
spills. In some instances, similar data are collected by more than one
agency. Also, jurisdiction of spills is uncertain in some cases.
In this study, spill statistics are separated into the following six
reporting areas:
(1) On!and pipelines for aboveground spills that are in excess
of 50 barrels and occur in a carrier's transportation sys-
tem.Includes inland spills that are on land and in water—
OPSO Department of Transportation.23>26
(2) Offshore pipelines carrying oil produced and owned by com-
panies that do not operate the pipeline. Statistics apply
only to transportation related pipelines—OPSO.
(3) Onshore and offshore facilities not related to transportation.
Includes inland pollution of water, streams, underground
water supplies, etc.—EPA collects information by a Spill
Prevention, Control and Counter-measure (SPCC) file that
was started in 1975.
(4) Inland spills that occur on Federal lands. Reporting sys-
tem began in 1974—USGS.27'28
(5) Federally controlled outer continental shelf. Reporting
system began about 1966 to provide information for regula-
tions for safe drilling and production practices in outer
continental shelf—USGS.
72
-------
(6) All spills from coastal, river and Great Lakes and all transpor-
tation-related facilities. Includes all offshore spills, no
matter what size, from all sources (including pipelines) out
to three miles and spills in excess of 50 barrels beyond three
miles. Also, oil spills that pose a threat to the water.
Pollution Incident Reporting Systems (PIRS) began in 1971 by
USCG.
PIRS data should include all OPSO-reported spills in the U.S. waters and
large spills reported by the USGS. Spill data for these areas and the
indicated Federal government sources were used for statistical analysis
of the spills in the United States.
For comparison purposes, oil spill statistics from various individual
states and some foreign countries were used. States include California,
Colorado, Louisiana, Mississippi, New Mexico, Oklahoma and Texas. These
sources are described in Appendix C of Reference 30. Foreign sources in-
clude spill statistics for cross-country pipelines in Europe31'37 by the
Stitching CONCAVE and in Canada38 by the Task Force on northern oil de-
velopment.
Caution must be exercised in using any of these statistics because a
number of deficiencies and defects exist both in the reporting of spills
and in the compiling of spill reports by the various Government agencies.
Some of these are:
• Errors in reporting
• Non-compliance with required reporting
• Different reporting requirements by Government agencies,
such as:
- Spill size
- Spill details
- Exclusion of spill reports for some types of line
pipes or systems
• Difference in interpretation of spills by Government
agencies.
A recent report1 by the U.S. General Accounting Office on pipeline safety
identifies a number of deficiencies and weaknesses in spill reporting and
analyses. Deficiencies in spill reporting are also identified in Refer-
ence 40. These problems are considered in this study and attempts are
made to minimize potential errors by various means such as comparing sta-
tistics from more than one source or estimating unreported events.
In the spill analysis carried out in this section, statistical data
are corrected for errors such as non-reporting of spills, pipelines not
covered by regulations, and spill sizes not covered in the reporting.
Possible errors and unusual trends in U.S. spill statistics are discussed
73
-------
also. A number of pertinent studies (References 14, 30, 41, 42, 43, 44, 45,
46, 47, 48, 49, 50) on risks and on oil spill trends have been carried out
and pertinent results are presented in this section.
Oil spill statistics on the incidents of failures of pipeline system
components and the causes of failures in the United States are examined in
Sections 4.6.2.1 through 4.6.2.5. Factors relating to the number and fre-
quency of these incidents (pipeline geometry, age, etc.), volume and sever-
ity of incidents (fatalities, injuries, etc.) are also presented. Particu-
lar emphasis is placed on the OPSO and USGS statistics for analysis of pipe-
line system spills. Statistics presented in Section 4.1.3 such as pipeline
mileage in-service by category and throughput are used in the examination
of spill statistics for various pipeline parameters. Statistics on re-
ported cross-country oil spillages in Europe and Canada are examined in
Section 4.6.2.6 and compared with U.S. spillage.
Results from this section provide the basic information for the oil
spill risk analysis in Section 6. Based on statistical data presented in
this section and in Section 4.6.1, the total oil spillage from pipeline
systems (see Section 4.1.3) is estimated in Section 4.7.
4.6.2.1 Loss Mechanisms—Inland and Offshore Spills in the U.S. (OPSO
Statistics) —
Statistics from the Office of Pipeline Safety and Operations of re-
ported liquid pipeline transportation system accidents are available be-
cause all petroleum transportation companies are required by the Code of
Federal Regulations, Title 49, Part 1957 (Appendix A) to notify OPSO if
there is a reportable spill incident and to submit a report on DOT form
7000-1 (see Appendix A, Figures A-l and A-2). Incidents defined by Title
49, Part 195.50 that must be reported for any failure in a pipeline system
where there is a release of commodity transferred resulting in:
(a) Explosion or fire not intentionally set by the carrier.
(b) Loss of 50 or more barrels of liquid.
(c) Escape to the atmosphere of more than five barrels a day
of liquified gas.
(d) Death of any person.
(e) Bodily harm to any person resulting in one or more of the
f ol 1 owi ng:
(1) Loss of consciousness.
Necessity to carry the person from the scene.
Necessity for medical treatment.
(4) Disability which prevents the discharge of normal
duties or the pursuit of normal activities beyond
the day of the accident.
(f) Property damage of at least $1,000 to other than the carrier's
facilities, based upon actual cost or reliable estimates.
It should be noted, however, that besides the 50 barrel spill minimum,
OPSO spill data does not include all inland and offshore pipeline system
spills. That is, it applies only to transportation-related pipeline and
74
V * /
(2)
(3)
-------
does not include, for example, spills occurring at offshore production
facilities such as a gathering net. Additionally, many gathering lines
are operated that do not come under regulation. These include:
• Transportation through a pipeline by gravity.
• Transportation through pipelines that operate at a stress
level of 20 percent or less of the specified minimum yield
strength of the line pipe in the system.
• Except for Subpart B of this part, transportation of petroleum
in onshore pipelines in rural areas between a production facil-
ity and a carrier's trunk!ine reception point.
• Transportation in offshore pipelines which are located upstream
from the outlet flange of each facility on the Outer Continental
Shelf where hydrocarbons are produced or where produced hydro-
carbons are first separated, dehydrated, or otherwise processed,
whichever facility is farther downstream.
Reportable information includes date, time, location, carrier informa-
tion, density of area, origin of release of commodity, cause, death, injur-
ies, damage material, dollar value of property damage, commodity, installa-
tion year, pipeline size, barrels lost, and other pertinent line pipe in-
formation.
A review of the actual reported information indicates a number of de-
ficiencies and discrepancies, such as:
• Gathering and trunk pipelines cannot be distinguished for
most incidents.
• Production and transportation lines cannot be distinguished
for most incidents.
• Leak duration is often not reported.
• Volume of oil spilled cannot be accurately assessed.
• Submitted reports do not always include answers to all
questions; thus the number of incidents are not consistent
with some categories.
Also, there appear to be slight discrepancies in data as reported by
USCG and USGS for some of the same incidents.38 Additionally, OPSO pro-
vides pipeline mileage data for gas lines but not liquid pipelines. Thus
data obtained from BOM such as miles of pipe in service (see Section 4.1.3)
were used as an aid in the analysis. None of these deficiencies and dis-
crepancies were sufficient enough to affect analysis and the outcome of
the conclusions on petroleum spills.
In the sections that follow, spill statistics are first examined to
determine the overall magnitude of the spill problem, i.e., the number and
volume of reported spills. Then the pipeline system components that fail
most often, i.e., line pipe, are also studied. This latter analysis is
75
-------
carried out because results of an analysis based on data from the total
pipeline system alone, as is done in other studies, can be misleading
and can culminate in incorrect conclusions and ineffective recommenda-
tions. Finally, factors relating to the probability of spill incidents
and severity are presented. The number of incidents and volume spilled
are examined for the following:
• Type of fluid
• Mileage
• Throughput
• Pipeline diameter
• Depth of cover
• Fatalities and injuries
• Property damage.
These and other factors such as:
• Pipeline coating
• Corrosion control
• Procedures
• Areas where spill occurred
• Elapsed time
• Miscellaneous
are also discussed in Section 6.3. This same general approach is used in
the study of spills based on data from other sources. (See Sections
4.6.2.2 through 4.6.2.6). However, not all factors are examined because
either available data are not suitable for analysis or there is no need
for further analysis.
4.6.2.1.1 Overall Magnitude of Spill Problem—Frequency and Volume of Spills
A summary of statistics on spills from petroleum pipeline systems for
all accidents reported to OPSO for the 9-year period between 1968 and 1976
is presented in Table 18. More detailed statistics are given in Appendix B.
Data show that a significant number of incidents were reported and a sub-
stantial volume of petroleum lost. The average number and volume of spills
between 1968 and 1976 were 317 incidents and 346,007 barrels respectively;
this decreased to 280 incidents and 319,628 barrels for the more recent
5-year period between 1971 and 1975. Figure 20 shows a rapidly declining
year-by-year trend for spill incidents and a slightly declining trend for
spill volume for the period between 1968 and 1976.
4.6.2.1.2 Pipeline System Component Failures—Incidents and Volume Spilled
Statistics on spills from components of the petroleum transportation
system are presented in Table 19 and illustrated in Figures 21 and 22. Data
76
-------
500
I 400
2
^ 300
I
a 200
i/n
1968 1969 1970 1971 1972 1973 1974 1975 1976 Avg Avg
Year 1968- 1971-
1976 1975
(A) SPILL INCIDENTS VERSUS YEAR
600.000
~ 500,000
wt
1 300,000
e
3»
~ 200,000
f»
*
100,000
I
1968 1969 1970 1971 1972 1973 1974 1975 1976' Avg Avg
Year 1968- 1971-
1976 1975
(B) VOLUME SPILLED VERSUS YEAR
Figure 20. Spill incidents and volume versus
year for petroleum pipeline system accidents
reported to OPSO between 1968-1976.
77
-------
TABLE 18. SPILL STATISTICS FOR PIPELINE SYSTEMS TRANSPORTING LIQUIDS IN THE U.S.
FOR INCIDENTS REPORTED TO OPSO BETWEEN 1968-19763
00
216
of Sjllls
~is5 14* "iol75 72 "IF
17 30 22 25 14 15
15 20 22 24 If 21
18 18 6 10 10 16 10 27 15 1.146 16,153 16.273 16.266 6.296 11.891 13,474 20,981 5.866
5
58
102
U
I 5 14 14 » 2 t 4.919 5.535 2.500 4.860 31.79* 22.975 6.299 11.356 4.365
46 56 62 56 29 26 20 95.758 99.899 45,265 65.646 51.948 169.510 63.485 94.440 61.847
75 73 86 74 M II 76 114.5/8 96.40* 81.631 74.907 123,696 63,600 120.347 116.273 102.567
20 16 13 II H 21 11 76.167 24.325 262.611 18.178 13.070 4.153 63.781 1.621 15.014
external
Corrosion
Internal Cor-
rosion
Operational
Causes*
foutpmt Mal-
function or Fall-
Severe Matter*' 4
Construction Pe- 80
facts* or Mate-
rial Failure' .
Ulernal tapact* 106
Miscellaneous t 46
Unknown
IOIM. 499
External Cor-
corlon
Internal Cor-
rosion
Operational
Causes*
fqulpmt Mil-
function or Fall-
Severe Heather* O.I 1.2 0.3 1.6 4.5 5.1 3.5 0.6 2.* I.I
Construction Dt- 16.0 14.4 13.3 11.2 20.1 20.6 II.1 10.0 9.6 24.4
rects* or Nate-
el
• of Spills (tbls)
57 41 61.904 70.871 45.232 21.125 40.475 32.637 21.399 37.073' 31.954
If 10 7,863 6.726 15.616 12.445 36,988 11.465 3.901 2.853 3,249
22 20 13,237 22,687 32.641 22.830 46.185 63.144 20.957 26.619 40.155
403 347 308 301 273 256 260 201 312.511 343.611 521.841 245.057 360.654 371.365 213,643 319,423 255.037
Spills by Cause as Percent Spills >y Cause as Percent
1*68 1*6* 1*70 1971
43.3 31.5 42.9 33.1
3.2 4.2 l.f 7.1
1972 1973
24.3 26.4
I.I 5.1
1*74 1*75 1976
20.3 21. t 19.6
6.1 6.2 4.6
1966
17.1
2.0
t.l
1969
20.6
1.7
5.3
1970
6.7
3.0
3.1
1*71
12.2
5.1
6.6
1972
11.2
10.6
1.7
1973
6.6
3.0
3.1
1974
7.3
1.3
4.6
1975
11.6
0.9
6.6
1976
12.6
1.3
2.3
rial Failure'
External Iniuct* 21.2
Miscellaneous t 1.0
UnknoMi
25.3 21.6 23.7 27.1 27.1 35.2 31.2 37.1 29.2
6.2 6.1 5.6 4.2 4.0 1.0 11.t 1.1 11.4
l.t
2*. I
20.1
7.1
0.6
0.7
15.6
54.2
2.0
26.8
30.6
7.4
1.8
16.6
34.4
3.6
IOIM
1001 1001 lOOt 1001 1001 lOOt 1001 lOOt 1001 lOOt
loot
loot
1001
6.1
44.7
16.8
I.I
1001
2.1
18.2
41.0
16.3
lOOt
3.6
29.6
36.4
1.1
I.I
20.3
40.2
6.9
lOOt IWt
'Source: Reference 12.
Incorrect operation by carrier personnel, surge of electricity and surge of flow.
'Malfunction of control or relief equipment, •ulfunctlon of yalve, pump failure, puep packing failure, and tank roof drain leaking.
*lleavy rains or floods, cold weather, lightning and landslides.
'Defective girth Held, failure of previously welded repairs, and defective weld.
Defective pipe se»n, failure In river crossing, ruptured or leaking gasket, threads stripped or broken, pipe failed due to buckling, ruptured or
leaking seal, pipe coupling failure, defective pipe, stress crack and wrinkle bend split.
Equipment rupturing line, rupture of previously damaged pipe, freight train derailment, vandalism, and explosives.
Percent figures my not total 1001 due to rounding off.
-------
10
TABLE 19. NUMBER AND VOLUME OF SPILLS FOR FAILURES OF PETROLEUM
PIPELINE SYSTEM COMPONENTS (IN PERCENT OF TOTAL)
1
Pipeline System Component
Line pipe
Pumping station
Delivery point
Tank farm
Bulk plant
Crossover
Test or safety
Underground storage
Reception point
... .
1971 1972 1973 1974
Spills Vol Spills Vol Spills Vol Spills Vol
90 90 79 79 81 78 81 72
43 10 3 83 54
16 11 21 2 10
35 8 17 8 19 11 30
-
2 2
3 1
1
1975 1971-1975
Spills Vnl Spills Vol
75 84 81 80
10 4 83
21 22
12 10 8 13
<1 <1
<1 <\
<1 <1
<1 <1
<1 <1
-------
500
450 .
400 •
350
300
- 250
St
200
150
100
50
• line pipe
a Pumping station
A Tank farm
o Delivery point
• Miscellaneous
A Total
() Percent of total
(90)
(83 r
1968 1969 1970 1971 1972 1973 1974 1975 1976
Year of Occurrence
Figure 21. Number of incidents versus year of occurrence
for carrier system components
80
-------
400,000
300,000
in
01
s.
2. 200,000
100,000
_ (90)
0
Line pipe
Pumping station
Tank farm
Delivery points
Miscellaneous
Total
Percent of total
1971 1972 1973
Year of Occurrence
1974
1975
Figure 22. Volume of spill versus year of occurrence
for carrier system components.
81
-------
show that most spill incidents and volume can be attributed to failure in
line pipe. During the 5-year period between 1971 and 1975, for example,
line pipe accounted for over 80 percent of the total number and volume of
spills. Figure 21 shows that the number of reported spills from line pipe
decreased dramatically between 1968 and 1976 but remained relatively the
same for other components. This rapid decrease in spill incidents of line
pipe was not observed in spill statistics of petroleum pipeline systems
from U.S. agencies for other areas or in statistics of gas leaks from gas
transmission and gathering lines. In contrast, Figure 22 shows a slightly
upward trend in the volume of spills from line pipe.
The tank farm was the only other component of the pipeline system
where spills and spill volume were high. Between 1971 and 1975, 8 percent
of the spills and 13 percent of the volume occurred at tank farms. There
was a slightly upward trend in the number of incidents and no consistent
trend in the volume of spills over this 5-year period.
Other major petroleum pipeline system components, including the pump-
ing station, delivery point, bulk plant, crossover, underground storage
and reception point, accounted for less than 7 percent of the spill volume
and 11 percent of the spill incidents. The reported number of incidents
and volume of spills were relatively the same, year-by-year.
4.6.2.1.3 Cause of Pipeline System Failures
Spills of 50 barrels or more from the pipeline system are due to a
variety of causes (see Appendix C and Table 20). Only five causes, how-
ever, occurred more often than 2 percent of the time. These five causes,
which accounted for over 78 percent of the spills and 66 percent of the
volume from 1968 through 1976 (Table 22), are:
(1) Equipment rupturing the line - 15 percent spills, 24 percent
volume
(2) Incorrect operation - 8 percent spills, 12 percent volume
(3) External corrosion - 29 percent spills, 12 percent volume
(4) Internal corrosion - 8 percent spills, 5 percent volume
(5) Defective pipe seam - 8 percent spills, 12 percent volume.
Variations in the number and volume of spills for these causes from year-
to-year are illustrated in Figures 23 and 24. For comparison purposes,
causes of spills are grouped into four main categories, i.e., outside
forces, corrosion, material or construction defect and other causes.
Corrosion (internal and external) was the major cause of spills, ac-
counting for an average of 37 percent of spill incidents and 16 percent
of the spill volume between 1968 and 1976. However, corrosion has not
been the major cause of spills since 1974 on a year-by-year basis. Spills
caused by external corrosion decreased dramatically over this 9-year pe-
riod (Figure 23). There appears to be no reason that adequately explains
the decrease.
82
-------
TABLE 20. NUMBER AND VOLUME OF ALL SPILLAGE INCIDENTS (IN PERCENT) FOR ALL
PIPELINE SYSTEM COMPONENTS, BY CAUSE REPORTED TO OPSO
BETWEEN 1971 AND 1975
1968
1.
11.
III.
IV.
Causes
Outside forces
1. Equipment rupturing line
2. Incorrect operation by
carrier personnel
Corrosion
3. External
4. Internal
Material or Construction Defect
5. Defective pipe seam
TOTAL I. II, III
Other
TOTAL 1, 11, III, IV
Spills
20
2
43
5
6
76
24
100
Vol
23
29
16
2
7
87
13
100
1969
Spills
22
3
38
4
8
75
25
100
Vol
25
4
21
2
2
54
46
100
1970
Spills
20
4
43
9
8
74
26
100
Vol
15
5
9
3
7
39
61
100
1971
Spills
22
7
33
7
10
79
21
100
Vol
25
9
12
5
15
66
34
100
1972
Spills
22
7
24
8
9
70
30
100
Vol
23
13
11
11
12
70
30
100
1973
Spills
24
6
26
5
8
64
36
100
Vol
12
17
9
3
27
68
32
100
1974
Spills
31
11
20
8
70
30
100
Vol
34
7
7
17
65
35
100
1975
Spills
29
9
11
22
6
77
23
100
Vol
26
8
7
12
14
72
28
100
1976
Spills
32
10
20
5
7
74
26
100
Vol
29
16
12
1
15
73
28
100
1968-1976
Spills
25
8
29
8
8
78
22
100
Vol
24
12
12
5
13
66
34
100
-------
225
200 -
150
a.
1/5
o
i.
100
50
• Corrosion external
Q Corrosion internal
± Equipment rupturing line
A Incorrect operation by carrier personnel
• Defective pipe seam
1963 1969 1970 1971 1972 1973 1974 1975 1976
Figure 23. Number of spills versus year of occurrence by cause.
84
-------
600,000
500,000
400,000
300,000
^ 200,000 i.
2 100,000 T-
•= 80,000
o 60,000
40,aoo
20,000
• Corrosion external
O Corrosion internal
± Equipment rupturing line
A Incorrect operation by carrijer personnel
• Defective pipe seam
4 All spills
1968 1969 1970 1971 1972 1973 1974 1975 1376
Figure 24. Volume of spill versus year of occurrence by cause.
85
-------
Outside forces were the major causes of spills since 1974. There has
been a slight decreasing trend in the number of incidents and a slight in-
creasing trend in the volume of oil spilled over this 9-year period. Equip-
ment rupturing the line and incorrect operation by carrier personnel were
responsible for an average of 33 percent of"the spills and 36 percent of
the spill volume between 1968 and 1976. Defective pipe seams were the ma-
jor cause of material failure or construction defects in the pipeline sys-
tem. This cause accounted for an average of 8 percent of the spills and
13 percent of the spill volume between 1968 and 1974. The number of inci-
dents has been fairly uniform and there has been no specific trend in the
year-by-year spill volume.
Causes of-Line Pipe Failure—Incidents and Spill Volume
Since line pipe accounts for most of the incidence of spills and spill
volume in the petroleum pipeline system (see Section 4.6.2.1.2), it is
treated separately. Statistics on the major causes of spills from line
pipe are presented in Table 21 and Figure 25. An average of 82 percent of
the total number of spills and 69 percent of the spill volume were due to
only four main causes for the 4-year period between 1971 and 1974. These
are:
(1) Outside Forces - 41 percent spills, 46 percent volume
(a) Equipment rupturing the line - 31 percent spills,
26 percent volume
(2) Corrosion - 39 percent spills, 18 percent volume
(a) External - 31 percent spills, 12 percent volume
(b) Internal - 8 percent spills, 6 percent volume
(3) Material Failure or Construction Defect - 17 percent
spills, 35 percent volume
(d) Defective pipe seam - 12 percent spills, 25 percent
volume
(4) Other - 3 percent spills, 1 percent volume.
Outside forces were the major cause of the incidence and volume of
spills for line pipe. This cause accounted for 41 percent of the total
incidents and 46 percent of the total spillage. Outside force incidents
(total number of incidents and percent of all incidents) have been in-
creasing steadily since 1971. Between 1971 and 1974, the number of in-
cidents increased from 88 to 103 and the percentage of incidents increased
from 34 to 53 percent. The spill volume, however, was relatively uniform
over this period. Equipment rupturing the line was the predominant cause
of spillage, accounting for 31 percent of the spills and 26 percent of the
spill volume. Equipment rupturing the line and excavation equipment ac-
counted for over 40 percent of the spill incidents and volume; these are
the only two causes from outside forces that potentially can be reduced
significantly by a spill prevention program.
86
-------
TABLE 21. NUMBER, VOLUME AND PERCENTAGE OF ALL SPILLAGE INCIDENTS FOR LINE PIPE,
BY CAUSE REPORTED TO OPSO BETWEEN 1971 and 1975
CD
1971
Mo. of Uol
Cause
A. ouisioE roiicts
Equip, rupturing MM
Incorrect operation by
carrier personnel
fire or explosion
f Kcavatton etfulpnent
flow control
malfunction
Natural causes
lank overflow or
operator error
IOIAI
n. CORROSION
fxlviMl
Internal
IOIAI
C. DtrtCIIVE PIPf
Sean
Ucli!
Rupture jasktt, seal.
packing and land
IOIAI
IOIAI (A.R.C)
II. OIHCR
IOIAI (A.B.C.D)
Spills Barrels
66
2
0
4
|
6
88
98
22
120
31
10
6
47
255
3
258
61.005
5.211
0
10.825
4,030
14,060
104,348
28.873
12.107
40.9M
42,501
13.818
5,840
62.IS9
207,487
5,213
212,700
tof
Spill
76
1
0
2
0
2
34
38
9
47
12
4
2
IB
99
1
100
total
19/2
No. of Hot
sl*ol Spills Barrels
29
2
0
S
2
,
49
14
6
20
20
7
3
30
98
2
too
66
|
1
C
1
9
84
72
14
91
28
10
5
43
218
17
235
76,924
4.712
1.030
33.157
1.242
8.007
130.415
38.710
36,613
75,323
43.303
6.4/2
7,462
57,237
262,395
9.2A7
27 1.682
TofTSIil
Spllls|«ol
28
0
0
3
0
4
36
31
B
39
12
4
2
18
93
7
too
28
2
0
12
0
3
48
14
13
27
16
2
3
21
96
4
Ion
1973
"NoTof ».» ««'!"l'«
Spills Barrels
66
0
9
2
II
90
66
12
78
20
8
5
33
291
3
204
45,991
0
I9.MI
S.S5S
21.686
93,931
30.949
6.460
37.409
102.603
27.215
18,419
MB. 217
279.577
1.156
2112, /I)
Spills |*ol
32
0
4
1
5
42
32
«
.38
10
4
2
16
96
4
too
16
0
7
2
•
35
II
2
13
36
10
7
5.3
99
1
100
19/4
No, of 1u\
Spills Barrels
76
1
0
8
2
II
1
101
47
14
61
20
8
3
31
195
7
202
56.936
I/O
«
13.168
225
13.913
loo
90,391
ID. 305
3,2/5
21.660
41.621
11.159
1.617.
63.312
175.451
5,116
180.567
lot
Spill
30
0
0
4
1
5
0
53
23
7
311
10
4
1
15
911
2
ion
InUI
'***' No. of »o1
1171
Average »
of lolal
s|«ol SplllsBarrels Spills >ol
32
i)
0
/
0
8
0
51)
10
2
12
?7
r.
1
34
•»
4
IIKI
274
4
1
77
ft
17
1
»,',
7111
41
132
11
16
11
IM
nr,i
77
nil
240,856
10.091
1.010
77.031
11.1152
57.666
(Oil
4I1.I»1
116.017
58.455
1/5,3/2
238.028
5B.CC4
11.313
111 .1)25
174.110
27.777
'H7.6D?
31
0
0
3
I
4
41
31
8
II
12
4
1
17
17
3
Mm
26
|
0
7
|
7
46
12
6
IH
25
6
4
15
11
1
IIKI
1174
Avfralr
No. of Vvl
Spills Barrels
60
1
0
7
1
1
0
11
71
17
11.1
75
1
5
18
717
7
775
60.714
2.521
25 257
I9.25R
5 7.763
.|iill Mi1"1
871
nyi
III III
7IIM
1047
14,416 1 IV»I
25 25
104.7/.1
?1,«I4
14.613
41.041
51.52/
14 .666
8.111
87. /56
731,227
5.6MI
236.170
IIKI
ll«»
111*1
II'IP
VI!
74111
ii.:"i
l,",1
7141
HIM
B4I
HIM
1975: No. Spills - 188; Vol. (Barrels) - 261, 751; other statistics not available.
-------
c
Ol
60
10
0
1971 1972 1973 1974 1975
Year of Occurrence
Figure 25. Percentage of number and volume of spills
by cause for line pipe.
-------
Table 21 shows that corrosion caused an average of 39 percent of the
incident spills and only 18 percent of the spill volume between 1971 and
1974. Corrosion was the major cause of spills between 1971 and 1973. How-
ever, there has been a downward trend of the spill incidents and volume
caused by corrosion since 1971. External corrosion was the major cause of
corrosion failures and accounted for 31 percent of the total spills and 12
percent of the spill volume.
Defective pipe caused an average of 17 percent of the incidence of
spills and 35 percent of the spill volume between 1971 and 1974. There
has been a downward trend in the reported incidence of defective pipe
since 1971; no trend is apparent for the spill volume. Defective pipe
seams was the major cause of defective pipe failures and accounted for
12 percent of the total incidents of spills and 25 percent of the total
spill volume.
Pipeline faults are caused mainly by corrosion and defective pipe.
These accounted for an average of 56 percent of the total number of spills
and 53 percent of the total spill volume between 1971 and 1974. The re-
ported incidents of all pipeline faults shown in Table 19 have been de-
creasing since 1971. In 1971 there were 167 incidents whereas in 1974
there were only 92 incidents. There is no explanation based on pipeline
construction, materials, or maintenance that adequately accounts for this
trend. Pipeline faults potentially can be reduced by line pipe inspections
and maintenance.
Outside force incidents by themselves are quite significant but are
not faults in the pipeline or maintenance procedures. They usually in-
volve human error and insufficient communication between pipeline opera-
tors and equipment operators. These incidents potentially can be reduced
by a prevention program that improves communication and/or detects or pre-
vents impacts from outside forces before damage to the pipeline occurs.
4.6.2.1.4 Factors Relating to Spills
Type of Fluid-Pipeline Systems
Data are given in Table 22 for crude and product spills and in Table
18 for all types of fluids from petroleum pipeline systems in the U.S. that
were reported to OPSO between 1969 and 1976. Figures 26 and 27 illustrate
the yearly variations in the volume spilled and number of incidents of
crude, products and LPG. Gasoline, fuel oil, jet fuel, diesel fuel, con-
densate, kerosene and anhydrous ammonia are included in the data as petrol-
eum products in Table 22 and Figure 25.
Reported total spill volume (Table 22) averaged 335,644 barrels for
the 9-year period between 1968 and 1976. The average volume of products
spilled was about 20 percent higher than crude or 1306 and 1085 barrels,
respectively. For a more recent period, 1971 through 1975, the average
spill volume was about the same. Reported total number of spills of crude
and products (Table 22) averaged 289 between 1968 and 1976. Spill incidents
89
-------
TABLE 22. SPILL VOLUME AND INCIDENTS FOR ACCIDENTS
REPORTED TO OPSO BETWEEN 1968 AND 1976 FOR
CRUDE AND PRODUCT PIPELINE SYSTEMS
Year
1968
1969
1970
1971
1972
1973
1974
1975
1976
1968-
1976
1971-
1975
Petroleum
Crude
Products
TOTAL
Crude
Products
TOTAL
Crude
Products
TOTAL
Crude
Products
TOTAL
Crude
Products
TOTAL
Crude
Products
TOTAL
Crude
Products
TOTAL
Crude
Products
TOTAL
Crude
Products
TOTAL
Crude
Products
Avg Crude
Avg Prod
Avg Total
Crude
Products
Avg Crude
Avg Prod
Avg Total
Commodity
(Barrels)
222.257
107,154
329,416
174,640
39.905
264,545
383,160
92.348
475,508
115,670
89,410
205,170
233,690
96.555
330,245
208,373
119,418
327,791
168.738
60.824
229.562
105.871
108.324
214.195
96.059
128.363
224,422
1.856,659
1,163.969
206.295
129.329
335.624
832.432
474,531
165,486
94.906
260,392
spins a
Nunoer
322
134
456
246
109
355
216
107
323
172
97
269
180
107
287
155
113
268
166
60
226
135
94
229
123
72
195
1.715
893
190
99
289
308
471
161
94
256
Commodity
Barrels/Spill
1,035
1.306
1.161
1,027
1,009
1,017
Note: LPG spills are not included in this table.
90
-------
500
A
0
O
Crude
Product
LPG
Total (Crude, Products)
Total all fluids (Crude,
Products and LPG)
1968 1969 1970 1971 1972 1973 197U 1975 1976
Year of Occurrence
Figure 26. Number of incidents versus year of occurrence for
type of fluid transported in pipeline system
accidents reported to OPSO between 1968-1976.
91
-------
550,000
Crude
Product
Total (Crude and Product)
Total all fluids (Crude
Product and LPG)
in
p—
OJ
CD
^
50,000 u-
1968 1969 1970 1971 1972 1973 1974 1975 1976
Year of Occurrence
Figure 27. Volume of spills versus year of occurrence for type
of fluid transported through pipeline system
accidents reported to OPSO between 1968-1976.
92
-------
averaged about 92 percent higher than products during this period, but
only 71 percent higher for the 5-year period between 1971 and 1975. Fig-
ure 26 shows a steep declining trend in the yearly reported spill volume
and incidents for crude. Trends for LPG and other product spills are dif-
ferent; a slight increasing trend is shown for the volume of spills and a
slight downward trend in the number of incidents in Figure 27. There ap-
pears to be no conclusive explanation for the steep declining trend for
spillage (incidents and volume) from crude pipeline systems.
Type of Fluid - Line Pipe
The incidents of line pipe spills of crude and product (including LPG)
reported to OPSO for the 5-year period between 1971 and 1975 are given in
Table 21 and illustrated in Figure 28. A slight decreasing trend is shown
for crude while no specific trend is evident for product or LPG spills.
Annual distribution of spill frequencies and barrels lost from line
pipe, including spills of less than 50 barrels, were estimated in Appen-
dix D (LPG spills were not included in the analysis). Pertinent results
are presented in Table 23. Annual incidents for spills of 50 barrels or
more are estimated to be 86 percent greater for crude lines than for pro-
duct lines. However, if all spill sizes are included, spill incidents
from crude lines are estimated to be six times greater than for product
lines. The larger number of spills from crude lines could be due to fac-
tors such as higher pipeline mileage, smaller diameter lines and age.
These factors are considered in subsections that follow. Annual spill
size for product line pipe is estimated to be 28 percent higher than for
crude line pipe for spills of more than 50 barrels. This relationship of
spill volume for crude and product lines is similar to that of the total
petroleum pipeline system. If all spill sizes are included, annual spill
size for product line pipe is estimated to be about three times greater
than for crude line pipe.
Mileage - Pipeline System
A summary of spill incidents and volume for the pipeline mileage of
the petroleum pipeline system during the 9-year period between 1968 and
1976 are tabulated in Table 24 and illustrated in Figures 29 and 30(A).
Pipeline mileage data were obtained from BOM statistics presented in
Table 8 and included the total mileage for crude and product transmission
lines and crude gathering lines. Because spills from some of these pipe-
lines are not required to be reported to OPSO, the actual mileage for cal-
culation loss rates .is somewhat less than used in these tables and figures.
Thus the actual barrels spilled per mile and spill incidents per mile are
higher than indicated.
Spill losses averaged 1.58 barrels per mile between the 9-year period
from 1968 to 1976 and 1.44 barrels per mile between the 5-year period from
1971 to 1975. Spill incidents during these two periods averaged 1.45 and
1.26 spills per one thousand miles, respectively. A rapidly declining
year-by-year trend for both spill incidents and spill volume per mile can
be seen in Figures 29 and 30(A).
93
-------
300
200
c
OJ
£ TOO
CL
A Crude
0 Product
Q LPG
• Total-All spill
---©---
£
-n.
-o-
©
LU
1971
1972
1973
1974
5^
1975 Avg
1971-
1975
Figure 28. Number of incidents versus year of occurrence for type
of fluid transported in line pipe, accidents reported to OPSO
between 1971-1975.
94
-------
TABLE 23. STATISTICAL ESTIMATE OF ANNUAL DISTRIBUTION OF PETROLEUM0
SPILL FREQUENCIES AND BARRELS LOST FROM LINE PIPE FOR
CRUDE AND PRODUCT
VO
tn
Frequency of Spills
Number
(Barrels)
0 -
50 -
100 -
200 -
400 -
800 -
1600 -
Total -All
50
100
200
400
800
1600
25000
Spills
Total-Spills
Greater than
50 Barrels
Crude
300
33
28
22
16
12
17
428
128
Product
7
7
11
13
13
11
14
76
69
Percent
Crude
70
8
7
5
4
3
4
100
30
Product
9
9
14
17
17
14
18
100
91
Volume of
Barrels
Crude
2
2
3
6
9
13
93
131
128
,796
,391
,969
,233
,262
,023
,326
,000
,204
Product
194
541
1,588
3,832
7,595
12,366
62,930
89,046
88,942
Spills
Percent
Crude
2
2
3
5
7
10
71
100
98
Product
0
1
2
4
9
14
71
100
100
LP6 spills are not included in this table.
-------
TABLE 24. SPILL DATA SUMMARY FOR PETROLEUM PIPELINE SYSTEMS TRANSPORTING
LIQUIDS IN THE U.S. FOR ACCIDENTS REPORTED
TO OPSO BETWEEN 1968-1976
VO
CM
Year
1968
1969
1970
1971
1972
1973
1974
1975
1976 ,
1968-1976
Total
Average
1971-1975
Total
Average
Through-
put
(Barrels
xlO9)
7.600
7.600
8.001
7.860
8.010
8.860
39.07
7.81
Length
(Miles)
209,478
212,000
215,000
218,671
220,000
222,000
223,535
224,400
225,000
1,970,084
218,898
1,108,606
221,721
Barrels
392,588
343,691
521 ,849
245,057
360,654
379,365
293,643
319,423
255,637
3,111,407
346,007
1,598,142
319,628
Loss
% of
Through-
put
0.0032
0.0047
0.0047
0.0037
0.0040
0.0029
0.020
0.004
Incidents
Barrels/
Mile
1.87
1.62
2.43
1.12
1.64
1.71
1.31
1.42
1.13
1.58
1.44
Spills
490
403
347
308
309
273
256
255
209
2,850
317
1,401
280
Spills
per
103 Miles
2.3
1.9
1.6
1.4
1.4
1.2.
1.1
1.1
0.9
1.4
1.2
1.3
-------
500 p A
? ' \
m I N
0)
•S 400 -
3
01
-a
J2 300
200
100
_ 3.00
ISI
Qi
•E 2.50
2.00f
o
o
o
O.
£ 1.50
•S i.oof-
u
- 0.50
•f—
Q.
^ 0
i
_L
1
1
1968 1969 1970 1971 1972 1973 1974 1975 1976 Avg Avg
1968- 1971-
Year 1976 1975
(A) SPILL INCIDENTS VERSUS YEAR
1968 1969 1970 1971 1972 1973 1974 1975 1976 Avg Avg
Year 196&- 1971'
1976 1975
(B) SPILL INCIDENTS PER 1000 MILES VERSUS YEAR
Figure 29. Spill incidents and spills per 1000 miles versus year
for pipeline system accidents reported to
OPSO between 1968-1976.
97
-------
10
S3
2.50
2.25 *
2.00 -
1.50
1.25 -
1.00 -
o 0.75 h
0.50 -
0.25
0
1968 1969 1970 1971 1972 1973 1974 1975 1976 Avg Avg
Year 1968- 1971-
(A) VOLUME SPILLED PER MILE VERSUS YEAR 1976 1975
0.006 r
| 0.005
51
I 0.004
« 0.003
•3
- 0.002
"a.
!/1
§ 0.001
3
"o
1968 1969 1970 1971 1972 1973 1974 1975 1976 Avg Avg
Year 1968- 1971
(B) VOLUME SPILLED AS % OF THROUGHPUT VERSUS YEAR 19?S
Figure 30. Spill volume for mileage and throughput versus
year for pipeline system accidents reported
to OPSO between 1968-1976.
98
-------
Mileage - Line Pipe
A summary of OPSO spill data for line pipe mileage is presented in
Table 25 and illustrated in Figure 31 for the 5-year period between 1971
and 1975. Annual spill statistics for line pipe mileage are presented in
Table 26. The latter statistics are based on the estimated (see Appendix
D) line pipe spills and spill volume, including those of less than 50 bar-
rels, presented in Table 23.
Table 25 shows that spill losses averaged 1.09 barrels per mile and
spill incidents averaged 0.98 spills per one thousand miles during the
5-year period from 1971 to 1975. These values are only slightly lower
(about 20 percent) than those for the entire petroleum pipeline system.
This is expected, of course, because over 80 percent of the total spill
incidents and volume are attributed to line pipe failures. Figure 31
shows a slight declining trend, year-by-year, of spill incidents per one
thousand miles of line pipe while no specific trend is evident for barrels
spilled per mile.
Estimated annual spill volume (Table 26) is 0.91 barrels per mile and
spill incidents is 2.28 spills per one thousand miles. If LPG spills (ap-
proximately 10 percent of total) are added to this estimated annual spill
volume, the resultant value would be about the same as the average value
of 1.09 barrels per mile from actual OPSO statistics. This is expected
even though the statistical estimate of annual spills includes spills of
less than 50 barrels. This is because there appears to be no significant
increase in the amount of 'spilled volume for spills of less than 50 bar-
rels. Table 23 shows that below 50 barrels about 70 percent of the spills
and only 2 percent of the spill volume are expected to occur for crude
lines and about 9 percent of the spills and less than 1 percent of the
spill volume are expected to occur for product lines.
Annual spill rate is estimated at 2.28 spills per mile for all petrol-
eum line pipe. This value is much higher than the value computed for
spills reported to OPSO. This increased rate is expected because of the
large number of spills (Table 23) that are estimated to occur below 50
barrels. In addition, inclusion of spills not covered by OPSO regulations
and estimation of unreported spills (to be discussed in later sections)
will further increase this spill rate.
Crude spills (Table 26) are estimated to occur three times as often as
product spills. A possible explanation for this is that crude lines are
typically smaller in dimension (diameter and wall thickness) and somewhat
older and thus have a higher frequency and a lower spill size. For product
lines, the 0.97 spill rate per mile is about the same as the 0.98 value in
Table 25 for all pipeline spills in excess of 50 barrels. This result is
expected because product spills generally would be larger than crude spills
because of larger pipeline diameter and thus are reported to OPSO. Also
more product lines are under OPSO reporting regulations than crude lines.
99
-------
TABLE 25. SPILL DATA SUMMARY FOR LINE PIPES TRANSPORTING
LIQUIDS IN THE U.S. FOR ACCIDENTS REPORTED TO
OPSO BETWEEN 1971-1975
Year
1971
1972
1973
1974
1975
1971-1975
Total
Average
Through-
put
(Barrels Length
xl<)9) (Miles)
7.600 218,671
7.600 220,000
8.001 222,000
7.860 223,535
8.010 224,400
39.07 1,108,606
7.814 221,721
Barrels
212,700
271 ,682
282,733
180,567
261 ,751
1,209,433
241,887
Loss
Incidents
% of
Through- Barrels/
put Mi 1 e
0.0028
0.0036
0.0035
0.0023
0.0032
0.0031
0.97
1.23
1.27
0.80
1.16
1.09
Spills
258
235
204
202
188
1,087
217
Spills
per
103 Mile
1.18
1.07
0.92
0.90
0.84
0.98
aMileage data obtained from Table 8.
TABLE
26. ANNUAL SPILL
LIQUIDS IN
3 STATISTICS FOR LINE
THE U.S. FOR ALL SPILL
PIPES TRANSPORTING
SIZES
Commodi ty
Petroleum
Crude
Product
Through-
put
(Barrels Length
xlQ9) (Miles)
7.81C 221,000
-4.7 s:143,000
=3.2 =78,000
Barrel £
220,046
131,000
89,046
% of
Through- Barrels/
put Mile
0.0028
0.0028
0.0028
0.91
0.91
1.14
Spillsb
504
428
76
Spills
per
103 Miles
2.28
2.99
0.97
Does not include LPG spills.
Based on spill analysis in Appendix D.
cBased on average value between 1971 and 1975.
100
-------
2.Or
"°^ 1 5 -
(DQ; I • 3 i
r™> »r—
•r- 2£
^3 1.0 h
Se.o.51-
1971
1972
1973
Year
(A) VOLUME SPILLED PER MILE VERSUS YEAR
1974
1975 Avg
1971-
1975
2.Or
4-J 0)
C i—
O) ^>
•O Z
•r-
O O
C O
H™< ^^
^^1 0.5
0.1—
1971
1972
1974
1973
Year
(B) SPILL INCIDENTS PER 1000 MILES VERSUS YEAR
*i
1975 Avg
1971-
1975
Figure 31. Spill incidents per 1000 miles and spill volume per mile
versus year for line pipe transporting liquids in the U.S.
reported to OPSO between 1971 and 1975.
101
-------
Throughput - Pipeline Sys.tem
Spill statistics based on the throughput of petroleum pipeline systems
are presented in Table 24 and illustrated in Figure 30(B). Throughput es-
timates are taken from BOM statistics (Table 5) and include domestic, for-
eign crude and natural gas liquids transported to the refinery input via
pipelines, and petroleum products turned into pipelines.
Table 24 shows that spill losses for the period between 1971 and 1975
averaged 0.004 percent of total throughput. A slightly declining, year-by-
year trend can be seen in Figure 30(B) for losses as a percent of through-
put.
Throughput - Line Pipe
Line pipe spill data for throughput are presented in Tables 25 and 26.
Losses (Table 25) for line pipe reported to OPSO during the 5-year period
from 1971 through 1975 averaged 0.0031 percent of total throughput. This
value was only slightly lower (about 20 percent) than for the entire
petroleum system. This is expected because about 80 percent of the total
spill volume is attributed to line pipe. There is no specific year-by-year
trend for this data.
Estimated annual losses (Table 24) are 0.0028 percent of throughput.
In comparison with the losses as reported to OPSO, this is a reasonable
value considering that: (1) LPG spills, which account for about 10 percent
of all spills, are not included in the estimated annual losses; (2) spills
of less than 50 barrels are included in the estimated annual losses but are
estimated to be less than 2 percent of the total volume; and (3) the esti-
mates are based on spillage trends that indicate a declining trend (based
on reported spills).
Diameter - Pipeline System
Spillage data for the petroleum pipeline system are broken down into
two categories, i.e., those greater than and those less than 12 inches in
diameter, in Figure 32. This was done because there was insufficient data,
statistically, for an accurate breakdown for each individual size of pipe-
line greater than 12 inches in diameter. The spill rate for each category
was estimated by dividing the total number of spills by the mileage exist-
ing at the beginning of 1974 and normalizing to one year. There were
190,331 miles of pipeline 12 inches in diameter or less and 34,296 miles
of pipeline greater than 12 inches in diameter as of January 1, 1974.
Table 27 shows the number of incidents and cause for these two categories .
for the 5-year period between 1971 and 1975.
The distribution of the sizes of the reported spills are plotted on
log normal probability coordinates in Figure 32. For the larger diameter
pipeline system, these data show the median spill size to be approximately
850 barrels. For pipeline systems less than 12 inches in diameter, the
median spill size is 360 barrels. For the larger diameter pipeline sys-
tems, 8 percent of the reported spills exceeded 10,000 barrels.
102
-------
o
to
TABLE 27. SPILL INCIDENTS OF PIPELINE SYSTEM FOR TWO PIPELINE DIAMETER
CATEGORIES FOR ACCIDENTS REPORTED TO OPSO
BETWEEN 1971-19759
Cause
External Corrosion
Internal Corrosion
Line Ruptured by Excavation
Equipment
Prior Damage by Excavation
Equipment
Defective Pipe Seam
Defective Held
Rupture of Gasket, Seal, etc.
Hre or Explosion
Flow Control Malfunction
Flow Control -Operator Error
Incorrect Operation by
Carrier Personnel
Natural Disasters (landslide.
Flood, Minds torn, Freezing,
Unknown
TOTAL INCIDENTS
aSource: Reference 14.
Pipelines
and Less
Installed Before 1950
(Number of Incidents)
297
40
142
6
83
22
23
6
3
23
2
etc.) 36
55
73ff
12 Inches
In Diameter
Installed After 1950
(Number of Incidents)
50
46
128
18
'I
11
56
14
11
12
11
24
67
S19*
Pipelines Greater Than
12 Inches In Diameter
(Number of Incidents)
9
8
28
3
19
7
2
0
0
0
0
8
0
"If
-------
105 -
CO
cc
a:
I
Q
Q_
CO
o-
,3
10
1
II I I I
Pipeline Greater than
12-Inches in Diameter
Pipelines 12-Inches
or Less In Diameter
0 P
J
L ____ I
I I I I I
L
I
I
I __ I __ L ___ I
0.01 0.08 0.3 0.5 1 2 5 10 20 30 40 50 60 70 80 90
PERCENTAGE OF SPILLS - ACCUMULATIVE
95 98 99 99.8 99.9
Figure 32. Size distribution of spills from U.S. terrestrial
pipelines transporting liquids. (Source: Reference 14.)
-------
Diameter - Line Pipe
Analysis was carried out according to individual sizes of line pipe.
Statistics of spill incidents and volumes for pipeline diameters3 reported
to OPSO for line pipe were used.
The mean spill size (Table 28) increased from 569 barrels per spill
for pipelines 4 inches or less in diameter to 6011 barrels for pipelines
30, 32 and 34 inches in diameter. Mean spill size increased with pipe
diameter to the exponent 2.xx and is essentially proportional to the cross-
sectional area of the pipe. These results are illustrated in Figures 33^
and 34 where mean spill size is plotted for individual pipeline diameters.
These results were particularly apparent for pipe diameters of 6, 8, 10 and
12 inches. Eighty-one percent of the spills and 78 percent of the spill
volume were reported for these pipeline dimensions. The large number of
incidents for these sizes was suitable, statistically, for accurate analysis,
The spill frequency (Figure 35) decreases rapidly as pipeline diameter
increases and then levels off for pipelines greater than 14 inches in diam-
eter. A peak rate of 2.14 spills/10^ miles/year is initially reached for
4-inch diameter lines; then the rate levels off at about 0.5 spills/103
mile/year for lines that are larger than 14 inches in diameter. Note that
when all gathering lines are included, the spill rate for 4-inch or less
diameter lines is obviously too low and does not follow the general trend
of spill frequencies and pipeline diameter. However, when gathering lines
are excluded in the mileage, a more realistic trend occurs, but the value
for the 4-inch or less diameter lines still appears slightly low.
Spill volume per mile (Figure 36) increases rapidly with pipeline
diameter. A peak of 3.94 barreIs/mile/year was reached for the 30, 32 and
34 inch diameter lines. Lowest spill volume, slightly above 1 barrel/mile/
year, occurred for the small diameter lines. When all gathering lines are
included in the pipeline mileage (Figure 36), the spill rate is obviously
too low and does not follow the general trend. When gathering lines are
excluded in the mileage, however, a more realistic trend occurs, but the
value for 4-inch diameter lines or smaller still appears too high.
The outside force incidents on line pipe with diameter are shown in
Table 29 for the four-year period between 1971 and 1974. Incidents caused
by line pipe faults are shown in Table 30 for various pipe diameters. The
frequency of spills caused by the equipment rupturing the line and corro-
sion (the major causes of damages by outside forces and line pipe faults)
are illustrated in Figure 37 for the year 1971.
Approximately 4 percent of the available data on reported spills did not
include pipeline diameters; thus, relationships that use this particular
statistic would be about 4 percent low. This, of course, does not af-
fect relative values or trends.
Combining more than one pipeline diameter appears to provide a clearer
representation of the effect of spillage with pipe diameter.
105
-------
TABLE 28. SUMMARY OF SPILL DATA FOR LINE PIPE DIAMETERS FOR LIQUID
PIPELINE ACCIDENTS REPORTED TO OPSO BETWEEN 1971-1975
1971-1975 Spills
Diameter
(Inches)
4 or less
6
8
to
12
14
0 '«
°> 18
20
22-24
26-28
30-32-34
36- up
unknown
i Total
Number
of
Spills
82
265
343
146
87
6
20
5
14
9
5
9
4
35
1,030
X of
Total
Spills
8
26
33
14
8
<1
2
1
<1
<1
<1
•1
3
X of
Volume Total
Barrels Volume
46.685
164.749
255.652
185,142
132,240
4.332
21.043
27.109
70.140
20,445
22,660
54.098
6.229
62,344
1.072,868
4
15
24
17
12
<1
2
3
7
2
2
5
1
6
Mean
Size
Barrels
569
622
745
1.268
1.520
722
1.052
5,422
5,010
2,771
4,532
6.011
1,557
1,781
1,041
All Pipelines
X of
Total
Miles Miles
59,891
36.664
52.892
22,592
19,512
3,402
7.664
3.269
6.053
4.703
1.871
2.750
2,272
223.535
27
16
24
10
9 .
2
3
3
2
1
1
1
Sol 11s/
103 Miles/
Year
0.26
1.4
1.28
1.3
0.9
0.36
0.52
0.3
0.46
0.38
0.54
0.66
0.34
0.92
Barrels
Spilled/
Mile/Year
0.24
0.9
0.96
1.64
1.36
0.26
0.54
1.66
2.32
0.87
2.42
3.94
0.54
0.96
Miles
7.637
26.454
47,753
21 ,652
18.886
3.384
7.622
3.268
6,044
4,699
1,867
2,750
2,272
154,288
All Pipelines*
Except Gathering
X of
Total
Miles
5
17
31
.14
12
2
5
2
4
3
1
2
1
Spills
103 Miles/
Year
2.14
2.14
1.42
1.32
0.92
0.36
0.52
0.3
0.46
0.18
0.54
0.66
0.34
1.34
Barrels
Spilled/
Mile/Year
1.22
1.24
1.07
1.71
1.4
0.26
0.54
1.66
2.32
0.87
2.42
3.94
0.54
1.39
Based on pipeline mileage in 1974.
-------
7000^-
6000
5000
(f>
JD
O!
M
4000
3000
C
(U
O)
/ O 9(30-32-34)
14(22-24-26-28) ©
r
45(14-16-18-20)
2000i—
87(12)
146(10) &/
lOOOj— /
>/343(8)
G^® 265(6)
F 82(4)
Ol i I
NOTE:
Number adjacent to data point
denotes number of spills making
up mean.
Number in () indicates pipe
diameters in inches used in
average.
10 15 20 25 30 35
Pipe Diameter D (inches)
40
45
50
Figure 33. Mean spill size and line pipe diameter for line
pipe accidents reported to OPSO, 1971-1975.
107
-------
10,000
^ 1,000
J3
0)
N
100
10
A United States Spill Data19/
1971-1975 5^ / * 9
A VA 5
NUMBERS ADJACENT TO DATA /
POINTS DENOTE THE NUMBER QTF q
SPILLS MAKING UP EACH MEAN *
146,
A 20
10
Pipe Diameter, D (inches)
100
Figure 34. Mean spill versus line pipe diameter for line pipe
accidents in the U.S. reported to OPSO
between 1971-1975.
-108
-------
•3.0
I 2.8
r
01
ro
O
2.6 _
2.4
42 2.2
E Total petroleum pipeline mileage in the U.S.
O Excludes gathering lines from pipeline mileaoe
in the U.S.
Number adjacent to data points denote number
of spills making up mean.
^
2.0 I-
1.8 I-
1.0
0.8
82 265
EJ B
O>
a-
V
u_
Q.
co
1
1
1
.6 h
i
•4 r- 21
.2
343
55(^>P
343
146
3 146
0.6
0.4
0.2
—
—
o82
n i i .
048
or
less
^JL^
6 5
i i i j i 1 i i
12 16 20 24 28 32 35 40
or.
greater
i i i i
44 48 52 56
Pipeline Diameter, D (inches)
Figure 35. Frequency of spills for pipeline diameters and mileage for
line pipe accidents in the U.S. reported to OPSO
between 1971-1975.
109
-------
UJ
(O
CO
4.0T
3.8
3.6
3.4
3.2
3.0 1-
2.!
T 2.1
1.6
"S. 1.4
oo
1.2
i.oL
s
0.8 j~
o.e'
0.4
0.2L
146
82
Numbers adjacent to data points denote number
of spills making up mean.
O Total petroleum pipeline mileage in the U.S.
A Excludes gathering lines from pipeline
mileage in the U.S.
20
0 4 8 12 16 20 24 28 32 36 40 44 48 52 56
Pipeline Diameter, D (inches)
Figure 36. Spill volume for petroleum pipeline diameters of line pipe
accidents reported in the U.S. to OPSO between 1971-1975.
110
-------
TABLE 29. OUTSIDE FORCE INCIDENTS ON LINE PIPE REPORTED IN
THE U.S. TO OPSO BETWEEN 1971-1974
UNf PIPE
Diameter
(Inchos)
4 or less
6
n
10
12
M
16
IR
20
22-24
26-2B
30-32-34
36 -lip
unknovn
TOIAIS
Number of Incidents
1971 1972 1973
1974 1971-1974
Mileage Equipment Excavation Equipment Excavation Equipment Excavation Equipment Excavation Total Spill;
AH All lines Rupturing Equipment Rupturtno Equipment Rupturing Equipment Rupturing Equipment Spills All
lines except Line line line Line lines
gathering
59,899
36,664
52,892
22,592
19,512
3.402
7,664
3,269
6,053
4,703
1,871
2,750
2,272
223,551
7,637 455
26,454 19 i 15 1 ,3
47,753 22 23 Z3
21.652 H 1 7 2 5
18,886 3 728
3,384 1 1
7,622 334
3,268
6,044 111 1
4.699 1
1,867
2,750 • 3 1
2,272
212 5
154,288 66 4 66 6 66
1 17 1 33 0.13
1 18 3 71 0.4R
4 14 2 80 0.41
1 12 1 40 0.44
1 7 2R 0.35
1 3 0.22
3 13 0.40
3 7 0.29
1 0.04
4 0.40
1 1 0.10
1 1 12
9 76 8 301 0.33
i 10s mlle-yr
All
Urns
except a
gatliorlng
1.07
O.fi7
0.46
0.46
0.37
0.?2
0.40
0.28
0.04
0.40
0.10
0.4P
a0.21 spills/103 mile-yr for line pipe greater than 12 inches in diameter.
-------
TABLE 30. FAULT INCIDENTS FROM LINE PIPE REPORTED IN THE U.S.
TO OPSO BETWEEN 1971-1974
r\>
Number of Incidents
1971 1972
Diameter
(Tn~che7r
4 or less
6
8
10
12
14
16
18
20
22-24
26-28
30-32-34
36- up
Unknown
TOTAL
Mileage Corrosion Defective
All All Tines External Internal Pipe HeTd
lines except Seam
guttering
59.891
36,664
52.892
22.592
19.512
3.402
7,664
3,269
6.053
4,703
1.871
2.750
2.272
223.551
7,637 9
26,454 29 7 3 1
47,753 39 8 16 3
21,652 12 3 5
18,886 5 153
3,384 i
7.622 i
3.268
6.044 i ,
4,699
1,867
2,750
2.272 ,
1 2
154.288 98 22 29 7
Corrosjon Defective Con
External Internal Pipe HeTd Extern?
Seam
75115
16 9 2 1 22
35 3 6 2 16
10 2 10 1 15
3 225
3
1
1
2
1 1
1 1 3
72 / 19 28 10 66
1973
rosion Defective
f internal Pipe Held
Seam
1 1
8 3
1 7 1
1 1 3
2 2
1
1 1
1
2
1
1
1
12 20 8
0.24 spttls/lO^ mfles-year for line pipe greater than 12 Inches In diameter
0.22 spills/103 miles-year for line pipe greater than 14 Inches In diameter
(continued)
-------
TABLE 30 (continued)
1971-1974
Diameter
TTScnesT
4 or less
6
8
10
12
14
16
18
20
22-24
26-28
30-32-34
36- up
Unknown
TOTAL
Mileage
All "ATT Tines
lines except
gathering
59,891
36,664
52,892
22,592
19,512
3,402
7,664
3,269
6,053
4,703
1.871
2.750
2,272
223,551
7,637
26,454
47,753
21,652
18,886
3,384 '
7,622
3.268
6,044
4.699
1.867
2,750
2,272
154,288
Total
Spills
30
106
127
52
15
1
3
1
4
1
1
9
350
Corrosion
Spills/MF'mlle-yr
All Exc gathering
lines lines
0.12
0.72
0.60
0.57
0.19
0.07
0.09
0.07
0.16
0.05
0.09
0.39
0.97
1.0
0.65
0.60
0.19
0.007
0.09
0.07
0.16
0.05
0.09
0.57
1971
-1974
Defective Pipe & Weld
Total
Spills
4
12
48
23
22
4
2
4
1
5
3
2
130
All
lines
0.02
0.08
0.23
0.26
0.28
0.29
0.06
0.16
0.05
0.66
0.27
0.22
0.15
.
Wl-1974
AH Une_ Pipe ; Faults
10' wlle-yr ,
ExcJatBiHng ^
0.13
0.11
0.25
0.27
0.29
0.30
0.06
0.16
0.05
0.66
0.27
0.22
0.21
33
118
175
76
37
5
5
1
8
2
5
3
3
13
480
Spills/ 10' mlle-yr
All
Lines
0.14
0.80
0.82
0.82
0.47
0.36
0.16
0.07
0.33
0.10
0.66
0.27
0.39
0.54
Exc. gathering
lines*
1.08
1.11
0.91
0.86
0.49
0.36
0.16
0.07
0.33
0.10
0.66
0.27
0.39
0.78
0.24 spills/ID^ miles-year for line pipe greater than 12 Inches In diameter
0.22 spills/103 miles-year for line pipe greater than 14 inches In diameter
-------
O)
>~
i
CO
CO
0)
o-
-------
Age - Pipeline System
Spill statistics for the frequency of spills as a function of age for
the pipeline system are presented in Table 31 and illustrated in Figures 38
and 39 for the period between 1971 and 1975. The number of miles of pipe-
line installed in each of the time periods was taken from Table 12.
The number of incidents per 1000 miles of pipeline is illustrated in
Figure 38. Results show that spill incidents are about the same, or about
0.8 spills per 1000 miles of line pipe, for pipeline systems installed af-
ter 1950. In contrast, spill rate increased rapidly as a function of pipe-
line age for lines built before 1950.
Table 27 shows that external corrosion caused over 40 percent of the
spills for pipelines installed before 1950 while internal corrosion caused
slightly less than 10 percent of the spills. Line rupture by excavation
equipment and defective pipe seams caused the following: (a) over 38 per-
cent of incidents for pipeline systems (12 inches in diameter or less) in-
stalled after 1950; (b) over 30 percent of the spills for pipeline systems
installed before 1950; and (c) over 55 percent of spills for pipeline sys-
tems greater than 12 inches in diameter.
Age - Line Pipe
Spill statistics of the frequency of spills and spill volume as a
function of age for the line pipe component of the pipeline system are pre-
sented in Table 32 and illustrated in Figure 40 for the 5-year period be-
tween 1971 and 1975. Additionally, major causes of spills for two typical
years (1972 and 1974) are illustrated in Figure 41 for line pipe age.
Results in Table 32 and Figure 40 show that spill incidents per 1000
miles increase with line pipe age. Lines built before 1950, in particu-
lar before 1930, show a significantly high incidence of spills, whereas
lines built after 1970 show a relatively low incidence of spills (about
six times lower than for lines built before 1930).
Depth of Cover
Number of incidents as a function of depth of cover of the line are
presented in Table 33. Mileage, geometry and age of line for each depth
are not reportable and not included in the table. A discussion of this
data is presented in Section 6.3.3.1.6.
Fatalities and Injuries
The number of fatalities and injuries resulting from petroleum inci-
dents over the 9-year period from 1968 through 1976 are given in Table 34.
Over this 9-year period the number of deaths averaged less than 4 and in-
juries less than 7. The number of fatalities and injuries occurring dur-
ing the transportation of crude and products was about the same.
115
-------
TABLE 31. SPILL INCIDENTS FOR PIPELINE SYSTEM AGE
AND ESTIMATED3 MILEAGE FOR ACCIDENTS REPORTED
TO OPSO BETWEEN 1971-1975
Year Installed
Before 1938
1930 - 1939
1940 - 1949
1950 - 1959
1960 - 1969
1970 - Present
Unknown
TOTAL
1971
Number Spill/
Spills 103 Miles
53 4.51
53 4.0
52 2.1.
60 0.9
50 0.7
11 0.4
28
307
X
of
Spills
17
17
17
20
16
4
9
Number
of
Spills
47
26
39
75
67
21
34
309
1972
Spill/
103 Miles
4.2
1.9
1.6
1.1
0.9
0.9
%
of
Spills
15
8
13
24
22
7
11
1973
Nu$*r Spill/
Spills 103 Miles
39 3.5
34 2.5
44 2.0
59 0.9
51 0.7
15 0.6
25
272
%
of
Spills
14
13
18
22
19
6
9
Year Installed
Before
1930 -
1940 -
1950 -
1960 -
1970 -
Unknown
TOTAL
1938
1939
1949
1959
1969
Present
Number
of
Spills
53
53
52
60
50
11
28
307
1974
spin/
103 Miles
2.8
2.9
1.8
0.8
0.8
0.9
X
Of
Spills
12
15
17
21
22
9
4
Number
of
Spills
36
35
48
32
68
26
10
255
1975
Spill/
103 Miles
3.2
2.6
2.0
0.5
1.0
1.1
1971-1975
X
Of
Spills
14
14
19
13
27
10
4
Total
Spills
206
187
232
279
291
95
108
1,398
X of
Total
15
13
17
20
21
7
8
Average
1 of S'
Spills 10J
41
37
46
56
58
19
21
280
>111/
Miles
3.7
2.6
1.9
0.8
0.8
9.7
Source: See Table 13.
116-
-------
400
300
200
TOO
c
QJ
•o
•?•
u
80
60
40
20
• All spills
A Before 1930
v 1930-1939
A 1940-1949
D 1950-1959
I 1960-1969
o 1970-1975
1
A
V
1971
1972
1973
Year
1974
197$
1971-
1975
Figure 38. Spill incidents by year of pipeline system installation
for accidents reported to OPSO between 1971-1975.
117
-------
LU
o
o
o
=1 2
3.
O
=:
1 —
A Before 1930
v 1930-1939
A 1940-1949
S 1950-1959
• 1960-1969
1970-1975
1971
1972
19/J
Year
1974
1975
Avg
1971-
1975
Figure 39. Pipeline system spill incidents per 1000 miles of pipeline
by the year of pipeline system installation for accidents
reported to OPSO between 1971-1975.
118
-------
TABLE 32. SPILL INCIDENTS AND VOLUME FOR LINE PIPE AGE FOR ACCIDENTS
REPORTED TO OPSO BETWEEN 1971 AND 1979
Year Installed Mlleige
Before 1930 11,172
1
1930 - 1939 13.412
1940 - 1949 24.588
19SO - 1959 69.295
1960 - 1969 71.531
1970 - 1979 53.000a
Unknown 8.941
1971
Number Spill I
of .03 „,
Jpflll 10 "'•
51 4.56
50 3.72
47 1.91
47 0.67
39 0.54
5 0.47
; (10.618
; Hlles) a
20 2.23
i
TOTAL
J259
Volume
of
Spills
(BbU)
26,432
30.514
36.556
54.637
53.414
4.366
6.781
212.700
»bl*
103 HI.
2.36
2.27
1.48
0.78
0.74
0.41
(10,618
Mllesja
0.75
1972
Number SpJIU
of ,03 M
Spill* I0 Hl-
42 3.75
25 1.86
32 1.3
65 0.93
38 0.53
S 0.31
(15.92{
MMesJ'fl
27 3.01
234
Volume
of
Spills
(Bbls)
41.705
15.846
35.764
99.350
69.236
6.003
12.578
280.482
»"»
103 Ml.
3.73
1.18
1.45
1.43
0.96
0.37
(15.927
Hlles) 3
1.46
1973
Number
Of
Spills
36
25
41
44
32
4
21
203
spin*
103 HI.
3.22
1.86
1.66
0.63
0.44
0.19
(21.236
Hlles) A
2.34
Volume
of
Spills
(Bbls)
134.942
15.544
40.894
118.341
47.896
5.345
7.271
370.233
Bbl*
10* HI.
12.07
1.15
1.66
1.70
0.66
0.25
(21.236
Hlles)a
0.81
continued)
Estimated mileage after January 1, 1970 over indicated period (based on new
pipeline mileage from 1971 through 1977).
Estimated.
-------
TABLE 32 (Continued)
ro
o
Year Installed Mileage"
Before 1930 11.172
1930 - 1939 13.412
1940 - 1949 24.588
1950 - 1959 69.295
I960 - 1969 71.531
1970 - 1979 53.000 a
Unknown 8.941
TOTAL
1974
Number Spills Volume Bbls
of H? Ml of in3 Ml
Spills SplUs
(Obis)
25 2.23 19.627 1.77
35 2.60 15.424 1.15
37 1.50 40.488 1.64
43 0.62 40.311 0.58
41 0.57 106.071 1.48
10 0.36 32.586 1.23
(26.S45 (26.545
Mlles)a Miles) fl
8 0.89 1.894 0.21
199 256,601
1975
Number Spills Volume
of lf,3 „, of
Spills IU "'• Spills
(Bbls)
31 2.77 7.436
31 2.31 20.225
42 1.70 28.742
27 0.38 60.810
44 0.61 106.071
8 0.25 32.586
(31.854
Miles) 3
3 0.33 2.579
186 258.449
Bbls
103 Ml.
0.66
1.50
1.16
0.87
1971
Total
Nunber Volume
of of
Spills SpllU
(Bbls)
185 230.342
166 97.553
199 182.444
226 373.449
1.48 194 382.688
1.02 32 65. 60S
(31.854
Hllesja
0.2B
79 31.103
1.081 1.378.465
- 1975
Average
SpUU Bbls
3.31 4.12
2.47 1.45
1.61 1.48
0.66 1.07
0.54 0.87
0.31 0.62
(21.200 (21.200
Miles) " Miles)8
1.76 0.69
Estimated mileage after January 1, 1970 over indicated period (based on new
pipeline mileage from 1971 through 1977).
Estimated.
-------
5 r
A Before 1930
V 1930-1939
A 1940-1949
Q 1950-1959
• 1960-1969
O 1970-7975
o
o
o
2? 3
2 2
OC.
LU
CC
©----
L
1971
©
I
1972
. -0--- .
©-•
1973
Year
1974
O
1975
O
—I
Avg
1971-
1975
Figure 40. Line pipe spill incidents per 1000 miles of pipeline
by the year of pipeline system installation for accidents
reported to OPSO between 1971-1975.
121
-------
YEAR 1974
-o
o
c
o
HI
"§
YEAR 1971
*5 •^••^
~ -Q g
Before 193° 1930-1939
AGE OF LINE
1950-1959 1960-1969
Figure 41. Age of line pipe and causes of spill incidents per
1000 miles of line pipe for accidents reported to OPSO
in 1971 and 1974.
122
-------
TABLE 33. NUMBER OF OUTSIDE FORCE INCIDENTS BY DEPTH OF COVER
;e?tfi of Cover
(incnes)
0
1 - 10
11-20
21 - 30
31 - 40
41 - up
Not applicable
Not re^ortea
Total
Number of
1963
6
14
13
30
30
--
5
98
1969
6
16
23
23
8
12
2
90
1970
3
10
16
20
12
9
70
1971
2
9
16
17
15
7
1
67
1972
2
10
14
16
16
10
1
69
Incidents
1973
0
9
9
23
13
8
4
66
1974
1
6
18
21
15
10
7
78
1975
• 1
6
14
12
18
18
4
73
1976
0
15
14
12
16
5
1
3
67
9-yr
21
95
137
175
143
79
4
24
678
Percent of
Total
3.1
14.0
20.2
25.8
21.1
11.6
0.6
3.5
123
-------
TABLE 34. FATALITIES AND INJURIES FOR PETROLEUM SYSTEM INCIDENTS
REPORTED TO OPSO BETWEEN 1968 AND 1976
Incidents3
Year
1968
1969
1970
1971
1972
1973
1974
1975
1976
1968-1976
1971-1975
Petroleum
Crude
Products
TOTAL
Crude
Products
TOTAL
Crude
Products
TOTAL
Crude
Products
TOTAL
Crude
Products
TOTAL
Crude
Products
TOTAL
Crude
Products
TOTAL
Crude
Products
TOTAL
Crude
Products
TOTAL
Crude
Products
Avg. Crude
Avg. Product
Avg. Total
Crude
Product
Avg. Crude
Avg. Product
Avg. Total
Number
of
Spills
322
134
456
246
109
355
216
107
323
172
97
269
180
107
287
155
113
268
166
60
226
135
94
229
123
72
195
1715
393
190
99
289
308
471
161
94
256
Injuries
12
3
20
1 •
0
1
1
1
2
2
6
3
.7
9
16
0
5
5
0
3
3
1
1
2
2
0
2
26
33
3
4
7
10
24
2
5
7
Deaths
1
1
2
1
0
1
1
7
3
0
1
1
2
1
3
0
0
0
10
0
10
1
0
1
0
5
5
16
15
2
2
4
13
2
3
1
4
LPG spills are not included in this table.
124
-------
Damage from Crude and Product
Damage figures from OPSO records include both property losses (dollars)
and loss of commodity (barrels spilled). Table 35 summarizes these crude
and product losses from petroleum pipeline system incidents for the 9-year
period from 1968 through 1976. Products include gasoline, fuel oil, jet
fuel, diesel fuel, condensate, kerosene and anhydrous ammonia. Ftgure 42
shows the variations in volume spilled, number of incidents, and property
damage between 1968 and 1976 for crude and product transferred in the
petroleum pipeline system. Table 30 shows that total property damage
losses average $1,246,405 yearly over this 9-year period and the average
loss for the 2608 spills was $430. Figure 42 shows a slight increasing
trend of the property damage even though the number of incidents and spill
volume showed a general declining trend.
Product spills were considerably more costly than crude spills. Re-
sults in Table 34 show that the average cost of property damage of an in-
cident for the 9-year period from 1968 through 1976 was about 60 percent
higher for products than for crude or $5786 and $3544, respectively. Be-
tween 1971 and 1975 the average cost of property damage for products was
almost double that of crude, $8090 and $4181, respectively. This occurred
even though the average spill size was about the same, 1009 and 1027 bar-
rels, respectively. Similar property damage costs occurred for crude and
product during these two periods, because pipeline systems incidents for
crude were almost double those for product. For example, 1715 incidents
for crude and 893 incidents for product were reported during the 9-year
period from 1968 through 1976.
4.6.2.2 Loss Mechanism—Oil Spills in Outer Continental Shelf-USGS
Statistics—
The U.S. Geological Survey is responsible for recording and investi-
gating oil spills in the outer continental shelf. Since 1971, all spills
in excess of 50 barrels have been recorded and a computer file is kept on
all spills of one barrel or more. Spills of less than one barrel are as-
sumed to be one-fourth barrel.
Total Spills—Volume, Throughput. Frequency
From January 1, 1971 to December 31, 1975, USGS27'28 recorded oil
spillage amounted to 51,421 barrels. Production during this period amounted
to 1,810,996,199 barrels. The barrels spilled each year averaged 10,284
and barrels spilled per barrel of throughput was 0.0028 percent. Table 36
shows a summary of the number and volume of spills each year. Of the 5,857
spills recorded, 85.5 percent of the total spill volume was contributed by
five spills; and 88.7 percent of the total spill volume was contributed by
the 27 spills that were 50 barrels or more. Figure 43 shows the number and
volume of 1 to 50 barrel spills for the years 1971 through 1975.
125
-------
TABLE 35. DAMAGE FROM PETROLEUM PIPELINE SYSTEMS IN THE U.S.
FOR INCIDENTS REPORTED TO OPSO BETWEEN 1968-1976
Year
1 968
1969
1970
1971
1972
1973
1974
1975
1976
1968-
1976
1971-
1975
Petroleum
Crude
Products
TOTAL
Crude
Products
TOTAL
Crude
Products
TOTAL
Crude
Products
TOTAL
Crude
Products
TOTAL
Crude
Products
TOTAL
Crude
Products
TOTAL
Crude
Products
TOTAL
Crude
Products
TOTAL
Crude
Product
Avg Crude
Avg Prod
Avg Total
Crude
Product
Avg Crude
Avg Prod
Avg Total
Loss3
Property
Commodity Property Damage
(Barrels) (Dollars) (I/Barrels)
222,257
107,154
329,416
174,540
89,905
264,545
383,160
92,348
475,508
115,670
89,410
205,170
233,690
96,555
330,245
208,373
119,4-18
327,791
168,738
60,824
329,562
105,371
108,324
214,195
96,059
123,363
224,422
1,356,659
•1,163,969
206,295
129,329
335,624
332,432
474,531
165,486
94,906
260,392
366,072
481 ,437
1,347,509
1,176,260
403,447
1,579,707
413,756
120,035
533,791
117,938
421,187
539,125
760,798
1,508,464
2.269,262
411,080
650,512
1 ,061 ,592
635,666
155,150
790,316
1,441,764
1,069,447
2,511,211
238,523
346,129
584,652
6,061,357
5,155,798
673,539
572,866
1 ,246,405
3,267,246
3,304,760
673,249
760,952
1,434,001
3.39
4.49
4.09
6.73
4.48
5.97
1.07
1.299
1.12
1.02
4.71
2.63
3.26
15.62
6.37
1.97
5.45
3.23
3.77
2.55
2.39
13.62
9.37
11.73
2.48
2.67
2.605
3.26
4.42
3.26
4.42
3.71
4.04
8.01
4.04
8.01
5.50
Number
322
134
456
246
109
355
216
107
323
172
97
269
180
107
287
155
113
268
166
60
226
135
94
229
123
72
195
1715
393
190
99
239
308
471
161
94
256
Spills3
Property Commodity
Damage Barrels/
(S/Spill) Spill
•
3,544 1.085
5,786 1.306
4,312 1,161
4,181 1,027
3,090 1,009
5,600 1,017
aNote: LPG spills and losses are not included in this table.
126
-------
500,000
t 400,000
IB
3 300,000
in
o
200,000
« 100,000
500
«j 400
•I
u 300
° 200
i 100
2,500,000
i 2,000,000
I
~ 1,500,000
CD
i
-------
TABLE 36. NUMBER AND VOLUME OF SPILLS EACH YEAR, 1971-1975, GULF OF MEXICO
OUTER CONTINENTAL SHELF9
r\>
oo
Spills of leas than 50 barrel*
Year
-- "
1971
1972
1973
1974
1975
1971
1972
1973
1974
1975
aSource
U T -I 1
Leas than
• 1 bbl
913
953
1,005
1,051
1,021
1^14
bbl
321
202
162
75
100
228 1,008
238 703
251 585
263 329
255 344
: Reference
25.
1 1 _ _j
15-49
bbl
11
4
4
3
5
257
91
85
75
112
c en u«
Total
1,245
1,159
'1,171
1,129
1,126
Total
1,493
1,032
921
667
711
50-499
bbl
Number of
11
2
1
6
2
Spills
500-999
bbl
apilla
0
0
0
0
0
vqliiM (barrels) of
1,285
150
240
675
266
0
0
0
0
0
of 50 barrels or nore"
1,000-
9,999 bbl
0
0
3
1
0
spills
0
0
21,935
2,213
0
10,000 bbl
or more
0
0
0
1
0
0
• o
0
19.833
0
Total
11
2
4
8
2
1,285
150
22,175 .
22,721
266
All
apilla
Total
1,256
1,161
1,175
1,137
1,128
2,778
1,182
23,096
23,388
977
/
and 1974.
5 spills in 1971, and 1 spill each in 1972
-------
IOOO
1400
1200
3 1000
i
u.
O
3
Ul
c
c
< 800
B
Ul
O
<
H
0 600
400
200
n
-
•
-
,
MB
15V
':>:•
*•".".
".". '•
.:'••;-
.'•:•••
•'.'O.
:>::
847
337
r/T1
'. ' ;
:":".-'!•
'.'.".•.
V-. j
V////////////A*
670
•
•;V
»'.';
W/////////A
1
•j Volume of spill!
Jj Number of spills
a
454 456
'••:'•
^
'•;•'•.'
105
I
-
-
—
IOUU
1400
1200
1000
in
d
a>
u.
800 °
C
Ul
|
Z
600
400
200
n
Figure 43. Number and volume of 10 to 50-barrel spills by year, 1971-75,
Gulf of Mexico outer continental shelf. (Source: Reference 25.)
129
-------
4.6.2.3 Loss Mechanism— Underwater OCS Pipeline Spills in the U.S. (OCS
Statistics) —
A total of 17 spiTls exceeding 50 barrels were reported for underwater
pipelines in the Gulf of Mexico Outer Continental Shelf between 1966 and
1976. It is estimated that there is 2500 miles of pipeline. Thus, the
spill frequency are estimated to be:
<> 50 barrels) .
A total of 236 spills occurred from these underwater pipelines between
1966 and 1976; 85 percent of the total reported spills in the Gulf of
Mexico were less than 50 barrels.
The spill frequency for all spills is estimated to be:
236 spills 1 = Q 4 y 10'3 a|JI"3 fall sums)
10 years * 2500 mfles x iu mile-year^311 SP1NS>
4.6.2.4 Loss Mechanism—Oil Spills in and About U.S. Waters (USCG Sta-
tistics)—
PIRS Statistics of oil spills in and about U.S. waters compiled by
the USCG include spills from all sources. These spill statistics cover
a broad area but are not as comprehensive as the OPSO data for pipeline
system spills.
Spills—Volume, Number, Location, Cause
From 1970 to 1975, total losses, number of spills and average volume
in and about U.S. waters29 are shown in Figure 44. The general areas and
location in which these losses occurred in 1975 are given in Table 37.
Sources and causes of losses are given in Tables 38 and 39.
4.6.2.5 Gathering/Distribution, Drilling and Production Spills Between
1970 and 1971—
Early spill data (see Tables 40-42), recorded primarily during the
2-year period between 1970 and 1971, were tabulated and reviewed in Refer-
ence 30 for U.S. and foreign data sources (see Table 43). Over 15,000 oil
spill events from gathering/distribution/drilling and production systems
were reviewed. Of these, 8,473 oil spills were applicable to the study.
Detailed information is provided on spills, such as component fail-
ures, causes of failures and spill distribution. However, factors relat-
ing to pipeline system spills, such as component age and geometry, and
information on total spill volume and throughput are not included in the
study.
Tables 40-42 provide a summary of component versus quantity spilled
category by system for gathering/distribution, drilling and production
130
-------
Gallons
of oil
discharged
i
i *
! o
r-^r-
\ o
j CM
! "•"
i <*">
\
! oo
i
o"
VO
0%
co
CM
to
r>.
VD
t
§
o
CO
CO
s
•^^^•M^H
j
co
CM
co
§
o
CM
O
CO
f— 1
CT>
CO
CO
to
cr>
CO
i— i
Number
of
discharges
!
CM
CM
If)
^
O
CO
CO
co"
""^^~~~"" 1
CO
c
o
f-H
r—t
0
1— 1
— «^^
r— <
2
Average
volume
per
discharge
i
co us r^ co co
r^. P-- CM CM co
CM «3- CO CO CO
co o vo T-H co
•!*• O r*> co CM
r— i o co co ^r
i— I CM <— I t—t i-f
|
i ' — -*~
J 1
r-4 CM O •«- ID t-( CM CO «i- U" r-4 CJ 0 «S- LT5
CT^ CTt O") CTt CT*t Cft C7> CTl G^ CTi CTi C"i CTi C7. CT^
,_! ,-H <— IT— 4 rH i— If— li— 1 i— 1 i— 1 i—ir— 1 i— 1 i— 1 r— 1
*Barrels
Figure 44. Discharge trends for oil only for calendar years
1970-1975. (Source: Reference 29.)
131
-------
TABLE 37. GENERAL AREAS AND LOCATIONS OF SPILLAGE IN AND ABOUT
U. S. WATERS3, REPORTED TO USCG IN 1975a
Item
General Area:
Atlantic Coast
Gulf Coast (West of Long. 83°15l)
Pacific Coast
Great Lakes
Inland U.S.
Total
Type of Location:
Inland Waters
Roadsteads
Ports
Beaches
River areas
Non-navigable areas
Total
Coastal Waters (including Great
Lakes)
Bays, estuaries and sounds
Ports
Beaches
River areas
Non-navigable areas
Open waters (Great Lakes or
territorial sea)
Total
Contiguous Zone
High Seas
Total
Number of
incidents
2,695
3,315
1,768
454
1,909
10,141
120
472
96
836
112
1,636
1,117
3,937
159
1,272
192
955
7,632
188
692
10,141
% of
Total
26.6
32.8
17.4
4.4
18.8
100.0
. 1.0
4.8
0.9
8.2
1.1
16.0
11.0
38.9
1.6
12.5
1.9
9.4
75.3
1.9
6.8
100.0
Volume in
gallons
1,465,689
5,430,212
440,923
307,772
6,795,001
14,439,597
37,164
1,106,611
88,537
3,816,371
303,752
5,352,435
3,227,563
2,173,847
108,588
2,008,819
214,447
453,683
8,136,947
5,725
894,490
14,439,597
% of
Total
10.2
37.6
3.0
2.1
47.1
100.0
0.3
7.7
0.6
26.4
2.1
37.1
22.3
15.1
0.8
13.9
1.5
3.1
56.7
0.0
6.2
100.0
Source: Reference 29.
-------
TABLE 38. CAUSES OF SPILLS IN ABOUT U.S. WATERS,
REPORTED TO USCG IN 19753
Item
Hull/tank rupture/leak
Transportation pipeline
rupture leak
Other structural failure
Pipe rupture/leak
Hose rupture/ leak
Valve failure
Pump failure
Other rupture/ leak
Other equipment failure
Tank overflow
Improper handling operation
Other personnel error
Bilge pumping
Ballast pumping
Other intentional discharge
Natural or chronic
phenomenon
Unknown/Mi seel 1 aneous
TOTAL
Number of
incidents
757
235
200
954
220
256
no
29
1,069
464
679
511
195
21
269
352
3,820
10,141
% of
Total
7.5
2.3
2.0
9.5
2.1
2.5
1.0
.2
10.5
4.5
6.5
5.1
1.9
.2
2.6
3.5
38.1
100.00
Volume in
gallons
6,359,433
1,200,004
2,744,656
1,322,385
120,203
251 ,608
39,499
94,683
221 ,956
473,454
276,226
504,976
14,821
754
196,089
144,115
474,735
14,439,597
X Of
Total
44.0
8.3
19.0
9.2
0.8
1.7
0.3
0.7
1.5
3.3
4.3
3.5
0.1
0.0
1.4
1.0
0.9
100.00
Source: Reference 29.
133
-------
TABLE 39. SOURCES OF SPILLS IN AND ABOUT U.S. WATERS,
REPORTED TO USCG IN 1975a
T Number of
Item incidents
Vessels:
1. Dry cargo ships
2. Dry cargo barges
3. Tank ships
4. Tank barges
5. Combatant vessels
6. Other vessels
Total
Land Vehicles:
1. Rail vehicles
2. Highway vehicles
3. Other/ unknown vehicles
Total
Noo-Transportation-Related
Eaci Titles:
1. Onshore refinery
2. Onshore bulk/storage
3. Onshore production
4. Offshore production
facilities
5. Other facilities
Total
Pipelines:
Marine Facilities:
1. Onshore/offshore bulk cargo
transfer
2. Onshore/offshore fueling
3. Onshore/offshore nonbulk
cargo transfer
4. Other transportation-related
marine facilities
Total
Land Facilities
Misc. Unknown
Total
277
31
643
757
202
1,143
ITblJ
27
263
20
~3To
176
305
233
1,243
762
2771T
564
250
74
19
80
"423
167
2,905
10,141
X of
Total
2.7
0.3
6,3
7.5
2.0
11.3
357T
0.4
2.6
0.1
~TT
1.7
3.0
2.3
12.3
7.5
2571
5.6
2.5
0.7
0.2
0.8
TT
1.6
28.6
100.00
Volume in
gallons
21 ,843
5,215
1,766,729
3,467,203
16,913
1,353,947
6,531,850
576,507
356 ,601
2,617
935,725
145,722
476,768
2,626,992
78,217
567,924
3,895,623
2,490,237
81 ,203
9,388
1,326
7,239
99,156
200,962
186,044
14,439,597
% of
Total
0.2
0.0
12.2
24.0
0.1
9.4
1571"
4.0
2.4
0.0
"BTT
1.0
3.3
18.2
0.5
4.0
TFQ
17.3
0.6
0.0
0.0
0.0
0.6
1.5
1.3
100.00
aSource: Reference 29.
134
-------
TABLE 40. COMPONENT VERSUS QUANTITY SPILL CATEGORY FOR GATHERING/
DISTRIBUTION SYSTEM, 1970-19713
Source: Reference 30.
135
-------
TABLE 41. COMPONENT VERSUS QUANTITY SPILL CATEGORY FOR DRILLING SYSTEM,
1970-19713
TSS?
5B"
1ST
4 . 10
11
Source: Reference 30.
136
-------
TABLE 42. COMPONENT VERSUS QUANTITY SPILL CATEGORY FOR PRODUCTION SYSTEM
AND. ALL SYSTEMS, 1970- 1971a
f#U Hmuarr
wan
sax.
Touil
•W
TBT
1
II
— • • !•-,•• 1 »' " •»/ •• ^ ••»•
il
¥
.22.
IT
-LL
M» KBJI*I»
TT
S3
• 5
I
15
TT
Source: Reference 30.
137
±
!!
TT
-iir
-------
TABLE 43. SOURCE DATA SUMMARY4
U. ««!,«.. nf
Item Records Used
Federal :
EPA Headquarters - OHM File
EPA Anchorage, Alaska
DOT/FRRC - Pipeline Carrier Accident Reports
U.S. Geological Survey, Conservation Division,
Washington, D.C. Office
States:
California - Department of Fish and Game
Commission
California - Water Quality Control Board
California - Division of Oil and Gas
California - City of Long Beach, Department
of Oil Properties
Colorado - Department of Health, Hater Pollu-
tion Control Division
Louisiana -' Department of Conservation,
Division of Minerals
Mississippi - Oil and Gas Board
New Mexico - Oil Conservation Coiiiulsslon
Oklahoma - Corporation Commission Oil and
Gas Conservation Division
Texas - Railroad Commission Oil and Gas Division
Other:
Alberta. Canada - Alberta Oil and Gas Conser-
vation Board
Source: Reference 30.
108
68
802
633
63
19
40
30
10
707
202
58
399
5094
235
Type of Report
Table
Form Narrative Number
X 0-1
X D-2
X 0-3
X D-4
X 0-5
X 0-6
X D-7
X D-8
X D-9
X D-10
X D-ll
X D-12
X D-13
X 0-14
X D-15
Systems
Reported"
G/D.
D.P
G/D
G/D.
G/D.
G/D.
G/D.
G/D.
G/D,
G/D,
G/D.
G/D.
G/D.
G/D,
G/D.
D, P
D. P
P
D, P
D. P
0, P
P
P
P
P
0, P
0. P
0. P
UG/D - Gathering/Distribution Systems
D - Drilling Systems
P - Production Systems
-------
systems during the 1970 through 1971 period. Figure 45 illustrates data
from Tables 40-42 for pipeline spills and total spills. These data show
that of the 8,453 total spills, over 7,900 spills greater than 2.5 bar-
rels occurred for all oil system components. Of the total~4,646 pipe
spills, 4,500 spills were greater than 2.5 barrels, 3,815 spills greater
than 10 barrels and 1,413 spills greater than 100 barrels. Pipe spills
represented over 81 percent of the 4,423 spills from the gathering/
distribution system and over 39 percent of the 3,993 from production sys-
tems. Spills from drilling systems were negligible, representing less
than 1 percent of the total number of spills.
4.6.2.6 Loss Mechanism—Cross-Country Oil Spillage in Europe and Canada-
Cross-country spillage data in Europe are compiled annually by
Stichting Concawe Oil Pipeline Special Task Force31'37. Reported spill
information includes date, cause, amount of spillage, damage, cost of
cleanup, pipeline dimensions and total pipeline length. Spill volumes
below about 1 barrel were not reported. Similar spillage data for Canada
pipelines is also compiled periodically.
Spill Volume for Mileage and Throughput
A summary of the reported spill data is tabulated in Table 44. Re-
ported spill volume between 1966 and 1976 averaged about 9,500 barrels of
petroleum a year. The spill volume rate was estimated by dividing the
average volume of spills (1966-1976) by the average pipeline mileage
(1971-1975):
9.538 barrels/year _ -, 06 barrels spilled
8,941 miles mile-year
Reported spill volume between 1971 and 1975 averaged 8,865 barrels. The
spill volume rate was estimated by dividing the average volume of spills
by the average pipeline mileage for those years:
8,865 barrels/year , m barrels spilled
8,772 milesmile-year'
These average volumes spilled are similar to that reported for the U.S.
pipeline.
The average percentage of the volume of oil spilled for the barrels
of throughput in Western Europe between 1966 and 1976:
104.966 barrels (spilled) _1 = Q QQ.« , ,, 111O
25,240,000,000 barrels (throughput) x 11 u-uutf/0 van spins;
This percent loss of throughput is an order of magnitude better than for
the U.S. pipelines.
139
-------
4000
3500
3000
2500
- 2000
o
i.
Ol
1 1500
1000
500
Pipe spills for
gatheri ng/di stri buti on.
Pipe spills for
production system.
Petroleum pipeline system
spills for both gathering/
distribution and
production system
0 1.1 2.6 11 101 239 501 1001 5GOO
to to to to to to to to to
1 2.5 10 100 238 500 1000 5000 10000
Barrels Spilled
10COO
or
Figure 45. Pipe and petroleum pipeline system oil spills in
U.S. for gathering/distribution and production systems
in the United States between 1970-1971.
(Source: Reference 30.)
the
140
-------
TABLE 44. CROSS-COUNTRY SPILL DATA FOR WESTERN EUROPE
Loss
Throughput
Year
1966
1967
1968
1969
1970
1971
1972
1973
1974
1975
1976
1966-
1976
Total
Average
1971-
1975
Total
Average
(Barrels)
1,333
1,408
1,484
1,559
1,572
1,940
2,723
3,509
3,290
3,037
3,390
25,245
2,295
14,499
2,899
,000
,000
,000
,000
,000
,000
,000
,000
,000
,000
,000
,000
,000
,000
,000
,000
,000
,000
,000
,000
,000
,000
,000
,000
,000
,000
,000
,000
,000
,000
Lengtha
(Miles)
NA
NA
NA
NA
NA
6,920
8,540
9,351
9,378
9,675
9,783
53,647
8,941
43,864
8,772
%
Through-
Barrels
9
2
3
4
17
5
6
12
2
3
104
9
44
8
,603
,075
62
,867
,522
,176
,660
,735
,260
,496
,165
,926
,538
,327
,865
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
put
00072
00015
000004
00025
00028
0009
0006
0002
0004
00008
00009
0004
0003
Barrels
Mile
NA
NA
NA
NA
NA
2.48
0.66
0.72
1.3
0.26
0.32
1.06
1.01
Incidents
Spills
5
5
2
6
n
n
21
20
18
20
14
133
12
90
18
ID'3
Spills
Mile
NA
NA
NA
NA
NA
1.59
2.4
2-1
1.9
2.1
1.4
1.3
2.0
Not reported in References 31-37.
141
-------
Spill Frequency for Pi pi line Mileage
The number of spillages reported between 1966 and 1976 averaged about
12 and showed a significant increase over this period". The spill rate was
estimated by dividing the average number of spills by the average pipeline
mileage (1971-1976):
12 spills/year _ . , n-3 spills
8,941 miles " ltd x iu mile-year '
Reported number of spills between 1971 and 1975 averaged 18. The
spill rate for the period was obtained by dividing the average number of
spills by the average pipeline mileage:
18/year 3 spills
8,772 miles ^UD x iu mile-year
This spill rate is similar to that reported in the U.S. if statistical
estimated of spills of less than 50 barrels are included in the U.S. total.
Spill Frequency for Pipeline Size
A relationship of mean spill size to pipe diameters based on European
(1971-1974) and Canadian (1965-1972) data is presented in Reference 37.
Figure 46, see Reference 39, shows a plot of the mean spill size and pipe
diameter. Also included in the plot are U.S. data (1971-1975) from OPSO
statistics. Results show that mean spill size was a function of pipe
diameter to the exponent 2.27 and essentially proportional to the cross-
sectional area of the pipe. Insufficient data were available to establish
a relationship of pipe diameter with either spill frequency or volume
transported.
Spill Size Distribution
Spill size distribution based on European and Canadian data (see Ref-
erence 39) is presented in Figure 47. The standard deviation for 1971-
1974 European data is used for both distributions. The mean spill size in
the Canadian data was 3,407 barrels, and 666 barrels for the European data.
Spill Probability for Various Pipeline Diameters
The recurrence and probability of spills from various pipeline diam-
eters can be obtained using the distribution and the following equations:
R(S) =_ - (1)
N[lOO-F(S)]
•JL[100-F(S)] TjfL
P(S) = l-e m = l-eR(S)
(2)
- FT J • (3)
. 142
-------
10,000 u
- o
— A
,000
CQ
CQ
UJ
a.
oo
100 _
r
i
10
kl <*
i1/
Western Europe Spill Data
1971-197425'28
Canada Spill Data
1965-1972"°
U.S. Spill Data
1971-197519
Numbers Adjacent to
Data Points Denote
the Number of
Spills Making Up 1
Each Mean
4?
—
S=2.06 D
Correlation Coefficient
= 0.84
j L
10
PIPE DIAMETER, D (INCHES)
100
Figure 46. Cross-country pipeline mean spill size versus pipe diameter
143
-------
10*
10"
I
ca -3 i
e ioj \-
UJ
101!-
10
0
Proposed Distribution Based
on Canada Data, 1965-1972^.
Arith. Mean Spill Size
= 3,407 BBL
Western Europe Spill ^
Data, 1971-19748-11
Proposed Distribution Based on
Western Europe Spill data, 1968-
19748-12. Arith. Mean Spill
Size = 666 3BL
i i i i i i i '
0.01 0.1 1
10 40 60 80
95 99
99.9 99.99
PERCENT PROBABILITY THAT A SPILL WILL 3E SMALLER THAN THE
SIZE INDICATED; F(S). (GIVEN THAT A SPILL OF
UNKNOWN SIZE HAS OCCURRED)
Figure 47. Cumulative size distribution for oil industry
cross-country pipeline spills, Western Europe and
Canada (assorted pipe diameters). (Source:
Reference 39.)
144
-------
where
R(S) = recurrence interval (years) for spills of size (> S (i.e.,
the average time interval between spills of size~greater
than or equal to S)
N = mean number of spills per year
F(S) = cumulative spill size distribution function (i.e., the per-
cent probability that a spill will be smaller than size S,
given that a spill of unknown size has occurred). F(0) = 0
P(S) = fractional probability of having at least one spill of size
> S during time T (assumes Poisson distribution for spill
frequency)
T = time interval of interest (years)
V = average annual spillage (bbl)
T = mean spill size (bbl).
4.7 TOTAL PETROLEUM LOSSES—PIPELINE SYSTEMS
Total petroleum losses from pipeline systems in the United States are
estimated to be approximately 0.16 percent to 0.22 percent of throughput.
The exact petroleum throughput in the U.S. is uncertain because of routing.
From BOM statistics of volume of petroleum transported in 1975 by pipelines
(see Table 5), approximately 8.0 billion barrels are transported. Crude
transported was about 4.7 billion barrels, or slightly greater than one-
half of the total petroleum transported. Assuming losses of about 0.22
percent of throughput (0.16 percent crude, 0.33 percent product), this re-
sults in an estimated loss of 17.2 million barrels by pipeline systems.
Cumulative estimates of annual losses for individual mechanisms are as
follows: about 0.05 percent from evaporation; 0.05 percent from admixture
shrinkage; and 0.02 percent from coefficient of expansion errors. Losses
from spills are uncertain because of the various reporting problems indi-
cated in Section 4.6.2. However, it appears that total annual spillage
losses can range from as low as 0.01 percent of throughput or about 800
thousand barrels to as high as 0.03 percent of throughput or about 2.4
million barrels.
145
-------
SECTION 5
PIPELINE SYSTEM LEAK DETECTION AND INSPECTION METHODS
Leak detection and inspection methods for petroleum pipeline systems
which might reduce the risk of oil spillage, i.e., reduce frequency of
spills or limit the volume of oil lost, are presented in this section. Be-
cause of the high incidence of pipeline (line pipe) spills, i.e., high risk
of oil spills, and the potential for significant reductions of the oil spill
risks, major emphasis is given to detecting and preventing leaks from line
pipe failures. However, methods are also included for other pipeline sys-
tem components.
A general discussion of leak detection, inspection and maintenance is
given in Section 5.1. Pipeline system component failures, leakage and dis-
charge rates are discussed, and difficulties associated with leak preven-
tion, detection and location are identified.
Section 5.2 provides a general description of leak detection and in-
spection methods. Type, mode, application, use and installation are defined
and discussed. Methods for detecting liquid leaks and failures that might
result in leakage are discussed. Preventing leakage by incipient (impend-
ing) failure detection and minimizing leakage volume through early detec-
tion of leakage is emphasized.
Specific leak detection and inspection methods are identified and com-
pared in Section 5.3. Each method is included in one of ten categories.
Section 5.4 presents the current status of these methods. Existing
domestic and foreign regulations and current recommended practices for in-
spection and leak detection of liquid and gas pipeline systems are reviewed.
5.1 GENERAL DISCUSSION
5.1.1 Definitions—Leak Detection, Inspection, Maintenance
In this study, leak detection methods are defined as methods that are
capable of detecting any pipeline system leakage. Inspection methods are
defined as methods that are capable of testing the integrity of the pipe-
line system. Inspection methods include ones that detect internal defects,
defects that cause insignificant small leaks such as leaking seals, or line
conditions that might result in failure and cause an oil spill. Thus, in-
spection methods, as defined here, may include some leak detection methods.
Hydrostatic pressure testing, for example, is considered to be both a leak
detection and an inspection method, whereas pressure deviation and flow de-
viation methods are considered as leak detection methods only.
146
-------
Maintenance is defined as keeping equipment, components, or systems in
a continuing state of proper working order. In some applications, main-
tenance may be considered as an inspection. For example, checking that the
cathodic protection system is operating properly is considered either main-
tenance or inspection of the pipeline system. Periodic cleaning of the in-
terior of the line pipe with pigs to inspect for foreign matter or remove
excessive water at low elevation areas is considered as both inspection and
maintenance. However, periodic cleaning of foreign matter from the interior
of the line pipe, scraping the line pipe walls to maintain an internally
smooth line pipe, or installing of interior or exterior protective line
coatings are considered as maintenance.
5.1.2 Pipeline System Component Failures
Major emphasis in this study is the prevention of pipeline failures
rather than failures of other pipeline system components. Justification
is based upon three main reasons. First, oil spill statistics (see Sec-
tion 4.6.2) show that most spills (over 80 percent) and fluid losses of
petroleum from pipeline systems are due to pipeline failures. Secondly,
pipeline spills currently present a much greater risk (see Section 6.2)
than other pipeline system components. For example, most onshore tank
storage facilities with associated components are protected by a secondary
containment system that virtually eliminates petroleum leakage outside the
facility. Finally, the most cost-effective means of reducing spill inci-
dents and volume, thereby minimizing environmental pollution, is expected
through the implementation of a viable and optimized spill prevention pro-
gram for pipelines. This would consist of scheduled inspections and/or
leak detection to test the condition of the line and check for the pre-
sence of leaks.
5.1.3 Leakage. Discharge Rates, and Spill Volume—Ruptures. Breaks,
Cracks, Pinholes
Petroleum leakage3 is difficult to prevent, detect and locate. These
difficulties exist for two main reasons. First, many variables contribute
to pipeline system failures that result in leakage. Secondly, actual lead-
age can occur in many different ways (critical failure, cracks, etc.), at
almost any location and at almost any discharge rate.
Pipeline leaks are usually caused by corrosion, hydrogen embrittlement,
defective pipe seams and welds, equipment rupturing the line, external im-
pacts, and natural causes. Results in Section 4.6.2.1 show that over half
of the pipeline spills and spill volume were caused by pipeline faults.
Note: The terms "leak" and "spill" are commonly used to describe the same
event and will be used interchangeably in this study. However, a leak is
normally considered as a gradual escape of a fluid, while a spill is norm-
ally considered an overflow or rapid escape of oil.
147
-------
These could be reduced significantly by new methods and/or increased
frequency of inspections. Pipeline accidents caused by outside forces,
i.e., equipment rupturing the line, anchor dragging, etc., account for
about one-third of the spill volume and less than one-third of the spill
incidents. These accidents can be reduced somewhat by implementing pre-
vention programs such as the One-Call System recommended in Reference 12.
However, effective continuous monitoring of leak detection and inspection
methods might also reduce the incidence of damage by outside forces.
There are several ways in which leakage might occur. These include:
• Small holes or hairline cracks
• Medium to large holes or cracks
• Large cracks that are continually growing in size
• Large cracks that have reached a critical size and rupture
• Large areas resulting from bursts or ruptures.
Small leaks are normally of pinhole size, are difficult to detect, and
often exist for long periods of time before detection. These types of
leaks account for a large number of spills, particularly for older pipe-
lines. Although data on reported spills (see Section 4.6.2) show that
small leaks account for a relatively low percentage of total spill volume,
there is evidence to indicate that this might not be entirely correct.
Also, even small leaks can be hazardous at certain locations. Large
leaks and ruptures are defined here as medium or major spills. These
account for most of the oil pollution incidents as well as a large per-
centage of the total spill volume. These leaks are normally detected
by visual observation, but in many cases detection is too late to avoid
serious pollution or large losses of oil. Spills at night, in water, or
in remote areas have continued for days without detection.
Leakage from other pipeline system components also occurs in many
different ways and varies widely. Leakage of very small amounts may oc-
cur through valve seals, gaskets, flanges, etc., or in large quantities
through very large areas such as over the top of storage tanks during
accidental overflow.
In most petroleum pipeline accidents, the spill discharge rate and
volume varies considerably depending upon factors such as the location
of the failed component (onland, underwater, depth of cover, etc.) type
of failure (crack, rupture, etc.), flow condition and pipeline opera-
tions. The spill discharge rate of petroleum leakage normally occurs
continuously, but in many instances discharge is transient or periodic.
Very small leaks, for example, can occur intermittently because they
have a tendency to become plugged by waxy particles or solids in the oil.
Also, leakage can be into rather than out of a pipeline. In subsea pipe-
lines where external pressures may be greater than internal line pressures,
seawater could be forced into a cracked line.
148
-------
Spill volume can vary greatly.. Typically, it ranges from the small
volume of spills of pinhole leaks to the large volume spills of pipeline
breaks or ruptures. Besides the cause, type of leak, and discharge rate,
spill volume also depends upon the detection of the leak and the time re-
quired to shut the system down. Oil spill statistics of incidents reported
to OPSO in Section 4.6.2.1.3 (Table 19) show that most small spills are
caused by discharge of oil at a low discharge rate. Statistics for this
type of spill indicate an average spill size of about 400 barrels. If es-
timated unreported spills were included, the average spill size would be
somewhat lower. Outside forces are the most frequent cause of medium to
large spills, and the average spill size is about 1,100 barrels. These
also cause considerable spillage over a short time and create the poten-
tial of serious pollution incidents. Major spills are usually due to de-
fective pipe, and the average spill size is about 2,100 barrels. In most
cases, a considerable amount of spillage occurs over a short time.
5.1.4 Leak Detection and Inspection Methods
A wide variety of leak detection and inspection methods can be applied
to detect petroleum leakage, internal defects that may result in leaks, or
line conditions that may result in failures and cause an oil spill. Cur-
rently, no single method has been found by governmental research, private
studies, or the oil industry that provides adequate protection against a
serious oil spill incident. However, combined inspections and/or leak de-
tection methods that verify the pipe condition and check for leakage have
been shown to be effective in preventing and detecting leaks for particular
pipeline systems. Methods such as visual observation of the line, hydro-
static tests that verify the tightness of the line, and on-stream leak de-
tection and inspection methods have been implemented. Two of the most ef-
fective methods in preventing leaks are scheduled pressure testing of the
line for detecting small pinhole leaks and inspecting the integrity of the
line with inspection pigs. The current trend in leak detection and inspec-
tion by the U.S. petroleum pipeline industry has been in the development of
low-cost continuous monitoring systems rather than inspections that require
shutdown of the line. The objectives of these developmental systems are to
provide rapid detection and location of a leak and shutdown of the line to
minimize the volume of oil spilled.
Many modern large-diameter pipeline systems in the U.S. have some form
of continuous monitoring leak detection system. However, variations in
the pipeline system operating and flow conditions often cause false alarms
from these systems. In order to account for this (reduce false shutdowns),
set points are typically raised. This results in a decrease in the leak
sensitivity and typically limits detection to only large spills. Also,
these existing monitoring systems cannot detect a leak until the flow (pres-
sure or flow rate changes) reaches a meter. Typically, meters might be lo-
cated up to 50 miles apart and a large rupture-type spills, for example,
aSmall spill < 2,100 gal (50 barrels); Minor spill > 10,000 gal (238 bar-
rels); Medium spill, 10,000 to 100,000 gal; Major spill > 100,000 gal.
149
-------
might leak many thousands of barrels before a leak indication is even de-
tected by a meter.
Variations in pipeline system size, age and other factors present prob-
lems in selecting the most effective leak detection and inspection methods.
In general, serious pollution spill incidents and large losses of petroleum
from modern pipeline systems are most likely to arise from medium to large
spills caused by outside forces and defective pipeline seams and welds.
Older lines have the additional problem of the high possibility of spill
incidents from external pipeline defects caused by corrosion. Thus, leak
detection and inspection that may be adequate for some lines are inadequate
for others. Furthermore, even small spills may present serious hazards to
segments of the population in some areas and cannot be tolerated. Thus,
even more sensitive or reliable leak detection and inspection methods may
be needed in these hazardous areas, e.g., see Section 7.4.2.
Because of the serious nature of pipeline spills, the many factors
that contribute to pipeline system failures and leakage, and the differ-
ent petroleum transportation systems in use, all potential leak detection
and inspection methods will be reviewed in the subsections that follow.
The most effective methods will be selected for evaluation in Sections 6
and 7.
5.2 GENERAL DESCRIPTION OF METHODS
In order to provide an overview of leak detection and inspection
methods, a general description is given. For comparison purposes, these
methods3 are classified according to type, operational mode, application,
installation and use in the following subsections.
5.2.1 Types of Methods
Two main types of leak detection and inspection methods exist. One
type checks the line for actual fluid leaks. The other type detects
either internal damage or line conditions that might results in a failure
and cause an oil spill. The latter type is defined here as incipient or
impending failure detection and is usually considered as an inspection
method to prevent leaks.
5.2.1.1 Fluid Leak Detection-
Leak detection and inspection methods for petroleum pipeline systems
detect either the actual oil leaking or events that are characteristic of
oil leakage. Typically, these events include:
• Pressure drops
• Pressure fluctuations
aSome methods might be applicable to more than one use in a specific cate-
gory. In these cases, the method is categorized as to the type of method
for which it is best suited.
150
-------
• Flow changes
• Temperature changes
• Sound level increases
• External pipe pressure
• Pipe variations
• Pipe acoustic energy releases
• Color changes
• Hydrocarbon vapors
• Vegetation changes.
Ideally, fluid leak detection and inspection methods should provide instan-
taneous detection and accurate location of leaks of any size to minimize
the volume of petroleum lost and damage to the environment. In actual
practice, the methods used are far from ideal. In fact, the only method
commonly used in the U.S. is periodic visual observations of the line. In
general, leaks are detected days after they occur, and even then precise
location is difficult. Remedial action normally is too late to avoid large
losses of oil and damage to the environment.
5.2.1.2 Incipient3 (Impending) Failure Detection-
Incipient (impending) failure detection is the detection of pipeline
flows, damage or external impacts that are just beginning to exist or ap-
pear and might result in a pipeline leak. This type of detection method
normally is considered as an inspection to check the condition of the line.
Some of the main ways these detection methods might be applied are:
• Detection of pipeline internal damage (cracks, corrosion,
etc.) that potentially can result in a pipeline leak.
• Detection of external pipeline damage, such as impacts
from outside forces, that can result in a pipeline leak.
• Detection of crack growths that can reach a critical size
and result in rupture.
• Detection of pipeline flaws by applying a stress that is
higher than what normally exists, i.e., proof testing at
elevated pressures. Enhancement of defects can occur with
high stresses and result in leakage or stress wave genera-
tion at a flawed area. Removal of the stress generally
causes the defect to diminish almost to its original state
and become undetectable by conventional flaw detection methods.
alncipient failure detection and impending failure detection have the same
meaning and are used interchangeably in this study.
151
-------
One of the most common impending failure detection methods is to tra-
verse the inside of the pipeline with an inspection pig to check for flaws
or corrosion damage in the pipeline. Data are normally tape recorded as
the inspection device is propelled through the line and then analyzed at
a later time.
Incipient failure detection, which prevents accidents, is the most ideal
way to minimize oil loss and oil pollution incidents. Although these meth-
'ods are effective in reducing the frequency of spills, in actual practice
this type of method is seldom used.
5.2.2 Operational Modes
Two operational modes, continuous and periodic, are possible with pipe-
line system leak detection and inspection methods. Generally, maximum ef-
fectiveness is obtained when both modes are implemented. In some instances,
this approach can be used to provide dual inspections or leak checks and re-
sult in an increase in the overall system reliability and effectiveness.
5.2.2.1 Continuous Monitoring--
Ideally, continuous monitoring for leaks is the most effective way of
reducing the volume of spillage, although not necessarily the most cost-
effective or practical. Most leakage volume occurs from a small number of
large spills (see Section 4.6.2). Continuous monitoring could immediately
detect and locate these large spills and shut the system down, thus mini-
mizing the volume of oil spilled. A few continuous monitoring methods could
also provide incipient failure detection and thus prevent a potentially
large spill from occurring.
Continuous monitoring systems currently used on modern lines are se-
riously limited. Only high spill-discharge rates and large spill volumes
can be detected. Also, leak location is either inaccurate or cannot be
done. One modern continuous leak monitoring methods, for example, consists
of metering the input and output volume of the pipeline and continuously
comparing the readings. When there occur significant differences in these
meter readings that cannot be accounted for, a leak is suspected. There
are a number of normal reasons for variance in these readings, such as tem-
perature, meter errors, flow changes, etc., that can be of the same magni-
tude and duration as a leak. Thus, there is difficulty in distinguishing
between a small-to-medium leak and a normal fluid variance. Other commonly
used continuous methods with similar limitations are pressure deviations
and flow rate comparisons. A few continuous monitoring methods which are
state-of-the-art or in the development or feasibility stage have fewer po-
tential limitations. The mathematical modeling method, for example, ac-
counts most normal variance and potentially can be an order of magnitude
more sensitive than most modern methods, i.e., volume comparison, etc., in
use today.
152
-------
5.2.2.2 Periodic Monitoring-Scheduled Inspections—
A scheduled inspection program using periodic monitoring methods poten-
tially can be a cost-effective way of adequately maintaining pipelines.
Many periodic leak detection and inspection methods are particularly effec-
tive in detecting small leakage rates or small internal defects that may
result in a leak. Pressure difference leak detection, for example, can de-
tect leakage rates to about 3 gal/hour. Periodic inspection of the pipe-
line with an electromagnetic inspection pig may be quite effective in check-
ing the line for internal damage and corrosion before a leak actually occurs.
Scheduled use of these and other methods may reduce significantly the fre-
quency and volume of leaks. The main disadvantage of using only a scheduled
periodic inspection program is that it provides no leak detection between
inspections and this could potentially allow huge volumes of petroleum
spillage to occur before detection.
5.2.3 Applications and Operating Conditions—Scope
The scope of this study is to investigate the capability of leak de-
tection and inspection methods to adequately control accidental spills from
new or existing operational pipeline systems by implementing periodic in-
spections and/or continuous monitoring. Essentially all pipeline system
operating conditions are expected to be available for implementation of
these methods. This would include normal operations or special conditions
for inspections such as shutdown of pipeline, filling and pressurization
with foreign gases or liquids.
Some of these methods may also apply for initial testing of new in-
stallations. A few of these methods could be effective in quality control
prior to construction phase or during construction. Specific guidelines
or recommendations for these applications are beyond the scope of the study.
It is assumed that new pipeline system components will be inspected for de-
fects, tested and qualified by the manufacturer prior to installation.
Furthermore, it is intended that these leak detection and inspection meth-
ods will not replace, or be used instead of, inspections currely required
by existing government regulations7 (see Section 5.4.1).
5.2.4 Uses
The primary use of these methods is to reduce the risk of oil spills,
i.e., frequency and volume of spills from the pipeline system. High risks
of oil spills from pipeline system components generally exist for older,
small diameter pipelines because of frequent spills or from large diameter
or underwater pipelines because of large volume oil spills. Implementation
of effective and practical leak detection and inspection methods potentially
can reduce these risks or oil spillage.
5.2.4.1 Reduce Frequency of Spills-
Periodic and/or continuous inspections to check the performance and
condition of system components can result in a reduction in the frequency
of spills. The purpose is to detect a condition that may lead to a leak
153
-------
so that suitable repairs or replacements can be made before a leak can oc-
cur. Detection would involve inspection for either insignificantly small
leaks, such as an oil drip, or internal defects that may lead to failure.
Detection would consist of examination of a particular component with one
or more inspection instruments. Then, based on a previously established
criteria, the suitability for continued service, repair or replacement
would be determined. Samples would include the periodic pigging of the
line to check the condition of welds, seams, corrosion, cracks, etc. Con-
tinuous inspection might be accomplished with an acoustic array to detect
either acoustic emissions generated from internal defects in a pipeline or
external pipeline impacts from outside forces.
5.2.4.2 Reduce Volume of Spills-
Continuous and/or frequent implementation of leak detection and in-
spections provide the potential of quickly detecting and locating a leak
so that the spill can be shut off, thereby minimizing the volume spilled.
For this purpose, continuous monitoring leak detection might include com-
parison of flows, pressures, volumes, etc., at both ends of the line while
frequent periodic inspection might include daily visual observation of
offshore lines.
5.2.5 Installation
Leak detection equipment can be installed in, on, or external to the
pipeline system. Essentially all installation schemes are considered in
this study. For analysis purposes, installation is separated into four
main categories and briefly described in the subsections that follow.
5.2.5.1 Non-interference with Pipeline Operation--
Ideally, leak detection methods that do not interfere with pipeline
operations are considered to be most desirable. However, these methods
generally are ineffective. For example, the two most common methods are
line walking and over and short accounting. Both are ineffective in early
detection of major oil spill incidents or prevention of leaks.
5.2.5.2 Limited Interference with Pipeline Operations--
Methods that cause limited interference with pipeline operations are
considered here as those that require temporary shutdown of the pipeline
or affect throughput. This would include methods such as inspection pigs,
continuous monitoring using equipment at each end of the line, mathematical
modeling, and most pressure testing schemes. Methods causing limited in-
terference are considered to involve only nominal costs. Furthermore, such
methods may be quite effective in preventing leaks.
5.2.5.3 Retrofitting—
Methods which require retrofitting require substantial line modifica-
tions and/or long-term shutdown. The necessity to uncover a significant
portion of the line to install a detector system or to provide equipment
154
-------
and power at a number of remote unmanned locations would be considered a
retrofit. Pressure difference methods would be considered in this cate-
gory because a number.of closely spaced block valves with controls and
sensor/transmitter units are required.
Depending upon specific use of a particular method, installation might
be considered a retrofit or limited interference with the line. For exam-
ple, a passive acoustic array system for detecting leaks and/or internal
pipeline damage requires retrofit installation consisting of a number of
closely spaced transducers installed along the line. However, an acoustic
array system to detect outside damage may allow large spacing between de-
tectors and require only limited interference with the line.
5.2.5.4 New Pipeline-
Installation of a leak detection and inspection system on a new line
generally can be done with minimal cost and no interference during the con-
struction phase. Also, modern pipeline systems are usually computer-
controlled with a number of remote monitoring stations and automatically
controlled equipment. Thus, equipment and installation for most leak de-
tection and inspection may require only nominal additions to existing
equipment. Some methods may be cost-feasible for new lines, while pro-
hibitive installation and equipment costs for existing lines would render
many of these same methods unacceptable.
5.3 COMPENDIUM OF LEAK DETECTION AND INSPECTION METHODS
Most basic principles such as acoustic, ultrasonic, electromagnetic,
nuclear, and optical can be applied by various methods or detecting and
preventing pipeline system leaks. A wide variety of applicable petroleum
pipeline leak detection and inspection methods exist that employ equipment
and detectors that are commercially available, in the engineering stage, or
in the feasibility/developmental stage. For completeness, applicable
methods, including many that are based on the same principle but involve
slightly different techniques, are identified and briefly compared in this
section. Only a limited number of methods are expected to be effective in
significantly reducing either the frequency or volume of spills. However,
many methods may be quite effective for particular lines or specific pipe-
line system components and may, if implemented, prevent a pollution inci-
dent. Thus, all applicable methods are included.
Most leaks from pipeline systems occur from buried onland and under-
water pipelines (see Section 4.6.2). Hence, leak detection methods are
geared to this problem.
In general, greatest risk of a serious pollution incident exists for
lines in or near water. This is true, even though the spill incidence
rate for underwater lines is similar to onland. A spill at or near water
usually causes a much greater pollution problem because the oil can be
spread over large areas, and costs such as those required for cleanup are
high. Thus, a number of leak detection and inspection methods are included
specifically for lines in or near water.
155
-------
Methods primarily intended for use at marine terminals and offshore
production platforms are also considered. Although fewer spills and lower
spill rates typically occur at these locations than onland, each individual
spill is usually much more serious.
5.3.1 Identification
A summary of all applicable leak detection and inspection methods are
listed on page 157. Methods are separated into two types in the list
presented on page 158. A summary that combines operational monitoring
modes and types is shown in Table 45.
5.3.2 Description and Comparison
Leak detection and inspection methods are described and compared in
Tables 46 through 55. Each method is included in one of ten categories3.
Additional details on methods selected for further analysis are given in
Appendix E. Section 7 supplies pertinent information on cost and equip-
ment.
5.3.2.1 Visual and Aided Visual Line Observations--
Visual inspections for petroleum pipeline systems are given in Table
46. These inspections do not interfere with line operations and are car-
ried out in a number of ways: oneland - by walking, driving or flying
the line; on water at marine terminals or pumping platforms - by inspec-
tion in a launch or an inspector located on the deck of the ship or pump-
ing platform; or underwater - by divers or submersibl.es (manned or remote).
For most lines, some form of periodic visual observation is required by
U.S. Government regulations.
Visual line observation, by line walking, was the-earliest form of
leak detection and inspection. Even today, it is the main method of de-
tecting pipeline system leaks. However, lower cost techniques such as
visual inspection from a light plane or truck have replaced actual walking
of the line. Simple line observation is considered by some to be the most
practical way of detecting small pinhole-size leaks.
Serious limitations exist for the visual inspections currently car-
ried out in the U.S. The most critical limtiations are:
• Pipeline faults resulting in leaks can be detected only after
a leak occurs and the oil surfaces.
• Leaks between the periodic line observations are not detected.
• There is only minimal protection against damage from outside
forces.
In certain cases, leak detection and inspection methods might be applica-
ble to more than one category. To avoid duplication, the information is
presented in only one main category.
156
-------
INSPECTION AND LEAK DETECTION METHODS FOR PETROLEUM PIPELINES'
1. Visual and Aided Visual Line Observations
a) Line walking, aircraft
b) From launch
b) On deck of ship
CO On terminal or pumping platform
(d Submersible
(d Diver
Visual aids
(d Hydrocarbon probe
(c) Remote oil spill detectors
(-0 Ultraviolet light source
id TV monitor
(c) Tape detection
2. Oil Spill Detectors on or Hear the Water
(c) Portable—on launch or ship
(c) Stationary—at marine terminals or on pumping platform
Id Land type—across rivers and channels or coastline
3uOy type
3. Internal Fluid Variations During Transfer
ib) Pressure deviation
(b) Volume comparison
(B) Flow rate comparison
(b) Flow rate deviation
(b) Over and snort
(c) Mathematical modeling
(c) 8Si* deviation
(d) Pressure difference with reference and threshold
(d) Fluid mass deviations
(c) Negative surge
(c) Inspection pig—passive ultrasonic
(d) Pressure fluctuation levels
(c) Production surveillance monitor
4.
Detection and Location of Leaked Oil on or at
a Short Distance from Pipeline
(d) Shroud with EHP pulsed coaxial cable
(e) Continuous thermistor
(e) Tape detection
(e) Oil soluble tubing
(c) Hydrocarbon probe
(c) Douole walled pipe
(c) Seal leak detector—capacitor type
(c) Nuclear tracers
(c) Nitrous oxide injection
(d) Acoustic/sonar
(d) Acoustic underwater
(e) Laser detection—underwater
(d) Passive ultrasonics
(c) External rods rfith acoustic sensors
(d) Passive acoustic array—leaks
5. Periodic Pressure Tests
(b) Pressure-static
(b) Hydrostatic
(b) Pressure difference
(c) Dye tracing
(c) Seal leak detector—joint type
(d) External hydrostatic
(c) Seal leak detector—thermistor type
(d) Reflected pressure wave
(d) Acoustic resonance
(c) Passive acoustic array
Acoustic emission
6. Corrosion Inspection
(c) Corrosion flow sampling
(b) Corrosion metering—Internal
(b) Corrosion metering—external
(b) Holiday detector
(b) Visual corrosion
See also NOT
See also inspection pigs
7. Standard Non-Destructive Inspection
(b) Active ultrasonics
(a) X-ray
(b) Gamma ray, radioactive isotopes
(d) Magnetic rubber, foil, tape
(d) Ultrasonic imaging
(b) Eddy current
(b) Penetrants
(b) 8. Inspection Pigs
9. Survey—Pipeline System Charting and Depth of Burial
(b) On land—radar
Underwater
(b) Sonar—sidescsn
(b) Sonar—sub-bottom profiling
(b) Microwave position system
(b) Scour
(b) Pneumofathometer
10. Miscellaneous
(d) Passive acoustic array for outside forces
(c) Passive acoustics for machinery damage
(b) Liquid level sensor
(e) Laser holographic interferometry
(c) Magnetic chip
(b) Control room monitoring, alarms shut-off
(b) Oil odor
"~ (e) Thermal or infrared
(e) Thermal paint
(e) Microwave
(e) Filtered particle
c Sonic
Status of methods for line pipe:
(a) Required by U.S. regulations
Common use
State-of-the-art
Developmental
Feasibility
w/
IS)
(d)
(e)
157
-------
TYPE OF LEAK DETECTION AND INSPECTION METHODS
FLUID LEAK DETECTION
Visual
Line walking, truck* aircraft
From launch
On deck of ship
On terminal or pumping platform
Submersible
Diver
Visual Aids
Hydrocarbon probe
Remote oil spill detectors
Ultraviolet light source
TV monitor
Tape detection
011 spill detectors on or near the water
Pressure deviation
Volume comparison
Flow rate comparison
Flow rate deviation
Over and short
Mathematical modeling
3S&U deviation
Pressure difference with reference and threshold
Fluid mass.deviation
Negative surge
Inspection pig—passive ultrasonic
Pressure fluctuation levels
Production surveillance monitor
Shroud with EMP pulsed coaxial cable
Continuous thermistor
Tape detection
Oil soluble tubing
Hydrocarbon probe
Double walled pipe
Seal leak detector—capacitor type
Nuclear tracers
Nitrous oxide injection
Acoustic/sonar
Acoustic underwater
Laser detection-underwater
Passive ultrasonics
External rods with acoustic sensors
Passive acoustic array—probe
Pressure-static
Hydrostatic
Pressure difference
Dye tracing
Seal leak detector-joint type
External hydrostatic
Seal leak detector-thermistor type
Reflected pressure wave
Acoustic resonance
Passive acoustic array
Acoustic emission
Liquid level sensor
Control room monitoring, alarms, shut-off
Thermal or infrared
011 odor
INCIPIENT FAILURE DETECTION
Passive acoustic array-acoustic emission
Reflected pressure wave
Acoustic resonance
Corrosion flow sampling
Corrosion metering-Internal
Corrosion metering-external
Holiday detector
Active ultrasonics
X-ray
Gamma ray, radioactive Isotopes
Magnetic particle
Magnetic rubber, foil or tape
Ultrasonic Imaging
Eddy current
Penetrants
Inspection pigs
Radar
Sonar-sldescan
Sonar-sub-bottom profiling
Microwave position system
Scour
Pneumofathometer
Passive acoustic array for outside forces
Passive acoustics for machinery damage
Laser holographic interferometry
Magnetic chip
Thermal point
Microwave
Filtered particle
Sonic
158
-------
TABLE 45. SUMMARY OF TYPE AND OPERATIONAL MODE OF LEAK
DETECTION AND INSPECTION METHODS
Periodic Monitoring—Scheduled Inspections
Fluid Leaks
Incipient Failure Jetection
• Visual and Aided Visual Line Observations
Line walking, truck, aircraft
From launch
On deck of ship
On terminal or pumping platform
Submersible
Di ver
Visual aids
Hydrocarbon probe
Remote oil spin detectors
Ultraviolet light source
Tape detection
• Internal Fluid Variations
3S&U deviation
Over and short
Inspection pig-passive ultrasonic
• Detection and Location of Leaked Oil on or
at a Short Distance from Pipeline ~
Hydrocarbon probe
Nuclear tracer
Nitrous oxide injection
Passive ultrasonics
Acoustic/sonar
Acoustl c-underwater
External rods with acoustic sensors
• Periodic Pressure Tests
Pressure-static
Hydrostatic
Pressure difference
Dye tracing
Seal leak detector-joint type
External hydrostatic
Seal leak detector-thermistor type
Reflected pressure wave
Acoustic resonance
Passive acoustic array
Acoustic emission
• Detection and Location of Leaked Oil
on or at a Short Distance from Pipeline
Passive acoustic array
Acoustic emission
• Periodic Pressure Tests
Reflected pressure wave
Acoustic resonance
Passive acoustic array
Acoustic emission
t Corrosion Inspection
Corrosion metering-external
Holiday detector
Visual corrosion
See also NDI
See also inspection pigs
• Standard Hon-Oestructive Inspection
Active ultrasonics
X-ray
Gamma ray, radioactive isotopes
Magnetic particle
Magnetic rubber, foil, tape
Ultrasonic imaging
Eddy current
Penetrants
« Inspection Pigs
• Survey-Pipeline System Charting and Deoth of Burial
Onland-radar
Underwater
Sonar-sidescan
Sona r-sub-bottom profi1i ng
Microwave position system
Scour
Pneumofathometer
• Miscellaneous
Oil odor
Thermal or infrared
Thermal paint
• Miscellaneous
Laser holographic interferometry
Filtered particle
Sonic
(continued)
159
-------
TABLE 45 (continued)
Continuous tfonitorina
Fluid Leaks
• Visual and Aided Visual Line Observations
TV monitor
• Oil Spin Oetectors on or Near she ilater
Portable-on launch or ship
Stationary-it marine terminals or on pumping
platform
Suoy type
Onland type-across rivers and channels or
coastlines
• Internal Fluid Variations During Transfer
Pressure deviation
Volume comparison
Flow rate comparison
Mathematical modeling
Pressure difference with reference and
threshold
Fluid mass deviations
Negative surge
Pressure fluctuation levels
Production surveillance monitor
• Detection and Location of Leaked Oil on
or at a Short 01 stance from Pipeline"
Shroud with EMP pulsed coaxial cable
Continuous thermistor
Tape detection
Oil soluble tubing
Double walled pipe
Seal leak detector-capacitor type
Laser detection-underwater
Passive acoustic array-probe
• Miscellaneous
Liquid level sensor
Control room monitors, alarms, shut-off
Incipient Failure Detection
• Detection and location of Leaked Oil
on or at a Short Distance from Pipeline
Passive acoustic array-acoustic emission
• Corrosion Inspection
Corrosion flow sampling
Corrosion metering-internal
t Miscellaneous
Passive acoustic array for outside forces
Passive acoustic array for machinery damage
Magnetic chip
160
-------
TABLE 46. VISUAL AND AIDED VISUAL LINE OBSERVATIONS
cr>
_. ._ ,
Method*
VISUAL Lint OBSERVATIONS
(al Iv walking. Plane. Truck
(b) From launch-Patrol HIM lines
ItMdenee or underwater
pipelines.)
Defects Measured
1. Oil leaks
2. Outside damage
1. Oil leaks
. •
Sensitivity11
1. Minor spills
1. Minor spill
Advantages0
1. Simple
of damage by outside
farces
3. Practical for minor
leaks
1. Simple
2. Practical for minor
leaks
• •
Disadvantages
1. High cost
inspections
3. Subject to personnel error
1. Medium cost
2. Cannot be carried out In bad
weather
I. Oil leaks
2. Component external
(c) from launch at Marine Terminal
or Offshore frodyctlon.
Platform
(d) On Deck of Ship at Marine I. Oil leaks
Terminal 2. Some external
(Mooring systems oil transfer defects on marine
system components-piping, hoses, hoses
valves, etc.) 3. Orientation, align-
ment, movement
problems that can
cause failure
I. Catastrophic
failures-early
detection
2. Small spills on
water
3. Minor underwater
(•Ills
4. Mooring system
failures
I. Catastrophic
failures-early
detection
2. Minor spills
I. Simple
1. Simple
2. Good for shipboard
connections
3. Practical
3. not adequate In darkness
4. Subject to personnel error
t. Difficulty in discriminating
between an oil spill and the
oil sheen from boat engines
or a few liters of oil leak-
age at a ship (this may
cause sheen over a wide area.)
1. Medium cost
2. Cannot be carried out In bad
weather
3. Subject to personnel error
4. Difficulty in discriminating
between an oil spill and thd
oil sheen from boat engines
1. Inadequate In bad fog or In
darkness except for ship break-
out
2. Subject to personnel error
3. Difficulty In discriminating
between minor spills and thin
oil sheen on surface of water
4. Medium cost
let On Pujnaina Platform or
Marine Terminal
(Platform, pipeline, etc.)
(f) Submersible-Manned or Remote
(Pipeline, pumping platform
support structure.)
(g) Otver or Scuba Olver5'
(Platform, pipeline, etc.)
AIDCO VISUAL
(h) line milling-Inspector with
Hydrocarbon Detector
4. Mooring system
failures at ship
5. Transfer system
failures
1. Oil leaks 1. Minor to medium
2. external piping, spills
(riser) damage
1. Oil leaks 1. Medium to major
2. teaks or defects spills
In underwater pipe-
line
3. Platform structure
damage
1. Oil leaks 1. Minor spills
2. External physical t. Small oil leaks
defects
1. Orientation, align-
ment and movement
problems that can
cause failure
1. Oil leaks on land 1. Small spills
1. Simple
1. Particularly useful
If defect or failure
Is expected based on
other Inspections or
If a major earth-
quake occurs
2. Fast Inspections
over wide erea
1. Underwater
inspection
1. Detects leaks that
may not surface
for long periods
of time
2. low cost
1. Difficult to discriminate between
thick oil spills and thin oil
sheen on surface of water
2. High cost
1. High cost
1 Medium cost
2 Inspection frequency Is limited
3 Cannot be used In fog. darkness
and rough water
4 Subject to personnel error
S Photographic records and video
tape records sometimes unreliable
or difficult to Interpret
(continued)
-------
TABLE 46 (continued)
cr>
ro
Method*
(j) Remote Oil Spill Detectors
Mane With Oil Spill
Detectors"'" Side ooktnfl
Airborne Detectors. Passive
Microwave laager. Camera.
Forward Looking Radar,
Infrared
(k) Man on Deck of Ship at
Marine ' erminal at Night with
Visual nspection but Using
Ultraviolet Light Source
(1) TV Monitor on Ship-
Low Light TV Camera
(Broad coverage of Marine
hoses, other oil transfer
system components and
mooring system. )
Defects Measured
1. Oil spills on land
or water
1. Oil leaks on water
1. Oil leaks
2. Orientation, align-
ment and Movement
problems that can
cause failure
3. Ship breakout
Sensitivity b
1. Small spills
1. Ultraviolet light
causes oil to
fluoresce which
enhances visual
Inspection
1. Catastrophic
failure-early
detection
2. Small or Minor
spills
1.
1.
2.
3.
1.
2.
3.
4.
5.
Advantages'" Disadvantages
Less subject to 1. High cost
personnel error
and Much more sensi-
tive than visual
observations only
Simple
Aids visual
inspection
Low cost
Simple 1. Inadequate in foy, bad weather
Wide area of or darkness
coverage
Operates at light
levels inadequate for
visual Inspection'
Cowuiercially
available
Aids visual inspection
on deck of ship
(in) TV Monitor on Humping
Platform-Low Light'l7 Camera
(Broad coverage of plaTfbrm,
pipeline, ship movement,
etc.)
1. Oil leaks
1. Minor spills
(n)
Tape Detection"
{Tape wrapped around hose.
flange, etc. leaked oil
will change color where
it conies into contact with
oil.)
1. Oil leaks
1. Small spills
1. Simple
2. Wide area of coverage
3. Operates at light
levels inadequate for
visual inspection
4. Commercially
available
5. Low cost
1. Simple
2. Good incipient
failure detection
3. Can be used for
continuous
inspection
4. Can be used
underwater
Inadequate in fog, bad weather
or darkness
1. Not rugged
2. Feasibility/development stage
3. High-cost
^Identifies iicferencu niwibcr
Small spill <2.100 gal. (50 barrels); Minor spill 10.000 gal. (238 barrels)} Medium spill 10,000 to 100,000 gal; Major spill >100 000
LVery rough estimate of Inspection costs for year for components that would typically be inspected- P >lou>00°
Low cost 0 - $20K; Medium cost $20K to $200K; High cost>$200K. "»H«.»ea.
-------
TABLE 47. OIL SPILL DETECTORS ON OR NEAR THE WATER
cr>
CO
Method
Defects Measured
Sensitivity"
OIL SPILL DETECTORS""61
(Continuous Monitoring)
(a) On-Boat-Infrared Type
(InfrareiTiype point sensor.
Transmitter projects an infra-
red light bean to surface of
water and reflected infrared
light is analyzed by receiver.
An alarm is activated when
oil Is detected.)
I. Oil spills on water 1.
(h) On Deck of Ship-Infrared
{Operation same as for on
a boat except devices must
be battery powered, porta-
ble and satisfy explosion
proof regulations.)
(c) At Marine Terminals or on
Platform-l nf rared~Type( d)
T.See operailon for on a
boat.)
(d) At_Mar1ne Terminals or on
Platform fluorescent
(Scanning fTuoriscense type
sensor but operation similar
to infrared type.)
Device has been
proven In USCG
tests to detect oil
spills In rough
water
Monitors a small
area typically less
than a square foot
1. Oil spills on water I.
2.
Located at bow of
ship
Can work at height
of about 30 meters
1. Oil spills on water
1.
Oil leaks on water
at night
1.
600 to 1000-foot
range
3.
Advantages
Disadvantages
Simple
Low cost
Reduces dependence
on visual Inspection
Some connierclal in-
struments can dis-
criminate between
thick film and on oil
sheen. This is dif-
ficult to interpret
by visual inspection
Some commercial
devices are explosion
proof certified
Simple
Medium cost
Works in fog
Aids visual inspec-
tions. Useful at
night.
1. Currently not used on boats
However, a few oil terminals
in Europe are considering pur-
chase of such devices
2. Alignment may be a problem in
very rough weather
3. Monitors a small area typically
less than a square foot
Simple
Low cost
Works In fog
Conine re tally
available
Aids visual
inspection
1. Works well at night
2. Medium cost
Commercially
available
Aids visual
inspection
I. No commercial version exists
for use on deck of ship but this
might be available in the near
future
2. Medium development funding
required for ship mounted version
3. Device must meet explosion
certification
4. Detector may not sec oil leaks
under certain conditions, i.e.,
tide can go In one direction
and wind in another direction
5. Requires a relatively stable
platform for a spot type
sensor
6. Sensor must allow for changes
in draft of ship
I. Requires about 4 sensors on
platforms or at terminal
1. No daylight operation
(continued)
-------
TABLE 47 (continued)
llethnd'*
llelpcts llrasured
Senslllvlly
Advanlaoi>sc
(c) On tluov-liilraied
(lnlrarcd~ lyjie point sensor.
transmitter projects a light
beam to surlace of water and
reflected Infrared light Is
analyzed by a receiver. An
alarm Is activated when oil
delected.)
(f) On Buoy-Scanning Infrared
(»)
Infrared
(See operation for In a
boat.)
(I) Across Channels and Rlvers-
kUUlH fcU«9 IUr»a.ea"
(Infrared' tiaiamitter and
receiver located on tall
timers across river or
channel. Device delects
hydrocarbons on water along
narrow path at distances u
inn feel.)
(H User Dclecllui On Water
Oil iiiK
(Laser placed on hl'jh land.
System traverses fixed plane
along pljiellne or hose slrlny.
Since laser absorption by oil
and water Is known, a leak can
hn dolecttnl.)
1. 4111 leaks In water
1. Oil IcaVs im water
1. llydrocarluns on
waltrr
1. III! li'.ils
I. Hide range around
buoy
cier
i ahoul I
1. SlKu'e
2. low cost
I. Hay require only
one sensor
I. Simple
2. low cost
llls.iuvaiitao.c-s
I. No r«mmM*ri:ial version available
I. feasibility stage only
2. Medium cast
3. KaiKjc uncertain at this linn
4. Complex
I. No comwrclal version available
?.. Utiuld re
source «ml pownr. w*lcr, OH $hut<-( OH
1.
2.
3.
4.
1.
2.
3.
1.
2~
Uws nut work well in Oaytlw:
Ho iUMwrclal wersfiMi «vtillat»le
Usually ic(|ulrns «n tmy a(
smsnri l» Iw v**ry cflertlve
Must IIP <>x|pliiili.ii |irnnf
H*t tumwiLtAl vcr&tun avaiUblo
Usually n'*|iitics a variety i>f
sunr.oit ta be oUoctlve
HuM l»(* CK|>IOSllM| |)|'Olr(
U«ter WMSI lirfvo HAvy stiriat,!
HirdiuK cu^i
Ilitfli co-il
Ni;t|ulin\ triiliif'il persiMiiiel
tklectur sens I- pUUoiM ur on deck
tlvlty aod traverse of slil|>
2. Hliwr or MrtlliM ?. Coiitlmwus
S|illl sensitivity iWMilt»rliii|
CX|M!C|l'd
3. Arfcrluil by li.nl we.i,!,<•»
4. Unlocts lo«ks only aftrr tlHry
IH;CMI aixl oil ilsrs t» Ihfi
surUcn
S. ltl?V«>l(l|MW!nt>ll Slrl^
lifa-uliflf. Ki'l" •• numVi
''Small spill ?.|IM gal (Ml luiielsli mlnm spill Ml.uMI i)al (2UI barrels); medium spill IO.MNI to IUI,IK»I ual; auinr spill IIKMI.IIIXI q.il.
cVery rougb estimate of lns|u>K. wdliim coil (?UK to $;'IKIt:;
hlqh cost t?«K
dllole: Hiuiy oil <:<«m,i«ules are currently ustwj these devices at supertanker terminals lor oil spill detection; usually Installed on pillna,
-------
TABLE 48. INTERNAL FLUID VARIATIONS DURING TRANSFER
CTl
01
"• "
Method1 Defects Measured
(a) Pressure Deviation Continuous 1. Oil leaks
iHijilfpr oT'llne'prcssuros at
various locations along pipe-
line. If pressure deviations
exceed a set point value, a
leak It assuKd. )
(b) tolune Couoaillstm (Flow 1. Oil leaks
tlo* Intervals of several
•Imltes duration to check
uttered barrels Into pipeline
against oxtered volume out.
Comparison of volune nasure-
•enll 'typically occur over
hourly periods. Ibis can bo
considered as a fast over and
short)
(cl flow lUte Ceooarlson |. oil leaks
toitlQfOit
If Id* rate oetsurenents
at each end of the pipeline
section are compared, tare*
differences In oeesumentl
Indicate posslhlllty of
leakage. )
(d| MM tato Devatlons l-5« 1. Oil leaks
(IKvleuoBlo flow rales are
Matured at specific flow
stations. II the change In
flew eiceeds a certain value.
l.o.i 1 to S percent, an
alar* Is sownled. )
•
Sensitivity*
1. Major oil leaks
2. Accuracy: I-SI
1. Medlun or «Jor
oil leaks
2. largo leaks over
i short period
of tin* and sooll
leaks ever * long
period of tlo*
3. Accuracy! 0.2 to
n
1. Major oil spills
2. Accuracy: 1-81
1. Major oil spills
2. Accuracy 2-M
Advantages'
1. low cost
2. Staple
3. Cenumly used
4. Commercially
available
S. Continuous
ennltorlng
(. Work well for large
leaks close to
pvnplng station
1. Staple
»r*^u*Mir*-t*llv
> WMMCrCltjIlJ
available
3. Continuous
oonlterlng
t. Staple
2. Detects catastrophic
failure
3. Coamerclally
available
4. ComoMly used
f. Continuous
ennltorlng
1. low cost
2. Coaojerclally
available
3. Continuous
awnltorlng •
4- Cceunnly used
S. Detects catastrophic
failures
1. Ci
2. Cl
li
pi
3. Al
el
1. ft
t. V
' tl
f
3. t
fl
4.?;
Sl
hi
Cl
S. «
f,
i. H
i. r<
2. ti
S
tl
3. D
4. P
•
S. M
•
1. Pi
2. ti
Si
tl
3. 0
161 fTerenci eelweea Barrels
received Into and delivered
out of pipeline systen and
correction for Inventory
changes and tevperatore.
typically carried oat weekly
to imUity Intervals.)
t. Oil teaks
Very snail leaki I. HlnlMl cost
that oilst for Ion* 2. fart of nonwl
period of tin- accoMitlit*
amths to years- 3. Caoomly wed
(.1 to 0.21 of 4. Vsefol for snail
throuoAput teaks
ttlsadvanta«i>s
Catastrophic failure only
2. Cannot detect soall leaks or
large leaks a lone distance from
pnp station
Affected hy changes In fluid
properties. Input end ootput tank
heads, lines vlth changing flow.
etc.
I. Detects leeks after they occur
t. •UM«U to detect slw leaks
that over a period of tie* *ay
result In a oajor oil spill
Cannot detect a catastrophic
failure In sufficient tin* to
prevent najor oil spill
tendency hy operators to raise
set-points to reduce alarm
because of line pack and other
considerations
Detects leak only once per hour
for enst comnmly used system
i. Hedlxa cost
roar leak detection
2. tendency hy operators to ralso
setpoInU to reduce alaros e*d
than decrease leak sensitivity
3. Detects leaks after they occur
4. Duos not respond to steady state
or slowly varying leaks
HIM five MamliM of large leak
only during the ttsw the line
dynamics change
Poor leak detection
2. tendency hy operators to ralso
sclpolnts to reduce alaraa and
thus decrease leak sensitivity
3. Detects leaks after they occur
I. Defects possibility of leaks only
after long periods of tin*
(continued)
-------
TABLE 48 (continued)
Method*
Defects Measured
Sensitivity"
Advantages
Disadvantages
cr>
en
(f) Mathematical Modeling 1. Oil leak
Continuous ~ 0. \% Accuracy
(Computerized Method of
modeling pipeline system
(components, sizes. Materials,
etc.) and flow characteristics
(temp., pressure, viscosity,
etc.) to optimize leak
sensitivity.)
(g) BStW Metering
(Periodic samples of petroleum
are taken at each end of the
line. Collective samples are
put in a centrifuge and bottom
sediment and water content is
determined.)
(h) Pressure Difference with
Reference and Threshold*'
(Continuously compartnjfThe
pressure difference between
the differential pressure in
the pipeline and a reference
pressure and a predetermined
threshold pressure. The thres-
hold pressure is the line
pressure at the start of each
cycle of comparison.)
1. Oil leaks
2. Leaks in underwater
pipelines
3. Volumetric errors
1. Oil leaks
1.
2.
3.
1.
2.
Detects a .IS of
actual flow rate
Small to medium
leaks
Detects leaks when
flow reaches flow
meters
1. Low cost
2. Computerized
reduction
3. Good failure
detection
4. Provides leak detec-
tion improvements
over conventional
hydrostatic pressure
tests
Medium leaks 1.
Less accurate than 2.
volume comparison
BS4U cannot be read
to better than 0.1X 3.
increments or IOX
of typical full
scale
1. Unknown
Simple
Provides check of
gross volumetric
leak detectors
Commonly used
Low cost
Computerized
reductions
Can be used In
conjunction with
supervisory control
system
Continuous
monitoring
1. Medium cost if transducers must
be added to existing supervisory
control system
2. Hay require trained personnel to
properly interpret data or
maintain system
1. Inaccurate
2. Medium cost
1. Insufficient information available
on actual on-line performance
(continued)
-------
TABLE 48 (continued)
CT)
(1)
(J)
(k)
(1)
(•)
Method* Defects Measured
Mass Deviations 1. Oil leeks
jjfgjtlva Surfi-Ueak Pressure 1. Oil leaks
(large rupture type leek
causes negative pressure wave
to travel along pipeline that
can be detected and located.)
Inspection Pip-Passive 1. leak from hairline
Ultrasonic Tne cracks or small
|m escaping fluid from a corrosion holes
pipeline leak emits sounds.
A passive ultrasonic detector.
mounted In an oil tight
container, detects the leek.)
Pressure Fluctuation levels 1. large oil leaks
(large leeks cause large In-
creases of pressure fluctua-
tion over a certain frequency
band at the leak location.
These fluctuations ere
travel lino along the pipeline.
They are detected by pressure
sensors with the tensing ele-
ment In contact with the flew.
Sensors ere located many miles
apart.)
Production Surveillance 1. large oil leeks
1* dynamic pressure transducer or well problems
measures small pressure
fluctuations that relate
directly to Mould flow nte-
for oil production two-phase
flow. )
Sensitivity*
1. unknown
1. leakage > MO
barrels per hour
t. leak location
within I miles
1. 1 to l gallons
per hour leaks
1. Unknown
t. lackfround noise
may seriously
limit sensitivity
Advantages'
1. [HmlMtes errors
Iran) both pressure
end temperature
1. Continuous
Inspection
t. Ceomtrclally
available
3. Supplements other
4. Continuous
monitoring
(. Instantaneous leak
detection end
location
1. locates leak within
a few feet
t. Should work well If
* ttak detector pig
Is built and dedi-
cated for a specific
pipeline
1. Simple
t. low cost
3. lap Id detection of
leak
4. Supplements other
leak detection
1. Proven In field use
t. low cost
1. Can be continuous
monitoring
4. Useful for
two-phase flow
Disadvantages
1. Not comnnly done
1. Negative surges caused by other
pipeline operations must be
accounted for
1. High cost
I. Not commercially available In the
U.S. because of difficult* In
applying device to a variety of
pipelines
1. nojulre* some development for
reliable results
4. background noise currently limits
leak resolution
1. Method currently In the feasi-
bility stage
hmall spill IOO.OOO gal.
'Very rough estimate of Inspection costs for year for components that would typically be Inspected:
low cost D - imi Medium cost 1201 to I20MC! III eh cost >ttMK.
-------
TABLE 49. DETECTION AND LOCATION OF LEAKED OIL ON OR AT A SHORT DISTANCE
FROM PIPELINE
0>
CD
Hetkod* Defects Hetsured
Sensitivity*
AdvMltoesc Dlstdvinttges
UAKID OIL
(•)
(k)
(c)
lilvfln« Is m-ipped
trouad tke line. c*verad trlln t
loose Jtcktt Md filled »ltk llould
under pressura. loss of prossura
Indlcites t letk.)
Hydroctrfcoa Probe (Sniffer) 1. Oil lent In iivallna 1.
INydroctrkon proba Is 2. Oil tklt nay settlt
instilled In touflsk tad on tka kottM
taxed t tw foot tkava tke
pipeline. Proba detects
kydroctrkons frat all latks.
A prassura senior In tmflsk
•enures deptk. )
Double Hilled Pipe 1. Oil latks 1.
(l*4ulr« eoubliiulled pipe
Hit* tka one trtnsferrlM 2.
fluid to ka centered Inside
tke other. A vtrlety of
•deauite all detectors could
ka Instilled on Ike Insldi
•f the outer pipe to detect
leiks fro» tka Inner pipe. )
Nlnor spills
Minute leiks CM
be datected end
tnen loctted
ultkln • fan ftet
ItaknoHn
Minor spills
Swll spills
1 • M~* Hi MS
per •! inter
Saull or eilnor
letls
Nor* sensitive
thin tny atker
Mtkod lor
pipelines
1. Slavic 1. Mnproven
2. Continuous 2. Relliklllty uncartiln
•MltarlMj }. Detects letkt tflar tlwy occur
». Good Incipient 4. Bedim cost
flllure detection
4. CM ka usad Hltk
supervisory control
systt*
1. Continuous 1. MedlM cost
•mltorliM 2. Fatslklllty stioa
1. Slnpla 1. Not ruuaed
t. Coed Inclpltnt C. retslfclllty/devalopwnt sttga
ft Hurt detection 1. Hlan cost
1. CM be used for
continuous
operttlon
1. Stable |. SystaB did not Hork on Irtns-
Alplna Pipeline
2l IM cast
3. CoeMrclelly
•Vllllbll
4. Good Incipient
flllure detection
S. Provides continuous
record
1. Slupla 1. >ery klaji cost
2. Cuccllant Incipient
ft II we detection
J. Continuous
Inspection
4. Conttlns tke leiked
•II tfcus preventing
split
S. Cost affective for
skort sections of
pipeline touted it
potentlll leil ireis
(continued)
-------
TABLE 49 (continued)
Hctluut'
Defects Measured
S«»ltl>lt«k
Advantaots
DIsldMntiftl
(h)
Nuclear Tracers " ."
(* limit! quantity of
II ft radioactive neter
(4) Soil Leak Petecton Capacitor
{insulated capacltlve wterlal
Is placed Insldt two comwct-
1n» flanges that contain a
stal. Air or water leaks
change capacitance and
Inptdanct of the capacttlve
Mtarlll.)
H ,7t
Short-
terlal which
Is soluble It the ell handled
Is Injected lett the pipeline
fluid. If a leak entsts. the
radioactive Mttrtal will
escape through tht leaking area
of the pipe to tht surrounding
sell. After the pipeline Is
cleared with a scrapper pig
and a suitable mount of non-
radloactfve product passed
through the line, an Inspec-
tion pig with a nuclear
detector or an Inspector
travel along the line and
anoltor any radiation leaked
into the soil.)
I. Oil, gas or water I. Better than 10'
leaks In flanges and scc/s
other types of seals
"'
1. lev cost
' failure detection
•rcltlly
I. Requires detector to he designed
as part of the pipeline system
I. Oil leak location
t. Method Is clalwed
to accurately
locate leak
S. Continuous
ml taring
I. Accurate leak
location
I. Nethod requires conplete renaval
of radioactive tracer from walls
of pipe
2. Special personnel required
(')
VO
"Hnxrt 0»la« iMpettlo.
(Iftt fM If nMIt I* vtttr
u4 til mt hibbln out of
selntltt tt Im prtMuref .
A Mill ejwitlty <> Injected
Into Uw lint. A* tnfrirtd
tmlynd pratt ll osed M
detect le*tt*| tu tltner
In tkt stll or ikon tht
ground tlong tht pipe. )
Oil letk>
I. Encollont leek
loutltn
I. faked preven ind
In me I* pipelines
I • Urjt nudier of bortholei nty
he required
2. H|e> coit
IMKEO til MOPACATIIW INTO
riUID HMIWI
U)
(A somr eevtce directs
•covstlc wives iloog the
Pith of • (otatroed pipeline.
The reflected wives ire
detected Md recorded, the
lecatlt* ef uoplne. liquid
Is then dtttratned fro. tht
recorded reflected tcoustle
wives. IM reflected
Hint result fro*
thedlffertiKO In acoustic
escapln* liquid and sewatcr.)
I. Oil leaks
I. Stevlo
Acoustic Undaraater"
(Leaks fro* mderatttr ell
pipelines ire detected and
located ky t tool It receiver
Mklch novel alone, a path
parallel U the pipeline and
detects the ultrasonic fre-
quencies In the 20 to 70
KHi ranee that art generated
by liquid escaping fro. a
leak.)
I. Oil leaks
I. Staple
2. low cost
(continued)
-------
TABLE 49 (continued)
MttllllM.'
(IHa.ej asleclien Unileiw.!.:,
I eaks
)*' laser systera would be
iwiunted on an underwater
pipeline or hose slrlmj.
the laser is attired parallel
to the pipe ur hose to t
detector Mounted further
away, light transa-lltance
would decrease with oil
escaping. Ibus a leal can
IK) delected.)
AlUuMli: IMMii UKtKAUM M It A*
•MOil AMI rtWAGAIIMi; AluNG IIH
I INC
C*) ("asstye Ultrasonic
(Portable-band held, ultra-
tunic detectors thai delecl
soinuls c«lttud at lejk source
on pipeline or valve.)
(u) l«le|«a| Mods With Acousljc
Sensors ~ "
(niilaVrods attached to pipe-
line at about every IMu leel
to penult th> locali/atlon of
leaks by Mamilorlttu, acoustic
energy generated at the leak
and. propagated to Hie rods.)
(o) Passive Acuusl{c Array-leaks
lArrayVf'aciusirc" ifjus-" "
queers peiwanenl ly Installed
OH pipeline to delect and
locate leak frura sounds
ewltled at the source ol a
leak. )
1*.
la UMK: kliih cost • SHlMr
bviecls lleasuriMl tiiittslllviiy
1. Ull leaks I. Impends on laser 1.
source and powui ,
deloctur sensi-
tivity and traverse
2. Minor spill sensi-
tivity «»pet.led
2. Valve damage 2. Detects taiidiuM 2!
3. Corrosion leaks up to about 3.
300 leill iway on
pipe) Inus
1. Sensitivity depends
upon S|/B ol leak, 4.
product In Una.
ftressorii In IfMt,
and toll conditions
around pipe.
1. leaks |. Minor leaks 1.
2.
1. till or lluuld leak 1. Uepends on taaler- 1.
deucllun and lal, compnniuil 2.
location ttrucluie. Irans-
ducer spacing. ).
etc.
detect a small. 4.
iliw leak (I.e..
few barrels p«r
hour) and locale
the leak wllnln
a few feet
3. Sensitivity in-
creased wltn ap-
plication of
higher test
pressures or a
test uas
0. .ta» mill *jm a-i (2M U^UK .41. SMI
cwUs fur irur lui- unmHMmth tWt wuuU lyMUdlly 1
Atlv>iHtau 4ttUtiw4 tu K>.tk-
CoHHHircUlly !M*J iirwcluru
«ivaill«fale for 2. lu*i luc-itluN iMily
Abuvii tMtui «ir 3. Cdiutut be uiej tu atmtdtwly
UMilcnMtitr u&u iktumtiiic' sevurUy uf teak
SllHH! llKl|»IUnt
failure tkteclimi
IUM cuil I. lujfc lucdituU wily
Slhutlu 2. Cdiuiut lit) usuil to Jt-'iuiwliiu
HiittiuJ U uruvuu icwuilty uf a leal
dtttl USC'ti IM
U|HJt4Un9 liiiuk
i»«Vl«-' 1. IV-Jlu- to- 1
CiuttlluMl luak 2. Sy&tuw in ikvclutMNiiit/uii^niuui iiiy
dutiiclliM pita&c
fUKlulrttk llltlu ur 3. (Utlltiblltiy «wJ |>t:i'lunuaiice
HA laltii'|irvldtluH k^Hictf Icdttuiib tiru uiM-crtalii
by fhtrsoMtel
CiUMfHiturucd. .iiitu-
»ate«i syslcw tdii |tu
A<|j(*li:d to exUliHtt
su|HirvUory cuntrul
&ystbM
til HI. 000 tu HMJ.tMW, Hujui L|.||) IUU.WM >ja|.
j« Iiuuet:te4. luw t»U U - ^1*1 wc.lium tu^l $i*UK
-------
TABLE 50. PERIODIC PRESSURE TESTS
Method"
Defects Measured
Sensitivity1"
Advantages0
LEAKS
(a) Pressure-Static
(Monitors line integrlty.
When a line segment ts
suspected of leakage, pres-
sure ts maintained and block
valves closed. Static pres-
sure between two valves is
monitored (less than an hour)
for any pressure drop.)
(b) Hydrostatic Pressure Drop
(Pipeline pressurized and
pressure gages used to
detect leaks by pressure
drops.)
(c) Pressure Difference
{Pipeline pressurized.
Pressure difference gages
Installed across a series
of block valves that Isolate
sections of pipeline are
used to detect a leak.)
(d) Dye Tracing
(insertion of dye under
pressure into pipelines, and
visually look for dye leaking
out.)
1. OH leaks
1. : 3 PBH 1.
2. Minor oil leaks 2.
3. Less sensitive than 3.
hydrostatic pressure
difference method
4.
1. Fluid or gas
leaks
2. Pipeline flaws,
cracks, etc.
1. = 0.5 BPH
Fluid of gas
leaks
Pipeline flaws.
cracks, etc.
1. * 0.05 barrels/
hour
1. Fluid leaks
1. Minor spills
Low cost
Simple
Quick check of line
for leaks, i.e..
less than 1 hour
Method In use on
some pipelines
1. Simple
2. Low cost
3. Provides good
incipient failure
detection
4. Widely used
Disadvantages
Extremely sensi-
tive method
Excellent incipient
failure detection
Method Is in use on
some pipelines
1. Simple
2. Low cost
3. Good incipient
failure detection
4. Commercially
available
5. Useful when residual
oil or oil from
external sources
cause difficulty In
detecting leak
1. Requires leak detection method
to locate leaks
2. Variations In temperature, etc.,
limit sensitivity
3. Difficult to detect slow leaks
4. Out-of-service operation required
5. Cannot be used with hot or very
viscous fuel oil or crude
6. Difficult In mountainous areas
1. Out-of-service operation
2. Potential damage to pipeline
3. Requires leak detection method to
locate leaks
4. Dowtime for temperature stabiliza-
tion can be of long duration
5. Use of petroleum in lines during
test Is risky and should be
avoided
6. Cannot be used with hot or very
viscous fuel oil or crude
7. Difficult In mountainous areas
1. Medium cost
2. Out-of-service operation
3. Use of petroleum In line during
test Is risky and should be
avoided
.4. Downtime for temperature stabili-
zations can be of long duration
5. Cannot be used with hot or very
viscous fuel oil or crude
6. Difficult In mountainous areas
1. Slow detection method
2. Requires out-of-servlce operation
3. Should be used only when other
visual Inspections cannot detect
a leak
4. Use of petroleum in line during
tests is risky and should be
avoided
(continued)
-------
TABLE 50 (continued)
ro
Method*
(c) Seal leak Detector Joint Type
(End element tubes wrapped
around a cylindrical fixture
are positioned on each side
of a joint and pressurized
to provide a good seal. Then
area in between the tubes is
pressurized and this pressure
is monitored with a gage.
Decrease in gage pressure will
indicate leak.)
(f) External Hydrostatic
(Device flls over and seals
small sections of piping.
External pressure is applied
and Monitored. If an Internal
leak exists, pressure decrease
will occur and the leak is
detected.)
(g) Seal Leak Detector
Thermistor Type
Defects Measured
1. Air or liquid
leaks Inside
pipe
1. Defects in
threaded or
welded connec-
tions of piping
2. Pipe leaks
1. Gas or air
leaks
Sensitivity1*
1 . Very smal 1
seal leaks
1. Better sensi-
tivity than
hydrostatic
testing
1. 10"8 scc/s for
nitrogen gas
Advantages'1
1. Simple
2. Low cost
3. Comnerc tally
available
4. Good incipient
failure detection
5. Reduces the need
to pressurize
lines
1. Low cost
2. Simple
3. Good incipient
failure detection
4. Quick test time
5. Commercially
available
1. Simple
2. Low cost
Disadvantages
I. Out-of-servlce operation
1. Must be applied externally
2. Not a replacement for convention
non- destructive testing such as
X-ray
1. Requires out-of-servlce operation
2. Developmental/engineering stage
(Portable heated thermistor
device senses gas leaks through
a seal that is placed over test
area. Thermistor exhibits
large changes in resistance
with small changes In tempera-
ture caused by the leaking
gas.)
3. Good incipient
failure detection
(continued)
-------
TABLE 50 (continued)
Method"
Defects Measured
Sensitivity0
Advantages
Disadvantages
CO
INTERNAL DEFECTS
(h) Reflected Pressure Have
(Pipeline must be blocked
off. emptied and filled with
nitrogen and then-pulsed with
a pressure wave. Pressure
wave reflects from cracks
before it reflects from end
of pipeline.)
1. Internal cracks
1.
Detects large
cracks of about
1 inch
1. Provides some
incipient failure
detection
2. Low cost
1. Out-of-service operation
2. Sensitive to internal surface
roughness
3. Experimental technique
(i) Acoustic Resonance 1. Cracks or gouges
{locates cracks in gas-filled in pipeline wall
pipelines by using the "organ
pipe" resonances of the gas
column in pipeline to measure
distance from pipe input to
crack in pipeline wall-
requires microphone and sound
generator at pipeline input.)
(j) Passive Acoustic Array- 1. Material defects
Acoustic Emission 2. External Impacts
(Array of acoustic transducers 3. Valve damage
(permanently or semi-
permanently Installed)
installed on pipeline to
detect and locate repetitive
acoustic emission sounds
generated at the source of the
flawed material under stress.)
1. Depends upon
pipeline radius
and wall thick-
ness and wave-
length of sound
1. Depends on mater-
ial, component
structure, trans-
ducer spacing.
etc.
2. Can detect and
locate Internal
flaws that are
growing and
prior to cata-
strophic failure
3. Sensitivity In-
creased with ap-
plication of
stresses higher
than normally
used
1.
2.
3.
4.
1.
2.
3.
4.
Low cost
Simple
Good incipient
failure detection
Particularly use-
ful In inaccessi-
ble area
Excellent incipient
failure detection
Computerized auto-
matic system can be
adopted to existing
control system
Comnerclal system
for periodic proof
testing of tanks,
pressure vessels,
etc. are available
Continuous
monitoring
1. Requires out-of-service operation
1. Incipient failure data subject
to
interpretation as to the severity
of defect but system does provide
defect location for further In-
spections by other means
2. Medium cost
3. Reliability and performance
.
Identifies Reference number.
bSmal1 spill < 2,000 gal (50 barrels); minor spill 10,000 gal (238 barrels); medium spill 10,000 to 100.000; major spill > 100.000 gal.
GVery rough estimate of inspection costs for year for components that would typically be inspected; low cost 0 - $20K; medium cost $20K
to $200K; high cost > $200K.
-------
TABLE 51. CORROSION INSPECTION
Equipment'
(a) Corrosion F!on 1.
seeoltno
(Corrosion rat*
conoons. laboratory
analysis of flo»
content, strainer
an* 019 trio eonl-
torlno.)
(b) Corrosion neterln* !.
TCorrosibn rat*
probe Installed
inside it aw tan
of glpcltne ottn
u noose* •le-
nient. Corrosion
causes » cnenee
In electrical
resistance of
tloesnt.)
(e) Corrosion me ten no T.
(Potential nasnir*
earns an* contl-
of uoosod iteel
of glpelino.)
(d) Mellde* Oeteetar 1.
(Oevlc* places a*
electric pot**C1al 2.
a* en eon pip* sec-
tion an* a* elec-
trode cast Is In
contact Kit* out-
Sid* coating of
pip*. Electrode
typically is a
called spHno
aramM pip*. Else-
trie Mtentlal 1s
set Did* emu** so
that a* arc In air
is produce* vn**
tftldmess of coat-
ing Is not
satisfactory.)
(*) Visual Cor-wlon 1.
Inspection
(f) See lisa non-
destroctlv*
eaulpMffi Cor
cprroslon
nessureBHt in
Table 3-11.
(g) See also inspsetio*
g1os far corrosion
n*aiuie*am In
Table >12.
Uef ects Measure*
Internal pipeline
corrosion
Internal pipeline
corrosion
External corra-
s1o* of sloeHne
Coatlnt dlscon-
tl mil ties
Virfac* daifekU
External corro-
slon of pipeline
Sensitivity*
1. Potentially can
detect nador
corrosion
2. Corrosion trend
indication
1. Potentially can
detect najor
cormslon
2. Corrosion trend
Indicator
1. Provides loe-
quat* aeisure
MMt Or CaWtfl*
tion* of cathodlc
protection
1. Location of gin-
hole or elcro-
scoplc six*
1. Poor
H»all spill < 2.000 «al (M Barrels): elnor spill 10.000 jel (OS
e*4or spill > 100.000 «*l.
Adventaoes' Olsaovantaaes
1. So** incipient 1. Hlfrlcalt ta determine
rallure detection quantitatively tlw
2. widely used ueunt of corrosion
3. Kediu* cost 2. 3oes not locate
4. Caaierctally corrosoln
ivatlable
1. So** indolent 1. Indicates corrosion
failure detection trend
2. Widely used 2. Does not locate
3. Lw con corrosion
4. Cunaarelally
available
1. Stele
2. Lov cost
3. widely used
4. fiood Incipient
failure detection
S. CoBMrclally
availael*
4. Catnodlc «e1tor-
Ine ea* be don* oy
aerial fyrteas
1. Sleele I. Pipe wst b*
2. So** Incipient >jncov*r**
failure ileisct Ion
3. rmesfrlally
avallabl*
*. Lax con
!. Indicates if coat-
In* less Uian
Ideal
I. Lou cost 1. Subject to genome)
2. Simile trror
3. So** Incipient 2. Xot auantltatlve
failure detection
barrels); «ediu» spill 10.000 ta 100.000:
'very rovae atlmt* af inspection casts for year for ccawnenti
> i SOU MtlOB cost Sat to S290K- *1~i cost > ^2
-auli typically is Inspected: low cos:
174
-------
TABLE 52. STANDARD NONDESTRUCTIVE INSPECTION
, a
•b)
•c)
d)
•e)
Equipment"
Active Ultrasonics
.Detector sends
ultrasonics wave
pulses througn or
along oipel ine and
then detects dis-
torted wave pulses
or reflection times
of returned waves
from defect areas.)
X-Say
(Radiographic method
uses X-ray source to
penetrate through a
material . Intensity
of penetration is
modified by passage
through material
and defects. Visi-
ole contrast an film
snows defect free
areas and defect. }
Radio Isotopes
Gamma Ray
(Same type of opera-
tion as (c) but a
radioactive source
is used instead of
X-ray source. Source
and detector film at
180°, for on land
inspections. Source
and detector can also
be on same side for
back scatter method.)
Magnetic Particle
(Magnetize inspected
component apply and
magnetic particles
to magnetized area.
If flaw close to
surface, magnetic
particles will de-
posit themselves
along flaw because
of leakage in mag-
netic flux at the
discontinuities. )
Ultrasonic Imaging
(Holographic)
(Similar to other
kinds of holography
but uses an array of
transducers for high
frequency coherent
sounds in order to
penetrate inside
structure for 3-
dimensional acoustic
image view of inter-
ior of solid.
Device includes
scanning head and
jltrasonic hologra-
phy computer signal
processor to ••ecre-
ats
-------
TABLE 52 (continued)
... .- - .-- _ . - — .- -
Equipment3
f) Eddy Current
(A cietector coil
carrying alternating
current" fs brought
near a material
specimen and eddy
currents are induced
by electromagnetic
induction. Dis-
continuities in
material chamje
the magnitude of
induced eddy
current. )
g) Penetrants
(A liquid penetrant
is applied to the
inspected surface.
After sufficient
penetrating time,
the excess pene-
trant 1s wiped off
and a wnfte powder
is typically applied
to the surface.
After a period of
time penetrant seeps
out of crack and re-
duces the whiteness
of the powder and
the defect can be
observed. )
"efects Measured
1 . Internal defects
2. Cracks
3. Thickness
4. Voids
5. Corrosion
1. Surface cracks,
laps
2. Internal defects
with surface
openings
Sensitivity
1. Cracks 0.2 to
0.4 nin sensi-
tivity
1. Cracks, defects,
etc. of about
1 mm width
2. Much more sensi-
tivity than
visual Inspection
1.
2.
3.
4.
5.
6.
1.
2.
3.
4.
5.
Advantages0 Disadvantages
Locating defects 1. Does not work in non-
near surface metals
Low cost
Simple
Commercially
available
Good incipient
failure detection
Works well on
thin materials
Simple 1. Will not work with
Covers a large surfaces covered with
area oil, grease, paint.
Low cost etc.
Commercially
available
Good incipient
failure detection
' Identifies Reference number.
bSmall spill < 2,000 gal (50 barrels); minor spill 10,000 gal (238 barrels); medium spill 10,000 to 100,000; major spill
> 100,000 gal.
cVery rough estimate of inspection costs for year for components that would typically be inspected; low cost 0 - $20K;
medium cost $20K to $?OOK; high cost > S200K.
176
-------
TABLE 53. PIPELINE INSPECTION PIGS
Ejuiorent"
'WPS.LS5 THROUSH _
'a. '-aCTttic "1u» 1.
iltctromegnetlc) 2.
/Changes in *aP
cnicxness affects 3.
iagnetle field. -.
Incuced uonttic
'ilia and utoc- :.
tlon accsmolisned
«itr tiectrs- =.
•acnets ;r oer-
lanent taanets.; 3.
,3i ••fi'ter 1.
•irt?er neena-
msm in 3ig
transmits 1.
ranges in oi3e
:1aneters to s 3.
enacting device
in oig .nousing.) 4.
5-
5.
!cl Active ultrasonics 1.
^ijitrsscnlcai ly
scans siotl in* in 2.
transverse dlrtc- 3.
tlan ultng. active *.
ultrasonic scanning
tool.;
(d) Passive Ultrasonics 1.
.An escaping fluid
from a ploeline
leak exits sounds.
A cassivt ultra-
sonic ottactor.
-iountea in an oil
"?nt container.
Detects tne 'eeii./
It) *V_ Camera '.
• insoection
caners «1tn low
' i ont "V camera
jnc /ioeo taoe
ar *v -somtor. )
r' f ) '«uc Star ' .
>>uc;ear source in-
stalled in inspec-
tion r-ig. Minute
radioactive suanr.1-
riei transmtteo
;.f,ctS,..sur,d
Corrosion 1 .
.Harcsoots. if a.
'laws
;irt*i >eids, aits
Citnocic
arotaction
I~3roser aends of 2.
oioel ine
"ouges 3 .
.-jrinnit oraos
.•iyoronen -11s ters
'4easurts changes ' .
'n inslae 31ot-
iine Jiameter
Detects aents.
auckles *.
3etects oostrjc-
tlons
Changes in wall
tnickness
clat soots.
Partially closed
ve 1 w^s
'iot nail thick- 1
ness
Laminations
Inclusions
Cracks
iensitivity3
Severity of car- 1.
roslon in tnree 2.
ranges - 3.
la-30* of vnll 4.
:3-50: of nal:
>50r, of wall 3.
Aooroxinataly
1/8 incn oefKt
Severity of
aittlng 3.
T
AOniot tnanfes 1.
•:n «antnt record 2.
•i«w units art cur- 3.
rently under dewt1>
aut are of neo. cost
Cisacvantages
•iign cast
Jtfficult to -nteroret
magnetic anomalies
•ieouires luman inter-
aretatlon
:ltctrontgnet1c type
cannot determne if
ieftct is inside or
autsioe :f oioe
:e>«tnent magnet tyee
can ^tt stuck in
aisellne and diffi-
cult to remove
Anomalies around
jirtfi weid 2lf* Icult
to cetect
Coes not cetect tnin
cracxs very «eil
••ucn lower sensitivity
t.tan tteonetic 'li-x
insoection Dig
mgn cost
31f*1eult ts interoret
Seouirei nuevn
interoretatlon
Net widely used
i rewire ftin. tntarore tart ens.
detection of 1
luld Hcaeing
tnrdugn nalrllne
cracxs or snail
corrosion nolts.
visually insoeets 1
inside of ciot-
Hne Jor cracks.
olts. ate. I
Small nole 1
tnrougn cracks
. 3 to 5 gallons 1.
jer nour leaks
2.
. Sligntly oetter '. .
then visual 2.
insoection 3.
. 360° vienlng
. Sensitivity 1.
jnctrtain
cnrougn a iean is sensed
iy external detectors. i
;oi .'Itrasoni: fcolo- :.
-raonic. .Tsoino
;-
-------
TABLE 53 (continued)
••luiastnt
"sasurw
MMrgy :
around tnt slot dr-
cuaftmct. A ;on-
jltudlnal ttrtss
corrosion crtcx
s ion nflacts wtrgy
3aeK :o a uttctor
Transductr. -2tv1ct
4s 3lowi tnrougn
:1st 2y 4 ;.as
>:r
'Icncontact .r-ns-
roncantact trans- til cMnning
:h»r« 1s :1f*<-
culty cauoiing
sound TO oiDtlint
:nrougn a Hcuia
Z. "ay »or« v^ll it
30 ion
stag*
ic
:r»«ar«. * floau-
3i« :*•< i«uor,
MITCH ii r»jponsiv«
a jrtssurt and
<«!cc1ty il!"tr«n-
r'lU cauiM sy i
;«««, it sov«a
tRrouon at 3i3tl1n«
along titn * 'laid.
5«nsor ttool sovt-
-ant sirougn csnouit
it "station it i
3«int1ally -art
!*nsl7iv« man
iccustlc frnotc-
•lon ?ig
ouc-of-s«rvica
:ni
Z. :.aauir«* •iavaetti
ni- 's 'itrosustd
itO I'M I'M !QS1-
'ontd 3y jnnolii?.
t rolnt af lt*
-------
TABLE 53 (continued)
E ou i amen ta
. mi Eddv Current " .
'.Eddy current 2.
changes in non- 3.
magnetic tuning 4.
caused by defects 5.
are detected by a 6.
recording Imped-
ance bridge. )
INSPECTION PIGS WITH MANNED
Defects Measured
«'al i tnicKness ' .
Pits
Cracus
Holes
Corrosion
Surface or near 2.
surface defects
Sensitivity13
.angitudinai
cracks .004 in
deep by .4 in
in length can be
detected
Changes in wall
thick"«s« of }",
in a 0.4 in length
can be detected
Auvdntagosc
1 . 'tedium cast 1 .
2. Good Incipient
failure detection
3. Widely used
4. Commercially 2.
available
5. Locates defects 3.
near surface
4..
"icauvantages
Insensitive to cir-
cumferential cracks,
short cracks and
shallow cracks
Requires out-of-
service operation
Non-magnetic materials
only
Limited to a few 30
foot lengths of pipe
INSPECTORS (PUSHED OR INTERNALLY
POWERED THROUGH LARGE PIPE
( n ) Inspection methods 1 .
Available
'See Table
(0) Ultrasonic (Holo- 1.
graphic Imaging)
(See discussion (g)
Method was applied
to ALYESKA pipeline
using manned inspec-
tors and i powered
vehicle.)
(p) Ultrasonic Riser7" ?!1.
Unit removed in-
side riser while
the ultrasonic
transducer rotates
to obta in helican
scan.
INSPECTION PIG TRACKING
;q; Tracking 1.
(Inspection pigs are
located in pipeline
by monitoring sig-
nal from nuclear,
acoustic pinger or
nuclear source in-
side inspection pig.
Also cleaning pigs
or an acoustic
pinger inside a
polyurethane sphere
are often used. )
LINES)
A11 internal 1.
defects and oipe-
line corrosion
Some as (g) 1.
Riser thinning, 1 .
cracks, etc.
Locates stuck 1.
inspection or
cleaning pig
caused by pipe-
line defects—
improper bending
or medium or
major leaks
Best overall sen-
sitivity of any
inspect method
?or pipeline
Same as (g)
Better then 20?
of thickness
Sensitive only
to large defects
in pipeline
1 . Best overall in- 1.
cipient insoec- 2.
tion technique 3.
for internal exam-
ination of pipe- 4.
line
2. Operational under
limited use
1. Excellent hard copy 1.
pictures of inter- 2.
nal flaw in pipe-
line
2. Excellent incipient 3.
failure detection
3. Simple data inter-
pretations 4.
4. Commercially
available
1. Small enough to 1 .
pass around Dipe
elbows without 2.
jamming.
2. Lightweight
3. Commercially
available.
1. Simple locating 1.
methods
2. Low cost for
polyurethane
spheres
3. Commercially
available
Very high cost
Slow
Requires out-of-
service operation
Feasibility stage
for powered type
vehicle
High cost
Device must be designed
and engineered for
specific pipeline
Reliability and speci-
fications are uncer-
tain at this time
Requires out-of-
service operation
Not self-pro-
pelled.
Out-of-service
inspection.
Insensitive to most
pipeline defects
-
'identifies ?>eforence nunbcr.
DSmal1 soil! < 2,000 gal (50 barrels); minor spill 10,000 gal (238 barrels); nediun sn-m 10,000 to 100,300; major :pi11
> 100,000 gal.
Very rough estimate of inspection costs for year for components that would typically be inspected; low cost 0 - S20K-
medium cost S20K to S200K; high cost > S200K.
179
-------
TABLE 54. SURVEY-PIPELINE SYSTEM CHARTING AND DEPTH OF BURIAL
00
o
UN IAHU
(.) K4d.ii"
(H4d4r IMimlse pissus thi-utujli
earth ta pipe. A clwinje in
conductivity, purnilUivllr
ur per«e4blllly causes im-
pulse to be reflected lu
dulociur.)
UNUU4AUH
(b) Sui.4r-Sj.fcst.in
(Inspects far ti^~^-r.
features and bare spuls In
pliwllnes. An uuust)C4l
transducer In a lowed flsk
(UavalUuu. C)UM! tu Ike
Sea lull lorn) sends 4 pulsed
llu> slrenijlk al the echo
provides sufficient Infur-
•utla* lu dete»lw si/o
and uricnlitlun ol* pl|n:liiie
bare SlMils. )
(Iwu acoustical Iransikicers,
one ueiuflriiln9 Iliruwjh
uuu llul reflecls fru> pipe-
In a tin. fisk. Irwsducers
send pulsed vertical beams.
Ike slrenylh uf reflecled
echus pruvldi; sufficient
dala tu determine depth ul
burial.)
M)
Sfi!(c HuJujH-Jiikjf"
used acoustic
IIISOUU.TS Ira-
se III II abuvu
sliuials received
milted tu lite sin
red lime uhere
an ismuetric CMI-
luur vien ul tht-
I llUlllM) 411.1
Uelm. is Measured
I. Precise uni^uund
piltellHti lui-llOU
1. Knife vpoU in Iku
2. fi|Hi)iHU urlenUtiuM
1. lk.*|iOi u| bwUI
2. IreAck dultNU-iUuii
3. PifMiliMtt locations
dtld OrltillttlllOtl
2. 1W-HU
3. (IdMugu IMHH
«iichftrs
4. Hf&4ilii
1. KttiMlirus sillied llllerprut4lilHi
• uf data
1. Heiuilrus sillied Illleipret4llun
Uf ll4t4
I. H,,I a.«»i
sIlMled iwi lull
2. lll.jl. L..SI.
(continued)
-------
TABLE 54 (continued)
00
SUNVlIINU-MIM'UniHI lOcAIIIHI
(t| giver iMwj MKIUW.I«IJ
Position Mfsieoi
Ifelvef wils'ifie pipeline
Mil will hold i taunt buoy
line. A Mlcruuave position-
ing system antenna Is Munlcd
over ike buoy and is used lu
accurately lucaie the |iosi-
tlw al t diver.)
(bfvur walks pipeline JIK
acoustic lr*m|MHii.t!i will
(lit Hlyer
ultli
lKIirer'uses~Kin
feet "1 water
depth
1. Precise pipeline
location to MM
feel uf water
depth
1. Visual
1. Visual
2. Accurate bottom
topography
measurement
1.
2.
1.
4.
».
0.
1.
2.
1.
2.
3.
1.
2.
3.
!*-«*»l.
Covers up tu altuul 1.
1/2 mile par day 2.
Can be used in
turbid waler
Provides visual
inspection by diver
Parllculary useful
for new pipelines
I lit el lent Incipient
failure detection
Inspection services
available
Same as for (d) 1 •
except item 4 2.
Diver can leave a
locating transponder
at an area where pipe
needs repair. Ibis
allows easy location
for later repairs
Simple 1.
low cost
Some Incipient 2.
failure detection
Simple 1.
low cost
Some Incipient 2.
failure detection
««.•«•••,«
High C..SI
Currently tWtlut lu aliuul
HI leet ol water tkylli
High cost
Uevi!lo|Hmj|il stage
Visual only and subject to
Interpretation
Cannot 1 Ind scoor areas that may
be covered up by current after
storm
Visual only and subject to
Interpretation
Cannot find scour arc-as that may
wealun structure but may be
II) CmllMw *'nl!orins
(IK'clrodc resistance
Oeplli Huasurlny1*
Hjynetlc Bti«conb4
tlectroMipietlc Heai.oii1-". )
I. lonliuuous scour
MNiltorlnij
I. Siour resulutici*
to within a tout
I. Detect scour areas
tnat my wealcn
Structure but *uy
be covered up by
stum
2. Continuous
•unllorlm
). (specially useful lu
deep water where
diver or tubMerslble
•chicle inspucllons
Are costly
4. fiood Incipient
failure detection
covered up after stom
I. Nedlun cost
2. Nol particularly advantageous
al siuirl depths (I.e.. ISO feet)
froai a cost consideration
1. Sysleib require development and
cngliKerlnii design
'likntlfles heference i
spill - I.OHO yal (bO ba.relsl; -l,«.r spill 10.000 «al (21* harrels); ledlu. spill 10.0110 tu 100.000; major spill -. IOO.UM gal.
cosls '"' "" "" "-*««"«« "«» •»"'< typically be Inspectedi low cost 0 - $20m Medluii cost »J(K
-------
TABLE 55. MISCELLANEOUS
00
ro
Detects Measured
1. Outside forces
Equipment
(a) Passive Acoustic Array
Outsideforces
lArray of acoustic transducers
installed on or near pipeline
to detect and locate acoustic
sounds generated at the source
of outside damage from equip-
ment impacts, digging, etc.)
(b) Passive Acoustic Machinery 1. Internal defects in
Damage1* " "" machinery
TAcoustic transducers hand held 2. Bearing damage
or permanently installed on 3. Valve damage
machinery detect abnormal
sounds emitted when internal
defects start to occur in
machinery. Strain transducers
or accelerometcrs can also be
used for this detection.
System usually provides con-
tinuous monitoring and alarms.)
1.
Sensitivity
Detects and
locates serious
third party
damage using
widely separated
sensors
Advantages
(c) Liguid Level Sensor
TOetects~leveT ofliquid in
tanks, piping, etc. using any
one of waiiy available methods-
ultrasonic, optical, microwave,
nuclear, etc.)
1. Oil levels
2. Oil leaks into
containers
1. Better than U
(d) laser Holography Interferometry 1. Small topographi-
ISurface Is Inspected
with a laser. High
resolution film is exposed by
laser light reflected from
surface. Developed film pro-
duces a 3-diiuensional hologram.
If surface undergoes a topo-
graphical change, another
exposure is made on the same
film and minute changes in
topography can be recorded.
Changes can be detected
liy fringe patterns shown in
holographic reconstruct ion.)
cal changes
1. Wavelength of
laser light
1. Excellent incipient
failure detection
2. Computerized auto-
matic system can be
adopted to existing
control systems
3. Can be used
underwater
4. Permanent records
5. Can be used at all
times
I. Commercial system
available
2. Good incipient
failure detection
3. Reduces machinery
maintenance cost
4. Permanent records
5. Continuous
moiiitorin
-------
TABLE 55 (continued)
00
CO
Equipment
Defects Measured
Sensitivity0
(e) Magnetic Chlj)
{Installed on machinery
and periodically examined.)
1. Internal defects
in machinery
2. Bearing damage
(f) Oil Odor
finspector detects oil from
smell.)
1. Oil leaks
I. Small leaks
Advantages
OTHER METHODS
(s) Integrated Machinery Inspection7'
(h) Control ROOM Monitoring. Alara Shut-off
(') Uy-H^LST Infrared
(J) ih§!mLF»JM
(k) Hicrowave
(1) Filtered Particle
(HI) Sonic
I. Simple
2. Low cost
3. Comit-rctally
available
4. Some Incipient
failure detection
5. Reduces Machinery
maintenance costs
6. Continuous moni-
toring possible
1. Simple
2. Can be used in dark-
ness or in bad
weather
3. Particularly useful
In darkness at sea
Disadvantages
1. Subject to personnel error
2. Hinds may cause oil odor not
be detected
3. Not quantitative
to
Identifies Reference number.
Small spill < 2.000 gal (50 barrels); minor spill 10.000 gal (238 barrels); Medium spill 10.000 to 100.000; major spill > 100,000 gal.
GVery rough estimate of inspection costs for year for components that would typically be Inspected; low cost 0 - $20K; medium cost J20K
to $?OOK; high cost > )200K.
-------
Because most pipelines are buried, oil escaping in either small or large
quantities may not be observed, or, if observed, only small quantities or
traces may be visible. Also, in some types of soi-1, the oil may take a
particularly long time to surface. These migration effects52»5^»55»ss
occur because the soil between the line and surface absorbs and spreads
oil below the surface. Often this causes the distance between the dis-
covered leak and the actual leak location to be quite large. When pipe-
lines are installed in gravel bottoms or sandy bottoms, this can become
a particular problem.
Underwater lines have other significant problems with visual inspec-
tions. If a leak occurs underwater, the oil can be spread by currents.
Many leaks cannot be detected by visual observation. For example, a small
crack in the pipeline may allow seawater to seep inside undersea pipelines
which are located at depths at which external pressures are greater than
internal pressures. This water can cause excessive internal corrosion.
In the extreme case, a crack can grow undetected to a critical size, then
rupture and cause a major pollution incident.
These and other factors make it difficult to detect and locate a pipe-
line leak. Furthermore, numerous instances of careless losses from small
quantities of oil, such as from dumping crackcase oil, losses from domestic
fuel oil tanks, or spills from ships present difficulties in discriminating
between a pipeline leak and other oil spillage.
Methods that aid visual inspection are quite useful in improving the
ability of an inspector to detect and locate a leak. Aided inspections
might be an inspector walking the line with a hydrocarbon detector, an in-
spection plane eqippped with remote oil spill detectors or an inspector on
the deck of a ship or at a marine terminal viewing the water with an ultra-
violet light source or low-light TV monitor. Use of a hydrocarbon detector
while walking the lines in populated or hazardous areas appears particularly
advantageous for detecting small leaks, such as could occur from corrosion.
These small-volume corrosion leaks may take either a long time to surface
or may only surface trace amounts. Inspection with a hydrocarbon detector
may provide detection of a leak before it can cause any surface damage and
possibly before it can cause damage to underground water supplies.
5.3.2.2 Oil Spill Detectors On or Near the Water-
Oil spill detectors such as those in Table 47 might be used to supple-
ment visual inspection. These detectors are of low to medium cost and do
not interfere with pipeline operations. A few advantages of these devices
are that they are not subject to personnel error and most provide contin-
uous detection. Many oil companies are currently using oil spill detectors
(Table 47(c)) at supertanker terminals and at other marine terminals. How-
ever, most of these oil spill detectors are still under development or
evaluation and are not in common use.
184
-------
Two oil spill detectors appear to have a particularly good potential
of effectively minimizing oil pollution. One is the platform infrared
type (Table 47(c)) for use at marine terminals. This may be particularly
beneficial when a considerably amount of petroleum is transported. The
other is the bistatic, active, infrared type (Table 47(j)) for channels
and rivers.
5.3.2.3 Internal Fluid Variations During Oil Transfer--
Unusual fluid variations in a pipeline during transport can indicate
possible pipeline system leaks. A variety of methods that detect fluid
variations are available for leak detection and inspection of the pipe-
Tine system during normal operations when oil is transported. These are
given in Table 48 and include:
• Pressure deviation
• Volume comparison (differentials)
• Flow rate comparison (differentials)
• Flow rate deviations
• Over-and-short accounting (long-term)
• Mathematical modeling
• BS&W deviations
• Pressure difference with reference and threshold
• Fluid mass deviations
• Negative surge
• Inspection pig-passive ultrasonic
• Pressure fluctuations
• Production surveillance monitor-pressure fluctuations.
These methods normally cause only minimal interference with line operations
and most are of low to medium cost. All methods provide continuous sur-
veillance of the line except the passive ultrasonic pig and long-term over-
and-short. Most methods primarily detect major leaks or catastrophic
failures. They are used to trigger controls to quickly shut the system
down and thus minimize the volume of spillage. None of these methods, ex-
cept possibly mathematical modeling, can reliably detect small leaks (a
few barrels per hour) over a short period of time (hours). A few methods,
however, are capable of detecting small leak rates over long time periods.
The first five methods are commonly used for pipeline leak detection
by the U.S. oil industry. In general, modern supervisory control systems
are computer controlled and many provide continuous monitoring of pressure
deviation, volume comparison, flow rate comparison and flow rate devia-
tions. Hardwired meter flow comparators are often installed in areas where
the risk of pollution is high. They are particularly effective where more
185
-------
of these leak detection methods might be implemented. A leak detection
system, for example, might continuously monitor pressure drop and compare
flow rates and flow deviations between two points. If one or more measure-
ments exceed preset values, an alarm would be sounded and appropriate
valves and equipment would automatically be operated to shut down the sys-
tem. Normally, an excessive pressure drop combined with a flow rate in-
crease are strong indicators of a leak.
Varying operating conditions, fluids, transients, temperatures, line
pack and other factors cause changes in these pipeline system measurements
that often result in false indications of leakage and cause false alarms
and shudowns. In order to account for varying conditions and minimize
false alarms, three things are normally done. First, a variety of highly
accurate metering equipment (i.e., pressure transducers, flow meters,
temperature gauges, etc.) are installed. Then sophisticated computer pro-
gramming with various scanning rates, error bands, checking schemes and
displays are used in the monitoring systems. Finally, the system sensi-
tivity is decreased and setpoints raised to minimize false alarms. For
most lines, this results in decreases in sensitivity that precludes de-
tection of small or minor spills.
Short-term losses (one hour or less) of from 1 to 5 percent of through-
put are normally detectable with these types of continuous monitoring sys-
tems. For a throughput of 10,000 bblh, this corresponds to a leakage rate
detection of 100 to 500 bblh. Long-term (24 hours or more) detection of
losses ranging from 0.2 to 1 percent of throughput might be expected; this
corresponds to leakage rate detection of 20 to 50 bblh. Supplementary
methods such as mathematical computerized modeling, which accounts for many
varying operating conditions and fluid changes, might be adapted to an
existing supervisory control system to provide more sensitive (up to a fac-
tor of 10) leak detection. If implemented, small and minor spills might be
detected. Mathematical modeling is currently being developed and evaluated
by a number of companies. The method has been implemented on a few crude
lines.
Long-term over-and-short accounting is the most common means of pro-
viding an indication of a leak. It requires only conventional metering,
i.e., LACT and tank gauging, and standard accounting of receipt and de-
livery tickets and inventory of the lines and tanks. This existence of
leakage is normally suspected when a shortage exceeds about 0.1 percent
of long-term (weekly, monthly or yearly) throughput.
On large lines, gross barrels in and out, then net barrels in and out,
are looked at on the basis of total volume. The only difference between
net and gross is in the BS&W. A loss/gain is run on water as is done with
oil because BS&W is a mirror image, i.e., BS&W gain is an oil loss or
vice versa. BS&W can provide both leak detection and a check or gross
volume to identify volumetric errors. This leak detection method is par-
ticularly useful in underwater lines where external pressures are higher
than internal pressures and seepage of water can occur into the line
through a crack. The measurement is quite inaccurate (see Section 4.3.2.3),
and only large leaks are capable of detection.
186
-------
Pressure difference with reference and threshold is a recently patented
method. At this time there is insufficient information for proper evalua-
tion.
Detection of leaks by monitoring the deviation in mass at the line in-
put and output is attractive because pressure and temperature changes do
not affect the measurement. The method requires that density be measured
directly. Fluid mass measurement is currently carried out on mixtures of
light products, i.e., natural gas mixtures. In this application, the coef-
ficient of expansion of light products is not known, and this is the only
method that has been used to successfully measure these fluids.
Negative surge leak detection is used on a few pipelines. Leak detec-
tion accuracies (about 600 bblh) are not as good as other methods. How-
ever, the method is quite simple and it can provide an inexpensive check
or backup system for large leaks.
The ultrasonic inspection pig is propelled through the line by the
fluid flow. It detects a leak by sensing the acoustic energy generated
at the pipeline leak source. This energy is propagated via acoustic waves
from the pipe inner wall through the fluid to ultrasonic detectors in-
stalled in the pig. Leak location is determined from tape recordings of
both the time the leak is detected and distance the device travels. The
method is currently used in a few lines. Leak detection capabilities of a
few barrels per hour are claimed. Because of the high cost for a single
inspection of the line, it is used at infrequent intervals ranging from
six months to five years. This type of inspection pig has a number of
limitations. It is particularly difficult to apply the same unit on a
variety of lines because the leak detection sensitivity varies with the
type of line, and leak detection background noise often presents false
indications of leakage.
Experimental work currently is being carried out on the pressure fluc-
tuation method. The basic concept is to detect the fluctuations with an
array of pressure or acoustic sensors inserted in the line. The fluid
turbulence generated by a leak is propagated in the fluid along the line
and sensed by the detectors. The method is attractive because of its sim-
plicity and the wide transducer spacing. The method is also attractive
for detecting and locating the wave fluctuations in the line that are gen-
erated by impacts from outside forces.
The production surveillance monitor measures flow rates in the typi-
cal two-phase flow in production lines. The monitor has been proven in
field tests and has been implemented on a number of lines. When used in
a continuous monitoring system, it can provide indications of anomalous
changes or well problems such as pipe leaks or pump malfunctions.
187
-------
5.3.2.4 Detecting and Locating Leaked Oil On or a Short Distance from the
Pipeline--
Methods for detecting and locating leaked oil using sensors located on
or at a short distance from a pipeline are given in Table 49. These methods
differ from those in the previous two sections because oil need not travel
to the surface of the soil or water for detection and the methods poten-
tially provide accurate leak location. A number of these methods are new
or are in the feasibility or development stage.
The most promising of these types of continuous monitoring leak detec-
tion methods are:
• Passive acoustic array
• Shroud with electromagnetic pulsed coaxial cable
• Continuous thermistor
• Tape detection.
The passive acoustic array (see also Section 5.3.2.3) potentially is
capable of detecting and locating the following: (a) leaks to a few bar-
rels per hour; (b) internal pipeline defects that may result in leaks; and
(c) (most importantly) outside damage and pipeline ruptures. The other
three methods are capable of detecting leaks ranging from a few barrels
per hour to pipeline rupture. However, these three methods are not con-
sidered to be as effective as the acoustic array and are expected to be
more expensive. The double walled pipe is too costly to be considered
except on short sections of the line in sensitive areas. The oil-soluble
tubing method has been a failure on at least one pipeline. All of these
continuous monitoring methods would require retrofitting the line, except
possibly the passive acoustic array when used for outside damage detection
only (see Section 5.3.273).
Other methods in Table 49 can also be applied to the entire line but
are normally used on a periodic basis for: checking the integrity of the
line, locating leaks, or locating and verifying a suspected leak. The
nuclear tracer method appears to be most effective, but very costly, for
periodic checking of the line and locating leaks. It has been implemented
on a few lines, and high leak location accuracy is claimed. The hydrocarbon
probe is considered to be quite effective in detecting and locating small
leaks in undersea or underwater lines. Rods installed permanently on the
pipeline and used in conjunction with acoustic sensors are used by a few
pipeline companies to locate and verify a suspected leak. Nitrous oxide
injection is costly but has been shown to provide very accurate leak loca-
tion. These methods normally cause only limited interference with the line.
The remaining methods in Table 49 are limited to checking for suspected
leaks of only sample areas of the pipeline. Some would require excessive
costs for inspecting a complete line, while others are not considered prac-
tical for the entire line. Although useful in locating known or suspected
leaks, these methods are limited and not considered as practical means of
significantly reducing petroleum leakage.
188
-------
5.3.2.5 Periodic Pressure Tests—
The periodic pressure tests given in Table 50 provide checks of the
pipeline for leaks and/or checks for internal defects that may result in
a leak. These methods require either limited interference with pipeline
operations and/or retrofitting. Pressure tests primarily for detecting
leaks are:
• Pressure static
* Hydrostatic
• Pressure difference
• Dye tracing
• Seal leak detector-joint type
• External hydrostatic
• Seal leak detector-thermistor type.
Pressure static, hydrostatic and pressure difference are the only
methods that may be capable of effectively testing the complete line for
leaks. However, these methods require that the lines be shut down and
they do not locate the leaks. They are generally considered as pressure-
volume-temperature-time (PVTT) test methods. Although less sensitive,
the pressure static method can be carried out in a relatively short period
of time. The other two methods require that the line reach temperature
stabilization, which may take as much as three days. Also pressure static
does not have the oil spill risk3 problem that exists when petroleum is
left in the line, as is often done, for the other two methods. Simple
hydrostatic proof testing is widely used in the petroleum industry to both
test new lines and to check existing lines. For typical lines, the test
is carried out every two to five years. Leak sensitivity is about 0.5 BBLH;
this is about 10 to 100 times more sensitive than the methods discussed
previously. The pressure difference method has a demonstrated leak sensi-
tivity of about 0.05BBLH on foreign lines. Despite its superior leak sen-
sitivity, the high cost of the method has generally precluded its use in
the U.S.
Even better leak sensitivities are expected if the PVTT leak detec-
tion systems are supplemented by mathematical modeling of the line. Should
a leak be detected, methods such as described in Section 5.2.1.3 could be
used for leak location. Additionally, insertion of dyes during these tests
can be advantageous in providing leak location, particularly when residuals
from previous spills cause difficulty in detection. The remaining three
methods can be implemented advantageously in certain limited applications.
Methods requiring elevated pressures present a serious spill risk problem
when the line is filled with petroleum. If a line should fail under pres-
sure the petroleum would present a serious hazard particularly in high
risk locations such as urban areas or near water.
189
-------
Pressure test methods for detecting and locating internal defects in-
clude:
• Reflected pressure wave
• Acoustic resonance
• Passive acoustic array - acoustic emission.
The capability of checking lines for internal defects has been demon-
strated on petroleum pipelines for the pressure wave and acoustic resonance
methods. Both methods require that the pipeline be emptied and filled with
a gas. The acoustic resonance technique appears best suited for testing
inaccessible areas of a pipeline. These methods typically cause only lim-
ited interference with the line. A permanently installed passive acoustic
array is in the development stage, but has been shown to be effective for
detecting leaks and incipient failure during experimental tests on actual
lines. Internal defects, however, can only be detected when the line is
pressurized. The method typically requires close sensor spacing and is
considered a retrofit. However, in high-risk areas, the method might be
considered despite high retrofitting costs.
5.3.2.6 Corrosion Inspection-
External pipeline corrosion is the major cause of spills from older
pipelines (those installed before 1950). Since the 1940's, improved tech-
niques, such as pipeline coatings and cathodic protection systems, have
reduced the number of oil spill incidents from new lines to a level about
equal to internal corrosion. This can be seen in Table 25, which identi-
fies causes of spills in the United States. In contrast, leaks from pipe-
lines caused by defective seams and welds were similar for both the old
and new pipelines.
A large number of inspection and leak detection methods are available
to detect pipeline corrosion damage (wall thinning, cracks, etc.) that can
result in leaks. The main ones (see Table 51) are:
• Flow sampling (internal corrosion) - periodic
corrosion rate coupons
laboratory analysis of petroleum
miscellaneous trap monitoring (pig traps, strainers, etc.)
• Corrosion metering (internal corrosion) - continuous
corrosion rate probe
• Corrosion metering (external corrosion) and eathodic protec-
tion monitoring
potential measurements
continuity measurements
manufacturer recommendations
• Holiday detector (external corrosion) - periodic
• Visual inspection - periodic
190
-------
• Standard non-destructive inspection equipment
active ultrasonics
X-ray
gamma ray
magnetic particle
ultrasonic imaging
eddy current
penetrants
• Pipeline inspection pigs
magnetic flux
active ultrasonics
TV camera
stereo pair
eddy current
electromagnetic noncontact transducer (EMAT)
manned inspections using standard NDI methods
• Other
passive acoustic array-acoustic emission.
External corrosion metering and monitoring of the pipeline cathodic
protection is carried out on most United States pipelines (see DOT Regula-
tion Title 49, 195.2 and 195.4). This has been highly successful in re-
ducing corrosion damage and the leaks that result from corrosion. New
techniques, such as those using computerized modeling, can define the po-
tential profiles and the need for additional anode protection. Trend in-
dications of internal pipeline corrosion can be obtained using flow samp-
ling methods and corrosion rate probe.
Standard non-destructive inspection (NDI) equipment is commercially
available for measuring actual corrosion damage. Active ultrasonic NDI
equipment is highly accurate (better than 1 percent thickness to flaw
size resolution) and reliable, and is often used for corrosion measure-
ments. Costs of excavating for inspections or boring through the soil to
the pipeline to provide contact of ultrasonic sensors to the line are quite
high. Thus, this equipment is limited to inspections of only sample areas
of the pipeline. High costs prevents this type of equipment from being
considered as a practical way of significantly reducing pipeline petroleum
leakage.
Only pipeline inspection pigs, used periodically, appear to have the
potential of substantially reducing pipeline leaks caused by corrosion.
This equipment provides excellent incipient failure detection of corrosion
damage. The ultrasonic imaging inspection pig (see Table 53(g) and (o))
provides a 3-dimensional view of the inside of the pipeline wall. It po-
tentially can detect all pipeline corrosion damage and eliminate most leaks
caused by corrosion. The equipment has been successfully used on the Alyeska
pipeline. However, ultrasonic imaging inspection pigs have not been devel-
oped to the extent that a device can be propelled through the pipeline by
fluid flow; this development is necessary for practical application to most
petroleum pipelines. Magnetic flux type inspection pigs (see Table 53(a))
191
-------
are commercially available from AMF Tuboscope and Vetco. These devices
have been used for a number of years and are capable of detecting most
pipeline corrosion damage. TV camera inspection pigs and other devices
can be used to complement the magnetic flux inspection by providing vis-
ual inspection of areas where interpretation of anomalies is uncertain.
A corrosion maintenance program used in conjunction with inspections
and existing cathodic protection improves the effectiveness of corrosion
prevention. For example, chemical inhibitors used at frequent intervals
are effective in preventing internal corrosion.
5.3.2.7 Standard Non-Destructive Inspection—
A wide variety of standard non-destructive inspection equipment
(Table 52) are available to periodically detect most pipeline system
damage. Actual leakage, as well as internal and external damage that
can lead to a leak, can be detected. Methods that are particularly sen-
sitive to defects in petroleum pipeline systems and are in wide use in-
clude:
• Active ultrasonics
» X-ray
• Gamma ray
• Magnetic particle
• Ultrasonic imaging
•• Eddy current
• Penetrants.
This equipment is effective and practical for inspecting pipelines under
construction or above ground. Cost considerations make the equipment im-
practical for use on buried pipelines. Since most of the petroleum pipe-
lines in the United States are buried, the use of this equipment is very
limited. Thus, these methods are not considered viable solutions to sig-
nificantly reducing the incidents or volume of spills.
5.3.2.8 Inspection Pigs—
In addition to detecting petroleum pipeline corrosion damage that can
lead to leaks, inspection pigs detect other pipeline damage, such as defec-
tive pipeline seams, pipeline welds and pipeline leaks. Because of high
cost, inspection pigs normally are used to inspect the line at intervals
ranging from six months to five years. These devices have inspected over
100,000 miles of pipelines.
A variety of inspection pigs (Table 53) with different operating prin-
ciples are available for internal inspection of the pipeline. Inspection
pigs include magnetic flux, electromagnetic, kaliper, active ultrasonic
passive ultrasonic, TV camera, nuclear, eddy current, ultrasonic imaging
and infrared units. Most pigs are propelled through the pipeline by oil
192
-------
or other fluids flowing through the pipeline. A few types are
pushed through the pipeline. In large-diameter pipelines, internal inspect
tions can be carried out with inspectors. In this type of inspection,
equipment and inspectors are propelled through the pipelines by vehicles
ranging from a simple dolly manually pushed through the pipeline to a
powered, controlled-environment inspection chamber with instrumentation
trailer.
A pipeline cleaning pig, typically a brush type, and a dummy inspec-
tion pig are normally run through the pipeline prior to actually running
the instrumented inspection pigs. A pipeline should be inspected with a
pipeline inspection pig before the oil is initially transported to obtain
background information. Calibration blocks for some types of inspection
pigs are put on the pipeline to provide operational and sensitivity checks.
Tracking systems for inspection pigs (Table 53(p)) are used in case the
inspection pig gets stuck in the pipeline and for identifying defective
areas.
The inspection pigs that are effective for detecting corrosion damage
are also the most effective for detecting other internal pipeline damage.
Five types of inspection pigs appear to be particularly effective. They
are:
• Magnetic flux Table 53{a)
• Kali per Table 53(b)
• Active ultrasonics Table 53(c)
• Ultrasonic imaging Table 53(g)
• EMAT Table 53{h).
A low-cost active ultrasonic device for detecting wall thickness only
was recently under development by Harry Diamond Laboratories for inspection
of U.S. Navy pipelines. The ultrasonic imaging inspection pig is capable
of providing a 3-dimensional view of line defects. The device is not cur-
rently available for long lines or wide range of pipeline diameters.
The EMAT inspection pig is in the feasibility/experimental stage.
Test results indicate that it can detect longitudinal stress corrosion
cracks and non-localized regions of corrosion.
At this time only the magnetic flux and kali per inspection pigs are
available for widespread pipeline inspection. The magnetic flux inspec-
tion pig has been used to inspect over 50,000 miles of pipeline. However,
the device is insensitive to longitudinal stress corrosion cracks and to
non-localized regions of corrosion. The manufacturer claims this equip-
ment to be 95 percent reliable in detecting pipeline defects. An example
of the capability of the magnetic flux pig can be demonstrated by the re-
sults of an inspection carried out in a line (large-diameter 200-mile
crude line) visited by project personnel during this study. In this case,
the magnetic flux pig detected and located over 200 defects.
193
-------
5.3.2.9 Survey-Pipeline System Charting and Depth of Burial--
Typical .inspection methods for charting movements and measuring depth
of burial and erosion of pipeline system components are given in Table 54.
All methods except those in Table 54(h) are carried out periodically. None
of these methods interfere with pipeline system operations.
Excessive movements of pipelines can cause abnormal stresses that may
result in a leak. These movements are generally due to natural causes.
Periodic inspections of pipeline movements provides data to chart the move-
ment history of the line. This information can be used to determine whether
excessive stressing of the line might be occurring and would allow remedial
action to prevent damage. Methods (a) through (e) in Table 54 can provide
such inspection. Of these, the radar and sonar methods appear to be the
most effective and are low cost.
Maintaining pipeline depth of burial is extremely important in the pre-
vention of pipeline damage from outside sources, particularly for under-
water lines. Proper coverage of the pipeline minimizes damage that might
occur from anchor dragging, machinery impacts, etc. The sonar method with
penetrating and sub-bottom profiling is currently used on a number of under-
water lines to effectively perform these inspections. Low-cost systems are
commercially available. The radar method is one of several that can pro-
vide depth of burial inspections for onland pipelines.
In general, excessive scouring does not pose a great risk for most
underwater lines and other components of pipeline systems. But inspections
are necessary because of the disastrous effects, such as an offshore pump-
ing platform collapse. Periodic inspections by divers are normally suffi-
cient.
5.3.2.10 Miscellaneous Methods-
Inspection methods for petroleum pipeline systems that are not covered
by the first nine categories are included in Table 55. One of these
methods has the potential of significantly reducing oil pollution inci-
dents. The others are either already required by existing regulations and
are adequate, or have only very limited application.
The passive acoustic array has the potential of being the most effec-
tive of any leak detection and inspection method. It is intended to reduce
the number of incidents of damage by outside forces. This is done by de-
tecting acoustical sounds generated external to the line before the pipe-
line can be impacted and damaged by outside forces. The method can be
adapted into existing supervisory control systems or it can be implemented
as a separate system. Implementation would require either limited pipe-
line installation or retrofitting. The method is currently in the devel-
opment stage.2
Currently under development by NDE Technology, Inc.
194
-------
Inspections of pipeline conditions such as overpressures and excessive
shock load devices are essential to all pipeline systems. To be effective,
these inspections should be carried out frequently, and systems should in-
clude alarms with suitable shutdown and relief devices. The United States
has adequate regulations for these types of inspections.
A number of inspection methods such as given in (b), (e), and (h) of
Table 55 can be effective in preventing shutdowns or for providing warn-
ings of impending failures of various equipment. These methods may re-
duce maintenance costs, but are not effective in substantially reducing
the oil spill risk because most pipelines have redundant or axuiliary
equipment in the event of failure.
5.4 STATUS
5-4.1 Current Status of Leak Detection and Inspection Methods
The current status of the leak detection and inspection methods are
identified on page . Status is separated into the following five
phases:
• Required by U.S. regulation
• Common use
• State-of-the-art
• Developmental
• Feasibility.
5.4.2 Pipeline Regulations and Recommendations - U.S. and Foreign
Countries'
Liquid pipelines in the U.S. are regulated by two main agencies, the
Department of Transporation and the Department of the Interior. Other U.S.
Government agencies (some by delegation), such as the U.S. Coast Guard for
deepwater ports1", also regulate liquid lines. Many foreign countries such
as the United Kingdom, Switzerland, Italy, the Netherlands and West Germany
also regulate their liquid lines. In some areas, foreign regulations are
more stringent than those in the U.S. In addition to existing regulations,
trade organizations from private industry often recommend practices based
on recent developments in industry.
Existing liquid pipeline inspection and leak detection regulations
have been reviewed and recommendations made in a few recent studies. One
study (Reference 51, pp. 4-1 through 4-301) sponsored by DOT provided a
detailed review of regulations and industry recommendations of offshore
and on!and lines. Another study71 sponsored jointly by DOT, DOE and DOI,
reviewed inspection/testing/monitoring and regulations for offshore struc-
tures and liquid lines. Details of specific government agency jurisdic-
tions are included in these reports and this information will not be re-
peated here. A third study11* conducted for USCG and DOT provided recom-
mendations of inspection methods and procedures for deepwater ports. Both
195
-------
onland and offshore lines were included in the study. An overall review
of U.S. pipeline regulations is presented in Reference 1. The study iden-
tifies a number of significant problems and weaknesses in existing U.S.
pipeline safety programs. No new regulations are specifically identified
for the Alyeska pipeline, but there has been a substantial effort in mini-
mizing the risk of serious spills. Studies with recommendations have been
made and leak detection and inspection methods developed specifically for
the Alyeska line. At this time, no in-depth report of this effort is pub-
licly available.
U.S. regulations for liquid pipelines have helped to improve the ac-
cident record of the petroleum industry. These regulations are normally
presently in the following six areas and are listed on page 197. Only
two of these areas, operation and maintenance and pressure testing3 are
pertinent to this study. There are a number of reasons for emphasis in
these areas; a few will be discussed. First, the main objective of this
study is to recommend a viable and cost-effective spill prevention program
that would significantly reduce spill incidents and volume. It is expected
that this can be accomplished by implementing scheduled line inspections
and/or leak detection. Implementation of a program that includes methods
for the other four areas is not expected to significantly reduce spill
volume because of the following: (a) current regulations; (b) recommenda-
tions by various pertinent organizations; and (c) oil company programs are
already quite effective in preventing spills.
Spills normally occur after the line is designed, constructed and
pressure-tested; and spill frequency depends on factors such as pipeline
age. It is in this time frame that leak detection and inspection methods
might effectively be used to prevent spills and minimize spill volume,
and where a void exists with current practices in the U.S. Thus, existing
regulations and recommendations for leak detection and inspection that per-
tain primarily to operation and maintenance'3 are reviewed in the subsec-
tions that follow. For comparison purposes, existing regulations and
recommendations for gas lines are also included. Regulations of U.S. and
foreign countries, recommendations, and also maintenance practices for a
typical well-maintained pipeline are included in the subsections that
follow.
Liquid Lines
• U.S. Regulations - Liquid Lines
1. DOT CFR 195. - U.S. Regulation LIQUID LINES
49 CFR 195 - Part 195, Title 49 Code of Federal Regulations,
Minimum Federal Safety Standards for Liquid Pipelines, ap-
pearing in 34 F. R. 15473, October 4, 1969, with ammendments
up to and including Amdt. 195-14, Federal Register Vol 43,
No. 84, May 1, 1978.
aThere are no U.S. regulations for pressure testing in-service of existing
operational lines.
Other regulations, recommendations and review are provided in Appendix A
and in the referenced studies.
-------
AREAS OF LIQUID PIPELINE SYSTEMS UNDER U.S. GOVERNMENT REGULATIONS
vo
GUKRM.
Scope
Definitions
taller Incorporated by reference
Acceptable petroleuM coowidltles
transportation of certain comvantfes
transportation of convmdltles III pipelines constructed wltk ether I
steel pipe
Responsibility of carrier for coopllance with this part
ACCINNI WhimlNG
Scope
lelcphonlc notice of certain accidents
Accident reporting
Instructions lor preparing DOI I'ona /QOO-1
Changes In or addlllant t« accident retnrt
Carrier assistance In Investigation
Supplies of accident repurt 001 fora NKJO-I
Scope
Design temperature
Variations In pressure
Internal design pressure
Internal pressure
External loads
No-pipe
(bed pipe
Valves
fittings
Changes In direction: Provision fur Internal passage
fabricated branch connections
Closures
Flange connection
Station piping
lafcrlcated aMrafelles
nbnve ground tanks
tOHSIIIUCIIUI
Scope
Coieillance xlth speclflutlons or standards
Inspection-general
Kilcrlll Inspection
Melding of supports and braces
flpellne location
tending of pipe
Melding: General
Melding, niter Joints
Heldlnn: Seam offset
CUB1WCJIMI (Continued)
Heidi: Filler aKtal
Meiers: testing
Melding: Heather
Melding: Arc burns
Melds andxctdlng Inspection: Standards or acceptability
Melds: Repair of defects
Mel*: ftnnval of defects
Melds: Nondestructive testing and retention of testing records
Internal corrosion protection
titental coating
Cathodlc protection systeui
lest Ifads
Installation of pipe In a ditch
Cover ever burled plyeliM
Clearance betneen pipe and underground structures
lack HI line
«bo»» ground cotiionentt
Crossing of railroads and hlifhmys
Valve*: General
Valves: location
fwnjilng e««l|nient
NH>M ground tanks
Construction records
Scope
General re«|ulraiients
testing ol components
tot ncdlum
letting of tie-Ins
Hecords
Scope
General requirements
Naps and records
HatluMi operating pressure
CoMunlcatlons
line Barkers
Inspnctlon of rlghls-of-tiay and crossings niihr navlaable *»t«rs
Calhudlc protection
Ixleiual corroslim control
Internal corrosion control
Valve Mlntenance
flpellne repairs
flpe amvenont
Scraiwr and sphere facilities
Overpressure safety devices
rireflghtlng eoulpnmt
Storage vessels
Slg«
Security of facilities
i nr open flanes
-------
2. PCS #9 - U.S. Regulation LIQUID LINES
OCS Order No. 9 - Oil and Gas Pipelines, established by the
U.S. Department of the Interior, Geological Survey, Conser-
vation Division. Gulf of Mexico Area, October 30, 1970.
• U.S. Recommendations - Liquid Lines
3. API - U.S. Recommendations LIQUID LINES
API RP 1111, Recommended Practice for Design, Construction
Operation and Maintenance of Offshore Hydrocarbon Pipelines,
1st Edition, March 1976, prepared by the Transportation De-
partment of the American Petroleum Institute.
• Foreign Regulations - Liquid Lines
4. DnV - Foreign Regulations LIQUID LINES
DnV - Rules for the Design, Construction and Inspection of
Submarine Pipelines and Pipeline Risers, January 29, 1976,
prepared by Det Norske Veritas, Norway.
• Foreign Recommendations - Liquid Lines
5. IP - Foreign Recommendations LIQUID LINES
IP - "Supplement to IP Pipelines Code, Submarine Pipelines,"
Petroleum Pipelines Safety Code, published in 1974 with
Supplements, prepared by the Institute of Petroleum, Great
Britain.
• Well-Maintained and Inspected Pipelines - Liquid Lines
6. Practices by pipeline company A.
Gas Lines
• U.S. Regulations - Gas Lines
7. DOT CFR 192. - U.S. Regulation GAS LINES
49 CFR 192 - Part 192, Title 49, Code of Federal Regulations,
Minimum Federal Safety Standards for Gas Lines, appearing in
35 F.R. 13257, August 19, 1970, with amendments up to and
including Amdt. 192-27A, 41 F.R. 47252, October 28, 1976.
• DOT CFR 191. - U.S. Regulation GAS LINES
49 CFR 191 - Part 191, Title 49, Code of Federal Regulations,
Leak Reporting Requirements for Gas Lines, appearing in
35 F.R. 320, January 8, 1970. (Also referred to as DOT regu-
lation).
A list of legislation and regulations, standards technical require-
ments, codes of practice for design construction and operation of oil pipe-
lines in Western Europe is included in Appendix A-2.
198
-------
3.4.2.1 Visual and Visual Aided Line Observations for Leakage--
Liquid Lines
1. DOT (U.S. Regulation)
CFR 195.412 "(a) Each carrier shall, at intervals not exceeding
2 weeks inspect the surface condition on or
adjacent to each pipeline right-of-way.
(b) Except for offshore pipelines, each carrier
shall at intervals not exceeding five years,
inspect each crossing under a navigable water-
way to determine the condition of the crossing."
2. PCS #9 (PCS Order)
l.E "All pipelines should be maintained in good operating
condition and inspected monthly for indications of
leakage using aircraft, floating equipment on other,
methods."
3. API
701.5 "Each pipeline operator should maintain a periodic
pipeline patrol program to observe conditions along
the pipeline route affecting the pipeline's safe
operations."
4. DnV
9.3.1.1 "The frequency and extent of surveys shall depend upon
factors such as
- type of survey
- transportation of product
- pipeline route (traffic density, depth of burial)
- operation of pipeline systems
- conditions of pipeline/riser as installed
- degree of pipeline/riser exposure to potential
damage and/or deterioration caused by corrosion,
wear, etc.
9.3.1.2 A periodic survey is normally to be carried out annually
if not otherwise agreed upon in accordance with 9.3.1.1
above.
9.3.1.3 The time for annual surveys may, under normal circum-
stances, be selected with due regard to weather condi-
tions and operation of the pipeline (Frequency 12 ± 3
months).
9.3.1.4 The society may, upon request accept a continuous sur-
vey in lieu of regular periodical surveys. Each part
of the system is then to be controlled as frequently
as in the case of regular periodical surveys.
199
-------
9.3.1.5 The owner is to notify the society on occasions when
such parts of the pipeline system, which are not
normally assessible for survey, may be examined.
5. IP.
11.9 "It is recommended that long submarine pipelines be
inspected regularly by aerial patrol."
6. Pipeline Company A
Ground patrolling is intended to prevent any deteriora-
tions in the line resulting from:
• vegetation, vultivation and construction inside
the right-of-way.
• erosion phenomena (landslides, gullies formed by
running water, etc.)
• the carrying out of various forms of work imme-
diately near the right-of-way (drainage, irriga-
tion ditches, pipe or cable laying, road building.
• this patrolling is carried out by hired staff; each
line walker covers 12.4 miles which he inspects
twice a month and keep in touch with various people
who are likely to be aware of anomalies or incidents.
Ground patrolling is supplemented by special air patrol
which is carried out about twice a month with low-flying
aircraft.
Gas Lines
7. DOT (U.S. Regulation)
CFR 192.705(a) Each operator shall have a patrol program to
observe surface conditions on and adjacent to
the transmission line right-of-way for indica-
tions of leaks, construction activity, and
other factors affecting safety and operation.
(b) The frequency of patrols is determined by the
size of the line, the operating pressures, the
class location, terrain, weather, and other
relevant factors, but intervals between patrols
may not be longer than prescribed in the follow-
ing table:
200
-------
Maximum interval between patrols
At highway and At all other
Class location of line railroad crossings places
1,2... . ....0.6 months 1 year
3 „ 3 months 6 months
4 .do 3 months
(Note: Offshore is Class 1)
49 CFR 192.706 "(a) Each operator of a transmission line shall
provide for periodic leakage surveys of the
line in its operating and maintenance plan.
(b) Leakage surveys of a transmission line must
be conducted at intervals not exceeding 1
year. However, in the case of a transmission
line which transports gas in conformity with
§ 192.625 without an odor or odorant, leakage
surveys using leak detector equipment must be
conducted—
(1) In Class 3 locations, at intervals not
exceeding 6 months; and
(2) In Class 4 locations, at intervals not
exceeding 3 months."
U.S. regulations of line observations at least every two weeks are
justified. However, more frequent or more effective line observations de-
pending upon factors such as volume of product transported, age of pipe-
line might be justified for some lines. For example, in South Europe, air
patrolling is used, and a staff is employed to walk the lines.
5.4.2.2 Oil Spill Detectors on or Near the Water—
Liquid Lines
1. DOT (U.S. Regulation)
No Regulation
2. PCS
No OCS Order
3. API
Not Discussed
4. DnV
Not Discussed
5. IP.
Not Discussed
6. Pipelines Company A
Not Discussed.
201
-------
Gas Lines
NOT APPLICABLE
5.4.2.3 Leak Detection by Continuous or Periodic Monitoring of Internal
Fluid Flow Variations during Product Transfer—
Liguid Lines
1. DOT
No Regulation
2. PCS
No OCS Order
3. API
Not Discussed
4. DnV
DnV 9.2.2.3 "It may be required that the product flow at de-
livering and receiving stations along the pipe-
line is measured continuously or at regular in-
tervals for the purpose of leakage detection."
5. IP
Not Discussed
6. Pipeline Company A
• Any substantial leaks are detected by volume com-
parators in permanent operation. Communication
equipment provides a permanent comparison of the
oil flow rates recorded at relief stations and at
delivery terminals. If the difference in volumes
measured upstream and downstream from a given
point exceeds a certain threshold, a visual and
acoustic alarm is triggered.
Gas Lines
7. DOT (U.S. Regulation)
No Regulation
5.4.2.4 Leak Detection and Location of Leaked Oil on or at a Short Dis-
tance from Line—
1. DOT (U.S. Regulation)
No Regulation
2. PCS
No QCS Order
202
-------
3. API
Not Discussed
4. DnV
5. IP
6. Pipeline Company A
Metal rods, attached to the pipelines by brazing
at approximate intervals of 500 m (1,640.5 ft)
permit the localization of leaks monitoring of
the oil flow carried out systematically by means
of appropriate equipment.
5.4.2.5 Leak Detection by Periodic Pressure Testing of Existing Pipeline
Systems—
Liquid Lines
1. DOT (U.S. Regulations)
No regulations of in-service oil pipelines.
There are regulations (see CFR 195.3) for newly
constructed replaced or otherwise changed.
2. PCS
No OCS Order
3. API
Not Discussed
DnV
See Section 5.4.1.5
IP
4.
5.
IP 9.2
"When the pipeline is not in operation, it is re-
commended that it be shut down under pressure,
except on occasions when the shut-down has been
arranged to allow maintenance on the line and a
careful record made of the pressure during the
shut-down period.
If the pipeline crosses areas where there are
particular dangers of water pollution being
caused by any leakage, then a static pressure
test as described above should be arranged over
a period of 24 hours once per year."
203
-------
6. Pipeline Company A
Very small leaks, which have limited immediate
consequences but the effects of which may be
noticed for a substantial period, are detected by:
• differential pressure measurements between
various sections of the pipeline.
This inspection involves the periodic verification
that pipelines are leakproof, by means of hydro-
static tests on each of the sections located be-
tween block valves (located 12.4 to 15.5 miles
apart).
Gas Lines
7. DOT (U.S. Regulation)
49 CFT 192.743 Pressure limiting and regulating stations: test-
ing of relief devices -
"(a) If feasible, pressure relief devices (except
rupture discs) must be tested in place, at
intervals not exceeding 1 year, to determine
that they have enough capacity to limit the
pressure on the facilities to which they are
connected to the desired maximum pressure.
(b) If a test is not feasible, review and calcu-
lation of the required capacity of the re-
lieving device at each station must be made,
at intervals not exceeding one year, and
these required capacities compared.with the
rated or experimentally determined relieving
capacity of the device for the operating con-
ditions under which it works.
(c) If the relieving device is of insufficient
capacity, a new or additional device must be
installed to provide the additional capacity
required."
8. DOT (U.S. Regulation)
No Regulation
204
-------
5.4.2.6 Corrosion Inspection—
5.4.2.6.1 Corrosion Inspection and Maintenance - Atmospheric
Liquid Lines
1. DOT (U.S. Regulation)
49 CFR 195.416"(D) Each carrier shall, at intervals not exceeding
5 years, electrically inspect the bare pipe in
its pipeline system that is not cathodically
protected and must study leak records for that
pipe to determine if additional protection is
needed.
(e) Whenever any buried pipe is exposed for any
reason, the carrier shall examine the pipe
for evidence of external corrosion. If the
carrier finds that there is active corrosion,
that the surface of the pipe is generally
pitted, or that corrosion has caused a leak,
it shall investigate further to determine the
extend of the corrosion.
(f) Any pipe that is found to be generally corroded
so that the remaining wall thickness is less
than the minimum thickness required by the pipe
specification tolerances must either be re-
placed with coated pipe that meets the require-
ments of this part or, if the area is small,
must be repaired. However, the carrier need
not replace generally corroded pipe if the
operating pressure is reduced to be commensur-
ate with the limits on operating pressure
specified in this subpart, based on the actual
remaining wall thickness.
(g) If isolated corrosion pitting is found, the
carrier shall repair or replace the pipe un-
less—
(1) The diameter of the corrosion pits, as
measured at the surface of the pipe, is
less than the nominal wall thickness of
the pipe; and
(2) The remaining wall thickness at the bottom
of the pits is at least 70 percent of the
nominal wall thickness.
(h) Each carrier shall clean, coat with material
suitable for the prevention of atmospheric
corrosion, and, maintain this protection for,
each component in its pipeline system that is
exposed to the atmosphere."
205
-------
2. PCS
No OCS Order
3. API
API 801.3 "Carbon steel pipe, valves and fittings exposed to the
atmosphere should be protected with an external coat-
ing, when such protection is necessary."
4. DnV
Not Discussed
5. IP,
Not Discussed
6. Pipeline Company A
Not Discussed
Gas Lines
7. DOT (U.S. Regulation)
No Regulation
8. DVT (U.S. Regulation)
No Regulation
5.4.2.6.2 Internal Corrosion Control
Liquid Lines
1. DOT (U.S. Regulation)
49 CFR 195.418 "(a) No carrier may transport any commodity that
would corrode the pipe or other components
of its pipeline system, unless it has inves-
tigated the corrosive effect of the commodity
on the system and has taken adequate steps to
mitigate corrosion.
(b) If corrosion inhibitors are used to mitigate
internal corrosion the carrier shall use in-
hibitors in sufficient quantity to protect
the entire part of the system and shall also
use coupons or other monitoring equipment to
determine their effectiveness.
(c) The carrier shall, at intervals not exceeding
6 months, examine coupons or other types of
monitoring equipment to determine the effec-
tiveness of the inhibitors or the extent of
any corrosion.
206
-------
(d) Whenever any pipe is removed from the pipe-
line for any reason, the carrier must in-
spect the internal surface for evidence of
corrosion. If the pipe is generally corroded
such that the remaining wall thickness is
less than the minimum thickness required by
the pipe specification tolerances^ the car-
rier shall investigate adjacent pipe to de-
termine the extent of the corrosion. The
corroded pipe must be replaced with pipe
that meets the requirements of this part or,
based on the actual remaining wall thickness,
the operating pressure must be reduced to be
commensurate with the limits on operating
pressure specified in this subpart."
2. PCS
No OCS Order
3. API
API 803 "NANCE RP-01-75 should be used to determine the need
for design, installation and evaluation of the re-
sults of an internal corrosion mitigation program.
Where necessary, internal corrosion may be mitigated
by one or more of the following: pipeline scraping,
pigging, or sphering at regular intervals, dehydration,
inhibition, bactericides,- oxygen scavengers, and pipe-
line internal coating. The variables and severity of
each case will determine the preventive methods that
should be used. A monitoring program should be es-
tablished to evaluate the results of internal corrosion
mitigation systems."
4. DnV
DnV 6.6.1.1 "Internal protection is to be considered for pipe-
lines and risers which during installation or op-
eration may be subject to corrosion. Information
about the possibilities of internal corrosion is
to be submitted for evaluation. Treatment of the
product to be transported may be utilized as a
means of controlling corrosion, likewise pigging
at regulat intervals and corrosion monitoring."
5.
IP 8.1 "The operation of pipelines depends to a large extent
upon the nature of the fluids being handled, and each
operator should formulate a procedure for safe pipe-
line operation. Particular emphasis should be laid
upon the following:
(e) Prevention of internal corrosion."
207
-------
Pipeline Company A
Experience has shown that in the case of crude oil pipelines,
•internal corrosion is practically non-existant on account of
the products transported. However, as a precaution, "corrosion
detectors" are installed at 62.1 mile intervals and examined
once every year.
Gas Lines
7. DOT (U.S. Regulation)
49 CFR 192.475 "(a) After July 31, 1972, corrosive gas may not
be transported by pipeline, unless the cor-
rosive effect of the gas on the pipeline has
been investigated and steps have been taken
to minimize internal corrosion.
(b) Whenever any pipe is removed from a pipeline
for any reason, the internal surface must be
inspected for evidence of corrosion. If in-
ternal corrosion is found—
(1) The adjacent pipe must be investigated
to determine the extent of internal cor-
rosion;
(2) Replacement must be made to the extent
required by the applicable paragraphs
of § 192.485, § 192.487, or § 192.489;
and
(3) Steps must be taken to minimize the in-
ternal corrosion.
(c) Gas containing more than 0.1 grain of hydro-
gen sulfide per 100 standard cubic feet may
not be stored in pipe-type of bottle-type
holders."
49 CFR 192.77 "If corrosive gas is being transported, coupons
or other suitable means must be used to determine
the effectiveness of the steps taken to minimize
internal corrosion. After July 31, 1972, each
coupon or other means of monitoring internal cor-
rosion must be checked at intervals not exceeding
6 months."
8. DOT (U.S. Regulation)
No Regulation
208
-------
5.4.2.6.3 Cathodic Protection System-External
Liquid Lines
1. DOT (U.S. Regulation)
49 CFR 195.414 "(a) After March 31, 1973, no carrier may operate
a pipeline that has an external surface coat-
ing material, unless that pipeline is cathodi-
cally protected. This paragraph does not ap-
ply to tank farms and buried pumping station
piping.
(b) Each carrier shall electrically inspect each
bare pipeline before April 1, 1975, to deter-
mine any areas in which active corrosion is
taking place. The carrier may not increase
its established operating pressure on a sec-
tion of bare pipeline until the section has
been so electrically inspected. In any
areas where active corrosion is found, the
carrier shall provide cathodic protection.
Section 195.416 (f) and (g) applies to all
corroded pipe that is found.
(c) Each carrier shall electrically inspect all
tank farms and buried pumping station piping
before April 1, 1973, as to the need for
cathodic protection, and cathodic protection
shall be provided where necessary.
(d) Notwithstanding the deadline for compliance
in paragraphs (a), (b), and (c) of this sec-
tion, this section does not apply to offshore
pipelines located between a production facil-
ity and a carrier's trunk!ine reception point
until August 1, 1977."
49 CFR 195.416 External Corrosion Control-
"(a) Each carrier shall, at intervals not exceed-
ing 12 months, conduct tests on each under-
ground facility in its pipeline system that
is under cathodic protection to determine
whether the protection is adequate."
2. PCS
No DCS Order
3. API
API 804 "The cathodic protection system should be maintained in
accordance with the recommendation in NACF RP-06-75.
The effectiveness of the cathocis protection system
should be evaluated at least annually. Voltage and
current output of impressed current systems should be
209
-------
5.
verified and recorded at two-month intervals. Inter-
ference bonds, where failure would jeopardize pipeline
protection, should be checked for proper operation at
two-month intervals."
4. DnV
DnV 9.3.2.2 "Control of the performance of the cathodic pro-
tection system is required. The extent of such
testing may be reduced to scrutinizing of the
Owner's potential measurement reports. If found
necessary, potential measurements are to be con-
ducted at the discretion of the Surveyor and in
accordance with 8.8.3."
IP_
IP 8.1 "The operation of pipelines depends to a large extent
upon the nature of the fluids being handled, and each
operator should formulate a procedure for safe pipe-
line operation. Particular emphasis should be laid
upon the following: (f) Regular checking of the cath-
odic protection installation."
IP 7.5.4 Operation and Maintenance-
"Regular attention to current consumption figures at
each installation is essential. Adequate protection
depends almost entirely on maintaining the safety
level of pipe-to-soil potential along the whole route.
Pipe-to-soil measurements should be taken at least
annually or whenever an abnormal condition is indi-
cated. Highly corrosive areas should receive special
attention. In addition, the pipe-to-soil potential
of other structures crossing the pipeline route should
be checked for any adverse changes, whether or not
they are bonded into the cathodic protection system."
6. Pipeline Company A
The pipelines must be effectively protected against external
corrosion by means of:
• Passive protection provided by appropriate
coatings;
• Active protection provided by a cathodic pro-
tection system.
In order to verify the satisfactory operation of these units and
consequently the effectivensss of the cathodic protection, 250
"potential measuring points" have been installed, especially at
intersections with buried metal pipes and at large crossings.
The line is protected against corrosion by keeping it at a nega-
tive potential with respect to the soil. This negative potential
is maintained by means of a system including supply stations,
which are checked every month, and insulating joints located up
and downstream of the pumping stations, upstream of the delivery
points and the crossings of large rivers.
210
-------
Gas Lines
7. DOT (U.S. Regulation)
49 CFR 192.475 "(a) Each pipeline that is under cathodic protec-
tion must be tested at least once each calen-
dar year, but with intervals not exceeding
15 months, to determine whether the cathodic
protection meeting the requirements of
§ 192.463. However, if tests at those in-
tervals are impractical for separately pro-
tected service lines or short sections of
protected mains, not in excess of 100 feet,
these service lines and mains may be surveyed
on a sampling basis. At least 10 percent of
these protected structures, distributed over
the entire system, must be surveyed each
calendar year, with a different 10 percent
checked each subsequent year, so that the
entire system is tested in each 10-year period.
(b) At intervals not exceeding 2 months, each
cathodic protection rectifier or other im-
pressed current power source must be inspected
to ensure that it is operating.
(c) At intervals not exceeding 2 months, each re-
verse current switch, each diode, and each
interference bond whose failure would jeopar-
dize structure protection, must be electrically
checked for proper performance. Each other in-
terference bond must be checked at least once
each calendar year, but with intervals not ex-
ceeding 15 months.
(d) Each operator shall take prompt remedial ac-
tion to correct any deficiencies indicated by
the monitoring.
(e) After the initial evaluation required by para-
graphs (b) and (c) of §192.455 and paragraph
(b) of §192.457, each operator shall, at in-
tervals not exceeding 3 years, reevaluate its
unprotected pipelines and cathodically protect
them in accordance with this subpart in areas
in which active corrosion is found. The op-
erator shall determine the areas of active
corrosion by electrical survey, or where elec-
trical survey is impractical, by the study of
corrosion and leak history records, by leak
detection survey, or by other means."
8. DOT (U.S. Regulation)
No Regulation
211
-------
5.4.2.6.4 Chemical Monitoring
Liquid Line
1. DOT (U.S. Regulation)
No Regulation
2. PCS
No OCS Order
3. API
Not Discussed
4. DnV
DnV 9.2.2.2 "It may be required that the concentrations of
various chemical or physical components in the
product are measured and recorded at regular
intervals."
5. If.
Not Discussed
6. Pipeline Company A
Not Discussed
Gas Lines
7. DOT (U.S. Regulation)
No Regulation
8. DOT (U.S. Regulation)
No Regulation
5.4.2.7 Standard Non-Destructive Inspections to Check Integrity of Pipe-
line (Flaws, Cracks, Buckling, Dents and Other Potential Failure
Conditions) —
Liquid Lines
1. DOT (U.S. Regulation)
49 CFR 195.418 "(d) Whenever any pipe is removed from the pipe-
line for any reason, the carrier must in-
spect the internal surface for evidence of
corrosion. If the pipe is generally corroded
such that the remaining wall thickness is less
than the minimum thickness required by the
pipe specification tolerances, the carrier
shall investigate adjacent pipe to determine
the extent of the corrosion. The corroded
pipe must be replaced with pipe that meets
the requirements of this part or, based on
212
-------
the actual remaining wall thickness, the op-
erating pressure must be reduced to commen-
surate with the limits on operating pressure
specified in this subpart."
2. DCS
No DCS Order
3. API
Not Discussed
4. DnV
Not Discussed
5. IP.
IP 11.10 "Regular underwater inspection should be carried out
where there is risk of damage to the pipeline or where
scour conditions may occur."
6. Pipeline Company A
Not Discussed
Gas Lines
7. DOT (U.S. Regulation)
Not Discussed
8. DOT (U.S. Regulation)
Not Discussed
5.4.2.8 Inspection Pigs to Check the Integrity of Pipeline (Flaws, Cracks,
Buckling, Dents and Other Potential Failure Conditions)--
Liquid Lines
1. DOT (U.S. Regulation)
No Regulations
2. PCS
No OCS Orders
3. API
Not Discussed
4. DnV
DnV 9.3.2.3 "Inspection by gauging-pig (e.g., caliper-pig is
required for detection of buckles or dents in the
pipeline/riser."
213
-------
DnV 9.3.2.5 "Thickness measurements may be required where
there is reason to believe that the pipe wall
thickness is being reduced due to external/
internal corrosion or internal erosion (e.g.,
the effect of sand content in the transported
gas).
5. IP.
IP 11.10 "Regular underwater inspection should be carried out
where there is risk of damage to the pipeline or where
scour conditions may occur."
6. Pipeline Company A
Not Discussed
Gas Lines
1. DOT (U.S. Regulation)
No Regulations
8. DOT (U.S. Regulation)
No Regulations
5.4.2.9 Survey-Pipeline Systems Charting and Depth of Burial--
Liquid Lines
1. DOT (U.S. Regulation)
49 CFR 195.404 "(a), Each carrier shall maintain current maps and
records of its pipeline systems that include
at least the following information:
(1) Location and identification of all major
facilities.
(2) All crossings of public roads, railroads,
rivers, buried utilities, and foreign
pipelines.
(3) The maximum operating pressure of each
pipeline.
(4) The diameter, grade, type, and nominal
wall thickness of the pipe.
(b) Each carrier shall maintain daily operating
records that indicate the discharge pressures
at each pump station and any unusual opera-
tions of a facility. The carrier shall re-
tain these records for at least 3 years.
(c) Each carrier shall also maintain for the use-
ful life of that part of the pipeline system
to which they relate, records that include
the following:
214
-------
(1) The date, location, and description of
each repair made to its pipeline systems.
(2) A record of each inspection and each test
required by this subpart."
2. DCS
PCS #9, l.A "The operator shall be responsible for the installa-
tion of the following control devices on all oil
and gas pipelines connected to a platform includ-
ing pipelines which are not operated or owned by
the operator... The operator shall submit records
semi-annually showing the present status and past
history of each device, including dates and details
of inspection, testing, repairing, adjustment, and
reinstallation."
PCS #9. l.E "... records of these inspections including the
date, methods and results of each inspection shall
be maintained by the pipeline operator and sub-
mitted annually by April 1 ..."
3. API
API 703 "The following records should be maintained for opera-
tions and maintenance purposes:
} Material and construction specifications.
'b) Route maps and alignment sheets.
(c) Coating and cathodic protection specifications.
(d) Pressure test data.
[e) Non-destructive inspection data.
(f) Necessary operational data.
(g) Pipeline patrol records.
(h) Corrosion mitigation records recommended in 805.1.
(i) Leak and break records and failure investigation
records.
[j) Records of safety equipment inspection.
[k) Records of other inspections, such as external or
internal pipe conditions when line is cut or hot
tapped.
The records recommended in (e), (f), (g) and (j) should
be retained for at least one year. Other recommended
records should be retained for the life of the facility
unless states otherwise in this Recommended Practice."
4. DnV
DnV 9.2.3.2 "Files on Owner's inspection reports are to be
available in connection with surveys required by
the Society."
DnV 9.2.3.3 "The Society's requirements for maintenance and
repair will be based on information obtained from
periodical surveys and special surveys if appli-
cable."
215
-------
5. IP.
IP 11.10 "Regular underwater inspection should be carried out
where there is risk of damage to the pipeline or where
scour conditions may occur."
6. Pipeline Company A
Not Discussed
Gas Lines
7. DOT (U.S. Regulation)
49 CFR 192.709 "Each operator shall keep records covering each
leak discovered, repair made, transmission line
break, leakage survey, line patrol, and inspec-
tion, for as long as the segment of transmission
line involved remains in service."
8. DOT (U.S. Regulation)
No Regulation
5.4.2.10 Miscellaneous-Leak Detection and Prevention by Equipment Inspec-
tion—
Liquid Lines
1. DOT (U.S. Regulation)
No regulations for oil lines. There are line regulations for
liquefied gases as follows:
DOT CFR 195.428"(a) Except as provided in paragraph (b) of this
section, each carrier shall, at intervals
not exceeding 12 months, or 6 months in the
case of pipelines used to carry liquefied
gages, inspect and test each pressure limit-
ing device, relief valve, pressure regulator,
or other item of pressure control equipment
to determine that it is functioning properly,
is in good mechanical condition, and is ade-
quate from the standpoint of capacity and
reliability of operation for the service in
which it is used.
(b) In the case of relief valves on pressure stor-
age vessels containing liquefied gas, each
carrier shall test each valve at intervals
not exceeding 5 years."
2. PCS
No OCS Order
3. API
Comments on safety equipment on nonproduction platforms are
presented in 701.6.
216
-------
4. DnV
Not Discussed
5. IP
Not Discussed
6. Pipeline Company A
Block valves are housed inside a sealed casing in order to avoid
any contamination of the surrounding area.
A center is provided with data concerning the hydraulic condi-
tion of the pipelines and the position of each of the block-
valves, which can be shut by remote control for safety"reasons.
Gas Lines
7. DOT (U.S. Regulation)
49 CFR 192.706 "(a) Each operator of a transmission line shall
provide for periodic leakage surveys of the
line in its operating and maintenance plan.
(b) Leakage surveys of a transmission line must
be conducted at intervals not exceeding 1
year. However, in the case of a transmission
line which transports gas in conformity with
§192.625 without an odor or odorant, leakage
surveys using leak detector equipment must
be conducted:
(1) In Class 3 locations, at intervals not
exceeding 6 months; and
(2) In Class 4 locations, at intervals not
exceeding 3 months.
5.4.2.11 Miscellaneous-Safety Equipment and Pressure Limiting and Reliev-
ing Devices—
Liquid Lines
1. DOT (U.S. Regulation)
49 CFR 195.426 Scraper and Sphere Facilities-
"No carrier shall use a launcher or receiver
that is not equipped with a relief device capa-
ble of safely relieving pressure in the barrel
before insertion or removal of scrapers or
spheres. The carrier must use a suitable device
to indicate that pressure has been relieved in
the barrel or must provide a means to prevent
insertion or removal of scrapers or spheres if
pressure has not been relieved in the barrel."
217
-------
49 CFR 195.428 Overpressure Safety Devices-
"(a) Except as provided in paragraph (b) of
this section, each carrier shall, at
intervals not exceeding 12 months, or
6 months in the case of pipelines used
to carry liquefied gases, inspect and
test each pressure limiting device, re-
lief valve, pressure regulator, or other
item of pressure control equipment to de-
termine that it is functioning properly,
is in good mechanical condition, and is
adequate from the standpoint of capacity
and reliability of operation for the ser-
vice in which it is used.
(b) In the case of relief valves on pressure
storage vessels containing liquefied gas,
each carrier shall test each valve at in-
tervals not exceeding 5 years."
2. PCS
No OCS Order
3. API
API 701.6 Safety Equipment on Non-Production Platforms-
"Pressure limiting devices, relief valves, automatic shutdown
valves, and other safety devices, except check valves should be
subjected to periodic inspections at a maximum interval of six
months. The inspection should verify that the device is in good
mechanical condition and properly performs the safety function
for which it was installed.
4. DnV
DnV 9.3.2.6 "Pressure limiting devices, relief valves, auto-
matic shutdown valves and other safety devices
should be tested and inspected. The inspection
should verify that the device is in good mechan-
ical condition and properly performs the safety
function for which it was installed."
5. IP
IP 8.1(a) "Safety devices for protecting the pipeline from pres-
sures in excess of those for which it was designed."
IP 8.l(b) "Instruments to give warning and shut down pumps in
case of damage caused to the pipeline by Act of God
or third party activities.
6. Pipel.ine Company A
Each valve is housed inside a sealed casing in order to avoid
any contamination of the surrounding area.
218
-------
Gas Lines
7. DOT (U.S. Regulation)
49 CFR 192.731 Compressor Stations: Inspection and Testing of
Relief Devices:
(a) Except for rupture discs, each pressure re-
lieving device in a compressor station must
be inspected and tested in accordance with
§§192.739 and 192.743, and must be operated
periodically to determine that it opens at
the correct set pressure.
(b) Any defective or inadequate equipment found
must be promptly repaired or replaced.
(c) Each remote control shutdown device must be
inspected and tested, at intervals not to
exceed 1 year, to determine that it functions
properly."
49 CFR 192.737 Pipe-Type and Bottle-Type Holders: Plan for In-
spection and Testing-
"(c) The pressure control and pressure limiting
equipment must be inspected and tested
periodically to determine that it is in a
safe operating condition and has adequate
capacity."
49 CFR 192.739 Pressure Limiting and Regulating Stations: In-
spection and Testing-
"Each pressure limiting station, relief device
(except rupture discs), and pressure regulating
station and its equipment must be subjected, at
intervals not exceeding 1 year, to inspections
and tests to determine that it is:
a) In good mechanical condition;
b) Adequate from the standpoint of capacity
and reliability of operating for the ser-
vice in which it is employed;
(c) Set to function at the correct pressure;
and
(d) Properly installed and protected from dirt,
liquids, or other conditions that might pre-
vent proper operation."
49 CFR 192.743 Pressure Limiting and Regulating Stations: Test-
ing or Relief Devices-
"(a) If feasible, pressure relief devices (except
rupture discs) must be tested in place, at
intervals not exceeding 1 year, to determine
219
-------
that they have enough capacity to limit the
pressure on the facilities to which they
are connected to the desired maximum pres-
sure.
(b) If a test is not feasible, review and calcu-
lation of the required capacity of the reliev-
ing device at each station must be made, at
intervals not exceeding one year, and these
required capacities compared with the rated
or experimentally determined relieving capa-
city of the device for the operating condi-
tions under which it works.
(c) If the relieving device is of insufficient
capacity, a new or additional device must
be installed to provide the additional capa-
city required."
8. DOT (U.S. Regulation)
No Regulation
5.4.2.12 Pipe Maintenance
1. DOT (U.S. Regulation)
49 CFR 195.402 General Requirements-
"(b) No carrier may operate or maintain its pipe-
line systems at a level of safety lower than
required by this subpart and the procedures
it is required to establish under paragraph
(a) of this section.
(c) Whenever a carrier discovers any condition
that could adversely affect the safe opera-
tion of its pipeline system it shall correct
it within a reasonable time. However, if
the condition is of such a nature that it
presents an immediate hazard to persons or
property, the carrier may not operate the
affected part of the system until it has
corrected the unsafe condition."
49 CFR 195.422 Pipeline Repairs-
"(a) Each carrier shall, in repairing its pipeline
systems, insure that the repairs are made in
a safe manner and are made so as to prevent
damage to persons or property.
(b) No carrier may use any pipe, valve, or fitting,
for replacement in repairing pipeline facili-
ties, unless it is designed and constructed as
required by this part."
220
-------
2. PCS
No OCS 'Order
3. API
API 701.1 "Written starting, operating and shutdown procedures
for pipeline facilities should be established and the
operating company should take appropriate steps to
see that these procedures are followed. These pro-
cedures should outline preventive measures and system
checks as required to provide for the proper function-
ing of safety, control and alarm equipment."
4. DnV
DnV 9.1.2.1 "If it is found that the pipeline/riser in some
way does not meet the Rules, the Society will re-
quire improvements, new surveys, or other mea-
sures found necessary in order to retain the
Certificate of Approval, regardless of previous
approvals."
DnV 9.2.3.1 "The Owner should provide such running inspection
of the pipeline system as to initiate maintenance
work necessary to retain the built-in safety."
DnV 9.3.2.1 "Survey of exposed parts of the pipeline, i.e.,
not buried parts, is required to verify that no
unacceptable damages have occured to the pipe,
the corrosion protection system, or the weight
coating (anchoring system)."
5. IP.
IP 9.3 "It is not always necessary to empty and gasfree a
pipeline or to use water plugs in order to carry out
maintenance or repair work which involves cutting and
re-welding the line. These operations may be safely
performed provided that the right conditions have been
achieved."
6. Pipeline Company A
Not Discussed
Gas Lines
7. DOT (U.S. Regulation)
49 CFR 192.631 "(a) Each operator shall have a procedure or con-
tinuing surveillance of its facilities to
determine and take appropriate action con-
cerning changes in class location, failures,
leakage history, corrosion, substantial
changes in cathodic protection requirements,
and other unusual operating and maintenance
conditions.
221
-------
(b) If a segment of pipeline is determined to
be in unsatisfactory condition but no imme-
diate hazard exists, the operator shall ini-
tiate a program to recondition or phase out
the segment involved, or, of the segment can-
not be reconditioned or phased out, reduce
the maximum allowable operating pressure in
accordance with § 192.619 (a) and (b)."
8. DOT (U.S. Regulation)
Not Discussed
222
-------
SECTION 6
ANALYSIS OF THE RISK OF ACCIDENTAL OIL SPILLS
FROM PETROLEUM PIPELINE SYSTEMS
An analysis of the risk of accidental oil spills from petroleum pipe-
line systems is carried out in this section. Both the risk that oil can
spill from a system and environmental problems created by spillage are
examined. First, the seriousness of oil spills is discussed in Section 6.1.
Then the difficulties in assessing the risk of oil spills are reviewed in
Section 6.2. Approaches accounting for these difficulties are presented.
Finally, an analysis and assessment of the risk of oil spills was carried
out in Section 6.3.
Since the risk of oil spills and the potential for risk reduction
were both found to be much greater for line pipe than for any other pipe-
line system component, an in-depth analysis of the risk of oil spills from
line pipe was carried out in Section 6.3. An analysis was performed on a
reference line (typical line), and correction factors for variations from
the reference line were developed so that the spill risk (potential) could
be established for most lines. Simplified tables and figures were provided
to enable an operator to estimate the spill potential of his own line.
6.1 SERIOUSNESS OF OIL SPILLS
Petroleum pipeline systems and line pipe, in particular, present a
continuous potential for serious accidents and spills. Spills are ex-
pected to increase as the U.S. liquid pipeline system ages3. These are
the obvious conclusions when data from accidental spills (Section 4.5.2)
and individual accident reports are examined and oil spill risks evaluated
(Appendix D). This situation exists even though petroleum pipeline trans-
portation systems produce relatively few reported spills for the large
quantities of oil transported.
Accidental spills are a serious matter and are detrimental to both
the national interest and the oil and gas industries. The major problems
created by these spills are:
• Subject segments of the population to potential catastrophies.
• Cause significant environmental damage.
• Cause losses of large quantities of petroleum.
aAverage age of the U.S. liquid pipeline increases yearly.
223
-------
Responsibility for the spill problem rests with both the government and
operating companies, who must use the best means available to minimize the
frequency and volume of spills and the resultant spill damage.
Various sources contribute to serious accidents in the petroleum pipe-
line system. Some of the major ones include:
• Properties and characteristics of the petroleum itself create
serious hazards (even losses of small quantities, i.e., few
barrels, can cause serious damage in some situations).
• Large volumes of petroleum transported through the system
potentially can result in large volumes of petroleum spilled.
• Inadequacies and/or limitations in the spill prevention and
detection measures.
• Variety of causes and types of pipeline system failures.
The potential for serious accidents is magnified when pipeline systems
are located in or near high-risk areas. Some of these are:
• High population density
• Heavy industrial areas
• Commercial areas
• Underground water supplies
• Underground facilities
• Offshore.
Although petroleum pipeline systems are relatively safe, serious spills
can and do occur for a variety of reasons. For example, large volumes of
petroleum are transported using lines that are not under continuous surveil-
lance. This can result in large volumes of hazardous fluids (thousands of
barrels) being spilled before the spill is detected and the line shut down.
Since explosions from crude and petroleum lines have occurred hundreds of
feet from a break58, the possibility exists, primarily in populated areas
(urban, commercial, industrial), for exposure of large numbers of people
and large amounts of property to potential hazards. The serious problems
that exist can be emphasized by examining a few case histories. In the
city of Los Angeles, for example, a network exists of approximately 200
miles of pipelines owned by 32 oil companies. One line which runs beneath
a school ruptured and exploded in 1976 and eight persons were killed72. In
another case, a crude leak caused an explosion near a heavily traveled free-
way. Fortunately, this crude line did not cause leaks in the gas and pro-
duct transmission lines located a short distance away. Other serious prob-
lems such as damage to water supplies are also possible. A recent study1*6
indicates that serious spill hazards to undergound water supplies exist and
identifies numerous cases of damage to underground water sources. The hun-
dreds of serious spills that have occurred within the last ten years con-
clusively illustrates the potential for serious spills. •
224
-------
These types of serious spills are often considered as highly improba-
ble, but serious accidents often do not follow statistical probabilities.
The nuclear powerplant accident in Pennsylvania in early 1979 and the ex-
plosion of the almost empty crude oil tanker in -the Los Angeles Harbor in
1978 are dramatic examples of highly improbable accidents. In the ship
explosion, a number of fatalities occurred and windows were broken by the
blast pressure wave at distances up to ten miles. In the nuclear reactor
accident, segments of the community were evacuated as a precautionary mea-
sure.
Comparisons are often made, such as in Figure 48, of the relative
safety of pipeline systems to other modes of transportation. Such compari-
sons are often misleading. For example, the significance of comparing
auto fatalities with pipeline fatalities is questionable. At least 95
percent of the U.S. population routinely uses automobiles, and thus the
expected fatalities are high. However, only a small percentage of the
population use pipeline facilities or are in close proximity to lines.
A further examination of the auto fatalities actually reveals that gasoline
is the cargo most frequently responsible for highway deaths (about 10 per-
cent). This latter information acutually demonstrates the hazard potential
of petroleum spills. If one were to use other comparisons such as the po-
tential of a major catastrophe from a single accident, the results would
be less favorable for pipeline systems.
6.2 PROBLEMS IN ASSESSING THE OIL SPILL RISK
It is difficult to assess quantitatively the actual oil spill risk
from U.S. petroleum pipeline systems. There are a number of reasons for
this. The main reasons are:
• Oil spills can result in a variety of risks that must
properly be identified and categorized.
• No industry standards exist for comparing and evaluating
spill risks.
• Limited information is available on oil spill statistics.
• Limited information on the variations of the complex pipe-
line systems.
• Lack of suitable analysis scheme for variations of pipe-
line systems.
These major difficulties can be accounted for. Each will be discussed
separately in the subsections that follow.
6.2.1 Identification and Categorization
Oil spills can result in a variety of significant risks such as the
substantial loss of petroleum and serious problems external to the line.
These risks must be properly identified and categorized to insure suitable
analysis of the problem so that needed spill prevention and control can be
identified. To satisfy these objectives, risks have been separated into
the following two categories:
225
-------
MARINE
GENERAL AVIATION
1,324
AIR CARRIERS
124
GRADE CROSSING
910
PEDALCYCIES
(Bieyeias)
900
Numfacrj are preliminary
estimates.
Figure 48. Transportation accidents in 1975, National Transportation
Safety Board 1975.
226
-------
• Risks that oil will escape from the pipeline system by
accidental spillage.
• Risks external to the pipeline system after petroleum es-
capes from the line.
The first risk category deals primarily with the actual loss of petro-
leum from the line. The risk, in general, is independent of line location.
The specific risks of spill incidents and spill volume can be determined
from nationwide oil spillage statistics that are based on most lines. Also,
the spill reduction capabilities of the available inspection and leak de-
tection methods can be established for the line itself. These spill re-
duction capabilities are generally not dependent on line location.
The second category deals with additional risks, such as property
damage or injuries that exist external to the pipeline system. These
risks, in general, are dependent on line location. Hence, this category
enables one to identify the various additional risks that may be charac-
teristic of a particular location. High-risk areas, for example, can be
identified and the extent of risk determined; the need for more effective
spill prevention methods can be established and the increases in cost ef-
fectiveness of the various options can be determined.
6,2.2 Comparison and Evaluation
There are no industry standards for comparing and evaluating the risk
that oil will escape from the pipeline system and risks external to the
line by accidental spillage. However, a variety of suitable means do
exist for assessing this risk.
6.2.2.1 Discussion of Risk of Oil Escaping from Pipelines—
We have selected two means that are most meaningful and practical in
this study for comparing and evaluating the risk that oil will escape from
the pipeline system. These are often used in oil, gas, and other industries,
The first means is to determine the risk of oil spills based on two
criteria:
• Frequency or probability that a spill may occur;
• Quantity of oil spilled.
Both of these are important. For example, the severity of spill accidents
based on injuries and property damage correlates almost directly with quan-
tity spilled, Figure 49. The frequency of spills is of significance, par-
ticularly in areas of high risk; in many areas even a small spill can be
extremely hazardous.
The second means, and the most meaningful, is to provide a measure of
the relative risk that accounts for both the frequency and volume of spills.
This is done by assessing the product (barrels per year) of the expected
227
-------
Injuries
Fatalities
.20-1
o>
OJ
«J
o
4)
ft
01
-------
frequency and volume of significant spills (typically greater than 50 bar-
rels). Although both the frequency and spill size are important in the
evaluation of spills, the product of both parts is typically the most useful
in comparing the spill risks of pipeline systems.
In general, a large number of small spills over a period of time is
normally expected to result in damage equivalent to one large spill81.
Based on fatalities alone, however, this equivalence does not apply. Large
spills (in excess of 500 barrels) are responsible for most of the fatali-
ties. This is apparent from spill statistics presented in Figure 49.
Evaluation of the measure of relative risk based solely on the mean
spill size is useful but somewhat limited. Spills between pumping stations,
for example, are expected to occur either as"leaks (small to medium spill
size) or ruptures (major spill size). Hence, an evaluation of the measure
of relative risk based on all three spill sizes (mean, leak, rupture) is
essential for a realistic evaluation of spills. It is particularly useful
in the development of a practical spill prevention program.
6.2.2.2 Discussion of Risks External to Pipeline System-
Risks external to a pipeline system are also assessed. No industry
standards exist for comparing and evaluating the risk of oil spillage after
escaping from the pipeline system. Furthermore, an in-depth assessment of
this risk is beyond the scope of this study. However, in this study, risks
are assessed at a level sufficient for the selection of methods for a spill
prevention program. Hence, sources of risks are identified and risk factors
assigned (see Section 6.3.2.5). It should be noted that an assessment of
the risk external to the pipeline system is extremely important in the
selection of optimum methods for a spill prevention program for a particular
line.
6.2.3 Applicable Spill Statistics
Available U.S. spill statistics are not ideal, and in many instances
are severely deficient regarding the development of spill risks of various
pipeline systems. Various government agencies have undertaken to collect
and compile spill statistics in order to understand and control the spill
problem. However, this information is not oriented towards analyzing and
reducing oil spill risks. For example, numerous spills and spill sizes
are reportable, but other vital information such as the hole size in the
pipe, flow rate, and shutdown time is not reportable. Also, oil spills in
certain areas and from certain gathering lines are not reportable. These
kinds of limitations and exclusions in spills reportable to U.S. agencies,
combined with deficiencies and noncompliance in spill reporting by operat-
ing companies (see Section 4.5.2), present limits and constraints on oil
spill risk analysis.
Despite these limitations, analysis of available spill statistics in-
dicates that sufficient data are available for developing guidelines for a
spill prevention program. Risks, for example, can be established for the
major and significant factors affecting oil spillage.
229
-------
6.2.4 Applicable Information on the Variations of the Complex Pipeline
System
The substantial amount of information that exists on U.S. petroleum
transportation is not oriented to the details of the various complex pipe-
line systems. Instead, information deals primarily with demand and supply-
quantities imported, produced, transported, etc.
Only a limited amount of information exists that is suitable for com-
paring individual lines to the U.S. total. Information is available on an
individual basis from each company, but there is no industry tool for many
of the variations between lines. For example, information is not available
on total mileage and age of line pipe for variations such as:
• Depth of burial
• Soil conditions
• Construction
• Materials.
This constraint somewhat restricts the analysis. However, sufficient in-
formation is available or can be estimated for the significant variations
that are essential for development of guidelines for spill prevention.
This is done in Section 6.3 for variations (e.g., line pipe age) of spe-
cific pipeline systems and the U.S. total.
6.2.5 Analysis Scheme for the Variations of Pipeline Systems
Risk assessment of the pipeline systems in the U.S. today is compli-
cated by the significant variations between systems. Although these vari-
ations can be estimated, they must be accounted for when assessing the
risks of oil spills of individual lines or the U.S. total. They also must
be accounted for so that an operator can assess the spill potential of his
particular line. A number of variations are important such as:
• Age
• Geometry
• Usage
• System shutdown time
• Throughput
• Other.
In order to handle this problem, the following scheme is used in this
study. Risks of spills are first established for a typical section of line
called a reference line. Then correction factors are established to account
for significant differences between the reference pipeline system and most
other lines. With this information, the spill risks for most lines can be
assessed.
230
-------
The ability to assess the risks of individual lines is particularly
useful for new lines, those in high risk areas, or where a spill problem
exists. An assessment capability enables one to determine the need and
the extent of a spill prevention program. It also aids in evaluating the
practicality and cost-effectiveness of such programs and the locations
where implementation may provide the greatest benefits.
6.3 ANALYSIS AND ASSESSMENT OF THE RISK OF ACCIDENTAL SPILLS
This section presents a detailed analysis and assessment of accidental
spills. The risk of oil accidentally spilling from a system and the re-
sultant environmental hazards are examined.
The major components of the petroleum pipeline transportation system
are first studied in Section 6.3.1. System component(s) with the highest
incidence of failure are examined. The risk of spills from line pipe is
found to be much greater than from any other petroleum pipeline system
components. Correspondingly, the potential for reducing line pipe spills
was also found to be much greater. Based on these results, oil spill
risks for line pipe were analyzed further.
Risks based on the frequency and volume of spills and their product
(relative measure of risk) are examined in Section 6.3.2. These risks
are estimated quantitatively for a reference line (a typical section of
line). Qualitative estimates are made for the risk of spill damage ex-
ternal to the reference line.
Correction factors that account for significant variations between
line pipes are considered in Section 6.3.3. Values are determined for
both line variations and the environment external to the line.
By applying these corrections to the reference line, the spill poten-
tial for most lines can be estimated. This is done is Section 6.3.4 for
the combination of all U.S. lines and a typical pipeline system. Simpli-
fied tables and figures are presented to enable a line operator to esti-
mate the spill potential of his own line.
The information from this section is used to estimate oil spillage
and cost effectiveness in Sections 6.5 and 6.6. It is also used in Sec-
tion 7 for the development of guidelines for a spill prevention program.
6.3.1 System Component(s) with the Highest Potential for Frequent and
Serious Spill Incidents
The petroleum pipeline transportation system model (see Sections 4.1.3
and 4.2.2) has been divided into the following four major components:
• Production
• Pumping stations
• Line pipe
• Storage facilities.
231
-------
These components account for all accidental spill incidents. Each major
component has been subdivided into a number of main system components
(see Figure 4 and Table 7). Each is considered in the evaluation of com-
ponents with the highest potential for frequent and serious spill inci-
dents.
Spills in petroleum pipeline systems have occurred at most components.
This is evidenced, for example, by the spill summary presented in Table 33.
However, careful review of the oil spill statistics (such as those pre-
sented in Section 4.5.2.1.2) clearly show that most spillage (over 80 per-
cent of the incidents and volume of spills) occurs in line pipe. Of the
remaining spills, only two components—tank farms and pumping stations--
account for more than two percent of the spill incidents and volume. Tank
farm spills typically account for about 13 percent of the spill volume and
and only eight percent of the incidents. Most storage tanks are surrounded
by dikes to contain accidental spills from tanks. This has significantly
reduced petroleum leakage outside the facility and loss of oil (oil can be
recovered). The incidence of spills from pumping stations is about eight
percent, while the volume spilled is only about three percent of the total.
This small spill volume is expected because of the constant vigilance of the
operating personnel. This enables almost immediate shutdown of the line or
other mitigating measures should a spill occur.
Further reduction of the spill risk of system components other than
line pipe appears difficult to attain. Existing U.S. regulations and op-
erating company practices insure effective and frequent inspections of pump
stations and tank storage facilities.
Once a line becomes operational, only limited and generally ineffective
line pipe inspection and leak detection methods are required by U.S. regula-
tions. These are primarily biweekly visual inspections and inspection and
maintenance of cathodic protection systems. Additionally, practices by
individual operating companies vary greatly, but methods other than pressure
monitoring at pump stations are not generally used. Fortunately, a signifi-
cant potential for spill reduction does exist because of available leak de-
tection and inspection methods. For example, some companies are using
existing continuous monitoring methods (pressure, flow, volume comparison)
or developing and implementing advanced methods. These methods are gener-
ally useful for identifying large sized breaks or ruptures. Additionally,
inspection of line pipe for internal damage or defects is often done and
offers great potential for reducing the frequency of spills.
Analysis indicates that implementation of a scheduled maintenance pro-
gram for line pipe offers the potential for significantly reducing the risk
of oil spills. Thus, the main objective and effort in this study is esti-
mating the spill risks and reduction of risks from line pipe by implementing
scheduled inspections and/or leak detection.
232
-------
6.3.2 Oil Spill Risks from a Typical Section of Line Pipe (Reference
Line)
Risk values for spills from a typical section of line pipe are estab-
lished in this section. The following approach is used:
• Typical section of line is defined (assumptions for this refer-
ence line are shown in Table 56).
• Failure modes are identified (failure modes are grouped into
three main causes of failure: line pipe faults, outside
forces and other causes. The advantages of grouping failure
modes by cause will be evident in the analysis carried out in
later sections that deal with spill prevention. Data for
failure analysis are based on the work carried out in Section
4.5.2.1.3).
• Fault tree11* (see Section 4.2) is constructed based on fail-
ure data'(For convenience, symbols used in this fault tree
are shown and defined in Figure 50).
• Fault tree is evaluated based on the values of the frequency
and mean value of accidental spilTTI
• Measure of relative risk is evaluated (based on the product
of spill frequency and volume for leaks, rupture and average
spills).
• Potential severity of the spill or the risk external to the
escaping oil is evaluated^
6.3.2.1 Risks—Frequency of Spills FST—
The expected frequency of spills F$y is the summation of the frequency
of spills for each of the individual causes of failures. Individual fail-
ures which could result in spill incidents from the typical section of line
pipe are identified in the fault tree9 of Figure 51.
The three main causes of failure are identified in Figure 51. They
are:
Line pipe faults (56 percent spills)
- Defective pipe (17 percent spills)
- Corrosion (39 percent spills)
Outside forces (41 percent spills)
- Impacts (34 percent spills)
- Non-Impacts (7 percent spills)
Other (3 percent spills).
Percentages indicated were obtained from Table 19.
233
-------
TABLE 56. ASSUMPTIONS FOR TYPICAL SECTION
OF LINE PIPE (REFERENCE LINE)
Age:
Oi ameter:
Length:
Commodity:
Flow velocity:
Flow rate:
Operating pressure:
Material:
Construction:
Line elevation:
Corrosion control:
Pump station
shutdown time:
Mainline valve
closure time:
External environment:
25 years
10 inches
1 mile
Crude oil
7 feet per second
2500 barrels per hour
1000 psi
Steel pipe
Butt weld joints
Buried underground at 3 feet depth
Horizontal
Coated with a cathodic protection system
2 minutes
72 minutes
Low risk
- on!and
- sparsely populated
- not near water
234
-------
The rectangle identifies an event
that results from the combination
of fault events through the input
logic gate.
The diamond describes a fault event
that is considered basic in a given
fault tree. The possible causes of
the event are of insufficient conse-
quence or the necessary information
is unavailable.
The circle describes a basic
fault event that requires no
further development. Frequency
and mode of failure of items so
identified are derived from
empirical data.
i t i
AND gate describes the logical
operation whereby the coexistence
of all Input events is required
to produce the output event.
The triangles are used as transfer
symbols. A line from the apex of
the triangle indicates a transfer
in and a line from the side denotes
a transfer out.
OR gate defines the situation
whereby the output event will
exist if one or more of the
input events exists.
Figure 50. Symbolism for fault trees.
235
-------
ro
CO
HEM SHU VOLUHC
A"
1.1 IMKIVUA*
SPILLS f ROH TYPICAL
LINE SECTION
(1 Nue LENGTH)
STILL FWOWIICV
A ,
/ \ '«
1.1 > 10 ' SPIUSWM
IBKHIAI
cowtosion
UllUttS
()1> SPILLS
US VOL.)
- DEFECTIVE FIFE
IMPACT DAMAGE
O.OS2 I 10° o.OZt I 10° ' 0.013 I 10°
O.Otl
-mi IMPACT DAMAGE-
HUE PIPE FAULTS
OUTSIDE FORCES
0 OH ( 10 "'
0 Oil
-MISCELIAHEOUS—"
I
Figure 51. Fault trees, VST and f^t spill frequency and volume from typical
section of line (reference line).
-------
Line pipe faults such as welds and seam defects and corrosion are respon-
sible for most spills from a typical section of line. These are failures
of the line pipe itself and potentially can be reduced by periodic inspec-
tions. Outside force incidents, primarily due to impacts, are significant
causes of the frequency of spills. These accidents usually involve human
error primarily due to insufficient communication between equipment and
pipeline operators. Such spills potentially can be reduced by a prevention
program that is currently recommended52and by leak detection systems that
detect impacts on lines before significant damage can occur.
The frequency of spills for the typical section of line pipe (10 in- 3
ches in diameter and 25 years old) is shown to be approximately 1.30 x 10
spins/mile-year, Figure 32. This is in reasonable agreement with both the
average (1.34 x 10-3 spins/mile-year) for all line pipe3, Table 26, and
the average (1.15 x 10~3 spins/mile-year) based on the line pipe age,
Table 28. Agreement would be expected because the typical line was assumed
to be the average diameter and age of line pipe in the U.S. Considering
that the U.S. statistics on the actual ages of all U.S. pipelines are not
known and mileages were based on estimates, these comparative values are
considered quite reasonable.
Data, such as given in Tables 19 and 21, indicate that "leak type"
spills (see Section 6.3.3.3) considered here as causing minor to medium
damage account for about 75 percent of all reported spills. The frequency
of all types of spills for the typical section of line pipe is 1.3 10-3
spins/mile-year. Thus, leaks are estimated to be 0.98 10" 3 spins/mile-
year. "Rupture type" spills (see Section 6.3.3.3) are considered to ac-
count for the remaining 25 percent of the reported spills. Thus, the es-
timated frequency of spills from ruptures is 0.32 x 10"3 spins/mile-year.
6.3.2.2 Risk—Volume of Spills V$T~
The expected spill volume VST is the summation of the spill volume for
lack of the individual causes of failures. Individual failures which could
result in spillage from the typical section of line pipe are identified
in the fault treeb of Figure 51.
The three main causes of failures are:
• Line pipe faults (53 percent volume)
- Defective pipe (35 percent volume)
- Corrosion (18 percent volume)
• Outside forces (46 percent volume)
- Impacts (33 percent volume)
- Non-Impacts (13 percent volume)
• Other (1 percent volume).
Excluding gathering lines.
Percentages indicated were obtained from Table 19.
237
-------
The total volume spilled from the typical section of line pipe is shown
to be approximately 1.3 barrels/mile-year, Figure 33. This result is in
agreement with the average volume spilled 1.4 barrels/mile-year for all
line pipe (Table 26) and the average volume spilled 1.3 barrels/mile-year
based on the average pipeline age (Table 28A).
6.3.2.3 Risks—Size of Spills VNLL, VNLR, VM—
"Leak type" spills are defined here as those spills that occur as a
slow escape of petroleum. These are caused by various line pipe faults
and result in hole-through failures ranging from pinhole to medium-sized
breaks. These spills are considered to cause small to medium damage (less
than 1000 barrels). The nominala largest leak spill size VNLL that could
be leaked from the reference line before being noticed and the operation
shut down is estimated at 400 barrels. This value, of course, can vary
widely depending upon the particular pipeline system. However, the value
is realistic considering the limitations of the following two main spill
monitoring methods in common use:
• Biweekly or ground patrol visual line inspections for direct
indications of leakage. This method, under worst-case condi-
tions, could allow leakage for the entire two-week interval
between inspections.
• Pipeline monitoring systems (i.e., pressure monitoring, flow
monitoring, etc.) typically can only detect a spill if devia-
tions exceed five percent of throughput.
"Rupture" type spills are defined here as those spills that occur as
a sudden escape of petroleum; they are normally caused by various line
pipe faults and result in hole-through failures ranging from medium-size
breaks to catastrophic failures. These spills are considered to cause
medium to major damage. The nominal largest rupture size VNLR f°r a major
pipeline rupture in the typical section of line is estimated to have a
maximum spillage of 7500 barrels. This is based on the assumption that a
rupture would be detected in about two minutes and about 72 minutes would be
required to close a mainline valve.
The mean spill size VM can be computed from the ratio of VST and F<-,..
The computation for the reference line is as follows:
V - v - 1.3 barrels/mile-year _ 1000 barrels
M" " 1.3 x 10'3 spins/mile-year
aBased on assumptions as to how the line will operate (e.g., frequency of
inspection, accuracy of flow meters, emergency shutdown, etc.), the cal-
culated volume of the pipeline and spill size data for typical lines.
238
-------
6.3.2.4 Relative Measure of Risks (Frequency x Spill Size) Leak, Rupture,
Mean—
The relative measure of risks, product of frequency and spill size,
for the typical section of line is shown in Table 57 and Figure 52. Val-
ues are provided for both spill frequency and the two largest nominal spill
sizes (leak and rupture) and mean size that can be expected from a single
accident. This information on leaks and ruptures is particularly impor-
tant in the evaluation of the potential damage of an accident in a high
risk area.
The relative measure of spill risk per year for "leaks" per year is:
RMR, _ = F. - x VNI , = °:98 SP".15 x 40QgS?ne1S - 0.392 barrels .
LT LT NIL 10-3 mile/year spl11
The relative measure of spill risk per year for "ruptures" is:
RMR - F x v - 0-32 7500 barrels = 2 4 barrels
KMKRT - hRT x VN] p -- 175 x cm 11 £•<*• oarreis
R1 Rr NLR 10 J mile-year sp1 "
Since the estimated total of all types of spills for a typical section of
line is 1.3 x 10~3 spills/mile-year, the relative measure of spill risk
per year for all spills is:
F-- x VM - *'3 SP111S x - 1-3 barrels
ST M 10-3 mile/year sPll]
6.3.2.5 Risks External to Line Pipe—
The risks (potential severity) external to the line from the acciden
tal spills have been grouped into the following ten categories:
• Fluid transported
• Undergound water supplies
• Water
• Industrial
• Population
• Commercial
• Underground facilities
• Surface transportation
• National preserves
• Spill volume.
239
-------
TABLE 57. RELATIVE RISK OF OIL SPILLS FROM
A TYPICAL SECTION OF LINE PIPE
Type of Spill
Leak
Rupture
Mean
Frequency
(Per Year)
0.98X10"3
(FLT)
0.32 X 10"3
1.3 X 10"3
Spill Size
Small to
Medium
Medium to
Major
Medium to
Major
Nominal
Largest Size
(Barrels)
400
<-VNLL)
7500
-------
"RUPTURE TYPE" SPILL FROM
"TYPICAL" SECTION OF LINE PIPE
(REFERENCE LINE)
MEM SPILL FMM "TYPICAL"
SECTION OF LINE PIPE
(REFERENCE LIKE)
"LEAK TYPE" SPIll FROM
"TYPICAL" SECTION OF LIKE PIPE
(REFERENCE LINE)
2.4 BARRELS/YEAR
1.3 BARRELS/YEAH
0.192 BARRELS/YEAR
ro
-pi
NOMINAL
LARGEST
SPILL SUE
FOR "RUPTUK
OF "TYPICAL"
SECTION OF
LINE
SPILL
FREQUENCY
OF "RUPTURES'
FROM "TYPICAL
SECTION OF
LINE
MEW SPILL
SIZE
SPILL
FREQUENCY
1000 BARRELS
1.3 » JO"3 SPILLS/YEAR
NOMINAL
LARGEST
SPILL SIZE
FOR "IE.W
OF "TYPICAL
SECTION OF
LINE
SPILL
FREQUENCY
OF "LEAKS"
FROM "TYPICAL
SECTION OF
LINE
7500 BARRELS
'W
0.12 X 10" 3 SPILLS/YEAR
400 BARRELS
0.99 I 10'3 SPILLS/YEAR
Figure 52. Relative measure of risk of oil spilled from typical section
of line (reference line)--leaks, ruptures, mean.
-------
The relationship of these potential failures is shown in the fault tree
of Figure 53. It should be noted that the risk value for failures in each
category is assigned a value of one because the typical line is assumed to
be located in a low-risk area. Correction factors that account for a com-
plete range of risks (i.e., low, medium, high), are provided in Section
6.3.3.4.
6.3.3 Correction Factors—General Use
The spill potential of most lines may be estimated by applying cor-
rection factors3 to the values developed for the reference line&. These
factors correct for the significant variations (e.g., geometry, age, etc.)
between lines. They are intended to provide a simple and practical means
of accounting for the overall effect of these variations on line pipe
failures.
Correction factors are provided for variations that have significant
effects on the four main items that are used in evaluating the risks of
oil spills. These are:
• Frequency of spills
• Volume of spills
• Size of spills
- Mean
- Leak
- Rupture
• External environment.
The first three items are primarily related to failures of the line itself.
The fourth item deals with damage to the environment, external to the line
pipe, that occurs after or during an accidental spill of petroleum. Cor-
rection factors for each of these four items are provided in the subsec-
tions that follow.
Because reportable information on accidental spills and operating com-
pany lines in the U.S. is limited (see Section 6.2.1), values for some
factors may be qualitative and should be considered as very rough es-
timates. These factors are intended for use in developing spill pre-
vention and control guidelines and should not be used for other analy-
sis purposes.
In this section it is assumed that the lines commonly use similar in-
spection and leak detection methods (see Section 6.3.1). Thus, cor-
rection factors are not included for these methods.
242
-------
SPILL DMWGE EXTERNAL
TO TYPICAL SECTION OF LINE PIPE
(REFERENCE LINE)
FLUID \ / UNDERGROUND
TRAIISPORTED /
-------
6.3.3.1 Frequency of Spills-
Many factors influence the frequency of spills from line pipes trans-
porting petroleum. Analysis and evaluation of available spill data indi-
cate, however, that only a few factors vary in such a manner as to signifi-
cantly affect the incidence of spills between different lines. These in-
clude:
• Age
• Geometry (diameter, wall thickness)
• Use
• Length.
Another important consideration is the minimum spill size. In this analy-
sis, spills of 50 barrels or more generally are considered to be signifi-
cant. In certain high-risk areas, however, even a spill of a few barrels
might be considered significant. Thus a correction factor is included for
spills between 1 and 50 barrels. Values for correction factors are pro-
vided for each of these five items in Figure 54. A fault tree for these
correction factors is shown in Figure 55. Note that the overall correc-
tion factor, CFp, is the product of the individual correction factors.
The five items are treated separately in the subsections that follow.
Other factors that are important include:
• Depth of burial
• Type of line
• Corrosion control
• Construction
• Material
• Soil condition
• Operating pressures
• Maintenance and inspection
• Overpressure.
The effect of this latter group appears to be minor. Effects of many of
these items are partially accounted for by the former group. For example,
the quality of line pipe construction and materials decreases with age.
Information that is necessary for a very accurate estimate of the effect
of these items in this latter group is not generally available (i.e., re-
portable to government agencies or available in the literature). For ex-
ample, the number of accidents versus depth of cover is reported, but the
miles of pipeline for each depth are unknown. The same holds true for type
of weld and grade of .pipe. Despite the lack of optimum information, reason-
able estimates can be made of the effects of these factors. Information
pertaining to the actual spills is reportable (e.g., Appendices A and B),
and reasonable estimates of the relative effect of these factors are possible.
244
-------
•a
CJ
I.
0
4-1
U
Correction f
3.0
2.5
2.0
1.5
1.0
0.5
0
—
i
_
-
J '
II 1 |l
t-
o
C
o
£
o
1.2
1.0
0.8
0.6
0.4
0.2
0
10 20 30 40
Age (Years)
J_ I I I I I I I I 1 1 1
50
8 12 16 20 24 28 32 36 40 44 48
Geometry Diameter (Inches)
ro
-P*
en
_400
u.
o
L.
',o 300
200
100
I
100 200 300 400
Length (Miles)
o
o
1.00
0.75
0.50
0.25
J_
"25 50 75
Use (Percent)
I.
8
u
C
O
J 3
10
8
6
4
3
100
1 2 34 6 8 10 20 30 40 50
Minimum Significant Spill Size
(Barrels)
Figure 54. Correction factors for frequency of spills--
age, geometry, length, use, size.
-------
Correction Factor
Frequency
Q
Spill Size
(CFSS)
Figure 55. Fault tree, CFp, correction factors for
frequency of spills.
246
-------
Additionally, important statistics required for analysis such as mileage of
line pipe by age can be estimated, and mileage of line pipe for each diam-
eter and location (state) is known.
6.3.3.1.1 Age
Corrections for line pipe age are provided in Figure 53. Results in-
dicate that the age of the line pipe has the most significant effect on the
estimated spill frequency.
Data such as presented in Figure 36 and Table 28A show a dramatic in-
crease in the spill frequency with pipe age. Results presented in Fig-
ure 36A and Table 25 clearly show that the age of the line pipe is a major
factor in the frequency of incidents. Lines built before 1950 appear to
have a much greater incidence of corrosion, while lines built before 1930
appear to have a much greater incidence of spills caused by outside forces.
Corrosion is a time-dependent process. Hence, older lines are ex-
pected to have a higher incidence of corrosion. Also, as the average age
of U.S. lines continues to increase, higher incidents of corrosion normally
would be expected. However, spills reportedly caused by corrosion have
decreased significantly in the past few years. Some of this decrease can
be accounted for by operator compliance with U.S. regulations that are in-
tended to reduce line pipe corrosion (i.e., Titles 49, Part 195.414,
195.415, and 195.416). Other decreases can be attributed to operator sys-
tems that exceed the requirements of government regulations.
6.3.3.1.2 Geometry-Diameter and Wall Thickness
Corrections for line pipe geometry (CFg) are provided in Figure 54.
These corrections indicate that the diameter of the line pipe has the
second most significant effect on the frequency of spills.
Data presented in Figure 32 show that the incidence of spills de-
creased rapidly with increasing line pipe diameter. This result occurred
for line pipe diameters of four inches to approximately 16 inches. The
incidence of spills increased slightly for line pipe diameters of 16 in-
ches to 32 inches. These results are in almost direct contrast to the ex-
pected incidence of spills based purely on the cross-sectional area of the
pipe; spill frequency normally would be expected to increase with increas-
ing cross-sectional area of the pipe.
There are a number of possible explanations to account for these con-
trasting results. Recent research82 on line pipe damage has indicated
that thin-wall line pipe is more easily damaged. Our analysis of spill
data indicates that this conclusion is reasonably correct for pipelines
transporting liquids. Thin-walled small-diameter lines, particularly
older ones (built before 1934), have a higher frequency of failures from
outside force incidents. Data indicates that the wall thickness effect
appears to become less a factor for line pipe diameters greater than 16
inches. At about this diameter, defective pipe and corrosion cause in-
creasingly more failures apparently because of the larger cross-sectional
areas.
247
-------
The age of the line pipe for each diameter is not known and the ef-
fects of age are uncertain. However, larger diameter lines (i.e., greater
than 16 inches in diameter) would be of newer vintage. Thus, low rates of
failure from corrosion and construction defects would be expected. Also,
the mileage of line pipe by diameter that is excluded from reporting (pri-
marily gathering lines) is unknown. Despite the limited data available, the
correction presented in Figure 54 is considered to provide an adequate
estimate of the overall effects of the various causes of line pipe failures
as related to diameter.
6.3.3.1.3 Use
Corrections for line pipe use (CFg) are provided in Figure 54. The
frequency of spills is assumed to be directly related to the use of the
line. For example, if the line is in operation 50 percent of the time, the
incidence of spills would be expected to be slightly more than 50 percent of
the value than if the line were operating at all times. Long periods of
shutdown would be expected to increase the frequency of spills, for example,
from internal corrosion.
6.3.3.1.4 Length
Corrections for line pipe length (CF-j) are provided in Figure 53.
The frequency of spills is assumed to be directly related to the length
of the line. For example, if a line is 100 miles long, the incidence of
spills would be expected to be 100 times greater than the one-mile sec-
tion of line.
6.3.3.1.5 Minimum Spill Size
The correction factor for the frequency of spills for spill size (CFSS)
of less than 50 barrels is shown in Figure 54. Various data and other con-
siderations were evaluated for this correction factor.
Spills of less than 50 barrels are difficult to estimate. Most spills
of less than 50 barrels are not reportable. Estimates of spills of less
than 50 barrels that are reported are also subject to criticism. Various
data bases (U.S. and foreign) on accidental spills are available for esti-
mating these incidents. For example, the mathematical analysis in Appendix
E, based on OPSO reported spills, indicates that at least 70 percent of
crude spills are less than 50 barrels. Data in Table 33 indicate that this
percentage is much lower. In contrast, careful reporting and analysis of
spills by one company, for a number of lines and for a large amount of
throughput, indicates that over 85 percent of the spills were less than 50
barrels. Other data bases such as Figure 42, indicate similar varying
results.
The size of small spills in the range of 50 barrels is especially dif-
ficult to estimate accurately for a number of reasons. No specific guide-
lines are in common use for accurately estimating the spill size. One would
normally expect a natural tendency to underestimate the size of a small
spill and thus many spills may go unreported.
248
-------
Talcing all these items into consideration, the correction factor for
minimum spill size is given in Table 54. This factor is based on the as-
sumptions that 75 percent of the spills are less than 50 barrels and the
size distribution is similar to the one presented in Figure 42.
6.3.3.1.6 Other Important Factors
Depth of Cover
The number of outside force incidents has been reported as a function
of the depth of burial of line pipe (Table 31), but the mileage, geometry,
and age of line at each depth are not reported. Thus, the effect of depth
of cover on the incidence of spills cannot be properly evaluated and a cor-
rection factor developed. Depth of cover is considered an important factor
in the incidence of spills, and suitable information is needed for proper
evaluation.
It should be noted that only 21.1 percent of the incidents occured
for a depth of cover from 31 to 40 inches (Table 31). Assuming3 a much
larger proportion of lines are under this depth of cover than the portion
of incidents reported, it could be inferred that a suitable depth of cover
is a significant factor in reducing the spill frequency.
The depth of cover for underwater lines is an extremely important fac-
tor in preventing spills. This is because most accidents from outside
forces are due to external impacts from such sources as anchor dragging.
A depth of cover of three to four feet is normally sufficient to eliminate
most damage from anchor dragging. In the case of onland lines, equipment
rupturing lines is the predominant cause of damage by outside forces and
evacuation equipment a minor one. A depth of cover of three to four feet
is considered sufficient to minimize spills caused by most equipment.
Depth of cover often becomes a problem because of soil erosion. For
example, lines buried underwater at three feet may become uncovered due to
ocean currents or onland lines due to erosion from water. Thus, even if a
line is buried, it is considered important to inspect the depth of cover.
This is particularly important for underwater lines where impact damage
cannot be easily detected. It may also be significant for onland lines,
particularly in high risk areas, because of potential damage from equip-
ment rupturing the line.
The depth of cover as required by existing U.S. regulations is con-
sidered adequate for line pipe. The possibility exists that the cover
above the line may substantially decrease, and in some instances the line
may become exposed because of soil erosion or other events. This is con-
sidered to be a factor in the incidence of spills. However, there is
aln 1971 and 1974, for example, there were 67 and 76 cases of equipment
rupturing the line and only four and eight cases of spills caused by
excavation.
249
-------
insufficient information pertaining to the depth of cover to establish a
suitable risk-correction factor. But this problem will be considered in the
development of a spill prevention program.
Various other factors have some bearing, but do not appear to be major
factors in the frequency of spills. A brief discussion of a few of these
follows.
Type, of Line
Crude gathering and trunk lines cannot be distinguished for most of the
incidents of spills reported to OPSO. Also, a large portion of the gather-
ing lines are exempt from spill reporting. Thus, a truly accurate correc-
tion factor cannot be developed. However, crude and product trunk!ines
appear to have similar incidence of spills. This is indicated in Figure 26
where crude spills averaged about 0.88 spills per mile, while product spills
average 0.92 spills per mile. Product lines are generally newer and of
larger diameter than crude lines; thus, a lower spill rate would be expected
for product lines. This latter observation is found to be correct when most
crude gathering lines are not included in the total mileage used in comput-
ing spills per mile. For this exclusion, crude spills average 1.3 spills
per mile. Crude lines are slightly older and of smaller diameter. However,
age and geometry effects have already been accounted for.
Based on these considerations, there appear to be no significant
differences for the type of line. Thus, no correction factors are provided.
Corrosion Inspection and Control
U.S. regulations exist for installation, maintenance and periodic
inspection of cathodic protection systems and for corrosion control (see
DOT 49 CFR 195.414, DOT 49 CFR 195.416 and DOT 49 CFR 195.418). These
regulations apply to most lines and are primarily for reduction of ex-
ternal corrosion. Since these regulations are quite comprehensive and
are applied equally to most lines, there appears to be no need for cor-
rection factors.
6.3.3.2 Volume of Spills—
A variety of factors influence the volume of spills from line pipe
transporting petroleum. However, only a few of the variations between
lines are considered to have the potential of significantly affecting the
spill volume. These include:
• Diameter
• Length
• Use
• Pumping station shutdown time
• Mainline valve closure time
• Line elevation.
250
-------
A fault tree for these correction factors is shown in Figure 56. The
overall correction factor CF$v is the product of the individual correction
factors. Values for correction factors are presented in Figure 57 for diam-
eter, length, use and pumping station shutdown. Correction factors are not
provided for mainline valve closure time and line elevation but these can be
computed using information provided in Table 48 and Reference 58 (see also
Section 6.3.1). These are considered to be less significant than the first
four factors. Estimates for these latter two factors are difficult to make
for two reasons. First, a number of variables must be considered. Second-
ly, necessary information for many of these variables is either not avail-
able or unknown.
6.3.3.2.1 Diameter
Corrections for the diameter of the line pipe (CF^) are provided in
Figure 37. Results indicate that the diameter of the Tine has a major ef-
fect on the estimated volume of a spill.
Data, such as presented in Table 26 and Figures 31 and 33, show a
dramatic increase in the average spill size with pipe diameter. The spill
size is essentially proportional to the cross-sectional area of the pipe.
These data follow approximately the same profile as the spill size data
for Western Europe and Canada (Figure 41). However, the mean spill size
in the U.S. for each diameter appears to be much larger than for Western
Europe, i.e., 1000 barrels to 666 barrels.
6.3.3.2.2 Length
Corrections for line pipe length (CF-j) are provided in Figure 57.
The volume of spills is assumed to be directly related to the length of
the line. For example, if a line is 100 miles long, the volume of spills
would be expected to be 100 times greater than the one-mile section of line.
6.3.3.2.3 Use
Corrections for line pipe use (CFU) are provided in Figure 57. The
volume of spills is assumed to be directly related to the use of the line.
6.3.3.2.4 Pumping Station Shutdown
Corrections for pumping station shutdown time (CFps) are presented in
Figure 57. This correction is an important consideration in the volume of
accidental spills. In the event a rupture or large-size break occurs, the
ability to rapidly shut3 the system down at the pumping stations is
Shutdowns not occurring at the proper time should be avoided. Incorrect
shutdown can create surge pressure that might cause leaks at other line
pipe locations.
251
-------
ro
tn
ro
CF,
Di ameter
\
/Mainline**
/ Valve
Closure
V (CFn,c> '
Line
Elevation^
(CF,
Figure 56. Fault tree, CFSy, correction factor for the volume of spills.
-------
8.0_
cn
co
2.0
£-
u
i; 1-5
1.0
0.5
12 16 20 24 28 32 36
Line Pipe Diameter (Inches)
40 44 48
6"~ 8
500
400 500
Length (Miles)
Shutdown time (Minutes)
Use (percent)
Figure 57. Correction factors for volume of spills—diameter, length, use, shutdown time.
-------
critical in minimizing the volume spilled. For example, if a line rupture
discharge rate3 were 5,000 barrels per hour, a two-minute pump shutdown time
would allow 166 barrels to be spilled, a six-minute shutdown time would
allow 500 barrels to be spilled.
A survey of pumping station facilities indicates that over 50 percent
of the facilities can shut their systems down within one minute, 70 percent
in two minutes, 96 percent in five minutes. Thus, a range of zero to five
minutes was used in the correction factor. Also, it was assumed for the ref-
erence line that rupture-type spills would be detected within ten minutes
and the line required two minutes for pump shutdown. Thus, if it required
five minutes to shut the pumps down in another line, the spill volume for
this line would be expected to be at least 25 percent greater than for the
reference line.
6.3.3.2.5 Mainline Valve Closure
Rapid closure of a mainline valve can significantly reduce the volume
of petroleum that escapes from an accidental spill. A recent study58 in-
dicates that additional spillage can occur unless the mainline valves are
closed rapidly after pump station shutdown. This study indicates, for
example, that a 14-inch line with a break (4-inch diameter hole) at the
bottom could decrease the 300 barrels per hour of spillage after pump sta-
tion shutdown. However, diffusion into the surrounding soil and eventual
buildup around the pipe until the static head is equalized is expected to
significantly reduce the spill rate.
Results of an industry survey58 indicated that it requires an average
of 72 minutes to reach and close a mainline valve and an average of 2.6
minutes for a remotely controlled valve. Ninety-five percent of the valves
surveyed required manual closure. Manual closure and a 72-minute closure
time was assumed for the reference line.
In addition to the effect of fluid buildup, a number of other factors
affect the spill rate. These include:
• Hole size
• Location of hole
• Line pressure
• Line elevation
• Soil
• Other.
Again, the effect of many of these factors on the spill rate is unknown.
Thus, an accurate estimate of this factor is difficult for individual lines.
It is extremely complicated to develop a general factor that could apply to
Line pipe rupture can result in spills in excess of throughput. Depend-
ing upon the line, spill rates can exceed flow rates by large factors
(i.e., two or three).
254
-------
most lines. Considering that the effect of mainline valve closure is of
much less significance than CFj or CFps and considering also the difficulty
in obtaining an accurate correction factor for CFmVc» a value of one is
normally assumed for this correction factor.
6.3.3.2.6 Line Elevation
Even after pumping station shutdown and mainline valve closure, spill-
age can occur because of the existing static head in an elevated line.
This problem is particularly significant in isolated and/or high risk areas.
The extent of the effect of line elevation depends upon a number of line
parameters (e.g., diameter, pressure, flow rate, etc.) and is not simply
determined. Values for the line elevation correction factor (CFie) are not
provided here (assume unity) but should be estimated for a particular line
where a potential problem exists. The relationship of line elevation to
other correction factors, however, is shown in the fault tree (Figure 56).
6.3.3.2.7 Other Factors
Flow Velocity
Flow velocities depend upon the size of the line and typically range
from 6 fps for six-inch lines to 12 fps for 36-inch lines. No correction
factor is provided since the factor is accounted for in the line diameter
correction.
Location of Mainline Valves
Check valves in uphill sections and remotely controlled valves in down-
hill sections are desirable. Since 1974, the ANSIS 31.4 Code requires re-
motely-controlled valves at a maximum spacing of 7.5 miles in commercial,
residential and industrial areas and 10-mile spacing of manual valves for
other liquids.
Valves generally cannot be installed in long length offshore lines.
This presents a serious problem in underwater lines located in elevated
areas.
6.3.3.3 Size of Spills-
Most of the factors that have a significant influence on the volume of
spills also influence the size of spills. These include:
• Diameter
• Pumping station shutdown time
• Mainline valve closure time
• Line elevation.
255
-------
These four factors have an effect on the size of spills resulting from rup-
tures or mean (average size) spill size. Fault trees for ruptures CFyR and
mean size spills CV\/M are presented in Figure 58. Values for these correc-
tion factors are the same as those presented for the volume of spills in
Table 57. Also, the discussion of each factor presented in Sections
6.3.3.2.1 through 6.3.3.2.4 applies to the size of spills and will not be
repeated here.
All four factors are considered to have a very minor effect on the
spill size for leaks. Thus, no correction factors of fault tree are pro-
vided for "leak type" spills.
6.3.3.4 External Environment-
There are many risks to the environment from escaping petroleum. These
risks are identified for each of ten categories in Table 58. Correction
factors that account for a complete range of risks (low, medium, high) are
also presented in this table. Many of the values9 for the correction fac-
tors are considered to be only qualitative estimates (see Section 6.2.1)
because of the absence in required spill reporting forms of suitable quan-
titative data relating to the external environment.
A fault tree for correction factors for risks to the external environ-
ment from escaping petroleum is presented in Figure 59. Note that the over-
all correction factor, CFgQ, is the product of the individual correction fac-
tors.
Rather than discuss each correction factor value in each category, ex-
amples of typical categories will be provided. Spills occurring in lakes
and rivers generally can be confined to small areas using suitable contain-
ment techniques; risk factors of 10 and 20 are assigned, respectively. Off-
shore spills, however, are quite difficult to contain and often (about 95
percent of the time) reach recreational areas along the shoreline; risk fac-
tor of 100 is assigned for spills located offshore. A recent study1*6 indi-
cates that a serious potential exists for damage to underground water sup-
plies in the United States. The study indicates that large supplies near
large urban populations are particularly vulnerable. Based on information
contained in the study, lines near medium to large volumes of underground
water supplies are assigned risk factors of 10 and 100, respectively.
6.3.4 Computation of Oil Spill Risk for Accidental Spillage
6.3.4.1 Risks that Oil will Escape from the Line Pipe--
Oil spill risks for line pipe for most petroleum pipeline systems can
be computed using information supplied in this section. Values are obtained
by applying the appropriate correction factors to the values determined for
the reference line.
aThese values are primarily intended to identify sources of high risk and
for use in developing guidelines in this study. They should not be used
in other analyses.
256
-------
A
V
&
CF,
Correction Factor
Spill Size—Rupture, Mean
Diameter \ /'
(CFd)
Pumping
Station
/ \ Shutdown
/ \CCFpjOx'
Mainline
Valve .
Closure
Figure 58. Fault tree, CFyR or CFy^, correction factors
for spill size—rupture and mean.
257
-------
TABLE 58. CORRECTION FACTORS FOR RISK OF SPILL DAMAGE
EXTERNAL TO LINE PIPE
Location or Coirmodlty
1. Fluid Transported
Crude
Products
LPG
LUG
Other
2. Underground Water Supplies
Zero potential
Snail volume
Hedlun voluise
Large volume
.>. uter
Zero potential
Streen
Lake -
River
Marshlands
Offshore
teaches
Terminals
High seas
4. Industrial
Zero
Light
1 Bedim
i Heavy
S. Population
Undeveloped
Rural
, Residential
Urban
6. Coemrclal
Zero
Light
Medlup
Heavy
7. Underground or Adjacent
Facilities
Zero
Electrical
fias
LPS
LN6
3. Surface Transportation
None
Road
Highway
Railway
Freeway
Sajor freeway
9. National Preserve
Zero
Snail
Medium
Major
10. Spill Volume
Fatalities
1-100 (barrels)
100-500 (barrels)
500-1,000 (barrels)
1,000-10,000 (barrels)
>10,000 (barrels)
Injuries
O-ICO (barrels)
100-500 (barrels)
500-1,000 (barrels)
1,000-10.000 (barrels)
>10,000 (barrels)
Correction Factor
Identification Low Risk Medium Risk Major Risk
I-' 5-9 ' 10-100
CFuw
M
CFi
"P
"c
CFuf
CFnp
CF,v
3 . '
10
2' ,
i«> i
i • i
10 !
wo :
id)
10 i
20 I
100 ;
100 i
100 '
100
2
I
1<»
l'1'
2
1
1»>
jdl
Z
5 :
10 :
100
s ;
10 j
i SO
s ;
10
100 1
i
5
10
20
50
S
10
50
5
10
s
10
Note: (1) Deference Line Pipe
258
-------
CORRECTION FACTOR
RISK OF SPILL DAMAGE
EXTERNAL TO LIME PIPE
in
vo
Figure 59. Fault tree, CFfp. correction factors for risks of spill
damage external to line pipe.
-------
Oil spill risks for frequency F$TC and volume V$TC of spills can be
computed using Table 59 and the figures indicated. Fault trees for F$TC
and VSTC are shown in Figure 60.
Oil spill risks of the nominal largest spill size of ruptures VNLRC
leaks V|\||_|_c and mean volume VMC, can be computed using Table 60 and the in
d fi
__
dicated figures. Fault trees for VNLRC> VNLLC ^ VMC are sn°wn i" Fig-
ure 61.
The relative measure of oil spill risks for rupture RMRR, leaks RMR\_
and means spills RMRw can be computed using Table 61. Fault trees RMRR,
RMR|_ and RMR|v| are provided in Figures 62 and 63.
6.3.4.2 Risks External to Line Pipe—
The relative severity of spills or the risks external to the line from
accidental spills can be computed using information suppl-ied in this sec-
tion.
First, the correction factor CF^o for risks external to the line are
computed using Table 62. Then the values for risks external to the line
can be computed using Table 63; this is accomplished by taking the product
of the risk correction factor CFgo and the previously computed risks (see
Section 6.3.4) that petroleum will escape from the line. Spillage risks
external to line pipe can be computed in Table 63 for the following:
• Spills
- Frequence F<-E
- Volume VSE
• Spill size
- Rupture (nominal largest)
- Leak (nominal largest)
- Mean VSME
• Relative measure of risk
- Rupture
- Leak
- Mean
If, for example, the nominal largest rupture size VNL.RC were 5000 bar-
rels for a particular line located in an urban area, the risk correction
factor is:
CFED = CFFT x CFUW x CFW x CFj x CFp x CFC x CFyF x CFSJ x CFNp x CFSC
= Ix IxlxlxlOOxlx 1x1 xl x 10 = 1000.
260
-------
TABLE 59. CALCULATION OF THE FREQUENCY FSTr AND VOLUME VcTr
OF SPILLS FOR LINE PIPE
IN* *r HM
Crudt Cithtrln* lint
Onlll*
Vndtnaltr
fruit InitkltM
OnltM
IMtmltr
rnxtucl frunUflit
Onltn*
IMtnattr
IOIM
SPIU FREOUENCr
Mftr-
MCt
«>
'si
Spllll/
Ittr
rrMjutncy Corrtctlon rtcUrl' '
Intflilduil (irlitlant
A*
«.
Ctawtrr
",
Itnttn
"l
u»
".
Spill
Sift
",,
CiMuUtlvt
cf. • cr.
> cr. * cf.
> «,, •
"r
u«t ri»t
'si""r-
'sic
DtftrtiiM
lint MM
(Spill
felim)
*SI
Itrnti/
Tt>r
SPIU VOLUnE
«>|IM CtrrtcIlM ricltn'"
l»4l>l«itl Vtrlitloni
OliiHltr
"<
Unflll
Ml
Hit
tr.
»«Vln(
SUIlM
SKul-
wxn
HP.
Nilnllnt
rti>«
ClMurt
"«
lint
lltM-
tlm
"l.
CimiltlKt
"< " "•
.cr.
'"P.
•««
»"i.
•««»
line Up.
»5l"
"$y-
•»TC
••rrcli/
Year
ro
bt<: III SM liklt •-!.
It) SM I(|nn 4-f.
Ill SM Mt«* 4-1.
-------
Spill Volume
Line Pipe
STCX
Spfl) Frequency
Line Pipe
ro
en
ro
Correction Factor
for Volume
of Spills
Spill
Volume
Spill Frequency '
of All Spills
from Reference Line
Correction Factor
Frequency
'ST
1.3 Barrels/Year
1.3 X 10"J Spills/Year
Figure 60. Fault trees, V$jc and F
volume of spills from
, frequency and mean
ne pipe.
-------
TABLE 60. CALCULATIONS OF THE NOMINAL LARGEST SIZE OF RUPTURES, VM1Dr!
LEAKSNLLC AND MEAN VMC FOR LINE PIPE SPILLS NLKL
Type of Line
Crude Gathering
Line
On land
Underwater
Crude Trunk! ine
Onland
Underwater
Product Trunk! Ine
Onland
Underwater
TOTAL
Reference
Line Pipel
(Nominal
Largest
Ruptured
Spill
Size)
VHLR
(Barrels)
NOW
1)
Diameter
CRd
WL LARGEST RUPTURE SPILL VNLRC
Volume Correction Factors'2*
Individual Variations Cumulative
Pumping
Station
Shutdown
"Ps
Mainline
Valve
Closure
"me
Line
E1eva- -
tion
CF,e
"d x CF
« ffmc *
CFle .
CFVR
Line
Pipe
VNLR *
CFVR -
V«IRC
(Barrels)
NOMINAL MEAN SPILL SIZE V^.
Reference
Line PtpeG
(Mean
Spill
Size)<3>
VM
(Barrels)
Cumula-
tive
Volume
Correc-
tion
Factor
CFVR
Line
Pipe
VH*
CFVH.
»nc
(Barrels)
NOMINAL
Reference.
Line Pipe1
(Nominal
Largest
Leak
Spill
Size)
VNLI
(Barrels)
LARGEST LEAK SPILL SUE VNU(.
1)
Individual
Variations
Line
Eleva-
tion
CFle
-
Cumula-
tive
CFle *
CFVL
Line
Pipe
VNLL *
CF¥L*
VNLLC
(Barrels)
ro
o>
OJ
Note: (I) S«« Table 4-1.
(2) See figure 4-9.
-------
ro
cr»
.
A.
i V
/ViCV
Urottl Spill Sl/f
«'
U»k»
Ml)
Correction Factor
Urottt Spill Sill
of loll
(Nu.lr.il)
,'Cf.,
.. -
X
spill Slit or P.f.
• lino l«k>
Correct lo
Spill V
of Rupl
1 14I9.1I Spill SIM
1 0»
C 1 nuptorot
& 1 (NlWlMl)
t
n factor 1 s' ... ^.
-I,.- 1 /nominal larootl^
„„ I V Spill Slia of lof
1 "VIIM duplyr.j.
1 "
^HC\~ 1 """ *
1
Cor reel to* factor
of Spllli
(««.)
IDUO Birr.ll
Figure 61. Fault trees, V^LLC* VNLRC» VMC> nominal largest size
of ruptures, leaks and mean for line pipe spills.
-------
TABLE 61. CALCULATIONS OF THE RELATIVE MEASURE OF OIL SPILL RISKS FOR RUPTURE RMRp,
LEAKS RMRL AND MEAN SIZE RMRM SPILLS FOR LINE PIPE K
Type of Line
Crude Gathering Line
On land
Underwater
i
Crude Trunk! ine
On land
Underwater
Product Trunk! ine
On land
Underwater
Total
Relative Measure of
Rupture Spill Risk RMRR
v (1)
NLRC
0.25 F$T(2>
CFf<3>
VHLRC
x O.Z5 FST
x CFf «
RMRR
Relative Measure of
Leak Spill Risk RHRL
v (1)
"HLLC
0.»Fs«>
(3)
CFF
VHLLC
x 0.75 FST
x CFF =
RMRL
Relative Measure of
Mean Spill Risk RMRM
„ (1)
>C
(2)
FST
(3)
CFf
VMC
x FST
xCFp =
RMRH
ro
o>
en
Note: (1) See Table 4-5.
(2) See Table 4-1.
(3) See Table 4-4.
-------
_ lypt" Splllt fro* LlM
(Itlttlvi Ntiwn of HUM
until spin SIM
lupturts
(Nwliul )
Spill Frtwwicy
o«
•uptyrtl
ro
en
luk Typt- Jpllli fro. Mm Clp.
(«tlttl» Nl»ur« of Kl
Figure 62. Fault trees, RMRp, RMRi , relative measure of risks (barrels/year) of
ruptures ana leaks from line pipe.
-------
Spilli from HIM
(RaUtlv* HMwn of Msk)
MOM Spill VoluM
Spill Fnqotney
0
Correction Factor
For HMH
of Spllli
Spill
Slzt
Spill Frtquwcy
of All Spllli
fro* Rofvrtneo Lino
Correction Fictor
Froquoney
1000 torroli 1.3 X 10"1
Figure 63. Fault tree, RMfyj, relative measure of risk
(barrels/year) of mean spills from line pipe.
267
-------
TABLE 62. CALCULATIONS OF THE CORRECTION FACTOR
CF£D FOR RISKS EXTERNAL TO LINE PIPE
Type of Line
Crude Gathering Line
On) and
Underwater
Crude Trunk) ine
On land
Underwater
Product Trunk) ine
On) and
Underwater
Individual Variations
Fluid
Transported
CFfT
Underground
Hater
Supplies
CFUH
Water
CFH
Industrial
CFl
Population
CFp
Conraercial
cfc
Underground
Facilities
CfUF
Surface
Transportation
Natural
Preserves
Spill
Volume
Cumulative
x CF., x CF.
x CFp x CFC
x CFuf: x CFjj.
x CF x CF
NP SV
'CFED
ro
cr>
c»
Note: See fable 4-1.
-------
TABLE 63. CALCULATIONS OF RISKS EXTERNAL TO LINE PIPE
Type or Line
Crude Gathering Line
Onland
Underwater
Crude Trunk! ine
Onland
Underwater
Product Trunk! ine
Onland
Underwater
Tola)
Spills
Frequency
FEO x FSTC
*FSE
Volume
CF£D x VSTC
* VSE
Spill' Size
Rupture
Norn. Largest
CFEO x VNLRC
'VSRE
Leak
Norn. Largest
CFED x VNLLC
IVSLE
Mean
CFEO x VMC
= VSME
Relative Measure of Spill Risk
Rupture
CFED x RMRR
" RHRSRE
Leak
CFED x RMRL
• RM«SLE
Mean
CFED X RHRH
* RMR,...-
irtt.
r\>
cr>
vo
-------
Hence, the relative severity or risk external to the line VSRE is:
VSRE = 500° barrels x 100° = 500,000 barrels
Thus, the indicated spill is considered to be equivalent to a 500,000 bar-
rel spill from the reference line.
270
-------
SECTION 7
ANALYSIS OF THE REDUCTION OF THE RISK OF OIL SPILLS FROM LINE PIPE
BY UTILIZATION OF LEAK DETECTION AND INSPECTION METHODS
The potential of selected leak detection and inspection methods to re-
duce the risk3 of oil spills (frequency and volume of spills and risks ex-
ternal to the line) from petroleum pipeline systems is estimated in this
section. Means are also provided to enable an operator to estimate the
risk reduction capabilities (effectiveness) of inspection and leak detection
methods for his own line. Since costs are a major consideration in a prac-
tical maintenance program for line pipe, costs of the various options (in-
spection and leak detection methods) and spillage are estimated and a cost-
effectiveness analysis carried out.
Means for reducing the risk of line pipe spills are examined in detail
in Section 7.1. First, the most promising inspection and leak detection
methods were selected based on a selection criteria. Then estimates made
of the capabilities of leak detection and inspection equipment to detect a
leak of a certain magnitude and/or impending failures. The methods are
evaluated based on their capability to reduce the frequency and volume of
"leak type" and "rupture type" spills for the reference line. Factors are
established which provide a simple means of comparing the capability of
methods to reduce the frequency and volume of spills. These factors help a
potential user select the most suitable method to satisfy the spill preven-
tion programs needs of a particular line.
The amount of oil spillage that can be prevented is examined in Section
7.2. Accidental oil spillage that exists for line pipe is first estimated.
Then estimates are made of spillage that can be prevented by implementation
of various scheduled inspection and leak detection systems. This is done
for all U.S. lines and the reference line. Means are provided so that
quantitative values of spillage and prevention of spillage can be determined
for most lines.
Section 7.3 contains a cost analysis of inspection and spillage.
A cost-effectiveness analysis of the selected inspection and leak de-
tection methods for various inspection schedules was provided in Section
7.4. Two measures of cost-effectiveness were used in the analysis.
Means are provided in Section 6 to enable an operator to estimate the
risk of oil spills for his own line.
271
-------
7.1 COMPARISON AND EVALUATION OF LEAK DETECTION AND INSPECTION METHODS
This section provides an in-depth comparison and evaluation of leak de-
tection and inspection methods that appear to be the most promising. An oil
spill risk reduction analysis is carried out for these selected methods.
This information is then used to estimate detectable and undetectable spill-
age for the reference line pipe and for all U.S. lines in Section 7.2.
The most promising methods are reviewed in the first subsection. Gen-
eral guidelines and criteria for method selection are presented. Methods
selected for further evaluation are identified and categorized by optimum
use, i.e., reducing one or both parts of the oil spill risk.
An analysis of the risk reduction capabilities of the selected methods
is carried out in the second subsection. Methods are analyzed as to their
capability of reducing the volume and frequency of "leak type" and "rupture
type" spills for both onland and underwater use. Risk reduction factors are
estimated for all methods and combinations of methods. In order to simplify
comparisons, normalized risk reduction values, based on a single reference
inspection, are provided. Using these values, for example, the effective-
ness of one inspection can be equated to a certain number of other inspec-
tions, i.e., one inspection pig survey to a number of visual inspections of
the line.
7.1.1 Selection of the Most Promising Leak Detection and Inspection
Methods
7.1.1.1 Guidelines--
No single inspection or leak detection method currently exists that is
capable of detecting all leaks in sufficient time to avoid an oil spill in-
cident. This situation exists for a number of reasons such as:
• Variety of line pipe systems
• Accidental spillage may occur at any location along the
line
• Wide variations in the frequency and volume of spills
• Numerous causes of spills.
Despite these difficulties, this report shows that the potential exists
for significant reductions in the volume and frequency of petroleum spillage
both for individual lines and those nationwide. In this study, the major
emphasis is for reductions of spillage nationwide. However, means are
provided in Section 6 for identifying and quantifying the risks of both
individual lines such as those located in potentially high risk areas and
lines nationwide and in Section 7 for estimating the reduction of risks
using the most promising methods. On a nationwide basis, maximum oil spill
risk reduction is accomplished by implementation of spill prevention pro-
grams incorporating methods that adhere to the following guidelines:
272
-------
• Adaptable to most lines
• Highly effective in reducing frequency of spills, particularly
those that are expected to result in ruptures
• Highly effective in limiting the spill size
• Cost-effective.
In high-risk areas, however, emphasis is placed on reducing the frequency
and volume of all spills to as low a level as practical for the specific
type of line. Hence, methods considered highly effective but, for example,
not adaptable to most lines are considered for use in individual spill pre-
vention programs for optimum spill reduction.
7.1.1.2 Selection Criteria--
Selecting the most promising inspection and leak detection methods was
based on the following criteria:
• Must be adaptable to most onland lines
• Must be adaptable to most underwater lines
• Must have a potential for a significant reduction in the
frequency or volume of spills
• Practical implementation and operation costs
• Limited interference with pipeline operations
• Rapid response for "rupture type" spills
• Inspections can be performed either by semi-skilied line
pipe personnel or inspection services
• High reliability or jiigh probability that the equipment
will operate satisfactorily during the inspection
• High design adequacy or probability that the inspection
equipment will accomplish its performance requirements.
7.1.1.3 Methods Selected for Further Evaluation—
Applicable methods were briefly described and compared in Section 5.3.
Of these, only a few leak detection and inspection methods were judged to
satisfy the general guidelines and selection criteria for line pipe. The
methods selected for further evaluation are identified in Table 64. Gen-
erally, the methods selected are highly effective in reducing one of the
two parts of the oil spill risks, i.e., reducing either the volume of
spills (Section 5.2.4.2) or the frequency of spills (Section 5.2.4.1). In
a few cases, a selected method may be effective for both parts. Therefore,
methods are identified as to their optimum use, i.e., reducing one or both
parts of the oil spill risk.
Although not selected, some leak detection and inspection methods may
be effective in solving atypical spill problems that are characteristic of
a particular line or small group of lines. For these special applications,
273
-------
TABLE 64. INSPECTION AND LEAK DETECTION METHODS SELECTED
FOR FURTHER EVALUATION OF LINE PIPE
Leak Detection and Inspection Method
1. Visual inspection by sir or ground patrol
in excess of required inspections
2. Visual inspection i, ;round patrol with
hydrocarbon detector :r other
comparable device
3. Hydrocarbon prone-towftsn or similar device
- Underwater lines 'inly
- All lines
4. Oil spill detectors
• Marine terminal or alatform
- Total line pioe rileege
5. Pressure deviations
a. Pump stations (existing method)
b. Along line
«. Flo, rate
a. Deviations
b. Comparisons (computerized)
7. volume comparisons (computerized)
a. Mathematical medelln; (computerized)
9. negative pressure surge (computerized)
10. Passive acoustic array (computerized)
11. External rods ulth passive acoustic
sensors
12. Pressure static
13. Hydrostatic
14. Pressure difference
IS. Change or add inhibitors as needed
ISr-lnsaeetlon of sample of line for veil
thickness by ultrasonic or comparable
technique
17. Survey by inspection pig-magnetic flux
type or other comoaraole device
18. Depth of cover inspection by sonar
(sldescan and penetrating) or other
comparable device j
- underwater lines only
- All lines
19. Charting of line pipe 1
- underwater Un« only
. All lines
20. Preventive program for outside
forces (Deference I2j-«ne call system
21. Pig line for water removal
22. Otner
- Catnodic protection system
Installation and maintenance
- visual inspection (biweekly) by air or
ground patrol (required inspection)
Reduction of
Size of Spill
Leaks j Ruptures
i
Reduction of Frequency of
Leaks
1
« j X «
X X
[NO
X X
X
X
X
! X
Spill (1)
! | Causes
Ruptures
X
Line Pipe Faults
Defects
X
X ' X
X i X
X '' X
;
X ;
X ; X
j
, X
I X
I
" i x
X
X
X
X
1
I
X
X
X
X
«
X
X
I NO
I NO
I
[NO
I
X
X
x
X
X
x x
X
X
X
X
X
X
X
X
X
X X
1(10 I*a
X X
X
I
X
I
External Internal
Corrosion Corrosion
X X
X X
X X
x ; x
Outside Forces
Equipment
Rupturing
Line
X
X
Damage byj
Excavation!
Equipment Other
X x
i
, ! x
;
X i X i X
X | X
X X
X
X i
!
X
X i
X X
1
X ! X
X
X
x x
| x \
X
x
X
X
X
X
X
mo
net discernible
274
-------
the operator should consider the potential of other methods described in
Section 5.3.
The most commonly used spill prevention methods for operating lines
currently required by U.S. government regulations are included in Table 64
for comparison purposes only. These methods, mainly bimonthly visual line
inspections and monitoring of cathodic protection systems, have already af-
fected some limited spill prevention. Cathodic protection systems, for
example, are considered to be responsible for some of the reduction of "leak
type" spills caused by external corrosion of line pipe.
7.1.2 Analysis of the Capabilities of the Selected Methods to Reduce
the Oil Spill Risk (Volume and Frequency of Spills)
For the purpose of this study, a spill reduction analysis can be best
accomplished by evaluating and comparing the methods that are applicable to
each part of the risk3. Therefore, the capability of selected methods to
reduce the volume and the frequency of spills are discussed and analyzed in
separate subsections.
Methods for onland lines are generally assumed in the analysis. These
methods normally apply to both onland and underwater lines.
Methods that might be particularly effective for only underwater lines
or lines close to water supplies are also included in the analysis. This
is done even though these methods would not contribute to a significant*3
reduction in the volume or frequency of spills nationwide. The reason
these methods are included is that high reduction of the frequency and/or
volume of spills from underwater lines can result in a significant reduc-
tion in the nationwide total of serious spills. This is because of the
high risk to the external environment from even low volume oil spills in
underwater lines.
A general discussion of factors that affect the capability to reduce
the volume and frequency of spills is presented in the subsections that
follow. Then a spill reduction analysis is carried out for methods that
reduce the volume and frequency of spills. Selected leak detection and
inspection methods are identified in tables along with inspection frequency,
Risks to the external environment are not directly accounted" for in the
reduction analysis based on only the frequency and volume of spills.
Underwater lines account for less than ten percent of the total line pipe
mileage. Hence, even if methods have the capability of significantly re-
ducing spillage from underwater lines, it would have a small effect of the
total nationwide.
275
-------
Other factors that should be considered include:
• System effectiveness
Reliability
Operational readiness
Design adequacy
• False alarms
• Other.
The effect of each of these factors varies widely depending upon the imple-
mentation of each method. It also depends on whether the method is used
for detection of rupture and/or leaks. The combined effects of these char-
acteristics along with these other considerations determine the overall
capability of a method to reduce the volume of a spill. Hence, each is
considered in the evaluation of selected methods.
Implementation of leak detection and inspection methods that are capa-
ble of adequately satisfying the required characteristics potentially will
not only prevent major pollution incidents but will significantly reduce
the annual loss of oil. Hence, implementation of methods that are effec-
tive in reducing the size of an accidental spill is an important considera-
tion in a spill prevention and control program. It should be noted, how-
ever, that methods effective in reducing spill volume generally are not
highly effective in preventing spills or reducing the frequency of spills.
Various leak detection and inspection methods to varying degrees can be
implemented to detect a spill before the volume becomes large enough to
cause a major pollution incident. Table 65 compares the characteristics of
selected methods for use on the reference line, assuming stable flowa.
Values are included for the four major system characteristics (sensitivity,
response, inspection frequency and leak location). For analysis purposes,
single values are presented in the table. However, wide variations in
sensitivity are possible in implementing certain methods. These variations
result from the variety of systems, equipment variations, operational pro-
cedures and other factors. Values given in the table and figures are not
optimum. However, these are considered typical of modern installations and
achievable for most lines. Many of the values presented in the table are
only rough estimates because the information available is insufficient for
more accurate estimates. This is often the case for new or experimental
methods. Additional details of specific systems are available from the
manufacturers indicated in Appendix E and from References 14, 58 and 60.
aDuring the periods when significant line transients exist, many methods
are much less effective than indicated.
276
-------
TABLE 65. COMPARISON OF THE SELECTED METHODS FOR DETECTION
AND LOCATION OF REFERENCE LINE LEAKS AND RUPTURES
Inunction or
Leak Detection Method
COXTINUOUS MMITOUW
Pressure deviations pue» station
along line pipe
Flow ratt deviation comparison
(cowuteriud)
VoluM comparisons (computerized)
NetMutical saddling (conputeHzei
Negative pressure surge
(cdeputertzed)
Passive acoustic array
- fta» 11M
- Retrofit
Oil spill detectors at narine
tenelnel or platfom
KXIOOIC
Pressure >ut1c
Hydrostatic
Pressure difference
External rods Kith passive
acoustic sensors
Leaks
Sensitivity
Thrauonout
us
is
1
0.3
0.6
1 0.1
0.2
HA
<.2
K
-
0.1
0.03
0.002
i 0.2
Visual Inspections by air or ' ..
around patrol for indication
of a spill
Visual inspection By ground patrol
irith nydracarbon probe or coeoara-
bla device for Indications of spill
< 0.002
larrels/
Hour
tt
HS
25
7.5
15
2.5
S.O
HA
2.0
US
-
3
0.7
0.06
5.0
-
<0.06
Response
Ti«e (1)
Lono-
Tera
ie
NS
1 hr.
2< tours
1 tour
24 tours
1 tour
DA
20 «1n.
US
15 «1n.
24 tours
72 tons
<2 nin. (}
288 tours
288 tours
loss
Sensitivity
TIM < 2
(barrels)
NS
NS
50
180
X
36
10
NA
<1
NS
SHll Spill
.75
16.8
4.3
-
•tinor spll
snail spil
Ruptures
Sensitivity
Percent
Throughout
20
2
>5
<4
<3
O.S
5
O.S
4
-
0.1
IK
NS
0.2
NS
< 0.002
Barrels/
Hour
500
50
125
100
75
25
125
25
100
-
3
NS
NS
5.0
NS
<0.06
Aesoonse
T1e» (il
rem
1 Hr.
10 hrs
Thr.
< 2 nin.
10 »1fl.
<2 «1n.
<2 «1n
1000
1000
250
6.6
25
< 1
31
<1
<*
SHll SPill
< 1
NS
NS
--
-
Inspections
Frequency
Continuous
Continuous
Continuous
Continuous
Continuous
Continuous
Continuous
Continuous
Continuous
After
indication
of laroe
lut
2
2
After
indication
of large
!••*
26
26
Location
Accuracy
I
Norn
None
None
None
t 2 riles
21 of
sensor
spacing
None
None
(etxeen
block
v«lve«
s: of
sensor
spacing
-
Feir feet
NOTES: (1) NS - Not suitable for indicated detection of leaks or ruptures.
(2) IHA- Information Is unavailable.
(3) Asswes systei operational after leak or rupture indication.
277
-------
The most important capability in reducing the volume of spills, for
both leaks and ruptures, is the sensitivity or accuracy of the particular
methods. A high leak sensitivity, i.e., a few barrels per hour, provides
the capability of detecting sptllage before a large volume of petroleum
can escape from the line.
High sensitivity, however, does not necessarily guarantee that a spill
can be detected and the spill volume reduced. Both the frequency of the
inspection and response time must also be considered. Frequent or contin-
uous inspection methods with good sensitivity are mandated for detecting
rupture type spills. In contrast, high sensitivity methods that require
longer response time and less frequent inspections are often suitable for
detecting small leaks. Rapid response time3 (few minutes) is necessary for
ruptures, whereas long-term response (1 to 24 hours) is usually suitable
for leaks. For example, a typical volume comparison system may be capable
of detecting a leak of 0.3 percent of throughput, but the leak must exist
for 24 hours before an alarm would be activated. In the case of the refer-
ence line, such a system would permit 180 barrels of spillage over a 24-
hour period before detection.
The capability of providing accurate leak location, i.e., within a few
feet, is also important. This capability is particularly advantageous in
certain high-risk areas. For example, spills from elevated lines in rough
terrain may continue (because of static head) even though the line is shut
down and the mainline valve closed. Location of a spill by conventional air
patrol might be quite time consuming and extremely difficult to carry out in
darkness, bad weather or dense foliage. A serious pollution incident could
occur. This could be avoided if a method that accurately locates leaks were
installed at the elevated sections of the line.
Other factors such as system effectiveness and false alarms are also
important considerations. False alarms, however, are generally accounted
for in the system sensitivity and response. System effectiveness depends
upon the specific system and will not be discussed further in this section.
It should be considered for each specific installation.
In addition to the capabilities of methods to reduce the volume of
spillage, sufficient information must be known about the actual spills.
This is important in selecting the most suitable methods for certain types
of spills. It is also important for defining spills so that estimates can
be made of the oil spill risk reduction that is achievable with each method.
The information needed for each accident includes:
• Geometry, dimensions and location on the line pipe of the
source of the escaping petroleum
• Volume of fluid lost as a function of time
• Spill volume
Response time is defined here as the time duration required to positively
identify a spill so that the system can be shut down.
278
-------
• Time the leak was detected and by what means
• Time the pumping station was shut down after detection of
spill
• Time the mainline valve closed
• Type of soil
• Other.
Much of this information is not available. In this study, however, esti-
mates and assumptions are made to account for this lack of information. For
example, leak-type spills are assumed to have low flow rates (less than a
few barrels per hour) with losses ranging from 1 to 400 barrels. Thus,
approximately ten percent of the reported spill volume is attributed to
leaks; the remainder is attributed to ruptures. (These percentages were
obtained from Table 19).
A reduction analysis of the spill volume for ruptures and leaks is
presented for the selected methods in the subsections that follow.
7.1.2.1.1 Estimation of the Capability of Reducing Rupture Volume Using
the Selected Methods
Selected methods for reduction of the volume of rupture type spills are
shown in Table 66. Frequency of inspection, percent reduction of spill vol-
ume, risk reduction factors and normalized values for risk reduction are also
included in the table. A single visual inspection in excess of the required
bimonthly inspections by air or ground patrol, is the reference inspection.
Bimonthly visual line inspections, periodic monitoring of cathodic protec-
tion systems, and the monitoring of pressure deviations at the pump stations
are considered to be the commonly used methods during the reference year for
detection of ruptures. Methods are evaluated primarily on the basis of the
product of sensitivity and short-term response time. However, effects of
line transients on the capability of a method to detect ruptures are also
considered.
In the analysis performed in this section, ruptures and large-size
breaks are assumed to account for approximately 90 percent of the volume
spilled. Also, ruptures and large size breaks are assumed to typically re-
sult in spill rates in excess of five percent of throughput. Hence, for the
reference line (flow rate of 2,500 barrels per hour), this type of spill
would result in a loss in excess of 125 barrels per hour.
Spill statistics in Section 4 (see Table 19) indicate that over 90
percent of the spill volume results from spills in excess of 400 barrels,
and over 80 percent result from spills in excess of 1,000 barrels. Thus,
methods that can limit the spill volume to less than 1,000 barrels (about
four percent of average hourly throughput) would conceivably eliminate
approximately 72 percent of the spill volume.
279
-------
TABLE 66. SPILL REDUCTION ANALYSIS FOR VOLUME OF RUPTURES
FOR TYPICAL SECTION OF LINE PIPE
Inspection or Leak Detection Method
1. Visual Inspections by air or ground patrol for
indication of a spill (Required inspection—
existing MtMd)
2. Visual inspections by air or ground oatrol In
encess of required inspections
3. Visual inspection by air or ground patrol with
hydrocarbon probe or comrable device
Indication of a spill
FffJQIMflcy/
rear
2S
26(Ref)
338
26
338
Ruptures' "'
J Reduction of
Spill volume
Indicated i Single
Inspections! Inspection
0 0
9 O.I9(R.f)
40 0.12
9 0.19
40 0.12
Risk
Factor*
1
0.95
0.6
0.95
0.6
Seduction
Normal tied'
Nft
l.O(Ref)
0.6.
1.0
0.6
4. Hydrocarbon prooe-towflsn or staiilar device
- UMenMter lines only
- Total line pipe «ileage
i. Oil spill detectors
- Marine terminal or platfora
- Total line pipe •iloaoa}
6. Pressure deviations
a. Puii) station (nisttng wtkod)
t. HaKf Une
7. F1« rate
a. Deviations
B. Canaarlsoin (ocwterized)
a. VotuM conarlsom (coeouteriied)
9. »oie»it1cal "odellng (coaputeritM)
10. Hegetive pressure surge (coeguterlted)
11. fiulve acoustic array (computerized)
12. Eiternel rods «lt» passive acoustic sensors
1). Pressure static
14. Hydrostatic
IS. Pressure difference
16. !!(•) and 13
17. 11(6) and 13
18. «(a). 7(b). *
19. <(a). 7(a). 7(o). a. 10
20. 10. 13
after Indi-
cation of
spill
after indi-
cation of
spill
continuous
continuous
continuous
continuous
continuous
continuous
continous
continuous
a. continuous*
b. eoHtinmui'
after Indi-
cation of
rupture*
after Indi-
cation of
metlire*
IB
«
continuous
continuous
continuous
continuous
continuous
I NO
!NO
30
[NO
0
10
20
40
70
90
33
SO
70
10
20
us
us
70
SJ
30
95
SO
[NO
[NO
30
[NO
0
10
20
40
70
90.
33
60
70
3.3
6.6
70
30
30
95
SO
[NO
[NO
0.7
(NO
m
0.9
0.8
0.6
0.3
0.1
0.7
0.4
0.9
0.8
us
NS
0.3
0.2
0.2
0.05
0.5
[NO
[NO
153
[NO
1U
S3
105
210
3M
474
174
315
36a
17.3
34.6
NS
us
122
237
4Z1
500
263
Nit-Net applicable since mathed MS cosamly used for yeer wwn risks xere determined.
NS-Hot suitable for indicated detection of leaks or ruptures.
3. lNO-lm»u»i»ent not discernible.
«. Msk of oil spills after applying CM Inspection otMd is CM product of t«U factor and tne risk
: and cosssgnly used Mtliods. The value is obtained by subtracting the percent n
from 1.0.
i. Risk redaction noramllted is the percent reduction froe a single inspection divided by the percent i
resulting free OM reference inspection.
i. Retrofit Installation.
7. Installation on new line.
a. Assam frequency of Inspection carried out three tines a year.
value «1th
eduction
280
-------
Bimonthly visual line inspections, the only major inspections required
by U.S. regulations, ar-e ineffective in reducing the volume of rupture type
spills. At a spillage rate of 125 barrels per hour, 42,000 barrels of oil
could be spilled between biweekly inspections or 21,000 barrels between
weekly inspections. Even for more frequent inspections, i.e., daily, 1,500
barrels of oil typically would be spilled before detection. Considering
that spills in excess of 1,500 barrels account for about 70 percent of the
total spill volume, reductions in spillage would be expected by more fre-
quent visual line inspections. However, many spills of this magnitude are
detected by outside observers. Thus, more frequent inspections are not as
effective as indicated; a risk reduction factor of 0.6 was assessed for
daily inspections and 0.95 for weekly inspections. Aided bimonthly visual
inspections such as using a hydrocarbon probe or other comparable device are
not expected to produce any additional reduction in the spill volume that is
discernible.
Although there is a trend towards continuous monitoring of lines using
advanced computerized methods, most operating companies only monitor pres-
sures at pump stations for indications of ruptures. This method is extreme-
ly insensitive and is capable of detecting only rupture type spills close
to the pump station; the sensitivity of the method decreases with distance
from the pump station. For this method, it has been estimated that, on the
average, ruptures resulting in 20 percent of throughput could go undetected
for greater than one hour.
Monitoring of pressure deviations along the line reduces the problem
of decreased sensitivity with distance. The method is considered only mar-
ginally adequate for rupture detection; a risk reduction factor of 0.9 is
assessed.
Simple monitoring of flow deviations at one or more stations for indi-
cations of rupture is often used by operating companies. Line transients
generally require that the alarm setpoint be quite high, i.e., five to ten
percent of throughput to eliminate false alarms. The method is considered
only marginally adequate for rupture detections; a risk reduction factor of
0.8 is assessed.
An approach to significantly reduce the volume of ruptures is to use
one or more of the methods that employ computerized continuous monitoring.
Based on the information supplied in Table 66, limiting the spill size to
less than 1,000 barrels appears to be relatively easy to accomplish using
many of the computerized methods. However, the various transients that oc-
cur in the line, such as line pack, pressure changes, flow changes, pump
shutdown, etc., make this somewhat more difficult to accomplish. These
transients are often accounted for by various measures such as lengthening
the time duration of monitoring for a leak alarm, raising alarm setpoints so
that the sensitivity is compromised, or using a computerized model that
automatically monitors and accounts for these variations.
The mathematical modeling method currently under evaluation on a number
of lines, has the potential of being the most effective of all methods
281
-------
surveyed. Despite transients, the method appears to be capable of detect-
ing leakage rates of 0.5 percent of throughput in about two minutes. This
would result in a loss of less than one barrel for a typical line before a
leak could be detected and verified. This method appears capable of de-
tecting essentially all rupture type spills; a risk reduction factor of 0.1
was assessed.
The flow rate comparison method is also effective in detecting rupture
type spills. The method requires that high accuracy meters (0.02 to 0.1
percent of flow) be installed at pumping stations and at inlet and outlet
locations. A leak between any two stations results in a characteristic in-
crease in flow rate upstream and a decrease downstream. Because of the
unique and characteristic rupture signal, the method is unaffected by many
line transients. The method is capable of detecting short-term leaks of
less than four percent of throughput in about two minutes. Primarily be-
cause the method appears capable of detecting most rupture type spills, a
risk reduction factor of 0.6 was assessed.
The computerized volume comparison method continuously monitors the
volumes at the input and output of the line. Corrections are normally made
for pressure, temperature and other factors. Modern systems incorporating
this method routinely detect ruptures of large-sized breaks at better than
three percent of throughput over a ten-minute interval. This method is con-
sidered to be an excellent means of indicating "rupture type" spills; a risk
reduction factor of 0.3 was assessed.
Negative pressure surge method is insensitive to many large-sized breaks
but should be capable of rupture detection. A risk reduction factor of 0.66
was assessed.
The passive acoustic array is an experimental method that has the dual
capability of rapid detection and accurate location of rupture or large-
sized breaks. Sensitivity, reliability and false alarms are uncertain.
Considering that the method is unproven on an operational line, a risk
reduction factor of 0.3 was assessed. Potentially, a risk reduction factor
of 0.1 is possible, primarily because of its ability to locate the rupture.
Two methods in use on a few lines are particularly useful after a rup-
ture or large-sized break is indicated by other means. Passive acoustic
sensors installed on rods permanently attached to the line would be a great
benefit in high risk areas because of the capability of locating the spill.
However, for use on a typical line, a risk reduction factor of only 0.9 is
warranted. The pressure static method is particularly well suited for veri-
fying the existence of a large leak which may be slightly below the sensi-
tivity of the leak indication method. A risk reduction factor of 0.8 was
assessed for the latter method.
Combinations of leak detection and inspection methods are capable of
affecting an improved reduction in the risk of rupture type spills. A few
of the more effective combinations are indicated for methods 16 through 20.
in Table 66.
282
-------
In summary, only a few methods are judged to have the potential for
significantly reducing the risks, i.e., have a risk reduction factor of
better than 0.5. These include:
• Mathematical modeling
• Passive acoustic array
• Volume comparison
• Combinations of pressure, flow and volume comparisons methods
• Pressure deviations, flow rate, volume comparison and negative
pressure surge.
Based on normalized risk reduction, the mathematical model method appears
most effective of those selected. One such inspection is estimated to be
equivalent to 474 visual inspections of the line. The volume comparison
method and passive acoustic array are also considered effective; these are
estimated to be equivalent to 368 visual inspections of the line.
7.1.2.1.2 Estimating the Capability of Reducing Leak Volume Using Selected
Methods
Selected methods for reducing the volume of "leak type" spills (less
than 400 barrels) are shown in Table 67 along with frequency of inspection,
percent reduction of spill volume, risk reduction factors and normalized
values for risk reduction. A single inspection in excess of the required
bimonthly visual line inspections is the reference inspection. Biweekly
visual line inspections and periodic monitoring of the cathodic protection
systems are assumed to be the commonly used methods for detecting leaks.
For leak reduction, continuous monitoring methods are evaluated primarily on
the basis of the product of long-term response time and sensitivity, peri-
odic methods are evaluated primarily on the_basis of sensitivity and fre-
quency of inspection.
In the analysis, leakage of less than 400 barrels is assumed to account
for about ten percent of the reported volume of spillage. Line pipe leakage
normally is assumed to occur continuously at very low rates (less than a few
barrels per hour). However, leaks may often occur intermittently. Reference
78, for example, indicates that for some lines leak rates of approximately
0.7 barrels per hour are required for hair crack type leaks to stay open.
Hence, intermittent leaks are also considered in this analysis.
Bimonthly visual line inspections are considered to be marginally ef-
fective in reducing the volume of "leak type" spills. For example, a con-
tinuous leak of two barrels per hour could spill 336 barrels before detec-
tion by biweekly visual inspections.
Assuming the leakage is normally continuous, more frequent visual line
inspections would be expected to reduce the volume lost by an amount almost
proportional to the inspection frequency. Thus, weekly inspections are
judged to reduce the spill volume by about 30 percent; a risk reduction
283
-------
TABLE 67. SPILL REDUCTION ANALYSIS FOR THE VOLUME OF LEAKS
FOR A TYPICAL SECTION OF LINE PIPE
I.I.I
; LEAKS
Inspection or Uak Detection Method
!
frnofrxyf |
y«ar ',
1. Visual Inspections By air or ground patrol for 25
Indication of > spill (Required insptction--
existlng method) '
2. Visual inspections By air or ground patrol in t. 2((Ref) ;
«Ctu of required Inspections s JM
3. Visual inspection By air or ground patrol with a. 2S(Ref)
hydroeiroeii proM or coraaraole dtvici h »
Indication of a soill , "
;c. 338
4. Hydrocarbon probe-tmrlsh or similar device
- Underwater lints only
- Total tine pipe mileage
5. Oil spill detectors
• NaHne terminal or platform
- Total KM pip* miMit
6. Prtssur* dtvittlom
a. Plav stltiom (Kilting nttlwd)
B. Along Hi*
1. flat ran
a. Otyiatiom
B. Coawrltam (comitirlzM)
8. VOIUM comnsons (coiouttrltid)
9. mtnmtlul aoMlinf (cMpuwHitd)
10. Ittjjtlv* prasnm turn (cowttrind)
S
s
continuous
continuous
continuous
continuous
continuous
contlnous
continuous
continuous
continuous
11. Pmtw s cuaiuuly ustd 'or ytar tKt1on of leaks or njptum.
3. llB-li»ii
-------
factor of 0.7 was assessed. For daily inspections, a risk reduction factor
of 0.3 was assessed.
Use of visual aids, such as hydrocarbon probes or other comparable de-
vices, by ground patrol are expected to further reduce the volume spilled.
These inspections provide the potential for detecting very small or inter-
mittent spills or spillage that may be trapped and not reach the surface
for a long period of time. A risk reduction factor of 0.8 was assessed for
aided biweekly inspections by ground patrol, 0.6 for weekly and 0.2 for
daily inspections.
Bimonthly periodic inspections of underwater lines using hydrocarbon
probes in a towfish system are expected to be quite effective. These in-
spections are expected to detect small leaks that may not be detected by
visual inspections. For example, small leaks that might otherwise be con-
sidered as underground seepage could be detected and located. A risk re-
duction factor of 0.5 was assessed for these inspections. Since only ap-
proximately ten percent of line pipes are underwater, the risk reduction
factor assessed on a nationwide basis for all lines is 0.95.
Continuous monitoring of internal fluid variations during fluid trans-
fer using pressure, flow rate, and negative surge are not considered suit-
able for leak detection. Normal operations generally produce line transients
that require an operator to raise alarm levels so as to preclude the de-
tection of small to medium sized leaks, i.e., few barrels per hour. The
volume comparisons method is only marginally capable of detecting leaks and
only under optimum conditions. These methods are expected to provide no
discernible improvement over biweekly visual inspections by ground patrol.
Of the remaining continuous monitoring methods, only mathematical
modeling and passive acoustic array are considered to be capable (margin-
ally) of detecting medium sized leaks. Risk reduction factors of 0.8 were
assessed~for these methods.
The method employing rods with passive acoustic sensors has poor sen-
sitivity for leak detection. However, it can be used to reduce the time
required to locate an expected leak. Thus, a risk reduction factor of 0.9
was assessed.
Periodic pressure testing methods, unless carried out often, are not
particularly beneficial. The pressure difference method is the most sensi-
tive of the inspection methods. However, frequent tests, i.e., monthly,
would be required to substantially reduce leakage. A risk reduction factor
of 0.1 was assessed for such frequent inspections. This method would be
particularly beneficial in certain high risk areas where even minute spills
can be hazardous. The pressure static method, by itself, would not appre-
ciably reduce leak spill volume because of its poor sensitivity. However,
after a medium sized leak is indicated by some other means, this method is
effective in verifying the existence of a leak. It was assessed a risk re-
duction factor of 0.8 assuming it would be used in conjunction with mathe-
matical modeling or the passive acoustic array. Hydrostatic tests carried
out even on a frequent basis are not expected to substantially reduce the
285
-------
leak spill volume. The method, however,_wou Id be quite effective in veri-
fying the existence of small to medium sized leaks. Risk reduction factors
of 0.6 and 0.7 were assessed for the two inspection schedules given in
Table 67.
Combinations of leak detection and inspection methods are capable of
affecting improved reduction in the risk. A few of the more effective com-
binations are shown in Table 67.
Only a few methods appear to be highly effective in reducing the volume
of leak type spills. Of all the methods, the pressure difference methods
appear to be the most effective. However, this inspection method requires
installation of closely spaced block valves and also requires line shutdown
during the inspection. Frequent visual and visual-aided line inspections
appear to be highly effective methods in reducing the volume of these oil
spills. However, almost daily inspections are required to produce risk
reduction factors better than 0.5. Twice yearly hydrostatic tests with the
passive acoustic array are judged to be effective. The main reason is that
the acoustic array can be used to locate the leaks. This is particularly
important during hydrostatic tests on long lines where a number of small
leaks may be erroneously accounted for by temperature or other effects.
Hydrocarbon probes may be substituted for the acoustic array if the line
is filled with petroleum or gas detectors or analyzers may be substituted
if the line is pressurized with helium of some other inert gas.
Since only about ten percent of the spill volume is attributed to leak-
type spills, no method presented can affect a significant reduction in the
spill volume as compared to the methods identified for rupture type spills.
Because even leak type spills can be serious, particularly in.high risk
areas, the volume of these spills should be reduced as much as possible.
7.1.2.2 Factors Affecting the Capability of Methods to Reduce the Fre-
quency of Spills--
The capability of a method to reduce and, if possible, eliminate the
occurrence of spills depends upon the ability of the method to detect im-
pending failures (internal defects) sufficiently early so that pipe line
repairs can be made and a spill prevented. Various methods can be imple-
mented to detect pipeline defects that may result in a spill. The following
three main factors affect the capability of these leak detection and inspec-
tion methods detect impending failure and thereby reduce the frequency of
spills:
• Sensitivity (detectable defect size)
• Location accuracy (pinpointing defect)
• Frequency of inspection.
The effect of each of these factors varies widely depending upon the imple-
mentation of each method. It also depends'upon whether the method is used
for reducing the frequency of leaks or ruptures or the frequency of spills
by cause. Thus the combined effects of these characteristics along with
286
-------
risk reduction factors3 and normalized values'3 for risk reduction. Spill
analysis risk reduction is based on the capability of an individual method
or combination of methods to reduce the oil spill risk below the value estj-
mated for the methods in common use on most lines for the reference year
1975. Inspection and leak detection methods for operational lines that are
assumed in the analysis for the reference year are:
• Biweekly visual inspections by air or ground patrol for
indications of a spill—(Required by U.S. regulations)
• Periodic inspections of the cathodic protection system—
(Required by U.S. regulations)
• Monitoring of line pipe pressure deviations at pump stations
• Voluntary One-Call System.
To simplify comparisons of the selected methods, a normalized risk re-
duction value is provided. This allows one to compare the number of inspec-
tions by various methods to achieve a certain reduction in the oil spill
risk. The normalized value is based on a single reference inspection. One
visual inspection in excess of the bimonthly inspections by air or ground
patrol is used for this purpose. Results indicate, for example, that one
inspection by an inspection survey pig is equivalent to 136 visual inspec-
tions of the line for reducing the frequency of leaks by 52 percent.
7.1.2.1 Factors Affecting the Capability of Methods to Reduce the Volume
of Spills-
Rapid detection and location of the onset of small quantities of
petroleum pipeline leakage so that system shutdown can be initiated are the
four most important factors in reducing the volume of accidental spills.
The four major factors (method characteristics or specifications) for re-
ducing the volume of spill are specifically:
• Response time
• Leak location accuracy
• Leak sensitivity
• Frequency of inspection.
Risk of oil spills, after applying the inspection method, is the product
of this factor and the risk value with currently required commonly used
inspections. The value is obtained by subtracting the percent reduction
from 1.00.
Risk reduction normalized is the value of the percent reduction of the
spill risk for a single inspection divided by the percent reduction of
the spill risk for one reference inspection.
287
-------
these other considerations determine the overall capability of a method to
reduce the frequency of spills. Hence, each is considered in the evalua-
tion. Section 7.1.2.1 provides a discussion of the significance of these
factors.
Implementation of leak detection and inspection methods that are capa-
ble of satisfying these required characteristics is considered to be the
optimum approach for minimizing the hazards of accidental spills from
petroleum line pipe. In general, prevention of ruptures would normally be
of higher priority than the prevention of leaks. However, the objective
should be to reduce all spill incidents. At this time, U.S. regulations
for installation and periodic inspections of cathodic protection systems in
the line pipe and depth of cover on line pipe are the main U.S. efforts in
achieving reduction of frequency of spills. Although these methods are
quite useful, they have not resulted in significant reductions in the fre-
quency of spills, particularly those resulting in "rupture-type" spills.
The major emphasis in a spill prevention and control program should be
directed towards the goal of eliminating or at least minimizing the fre-
quency of spills. The reason for this conclusion is that methods that re-
duce the volume of spills (detect actual spills) do not stop spills. These
methods have two inherent limitations. First, even with rapid detection of
medium to large size spills, a certain amount of hazardous fluid can escape
from the line before the spill is totally contained. Secondly, many "leak-
type" spills and many "rupture-type" spills cannot be detected.
The sensitivity of the method is the most important capability in re-
ducing the frequency of spills. Sensitivity is based on the ability of the
method to detect minute internal defects that may lead to a leak or minutely
small leaks (drip type) that may lead to failure.
The frequency of inspection is also an important consideration. For
example, increased frequency of visual inspection of the line for the pur-
pose of eliminating third party activities such as excavations that might
result in damage to the line pipe, reduces the risk of a line pipe spill.
The capability of providing accurate leak location, i.e., within a
few feet is also important. For example, the ability to pinpoint a small
leak during hydrostatic tests greatly enhances the effectiveness of a method.
It is difficult to assess quantitatively the ability of a method to
reduce the frequency of spills, for a number of reasons. These include:
• The geometry and dimensions of an interval defect that will
result in a spill incident are not precisely known.
• The geometry and dimensions of leaking areas are not reporta-
ble. Thus, information as to minimum or typical hole size
that would result in a leak or rupture type spill is not pre-
cisely known.
• Quantitative data suitable for detailed comparisons of many
of the selected methods are not available.
288
-------
• Effectiveness of some inspection methods often depend upon
such factors as line pipe dimensions (i.e., diameter) and
location (i.e., on!and, underwater, etc.).
Nevertheless, sufficient information is available from manufacturers and
user experience (see Tables through and Appendix E) to make qualita-
tive estimates that are suitable for analysis.
For typical lines, the capability of a method to reduce the frequency
of spills by cause is of prime importance. For certain line pipe, analysis
of the capability of a method to reduce incidents of spills based on type,
leak or rupture, can be particularly advantageous. For example, older lines
typically have high potential for leaks whereas large diameter new lines
have an inordinately high potential for ruptures.
Analysis of the reduction of the frequency of spills is carried out in
the subsections that follow. Analysis by cause and type (leaks, rupture)
are treated separately.
7.1.2.2.1 Estimation of Capability of Reducing Spill Incidents from All
Causes Using Selected Methods
Selected leak detection and inspection methods for reducing the fre-
quency of spill incidents for each of the major causes of spills and all
causes combined are presented in Table 68. Methods are shown along with
frequency of inspection, percent reduction of incidents, risk reduction
factors and normalized values for risk reduction. A single inspection in
excess of the required bimonthly visual inspections is the reference inspec-
tion. It is used in the computation of the normalized values for risk
reduction. Biweekly visual line inspections and periodic monitoring of the
cathodic protection system are assumed to be the commonly used methods dur-
ing the reference year for spill prevention.
Capability to prevent line pipe faults is based primarily upon the
overall ability of a method to detect insignificant leaks or line pipe de-
fects that are generally indicative of impending failure. Capability to
prevent damage by outside forces is based on the overall ability of a
method to detect activities of outside forces (usually impacts) before a
failure can occur.
The number of incidents for the causes of line pipe failures were
obtained from Table 19. These are:
• Line pipe faults - 56 percent of spills
• Outside forces - 41 percent of spills
• Other causes - 3 percent of spills.
These values are used in the computation of the percent reduction of spill
incidents identified under all causes. For example, a method that accounts
for a 50 percent reduction of spills from the line pipe faults but no other
reductions would account for a 28 percent reduction in the overall causes
of spills.
289
-------
TABLE 68. REDUCTION ANALYSIS FOR THE FREQUENCY OF OIL SPILLS BY CAUSE
FOR A TYPICAL SECTION OF LINE PIPE (REFERENCE LINE)
I
Inspection Or Leak Detection Method
1. Vlsiul Inspection by >ir or ground
patrol (Seoul red Inspections)
Z. Visual Inspection by air or ground .
Mini in excess of required h
Inspection*
niptctlom
ronmey/
Yttr
2S
. 26 !D1pt
FaulU
(S6t of
Spills)
0
OutHdt
Forcas
(405 of
Spills)
0
5 | JO-
ll) : 80
Outer
(3 -. of
Spills)
0
[NO
INO :
AH Causes
Indicated
nsaectlons
0
u
38
30 . [NO !NO ib
35 20 '.HO 28
40 80 ! :iO 55.
I
1
SO IW
5 INO
I!0
IW
I HO
IK)
WO
10
SO
60
70
90
10
w
M
7S
ss
10
10
80
90
9t
30
10
mo
[NO
70
90
IW
IW
IW
I»
[Ml
[in
I HO
im
IW
INO
so
90
90
95
[NO
INO
3
0
[NO
INO
MO
IW
IW
I NO
INO
IW
in
IW
INO
IW
IW
[HO
IW
IW
IW
IW
28
2.3
32
4.1
3.4
[NO
2*
43
28
34
39
SO
s.s
u
37
42
47
S.S
30
32
37
92
Single
nsoeetlon
a
0.42(Ref
3.11
3.2
:.os
0.16
».7
3. 47
a
i
5.4
[NO
28
43
28
34
39
12. S
1.9
11
37
42
47
1.9
30
32
0.37
3.54
Disk
Reduction
Factor* It
HA
0.39
0.62
0.34
0.72
5.45 j
Risk
Deduction
braallted'
NA
1.0 (Ref)
0.3
7.62
2.52
0.38
1
0. 72 11.2
0.97 i !.2
0.63
0.96
0.9
INO
0.72
0.57
0.72
0.66
0.61
0.50
0.95
0.89
0.63
O.S8
0.5}
0.95
0.70
9.18
0.13
0.08
19
2.4
20
IW
67
107
57
40
n
46
4.5
26
88
100
112
4.5
71
195
207
8.43
Notei: (l) m - Mt aooltcable since netnod M« convmly used «ien soil! frequencies «ere determined,
(2) IW - uionwtvent not discernible.
(3) Deduction of spill incidents, after applying inspection and/or leak detection iteund is cite sun of tfle produces of tne reduction
factors and Uw percentage of spills resulting from casn cause (line oipe - Sts, outside forces 41:. other 3D ?or cue reference!ins.
(4) Disk of spills, after applying Oie inspection method is tne product of Ulis factor and tne risk «alu< oitn current required and
ccneonly used inspections, cue «akie is obtained by subtracting tne '. reduction fro i.o.
(5) Risk nonialKed is the percent reduction fro» a single inspection divided by tne percent reduction resulting fro* one reference
Inspection..
Retrofit Installation.
He* line Installation.
1:1
Altwe freeuency of Inspection carried out three ti«n a rear.
290
-------
Biweekly visual inspections carried out by air or ground patrol are
considered ineffective in preventing spills, even spills caused by outside
forces. Analysis of spill statistics (see Figure 22) indicates little or
no change in the reported incidents by outside forces, such as equipment
rupturing the line, between 1970 and 1976. Spill prevention by this in-
spection method is considered to be minimal.
Periodic inspections of cathodic protection systems are considered to
be responsible for some reduction in the spill incidents from external
corrosion. Data such as presented in Figure 22 indicate substantial reduction
in external corrosion between 1970 and 1976. Some of this reduction can be
attributed to cathodic protection systems. It should be noted, however,
that external corrosion generally results in the smallest spill size for the
major causes of spills (see Table 19). Thus, a significant reduction in
external corrosion does not affect a significant reduction in the total
volume of petroleum spilled. Also, these cathodic protection systems are
ineffective for reducing the incidence of internal corrosion and other major
causes of spills.
Visual inspections by air or ground patrol carried out on a weekly
basis are expected to reduce incidents of damage by outside forces by
about 20 percent and line pipe faults by about five percent. For these
26 additional inspections, a risk reduction factor of 0.9 was assessed.
For daily inspections, a risk reduction factor of 0.6 was assessed.
Use of visual aids, such as hydrocarbon probes or other comparable
devices, by ground patrol is expected to improve the detection of insig-
nificantly small leaks that may result in a reportable spill. Respective
risk reduction factors of 0.84, 0.72 and 0.45 were assessed for aided in-
spections by ground patrol that are carried out biweekly, weekly and daily.
Bimonthly inspections of underwater lines using hydrocarbon probes or
other equivalent devices in a towfish system that traverses the line is
expected to be quite effective in detecting and locating small leaks that
occur from corrosion and weld defects. These inspections are expected to
detect very small leaks that may lead to failure and that could not be de-
tected by other means. A risk reduction factor of 0.72 is assessed for
underwater lines, and a factor of 0.97 when all line pipe mileage is in-
cluded. Visual inspections at the surface for underwater lines are con-
sidered to provide no discernible prevention of line pipe faults.
Quarterly inspections of the line pipe for the depth of cover may re-
duce significantly the frequency of ruptures. Since most undersea pipeline
ruptures are caused by external impacts such as anchor dragging, a proper
depth of cover will minimize the potential of damage from these sources.
These inspections are judged to reduce the incidents of spills from outside
forces by about 30 percent. The overall risk reduction factor is 0.68.
For all line pipe, these surveys were judged to have a risk reduction fac-
tor of only 0.97.
291
-------
Charting the movements underwater lines enables one to detect abnormal
movements which could excessively stress the line. Inspections every two
years were judged to reduce the incidents of line pipe faults by about 15
percent; a risk reduction factor of 0.91 was assessed for underwater lines
and negligible risk reduction factor when all lines are included.
Use of a continuous monitoring passive acoustic array is expected to
be highly effective in reducing the incidents of damage by outside forces
and slightly effective in detecting impending line pipe failures, primarily
ruptures. A risk reduction factor of 0.57 was assessed for this method.
Yearly hydrostatic tests of the line pipe were judged to be capable of
detecting a majority of serious line pipe faults, i.e., minute leaks. For
this method, an overall risk reduction factor of 0.66 was assessed. If used
with the acoustic array, a much greater effectiveness can be achieved since
the line pipe response to changes in stress levels can be monitored. The
risk reduction factor of 0.18 is assessed for the combined methods.
Periodic pressure difference tests are judged to be the most effective
inspection method for detecting small leaks that may lead to failure. For
inspections carried out yearly and quarterly, risk reduction factors were
assessed at 0.61 and 0.50, respectively.
Yearly inspections of the condition of the interior of the line pipe
using inspection pigs are considered to be the best method overall for de-
tecting line pipe faults. Commercially available devices, such as equip-
ment from AMF Tuboscope, Inc. and Vetco, Inc. can be used in lines ranging
from six to 36 inches in diameter. A risk reduction factor of 0.53 was
assessed for yearly inspections with this device.
The prevention program for outside forces, the One-Call System, as
recommended in Reference 12, is considered to have the potential of signifi-
cantly reducing the damage by outside forces. An overall risk reduction
factor of 0.70 was assessed for this program.
Pigging the line for water removal is an important inspection/mainte-
nance activity that should be carried out on a frequent basis. This method
reduces the risk of internal corrosion and other causes of line pipe spills.
Since the method is often used on many lines, it has affected some reduction
in the spill risk. Thus, a risk reduction factor of only 0.95 was assessed.
Inspections of sample sections of the line for wall thickness changes
caused by corrosion can be effective in reducing the risk of oil spill in-
cidents, particularly for lines with corrosion problems. It is only prac-
tical to sample a small area of a line. This small sampling is judged
to be inadequate and thus no significant reduction in the risk can be at-
tained; a risk reduction factor of only 0.89 was assessed.
Combinations of leak detection and inspection methods are capable of
significantly reducing the risk of spill incidents. A few of the more ef-
fective combinations are shown in Table 28. Pigging the line for preven-
tion of line pipe faults used in conjunction with a One-Call System program
292
-------
to prevent outside forces is expected to produce an excellent risk reduc-
tion factor of 0.13. If weekly visual inspections added by the use of
hydrocarbon probes are also carried out, the risk reduction factor further
improves to 0.08.
7.1.4.2.2 Estimation of the Capability of Reducing the Frequency of Leaks
Using Selected Methods
Results of a reduction analysis for the prevention of "leak type"
spills are given in Table 69. Selected methods and major causes of spills
are the same as in the previous section. However, the values indicated in
the table are specifically for the reduction of "leak type" spills. These
values have been based on two estimates9. First, "leak type" spills are
considered to account for 75 percent of the reported accidental spills.
Second, the major causes of spills are expected to account for the follow-
ing incidents of leaks:
LEAKS (75 percent of all spills)
• Line pipe faults - 61 percent of leaks (46 percent of all spills)
corrosion - 52 percent of leaks (39 percent of all spills)
defective pipe seams, welds, etc. - 9 percent of all leaks
(7 percent of all spills)
• Outside forces - 35 percent of leaks (26 percent of all spills)
equipment rupturing line - 28 percent of all leaks (21
percent of all spills)
miscellaneous - 7 percent of leaks (5 percent of all spills)
• Other - 4 percent of leaks (3 percent of all spills).
Only a few methods appear to have the potential of significantly re-
ducing the line pipe faults which account for most (over 61 percent) of the
incidents of leaks. One method, survey by inspection pig, appears to be
the most promising. Other methods judged to have the potential of signifi-
cantly reducing line pipe faults include:
• Hydrocarbon probe with towfish for underwater lines
• Hydrostatic pressure testing
• Pressure difference pressure testing.
In preventing damage by outside forces, only these methods appear to be
effective. These include:
• Acoustic array
• Depth of cover inspection of underwater lines
• Frequent, e.g., daily, visual inspections of the line
• Prevention program for outside forces--One-Call System.
aThese are rough estimates and suitable for the evaluation purposes in this
study only.
293
-------
TABLE 69. SPILL REDUCTION ANALYSIS FOR THE FREQUENCY OF LEAKS
FOR A TYPICAL SECTION OF LINE PIPE
I
Inunction or Leak D*t*ct1m Mttlnd
1. iMiuil Inspection by air or ground
Htral (KtoulrM Inspections)
Z. Vlutl Inspection by air or ground
patrol In excess of required :
Inspections *
nsptctlOM
•rtoumcy/
Ytar
26 :
. 25(Rtf)
. 338
3. »1sual Inspection by ground patrol a. 26
[NO
a
0
:m>
[no
im>
uto
INO
I»
IW
IM)
INO
IW
IW
IW
IW
IM)
IW
INO
IW
IW
IW
30.5
1.0
28
2.3
9.2
INO
21.5
37.6
30.5
36.6
42.7
67.3
6.1
12.2
40.3
45.8
51.8
6.1
27.1
30.3
36.4
91.2
36.4
SlmjU
nsptctlon
a
0.38(Ref
0.10
3.66
1.04
0.13
5.08
.51
7
.7
9.2
IW
21.5
37.6
30.5
18.3
42.7
17
2.03
12.2
40.3
45.3
51.3
2.03
27.1
30.3
36.4
3.51
36.4
Risk
Reduction
Factor* H
(«RFaj ;
W
Risk
Reduction
arnaltitd*
ill Causes
Single
nsptctlon
1A
0.90 l.O(Ref)
0.66 1 0.3
0.32 9.63
0.72 ! 2.37
a. 55 ' 0.34
0.70 !
0.97
0.72
0.97
0.91
IW
0.79
0.62
0.70
0.6T
0.57
0.32
0.94
0.88
0.60
0.54
0.48
0.94
0.73
0.20
0.14
0.09
0.14
13.1
1.3
18.4
1.3
24.2
IW
57
99
-.0.6
48.6
112
44.7
5.34
32.1
106
120
136
5.3
71.3
21 1
227
9.2
227
Notes: (i) *A . not applicant* since method M«S coomnly ustd vh«fl spill frequencies «ere d*temhn«d.
(2) INO - ImproveMent not discernible.
(3) Reduction of SQill incidents, after aoplying. imptctton and/or leak detection method is the sum of tne products of the reduction
factors and the percentage of spills resulting fron cash cause (Tine pipe - 61'. outside forces 35*. other 4?) for tne reference Hr>
(*) Risk of spills, after applying the inspection method is the product of this factor and tne risk value with current required and
coMonly used inspections, tne value is obtained by subtracting tne • reduction fron 1.0.
Risk nonNlizeti is the percent reduction fro* « single inspection divided Oy trie percent reduction resulting from one reference
inspection..
Retro/It installation.
New line Installation.
(5)
(6)
(7) .
(8) AHUM frequency of inspection carried out three t1«et a year.
294
-------
These methods are most effective in reducing either line pipe faults or
outside incidents. Hence, no individual method appears to have the poten-
tial of eliminating most of the leak type spills. Only two methods are
judged to have risk reduction factors better than 0.5. Bi-yearly surveys by
inspection pigs are assessed at slightly better than 0.5. Frequent pressure
difference tests are assessed at 0.32. By using combinations of inspections,
however, risk reduction factors of better than 0.20 are potentially possi-
ble. For example, using the survey inspection pig and either a prevention
program for outside force or a passive acoustic array system are assessed
excellent risk reduction factors of 0.14.
Based on normalized risk reduction factors and single inspections,
five methods stand out for reducing risks. These include:
• Survey by inspection pigs
• Pressure difference
• Passive acoustic array
• Hydrostatic
• Prevention program for outside forces One-Call System.
7.1.4.2.3 Estimation of the Capability of Reducing the Frequency of Rup-
tures Using Selected Methods
Results of reduction analysis for the prevention of "rupture type"
spills are given in Table 70. Selected methods and major causes of spills
are the same as those identified in Section 7.1.4.2.1. Values indicated in
the table are computed using the same approach followed in the reduction
analysis of the frequency of "leak type" spills. "Rupture type" spills
are estimated to account for 25 percent of the reported accidental spills.
The major causes of spills are estimated to account for the following inci-
dents of rupture:
RUPTURES (25 percent of all spills)
• Line pipe faults - 40 percent of ruptures (10 percent of all
spills)
defective line pipe - 40 percent of ruptures (10 percent
of all spills)
• Outside forces - 60 percent of ruptures (15 percent of all
spills)
Equipment rupturing line - 40 percent of ruptures (10
percent of all spills)
Excavation equipment - 12 percent of ruptures (3 percent
of all spills)
Miscellaneous - 8 percent of ruptures (2 percent of all
spills).
295
-------
TABLE 70. SPILL REDUCTION ANALYSIS FOR THE FREQUENCY OF
RUPTURES FOR A TYPICAL SECTION OF LINE PIPE
Inspection or Ltik Dtttctlon Method
1. y
-------
Methods that appear most promising for reducing line pipe faults and
outside forces for "rupture type" spills are similar to those previously
identified for "leak type" spills. Whereas line pipe faults account for
most leak incidents, outside forces are the major causes of ruptures. Thus,
there are some differences in the risk reduction factors and normalized risk
values for the most effective methods for "rupture type" spills. Two meth-
ods judged to have risk reduction factors better than 0.5 are:
• Frequent, e.g., daily, visual inspections by ground patrol
with hydrocarbon detectors or other comparable devices
• Passive acoustic array.
By using a combination of inspections, risk reduction factors of better than
0.20 are potentially possible.
7.2 OIL SPILLAGE THAT POTENTIALLY CAN BE PREVENTED USING THE SELECTED LEAK
DETECTION AND INSPECTION METHODS
Analysis of the spillage that potentially can be prevented (or go unde-
tected) by implementation of the selected inspection and leak detection
methods is carried out in this section. From this analysis, it is possible
to estimate the expected reduction of spill incidents and spill volume for a
scheduled series of inspections. Estimates are based on the results of the
analysis of the oil spill risks in Section 6.3 and the capabilities of
selected methods to reduce the oil spill risks in Section 7.1.2.
The approach and necessary equations for estimating spillage are pre-
sented in Section 7.2.1. Sufficient information is provided so that quan-
titative values can be determined for most lines.
Values for spillage prevented using selected leak detection and inspec-
tion methods are computed and results discussed for the reference line in
Section 7.2.2. Spillage values are computed for all U.S. lines in Section
7.2.3.
7.2.1 Means of Calculating Spillage that Potentially can be Prevented
Quantitative values of the spillage that potentially can be prevented
(or go undetected) can be computed for most lines using the equations pro-
vided in Sections 7.2.1.2 through 7.2.1.4. The frequency and volume of
spillage can be estimated for all spills and also for "leak type" and "rup-
ture type" spills. Values for existing spillage that must be used in these
equations can be obtained from information provided in Section 7.2.1.1. The
risk factors for the various leak detection and inspection methods that also
must be used in these equations can be obtained in Table 71a.
aThis table summarized the risk reduction factors from Tables 65, 67, 69
and 70.
297
-------
TABLE 71. SUMMARY OF RISK REDUCTION FACTORS FOR SELECTED
INSPECTION AND LEAK METHODS
Oil S>111 Mtt Rtductlon Ftcton1"'1
Inspection uid/or Ltak Detection nttmos
1. V1SW1 Inspection of lint by «1r or
ground patrol (required Inspections)
for Indication of spill
2. Visual Inspection of line by air or
ground patrol in txcess of required
inspections 'or indication of 50111
3. yisual Inspection of I1n* or jround
Mtroi «IOi hydrocarbon probe or
cosmrable 3t»lct for indication of
jplll
4. HydraeirMn proM-CMflsn or jimlar
device
-Underwater llnM only
-All 11m*
5. f ratlin deviations
(.Pin* stations (eitstlno notnod)
b.Alo«« Ifnt
«. no. tatt
• .Oavlttlom
i.Co^artioin (comittrtM)
7. Mmm eo«wrtjont(a»p<.t«rltM)
3. mtlMHtlul adtitng (omtarlad)
9. Nfgitlw prntyr* surot (cowttrlztd
10. PtutM Mowttc S
O.TO e
o.so us
0.40 NS
n. m «
« NS
IS NS
NS NS
NS NS
NS NS
is e
NS NS
NS NS
NS «
(continued)
298
-------
TABLE 71 (continued)
20
21
22
23
24
25
26
27
28
29
Inspection and/or Leek Detection Methods
. Preventive program for outside forces
(Reference 12)
. P1g line for weter removal
. Oil spill detectors
-Herlne temlna! or platform
-Total line pipe «1!ea«e
. 10(b) and )3(a;
. 17(c) em) 20
. 3(a). 17(c) and 20
. HXb) and 17(c)
. 5(a). 6(a). 7
. 5(a). «(a). 6 (t). 7 and 9
. B end 17(c)
' Oil
Inspections i frequency
'-""" ; %%
cont. 0.73
3' 0.94
continuous IW
continuous INO
0.20
0.14
0.09
0.10
0
0
0.48
Spill Disk deduction Factors'"-'
Ruptures
0.60
0.96
1C
MS
0.14
0.10
O.OS
0.14
0
0
0.6C
Volume
Leaks
NS
•6
0.50
0.95
0.50
NS
o.co
0.80
1X0
IW)
0.80
Ruptures
NS
«
IB
NS
0.70
NS
0.95
0.30
0.20
O.OS
0.10
(1) M - Hot ippllciblt slnct nthod MI umaiiily uud for ynr «ftw ritkt wrt dtuminta.
(2) NS - Not lulUblc for indiuuo dttietlon of luki or ruptures. Thtrtfore < risk
reduction factor of 1 1s «ss1ontfl.
(3) [NO - Immanent not discernible. Tnerefore i risk reduction fictor of 1 is
assigned.
(4) Retrofit Installation.
(5) Installation of nex lines.
(6) Assune frequency of Inspection carried out tnree times a year.
299
-------
Particular attention should be given to the means of computing the
total volume of spills (Section 7.2.1.4) that can be prevented. Total
volume depends on the capabilities of a particular method to reduce both
parts of the risk, i.e., preventing spills and reducing the size of each
spill. This results in more complex equations than required in computing
spillage for methods that reduce only one part of the risk.
7.2.1.1 Computation of Existing Annual Spillage—Frequency, Volume and
Spil-1 Size—
The oil spill risk, total volume spilled V$TC and frequency F$TC» is
computed for the line using an approach similar to the one presented in
Section 6.3.4. First, the spill risk for the reference line, frequency
FST and volume V$j, are obtained from Table 57. Then values for correction
factors which are used to account for variations from the reference line
are obtained. Correction factors for the frequency CFp (age, geometry,
length, use, spill size) and volume CFyM (diameter, length, use, pumping
station shutdown, mainline valve closure, line elevation) are estimated
using Figures 54 and 57. Annual volume V$TC and frequency FSTC °f spills
from the line are then calculated using Table 59.
The same approach can be used in computing the nominal largest spill
size for ruptures V^i RC» leaks VNM c and mean spill size VJVJQ. Table 60,
however, is used instead of Table 59.
7.2.1.2 Computation for Reducing the Frequency of Spills—
The number of spills that can be prevented (or go undetected) can be
computed for the methods installed specifically for the reduction of oil
spill incidents. This can be done since the oil spill risk FSTC and the
risk reduction factors, RRFpi_ and RRFpR, for the methods are known. Assum-
ing 75 percent of the spills can be attributed to "leakage-type" and the
remainder to "rupture type," the following can be used to estimate annual
frequency of spills:
FLp = (0.75 x FST(.) (1 - RRFFL) (1)
FRp = (0.25 x FSTC) (1 - RRFFR) (2)
FP =FLP+FRP
FU " FLP ' FRP + RSTC
where
FSTC = total frequency of spills corrected for line pipe
F, p = frequency of leaks prevented after installation of leak
detection and/or inspection method(s)
300
-------
FRp = frequency of rupture prevented after installation of leak
detection and/or inspection method(s)
Fp = total frequency of spills prevented after installation of
leak detection and/or inspection method(s)
RRFp, = risk reduction factor for frequency of leaks after installa-
tion of leak detection and/or inspection method(s) - see
Table 71
RRFpn = risk reduction factor for frequency of ruptures after in-
stallation of leak detection and/or inspection installed -
see Table 71
F.. = total frequency of spills undetected after installation
of leak detection and inspection method(s).
7.2.1.3 Computation of the Reduction of the Volume of Spills Using Methods
that Reduce Only the Spill Size—
The value of volume of spills that can be prevented or go undetected
can be computed for the methods installed specifically for spill size re-
duction. This can be done since the oil spill risk for the volume spilled
VSTC and tne ris|< reduction factors, RRFyF and RRFvi_» for the methods are
known. Assuming that 90 percent of the spill volume can be attributed to
"rupture-type" and the remainder to "leak-type," the following can be used
to estimate annual volume of spills:
VLR = 0.10 VSTC (1 - RRFV|_) (5)
VRp = 0.90 VSTC (1 - RRFVR) (6)
Vp =VLp+VRp (7)
VU ' VLP = VRP + VSTC
where
VSTC = total vo^ume of spills corrected for line pipe
V.p = volume of leaks prevented after installation of leak
detection and/or inspection method(s)
VRp = volume of ruptures prevented after installation of leak
detection and/or inspection method(s)
RRFy, = risk reduction factor for volume of leaks after installa-
tion of leak detection and/or inspection method(s) - see
Table 71
RRFVR = risk reduction factor for volume of rupture after installa-
tion of leak detection and/or inspection method(s) - see
Table 71
301
-------
Vp = volume of spills prevented after installation of leak de-
tection and/or inspection method(s)
V.. = volume of spills undetected after installation of leak de-
tection and/or inspection method(s).
7.2.1.4 Computation of the Reduction of Volume of Spills for Methods that
Reduce Size and/or Frequency of Spills--
The volume of petroleum spillage that can be prevented (or go undetec-
ted) depends upon the capability of reducing both parts of the spill risk,
i.e., preventing spills or reducing the size of each spill3. As indicated
in the previous section, methods that reduce spill size cause the total
spill volume to decrease. Also, methods that reduce spill size often affect
some reduction in the frequency of spills. However, methods that reduce the
frequency of spills also perform a secondary function; they eliminate the
spillage volume that would have occurred if the spill had not been prevented.
Generally, a spill prevention and control program would be expected to
include a combination of methods that result in the reduction of both the
frequency and volume of spills.
The following equations can be used to compute the total spillage pre-
vented Vjp and spillage undetected Vju for one or more methods that reduce
one or both parts of the oil spill risk:
VSMC = VSTC/FSTC (9)
VTRp = VSMC x FSTC (1 - 0.9 x RRFVR x RRFFR - 0.1) (10)
\i — \i v C M n 1 v BDF Y BPF -DCn M "M
VTLP " VSMC x FSTC U ' °a X RRFVL x RRFFL °'9; U1)
VTP = VTRP x VTLP = VSMC = VSTC (1 " °'9 x RRFVR
x RRFFR - 0.1 RRFVI_ x RRFF1_) (12)
VTU = V$TC - VTp (13)
where
VCMP = Mean spill size corrected for line pipe
oMv,
= Total volume of spills corrected for line pipe
= Total frequency of spills corrected for line pipe
VTRp = Total annual volume of ruptures prevented after installa-
tion of leak detection and/or inspection tnethod(s)
aln the previous approaches for computing the reduction of the frequency and
mean volume of spills, methods were assumed to reduce only one part of the
risk.
302
-------
VTP = Total annual volume of spills prevented after installa-
tion of leak detection and/or inspection method(s)
V,... = Total annual volume of spills undetected after installa-
tion of leak detection and/or inspection method(s).
As an example, a method may have the following risk reduction factors:
RRFyR =0.5
RRFVL =1.0
RRFFR =0.5
RRFFL « 1.0.
Assuming VSMC = 1000 barrels/spill and F^JQ = 100 spills/year, then:
VSTC = 100° barrels/year x 100 spills/year = 100,000 barrels/year
VTD = 1000 barrels/year x 100 spills/year x (1 - 0.9 x 0.5 x 0.5
IK - 0.1 x 1 x 1) - 100,000 barrels/year (1 - 0.225 - 0.100)
= 67,500 barrels/year
V,.. = 100,000 barrels/year - 67,500 barrels/year = 32,500 barrels/
year.
7.2.2 Estimated Existing and Potential Reduction of Spillage for the
Reference Line
Accidental spillage (incidents and volume) from the reference line that
can be prevented (or go undetected) is estimated based on information pre-
sented in Section 7.2.1. Results for the selected methods are given in
Table 72. Estimates are included for leaks and ruptures. In order to pro-
vide a general comparison of the capabilities of the selected methods,
spillage that can be prevented (or go undetected) is indicated in percen-
tages rather than barrels spilled or number of incidents.
Only a few of the selected methods appear to have the potential of
substantially (33 percent or better) decreasing both the incidents and the
volume of accidental spillage. These include:
• Visual inspection of the line
daily by air or ground patrol
daily by ground patrol with hydrocarbon probe or
comparable device for indication of the spill
• Passive acoustic array
• Pressure difference
• Survey by inspection pig
303
-------
TABLE 72. ANALYSIS OF OIL SPILLED FROM THE REFERENCE LINE
THAT CAN BE PREVENTED OR GO UNDETECTED
Inspection and/or Leak Detection MetMds
SplUaoe
i
Inspection L!ikl
Frequency/ f'J>
rear .10.
1. Visual line inspection by air or ground 26 ! 0
patrol (required Inspections) far
indication of spill
2. Visual line inspection by air or general i. 26 (Ref) 22
patrol in excess of required inspection . .. ... ,.
for indications of a spill | "• 3M "
3. Visual inspection by ground patrol »itn ' a. 26 Oeq) 39
hydrocarbon probe or comparable device : K .. ..
for indication of spill (onland lines *• » •»
only c. 338 , 98
4. Hydrocarbon probe-towfish or similar
device
- Unaarwter lines only
- All lines
5. Pressure deviations
a. Puep stations (existing net nod)
b. Along, line
6. F1o» rate
a. Deviations
b. Comparisons (comouteritedl
7. volume comparisons (comterlteO
8. Matlosetlcel modeling (computerized)
9. negative pressure surge (computerized)
10. Passive acoustic array (computerized)
a. Retrofit
b. Hex line
11. External rods vith passive acoustic
sensors
12. Pressure static
13. Hydrostatic
6 «
6 ' S
continuous ! M
continuous 0
continuous j 0
continuous : 0
t
continuous 0
continuous 0
continuous 1 0
1
a. continuous 1 46
b. continuous ' 32
after indication 0
of spill1
after indication 0
of spill1 :
a. 1 85
b. 2 69
c. after indication 80
of spill' i
Prevented 1
Spills
Ruptures^ Total
.10. ' NO.
0 0
10 32
35 106
! 9 48
i 19 90
i 47 MS
1 7
1 I 7
M M
0 0
1
0 | 0
•oo
0 0
0 0
' o i o
1
25 I 71
: 42 i 124
o ! o
0 0
IS 80
It 88
i 22 102
t
volute
Leaks ;
''TU>
( bbls
Ruptures i Total
V : tf
*TRP *TP
K bbls i K bbls
0 0 ! 0
U 40 57
23 186 209
9 39 48
17 i 78 95
26 | 206 232
2
2
0
o
0
0
6
10
10
M
27
53
10S
183
23S
< 0 | 80
: 6
14
194
230
12
12
M
27
S3
10S
183
241
90
200
244
3 i 26 24
6 i 52
9
17
17
52
63
78
58
61
80
95
Undetected
Spills j Volume
Total Total
fu ''u
No. ic bbls
290 | 291
258 233
184 31
242 243
219 198
MS 58
I
22 j 17
283 \ 277
s
M i M
290 | 264
I
290 ! 237
290 | 185
290 ( 107
290 : SO
290 | 211
219 92
166 47
290 ! 262
290 : 233
I
210 ; 230
208 ! 211
188 | 196
[continued)
304
-------
TABLE 72 (continued)
Inspection and/or leak Detection Methods
Inspoctlon
Frtflotncy/
»oar
14. Pressure difference >. 1
; b. Z
[ C. 4
d. 12
Suilljoe
Spills
Ltaks
ri>
no.
93
108
148
152
RuOturtl
FRP
Ho.
20
23
26
29
15. ChMO* or add Inhibitors « needed :* 13 J
16. Inspection at sarnie of line for Mil 1 26 6
tftlckneu By ultriwnic or CMMrUlt
technique i
17. Surny by inspection p1a-Menet1c i. 0.25
flux typt or other comreble K „ „
device '• O'w
1 c. 1
18. Den* of covtr Inspection by sonar
(lidttcwi and penetrating) or otMr
coBOarablt oe«1ce
- UndOTMter 11iw> only 4
- All Unas 4
19. Charting of lino P1PO
- Underwater linn only 0.5
- All MUM O.S
20. ProMfttlM proerw for outside forcts
(•efefOMO 12) One-Call SystM , ""•
21, Pig KM for ottor rooval
22. Oil iplll aittcurs
- Mrino tomlnal or platform
- Totil Hit pipt mi lost
23. 10(0) and 13)«)
24. 17(e) Md 20
2S. ](»). 17(c) and 20
26. 10(b) and 17(c)
27. 5(6), 6(«) and 7
2S. S(a), <(t). 6(6). 7 and 9
29. a and 17(c)
J1
continuous
continuous
87
100
113
6
6
2
2
59.
13
19
22
24
4
4
0
0
29
3
0 0
0 0
174 (3
; 187 68
197 69
i 195 63
°
°
113
«
0
IS
Prawtttd
Total Latks
% "TIP
NO. K DblS
113 17
131 22
174 26
181 29
16 3
32 3
106 12
122 13
128 15
10 j 1
10 | 1
2 i 0.3
2 < 0.)
88 i 10
1
K , 3
0 1.4
0 1.4
237 26
253 26
266 28
258 29
0 0
o : o
128 i 17
| unditoctod
Volusc
Duptum
"TUP
K bbls
Total
"TP
K bbls
73 i 90
84 ; 106
93 ! 119
105 133
Spills
Total
:•.
177
159
116
109
VolUM
Total
vu
K Obi!
201
185
198
186
12 15 274 276
20
68
79
90
12
12
1.4
1.4
23 ' 258 268
80 184
92 168
105 : 162
13
13
1.7
1.7
10S 120
12 15
0
0
1.4
1.4
235 261
19
210
27
m
202
274
29
290
53
235 | 261 > 37
250 : 278 i 24
!
250
209
2*8
279
209
42
290
248 ! 2K
'• i
244 ! 261 i 1(2
211
199
196
16
278
27.3
2*0
170
276
27.4
290
30
30
13
12
82
43
30
(1) EstlMtod annual sg111a
-------
• Preventive program for outside forces
• Depth of cover by sonar or other comparable device .
Methods that have the potential of significantly (60 percent or better)
decreasing the spill volume include:
• Visual inspection of the line
daily by ground patrol with hydrocarbon probe or
comparable device
• Volume comparison
• Mathematical modeling
• Passive acoustic array.
Methods that have the potential of substantially (40 percent or better)
decreasing the incidents of spills include:
• Visual inspections of the line
daily by ground patrol with hydrocarbon probe or compara-
ble device
• Passive acoustic array
• Pressure difference
• Survey by inspection pig-magnetic flux type or other compara-
ble device.
Overall, mathematical modeling and passive acoustic array appear to be the
most promising for preventing the greatest volume of spillage; periodic sur-
veys by inspection pigs and the pressure difference method appear to be the
most promising for reducing the frequency of spills.
Most methods are highly effective in reducing either the frequency or
volume of spills, and in some instances are moderately effective in reduc-
ing both parts of the risk. Results indicate that the overall reduction of
both the frequency and volume of spills is never better than 50 percent for
any method. However, various combinations of as few as two methods result
in potential reductions of greater than 80 percent of both the frequency and
volume of spills; combinations of three methods or more can result in poten-
tial reductions of greater than 90 percent for both the frequency and volume
of spills.
7.2.3 Estimated Spillage and Prevention for All U.S. Lines
Total accidental spillage from U.S. lines that can be prevented or go
undetected is estimated based on information presented in Section 7.2.1.
Results for the selected methods are given in Table 73. Values are provided
aUnderwater lines only.
306
-------
TABLE 73. ANALYSIS OF OIL SPILLAGE THAT CAN BE PREVENTED OR GO UNDETECTED FOR ALL U.S. LINES
CO
o
o
o
3
n>
o.
Inspection and/or leak Detection Methods
1. Visual Inspection of line by air or
ground patrol (required Inspections)
2. Visual Inspection of line by air or
ground patrol In excess of required
Inspections
3. Visual inspection of line by ground
patrol with hydrocarbon probe or
comparable device for Indication of
spill
4. Hydrocarbon probc-tOHf Ish or similar
device
' Underwater I inns only
- All lines
5. Pressure deviations
a. Pump stations (existing method)
b. Along line
6. flow rate
a. Deviations
b. Comparisons (computerized)
7. Volume comparisons (computerized)
S. Mathematical modeling (computerized)
9. Negative pressure surge (computerized)
10. Passive acoustic array (computerized)
11. External rods with passive acoustic
sensors
12. Pressure static
)3. Hydrostatic
Inspections •
F i equency/
Vcar
26
Spillage
Prevented
Spills
Leaks
flP
I
0
a. 26(rcf) ' 10
b. 338 33
ji
». 26(reg) '
b. 26
c. 338
6 !
6
18
28
45
*
30
3
continuous • 0
continuous 0
continuous
continuous
continuous
continuous
continuous
a. continuous*
k- cent!n«Q«j'
after Indi-
cation of
spill*
after indi-
cation of
rupture*
« I
b. 2
c. after Indi-
cation of
rupture*
0
0
0
0
0
21
39
0
0
sr •
32
37
Ruptures
'HP
".
0
14
49
12
26
64
20
1
0
0
0
0
0
0
0
34
.58 __.
0
0
- ?o
24
30
total
fP
S
0
11
37
17
26
50
24
2.4
0
0
0
0
0
0
0
30
13
0
6
28
30
35
Volume
Leaks
y
0
3D
79
31
59
90
69
10
0
0
0
0
0
21
0
21
48
10
21
n
59
59
Ruptures
V1RP
0
18
71
15
3D
79
38
4
0
10
20
40
70
90
30
M
10
Total
y
0
20
72
16
33
Fd
41
4.1
0
9
18
36
63
83
27
70"
84
10
20 ( 20
I
20
24
30
21
27
33
Undetected
Spills 1 Volume
Total ! Total
1 u : vu
•.
100
89
63
83
72
to
76
98
100
100
100
100
100
100
100
"W
'57
100
ion
W~"
70
65
100
80
28
84
67
i'C
59
95
100
3?
81
64
37
17
73
XT '
16
90
W
»-
73
67
-------
TABLE 73 (continued)
CO
o
00
Inspection and/or leak Bisection Hulhods
14. fitiiiurt dlllcrcniB
IS. Change or *dd Inhllillnrs *s needed
16. Inspection of ta*ple ol Hue for mil
Ihlcknctl by ultrasonic or compartbU
technlaui
17. Survey by Inspection pl|-«uttnetlc flux
typt or oilier c»«|>arabl* devlct
II. Ueptb of cover Usueil Ion - Underwater
- All lines
19. Clurtlmol line pipe - Und«n«ttr
- All lints
20. freventlve program lor outside forcel
(lefcrenct 12) One -Oil iyslea
21. fig line for water rcwivtl
22. Oil spill detectors
- (Urine tei«ln
Leaks 1 Ruptures
»ar 1 '•w
i t
M
76
90
97
10
10
— «r
46
14
1
10
1
11
10
40
&
90
90
97
99
0
0
S9
It
12
»
40
S
76
u
10
. M
46
6
4.6
1
40
(
0
0
90
90
9S
9S
80
9S
91
loUl
t
11
16
41
46
S
a
12
16
U
4
4
1
19
&
6
0
90
90
96
96
72
as
90
undetected
•nii •«
Total
1
6t
SS
40
18
9«
89
(4
S6
66
97
94
99
70
94
100
100
IB
11
14
It
100
100
56
Tolil
»U
t
»
64
68
64
91.
9?
74
70
. M
Si
9t
9f
V'J
61
9S .
94
100
10
10
4 .
4
21
IS
10
^,
(I) Ammal tplllaye: lulal vulunc (¥,_,) • 1.1 barrels! Ruptures (0.90 »s|(.| • 1.17 barrels; leaks (O.I «s|[|
1.1 > 10 '( Ru|>lurct (0.2S f$|) • 0.12 » 10°; teaks (0.7S fil(.J
• 0.11 Until.
total spills (f,
• 0 Id » 10°.
(2| Retrofit Im
(I) Nm line In-.Ullillim.
(4) (requcncr ol lni|»!< I Ion It issiwiud
-------
for both the frequency (number of incidents) and volume (barrels of oil) of
spills that can be prevented (or go undetected). Estimates are included for
leaks and ruptures.
The relative effectiveness of the methods is, of course, similar to
results presented in the previous section for the reference line pipe.
Annual inspections of the line can be quite effective. For example, yearly
surveys by an inspection pig can potentially prevent 128 spills and elim-
inate 100,000 barrels spilled. On a comparative basis, this particular
inspection potentially could result in reducing the frequency and volume
of spills by a factor of at least three times better than cathodic protec-
tion systems. Use of more than one method can result in substantial reduc-
tions in the reported spill incidents and volume. For example, annual sur-
veys by inspection pigs and implementation of a mathematical modeling system
on the lines can potentially result in a reduction of 128 spills and pre-
vention of spillage of 244,000 barrels yearly. The dual prevention capabil-
ity is particularly desirable for use in older smaller diameter lines where
the incidents of spills are high, but spill volume relatively low. In cases
of large-diameter lines, where high throughput presents a continuing poten-
tial danger of a major spill, methods that are particularly effective in
reducing the volume of spills are highly desirable. For example, computer-
ized mathematical modeling or volume comparison methods are expected to
reduce the annual spill volume by 240,000 barrels and 185,000 barrels,
respectively.
7.3 COST ANALYSIS
Costs of inspections and spillage are extremely important in the de-
velopment of a practical spill prevention program. All significant costs
must be obtained to properly evaluate the cost and cost-effectiveness of
the various options (methods and schedules). When a number of options are
suitable for spill prevention and control, inspection costs would normally
be the most important factor in recommending a particular method. When high
spillage costs are expected, such as in high risk areas, additional inspec-
tions or methods may be justified. Costs for inspections and leak detec-
tion and spillage are discussed in the following subsection.
7.3.1 Inspection and Leak Detection Costs
Significant cost items for implementation of each inspection or leak
detection method are included in the estimate of the total inspection cost.
Costing includes such items as:
• Equipment
• Installation
• Interest
• Operation
• Maintenance
• Facility downtime
309
-------
• Training
• Inspection services.
Cost estimates, with a baseline year 1978, were made using data supplied by
inspection services, equipment manufacturers, operators, industry surveys,
trade journals and from a number of references in this study.
The reference line is assumed for estimating the costs of the various
inspection and leak detection methods. The assumptions for the reference
line are given in Tables 56 and 74. The reference line is assumed to con-
sist of 300 miles of on!and and 30 miles of underwater lines. Costs for
the reference line are multiplied by a factor of 676 (mileage of U.S. lines/
mileage of reference line) to obtain an estimate for all U.S. petroleum
pipelines. Typical costs of equipment and/or inspection services that are
assumed for the cost analysis are shown in Table 75.
Costs for methods requiring purchase of equipment and/or permanent or
semi-permanent installations were obtained by first estimating the equip-
ment costs. Added to this value was a 25 percent cost for installation.
Yearly maintenance and operational costs were assumed to be five percent
of the cost of equipment and installation. The equipment was assumed to
have a life of ten years with no salvage value. Total annual costs were
computed by amortizing the equipment and installation costs over a ten-
year period and then adding the maintenance and operational costs.
Estimates of the yearly costs3 (equipment, installation, maintenance,
operation, etc.) of the selected leak detection and inspection methods for
the reference line and all U.S. lines are presented in Table 76. These
estimates are presented on an annual basis and do not include interest and
inflationary costs"; this is done to help simplify the comparison of methods
that may not require equipment or installation, such as line walking, with
those methods that do, such as flow deviation monitoring. It is normally
assumed in the costing that implementation of the various methods can be
accomplished either by limited interference (retrofitting) or not inter-
ference with the line.
aCosting is developed at a level suitable for evaluation purposes of this
study only. Cost estimates are based on an assumed reference line. De-
tailed costing and optimum accounting for each method is beyond the scope
of this study. Costs for a specific line may vary considerably.
Interest and inflationary costs are not needed for the cost analysis.
Equipment costs (once purchased) are fixed and thus affected by inter-
est costs and not significantly affected by inflationary costs. Labor,
rentals, or inspection services are affected by inflationary costs but
not significantly affected by interest costs.
310
-------
TABLE 74. ASSUMPTIONS FOR REFERENCE LINE
Mileage:
On!and - total 300 miles
- crude gathering lines
- crude trunk!ines
- product trunk!ines
Marine Terminal:
Underwater - total 30 miles
- crude gathering lines 20 miles
- crude trunklines 10 miles
- product trunklines
Pump Station:
Number - 6
Spacing - 50 miles
Inlet and Outlet Meter:
Number - 3
Spacing - 100 miles
Corrosion Monitor:
Number - 30
Spacing - 10 miles
311
-------
TABLE 75. EQUIPMENT AND INSPECTION SERVICE COSTS FOR SELECTED
LEAK DETECTION AND INSPECTION METHODS
item No.
1
2
j
4
5
6
7
i
9
10
u
12
13
14
15
16
17
13
19
20
21
2:
23
24
25
26
27
23
29
Equipment or Inspection Service
Alp patrol line pipe inspection service
Ground patrol
Hydrocarbon probe or equivalent
Ground patrol with hydrocarbon probe or other comparable device
Hydrocarbon prooe (towfisn type) or otner camparaole device
- lease or rent system
• operator
Oil spill detector system
Line pressure meter
Telemetry for remote monitoring
Telephone lines for remote monitoring
Leak detection display, control and alarms for pressure monitoring
and/or flow monitoring
Flow meter system
Correction equipment for flow transients
Inlet and outlet meters modified for leak detection
Leak detection display, control and alarms - flow rate comparison
- master control room
Central minicomputer
Real-time pipeline model software
Negative pressure surge monitor
Acoustic detector (portable)
Acoustic analyzer and display
Acoustic detector and signal conditioning (permanently installed)
Acoustic transmission/receiver cabling - on land
- underwater
Acoustic master unit
Acoustic central minicomputer
Remotely controlled block valves witn pressure difference
transducers (includes valve, transducers, power,
telemetry units, connectors to telephone lines)
Laboratory equipment for analysis of fluid in line
Survey inspection pig-magnetic flux type or equivalent
-by inspection service
Depth of cover inspection of underwater lines by .inspection service
Sonar system (sidescan and penetrating) towfish type or other
comparable device
- lease or rental of equipment
- operator
Charting of underwater lines by inspection service
Estimated Cost
SO. SO / mile
$4.00 / mile
$2,000 / meter
$250 / day
$250 / day
$200 / day
$7,000 / unit
$2.000 / meter
S3, 000 / unit
$5,000 / station
$3,000 / pump station
$15,000 / pump station
$10,000 / pump station
$5,000 / meter
$5,000 / pump station
J10.000 /line
$100,000 / line
150,000 /line
$5.000 / unit
$500 / detector
$3,000 / unit
$200 / unit
$2.500 / mile
$10.000 / mile
$10.000 / station
$100,000 / line
$20,000 / valve
$2,000 / unit
$500 /mile
$300 / mile
S500 / day
$200 / day
$300 /mile
312
-------
TABLE 76. COST OF INSPECTION AND LEAK DETECTION METHODS
Inspection and/or Leak Detection Methods
I. Visual line inspection by air or
grouna patrol (Required inspections
for indication of spill )
2. visual line inspection by air or
general patrol in excess of re-
quited inspection for indications
of a spill
3. Visual inspection by ground patrol
»itn nydrocarbon prooe or compar-
able device for indication of spill
(on land lines only)
4. Hydrocarbon probe-towfish or sim-
ilar device
-Underwater lines only
5. Pressure deviations
a. Pump stations (existing method)
b. Along line
6. Flow rate
a. Deviations
b. Comparisons (computerized)
7. Volume comparisons (computerized)
o. Mathematical modeling (computerized)
9. Negative pressure surge (computerized)
10. Passive acoustic array (computerized)
a. Retrofit
b. New line
11. External rods with passive acoustic
sensors
12. Pressure static
13. Hydrostatic
INSPECTIONS
Frequency/
Year
26
a. 26 (Ref.)
b. 338
a. 26 (Reg.)
b. 26
c. 338
6
ANNUAL COSTS
Reference Line '
Per(1) Total
Mile
$ S K
All U.S. Lines
Total
S Million
14 4.8 3
!
14
187
110
4.8 3
62 i 42
33 22
220 66 i 44
1,420
1,200
429 | 286
36 24
continuous NA IJ'
continuous 85
continuous
continuous
continuous
continuous
continuous
continuous
continuous
after indica-,,,
tion of spillu'
after indica-.,.
tion of spill^'
i
a. 1
b. 2
s- after indica-,,,
I tion of spilT^'
28 19
68
185
185
212
159
23
61
61
70
52
210 70
954
59
9
36
315
18
3
12
72 24
15
41
41
47
35
46
210
12
2
8
16
• !
36 12
8
313
(continued)
-------
TABLE 76 (continued)
Inspection and/or Leak detection Methods
14.
15.
16.
17.
18.
19.
20.
21.
22.
n.
24.
25.
26.
27.
28.
29.
Pressure difference
Change or dad inhibitors is needed
Inspection of sample of line for wall
thickness by ultrasonic or comparable
technique.
Survey by inspection pig-magnetic flux
type or other comparable device
Depth of cover inspection by sonar
(sidescan and penetrating) or other
comparable device
-Underwater lines only
Charting of line pipe
-Underwater lines only
Preventive program for outside forces
(Reference 12) One-Call System
Pig line for water removal (low
elevation only)
Oil spill detectors
-Marine terminal or platform
-Total line pipe mileage
INSPECTIONS
Frequency/
Year
a. 1
b. 2
C. 4
d. 12
jlZ)
1
a. 0.25
b. 0.50
c. 1
4
O.S
Cont.
3<2>
continuous
continuous
10 (b) and 13 (a)
17 (a) and 20
3 (a), 17 (c) and 20
10 (o) and 17 (c)
5 (b). 6 (a). 7
5 (a). 6 (a). 6 (b), 7 and 9
3 and 17 (c)
ANNUAL COSTS
Reference Line
Per'1' Total
Mile
5 S K
290
327
400
690
4
87
125
250
500
1,200
150
45
15
16
16
1,000
1,080
303
454
712
96
108
132
228
1.2
26
42
83
165
36
4.5
15
5
5.2
5.2
327
357
100
150
235
All U.S. Lines
Total
S Million
64
72
38
152
0.8
18
28
56
111
24
3
10
3
3.5
3.5
218
238
67
100
156
NOTE: (1) Cost per mile for single inspection can be computed by dividing the cost by number of inspections/year
(2) Frequency of inspection is assumed at three per year
(3) NA - Not applicable since method is commonly used
314
-------
Typical construction costs per mile for offshore pipeline are shown in
Table 77 for various diameters. This information is useful in comparing
the cost of the pipeline and the cost of spill prevention.
Pertinent items assumed for the selected methods are provided in the
subsections that follow.
7.3.1.1 Cost of Selected Visual and Visual-Aided Line Inspections--
Visual inspections of the line pipe are normally carried out in the
U.S. by air patrol inspection services. Cost estimates are based on in-
spections by air patrol services with some adjustments for the limited use
of ground patrol by the pipeline company personnel. Air patrol inspections
services3 are estimated at 50 cents per mile for each inspection. Ground
patrol inspections are estimated at four dollars per mile. Inspections
carried out every two weeks are estimated to cost approximately 3.2 million
dollars each year for all U.S. lines.
Ground patrol inspections with a hydrocarbon probe or other comparable
devices are estimated at 4.2 dollars per mile for each inspection. Inspec-
tions carried out every two weeks are estimated to cost approximately 22
million dollars each year for all onland U.S. lines.
Inspections of underwater lines using a hydrocarbon towfish or other
comparable devices are expected to cost approximately 200 dollars per mile
for each inspection. These inspections are normally performed by inspec-
tion services that provide both the equipment and the operator. A boar
(with a winch) and support crew are also required. Inspections carried
out every two months are estimated to cost approximately 36,000 dollars
for the reference line and 24 million dollars each year for all underwater
U.S. lines.
7.3.1.2 Cost of Selected Method for Oil Spill Detectors on or Near~"the
Water—
The method employing oil spill detectors at marine terminals and off-
shore platforms would typically require about four detectors at each loca-
tion. Annual costs are estimated at 5,200 dollars for the reference line
and 3.5 million dollars for all U.S. lines.
aCost of electronic monitoring by air patrol of the corrosion monitors
installed on the line, typically spaced at 5 to 15 miles, are not in-
cluded. However, this service is estimated at 50 cents per mile as-
suming bi-weekly inspections. Typically, the monthly charge for each
monitor is about ten dollars.
315
-------
TABLE 77. OFFSHORE PIPELINE CONSTRUCTION COSTS
(a)
Diameter
(inch)
6
8
10
12
14
16
18
20
22
24
26
28
30
32
36
40
42
44
48
Cost
Installation
$K/mile
250
262
278
300
325
350
385
420
465.
510
576
642
720
810
1,010
1,290
1,440
1,660
2,210
Yearly(b)
$K/mile
5.0
5.24
5.56
6.0
6.5
7.0
7.7
8.4
9.3
10.2
11.5
12.8
14.4
16.2
20.2
25.8
28.8
33.2
44.2
aSource: CEQ 1974, Inflated 31 percent (Ocean Industry pipeline
inflation factor) to convert 1972 dollars to 1975 dollars.
Assumes 50-year life with no salvage value.
316
-------
7.3.1.3 Cost of Selected Methods for Continuous Monitoring of Internal
Fluid Variations During Transfer—
The method employing continuous monitoring of pressure deviations
along the line is assumed to include the existing pressure monitoring sys-
tems at each pump station and new installations at two locations (approxi-
mately 16 miles apart) between each pump station. Twelve of these sta-
tions would typically be installed on the reference line. Other equipment
includes telemetry and telephone lines at remote monitoring stations and
leak detection displays, controls and alarms at each pump station. Annual
costs are estimated at 28,000 dollars for the reference line and 19 million
dollars for all U.S. lines.
The computerized flow rate comparison method requires the installa-
tion of flow meters at each pump station and modification (for leak detec-
tion) of inlet and outlet meters typically at three locations. Other
equipment includes electronics that connect flow meter to telephone lines,
leak detection monitoring units with control and alarms and correction
equipment for flow transients at each pump station, and a central mini-
computer at the main control room. Annual costs are estimated at 61,000
dollars for the reference line and 41 million dollars for all U.S. lines.
Costs for implementing the computerized volume comparison method are
about the same as for the flow comparison method. Similar equipment is
required.
The mathematical modeling method requires equipment similar to the
volume comparison method. However, some additional equipment is required
for implementing the real-time pipeline model. Typically, the equipment
is implemented in the central computer as a software implemented system.
Annual costs are estimated at 70,000 dollars for the reference line and
47 million dollars for all U.S. lines.
Implementation of the negative pressure surge method primarily re-
quires installation of pressure surge monitoring stations at the pump sta-
tions and at various locations, typically at two stations approximately
16 miles apart, between each pump station. A total of 18 stations typi-
cally would be required. Other equipment includes telemetry and telephone
lines for remote monitoring and a central minicomputer at the central con-
trol room. Annual costs are estimated at 50,500 dollars for the reference
line or 35 million dollars for all U.S. lines.
7.3.1.4 Cost of Selected Methods for Detection and Location of Spills
on or at a Short Distance from the Line Pipe—
The major installation required for the method employing external rods
with passive acoustic sensor are the metal rods permanently attached (by
brazing) to the lines. The rods are typically spaced approximately one-
quarter mile apart. Portable acoustic sensors with associated analyzing
equipment normally would be stored at each pump station. Annual costs are
estimated at 17,600 dollars for the reference line and 12 million dollars
for all U.S. lines. This estimate does not include installation on under-
water lines.
317
-------
Two cost values are provided for the passive acoustic array inspection.
One cost is based on a typical system for retrofit installations. The
second cost is based on a typical system for new lines. A system retrofit-
ted to the line is expected to be capable primarily of preventing damage by
outside forces and detecting ruptures, while a system for a new line is
expected to be capable of preventing most failures and detecting most leaks.
Costs for this latter system are expected to be too high for most existing
lines because of the large number of sensors and associated equipment and
the high expense to uncover the line for equipment installation. However,
this system should be considered for short sections of lines in high risk
areas, i.e., underwater.
For retrofitting on existing lines, the passive acoustic array method
is expected to require installation of acoustic transducers at the pump sta-
tion and at various locations, typically at four stations (approximately
ten miles apart), between each pump station. Other equipment includes sig-
nal conditioning, signal processing, and telemetry and telephone lines for
remote monitoring. Annual costs are estimated at 69,600 dollars or 46 mil-
lion dollars for all U.S. lines.
For installation on new lines, acoustic sensors with signal condition-
ing and signal processing equipment are expected to be installed at approxi-
mately five locations each mile. Typically, a multiconductor cable running
the length of the line would provide both power and transmission of the sig-
nal from each acoustic sensor. (Numerous other means, such as telemetry
data link, might also be used for transmission of the acoustic signal from
line pipe to the control room.) Master units supply the power, signal pro-
cessing and control. These units would be located at each pump station and
at two locations between each pump station. • Annual costs for a new line
are estimated at 315,000 dollars or 210 million dollars for all U.S. lines.
7.3.1.5 Cost of Selected Methods for Periodic Pressure Tests-
Pressure static tests are expected to be of low cost. For a typical
test, the line would be held at normal line pressure for a short time interval
(less than one hour). Normally long sections of the line would be tested
separately and a few hours would be required for a complete test. It is
assumed that mainline valves would require manual closure. Yearly costs2,
assuming tests are carried out three times, are expected to be about 3,000
dollars for the reference line and 2 million dollars for all U.S. lines.
Hydrostatic tests are based on test section spacings of about 16 miles
and hold times of about 24 hours. Annual inspections are estimated to cost3
approximately 12,000 dollars for the reference line and 8 million dollars for
all U.S. lines. Hydrostatic tests carried out only after an indication of a '
leak are expected to be about the same cost. For this latter test scheme,
only a limited number of line sections would be tested. Inspections are
expected to be carried out about three times a year.
\ine downtime and leak location costs are not included in the estimate
for this inspection.
318
-------
7.3.1.6 Costs of Selected Corrosion Inspection Methods-
Inspections forMnternal pipeline corrosion through laboratory analysis
of fluid transported could be carried out by either operating company per-
sonnel or outside inspection services. Equipment for analysis of the fluid
can be either purchased or leased. Costs are based on the assumptions that
the inspection is done by an outside service three times a year. Annual
costs are estimated at 1200 dollars for the reference line and 800,000 dol-
lars for all U.S. lines.
7.3.1.7 Costs of Selected Standard Non-Destructive Testing Methods-
Inspections of sample sections of the line for wall thickness changes
by ultrasonic or comparable techniques are assumed to be carried out by an
inspection service. The cost estimate is based on inspection of small
areas of the line pipe at a spacing of approximately two miles. Annual
costs are estimated at 26,000 dollars for the reference line and 17.6 mil-
lion dollars for all U.S. lines.
7.3.1.8 Cost for Selected Inspection Pigs-
Inspection of the line using inspection pigs (magnetic flux type or
other comparable devices) are normally carried out by inspection services.
Costs are estimated at 500 dollars per mile per inspection. Annual costs
are estimated at 165,000 dollars for the reference line and 111 million
dollars for all U.S. lines.
7.3.1.9 Cost of Selected Methods for Line Pipe Charting and Depth of
Cover-
Inspections of the depth of cover of underwater lines using a sonar
(sidescan and penetrating) towfish or other comparable device would typi-
cally be carried out by inspection services that provide equipment and an
equipment operator. A boat (with winch) and a support crew are also re-
quired. These inspections are expected to cost approximately 300 dollars
per mile for each inspection. Inspections carried out every three months
are estimated to cost 36,000 dollars for the reference line and 24 million
dollars each year for all underwater lines in the U.S.
Charting the location of underwater lines is normally carried out by
inspection services. A diver, support crew and a boat are the major cost
items. Inspection costs are estimated at 300 dollars per mile. Annual
costs are estimated at 4,500 dollars for the reference line and 3 million
dollars for all underwater lines in the U.S.
aUnderwater lines are not included in this cost estimate.
319
-------
7.3.2 Spillage Costs
Spillage costs in the study are separated into the following three major
categories:
* Petroleum spilled from line pipe
• Spill cleanup (function of the location and means of
spill cleanup)
• External damage (function of the severity or risks
external to the line).
Separation into these three cost categories is particularly advantageous for
both the cost-effectiveness analysis carried out in Section. 7.4 and the se-
lection of methods for lines located in high risk areas. Spillage cost for
each category will be discussed in the subsections that follow.
7.3.2.1 Petroleum Spilled—
The cost (value) of petroleum spilled from line pipe is difficult to
estimate because of the wide range of petroleum prices that exist for crude
(imported, new production wells, old production wells, etc.) and product
(various types). To avoid this problem, the costs of petroleum spilled are
evaluated based on the quantity (in barrels) rather than the actual cost
(dollars). Conversion from barrels lost to dollar values can be made di-
rectly for any specific line or nationwide based on average petroleum costs.
7.3.2.2 Spill Cleanup—Cost Factor (Function of the Location and Means of
Spill Cleanup) —
Spillage costs for spill cleanup vary widely. However, these costs
generally depend on the location and means of cleanup of the spill. For
example, a spill located in a remote area at large distances from roads
would be very costly to clean up using typical oil recovery methods, but
the cost of burning the lost oil would be much lower. In this study, stan-
dard oil recovery methods will be assumed.
Values for the cost factors SCF$C that account for the location and
means of spill cleanup are provided in Table 78. These factors are quali-
tative estimates based upon information obtained from companies carrying out
spill cleanup, from a typical operating company and information available in
the literature. It should be noted that these cost factors can vary widely
depending upon company practices or local regulations for a particular line.
For example, if spilled petroleum were burned on!and costs of cleanup would
be minimal.
7.3.2.3 External Damage—Cost Factor (Function of Severity or Risks Ex-
ternal to the Line)--
Spillage costs, for damage external to the line from a spill also vary
widely. However, these costs generally depend upon the risks that exist
external to the line.
320
-------
TABLE 78. SPILL CLEANUP COST FACTORS AS A FUNCTION
OF THE LOCATION OF SPILL
Location Cost Factors SCFSC
Onland
Easy access 1
Moderate access 2
Difficult access 5
Underwater
Confined area 50
Offshore 100
321
-------
Values for the cost factor SCFgo that account for severity of the spill
or risks external to the line can be obtained from Table 58. This cost fac-
tor is assumed to be identical to the overall risk correction factor
presented in Section 7.3.3.4.
7.4 COST-EFFECTIVENESS ANALYSIS
An analysis of the cost-effectiveness of the selected inspection and
leak detection methods for various inspection schedules is carried out in
this section. The analysis accounts for both effectiveness (the capability
of methods to reduce the incidents and volume of spillage) and the costs
(inspection methods and spillage). The cost-effectiveness of the various
options is the most important consideration in the development of a practi-
cal spill prevention and control program for line pipe. The need for eval-
uation based on cost-effectiveness arises because of the variety and number
of options at various costs that are available for reducing the risks of
oil escaping from the line.
A general measure of cost-effectiveness C£QM based on the ratio of the
reduction of barrels of oil spilled to the inspection costs is presented in
Section 7.4.1. This measure applies to most lines. Values are computed
for the selected methods and results discussed.
A second or specialized measure of cost-effectiveness CE$M is presented
in Section 7.4.2. It accounts for spillage from lines located in areas
where the potential exists for high spillage cleanup costs and/or high spill
risks external to the line. This measure should be used for individual
lines where special spill prevention and control may be needed because of
the potential of serious and/or costly spillage.
7.4.1 General Measure of Cost-Effectiveness
The general measure of cost-effectiveness CEQM accounts for the volume
or incidents of spillage (barrels) and inspection costs. It is defined as:
For Volume Spilled
rp _ spillage (barrels of oil prevented)
GM ~ inspection costs
or
For Incidents
PP _ spillage (number of incidents prevented)
GM ~ inspection costs
Values for the selected methods are given in Table 79.
Two main reasons for using the general measure of cost-effectiveness
are:
322
-------
TABLE 79. COST-EFFECTIVENESS ANALYSIS FOR ALL U.S. LINES
Insptctton "id/or LMt IKUctlM NttMOS
1. Visull lint insptction by «1r or
grmia p«roJ (Rwuirtd fnsptcttons
for IMIutton of spill)
2. VtiiMl HIM Inspection by . 1
b. 2
c.
-------
TABLE 79 (continued)
[nsptetlon ind/or Luk Otttctlon Htuiods
14. Prtssurt dfffirtnct
IS. Cktnat or idd initiators a nttdtd
16. Insotctlon of stcplt of lint for Mil
talckntss by ultrasonic or comnolt
ucMlqut.
17. Sunny by Insptctlon plg-Montttc flu
tyet or otlwr cotptnolt dnict
1». Otpta of covtr inspection by sontr
(sldtfcn ind ptnttritinq) or otMr
coipirult dtvlct
-UMtnwttr lints only
-All HIM
19. ChMtlno. of lint plpt
-Unatnnttr 11nt< only
-All lints
20. Prntntlw pratrM for outltdt fonts
(Ktftrtnct 12) Ont-Ctll Systn
21. fit, lint for
1
I. 0.2S
b. 0.50
C. 1
4
4
0.5
O.S
cant.
3'"
continuous
continuous
I
IMUAL COSTS
U.S. Lints
rotil
S Million
S4
72
33
152
0.8
SPILLAGE PREVENTED COST -FflCTIVENESS
Inc1 dints
Itductd
Ha.
113
131
174
181
16
18
28
56
111
24
24
3
3
10
3
3.5
3.5
218
238
67
too
156
32
106
122
128
10
10
2
2
88
16
0
0
237
253
266
258
0
0
128
oluot
Sivtd
bbls
90
106
119
133
15
23
80
92
105
13
13
1.7
1.7
120
15
1.4
1.4
261
261
278
279
209
248
261
nc1 dtnts
»^«" ,
no.
TIT
1.3
1.3
2.0
1.2
20
1.8
3.8
2.2
1.15
4.2
4.2
0.7
0.7
8.8
5.3
0
0
1.1
olum
JVtd
cost
1.4
1.5
1.3
0.9
19.8
12.8
2.9
1.6
0.9
0.5
0.5
0.7
0.7
12
3
0.4
0.4
1.2
,
1.1 : 1.2
0 3.1
0 2.4
3.3 I 1.7
55:
(1) W - Mt 4pp!1cibtt ilnct «ttnod
-------
• Can be applied to most lines
• Independent of line location, i.e., depends only upon the
volume of petroleum escaping from the line.
This measure is particularly useful in the consideration of guidelines
or recommendations for a general spill prevention and control program that
might be developed for a typical line by the operating company or for most
U.S. lines by a nationwide program.
7.4.2 Special Measure of Cost-Effectiveness
High spillage costs are possible in certain locations where high spill
cleanup costs are expected or in areas where high oil spill risks to the en-
vironment exist external to the lines. These two sources of spillage costs3
are not directly accounted for in the general measure of cost-effectiveness.
In order to account for both of these spillage costs, a second or special
measure of cost-effectiveness CF$M is provided. It is defined as:
and
CESM = CEGM x SCF
SCF = SCF$C x SCFED
where
CE_M = special measure of cost-effectiveness
CEGM = general measure of cost-effectiveness
SCF = spillage cost factor
SCFSC = spillage cost factor for spill cleanup as function of
the location of spill (see Section 7.3.2)
SCFrn = CFPn = spillage cost factor for risks external to the
LU tu line (see Section 7.3.3).
Spillage costs due to risks of the external environment are accounted for
when the risk correction factor CF£0 for risks external to the line is
used in estimating the oil spill risk of a particular line or group of
1i nes.
If the.risk correction factor CF£D was used in estimating the oil spill
risk the spillage cost factor SCF£D for risks external to the line is
given a value of 1.
325
-------
The special measure of cost-effectiveness requires only the inclusion of the
spillage cost factor. Thus, the relative cost-effectiveness of the avail-
able options (see Section 7.3.1), based on the general measure of cost-
effectiveness, does not change.
As the costs of spillage increases, the cost-effectiveness3 of the
various options becomes greater. For example, if a line were located off-
shore the cost-effectiveness of each method would be substantially greater
than for a typical onland line. This obvious improvement in the cost-
effectiveness is indicated in the special measure of cost-effectiveness.
Improvement in cost-effectiveness often allows the user greater range of
selection, from a cost basis, of the various available options.
The special measure of cost-effectiveness should be considered in the
development of spill prevention and control programs for lines located in
areas where the following spillage problems exist:
• Potential of serious spill problems or high risks external
to the line pipe exists, i.e., the risk correction factor
C^rn is greater than 3.
• Spill cleanup costs are expected to be high, i.e., the cost
correction factor is greater than 3.
When the potential of these spillage problems exists, a more effective spill
prevention and control program is often needed. This might require more
effective methods (greater spill reduction capability) and/or more frequent
inspections.
aFor many lines, the additional costs of spillage may be so high that an
adequate spill prevention program may, in fact, result over the long-
term in substantial cost savings and other benefits.
326
-------
SECTION 8
RECOMMENDATIONS
8.1 DISCUSSION
Results of this study show that the potential exists for serious spills
from the line pipe of petroleum pipeline systems and that various options,
inspection and leak detection methods and schedules, are available to sig-
nificantly reduce the oil spill risk. Based on the results of this inves-
tigation, preventive maintenance programs for line pipe are recommended.
These recommendations are based primarily on qualitative3 estimates of the
oil spill risks that exist and the capabilities of various methods to reduce
the risks. However, recommendations are also based on practical considera-
tions such as costs of inspections and spillage.
Recommendations are primarily intended to aid the user both in evaluat-
ing the need for, and actually developing, a spill prevention and control
program for his own line pipe. The recommended programs are expected to
significantly improve the prevention and control of accidental spills of
petroleum from operational lines. The programs include scheduled inspec-
tions and/or in situ leak detection (some functioning continuously) that
could be effectively developed and implemented. Overall, the programs pro-
vide means of estimating the risks of spills from line pipe and the poten-
tial for preventive maintenance.
It should be noted that the statutory requirements and voluntary pre-
ventive maintenance programs by operating companies have already resulted in
a good safety record for line pipe. In many cases, voluntary individual
programs, such as implemented on many large-diameter lines or underwater
lines, exceed statutory requirements. Voluntary programs have helped to
significantly reduce the spill risk of these individual lines and also have
contributed to the good safety record nationwide. Despite these efforts,
spills that are costly, hazardous, or could result in a major pollution
The intent in this study is to provide the necessary background information
and approach so that recommendations are developed from a purely qualita-
tive basis. This is in contrast to most spill prevention plans that are
developed based on educated guesses using inadequate and possibly incorrect
information derived partly from the literature, manufacturers, experience,
hearsay, and quantitative estimates.
327
-------
incident can and do occur. The risk of such serious spills is particularly
high for certain lines or lines in certain areas. Results of this study
indicate that means are available to reduce the possibility of these serious
accidental spills and improve the safety record nationwide.
Since there are many variations between lines and wide ranges in the
oil spill risk, it is beyond the scope of this study to recommend programs
for particular lines. Hence, recommendations are generalized and do not
include specific details for implementing methods for particular lines.
However, recommended programs are developed that apply to typical lines.
Also, sufficient information is provided (see Sections 6 and 7) so that a
user may develop a program for his own line.
In order to evaluate the potential of spill prevention programs for
pipelines, spill records, the oil spill risks from petroleum pipeline sys-
tem preventive maintenance programs (statutory, required, voluntary),
and the risk reduction capabilities of the various available options (in-
spection and leak detection methods) were investigated. Costs of inspec-
tions and spillage were also studied. Since inadequate quantitative infor-
mation of the development of spill prevention programs was available for
many of the areas investigated, qualitative estimates3 were made to account
for these deficiencies. Then measures were developed for evaluating line
pipe oil spill risk (frequency and volume of spills), capabilities of meth-
ods to reduce the frequency and volume of spills (effectiveness), costs of
inspections and spillage, and cost-effectiveness. Values for these measures
were estimated. Finally, the recommended maintenance programs (Section 8.2)
were developed based primarily on analysis of the results of these measures.
The cost-effectiveness of the various available options (inspections and
leak detection methods and schedules), however, was the most important
consideration in the recommendations.
Benefits of a preventive maintenance program are discussed in Section
8.2. Although no inspectton or leak detection method can be justified
purely on the cost savings resulting from the value of oil that would be
saved, other factors that might justify preventive maintenance are pre-
sented.
A recommended preventive maintenance program for the line pipe of op-
erational line is presented in Section 8.3. Implementation of such a pro-
gram is expected to decrease substantially the total quantity of petroleum
lost, the number of spill incidents and the number of major spill incidents.
A recommended approach for the development of a spill prevention program
for individual lines is presented in Section 8.3.1. A general spill pre-
vention program is presented in Section 8.3.2 that produces the confidence
that lines will have no more than (X) barrels spilled and (X) number of in-
cidents per year. The program consists of a series of recommendations of
aAn in-depth analysis of these areas for the purpose of obtaining highly
accurate estimates is beyond the scope of this study.
328
-------
scheduled inspections and/or in situ leak detection. Finally, recommenda-
tions of a specific scheduled preventive maintenance program for all U.S.
lines and for special lines (lines with a serious potential that petroleum
will escape or those located in high risk areas) are presented in Section
8.3.3. Recommended developmental inspection and lead detection methods
are discussed in Section 8.4.
8.2 BENEFITS
No inspection or leak detection methods can be justified purely on the
cost savings resulting from the value (dollars) of the petroleum saved.
However, when other benefits are considered, preventive maintenance is often
justified. Overall, damage from spills is becoming more costly and the
benefits of spill prevention are evident.
Reduction of spillage costs (i.e., spill cleanup and external damage
(see Section 7.3)) is a major consideration in justification of preventive
maintenance. This is particularly true of lines located in high risk areas
(see Section 7.4) or lines that have a high risk that petroleum will escape
(i.e., large diameter lines, old lines, etc.). For many of these lines, the
additional costs of spillage may be so high that effective preventive main-
tenance programs may, in fact, result in substantial cost savings over the
long-term.
Benefits such as reduction of risks to personnel or others, reduction
of line shutdown, improved public relations, and the saving of the important
and valuable petroleum resources, further justify preventive maintenance.
Additionally, another important benefit is the reduction of the various
socio-economic-legal aspects of spills. For example, in evaluating real or
personal property of damage from a spill83, items such as diminished value,
temporary non-use, profit and loss, loss of taxes, penalties, etc., are
considered.
8.3 RECOMMENDED PREVENTIVE MAINTENANCE PROGRAM
8.3.1 Individual Lines
The recommended approach for developing a spill prevention program for
an individual or group of lines is based on the following steps:
• Estimate the risk that oil will escape from a line (see
Section 6.3.4.1)
• Identify locations of a line where a spill may create
special serious problems (see Sections 6.3.3.4 and 6.7.2)
such as the possible damage to the external environment
or the cost of spill cleanup.
• Estimate the level of corrective action to reduce the spill
risk to an acceptable level.
329
-------
• Identify the various options of a spill prevention program
that are available to achieve the necessary reductions in the
spill risk (see Sections 8.3.2.2 and 7.1.2).
• Select the optimum methods based on the cost-effectiveness
and other considerations of the various suitable options.
8.3.2 General Schedule
The general schedule produces confidence that an individual line or
group of lines will have no more than (X) barrels spilled and (X) number of
incidents per year. The recommended approach is to first estimate the risk
that oil will escape from the line (see Section 6.3.1.1). Then select one
of the specific preventive maintenance schedules:
• Reduction of frequency of spills - Table 80
• Reduction of volume of spills - Table 81
• Reduction of frequency and volume of spills - Table 82.
Using the indicated approach and schedules, selectable spilling reductions
of greater than 25 percent, 50 percent and 75 percent may be achieved.
8.3.3 Specific Schedule
8.3.3.1 Most U.S. Lines-
Recommendations for typical U.S. lines include:
• Visual inspections of the line by air or ground patrol -
weekly
• One-Call System (Reference 12)
• Survey by inspection pig (magnetic flux type or other
comparable device - every four years
or
• Hydrostatic tests
yearly, or
after indication of spill.
Increasing the frequency of visual inspections of the line by air or
ground patrol from bi-weekly to weekly is expected to result in annual sav-
ings of approximately 57,000 barrels of petroleum (20 percent of total) and
decrease in number of incidents by 32 (11 percent of tota.l). These reduc-
tions are expected to be achieved at a nominal cost of about 3 million dol-
lars. These inspections are the most cost-effective (19 bbls/$K) of any of
the selected methods. Assuming the average cost of petroleum (crudes and
products) at 25 dollars per barrel, cost savings of the value of petroleum
alone almost justifies these recommended inspections.
330
-------
TABLE 80. PREVENTIVE MAINTENANCE SCHEDULES FOR THE REDUCTION OF THE
FREQUENCY OF ACCIDENTAL SPILLS FROM LINE PIPE
Inspection ino/or ^tu »tttction "ttftoai
J*lL4l lint insetctlon Sy lir or ground
iitrol >n ticni of rtouirto 'nspKtions
/iiuil insatctim or ;round Html »iui
lyoncjraoii 9nM or joapjrieit 4tvtct
for inoiutlon of tpill (onlim Itnti
4yareciroan pra&**to«fiui or si»11«r
OKICt-JlOtrMUr Him only
Pxstn icaustlc imy iceouttriita)
i. JttrgfU
a. tta lint
4yerefUCic
Uun?t or ted timipiton u ntMM
In«ptctio255 ScntOult' ; >SOS SOlMult' >75S SclWdull'
A S t 0 E f 5' «' 4 i C D i1 A 3 C D1 £' f
XXX XXX IX
x !
X X
X X XXX
X XX
X X
I X X
i
i
x xx
.< X
x x xxx
J
X X
X I X I X , X X X
i
'
Wut:
!. ?ction4l scntaults co tcconoHin inaluttd otrcmt rtductlon of frtqutney of soills art idtntifftd in columt !A, 3. C.... tec.) &tlon.
I. :rtau«ncy of intotctisn *i tisurao it turn cim Mr yttr.
3. :no*n«ttr 1 (net only. na
-------
TABLE 81. PREVENTIVE MAINTENANCE SCHEDULES FOR THE REDUCTION OF THE
VOLUME OF ACCIDENTAL SPILLS FROM LINE PIPE
3*pct/it iMuc:ioA of voliot af Saflls
>2£i Scntcuie1
>5QS Scntfluli1
>?55 Scitcult*
1 3 - 3 = f : i 1 J' <' ' ; j 5 2 r : S'H'IM* i C 3 • ;'S'
v/crzcir«cn :roo« or csssiracl* civict :,
*sr ^;icjt:cn jr seili (ontind Itnts
anl,, . -. 133
-/srtcarccn sract-ccwf^sft 3r similar i
^ivut-vnctrMur hn«s snly
i. :--: sutions :.uitcin; -ttro;) , ^ntinuous
3. 4;acq : tot : continuous
:. l^c=»imons xC3rou: i. 0.25
-./;* sr 3C.-.tr cs=Cirw!« avict , , JG
. c. 1
r"lisar:Kr:w«rM-';;;:r"t«r
< (
i
X
1 ' x
X ; X
I
X X
j
1
X ' XX
X XX
i
X
X X
X
X
I
X X
:.nrvrj :r 1 ,„ ,:»--«•«*:» linti ! :.=
•-9v«rt:«« creqrsn for 3u£Slct forexs j continuous
= i{ -,!-.« '5f«:tr -MM! ' 2'
;il ioitl Kttctsrs^*iriit timinal , *tc. • cofttlnuous
j
< < < t 1 < x x x
1
1
1
rious ;a tccaireiitn inaiciud otrctnt 'tgucc:cn er
3^ intMc;1on is ittunM i: :nr»t :imts aer /«*c.
* lints only. ^OM*v«r, otntr tcnrruict s.iown n«y lisa
IUM af JBills «r« iatntH1« in cotuwt i.4. 3. C,...*tc.l Mlcw.
332
-------
TABLE 82. PREVENTIVE MAINTENANCE SCHEDULES FOR THE REDUCTION OF THE
FREQUENCY AND VOLUME OF ACCIDENTAL SPILLS FROM LINE PIPE
Inunction mayor Ltik ClCKtion Kttmes
pit ml in ucus of rtquirto inSMetlMK
visutl insstction ay ground sjtrol men
lyoroctroon orooo or ca*io«rtelt otvict
for ineiutlon of loll! (cnlwd linn
only)
Hyaroctroofl aroo«*cowfisn or siniltr
Stvict-Unotmtttr linn only
'rtsmrt anittians
*. Pu»o sutiMs (taistine. attnoa)
3. Alone, line
Ho* rltt
«. 3un turf* (EgrmiurttMl
?>s>i« iC3ultlc irriy (cn«»uttrut«l
i. Ditrafic
a. M>t lint
£ittn*l rgcs •ttn MUlvt tcouidc
Pntsun suite
«y«nml« of MM for »«U
cnteuins ay uitriMnle gr caiMriolt
ciemiout
Sarv«r Oy inioKtton gionMgiwcic 'lux
;yst or atntr comotraolo oovlct
>otn of cour insMCtior sy ion«r
l>19Ueu> >na otnttrttin?) or otMr
smtrtolc jnici-unainMur Un$ only
Durtfn; of I in* piw-unotnaur linn
?rt»«ntivt orognm for oyuia fsreis
Uifiraftci 12} Ono*CiU Sysun
?1j Itm for Mtir rMMl
311 tpill aitKUrs-idrtM ttrmntl , «tc.
Insottf.on
fur
>. 338
1. 26
3. 52
c. 333
6
continuous
continuous
continuous
continuous
continuous
continuous
continuous
continuous
continuous
•fttr inaic**
cion af set 11'
iftor indict-
don of spill'
s. 1
a. 2
c. tfttr i nota-
tion it spill1
i. 1
D. 2
C. 4
d. 12
31
i. C.2S
s. 0.50
C. *
«
3.5
continuous
3'
continuous
Ptrctnt aMuctlM of ValuM of Soilli
>2S: Scntdult' ><0t SOM«ilt'
A t : B £ f 5' H' ! » s : g E s1 n'
I I X I
.<
f
' i «
i
X j XX
,
• x
; * *
I
1 i "
1
1
!
i
i
X 1
i
1
:
1
X , I X X
:
x ; xx
>75«
Scntault'
* B C
XXX
X X
I
X
x
XXX
39tiM) SCMdulM to 4COMtt)lttn inaictua otrewt rfauctton gf vQim.* of spills art identified
in calums (A, 3. C....KC. } otto*.
Frequency of insoeccion is iisuMd *t cnree tines per year.
^noenMCer lines only, noitever, ocner ser.eoules shown my tUo
-------
Implementation of the One-Call System as recommended in Reference 12
is expected to result in annual savings of approximately 120,000 barrels
of petroleum (40 percent of total). This is also expected to decrease the
number of incidents by 88 (33 percent of total).
Survey of the line by inspection pigs (magnetic flux type or other com-
parable device) carried out every four years is expected to save approxi-
mately 80,000 barrels of petroleum (28 percent of total) and decrease the
number of incidents by 106 (36 percent of total). This inspection would be
particularly effective in preventing spills caused by line pipe faults.
Hydrostatic tests are recommended for lines where inspection pig sur-
veys are impractical because of line pipe size or other considerations.
Either yearly tests or tests after indication of a spill (whichever sched-
ule is most effective) are recommended. These inspections are expected to
save approximately 60 to 90 thousand barrels of petroleum (20 percent to 33
percent of the total) and decrease the number of incidents by 80 to 100 (28
percent to 35 percent of the total).
Overall, these inspections are expected to reduce the spill volume and
spill incidents by more than 66 percent. Total costs are estimated at less
than 40 million dollars per year. This cost is significantly lower than the
cost of preventive maintenance using cathodic protection systems. Yet, the
reduction of the number of incidents and spill volume is expected to be much
greater.
8.3.3.2 Underwater Lines--
Recommendations for underwater lines include:
• Visual inspections of the line by air or ground patrol - weekly
• Depth of cover inspection by sonar or other comparable device -
every two years
• Hydrocarbon probe (towfish or similar device) inspection -
yearly
• Survey by inspection pig (magnetic flux type or other compara-
ble device) - every two years
or
• Hydrostatic test - yearly.
(See also developmental and new methods Section 8.4).
8.3.3.3 Large Diameter Lines--
Recommendations for large diameter lines (greater than 16 inches in
diameter) include:
• Visual inspections by air or ground patrol - weekly
• One-Call System (Reference 12)
334
-------
• Survey by inspection pig (magnetic flux type or other com-
parable device) - every four years
• Volume comparison, flow rate comparison and pressure deviation
or
• Mathematical modeling.
(See also developmental and new methods Section 8.4).
8.3.3.4 Old Lines-
Recommendations for old lines (30 years or older) include:
• Visual inspections by air or ground patrol - weekly
• One-Call System (Reference 12)
• Survey by inspection pig (magnetic flux type or comparable
device) - every three years
or
• Hydrostatic tests
yearly, or
after indication of spill.
8.3.3.5 Lines with Two or More Reportable Spill Incidents within One Year--
Recommendations for lines with two or more reported spill incidents
within one year include:
• Visual inspection by air or ground patrol - weekly
• One-Call System (Reference 12)
• Survey by inspection pig (magnetic flux type or comparable
device) - every year
• Hydrostatic tests - yearly.
This schedule should be in effect until approximately three years after
spill incidents are reduced to a normal rate.
8.4 RECOMMENDED DEVELOPMENTAL AND NEW INSPECTION AND LEAK DETECTION
METHODS
Although a number of developmental and new inspection and leak detec-
tion methods are capable of reducing the potential of spills, only a few
are considered to have the potential of significantly reducing the oil
spill risk. These include:
335
-------
• Mathematical modeling
• Passive acoustic array
retrofit (prevent damage from outside forces)
new lines (prevent damage and detect failures).
Mathematical modeling of the line appears to be a proven new technique
but should be thoroughly reviewed and evaluated before it can be recommended
for most lines. It is recommended at this time for use on .lines that are
in high risk areas, such as underwater lines, and/or lines with a high risk
that oil may escape from a line such as those of large diameter. The
method, if used alone, is expected to reduce the expected volume of spills
by about 65 percent. It also appears to be one of the most cost-effective
methods available (see Table 78). No conclusive evidence is available to
prove that the method can prevent spills, although such events as line
pressure surges may be reduced with this method. The method should be con-
sidered with other options, particularly ones that prevent spills. Use of
this method on a large number of lines would be difficult to implement at
this time because only a few small sized commercial companies and a few oil
companies are actively involved in this area.
The passive acoustic array system for retrofitting on existing lines is
a developmental method that shows promise for most lines. It is recommended
that this method be investigated further to verify its capability of prevent-
ing damage from outside forces and rupture detection. The method, if used
alone, is expected to reduce the frequency of spills by about 30 percent and
the volume of spills by about 70 percent. The method appears to be one of
the most cost-effective methods available (see Table 78).
The passive acoustic array system for new lines is a developmental
method that is attractive for lines in certain high risk areas, i.e., under-
water lines, populated areas, etc. It is recommended that this method be
investigated further to verify its capability to both prevent failure from
outside forces and other causes and detect leaks. The method, if used
alone, is expected to reduce the frequency of spills by about 43 percent
and the volume of spills by about 84 percent. Since the method appears to
be of relatively low cost-effectiveness, its application should be limited
to lines in high risk areas.
336
-------
REFERENCES
1. Pipeline Safety Need for a Stronger'Federal Effort. PB-280321, U.S.
General Accounting Office, 1975.
2. Leggett, N., and M. Placet. A Comparison of National Air Residual in
; NEDS and SEAS. Working Paper IR&T-18, International Research and
: Technology Corp., 1978. j
! 3. Crude Petroleum, Petroleum Products and Natural Gas Liquids in the U.S.
I Annual Reports. U.S. Department'of Interior (Bureau of Mines), 1971-
1977.
i ;
i
j 4. Hittman Associates, Inc. Environmental Impacts, Efficiency, and Cost
| of Energy Supply and End Use. Volume 1. PB 238-784. Council on En-
j • vironmental Quality, Washington, ,D.C., 1974.
5. Association of Oil Pipelines. Press Release. June 7, 1975: May 26,
! 1975. | !
1 , !
6. Transport Statistics in the U.S.,' Part 6, Oil Pipeline. Annual Re-
ports. Interstate Commerce Commission (Bureau of Accounts). 1965-
1976. i
• i
1 7. Regulations on Transportation of Liquids by Pipeline. Title 49 of
Code of Federal Regulations, Part 195. Department of Transportation,
1977. ,' |
8. Petroleum Extension Service. Introduction to the Oil Pipeline Indus-
try. The University of Texas, Austin, Texas, 1966. 84 pp.
i ' '
i 9. Mineral Industry Surveys: Crude Oil and Refined Products Pipeline
I Mileage in the United States. U.S. Department of Interior (Bureau
! of Mines), 1971. I
i !
j 10. Mineral Industry Surveys: Crude Oil and Refined Products Pipeline
I •• Mileage in the United States. U.S. Department of Interior (Bureau
j of Mines),, 1974.
i ' •
i 11. Energy Data Reports: Crude Oil and Refined Products Pipeline Mileage
: in the United States. U.S. Department of Energy (Bureau of Mines),
< 1977. : !
337
-------
12. Courtney, W.J., G. Yie, and D. Kalbbrenner. Effectiveness of Programs
for Prevention of Damage to Pipelines by Outside Forces. Report DOT/
MTB/OPSO-77/12, ITT Research Institute, 1977.
13. Petroleum Storage Capacity, National Petroleum Council, 1974.
14. Mastandrea, J.R., O.A. Simmons, P.B. Kimball, and K.J. Gilbert. Deep-
water Port Inspection Methods and Procedures. Report No. CG-D-31-78,
United States Coast Guard, Department of Transportation, 1978.
15. Volumetric Shrinkage Resulting from Blending Volatile Hydrocarbons with
Crude Oils. Bull. API 2509C. American Petroleum Institute, Washington,
D.C., 1967.
16. Lease Automatic Custody Transfer. Bull. API Standard 2502. American
Petroleum Institute, Washington, D.C., _1967.__
.
17. Brooks - Bi Rotor Meters. Bulletin (S)DS-(B-42DEB). Brooks Instrument
Division, Emerson Electric Company, 1978.
18. Smith-Meter Systems. Bulletin 271. A. 0. Smith Company.
19. Karabelas, A..J. Recent Studies Improve Velocity Criteria Used for
BS&W Sampling. Oil and Gas Journal, April 17, 1978. pp. 98-104.
20. Reproducibility Between Laboratories. ASTM Standard D 287.
21. Lipkin, M.R. and S.S. Kurtz. Temperature Coefficient of Density and
Refractive Index for Hydrocarbons. Ind. Eng. Chem. Anal., Ed. 13,
1941. pp. 291-295.
22. Downer, J.L.' and F.A. Inkley. Need Shown for Separate Thermal-Expansion
Tables. Oil and Gas Journal 70:25, June 19, 1972. pp. 52.55.
23. Form P—Pipe Line Operation Expenses for Carrier Pipe Line Companies in
the U.S. Annual Reports. Interstate Commerce Commission (Bureau of
Accounts), 1965-1976.
24. Capital Systems. Digest of Pipeline Rates on Gasoline and Petroleum.
Capital Systems Publications, Rockville, MD, Annual Report.
25. Liquid Pipeline Accidents Reported to the Department of Transportation
on DOT-Form 7000-1. Annual Reports. Department of Transportation
(Office of Pipeline Safety), 1969-1977.
26. Ulrich, L. Detailed Information for Each Accident During the Period
1971 through 1975. Office of Pipeline Safety, 1977.
27. Danenberger, E.P. Oil Spills, 1971-1975, Gulf of Mexico Outer Con-
tinental Shelf. Circular 741. U.S. Geological Survey, 1976.
338
-------
28. Conservation Division. Accidents Connected with Federal Oil and Gas
Operations on the Outer Continental Shelf. U.S. Geological Survey,
Conservation Division, 1977.
29. Pollution Incidents in and Around U.S. Waters. U.S. Coast Guard Publi-
cation, Pollution Incident Reporting System (PIRS). CG-487. 1975 and
1976.
30. Ritchie, J.E., Jr., et al. Petroleum System Reliability Analysis.
Volume I and II. Report No. EPA-R2-73-820b, U.S. Environmental Pro-
tection Agency, Washington, D.C., 1973.
31. Leo, J.E. and G.G. Kruijer." Spillages from Oil Industry Cross-Country
Pipelines in Western Europe, Statistical Summary of Reported Incidents,
1966-1969. Stichting CONCAWE Oil Pipelines Working Group, 1971.
32. King, E.M. and G.G.Kruijer. Spillages from Oil Industry Cross-Country
Pipelines in Western Europe, Statistical Summary of Reported Incidents,
1971. Stichting CONCAWE Oil Pipelines Advisory Group, 1972.
33. King. E.M. and P. Rogier. Spillages from Oil Industry Cross-Country
Pipelines in Western Europe, Statistical Summary of Reported Incidents,
1972. Stichting CONCAWE 011 Pipelines Special Task Force 1, 1973.
34. King, E.M. and P. Rogier. Spillages from Oil Industry Cross-Country
Pipelines in Western Europe, Statistical Summary of Reported Incidents,
1973. Stichting CONCAWE Oil Pipelines Special Task Force 1, 1974.
35. King, E.M. and P. Rogier. Spillages from Oil Industry Cross-Country
Pipelines in Western Europe, Statistical Summary of Reported Incidents,
1974. Stichting CONCAWE Oil Pipelines Special Task Force 1, 1975.
36. King, E.M. and P. Rogier. Spillages from Oil Industry Cross-Country
Pipelines in Western Europe, Statistical Summary of Reported Incidents,
1975. Stichting CONCAWE Oil Pipelines Special Task Force 1, 1976.
37. King, E.M. and V. Baradat. Spillages from Oil Industry Cross-Country
Pipelines in Western Europe, Statistical Summary of Reported Incidents,
1976. Stichting CONCAWE Oil Pipelines Special Task Force 1, 1977.
38. Mackay, D., et al. The Physical Aspects of Crude Oil Spills on Northern
Terrain. Task Force on Northern Oil Development. Report 73-42; In-
formation Canada Catalog R72-9173, 1974.
39. Beyer, A.M. and L.J. Painter. Estimating the Potential for Future Oil
Spills from Tankers, Offshore Development and Onshore Pipelines. In:
Proceedings of the 1977 Oil Spills Conference, Prevention, Behavior,
Control, Cleanup. Am. Petroleum Institute, U.S. Environmental Protec-
tion Agency, and the U.S. Coast Guard, New Orleans, LA, 1977.
339
-------
40. Stewart, Robert J. The Tanker/Pipeline Controversy. Proceedings of
the 1977 Oil Spills Conference; Prevention, Behavior, Control, Clean-
up, Petroleum Institute, U.S. Environmental Protection Agency, and the
U.S. Coast Guard, New Orleans, LA, 1977. pp. 95-99.
41. Operations Research, Inc. Spill-Risk Analysis Program: Interim Report
E-2: Facilities Analysis, 1975.
42. Young, G.K., D. Evans, and R.U. Jettmar. Analysis of Oil Spill Trends.
Report to the Council of Environmental Quality, Alexandria, VA, 1976.
43. Bayaert, Bruce. Analysis of Oil Spill Accidents for Environmental Im-
pact Statements. In: Proceedings of the Conference on Prevention and
Control of Oil Pollution. American Petroleum Institute, U.S. Environ-
mental Protection Agency, and U.S. Coast Guard, San Francisco, CA, 1975.
pp. 39-45.
44. Paulson, A.S., A.D. Schumaker, and W.A. Wallace. A Risk-Analysis Ap-
proach to Control of Large-Volume Oil Spills. In: Proceedings of the
Conference on Prevention and Control of Oil Pollution. American
Petroleum Institute, U.S. Environmental Protection Agency, and the
U.S. Coast Guard, San Francisco, CA, 1975. pp. 301-106.
45. Wilson, J.D. A Statistical Estimate of Pipeline Leakage. Report SAI-
79-162. Santa Ana, CA, 1977.
46. Col burn, C. Methods of Preventing, Detecting and Dealing with Surface
Spills of Contaminants which may Degrade Underground Water Sources.
Report EPA 68-01-4620, U.S. Environmental Protection Agency, Office of
Water Supply, Washington, D.C., 1978.
47. Boyd, B.D., C.C. Bates, and J.R. Harrald. The Statistical Picture Re-
garding Discharges of Petroleum Hydrocarbons in and Around United
States Water. Sources, Effects and Sinks of Hydrocarbons in the
Aquatic Environment. In: Proceedings of the Symposium, American In-
stitute of Biological Sciences, American University, Washington, D.C.,
1974. pp. 38-53.
48. Petroleum in the Marine Environment. National Academy of Sciences,
1975.
49. Kiefner, J.R. and R.B. Smith. An Analysis of Reportable Incidents for
Natural Gas Transmission and Gathering Lines, 1970 through 1975. Re-
port NG-18, Report No. 106 to the Pipeline Research Committee, American
Gas Association, 1977.
50. Marine Environmental Protection Program: An Analysis of Mission Per-
formance. United States Coast Guard, Department of Transportation,
Washington, D.C. 1975.
340
-------
51. Funge, W.J., K.S. Change, D.I. Juran, et al. Offshore Pipeline Safety
Practices, Volume II - Main Text. Report DOT/MTB/OPSA-77/14, 1977.
52. Jones, S.C. and Roszelle, W.D. Graphical Techniques for Determining
Relative Permeability from Displacement Experiments. Journal of
Petroleum Technology, 1978.
53. Blocker, P.C. Migration of Oil in Soil. Paper Presented at Interna-
tional Conference, "Antinguinamento 71," 1971.
54. Dan J. Van. The Migration of Hydrocarbons in a Water Bearing Stratum
Of Stichting CONCAWE, The Hague, 1966.
55. Duffy, J., M.F. Mohtadi and E. Peaka. Subsurface Persistance of Crude
Oil Spilled on Land and Its Transport in Groundwater. Proceedings 1977
Oil Spill Conference. American Petroleum Institute, U.S. Environmental
Protection Agency, and U.S. Coast Guard, New Orleans, LA, 1977.
56. The Migration of Petroleum Products in Soil and Groundwater. Publica-
tion No. 4149, American Petroleum Institute, Washington, D.C., 1972.
57. Goodfellow, R. Underwater Engineering. Tulsa, OK, 1977.
58. Platus, D.L., et al. Rapid Shutdown of Failed Pipeline Systems and
Limiting of Pressure to Prevent Pipeline Failure Due to Overpressure.
Report PB 241-325, U.S. Department of Transportation, 1974.
59. McFarlane, C. and R.D. Watson. The Detection and Mapping of Oil on
a Marshy Area by a Remote Luminescent Sensor. Proceedings 1977 Oil
Spill Conference. American Petroleum Institute, U.S. Environmental
Protection Agency, and U.S. Coast Guard, New Orleans, LA, 1977.
60. Jackson, P. Leak Detection in Underwater Oil Pipelines. Report NMRC
272-23100-R2, National Maritime Research Center, 1973.
61. Oil Pollution Detection and Sensing - Bibliography with Abstracts.
Search Period 1964-July 1976. Report NTIS/PS-76-0701, NTIS, 1976.
62. Rambie, G.J., Jr., R.H. Morgan and R.J. Jones. Feasibility of Contin-
uous Monitoring for Oil Pollution Across Channels and Rivers. Proceed-
ings 1977 Oil Spills Conference, Prevention, Behavior, Control, Clean-
up, American Petroleum Institute, U.S. Environmental Protection Agency,
and U.S. Coast Guard, New Orleans, LA, 1977.
63. Alford, B.J., M.M. Paterson, and F.A. Womack. Development and Field
Evaluation of the Production Surveillance Monitor. Journal of
Petroleum Technology, 1978.
64. Jouve, P. Process and Apparatus for Detecting a Fluid Leak from a
Pipeline. Belgian Patent No. 823,862, 1977.
341
-------
65. Holland, W.E. and G.R. Burrell. Acoustic Method for Detecting Leaks
from Submerged Pipelines. U.S. Patent No. 4,001,769, 1977.
66. Ells, J.W. and A. Roberts. Underwater Pipelines. U.S. Patent
No. 3,992,924, 1976.
67. Boyens, V.C. Pipeline Leak Locator Method. U.S. Patend No.
4,016,748, 1977.
68. Bielawa, R. and H. Howe. Radar Unit Locates Buried Plastic and Metal
Lines. Pipeline Industry, July 1977.
69. Farmer, W.M., et al. Feasibility of Scour Monitoring. Report 1-SAI-
75-503-77, U.S. Department of Transportation, Washington, D.C., 1975.
70. Mastandrea, J.R. and J.M. Zuieback. Feasibility of Scour Monitoring
Instrumentation. Report SAI-260-76-518-LA for U.S. Department of
Transportation, Washington, D.C., 1976.
71. Underwater Inspection/Testing/Monitoring of Offshore Structures. Con-
ducted by R. Frank Busby Associates. Sponsored by DOT, DOE and DOI,
Contract 7-35336, February 1978.
72. Los Angeles Times.
73. Phil pot, F.V. and A.D. Higham. Pig and Leak Location Undersea Pipe
Lines. Presented to the Oceanology International Conference, Brighton,
England, 1978.
74. These Inspection Probes Check Internal Rise Corrosion. Ocean Industry,
February 1978.
75. Ultrasonic Rises Inspection Tool Successful. Ocean Industry, August
1978.
76. Metzler, J.A. Acoustic Holography may Permit Buried Pipe Line Inspec-
tion. Ocean Industry, May 1978.
77. Dodd, V.R. Integrated Machinery Inspection Program Cuts Maintenance
Costs. Oil and Gas Journal, April 1978.
78. Boor-man, R.D. and A.R. Walledge. Current Method of Routine Testing
and Leak Detection in Operating Oil Line. Report No. 5/73, Stichting
CONCAWE, The Hague, 1973.
79. Catagnet, A.C. Application of Radioisotopes in Oil, Gas and Petro-
chemical Industries - Transport of Hydrocarbons. Institute de Energia
Atomica, Sao Paulo (Brazil). Div. de aplicacao de Radioisotopos na
Engenharis e na Industrie, August 1976.
342
-------
80. Published Regulatory Guidelines of Environmental Concern to the Oil
Industry in Western Europe. CONCAWE Report No. 2/77.
81. Baker, J.M. Marine Ecology and Oil Pollution: The Work of the Oil
Pollution Research Unit.
82. Annual Report to Congress by National Transportation Safety Board, U.S.
Government Printing Office, Stock No. 5000-00084, 1975.
83. Dubiel, E.J. The Practical Aspects of Litigating an Oil Spill (Plain-
tiff's Viewpoint). Proceedings 1979 Oil Spills Conference, American
Petroleum Institute, U.S. Environmental Protection Agency, and U.S.
Coast Guard, Los Angeles, CA, 1979.
343
-------
APPENDIX A
EXISTING U.S. AND FOREIGN REGULATIONS FOR TRANSPORTATION
OF LIQUIDS BY PIPELINE
A.I TITLE 49-U.S. TRANSPORTATION OF LIQUIDS BY PIPELINES
TITLE 49 PART 145
A.I. TITLE 49-U.S. TRANSPORTATION OF LIQUIDS 8Y PIPELINES
OffAOTMENT OF TRANSPORTATION
MAI6UAU TRANSPORTATION 3U8EAU
WAiMMOTOM. 3-C HIM
no***^"«aiiapaftt*1aii "OtmatioB of ta* ConaMBtB" Motion
awmx t—MATDBAts TMNWOajTA. of -JO* snamal*. Cammnui vatca air-
«ug«»u. amutnum or mod m* c&aatai ouoid* ta* «ap* ot
ta* ui imuiaa noae» an act ilUmiMil big
say b* coaattcnd ay ta* 'ntanili
^•acporcaaaa "Inratn "» «* 0°*•"•* «•» nanane* ot ta*
URAlANOOTHCltaMSr MmiNK ptopoaU an aot dUeocMd Swat
.-_,_, «_ .__ OUemtnon at COHMWRCK. Wtta refart
^^^^ ^T2U2T JSSLi*"1' nl11"* ca ca> utm -small" la ta* tnapoHd
""•"• "•"•""• iiaa.tuui.an <
11II ill I •JitaniliTiiiiinaiiiliiii nil ta*ax«
.mfc Offle* of ttgtUa* Sattcr Opat»> to* t*i
QODA Qcpa^Bian of TiaiuuanaooA* *uuJd b* OD*B to ladtttdual l
ACSOail
P«naM» ttrdr. taawc
aanatatCTttaiaaaiaintBaaiaQlaiBa tatra* of ta* L.
pipvUfiiB wtt&om cntiBs, csctiocuo pro* toctui or !•• t& c
•MCMaV ISd tBOHlfmttlty VtM& •\dVQUftW up !••• 4« I*MMLH T^M (
. ___L|l_i_i_||_i. -nnr^i I i. i •• .1 «7 J t»M ^^ l^*»»«* *•*• »
nCVUw CORIwHB UMAUUI U LUUfMMQ OT
ailorafa. Tb* fatt Mfjop and •BBBOBB* .
tdvaBtac* of ta*H stciavi "*fflT*** b* •^••i*^ ^pi^u^ ^ ^«»i^.y ^ 'li'iiiiin^ur
nattnd ujular t&* PVCMBS rala b*caua Jaalaud. ?sra ot&ar oooiB*Bt*n iraiTntf
of t&* -«a» tag batttaa of srenatn* • tSat-aay .—"y""-i of atoao jaoold
S^BC^^nB OA7E: "Hsla •H**'1'"^'1*
I on ta»oj*
of plattle tttp* la CM nmea by ta* arail-
_.__. aU* lomsu nwtooda aad by ta* COM of
""-1- TH"rmi- iPmiutly. 13-inca dJaautcr
(tn-Ot-nm- plame ntp* u tb* lan*n nomaily ua«d
mmtiwrmw' * «*• MnrtMJ TlMM coouacaua alM
CTOOKMATIDW. pottMd out taat ca* nun ntaraat tan
of iaf«y woald b* mwcta* ta* BtopoMd
nojusvBMBt t* d*BMoatfac* taat umo*
rt«. (• nMB « onotenu yirthar. o& fH<*
mn'T.' ^^« SS^'SST'SS.S S««. ^« TSSeal Ptoalla* Saf«y
!22LdiB^2S53St^£SSK «*3arta gamma* <0nc^MH4-
i^4iTr.w,iSf^ar^sssss« tSsBS^Ss^H
by nmttus ta* opantar ta >aow by
tana, laraBUaaoo. or ttpirnnr* taat
adfqnat* oaoeaMB ooosoi ta obtain**
On ScpumMr a. HIS. tt»
Taaponattoo 3onM (MTB> mud s
aooo of .uugumit retaoaldac. Mode*
NO. TS-l. (41 PR 4381. StOCHBlMr TT.
not laakac* i
noaaitan*. aad taat ta* opanor «•••
J^Sd.fbV'lISSa^'SSSSS^ -firSTS-m J«3 ao-d tnat tj»
Mntw«r»gtTitida«nanUf»ntt«ndaca.
nir... or tivooaa by Wmobar I. iTtte'tam or toy eompanal* rmtOf
Tn^^n^p.^, *.«— S^tSSSS^
stae. cnuluorr aonete. I tnd. ajj SSslSar'e.^^SS
naoon. and l maanricrorp of imnn SS'.SSU^S^fSSSTSLUL^I
AU :, ceomuuta «m faTCnbla, a ISSSfySrV^-SS^Ski^II
', to ta* uruuuata aauBdmcBi. A iitomlnn of ta* ana -naaU could r*»
«« tB*nSSaatM^a«Ba mlt in noaoaifana apoUcaHon of ) IS2.-
wmca «.^M aTdSSL't.'SS. * •«>trgSSffSS* »
rn rn« nuiiiimii iniTntlmim* "v< "I-^T- AUO. a ax* ilnnfafinn 11 tgieuiuy en-
WHHHBH •jiiwuimim* ami *om MI*~ ~~, . _ _ t . - _j i*n_
of tao** eonmuBia a drrdoo. *"• « snaiat ^irmototr aau ««o»
iaal mliuoonaiMd am* acoue* ™»aMd to t&* us* of pla.de plpa.
344
-------
More significant than size, however.
in protecting against corrosion is the
fact that as discussed hereafter the op*
erator would be retmired under 5 192.455
?3) is intended to;
provide for anv future inspection, repair.
or replacement that might be required as
a result of future rulemaking should any
new information indicate a need for such
remedial action.
In addition, MTB requests that oper-
ators voluntarily report the condition of
any alloy fitting installed under } 192.455
(f) which is uncovered for any reason.
MTB Is interested in receiving reports on
corrosion performance of the fittings, es-
pecially any leakage of a fitting that is
not required to be reported under
1} 191J and 191J of this chapter, and the
number of fittings installed. These re-
ports could be submitted by operators in
letter form and need not be submitted
more often than once a year, unless the
operator desires to report more fre-
quently. MTB expects that information
obtained through the voluntary report-
ing may serve as a basis for a future rule-
making action either to relax the restric-
tions applicable to exemption under
§ 192.455 (f) or to prescribe any necessary
remedial measures, as the case mav be.
345
-------
Regarding the proposed { 182.459(1)
(2). the TPSSC further suggested that
the term "corrosion pitting" be replaced
by "corrosion attack." This comment was
not adopted for the sake of consistency
since the term "corrosion pitting" is used
elsewhere in Pan 192.
Another commenter thought that an
operator should not have to use tests. In-
vestigation. or experience "In the area of
application" to show under 3 192.455(f)
(1) that alloy fittings provide adequate
corrosion control. This commenter al-
leged that the testing, investigation, or
experience in the corrosion studies re-
ported in the National Bureau of Stand-
ard's (NBS) Circular No. 579 and two
California field studies mentioned in the
Notice are sufficient to allow a general
exception without the need for an indi-
vidual finding by each operator.
MTB does not agree. The NBS study
compares the performance of certain
materials under a limited number of en-
vironments. It did not establish a means
to quantitatively measure the corroslvitv
of anv environment in which a material
might be used. Also, the-two field studies
conducted In California do not have uni-
versal application to all soils. Those stud-
ies are more indicative of local condi-
tions. They include the type of testing
and investigation that an operator might
conduct in an area to determine whether
flttlntts are adequately nrotected affainst
corrosion by allovage. For these reasons.
MTB did not adopt the suggested change
in the final rule.
or THE TECHNICAL Pirsum
SAFETY STANDARDS COMMITTEE
Section 4(b) of the Natural Gas Pipe-
line Safety Act of 1968 requires that all
proposed standards and amendments to
such standards pertaining to gas pipe-
lines be submitted to the Committee and
that the Committee be afforded a rea-
sonable opportunity to prepare a report
on the "technical feasibility, reasonable-
ness, and practicability of each proposal."
The proposed amendment was submitted
to the Committee as Item A-l in a list of
two proposed amendments at a meeting
In Washington. D.C., on December 16 and
17, 1978. A minority report was not sub-
mitted.
On February 3. 1977. the Committee
filed the following favorable report:
This communication 13 the official report
of the Technical Pipeline Safety Standards
Committee concerning the Committee's ac
tion on two amendments to 49 CFR Part 192
proposed by the Office of Pipeline Safety Op-
erations and other matters which the Com-
mittee decided should be brought to the at-
tention of the Department of Transportation.
The fbllowlng described actions were taken
by the- Committee at a meeting held In Wash-
ington. D.C.. on December 18 and 17, 1976.
Item A-l of the agenda was a proposal by
OPSO to revise i 192.455. External corrosion
control. By an affirmative vote of 12-1 the
Committee found that the following lan-
guage for i 192.455 Is technically feasible.
reasonable, and practicable.
(The language suggested Is adopted In the
flnal rule except as discussed in the "Dis-
cussion ot Comments Section" above.)
PRINCIPAL AUTHORS
Ralph T. Simmons. Regulations Soe-
cialist. Georee Mocharko, Staff En-
gineer, and Robert L. Beauresard. Attor-
ney, Office of the General Counsel.
In consideration of the foregoing.
i 192.455 of Title 49 of the Code of Fed-
eral Regulations is amended bv amend-
ing paragraph (a) and adding a new
paragraph (f) to read as follows:
§ 192.455 External corrosion control:
buried or submerged pipelines in-
stalled after July 31,1971.
(a) Except as provided in paragraphs
(b). (c>, and (f) of this section, each
buried or submersed pipeline installed
after July 31. 1971. must be protected
against external corrosion, including the
following:
• • • • •
(f) This section does not apoly to
electrically isolated, metal alloy fittings
in plastic pinelines if—
(1) For the size fitting to be used, an
ooerator can show by tests, investiga-
tion, or experience in the area of appli-
cation .that adequate corrosion control
is provMud h" n.Uova»e:
(2) The fitting is designed to prevent
leakage caused by localized corrosion
pitting; and
(3) A means is provided for identifying
the location of the fitting.
(49 USC 1672: 49 CFR 1.53 (a).)
Issued in Washington. D.C., on July
1-. 1977.
ALAN A. BCTCHMAN,
Acting Director. Materials
Transportation Bureau.
[PR Doc.77-19421 Filed 7-8-77:8:45 am)
346
-------
the Notice but feels that It is within the
broad scope and Intent of the Notice.
and therefore it is appropriate to include
it in the final rules. However, in keeping
with MTB's policy to ensure that the
public has full opportunity to participate
in the rulemaUing process, MTB is delay-
ing the effective date of 5 192.313 (a) (4)
(B) and 1 19S.212(b)(3)(B) until No-
vember 3, 1977 to permit any interested
person the opportunity to comment be-
fore the rule becomes effective. If no ad-
verse comment is received that raises
substantial doubt as to the desirability of
the amendment, it will become effective
November 3. 1977 as written.
RCFOKT or TXC TECHNICAL PXTEUNX
SAFETY STANDARDS COMMITTED
Section 4O» of the Natural Gas Pipe-
line Safety Act of 1968 requires that all
proposed standards and amendments to
such standards pertaining to gas pipe-
lines be submitted to the Committee and
that the Committee be afforded a rea-
sonable opportunity to prepare a report
on the technical feasibility, reasonable-
ness. and practicability of each- proposal.
Ibis amendment to Part 192 was sub-
mitted as Item A-2 in a list of two pro-
posed amendments at a meeting in
Washington, D.C.. on December 16 and
17. 1976. On January 12. 1977. the Com-
mittee filed the following favorable re-
port* A Tnfajmi'lt^ report was not ftT^fl
Thl» communication is the official report
of the Technical Pipeline Safety Standards
Committee concerning tb* Committee '» ac-
tion on two amendment* to 48 cm Part 183
propoied by the Office of Pipeline Safety Op-
erations and other matteri which the Com-
mittee decided should be brought to the at-
tention of the Department of Transporta-
tion.
Tb* following described action* were
taken by the Committee at a meeting held
in Washington. O.C. on December 18 and 17.
1878.
Item A-2 wa* a proposal by OPSO to re-
vise 1 182.313 (a) (4), Bend* and elbow*. By
an affirmative vote of 13-1 the Committee
found that the following language for I 193.-
313(a) (4) If technically feasible, reasonable.
and practicable.
• • • • •
(The language suggested is adopted In the
final rule M diaeimed In the "Discussion of
Comment* Section" above.)
After additional discussion* of agenda Item
A-3, by an affirmative vote of 13-1, the Com-
mittee further recommended that I 183.313
be further modified to provide that for pipe
with a D/t ratio less than 70. the location of
the longitudinal seam may be at the discre-
tion of the operator.
PMWOTAI AUTHORS
Ralph T. Simmons. Regulations Spe-
cialist, and Robert L. Beauregard, Attor-
ney Office of the General Counsel.
In consideration of the foregoing.
Parts 192 and 195 of Title 49 of the Code
of Federal Regulations are amended as
follows:
1. Section 192.313(a><4> is amended
to read as follows:
§ 192-313 Bend* and elbow*.
(a) * ' '
(4) On pipe containing a longitudinal
weld, the longitudinal weld must be as
near as practicable to the neutral axis of
the bend unless—
(i) The bend is made with an Internal
bending mandrel; or
(11) The pipe is 12 inches or less ih"out-
side diameter with a diameter to wall
thickness ratio less than 70.
(See. S. Pub. L. 80-481. 83 Stat. 731. 48 T7SO
1672: for offshore gathering lines. Sec. IDS.
Pub. b. 83-833. 88 Stat. 21S7. 48 USC 1804: 40
TB 43801.48 CPU 1.53.)
2. Section 195.212 (b) (3) is amended
to read as follows:
8195.212 Bending of pipe.
• • • • •
(bJ ' • •
(3) On pipe containing a longitudinal
weld, the longitudinal weld must be as
near as practicable to the neutral axis of
the bend unless—
(i) The bend is made with an internal
bending mandrel; or
(11) The pipe is 12 inches or less in out-
side diameter with a diameter to wall
thickness ratio less than 70.
• • • • •
(See. 8. Pub. L. 89-670, 80 Stat. 937. 48 U.S.C.
1655: 18 U.S.C. 831-835; 40 PR 43901. 49 CFR
1.S3.)
JOHN J. FEARNSIDES.
Acting Director.
Materials Transportation Bureau.
[PR Doc.77-24303 Piled 8-24-77:8:45 amf
348
-------
DEPARTMENT OF TRANSPORTATION
MATERIALS TRANSPORTATION 3U8EAU
WASHINGTON. O.C 20390
CHAPTER I—MATERIALS TRANSPORTA-
TION BUREAU. DEPARTMENT OF
TRANSPORTATION
suaoumft o nnutn swcrv
[Atoota, iS3-» ua.13: OookM Mo.
Taere wen six person* who responded
and suamitted written comments to No-
tice 75-i Tarea won from EU dlambu-
tton companies sad three -we from
trade mmrntlOTM. A *<««M«i««q of the
comments aad the reeom-
PART 193—TRANSPORTATION OP
NATURAL AND OTHER OAS 8Y PIPELINE
PART 136—TRANSPORTATION OF
LIQUIDS 3Y PIPELINE
lonejuidlnei Seems In Plpa-Benda
AGENCY: **>***tnli Ttaasportaaaa
ACTION: Fmalruift.
beads of steel
pipe to be placed other than near the
mandrel is used or when bending pipe of
12 inches or leu m outtttde diameter that
baa a diameter to wall fM* (3) (B) do
not beeooM efiiettra anal November 3.
13TT.
flBrir*ny to con*
meat oa Section 19Z313(am> or
Section lS5-2ia(b)(3) (B> snonld com-
seat la wiiuiic to*
Dtreetor. Qfle* of TIpamw Safety Op-
endoBi. Department of Traasportaaan.
noo Second Street. 3.W, Waiam«toa.
D.C. 20890.
Comments win be avmUabl* at Docket
Room 3SOO. 2100 Seeoad Street, S.W..
Waabiaiton. O.C.
An of tae coauaeBten aad *^^ T?SSC
supported the proposal as published la
the Notice. Tin* reasons wen that the
particularly wita thtt use of **w lateraalr
bending mandrel, have made the resale-
is, a b^tut uaaacessary. Ibey concluded
that operators aad carriers should be
allowed to utilize this Improved teca-
to produce a high quality pipeline aad
shown the old requirement to be obsolete.
MTB n gr lies wita the eommenters aad
tae TFSSC. that operators should be al-
lowed to take full advantage of improved
«eidm« tad bendiag technology that is
not *F»"»«-teat with pipeline safety.
tton, MTB la aot aware of any failures
in the '""r*"^"-* weld seam of pipe
weld seam placed oth*r than near the
neutral axis. *•<'"*"—"r MTB is of the
aura ta I! 19X313 ""1 198J12 are sof-
aelent to ensure that say pipe wita a
damaged wvid seam would be detected
and rejected before being placed la serv-
ice. Tae edvaaeea la pipe maaufaccur
TOR P
TACT:
IOTOHMA33OK CON-
(202) 428-2392.
SDPP12MENTAKT
Toe Matertiilt Transportation Bureau
(MTBX lamed a node* of proposed nue-
maktai; Node* No. T8-3 (41 PR 4««3.
October 21, 1978), proposing to amead
i 19Z313(a> (4) of tae ?adenl gas pipe-
line safety standards and ! 13i212(b> (3)
of toe Federal liquid pipeiia* safety
standards to permit tne field beadtac of
steel pipe with lonattudJnal -wins -*»tn-
oat p«^*n»f; tae loncttodmal weld n^ir
tae neutral aids of tne bend If aa inter-
nal bendrns mandrel is used. Interested
person* were invited to peraetpate la
rulemaklai action by
make the i smilre
meat for piaclag the longitudinal wvid
la a neutral axis when bending wita aa
internal bending mandrel uaaeeessex*
Tae rrtaTaBnn
_
One ^*t>*l'tl*lf'*T agreed witb tae pro*
2CXB •rTnf'But lt» tnvestigaUoa to sub-
stantiate tag need fop tag reQuiremeat
^^ a longitudinal weld be placed near
tae neutral axis during bending when
aa Internal >«t»»«ny mandrel Is not used.
T&e reason given was that the proposal
wul provide the operator with a choice
of methods for sett bending of pipe.
MTB win continue tta investigation
tton from inowledgeable sources. Such
information will be considered by MTB
for future rttlemaJdag proceedings oa
pipe beadiac.
Three eommenters aad the t?SSC
suggested that aa internal ^"t'tlT'g man-
drel la ^^*rftyrrrffrtatfe for bendiag small
pipe, aad use of the mandrel should not
be adopted sa a condition to not placing
the Tungitudmal seaax n^ti* the neutral
axis. Their argument waa that for large
aa internal bending mandrel is not un-
reasonable because latemal n^n^^g
mandrels an required to obtain accept-
able field bends. However, for small dl-
ametar pipe, particularly 12 tnchy ****
t^Ha^ with a diameter to wall thickness
(D/t) ratio of less than 70. internal bend-
Ing ********* an aot needed to achieve
acceptable beads. Taey farther stated
that, when using electric resistance weld-
ed pipe that has been weathered, cleaned.
and coated, ft la very difficult to locate
the '""T**1"""*1 weld seam.
available information. MTB believes that
safe beads &x steel pipe 12 laches or
lesa la outside diameter with a D/t ratio
of leas than 70 can be made without using
fchfwyK rt^ loagltttdiaal ^^m is not
placed near taa neutral axis of the bead.
Farther. MTB baa aot jeeelved any re-
ports of failure of beat pipe of 12-lach
diameter or less with a D/t ratio of lesa.
than 70 that can ba attributed to the
was aot used or that taa location of the
Irnigltncltnal weld seam la bending waa
a contributing factor.
Therefore, la view of the favorable m-
fftrmalilTfi *"*t ^^ aheencti of aay mfor—
marton to tae contrary. MTB is c-f the
opiaioa fhat omittiag the proposed coa-
'littftn fhat »^ lateraal bendiag niaadrel
be used whea bending pipe of 12 Inches
or less la outside diameter with a D/t
ratio leas, than 70 as aa alternative to
placing the weld seam near the neutral
a^jfl jg OQ( contrary to the public inter-
est nor inconsistent with, pipeline safety.
For the foregoing reasons. MTB has
further amended M l92J13(a><4) and
19B.212(b><3) by adding to each section
a aew subdivision (B> to allow the field
bending of small diameter steel pipe with
a longttndiaal seam without placing the
seem near the neutral axis liTeapeutive
of whether a h*"'<<''g maadrel is used.
UTS is cognizant of the fact that this
Issue wea not specifically addressed la
written data. Ttews. or arguments aot
later than JTovenber 8. 1978.
347
-------
DEPARTMENT OF TRANSPORTATION
MATERIALS TRANSPORTATION BUREAU
WASHINGTON, D.C 20390
[Amdts. 182-29; 195-12; Docket No. OPSO-38]
PART 192—TRANSPORTATION OF
NATURAL AND OTHER GAS BY PIPELINE
PART 195—TRANSPORTATION OF
LIQUIDS BY PIPELINE
Longitudinal Seams in Pip* Bends;
Correction
AGENCY: Materials Transportation Bu-
reau. DOT.
ACTION: Correction.
SUMMARY: This document corrects a
final rules document that appeared at
page 42865 in the FEDERAL REGISTER of
Thursday, August 25, 1977 (FR Doc. 77-
24303).
EFFECTIVE DATE: November 3.1977.
FOR FURTHER INFORMATION CON-
TACT:
Peggy Hammond, 202-426-0135.
SUPPLEMENTARY INFORMATION:
By Amendments 192-29 and 195-12, new
{{ 192.313(a) (4X11) and 195.212(b><3>
(ii) were added, respectively, to Parts 192
and 195 to provide t-**frt the longitudinal
seam of steel pipe need not be placed
near the neutral axis during bending
if—
"The pipe is 12 inches or less In outside
diameter with a diameter to wall thickness
ratio less than 70."
As stated in the preamble, the ra-
tionale for adopting this provision was
that "safe bends in steel pipe 12 inches or
less in outside diameter with a D/t (di-
ameter to thickness) ratio of less than
70 can be made without using an inter-
nal bending mandrel even though the
longitudinal seam is not placed near the
neutral axis of the bend." This rationale
purportedly was based on comments re-
ceived on Notice 76-2 (41 FR 46463, Oct.
21, 1976), which proposed to remove the
requirement for placement of the longi-
tudinal seam near the neutral axis when
a bending mandrel is used. Recently,
however, several interested persons have
pointed out that both the final rule and
the rationale incorrectly reflect the writ-
ten comments in the docket and the po-
sition of the Technical Pipeline Safety
Standards Committee (TPSSC). These
persons have stated that the view of com-
menters and the TPSSC was that pipe
12 inches and under in diameter can be
bent safely without a mandrel and with-
out placing the longitudinal seam near
the neutral axis, irrespective of the D/t
ratio. In addition, they stated the record
shows that any size pipe with a D/t ratio
of less than 70 can likewise be bent
safely.
After thoroughly reviewing the mat-
ter, it appears that Amendments 192-29
and 195-12 are in fact inconsistent with
the record as the interested persons have
stated.
Accordingly, the following corrections
are made:
1. Section 192.313side diameter or has a diameter to wan
thickness ratio less than 70.
* • * * *
(Sec. 6, Pub. L. 89-670, 80 Stat. 937. 48 USC
1655; 18 USC 831-885; 49 CPR 1.53.)
Issued in Washington. D.C.. on Novem-
ber 18, 1977.
It. D. SAMTMAH,
Acting Director,
Materials Transportation Bureau.
[PR Doc.77-33914 Filed 11-23-77:8:45 am]
349
-------
DEPARTMENT OF TRANSPORTATION
RESEARCH AND SPECIAL PROGRAMS DIRECTORATE
WASHINGTON. D.C. 2O59O
into
CHAPTER I— MATERIALS TRANSPOR-
TAT1ON BUREAU, DEPARTMENT OF
TRANSPORTATION
SWCHAFTB D-miM SAftlY
Umdtt, 193-92,198-14: Docket Wo. 77-10]
PART 192—TRANSPORTATION OP
MATURAL AND OTHER GAS BY
P1PEUNE
PART 195—TRANSPORTATION OP
LIQUIDS BY PIPELINE
Qualification and D«*ia.n of Stad
Pip*
AGENCY: Material* Transportation
Bureau.
ACTION: Final rale.
SUMMARY: This amendment updates
the existing incorporation by refer-
ence of API Standard 5LS. "API Speci-
fication for Spiral-Weld Line Pipe."
and API Standard 5LX. "API Specifi-
cation for High-Test Line Pipe." to in-
clude in Part 192. the March 1976 Sup-
plement and the 1977 edition of each
document and in Part 195. the 1977
edition of each document.
DATE: This amendment •becomes ef-
fective June 1.1978.
FOR yutmmK INFORMATION
CONTACT:
Frank E. Fulton. 202-426-2082.
SUPPLEMENTARY INFORMATION:
This amendment TTIP^M Parts 192 and
198 conform with recent developments
in the manufacture and design of steel
pipe. These subjects are now regulat-
ed, in part, through an incorporation
by reference of API Standard SIS and
API Standard 5LX. At present, the
1975 editions are the latest applicable
editions of-API 5LS and SIX listed in
Parts 192 and 195. This amendment
updates the lists to include the 1977
editions in both parts and the March
1976 Supplements in Part 192.
Of particular importance is that by
referencing the March 1976 Supple-
ments and the 1977 editions of API
5LS and 5LX, pipeline operators will
be permitted to use Grade X-70 pipe
In the transportation of gas. Grade 3C-
70 is more economical for certain uses
than other available grades of steel
pipe because of its high strength.
which permits the use of thinner
walled pipe. It is projected for use in
the pipeline approved under the
Alaska Natural Gas Transportation
Act of 1976 (15 U.S.C. 719) to trans-
port gas from the North Slope to the
lower 48^States.
The Office of Pipeline Safety Oper-
ations proposed adoption of the later
editions of API 5LS and 5LX in a
Notice of Proposed Rulemaking
(Notice 77-7) issued on December 7,
1977 (42 PR 62397, Dec. 12. 1977). In-
terested persons were invited to par-
ticipate in the rulemaking proceeding
by submitting written data, views, or
arguments by January 12,1978. In ad-
dition, in accordance with Sec. 4(b) of
the Natural Gas Pipeline Safety Act of
1968 (49 U.S.C. 1673
-------
an immediate need to pursue the
matter further. It is recognized, how-
ever, that under certain conditions of
high stress and low temperature, the
potential for fractures may increase in
thin walled pipe. Under these condi-
tions, operators are already required
by the general design requirements of
Parts 192 and 195 to take additional
precautions (see §§ 192.53 and 195.102).
Further consideration will be given to
this problem in a future rulemaking
proceeding on pipelines operating in
low temperature environments.
Several commenters suggested that
the introductory language in Section
II of Appendix A and Section I of Ap-
pendix B to Part 192 and in § 195.3(a)
be amended to permit the use of com-
ponents which provide a comparable
level of safety but do not comply with
any of the listed editions of the docu-
ments incorporated by reference.
Since the safety standards were adopt-
ed, the use of items in stock or the
reuse of salvagable items has been per-
mitted only if the items meet the re-
quirements of Part 192 or Part 195. as
the case may be, and where applicable.
the requirements of a listed edition of
a referenced document. Although this
requirement may be too Inflexible in
certain situations, it was not proposed
to be changed in Notice 77-7 and thus
cannot be changed in the final rules.
However, the problem will be given
further attention in the future when
action is taken on a petition for rule-
making filed by the Interstate Natural
Gas Association of America. This peti-
tion proposes that criteria be estab-
lished for the use of materials that do
not conform with any listed edition of
a listed document.
REPORT OP THE TECHNICAL PIPELINE
SAFETY STANDARDS COMMITTEE
Section 4(b) of the Natural Gas
Pipeline Safety Act of 1968 requires
that all proposed standards and
amendments to such standards per-
taining to gas pipelines be submitted
to the Committee and that the Com-
mittee be afforded a reasonable oppor-
tunity to prepare a report on the tech-
nical feasibility, reasonableness, and
practicability of each proposal. This
amendment to Part 192 was submitted
as Item A in a list of items before the
Committee at a meeting in Washing-
ton. D.C.. on January 17 and 18, 1978.
On March 10. 1978, the Committee
filed the following favorable report. A
minority report was not filed.
This communication is the official report
of the Technical Pipeline Safety Standards
Committee concerning the Committee's
action on one amendment to 49 CFR Part
192 proposed by the Office of Pipeline
Safety Operations, on revisions of Subpart I
of Part 192, as proposed by Committee
members, and on other matters the Com-
mittee decided should be brought to the at-
tention of the Department of Transporta-
tion.
The following described actions were
taken by the Committee at a meeting held
In Washington. D.C.. on January 17 and 18.
1978.
A. PROPOSAL BY OFSO TO AMEND THE UQVXXZ-
MZHTS Of RESPECT TO QUALIFICATION AND
DESIGN OP STEEL PIPE
OPSO proposed to amend the require-
menu of Appendix A and Appendix B of
Part 192 pertaining to qualification and
design of steel pipe as published In Notice
77-7, Docket No. 77-10. By a unanimous af-
firmative vote of the nine member* present,
the Committee found the following lan-
guage to be technically feasible, reasonable
and practicable.
[The language suggested is adopted in the
final rules.]
NOTE.—It has been determined that this
document does not contain a major regula-
tion requiring preparation of a Regulatory
Analysis under DOT procedures or Execu-
tive Order 12044.
In consideration of the foregoing.
Parts 192 and 195 of Title 49 of the
Code of Federal Regulations are
amended as set forth below.
1. Section n of Appendix A to Part
192 is amended to read as follows:
APPENDIX A—INCORPORATED BT REFERENCE
IL Documents Incorporated by reference.
Numbers in parentheses indicate applicable
editions. Only the latest listed edition ap-
plies except that an earlier listed edition
may be followed with respect to pipe or
components which are manufactured, de-
signed, or installed in accordance with the
earlier edition before the latest edition is
adopted, unless otherwise provided in this
part.
A.***
(5) API Standard 5LS "API Specification
for Spiral-Weld Line Pipe" (1967, 1970. 1971
plus Supp. 1. 1973 plus Supp. 1. 1975 plus
Supp. 1. and 1977).
(6) API Standard 5LX "API Specification
for High-Test Line Pipe" (1976, 1970, 1971
plus Supp. 1. 1973 plus Supp. 1, 1975 plus
Supp. 1. and 1977).
2. Section I of Appendix B to Part
192 is amended to read as follows
APPENDIX B—QUALIFICATION or PIPE
I. Listed Pipe Specifications. Numbers in
parentheses Indicate applicable editions.
Only the latest listed edition applies except
that an earlier listed edition may be fol-
lowed with respect to pipe or components
which are manufactured, designed, or in-
stalled In accordance with the earlier-edi-
tion before the latest edition Is adopted.
unless otherwise provided in this part.
351
-------
API 5LS. Steel pipe (1967, 1970. 1971 plus
Supp. 1. 1973 plus Supp. 1, 1975 plus Supp.
1. and 1977).
API SLX. Steel pipe (1967. 1970. 1971 plus
Supp. 1. 1973 plus Supp. 1. 1975 plus Supp.
1. and 1977).
(See. 3. Pub. L. 90-481. 82 Stat. 721. (49
UJS.C. 1872): for offshore gathering lines.
see. 105. Pub. L. 93-633. 88 Stat. 2157. (49
U.S.C. 1804); 49 CFR 1.53 and App. A to
PartU
3. In §195.3. paragraphs (a) and
(c)U) (Iv) and (v) are revised to read as
follows:
5 195.3 Matter incorporated by reference.
(a) There are incorporated by refer-
ence in this part all materials referred
to In this part that are not set forth in
full in this part. These materials are
hereby made a part of this regulation.
Applicable editions are listed in para-
graph (O of this section in parenthe-
ses following the title of the refer-
enced material. Only the latest listed
edition applies, except that an earlier
listed edition may be followed with re-
spect to components which are manu-
factured, designed, or installed in ac-
cordance with the earlier edition
before the latest edition is adopted,
unless otherwise provided in this part.
• * *
(I)"*
(iv) API Specification 5LS "API
Specification for Spiral-Weld Line
Pipe" (1969.1975, and 1977).
(v) API Specification 5LX "API
Specification for High-Test Line Pipe"
(1969,1975. and 1977).
(Sees. «. Pub. L. 89-670. 80 Stat. 937. (49
U.S.C. 1655): <18 D.S.C. 831-835V. 49 CFR
1.53 and App. A to Part 1.)
Issued In Washington. D.C., on April
25. 1978.
L. D. SANTMAW,
Acting Director,
Materials Transportation
Bureau.
352
-------
available for inspection in the Materials
Transportation Bureau, Washington,
D'.C. In addition, materials incorporated
by reference are available as follows:
(1) American Petroleum Institute
( API). 1801 K Street, N.W., Washington.
D.C. 20006, or 300 Corrigan Tower Build-
ing. Dallas, Texas 75201.
(2) The American Society of Mechani-
cal Engineers (ASME). United Engineer-
ing Center, 345 Bast 47th Street. New
Vork. N.Y. 10017.
(3) Manufacturers Standardization
Society of the Valve and Fittings In-
dustry (MSS), 1815 North Fort Myer
Drive. Arlington, Va. 22209.
(4) American National Standards In-
stitute (ANSI). 1430 Broadway. New
York. N.Y. 10018. (Formerly the United
States of America Standards Institute
(USASI). All current standards issued
by USASI and ASA have been redesig-
nated as American National Standards
and continue In effect.)
(5) American Society for Testing and
Materials (ASTM). 1916 Race Street
Philadelphia. Pa. 19103.
(c) The full title for the publications
incorporated by reference in this part
are as follows:
<1) American Petroleum Institute:
API Standard 6D "API Specifica-
tion for Pipeline Valves," which may be
obtained from the Dallas office (1968.
1974).
(lit API Standard 1104 "Standard for
Welding Pipe Lines and Related Facil<-
Ues" (1968. 1973).
(ill) API Specification 5L "API Spec-
ification for Line Pipe" (1969,1975).
API Specification 5LS "API Spec-
ification for Spiral-Weld Line Pipe"
(1969. 1975).
(v) API Specification 5LX "API Spec-
ification for High-Test Line Pipe" (1969,
1975).
<2) ASME Code is the American So-
ciety of Mechanical Engineers BoUer
and Pressure Vessel Code. Section Vm.
"Pressure Vessels. Division 1" (1968.
1974>.
< 31 Manufacturers Standardization
Society of the Valve and Fitting Indus-
try:
(i) MSS Standard practice SP-48
"Steel Butt-Welding Fittings <26 inch
and larger)" 11969).
Ui> MSS Standard Practice SP-63
"High Strength Wrought Welding Fit-
ting" H969).
"Specification for High-Test Wrought
Welding Fittings" (1973).
(4> American National Standards In-
stitute:
(i) ANSI B16.9 "Factory Made
Wrought Steel Butt-Welding Fittings"
(1964. 1971).
(ii) ANSI B31.4 "Liquid Petroleum
Transportation Piping Systems" (1966.
1974).
(5) American Society for Testing and
Materials:
i i) ASTM Specification A53 "Standard
Specification for Welded and Seamless
Steel Pipe" (1968.1973).
(ii> ASTM Specification A106 "Stand-
ard Specification for Seamless Carbon
Steel Pipe for High-Temperature Serv-
ice" (1968, 1972a).
(iiii ASTM Specification A134 "Stand-
ard Specification for Electric-Fusion
i Arc)-Welded Steel Plate Pipe. Sizes 16
in. and Over" (1968. 1973).
(iv) ASTM Specification A135 "Stand-
ard Specification for Electric-Resist-
ance-Welded Steel Pipe" (1968. 1973a).
(v) ASTM Specification A139 "Stand-
ard Specification for Electric-Fusion
(Arc) -Welded Steel Pipe, (Sizes 4 in. and
Over)" (1968, 1973).
(vi) ASTM Specification A155 "Stand-
ard Specification for Electric-Fusion-
Welded Steel Pipe for High-Pressure
Service" (1968, 1972a).
(vii) ASTM Specification A211 "Stand-
ard Specification for Spiral-Welded Steel
or Iron Pipe" (1968. 1973).
(vui) ASTM Specification A333
"Standard Specification for s*^m'"f«
and Welded Steel Pipe for Low-Tem-
perature Service" (1968. 1973).
(ix) ASTM Specification A381 "Stand-
ard Specification for Metal-Arc-Welded
Steel Pipe for High-Pressure Transmis-
sion Systems" (1969, 1973).
140 FR 43901, 49 CFR 1.53) |34 PR 15473.
Oct. 4. 1969. as amended by Amdt. 195-2.
35 FR 17184. Nov. 7. 1970: Amdt. 195-9. 41 FR
13592. Mar. 31, 19761
§ I95.-1 Acceptable petroleum comtnod
ilir*.
No carrier may transport any petro-
leum or petroleum product unless the pe-
troleum or petroleum product is chemi-
cally compatible with both the pipeline.
including all components, and any other
commodity that it may come into contact
with while in the pipeline.
353
-------
pipeline in Interstate and foreign com-
merce of Hazardous materials that are
subject to Parts 172 and 173 of this chap-
ter, petroleum, and petroleum products.
This sort does not apply to—
n*PMirfflvynt
equal to 42 U.S. standard gallons.
"Carrier" means a pipeline carrier
subject to sections 831-435 of title IS.
United States Code.
"Commodity" means a hazardous ma-
terial that is subject to Parts 172 and
173 of this chapter, petroleum, and
petroleum products.
"Component" means any part of a
pipeline which may be subjected to pump
pressure including, but not limited to.
pipe, valves, elbows, tees, flanges, and
closures.
"Line section" means a continuous run
of pipe between adjacent pressure pump
stations, between a pressure pump sta-
tion and terminal or working tankage.
between a pressure pump station and a
block valve, or between adjacent block
valves.
"Nominal wall thldmMi1* means the
wall thickness listed In the p*pe
specifications.
"Offshore" means beyond the line of
ordinary low water along that portion
of the coast of the United States that
is in direct contact with the open seas
and beyond the line marking the sea-
ward limit of inland waters.
"Pipe" or "line pipe" means a tube.
usually cylindrical, through which a
commodity flows from one point to
another.
"Pipeline system" or "pipeline" means
all parts of a carrier's physical facilities
through which commodities move In
transportation that Is subject to this
part, including, but not limited to. line
pipe, valves and other appurtenances
connected to Une pipe, pumping units.
fabricated assemblies associated with
pumping units, metering and delivery
stations and fabricated assemblies there-
in, and carrier-controlled breakout
tankage.
"Secretary" means the Secretary of
Transportation or any person to whom
he has delegated authority in the matter
concerned.
"Specified minimum yield strength"
means the muViuim yield strength, ex-
pressed in pounds per square inch, pre-
scribed by the specification under which
the material is purchased from the
manufacturer.
"Stress level" means the level of
tangential or hoop stress, usually ex-
pressed as a percentage of specified
minimum yield strength.
"Surze preuure" means pressure pro-
duced by a change in velocity of the
moving stream that results from shutting
down a pump station or pumping unit.
closure of a valve, or any other blockage
of the moving stream.
{34 FR 15*73. Oct. 4. 1989, u amended by
Amdt. 195-5. 38 FR 2977. Jan. 31. 1973)
§ 195.3 Matter incorporated by refer*
ence.1
'a) There are incorporated by refer-
ence in this part all materials referred
to in this part that are not set forth
in full in this part. These materials are '
hereby made a part of this regulation.
Applicable editions are listed in para-
graph ic' of this section in parentheses
fallowing the title of .the referenced ma-
terial. Only the latest listed edition ap-
plies, except that an earlier listed edition
may be followed with respect to compo-
nents which were manufactured, de-
signed, or installed before July 1. 1975.
unless otherwise provided in this part.
• MOTT: Incorporation by reference provi-
sions approved by the Director at *he Fed-
eral Register. March 28. 1978.
354
-------
(b) AU incorporated materials are
available for inspection in the Materials
Transportation Bureau. Washington.
D.C. In addition, materials incorporated
by reference are available as follows:
a) American Petroleum Institute
(API). 1801 K Street. N.W.. Washington.
D.C. 20006. or 300 Corrigan Tower Build-
ing. Dallas. Texas 75201.
12) The American Society of Mechani-
cal Engineers (ASME). United Engineer-
ing Center. 345 East 47th Street. New
Vorle. N.Y. 10017.
(3) Manufacturers Standardization
Society of the Valve and Fittings In-
dustry (MSS). 1815 North Fort Myer
Drive. Arlington. Va. 22209.
(4) American National Standards In-
stitute (ANSI). 1430 Broadway. New
Vorfc. N.Y. 10018. (Formerly the United
States of America Standards Institute
lUSASI). All current standards issued
by USASI and ASA have been redesig-
nated as American National Standards
and continue in effect.)
(5) American Society for Testing and
Materials (ASTM), 1916 Race Street
Philadelphia, Pa. 19103.
American Petroleum Institute:
(i) API Standard 6O "API Specifica-
tion for Pipeline Valves." which 'may be
obtained from the Dallas office (1968.
1974).
Hi) API Standard 1104 "Standard for
Welding Pipe Lines and Related Facili-
ties" '1968, 1973).
'iii> API Specification 5L "API Spec-
ification for Line Pipe" < 1969. 1975).
i iv> API Specification SLS "API Spec-
ification for Spiral-Weld Line Pipe"
11969. 1975).
(v> API Specification 5LX "API Spec-
ification for High-Test Line Pipe" (1969.
1975).
12) ASMS Code is the American So-
ciety of Mechanical Engineers Boiler
and Pressure Vessel Code. Section VIII.
"Pressure Vessels. Division 1" ' 1968.
1974).
' 3 > Manufacturers Standardization
Society of the Valve and Fitting Indus-
* rv *
U> MSS Standard practice SP-48
"Steel Butt-Welding Fittings '26 inch
and larger) " < 1969).
MSS Standard Practice SP-63
"High Strength Wrought Welding Fit-
ting" i1969).
liii) MSS Standard Practice SP-75
"Specification for High-Test Wrought
Welding Fittings" (1973).
(4) American National Standards In-
stitute:
a) ANSI B16.9 "Factory Made
Wrought Steel Butt-Welding Fittings "
(1964. 1971).
(ii) ANSI B31.4 "Liquid Petroleum
Transportation Piping Systems" (1966,
1974).
(5) American Society for Testing and
Materials:
U) ASTM Specification A53 "Standard
Specification for Welded and Seamless
Steel Pipe" '1968. 1973).
(ii) ASTM Specification A106 "Stand-
ard Specification for Seamless Carbon
Steel Pipe for High-Temperature Serv-
ice" (1968. 1972a).
(iiii ASTM Specification A134 "Stand-
ard Specification for Electric-Fusion
(Arc)-Welded Steel Plate Pipe. Sizes 16
in. and Over" (1968, 1973).
(iv> ASTM Specification A135 "Stand-
ard Specification for Electric-Resist-
ance-Welded Steel Pipe" '1968. 1973a).
(v> ASTM Specification A139 "Stand-
ard Specification for Electric-Fusion
'Arc) -Welded Steel Pipe. (Sizes 4 in. and
Over)" (1968. 1973).
i vii ASTM Specification A155 "Stand-
ard Specification for Electric-Fusion-
Welded Steel Pipe for High-Pressure
Service" (1968. I972a>.
(vii) ASTM Specification A211 "Stand-
' ard Specification for Spiral-Welded Steel
or Iron Pipe" '1968. 1973).
(viiti ASTM Specification A333
"Standard Specification for Seamless
and Welded Steel Pipe for Low-Tem-
perature Service" '1968. 1973).
<«) ASTM Specification A381 "Stand-
ard Specification for Metal-Arc-Weided
Steel Pipe for High-Pressure Transmis-
sion Systems" <1969. 1973).
140 FR 43901. 49 CFR 1.S3) 134 TO 15473.
Oct. 4. 1989. u "amended by Amdt. 195-2.
35 FR 17184. Nov. 7. 1970: Amdt. 195-9. 41 PR
13592. Mar. 31. 1976!
$ 195.1 Acceptable petroleum romrnod.
iliro.
No carrier may transport any petro-
leum or petroleum product unless the oe-
troleum or petroleum product is chemi-
cally compatible with both the pipeline.
including all components, and any other
commodity that it may come into contact
with while in the pipeline.
355
-------
(13) Part G. Item I. State the com-
monly used name of the commodity, sucb
as fuel oil. regular gasoline, liquefied
petroleum gas. It the commodity name
Is one not commonly used, state the name
here and give a brief description of it
under "Account of Accident by Respon-
sible Official of Carrier."
(14) Part G. Item 3. State the year
facility was installed or the best estimate
possible. Pipe is excluded as the year of
installation Is required In Item 4 of
PartH.
(19) Part H. Mark appropriate boxes
and state Information required In all
items of ***« part only if the accident oc-
curred In line pipe. If the accident oc-
curred In any other part of the pipeline
system, omit this part.
(16) Part 1. Mark appropriate boxes
and state Information required In all
Items of this part if the accident was
caused by corrosion In any component of
the pipllne system. In Item 4. state the
length of time between the type of tests.
such as pipe-to-soil potential, stated In
Item 5.
117) Part J. Complete all three items
only If the accident was caused by equip-
ment rupturing the pipeline. In Item 2.
all the information stated on the closest
line marker must be shown.
(b) In addition to the requirements of
paragraph (a) of this section. In the
space provided after Part J. the carrier
shall enter an account of the accident
containing the most reliable Information
to which the carrier has access at the
time of reporting, sufficiently detailed
and complete to convey an understand-
ing of the accident. This account may be
continued on an extra sheet of paper If
more space is needed.
(c) At the bottom of the back of DOT
Form 7000-1. the carrier shall state the
name and title of the pipeline official
responsible for compiling and flllng the
report along with the telephone number
at which this official can be reached, and
the date the report was completed.
5 195.38 dinner* in nr n«lililion» to or.
ciJcnl report.
Whenever a carrier receives any
changes In the information reported or
additions to the original report'on DOT
Form 7000-1 tt shall immediately flle a
supplemental report with the Director.
Office of Pipeline Safety. Department of
Transportation. Washington. D.C. 20590.
[34 FH 15473. Oct. 4. 1969. as amended by
Amdt. 195-5. 38 7R 2978. Jan. 31. 19T31
§ 195.60 Carrier uuulance in investiga-
tion.
If the Department of Transportation
Investigates an accident, the carrier In-
volved shall make available to the rep-
resentative of the Department all rec-
ords and Information that In any way
pertain to the accident, and shall afford
all reasonable assistance in the investi-
gation of the accident.
$ 19S.62 Supplies of accident report
DOT Form 7000-1.
Sach carrier shall maintain an ade-
quate supply of forms that are a facsimile
of DOT Form 7000-1 to enable It to
promptly report accidents. The Depart-
ment will. upon request, furnish speci-
men copie* of the form. Requests should
be addressed to the Director. Office of
Pipeline Safety. Department of Trans-
portation. Washington. D.C. 20590.
[34 PR 15473. Oct. 4. 1969. as amended by
Amdt. 195-5. 38 PR 2978. Jan. 31. 1973)
Subpart C— Design Requirements
i 195.100 Scope.
This subpart prescribes minimum de-
sign requirements (or new pipeline sys-
tems constructed with steel pipe and for
relocating, replacing, or otherwise chang-
ing existing systems constructed with
steel pipe. However, it does not apply to
the movement of line pipe covered by
i 195.424.
§ 195.102 Design temperature.
Material for components of the system
must be chosen for the temperature en-
vironment In which the components will
be used so that the pipeline will maintain
Its structural integrity.
§195.104 Variations in pressure.
If. within a pipeline system, two or
more components are to be conaected at
a place where one will operate at a higher
pressure than another, the system must
be designed so that any component oper-
ating at the lower pressure will not be
overstressed.
§ 195.106 Internal «lesi<:n pressure.
(a) Internal desi?n pressure Tor the
pipe In a pipeline is determined In ac-
cordance with the following formula:
Ps Internal design pressure la poundj per
square Inch gauge.
356
-------
§ 195.O Transportation of certain com-
modities.
(a) Except for petroleum, petroleum
products, natural gasoline, and liquefied
petroleum gases, no carrier may trans-
sort any commodity n«i*«f the carrier
notifies the Secretary in writing, with
the information listed in paragraph
>b) of this section, at lean 90 days be-
fore the date the transportation is to
begin. If the Secretary determines
that the transportaion of the commod-
ity in the manner proposed would be
unduly hazardous, he will, within 90
days after receipt of the notice, order
the carrier. In writing, not to transport
ihe commodity in the proposed manner
until further notice. As soon as practi-
cable after Issuance of such an order.
the Secretary will initiate appropri-
ate action to determine whether and
tn a hat manner the commodity may be
transported without undue hazard.
The notice submitted to the Ad-
ministrator by the carrier must state the
chemical name, common name, hazard
classification determined in accordance
with Pan 173 of this chapter, properties.
and characteristics of the commodity to
be transported. It must also Include
design specifications, including mate-
rials used in construction of the pipeline
and the maximum operating pressures
for the pipeline through which the com-
modity is to be transported.
[34 FR 15473. Oct. 4. 1969. a* amended by
Amdt. 196-1. 35 FR 5333. Mar. 31, 1970: Aradt.
195-3.38 FR 2978. Jan. 31.1973 ]
3 105.8 Trn importation of ronnnodiliei
in pipeline! constructed with other
than steel pipe.
No carrier may transport any com-
modity through a pipe that Is con-
structed after October 1. 1970. of mate-
rial other than steel unless the carrier
has notified Che Secretary in writing
at least 90 days before the transporta-
tion is to begin. The notice must state the
chemical name, common name, hazard
classification (if any) determined in ac-
cordance with Pan 173 of this chapter.
properties, and characteristics of the
commodity to be transported and the
material used In construction of the
pipeline. If the Secretary determines that
the transportation of the commodity in
the manner proposed would be unduly
hazardous, he will, within 90 days after
receipt of the notice order the carrier, in
writing, not to transport the commodity
in the proposed manner until further
notice.
[Amdt. 195-1. 3S F.R. 5333. Mar. 31. 1970.
as amended by Amdt. 195-2, 36 F.R. 17184.
Nov. 7. 1970: Amdt. 195-5. 38 FR 2978. Jan. 31.
19731
§ 195.10 Rc»pon»ibililY of carrier for
compliance with ihu Part.
A carrier may make arrangements with
another person for the performance of
any action required by this part. How-
ever, the earner is not thereby relieved
from the responsibility for compliance
with any requirement of this part.
Subpart B—Accident Reporting
5 195.50 Scope.
This suupart prescribes roles govern-
ing the reporting of any failure in a pipe.
line system subject to this pan In which
there is a release of the commodity trans-
ported resulting tn any of the following:
(a) Explosion or fire not intentionally
set by the carrier.
(b) Loss of 50 or more barrels of
liquid.
(c) Escape to the atmosphere of more
khan five barrels a day of liquefied petro-
leum gas or other liquefied gas.
(d) Death of any person.
i e) Bodily harm to any person result-
ing tn one or more of the following:
(l> Loss of consciousness.
(2) Necessity to carry the person from
the scene.
<3> Necessity for medical treatment.
(4) Disability which prevents the dis-
charge of normal duties or the pursuit
of normal activities beyond the day of
the accident.
a) At the earliest practicable mo-
ment following discovery of a release of
the commodity transported resulting in
an event described in 3 195.30. each car-
rier shall give notice, in accordance with
paragraph 'b> of this section, of any
failure that—
11) Caused a death or a personal in-
jury requiring hospitallzation:
(2) Resulted in either a Ore or explo-
sion not intentionally set by the carrier:
'3) Caused estimated damage to the
property of the carrier or others, or both.
of a total of SS.OOO or more;
357
-------
strength in pound* p*r square
inch determined in accordance with
paragraph (b) of this station.
tsNomlnal wall tblcsness of to* pip* in
Inch**. If tali Is unknown. It Is deter-
mined in iHW^ft^nc^ with paragraph
Tne yield strength to be used in
determining internal design pressure us*
der paragraph (a) of this section is the
specified "'*Tl*'T'"rn yield strength. If the
specified minimum yield strength is not
known, the yield strength ia determined
by performing all of the tensile test* at
either API Standard 5L. SLS. or 5LX on
randomly selected test specimens with
the following number of tests:
Pip* tut ATwnocr of tot*
U**i than 8 laches in One test lor each
outside diameter. 200 lengths.
4 laches through 12% On* test for esar>
inches in outside 100 lengths
diameter.
Larger Chan 13% One test tor eaeb
inches in outside JO length*.
diameter.
U the average yield-tensile ratio exceeds
0.85. the yield strength of the pipe is
taken as 24,000 PJX If the average yield-
tensile ratio is 0.83 or less, the yield
strength of the pipe Is taken as the lower
of the following:
(1) Eighty percent of the average
yield strength determined by the tensile
tests.
(2) The lowest yield strength deter-
mined by the tensile tests.
(c) If the nominal wall thickness to
be used in determining internal design
pressure under paragraph (a) of this
section Is not known, it Is determined by
measuring the thickness of each piece of
pipe at quarter points on one end. How-
ever. If the pipe is of uniform grade, size.
and thickness, only 10 individual lengths
or S percent of all lengths, whichever is
greater, need be measured. The thickness
of the lengths that an not measured
i2> class designation or the maxi-
mum working pressure to which the
valve may be subjected.
<3> Body material designation (the
end connection material. if. more than
one type is used).
14> Nominal valve size.
(40 PR 43901. 49 CFR 1.33) (34 PR 15473.
Oct. 4. 19T9. as amended ay Anult. 199-2. 35
PR 17186. ITov. 7. 1970; Amdt. 195-3, 41 PR
13592. Mar. 31. 19781
§ 195.118 Fittings.
There may not be any buckles.
dents, cracks, gouges, or other defects In
> he fitting that might reduce thr
strength of the fitting.
(o The fitting must be suitable for
ihe Intended service and be at least as
strong as the pipe and other fittings In
the pipeline system to which It la
Attached.
140 FR 43901. 49 CPU 1.33) [34 PR 15473.
Oct. 4. 1989. as amended by Amdt. 195-9
41 PR 13592. Mar. 31. 1978)
S 105.120 Clumge* in ilirmiim: Provi-
sion for intenwl ptuwugr.
Each component of a main line sys-
tem, other than station and terminal
manifolds, that change direction within
the pipeline system must have a radius
of turn that readily allows the passage
of pipeline scrapers, spheres, and in-
ternal Inspection equipment
§ 195.122 Fabricated Itranrlt connw-
tions.
Each pipeline system must be designed
so that the addition of any fabricated
branch connections will not reduce the
strength of the pipeline system.
§ 195.124 Closures.
Each closure to be installed in a pipe-
line system must comply with the ASMS
Boiler and Pressure Vessel Code. Section
VHI, Pressure Vessels, Division l. and
must have pressure and temperature rat-
ings at least equal to those of the pipe
to which the closure is attached.
I4f '"R 43901. 49 CFH 1.53) [Amdt. 195-9.
..-•,« .».. ,, ,0-Tfil
359
-------
(4> Resulted In pollution of any
stream, river, lake, reservoir, or other
similar body of water that violated ap-
plicable water quality standards, caused
a discoloration of the surface of the
water or adjoining shoreline, or deposited
a sludge or emulsion beneath the surface
of the water- or upon adjoining shore-
lines : or .
< 5) In the judgment of the carrier.
was significant even though tt did not
meet the criteria of any other subpara-
sraph of this paragraph.
>t» Reports made under paragraph
'a> of this section are made by telephone
to area code 202, 426-0700 and must
include the following information:
1 1) Name and address of the carrier.
(2) Name and telephone number of
the reporter.
'3) The location of the failure.
(4) The time of the failure.
'5) The fatalities and personal in-
juries. If any.
(6) All other significant facts known
by the carrier that are relevant to the
cause of the failure or extent of the
damages.
(38 PR 7121. Mar. 18. 19731
S 1 1.".." I Accident reporting.
Each carrier that experiences an acci-
dent that is required to be reported
under this subpart shall, as soon as prac-
ticable but not later than 15 days after
discovery of the accident, prepare and
flle an accident report, on DOT Form
7000-1 or a facsimile, with the Director.
Office of Pipeline Safety, Department of
Transportation. Washington. D.C. 20590.
The carrier shall flle two copies of each
report and shall retain one copy at its
principal place of business.
[34 PR 13473. Oct. 4. 1909. u amended by
Amdt. 195-3. 38 PR 2978. Jan. 31. 1973 1
tnntntctinnn foe prvpncing DOT
Form 7000-1.
Each carrier shall prepare each
report of an accident on DOT form
Tnoo-l or a facsimile, in accordance with
the following instructions:
(1) General. Each applicable item
must be marked or filled in as fully and
as accurately as information accessible to
the carrier at the time of filing the report
will permit.
(2) Part A. Enter name as it Is filed
with the Interstate Commerce Commis-
sion. If the carrier's name Is not filed
with the Commission, enter the complete
corporate name of the carrier. Enter the
address of the carrier's principal place
of business Including zip code.
(3) Part S. Item l. Eater the data the
accident occurred or was discovered. If
the accident was not discovered on the
date It occurred, state this fact on the
back of the form.
<4) Part B, Item 2. Enter the exact
time la hours and minutes (La, 10:19) if
known or a time range 'i.e.. 10:11) if
exact time Is not known. If the accident
was not discovered on the date it oc-
curred, enter the time It was discovered
and state this fact, on the back of the
form as in Pan B. Item 1.
(5) Port B. Item 3. Eater all three
names. State, county, city, or town, la or
near which accident occurred.
(8) Part B. Item 4. Mark the appro-
priate box. If "other" is marked, state
clearly oa form what part of the pipe-
line system.
(7) Part B, Item S. If the accident oc-
curred in an uninhabited area, such as
woods, cultivated field, swamp, etc.. so
itate clearly oa the form under Item S.
If not. attach a sketch to the form show-
lag the part of the pipeline system where
the accident occurred, and the location
of the accident as related to significant
landmarks. Each Item shown on the
sketch must be clearly and distinctly
marked to identify It. Approximate dis-
tances from accident location to all land-
marks shown on the sketch must be
indicated.
(8) Part c. Mark the appropriate box.
or boxes. If applicable, mark more than
one box. If "other" is marked, state
clearly on form the exact origin of the
release of commodity.
(9) Part D. Mark the appropriate box.
If -other" Is marked, clearly state the
cause of the accident.
(10) Part S. Indicate a number under
each heading Including "0" If none. Re-
port deaths, even if previously reported
in accordance with } 195.52.
(11) Part r. Items 1 and. 2. Report
only material In the pipeline system that
was actually damaged such as pipe.
valves, or fittings. Do not Include cost of
commodity which was lost due to the
accident or fittings used during repair
which became permanently attached to
the system. The dollar value of damage
should be based on replacement at pres-
ent day costs.
<12) Part F, Items 3 and 4. This Is
damage to property other than that of
the carrier. Dollar value must be actual
or the best estimate available.
358
-------
41 PR 34040. Aug. 12. 1978. effective Aug. 1.
1977. For tne convenience of the user, the
superseded text is set out below:
r=A design (Mtor of 0.73. except that a
design actor of 0.54 U uaed for pipe)
that oae beea cold worked to meet
the «peeifled mlninmm yield strength
and is subsequently betted, other
than br welding, to 400* P. or mere.
S 195.108 External preMure.
Any external pressure that will be ex-
erted on the pipe must be provided tor in
designing a pipeline system.
§ I9S.110 External lowi*.
i
a) Anticipated external loads (e.g.),
earthquakes, vibration, thermal expan-
sion, and contraction must be provided
(or in designing a pipeline system. In
providing for expansion and flexibility,
section 419 of ANSI B31.4 must be fol-
lowed.
The pipe and other components
must be supported In such a way that
the support does not cause excess local-
ized stresses. In rf*««gm"g attachments
to pipe, the added stress to the wall or
the pipe must be computed and com-
pensated for.
:40 PR 43901. 49 CPB 1.S3) [34 PR 15473.
Oct. 4. 1909. ae amended by Amdt. 195-9.
41FH13S92. Mtr. 31. 1978 1
5 195.112 New pip*.
Any new pipe Installed in a pipeline
system must comply with the following:
(a) The pipe must be made of steel of
the carbon, low alloy-high strength, or
alloy type that Is able to withstand the
internal pressures and external loads and
pressures anticipated for the pipeline
system
' b> The pipe must be made In accord-
ance with a written pipe specification
that sets forth the chemical requirements
for the pipe steel and mechanical tests
for the pipe to provide pipe suitable for
the use intended.
Each length of pipe with an outside
diameter of 4 Inches or more must be
marked on the pipe or pipe coating with
the specification to which it was made.
the specified minimum yield strength or
grade, and the pipe size. The marking
must be applied In a manner that dees
not damage the pipe or pipe coating an and the following:
(a> The pipe must be of a known spec-
ification and the seam Joint factor
must be. determined In accordance with
I !9S.l06 or There may not be any—
(l) Buckles:
< 2) Cracks, grooves. Rouges, dents, or
other surface defects that exceed the
maximum depth of such a defect per-
mitted by the specification to which the
pipe was manufactured: or
(3) Corroded areas where the remain-
Ing wall thickness Is less than the mini-
mum thickness required by the toler-
ances in the specification to which the
pipe was manufactured.
However, pipe that does not meet the
requirements of subparagraph (3) of
this paragraph may be used if the oper-
ating pressure Is reduced to be commen-
surate with the remaining wall thickness.
134 P.R. 15473. Oct. 4. 198*. a* amended by
Amdt. 195-3. 38 P.R. J71SS. NOT. 7. 19701
5 19S.11* Vulvr*.
Each valve installed In a pipeline sys-
tem must comply with the following:
(a) The valve must be of a sound
engineering design.
Materials subject to the Internal
pressure of the pipeline system. Including
welded and flanaed ends, must be com-
patible with the pipe or fittings to which
the valve Is attached
d> Each valve must be both hvdro-
statically shell tested and hydrostatically
seat tested without leakage to at least
the requirements set forth in section 5 of
API Standard 6D.
>e> Each valve other than a check
valve must be equipped with a means for
clearly indicating the position of the
valve 'open, closed, etc.).
>t i Each valve must be marked on the
body or the nameplate, with at least the
following:
' 1> Manufacturer's name or trade-
mark.
360
-------
<2> Class designation or the maxi-
mum working pressure to which the
valve may be subjected.
<3> Body material designation (the
end connection material, it more than
one type is used).
14 > Nominal valve size.
(40 PR 43901. 49 CFR 1.53) [34 PR 13473.
Oct. 4. 19T9. aa amended sy AmUt. 195-2. 35
PR 17185. Nov. 7. 1970: Aradt. 195-9. 41 PR
13592. Mar. 31. 1978)
§195.118 Fittings.
ia) Butt-welding type fittings must
meet the marking end preparation and
the bursting strength requirements of
ANSI B18.9 or MSS Standard Practice
SP-75, except that fittings manufac-
tured, designed, or installed before July
1. 1976; may meet the requirements of
MSS Standard Practice SP-48 or MSS
Standard Practice SP-83.
There may not be any buckles.
(tents, cracks, gouges, or other defects In
the fitting that might reduce the
strength of the fitting.
The fitting must be suitable for
ihe Intended service and be at least ai
strong as the pipe and other fittings In
the pipeline system to which It U
attached.
(40 PR 43901. 49 CPR 1.53) [34 PR 15473.
Oct. 4. 1909. a* amended by Amdt. 195-9
41 PR 13592. Mar. 31. 19781
§ I OS. 120 Quiit ~e» in ilirmioni Provi-
sion for internal pannugr.
Each component of a main line sys-
tem, other than station and terminal
manifolds, that change direction within
the pipeline system must have a radlu*
of turn that readily allows the passage
of pipeline scrapers, spheres, and in-
ternal Inspection equipment
§ 195.122 Faliricaleil liranrli connec-
tions.
Each pipeline system must be designed
so that the addition of any fabricated
branch connections will not reduce the
strength of the pipeline system.
§ 195.124 Closures.
Each closure to be installed in a pipe-
line system must comply with the ASME
Boiler and Pressure Vessel Code. Section
Vin. Pressure Vessels, Division 1. and
must have pressure and temperature rat-
ings at least equal to those of the pipe
to which the closure is attached.^
140 PR 43901. 49 CPR 1.53) (Amdt. 195-9.
41 FR 13592. Mar. 31. 19161
§ 195.126 Flange connection.
Each component of a flange connec-
tion must be compatible with each othsr
component and the connection as a
unit must be suitable for the service in
which it is to be used.
§ 195.128 Station piping.
Any pipe to be installed In a. station
that Is subject to system pressure must
meet the applicable requirements of this
subpart.
5195.130 Fabricated auemblies.
Each fabricated assembly to be In-
stalled in a pipeline system must meet the
applicable requirements of this subpart.
§195.132 Above ground tank*.
Each above ground tank must be de-
signed to withstand the internal pres-
sure produced by the commodity to be
stored therein and any anticipated ex-
ternal loads.
Subpart 0—Construction
§ 195.200 Scape.
This subpart prescribes minimum re-
quirements for constructing new pipe-
line systems with steel pipe, and for
relocating, replacing, or otherwise
changing existing pipeline systems that
are constructed with steel pipe. However.
this subpart does not apply to the move-
ment of pipe covered by 5 195.424.
§ 195.202 Compliance with «pccific«-
tiona or stannania.
Each pipeline system must be con-
structed in accordance with comprehen-
sive written specifications or standards
that are consistent, with the requirement;
of this part.
§ 195.204 Inspection—general.
Inspection must be provided to ensure
the installation of pipe or pipeline sys-
tems'in accordance with the require-
ments of this subpart. No person may
be used to perform inspections unless
that person has been trained and is
qualified in the phase of construction
he is to inspect.
§195.206 Material inspection.
No pipe or other component may be
installed in a pipeline system unless it
has been visually inspected at the sit?
of installation to ensure that it is not
damaged in a manner that could im-
pair its strength or reduce Its service-
ability.
361
-------
§ 195.208 Welding of support* and
braces.
Supports or braces may not be welded
directly to pipe that will be operated at
a pressure of more than 100 pjj.gr.
§ 195.210 Pipeline location.
(a) Pipeline right-of-way must be
selected to avoid, as far as practicable.
areas containing private dwellings. In-
dustrial buildings. and places of public
assembly.
(b> No pipeline may be located within
50 feet of any private dwelling, or any
industrial building or place of public
assembly in which persons work, congre-
gate, or assemble, unless it is provided
with at least 12 inches of cover In addi-
tion to that prescribed in 3 195.248.
§195.212 Bending of pipe.
(a) Pipe must not have a wrinkle bend.
(b) Each field bend must comply with
the following:
(1) A bend must not impair the serv-
iceability of the pipe.
(2) Each bend must have a smooth
contour and be free from buckling.
cracks, or any other mechanical damage.
(3) On pipe containing a longitudinal
weld, the longitudinal weld must be as
near as practicable to the neutral axis of
the bend.
(c) Each circumferential weld which
is located where the stress during bend-
ing causes a permanent deformation in
the pipe must be nondestructively tested
either before or after the bending
process.
(40 FR 43901. 49 CFR 1.53) (Amctt. 195-10.
41 FR 26018. June 24. 1976)
§ 195.214 Welding: General.
(a) Welding must be performed In
compliance with this section and
5f 195.218 through 195.234.
• b> Welding must be performed in
accordance with established written
welding procedures that have been tested
to assure that they will produce sound.
ductile welds that comply with require-
ments of this subpart. Detailed records
of these tests must be kept by the ear-
ner involved.
§195.216 Welding: miler joints.
A miter joinc is not permitted mot in-
cluding deflections up to 3 degrees that
are caused by misalignment).
(40 FR 43901. 49 CFR 1.53)
41 FR 26018. Junt 24. 19781
I Amdt. 195-10.
§ 195.218 Welding! Seam offset.
Seams on adjacent pipe lengths must
be offset.
§ 195.220 Weld.: Filler metal.
Filler metal must be at least equal In
strength to the highest specified mini-
muni yield strength of the pieces being
welded and must fuse the pieces together.
§195.222 Welders: Testing.
Each welder must be qualified in ac-
cordance with one of the following edi-
tions of Section 3 of API Standard 1104:
(a) The 1973 edition, except that a
welder may be qualified by radiography
under subsection 3.51 without regard for
the standards in subsection S.9 for depth
of undercutting adjacent to the root bead
unless that depth is visually determined
by use of a depth measuring device on
all undercutting along the entire circum-
ference of the weld: or
(b) If a welder is qualified before
March 20. 197S. the 1968 edition, ex-
cept that a welder may not requalify
under the 1968 edition.
[Amdt. 195-3. 40 FR 10183. Mar. 5. 197S. aa
amended by Amdt. 19S-3A. 40 FR 27233.
Juna27, 19751
§ 195.224 Welding: Weather.
Welding mutt be protected from
weather conditions that would impair
the quality of the completed weld.
§ 195.226 Welding: Are burn*.
(a) Each arc burn must be repaired.
(b) An arc burn may be repaired by
completely removing the notch by grind-
Ing, if the grinding does not reduce the
r»m..
-------
in subsection 6.9 for depth of undercut-
ting adjacent to the root bead apply
only if—
(l) That depth is visually determined
by use of a depth measuring device on all
undercutting along the entire circumfer-
ence of the weld: and
(2) Visual determination of internal
undercutting is made in all pipe of the
same diameter in a pipeline, except
where unpractical at tie-in welds.
(Amdt. 19S-8A. 40 FR 27323. Jun* 27. 1973)
§ 195.230 Weld*: Repair of defect*.
(a) Except as provided in paragraph
of this section, a weld that is found
unacceptable under § 199.228 may not be
repaired unless*-
H) There are no cracks in che weld:
(2) The segment of the weld to be re-
paired was not previously repaired: and
(3) The weld is inspected after repair
to assure its acceptability.
(t» In the case of offshore pipelines, a
weld on a pipeline being installed from
a pipeiay vessel may be repaired if the
repair is made in accordance with es-
tablished written welding procedures
that have been tested under 3195.214 to
assure that they will produce sound
ductile welds.
(40 Pit 43901. 49 C7B 1J3) jArndt. 195-11
41 PR 34040. Aug. 12. 19781
§ 195.232 Weld*: Removal of defects.
Except for offshore pipelines being in-
stalled from a pipeiay vessel, a cylinder
of the pipe containing the weld must be
removed and the ends rebeveled when-
ever—
(a) The weld contains one or mom
cracks:
(b) The weld is not acceptable under
3195.228 and is not repaired: or
The weld was repaired and the re-
pair did not meet the requirements of
§ 195.228.
I 40 PR 43901. 49 CFR 1.53) |Amdt. 193-11.
41 FR 34040. Aug. 12. 19761
§195.234 Welds s Nondestructive test-
ing and retention of testing record*.
(a) A weld may be nondestructlvely
tested by any process that will clearly
indicate any defects that may affect the
integrity of the weld.
(b) Any nondestructive testing of
welds must be performed—
(1) In accordance with a written set
of procedures for nondestructive testing;
and
(2) With personnel that have been
trained is the established procedures
and la the use of the equipment employed
In the testing.
During construction, at least 10
percent of the girth welds made by each
welder during each welding day must
be nondestructlvely tested over the en-
tire circumference of the weld.
(e) In the following locations, 100 per-
cent of the girth welds must be non-
destructlvely tested:
(1) At any onshore location where a
loss of commodity could reasonably be
expected to pollute any stream, river.
lake, reservoir, or other body of water.
and any offshore area unless impractica-
ble, in which case only 90 percent of each
day's welds need be tested.
<2> Within railroad or public road
rights-of-way.
(3) At overhead road crossings and
within tiirtntf'a
(4) At pipeline tie-ins.
(5) Within the limits of any incorpo-
rated subdivision of a State government.
(6) within populated areas, including
but not limited to. residential subdivi-
sions, shopping centers, schools, desig-
nated commercial areas. Industrial
facilities, public institutions, and places
of public assembly.
When installing used pipe. 100 per-
cent of the old girth welds must be non-
destructlvely tested.
<«> A record of the nondestructive
testing must be retained by the carrier
who Is Involved. Including (if radi-
ography is used) the developed film with.
to far as practicable, the location of the
weld. This record must be retained for
3 rears after the line Is placed In
operation.
(40 PR 43901. 49 CPR 1.33) |34 PR 13473,
Oct. 4. 1969. u amended by Amdt. 195-1.
33 FR 5335. Mar. 31. 1970; Amdt. 193-11. 41
FR 34040. Aug. 12. 1978|
§ 193.236 Kxtcrnul corrosion protec-
tion.
Each component in the pipeline sys-
tem must be provided with protection
against external corrosion.
§ 195.238 External coating.
(a> No pipeline system component may
be buried or submerged unless that com-
363
-------
ponent has an external protective coat-
ing that—
(1) Is designed to mitigate corrosion
of the buried or submerged component;
<2) Has sufficient adhesion to the
metal surface to prevent underfllm mi-
gration of moisture:
(3) it sufficiently ductile to resist
cracking;
(4) Has enough strength to resist
damage due to handling and soil stress:
and
(5) Supports any supplemental ca-
thodlc protection.
In addition. If an insulating-type coating
is used It must have low moisture ab-
sorption and provide high electrical
resistance.
(b> All pipe coating must be inspected
just prior to lowering the pipe into the
ditch or submerging the pipe, and any
damage discovered must be repaired.
(40 PR 43901. 49 CFR 1J3) [34 PR 15473.
Oct. 4. 1969, a* uatnded by Amdt. 190-11.
41 PR 34040. Aug. 12. 1978)
§ 195.242 Calhodic protection iriiem.
(a) A cathodic protection system must
be installed for all burled or submerged
facilities to mitigate corrosion that
might result in structural failure. A test
procedure must be developed to deter-
mine whether adequate cathodic protec-
tion has been achieved.
i b> A cathodic protection system must
be Installed not later than 1 Tear after
completing the construction.
(40 FR 43901. 49 CFR 1.S3) |34 FR 1S473.
Oct. 4. 1969. H amended by Amdt. 193-U.
41 FR 34040. Aug. 12. 19781
§ 195.244 Test Icada.
'a) Except for offshore pipelines.
electrical test leads used for corrosion
control or electrolysis testing must be
Installed at Intervals frequent enough
to obtain electrical measurements Indi-
cating the adequacy of the cathodic
orntectlon.
(b) Test leads must be installed as
follows:
i'l) Enough looping or slack must be
provided to prevent test leads from being
unduly stressed or broken during back-
filling.
< 2) Each lead must be attached to the
pipe so as to prevent stress concentra-
tion on the pipe.
<3> Each lead Installed la a conduit
must be suitably Insulated from the
conduit.
§ 195.246 Inslallalion of pipe in a
ditch.
(a) All pipe installed in a ditch must
be installed in a manner that minimizes
the introduction of secondary stresses
and the possibility of damage to the
Pipe.
amended by Amdt. 195-11.
41 PR 34040. Aug. 12. 1976]
Zmcnvz DAT*: In i 196.246. the existing
text wt* designated at paragraph Less cover than the minimum re-
quired by paragraph (a) of this section
and $ 195.210 may be used if—
(1) It Is Impracticable to comply with
the minimum cover requirements; and
364
-------
(2> Additional protection la provided
that la equivalent to the inHttmnm re.
quired cover.
(40 FR 43901. 49 CFR 1.53) 134 FR 1S4T3.
Oct. 4. 1969. as amended by Amdt. 196-11.
41 PR 34040. Aug. 12. 1978)
DAT* NOTE: la i 199.248. para-
graph lai was revised by Amdt. 195-11 at
41 PR 34040. Aug. 12. 1978. effective Aug. 1.
1977. For the convenience of the user the
superseded test a an out below.
i at Unless specifically exempted in this
subpart. all pip* must be burled so that it Is
below the level of cultivation. Except as pro-
vided in paragraph end railraedf
Amy atfter we*.
M
30
M
It
' Rack acavation is any •seavatlan thai raqutfe*
MeMlns er raooval hy eoulvalent meanii.
5 IQS.250 Clearanrr hrtvern iiipr »n
underground .•trurturr*.
Any pipe Installed underground must
nave at least 13 Indies of clearance be-
careen the outside of the pipe and the
extremity of any other underground
structure, except that Tor drainage tile
the minimum clearance may be less than
12 Inches but not less than 2 inches Row-
ever, where 12 inches of clearance is Im-
practicable, the clearance may be re-
duced If adequate provisions are made
for corrosion control.
S 195.252 rUekiillinc.
Backflning must be performed In a
manner that protects any pipe coating
and provides firm support for the pipe.
3115.254 Abow ground romponmtn.
'a) Any component may be Installed
above ground In the following situa-
tJons. If the other applicable require-
ments of this pan are complied with-
a> Overhead crossings of highway*
railroads, or a body of water.
(2) Spans over ditches and gullies.
(3) Scraper traps or block valves.
<4> Areas under the direct control ol
the carrier.
15) In any area Inaccessible to the
pubUc.
fb> Each component covered by this
section must be protected from the
forces exerted by the anticipated loads.
5 195.256 Crnming of railroad* .and
highways.
The pipe at each railroad or highway
crossing must be Installed so as to ade-
quately withstand the dynamic forces
exerted by anticipated traffic loads.
5 105.253 VnKf*: Crf-twiil.
i at Each valve must be installed in a
location that is accessible to authorized
employees and that is protected from
damage or tampering.
< b) Each submerged valve located off-
shore or in inland navigable waters must
be marked, or located by conventional
survey techniques, to facilitate quick lo-
cation when operation of the valve is
required.
(40 FR 43901. 49 CFR 1.531 [34 FR 13473.
Oct. 4. 1989. as amended by Amdt. 195-11.
41 FR 34041. Aug. 12. 1978)
Emcnvx OATS NOTE: In $ I93.2S8. the ex-
isting text was designated as paragraph ia»
and paragraph* •. b I was added by Amdt. 193-
11 at 41 FR 34041. Aug. 12. 1978. effective
Aug. 1. 1977.
S 193.260 Vnl'rv: Lorn lion.
A valve must be installed at each of
the following locations:
(a) On the suction end and the dis-
charge end of i pump station In a man-
ner that permits Isolation of the pump
station equipment In the event of an
emergency.
'b> On each line entering or leaving
a tank farm in a manner that permits
Isolation of the tank farm from other
facilities.
On each mainline at locations
along the pipeline system that will mini-
mize damage or pollution from acci-
dental liquid discharge, as appropriate
tor the terrain in open country, for off-
shore areas, or for populated areas.
On each lateral takeoff from a
crunk line In a manner that permits
shutting off the lateral without Inter-
rupting the flow In the trunk line.
On each side of a water crossing
'.hat is more than 100 feet wide from
365
-------
high-water mark to high-water mark
unless the Secretary finds in a particular
case thac valves are noc justified.
On each side or a reservoir hold-
Ing water Tor human consumption.
140 FB 43901. 49 CFR 1.23) [34 FR 15473.
Oct. 4. i960, as amended by Amdt. 195-3.
38 FR 2973. Jan. 31. 1973: Amdt. 195-11.
41 FB 34041. Aug. 12. 19761
EFFCCOTK DAT* MOTS: In S 195.280. para-
graph id was revised by Amdt. 195-11 at
41 PR 34041. Aug. 12. 1978. effective Aug. 1.
1977. For the convenience of che user, toe
superseded text is sec out Below:
. • • . •
The following must be provided la
each pump station:
f 1) Safety devices that prevent over-
pressuring of pumping equipment. In-
cluding the auxiliary pumping equip-
ment within the pumping station.
(2) A device for the emergency shut-
down of each pumping station.
• 3) If power is necessary to actuate
the safety devices, an auxiliary power
supply.
Each safety device must be tested
under conditions approximating actual
operations and found to function prop-
erly before the pumping station may b*
used.
fd) Except for offshore pipelines
pumping equipment may not be In-
stalled—
' 1 > On any property that will not be
under the control of the carrier: or
<2> Less than SO feet from the bound-
ary of the station.
'e> Adequate fire protection must b*
Installed at each pump station. If the
flre protection system installed require*
the use of pumps, motive power must be
provided for those pumps that is sep-
arate from UK power that operates thr
station.
S 195.264 Above ground lank*.
(a) A means must be provided for
containing liquids In the event of spillage
or tank failure.
Tankage areas must be adequately
protected against unauthorized entry
(c> Normal and emergency relief vent-
Ing must be provided for each tank.
§ 195.266 Construction records.
A complete record that shows the fol-
lowing must be maintained by the carrier
Involved for the life of each facility:
(a) The total number of girth welds
and the number nondestructively tested.
Including the number rejected and the
disposition of each rejected weld.
(b) The amount, location, and cover
of each size of pipe installed.
(c) The location of each crossing of
another pipeline.
(d) The location of each buried utility
crossing.
(e) The location of each overhead
crossing.
(f) The location of each valve.
weighted pipe, corrosion test station, or
other Item connected to the pipe.
Subpart E—Hydrostatic Testing
AUfHomrr: The provisions of this Subpart
E Issued under sees. 831-835. Title 18. United
States Code: tees. 8 («)(4), (f)(3)(A). De-
partment of Transportation Act (49 U.S.C.
1855 (e><4). (DOHA)): 5 l.-Md) (8). Regu-
lations of the Office of the Secretary of
Transportation.
3ouxot: The provisions of this Subpart K
are contained In Amdt. 195-3. 35 VA 17185.
Nov. 7, 1970. unless otherwise noted.
§ 195.300 Scope.
This subpart prescribes minimum re-
quirements for hydrostatic testing of
newly constructed steel pipeline systems
and for hydrostatic testing of existing
steel pipeline systems that are relocated.
replaced, or otherwise changed. However,
this subpart does not apply to the move-
ment of pipe covered by 3 195.424.
§ 195.302 General requirements.
'a) Each new pipeline system, each
pipeline system in which pipe has been
relocated or replaced, or that part of a
pipeline system that has been relocated
or replaced, must be hydrostatically
tested in accordance with this subpart
without leakage.
(b) The test pressure for each hydro-
static test conducted under this section
366
-------
must be maintained for at least 24 hours
throughout the part of the system that
is being tested.
§ 195.304 Testing of component*.
(*) Each hydrostatic test under
319S.302 must test all pipe and attached
fittings, including components, unless
otherwise permitted by paragraph A component that is the only item
being replaced or added to the pipeline
system need not be hydrostatically
tested under paragraph of this sec-
tion if the manufacturer certifies that
either—
(1) The component was hydrostati-
cally tested at the factory: or
(2) The component was manufactured
under a quality control system that en-
sures each component is at least equal
in strength to a prototype that was hy-
drostatically tested at the factory.
9 19&306 Test medium.
ta) Except as provided in paragraph
(b) of this section, water must be used
as the test medium.
(b) Except for offshore pipelines, liquid
petroleum that does not vaporize rapidly
may be used' as the test medium if—
(1) The entire pipeline section under
test is outside of cities and other popu-
lated areas:
(2) Each building within 300 feet of
the test section is unoccupied while the
test pressure is equal to or greater than a
pressure which produces a hoop stress of
50 percent of specified minimum yield
strength:
(3) The test section is kept under sur-
veillance by regular patrols during the
test: and
<4> Continuous communication is
maintained along entire test section..
(S«c. 8. Pub. L. 89-870: 40 PR 43901. 49 CFR
1.53) [Amdt. 195-2. Nov. 7. 1970. u
amended by Amdt. 199-11. 41 FR 43152.
Sept. 30. 1978)
§195.308 Testing of tie-ins.
Pipe associated with tie-ins must be
hydrostatically tested, either with the
section to be tied in or separately.
§ 195.310 Records.
(a) A record must be made of each
hydrostatic test and that record must
be retained as long as the facility tested
is in use.
(b) The record required by paragraph
fa) of this section must inciude the re-
cording gauge charts, dead weight tester
data, and the reasons for any failure
during a test. Where elevation differ-
ences in the section under test exceed
100 feet, a profile of the pipeline that
shows the elevation and test sites over
the entire length of the test section must
be included. Each recording gauge chart
must also contain—
. (1) The carrier's name, the name of
the person responsible for making the
test, and the name of the test company
used, if any:
(2) The date and time of the test:
(3) The minimum test pressure:
(4) The test medium:
<5> A description of the facility
tested: and
18) An explanation of any pressure
discontinuities that appear on any chart.
Subpart F—Operation and
Maintenance
§ 195.400 Scope.
This subpart prescribes minimum re-
quirements for operating and maintain-
ing pipeline systems constructed with
steel pipe.
§195.102 General requirement*.
(a) Each carrier shall establish and
maintain current written procedures:
(1) To ensi're the safe operation and
maintenance of Its pipeline system In
accordance with this Part during normal
operations.
(2> To be followed during abnormal
operations and emergencies.
(b) No carrier may operate or main-
tain its pipeline systems at a level of
safety lower than that required by this
subpart and the procedures It Is re-
quired to establish under paragraph (a)
of this section.
fc) Whenever a carrier discovers any
condition that could adversely a.Toct the
safe operation of Its pipeline system it
shall correct it within a reasonable time.
However, if the condition Is of such a
nature that It presents an Immediate
hazard to persons or property, the car-
rier may not operate the affected part
of the system until it has corrected the
unsafe condition.
No carrier may operate any part
of a pipeline system upon which con-
struction was begun after March 31.1970.
or in the case of offshore pipelines lo-
cated between a production facility and
a earner's trunkline reception point, af-
367
-------
ter July 31. 1977, unless it was designed
and constructed as required by this part.
(40 PR 43901. 49 CTR 1.53) [34 PR 15473.
Oct. 4, 1963. as amended by Amdt. 195-1 1.
41 PR 34041. Aug. 12. 1976)
§ 195.-I04 Maps and record*.
(a) Bach carrier shall m«ft«faHn cur-
rent maps and records of its pipeline
systems that include at least the follow-
ing information:
(l) Location and Identification of all
major facilities.
<2) All crossings of public roads, rail-
roads. rivers, buried utilities, and. foreign
pipelines.
(3) The maximum operating pres-
sure of each pipeline.
(4) The diameter, grade, type, and
nominal wall thickness of all pipe.
(b) Each carrier shall maintain dally
operating records that indicate the dis-
charge pressures at each pump station
and any unusual operations of a facility.
The carrier shall retain these records
for at least 3 years.
(o) Each carrier shall also maintain
for the useful life of that part of the
pipeline system to which they relate, rec-
ords that include the following:
(1) The date, location, and descrip-
tion of each repair made to its pipeline
systems.
(2) A record of each Inspection and
each test required by this sunpart.
(34 P-R. 15473, Oct. 4. 1909. as amended oy
Amdt. 195-1. 35 PJL 5333. MmT. 31. 1970:
Amdt. 195-2. 35 KB. 17180, Nov. 7, 1970)
9 195.406 Maximum operating pressure.
(a) Except for surge pressures and
other variations from normal operations.
no carrier may operate a pipeline at a
pressure that exceeds any of the
following:
(1) The internal design pressure of
the pipe determined in accordance with
! 199.108.
(2) The design pressure of any other
component of the pipeline.
'3> Eighty percent of the test pres-
sure for any part of the pipeline which
has been hydrostatically tested under
Subpart E of this part.
•4> Eighty percent o£ the factory test
pressure or of the prototype test pressure
for any individually installed component
which is accepted from testing under
5 195.304.
(b) No carrier may permit the pres-
sure in a pipeline during surges or other
variations from normal operations to
exceed 110 percent of the operating pres-
sure limit established under paragraph
(a) of this section. Each carrier must
provide adequate controls and protective
equipment to control the pressure within
this limit.
[Amdt. 195-3. 35 PJL 17180. NOT. 7. 1970]
§ 195.408 Communication*.
Each carrier shall have a communi-
cation system that insures the trans-
mission of information required for the
safe operation of its pipeline systems.
§ 195.410 Lin« markers.
(a) Except as provided in paragraphs
(b) and of this section, each carrier
shall placs and maintain line markers
over each buried "«• in accordance with
the following:
(1) Markers must be located at each
public road crossing, at each railroad
crossing, and In sufficient number along
the remainder of each buried line so
that its location Is accurately known.
(2) The marker must state at least
the following: "Warning" followed by the
words "Petroleum 'or the name of the
commodity transported) Pipeline" Une markers are not required In
heavily developed urban areas such M
downtown business centers where—
(1) The placement of markers Is Im-
practicable and would not serve the pur-
pose for which markers are Intended:
and
'2) The local government maintain.
current substructure records.
(c) Line markers that have been In-
stalled before April 1. 1970. may be used
until Anrtl 1.1973.
td> Each carrier shall provide line
marking at locations where the Une is
above ground In areas that are acces-
sible to the public.
368
-------
3 105.412 Inspection of righu-of-way
»nd crmeinfcs umler nuvifable w»ler».
(a> Each carrier shall, at Intervals not
exceeding 2 weeks, inspect the surface
conditions on or adjacent to each pipe-
line right-of-way.
i b) Except for offshore pipelines, each
carrier snail, at Intervals not exceeding
5 yean. Inspect each crossing under a
navigable waterway to determine the
condition of the crossing.
5 195.414 Cullimlir prelection.
After March 31. 1973. no carrier
may operate a pipeline that has an ex-
ternal surface coating material, unless
that pipeline Is cathodicaUy protected.
This paragraph does not apply to tank
farms and buried pumping station piping.
ibi Each carrier shall electrically In-
spect each bare pipeline before April l,
1975. to determine any area* In which
active corrosion Is talcing place. The car-
rier may not Increase Its established
operating pressure on a section of bare
pipeline until the section ha* been so
electrically Inspected. In any area*
where active corrosion Is found, the car-
rier shall provide cathodlc protection.
Section 199.418 and 'g> applies to all
corroded pipe that Is found.
(e> Each carrier shall electrically In-
spect all tank farms and burled pumping
station piping before April 1. 1973. as to
the need tor oathodJc protection, and
cathodlc protection shall be provided
where necesnary.
(d> Notwithstanding the deadlines for
compliance in paragraphs (a), 'b), and
ic> of this section, this section does not
apply to offshore pipelines located be-
tween a production facility and a car-
rier's trunkiine reception point until Au-
gust l, 1977.
(40 ?R 43901. 49 CTR 1.53) [34 PR 15473.
Oct. 4. 1969. aa amended by Amdt. 195-2. 35
PR 17188. Nov. 7. 1970: Amdt. 195-11. 41 PR
34041. Aug. 12. 19781
EFTTCTTVT DATX NOTE: In s 195.414. para-
graph id) was added by Amdt. 195-11 at
41 FR 34041. Aug. 12. 1978. effective Aug. 1.
1977.
§195.416 External rorrcMton control.
(a) Each carrier shall, at Intervals
not exceeding 12 months, conduct tests
on each underground faculty in its pipe-
line systems that Is under cathodlc
protection to determine whether the
protection is adequate.
(b) Each carrier shall maintain the
test leads required for cathodlc protec-
tion In such a condition that electrical
measurements can be obtained to en-
sure adequate protection.
(c> Each carrier shall, at Intervals
not exceeding 2 months. Inspect each of
Its cathodlc protection rectifiers.
Each carrier shall, at Intervals
not exceeding 5 years, electrically in-
spect the bare pipe In Its pipeline system
that Is not cathodlcally protected and
must study leak records for that pipe to
determine If additional protection Is
needed.
'e) Whenever any buried pipe is ex-
posed tor any reason, the carrier shall
examine the pipe for evidence of external
corrosion. If the carrier finds that there
Is active corrosion, that the surface of
the pipe Is generally pitted, or that cor-
rosion has caused a leak. It shall investi-
gate further to determine the extent of
the corrosion.
(f) Any pipe that la found to be gen-
erally corroded so that the remaining
wall thickness Is less than the minimum
thickness required by the pipe specifica-
tion tolerances must either be replaced
with coated pipe that meets the require-
ments of this part or. If the area Is small.
must be repaired. However, the carrier
need not replace generally corroded pipe
If the opera tlr.g pressure Is reduced to be
commensurate with the limits on operat-
ing pressure specified In this subpart.
based on the actual remaining wall
thickness.
(g> If Isolated corrosion pitting Is
found, the carrier shall repair or replace
the pipe unless—
(I) The diameter of the corrosion nits.
Is less than.the nominal wall thickness
as measured at the surface of the pipe,
of the pipe: and
f2> The remaining wall thickness at
the bottom of the pits Is at least 70 per-
cent of the nominal wall thickness.
fh> Each carrier shall clean, coat with
material suitable for the prevention of
atmospheric corrosion, and. maintain
this protection for. each component In Its
pipeline system that Is exposed to the
atmosphere.
§195.118 Internal corrosion control.
(a) No carrier may transport any
commodity that would corrode the pipe
or other components of Its pipeline sys-
tem, unless It has Investigated the corro-
369
-------
sive effect of the commodity on th» sys-
tem and hma taken adequate steps to mit-
igate corrosion.
(b> If corrosion Inhibitors are used to
mitigate Internal corrosion the carrier
shall use Inhibitors In sufficient quantity
to protect the entire pan of the 'system
that the Inhibitors are designed to pro-
tect and shall also use coupons or other
monitoring equipment to determine their
effectiveness.
The carrier shall, at Intervals not
exceeding 8 months, examine coupons
or other types of monitoring equipment
to determine the effectiveness of the In-
hibitors or the extent of aoy corrosion
(d) Whenever any pipe Is removed
from the pipeline for any reason, the
carrier must inspect the internal surface
for evidence of corrosion. If the pipe la
generally corroded such that the remain-
Ing wail thickness Is less than the mini-
mum thickness required by the pipe
specification tolerances, the carrier shall
Investigate adjacent pipe to determine
the extent of the corrosion. The cor-
roded pipe mutt be replaced with pipe
that meets the requirements of this
part or. based on the actual remaining
wall thickness, the operating pleasure
must be reduced to be commensurate
with the limits on operating pressure
specified in this subpart.
134 PJL 15413. Oct. 4. 1909. u saonded by
Amdt. 196-1. 33 ML 8333. MM. 31. 1010]
8 195.420 Vnlv* nuintrnnnr*.
(a) Bach carrier shall maintain each
valve that Is necessary for the safe opera-
tion of Its pipeline systems in good work-
Ing order at all times.
(b) Kach carrier shall, at Intervals not
exceeding 8 months. Inspect each main
line valve to determine that it is function-
Ing properly.
ic) Each carrier shall provide protec-
tion for each valve from unauthorized
operation and from vandalism.
§ 19S.422 Pipeline repair*.
(a) Each carrier shall. In repairing ito
pipeline systems. Insure that the repairs
are made In a safe manner and are made
so as to prevent damage to persons or
property.
(b) No carrier may use any pipe, valve.
or fitting, for replacement In repairing
pipeline facilities, unless U Is designed
and constructed as required by this part
§ 195.424 Pipe
(a) No carrier may move any line pipe.
unless the pressure In the line section In-
volved Is reduced to not more than 50
percent of the maximum operating
pressure.
(b) No carrier mar move any pipeline
containing liquefied gases where mate-
rials in the line section involved are
joined by welding unless—
(1) Movement when the pipeline does
not'contain liquefied gases is Impracti-
cal:
(2) The procedures of the carrier
under 5195.402 contain precautions to
protect the public against the hazard in
moving pipelines containing liquefied
gases, including the use of warnings.
where necessary, to evacuate the area
close to the pipeline: and
(3) The pressure in that line section is
reduced to the lower of the following:
(1) Fifty percent or less of the maxi-
mum operating pressure: or
(11) The lowest practical level that will
maintain the commodity in a liquid state-
with continuous flow, but not lest- than
50 pj±t. above the vapor pressure of the
commodity.
(c) No carrier may move any pipeline
containing liquefied gases where mate-
rials in the line section involved are not
joined by welcUng unless—
(1) The carrier compiles with para-
graphs (b) (1) and (2> of this section:
and
(2) That Hn^ section is isolated to pre-
vent the flow of commodity.
[Amdt. 195-2. 39 PR 17185, Xov. 7. 1970. u
amended by Amdt. 185-7. 39 PR 18781. June
4. 1874]
§ 195.426 Semper and uphere ficilities.
No carrier may use a launcher or re-
ceiver that Is not equipped with a relief
device capable of safely relieving pres-
sure In the barrel before Insertion or
removal of scrapers or spheres. The car-
rier must use a suitable device to Indi-
cate that pressure has been relieved In
the barrel or must provide a means to
prevent insertion or removal of scrapers
or spheres If pressure has not been re-
lieved in the barrel.
9 195.428 Overpressure safety devices.
(a) Except as provided In paragraph
(b) of this section, each carrier shall.
370
-------
at intervals not exceeding 12 months, or
8 months in the case of pipelines used
to carry liquefied gases. Inspect and test
each pressure limiting device, relief
valve, pressure regulator, or other item
of pressure control equipment to deter-
mine that it Is functioning properly, is
tn good mechanical condition, w^ is mie
quate from the standpoint of capacity
and reliability of operation for the serv-
ice in which it is used.
(b> In the case of relief valves on
pressure storage vessels containing liq-
uefied gas. each carrier shall test each
valve at intervals not exceeding 3 years
I Amdt. 195-1.37 F.B. 18733. Sept. 18.1979]
§ 195.430 Firefighting equipment.
Each carrier shall maintain adequate
flrefighting equipment at each pump sta-
tion, terminal, and tank farm. The
equipment must be—
(a) m proper operating condition at
all times:
(b) Plainly marked so that Its Identity
as flrenghting equipment is clean and
(c) Located so that it is easily acces-
sible during a flre.
§195.432 Storage renal*.
Each carrier shall, at Intervals not ex-
ceeding 12 months, inspect each storage
vessel '<""i"rf*"g atmospheric ?nfl pres-
sure tanks).
§19&434 Sign*.
^sflli carrier *^*^ WWJM+^JM ffgtiit visi-
ble to foi public around each pumping
station, terminal, or tank farm. Each
sign must contain the name of the car-
rier and an emergency telephone number
to contact.
§ 195.436 Security of facilities.
Bach carrier shall provide protection
for **"h pumping station, terminal, and
tank farm and other exposed facility
(such as scraper traps) from vandalism
and unauthorized entry.
3 195.438 Smoking or open flame*.
Each carrier shall prohibit smoking
and open flames tn each pump station
area and each terminal or tank farm
area where there is a possibility of the
leakage of a flammable commodity or of
the presence of "•"""*hl* vapors.
371
-------
A. 2 FOREIGN - TRANSPORTATION OF LIQUIDS BY PIPELINE
Reference list of ligislation and regulations, standards, technical
requirements, codes of practice for design construction and operation of
oil pipelines in Western Europe80.
372
-------
TRANSPORTATION OF LIQUIDS BY PIPELINE (FOREIGN REGULATIONS)
CO
~-J-
CO
government legislation/regulations
- Law/Decree/Application order
= Pei mil or approval required
- Law ol 1976.07 3 on transport ol pioducu by pipe-lines
- Trade Rights Peimil (Oeslerreichische Gewerbeordnung) from
Ministry ol Trade
= Water flights Permit (WasseirechlsgeseK 1969. revised in I960)
from Provincial Authority lor each Province crossed
- Law ol 1965 04.12 on transport by pipus ol gaseous pioducls
and others
Royal Decree ol 1967.06. IS dealing with transportation by
pipeline ol liquid and/or liquefied hydrocarbons other than
those cited in the 1965 Law
- Royal Ducree of 1967.07. IS specilying security measures
to be taken during construction and operation ol installations
ol transportation by pipeline ol products affected by the
above decree.
- Operating peimil hom special inspectorate alter its
cifi tilication ol the completed installation
Pipelines are normally built in accordance with a declaration ol
"National Interest." according to:
• Finance Law of 1958.03 supplemented by
- Decree ol 1959.06 16 (and Ministerial circular dated
1961 09.14) which sels up Hie code lor "national interest"
pipelines
— Decree ol 1959.08.14 on salely regulations lor pipelines lor
liquid liydiocarbons or liquefied hydrocarbons under pressure
- "Anile" dated 19S9.10 I and modifying "arretes" of
1961 OB 11.1962 07 2. and 1966 1130 which sets
the salely code for liquid pipelines and liquefied petroleum gas
pipelines
Oecievol 1961 OH 1 modified by decree of 1967.12.10
lor the protection ol water catchments and location ol pro
lected areas
- Ministerial circulai dated 1970.06.29 on changing salely
category during operations
- Approvals Irom "Direction des Carburanls" (DICA), "Service
des Mines" and "Piefels" (local councils) lor design and
operations
Rel.
A1
B I
B2
B3
f-l
F2
F3
F4
F6
F6
standards
techn. requirements/code of practice
"Austrian Pipeline Code" issued by the
Austrian Association for the Mineral Oil
Industry, based on IP Coda, but not
acknowledged by the aullioriiies (Regeln
lur Bau und Betrieb von Ferntuitungen lui
Erdol und llussige Eidolprodukle 1964)
Minimum design standards, allowable stress, specilic limits for con
slruclion near lo water catchment areas embodied in Decree of
1967.07
"Rules and Regulations lor pipelines carrying liquid hydrocarbons •
and hydrocarbons liquefied under pressure", til 1968. by Ministry
of Industry. "DuecliondesCarburants" (DICA)
Tills document specifies the negociating piocuduies as well us the design
and construction criteria incoiporaled in the deciees and "arreles"
menliorMid
Common ptescr iptions apfilicaltle to lite
contracts ol public winks signed by the
government: "Fascicule 72 — Oil and Gas
Pipeline" put into ellect by decree dated
1968 1024
They desciibc the rules ol an generally
agreed by the pipeliners
Rei.
A2
F7
F8
-------
CO
>
c
E
j.
o
-
t
•o
it
|
z
government legislation/regulations
- Law/Decree/Applicalion order - Permit or approval required
* "Planning approval" lor *ach "Land" concerned (Verordnung
ganehmigungsbedurtlige Anlayen 4. August 1960) according lo
Chapter 16 and 24 of the Industrial Code (Gewerbeordnungl
« Permit requited lor pipelines transporting crude oil, gasoline.
diesel oil and fuel oil and other water endangering liquid or
gaseous substances
(Wasserhaushallsgeseu - WHC, 2nd amendment of 1964.08.6
' Water endangering substances as specilied by regulation
(Verocdnuna uber wasseigiilaludenda Siolte bei der tUldrderung in
Rohrleilungen. vom 19, Pezember 1973, BGB1I. vom 22. Dezem-
ber 1973 p. 1946)
'Water Right Permit (Wasseihaushallsgeseu • WHG
issued by "Land" authorities
No specific Law
- General approval Irom the Mimuy ol Economic Allairs
- Technical appioval from each ol the provincial authorities and
the local Councils (poldeis, dykes, etc.)
flel
0-1
02
D3
ilandardi
DIN 2413 10.6)
Sleet pipes, for-
mula lor calcula-
ling thickness of
pipes subjected
to internal pies-
stir* 1966 06 re
vised in 1970
and IMI iuuad
in 1972.06
DIN 17172 (0.6)
Steel pipes lor
pipelines for com-
bustible liquids
and gases, technical
specifications.
1966.10
techn.requuemanu/code of practice
Guidelines for pipelines lor lite trans-
portation of hazardous fluids:
a) "Richilinien liir Fernleitungen
lum belordern gefiihrdender
Fliissigkeilen" (REF), published
on 1968.12.6. This document was
revised and issued on 1971.109 22
asTRuFMI (see 0.4b|
b| "Technische Regeln liir brennbare
Fliissigkeilen" (JRbF 301) issued on
1971.10.22 as part of the technical
regulations telatud to combustible liquids;
TflbF issued in 1970 06
Amended by 1973 04 and 1974.02
Pipeline Code 1972 lor the construction
and operation ol pipelines lor the Hans
ponalion of gases and liquids with le-
quiremenls pul for walet protection (pijp
leidingcode 1972) published by the Provin-
ical Authorities ol South Holland,
1972.01.
Guidelines lor oil pipelines (Leidraad
voor Oliepijpleidingenl published in 1973
and prepared by a national comiuilltc on
storage ol dangerous substances (COGS).
National Pipeline Code for dyke crossing
under study by TAW (Technische Adviev
comrniisie voor de Walerkeringen)
Ref.
04
04
NL
NL
-------
GO
~~J
en
Pemui ut approval iec|uiied
government legislation/regulations
LuvWDecieti/AtHititaliuti oitlei
Nu specific Law
- Miimteiial Dcciee I9G6 07 12. revised on 1971.02.23
on technical standards tot pipelines carrying liquids and gas
acioss and parallel 10 railways, uamways and branch lines
- Pei mil loi construction and operation Irum Ministry of
Tiade and Industry on basis of a general decree of 1934 07.31,
"Rules loi Ihe Input, Refining. Storage and Distribution of
Mineral Oils"
Federal Law ill 1963 10.4, "Pipeline Installations lor the
.Tiamporlaliun of Liquid o< gaseous Fuels"
Application Order dated 1968.09. 1 1 of the above Law
- "Arrete" uf the Federal Council ol 1968.09. 1 1 on the
technical control of the installations ol pi|ieline
transput laliun
Ordinance til 106607.1 on safely rules lor the installation
ol pipeline lianspoilahon
- F edei at concession to build and opei ale a pipeline under the
1963 Fedei al Law subjected to Hie above I960 Ordinance on
salely lules
Pipe line! Act 1962 which empowers the Secietary of Slate
lor Energy to control all pipeline construction and operation.
other than thuse ol the Statutory Undertakings.
Land Drainage Act 1961
The Petroleum and Submarine Pipe lines Act 1975 empowers
the Secietary ol Stale for Energy to control the construction
and use ol pipelines in territorial and continental shelf waters,
other than a pipeline which has neither initial nor terminal
point wrilnn those waters
Coast Protection Act 1949
Continental Shell Act 1964
Town and Country Planning Act 1962, for that part of a
submarine pipeline above sea level ILWMOST). provided it is less
than 10 miles in length
del
F I
CHI
CH2
CH3
CH-4
UK I
standards
teclin. requirements/code ol practice
Design standards to be discussed with the authorities lui each piojeci
Specific technical standards embodied in Decree ol 1966 07 ANSI
B 31 A are commonly recognised.
"Regulations on the pipeline installations tor Hie transportation ol
liquid or gaseous fuels" dated 1971.04 30 Federal Board ol
Energetic Economy.
This document brings together the procedures and regulations
incorporated in the mentioned Fedcial legislation
British Standard Coda ol Practice No. CP 2010
"Pipelines" Pans 1 and 2
Institute ol Petroleum Model Code ol Sale Practice
Petroleum Pipelines • Part 6
BS 4616. Specification loi held welding ol caibon
steel pipelines
Ret
CH6
UK 2
-------
DEPARTMENT Of TRAMIPORTATIOM
PIPaiNE CARRIER ACCIDENT REPORT
tnstruc-
tions
A Carrier
nfomtiqr
B
Fine and
Location
of
Accident
i Origin oi
.iguid or
apor
te lease
) Cause of
Accident
E Death
or |n>ry
F
prcoerty
GGereral
nfonaticn
instruc-
tions — »
H
Occurred
Li* .
DIM
MBCET tunCAU «.
004R56ni
Complete ia duplicate. If ike apace provided (or ••)> qucalion ia net adequate, ellack aa additional
akccl. Definition of a reporuble accident ia elated ia ike Code of Federal HcguUlionn, Title 49,
dopier 1. Pert 195. File balk copie* of tkt* report oitkin IS dora after discover,, of ike accident
vitk ike Adaiiaialralor. Federal Railroad Adminiatraiion, Department of Trananonalion. Kaakiagloa.
D. C. 20591 (liefer: Sec. 19S.S4).Oeuiled iaatruciinaa for preparing tkia form are found ia Part 195
Specimen copica of ikia. form trill be xipplicd gpoa rcqiieal oilkoul ckarte. Additional copiea may
be reproduced uaing tka aanto formal an. conrietMAiiON AT POINT
n siiAicMr n «Ae
10. COVER, ir BELOW CHOUNO
NON-CMPLOYCES
• ITEMS 0«»ICCO
•* ITEMS OAtuCCO
I. ESTIMATED LOSS CUE 1' YEAH
ro ACCIDENT INSTALL
•Ao*n e
-<, » ,*., .„„ ,. ,». p,
»s *',-£,"' "s
G3ct"" H '•••• ^..u.i
r^ M.T /-.i.fii rn«*--.
SURE AT TIMC tLKAIICn 11. HO TiiME BICN A P4CS
ICNT Sfl HSI ON SYitC-1
TION OF '«• MAIIMJ4 !'• OAIE OF LAICST '1ST
IEST PKEUUU
NI« r«le
DOT form 7000-1 (8-69)
Figure A-l. DOT 7000-1 (Page 1).
376
-------
&w?£d by
terra* ion
Tauiumt
Kiiotirins
Fiselinr-
>• nrrc or CCMWKICX
a '
ST LINC MMKCft
a«.
^TTieiLif* IMM.II
r>Txjoic notccTiONl.
CJ ttt
O..
conation TUTS
i- IVfL U lei! i,,^
i- LtNCIM 0' 1IVI U.lM.i.fl
PAnOL ON itCIICH
or «ccioo.r or HCSKMSIO.I ornci-u. or OUMIU
J*t AM3 TITLE Of CMflltft
OFflCIM. FILING THIS UKMT
TTLtPnOwC NO- ('•^LU*K
UCA C**C|
CAtt
Figure A-2. DOT 7000-1 (Page 2)
377
-------
R«f
Austria A-l Sund*sges*tz von 3. Juli 1975 ub*r di* gcwerbsmaszige Seforderung von Gutern in Rohrlaitungen
(Rohrleitungsgesetz).
Ausgegeben.un31.Juli 1975-Nr. 411 p 1759- 1771
A-2 Fachv*rband d*r Erdblindustrie Oesterreichs.
"Regeln fur Bau und Betrieb von Ftrnleitungen fur Erdol und Flussige Erdolprodukte • 1964"
Belgium B-1 Loi du 12 Avril 1965 relative au. transport de produits gazeux tt autres par canalisations - Moniteur
Beige du 7 Mai 1965.
3-2 Ministere des Affaires Economiques et Ministere des Transports et PTT.
a. Deem Royal du 15 Juin 1967 Monittur Beige 137:118. 22 Juin 1967. pp. 6753-6754.
Extension d* certain** dispositions d* la loi du 12 Avril 1965 r*lativ* au transport da produits
gazeux tt autres par canalisations, au transport par canalisations d'hydrocarbures liquid** et/ou
d'hydrocarbures liquefies, autres que ceux vises par ('article ler, littera a, de cette loi.
b. Oecret Royal du 25 Juillet 1967 Moniteur Beige 137:169, S Septembre 1967, pp. 9312-9323.
Determination des mesures d* leeurite a prendre lors de I'ttablissement et dans Sexploitation des
installations de transport par canalisations d'hydrocarbures liquides tt/ou d'hydrocarburcs
liquefies, autres qu* ceux vises par ('article ter, littera a, de la loi du 12 Avril 1965, relative au trans-
port de produits gazeux « autrn par canalisations.
Franc* F-1 Loi d* Finance no. 58-336 du 29 Mars 1958 (article II).
Journal Official (J.O.) du 1 avril 1958 p. 3170.
F-2 Oecret nr. 59 645 du 16 Mai 1959 J.O. du 21 Mai 1959 p. 5178-5132. modifie par decret no. 66-550
du 25 Juillet 1966 (art 7 et 33) tt Circulate Ministerielle du 14 Septambre 1961 nr. 2.905
F-3 Oecret nr. 59-998 du 14 Aout 1959 J.O. du 23 Aout 1959 p. 3412.
F-4 Arrete du ter Octobre 1959 J.O. du 3 Octobr* 1959 p. 9SS7 tt arretes modificatifs du 11 Aout 1961
(art 24 liquides), 2 Juillet 1962 (art 49. liquides; art. 47, liquefies) et 30 Novembre 1965 (art 7,
18 tt 48. liquid**; art 10 *t 14, liquifies).
F-5 Oecret nr. 61-859 du 1 Aout 1961 J.O. du 5 Aout 1961 modifi* par d*cr*t nr. 67-1093 du 15
Oecembre 1967 J.O. du 10 Oectmbrt 1967.
F-6 Cireulair* Ministtricil* du 29 Mai 1970 (minister* du Oeveloppement Industrie! et Scitntifiqut,
Direction des Carburants) nr. 3145 en application du decrtt nr. 59-998 du 14 Aout 1959.
F-7 Minister* d* I"Industrie. Direction des Carburants "Legislation et Regtementation des Pipelines
i hydrocarburn liquid** ou liquifiit sous pression". Imprimerie National*. Paris, 1968, pp. 1-106.
F-3 March* d* CEtat March* d*s Travaux Publics.
Cahiw das prescriptions commune) "Fasciculi nr. 72: Oltoducs — Gazoducs" mis tn application
par decret no. 68.1003 du 24 Octobre 1968.
Germany 0-1 Gcwerbtordnung (GtwO) § 16 indtr Neufassung vom I.Juni 1960.
0-2 Vtrordnung ub*r di* Errichtung und den Bttritb von Anlagtn zur Lagtrung, Abfullung und
3cford*rung brcnnbarsr Flussigkaiten zu Land* (Vtrordnung ubtr brtnnbare Ftusjigkeiten -
VbF vom 5. Juni 1970 (BG8I. IS. 689). S 9 Abs. 1 nr. 4 und Abs. 2 und 13 Abs. 1.
0-3 Ge*etz zur Ordnung des Wasserhaushatts (Wassarhaushaltsgtsetz - WHG) vom 27. Juli 1957 (8G8I
1 S. 1110). zuletzt gtindtrt durch da* Gtsetz vom 6. August 1964 und vom 23. Juni 1970 (BGBI.
1 S. 305). § 19 a ff.
0-4a 8und*sminist*rium fur Arbeit, Bonn.
RFF: Richtlini* fur Ftrnleitungen zum Befordern gtfShrdtnder Flussigkeiten.
Bek. des 8MA vom 6. Oezember 1968.11 Ib4-3893.016-4940/68. Arbeitsschutz nr. 12/1963. pp.
347-365.
1- ^tfensnct 30
378
-------
0-4b Tacnnische Regain fur brennbare Flussigkeitan — TRbF - aufgestellt vom 0«utichen Aunchuts
fur brennbare Ftussigkeiten (DAbF) und veroffentticht vom Bundesminister fur Arbeit und
Sozialordnung im Bundesarbeitsblan, Fachteil "Arbeitsschutz" (nacfi der Allgemeinen
Verwaltungsvoncnnft zu § 8 dtr Verordnung ubtr brennbara Flussigkaiten— VbF-vom 12. Mai 1970).
TrbF 301 Richtiinie fur Ftmleitungan zum Befordem gefahrdender Flussigkeiten (RFF) Ausgafae
September 1971. Revised on 1973.04.9 and 1974.04.1.
0-5 DIN 2413 Stahlrohre. Berechnung dar Wanddicka gegen Innandruck. in dar Faming vom Juni 1972
OK 621.843.23: 669.14-462.
0-6 DIN 17 172 Stahlrahra fur Famlaitungan fur brannbara Fluaigkaitan und Gate. Technischa
Lieferbedingungan. Oktobar 1966.
Italy- 1-1 Oecreto Ministerial* 12 luglkj 1968 — Norm* Tecniche per gli attraversamanti ed i parallelismi di
condotta di liquidia a gaa eon linaa farroviaria, tranviaria at eon binari di raceordo. Gazetto
Ufficiala nr. 221 del 6 sattambra 1966. pp. 3-15.
Raviiad 23 faobraio 1971 - Gazatto Uffieiala dal 7 maggio 1971.
NatnarlandiNL-1 Provinciate Watantaat in Zuid-Holland. "Pliplaidingeoda 1972". Ijanuari 1972 • revuie 3. 1 januari
1974) eitan ta statlan aan pijplaidingan voor net transport van gassan an vlotutorfen mat batrakking tot
da Watamaatkundiga vailighaid.
Studiagroap Pifplaidingan voor gattan en vloaiitorfan.
NL-2 Commmia Optlag Gavaartijka Scoffan (COGS).
"Uidraad voor Oliapijplaidinoan" - 1973 publisnad by Ministry of Social Affairs (Directoraat-
Ganaraat van da Arbatd van hat Ministaria van Soeiala Zakan, Voorburg).
Swnzariand CH-1 Loi Fadaraladu4Avril 1969surlaiiratallationsdatran>portpareonduita$dacombustibl«ou
carburants liquidaf ou gazaux (loi sur In installatiora da tramport par conduitat). RO. 1964 95
CH-2 Ordonnancad'exacutiondu 11 Saptamora 1968 da la loi fadarala sur IM installations da transport par
cooduitas da combustibtai ou carburana liquidas ou gazaux.
RO. 196811621404
CH-3 Arrata du Contail Fadaral du 11 Saptambra 1968 concamant la survaillanca tachnigua das
installations da transport par conduitas. RO. 1968 1185.
CH-4 Ordonnanca du 1ar Juillat 1966 concamant las prescriptions da sacurita our las installations da
transport par conduitas. RO. 1966 497.
CH-5 Offtaa Ftdtr^l da I'Economia Enargiqua.
Prescriptions sur las installatiora da transport par conduitas da combustibles ou carburants
liquidas ou gazaux. Etat la 30 Avril 1971 pp. 1-71.
United UK-1 Her Majesty ^Stationary Office. London. Pipelines Act. 1962. 10 & 11 Eliz 2. CH 58. pp. 1-66
Kingdom
UK-2 British Standards Institution (BSD London. Coda of Practice for Pipelines. CP 2010.
Part 1 "Installation of pipelines in land" - 1966. pp. 1-22
Part 2 "Design and construction of steal pipelines in land" 1970. pp. 1-S6.
USA (For information)
ANSI: American National Standards Institute.
ANSI 8 31.4 "Liquid Petroleum Transportation Piping Systems".
API: American Petroleum Institute. New York.
API-SI; Specification for Line Pipe
API-5LX: Specification for high-test Line Pipe
API-SLS: Specification for spiral welded Line Pipe
API-1104: Specification for Field welding of pipelines
API-1105: Recommended Practice on construction of steal pipelines
Code of Federal Regulations
"Minimal Federal Safety Standards for Liquid Pipelines" related to transportation of
liquids by pipeline. (Part 195, Title 49 of ttie Code).
379
-------
APPENDIX B
PIPELINE MILEAGE DATA
ptp«lUtn» activity La ch« Ualud Seaui,
I, 1974 n JMMIT I. 1977
(Mil..)
Seme*
p|
Wyoming
Total
Total
In pi*c«
Jan. 1, 1974
I 759
124
I 352
Z 997
10 183
2 ISO
&
273
1 386
633
11 405
3 951
2 326
354
220
£/ 272
£/ 2 369
3 389
& 928
2 906
3 283
198
108
r/ 453
5 927
I 939
900
I 741
r/ 20 762
673
17
642
707
r/ I 048
'77
334
760
942
6,397
£/ 223,535
?i3
10
j.1
~ 16
597
24 , 006
Tocai
la olac*
Jan. 1, 1977
;<»3
3 071
10 352
2 396
3
2 024
633
10 914
4 552
4
-------
F«en»Uu» ptp,Unin« «rtvtey in ch. Unit.d Sucu
January 1, 1971 Co J.«a.ry 1, 1974
CMtU.)
State
Tool
Total
la plaea
Jan. 1, 1971
1 237
87
I 071
2 397
r/ 9 358
1 756
92
3
87
1 728
640
r/ 11 096
4 495
3 389
16 013
2 311
7 956
353
219
242
1 744
2 955
3 058
6 295
3 054
3 341
328
108
568
5 941
1 673
834
1 664
6 909
22 308
689
3 291
17
635
640
629
65 259
I 042
177
322
762
3 612
942
6.644
r/ 218,671
Mpe
H«v
421
35
272
565
717
275
213
184
6
1 254
283
376
500
3
1 296
94
156
107
435
562
127
149
189
2
1 139
74
70
63
420
1 401
11
87
71
78
1 436
39
3
129
41
644
13,957
laid
Second-
hand
101
4
9
408
760
236
3
4
72
635
1 310
834
1
1
11
297
106
78
336
188
182
2
38
1
223
247
113
123
292
12
162
37
1
3 429
504
11
3
548
1
136
12.326
?tp«
Cakatt
up
2
373
1 152
87
27
30
13
221
314
1 241
1 503
1 295
268
69
182
265
563
389
321
38
99
1 376
55
4
99
487
3 454
39
502
37
35
1
4 652
744
7
134
661
1
597
22,599
Total
In placa
Jan. 1, 1974
1 759
124
1 352
2 997
10 133
2 180
92
5
273
1 386
633
11 405
4 &29
3 951
15 907
2 326
3 791
354
220
347
3 929
3 099
3 389
6 928
2 306
3 283
198
108
472
5 927
1 939
900
1 741
6 965
20 547
" 673
17
669
642
707
65 472
1 341
177
334
760
3 540
942
6.897
Z2Z.3SS
381
-------
Petroleum pipelining activity in ch« United States,
January I, 19«8 to January 1, 1971
(Miles)
Seat*
Total
In place
Jan. 1, 1968
Pip* laid
Second-
hand
taken
up
Total
IB place
Jan. 1, 1971
1,121
Alaska .............................. 74
Arizona 1,064
Arkansas 2,334
California 9,336
Colorado 1,739
Connecticut ......................... 91
Delaware 3
Florida ............................. 35
Georgia 1,567
Idaho 639
IlUnol 10,707
Indiana 4,309
Iowa 3.169
Kansas 13,486
Kentucky 2,219
Louisiana 7,373
Mala* 544
Maryland and District of Columbia ... 235
Massachusetts ....................... 239
Michigan 3,797
Minnesota 2,909
Mississippi 2,356
Missouri 6,434
Montana 2,671
Nebraska 3,114
Nevada 331
Wev Hanpahire 109
New Jersey 545
Nev Mexico 5,333
Sew York 1,368
North Carolina 333
North Dakota 1,723
Ohio 7,168
Oklahoma 22,250
Oregon .««....«...«.....««••...«...«. 402
Pennsylvania ........................ 3,943
Rhode Island 17
South Carolina 567
South Dakota 575
Tennessee ...............•.....•••.•• 519
Taxas 60,316
Utah 1,362
7eraont 177
Virginia 307
Washington 672
Uest Virginia 3,714
Wisconsin S94
Wyoming 5.883
Total 209,478
258
641
153
1
2
165
1,262
386
753
1,753
93
1,015
3
562
46
758
112
497
422
2
304
23
1
37
687
1,313
286
212
74
65
112
6,978
45
18
155
99
346
957
22
20
9
43
402
125
50
129
1
91
23
3
236
6
322
8
1
28
129
14
5
162
22
3
27
329
68
5
6
81
392
1
171
1
1
1
2,306
6
1
147
2
225
12
2
238
711
261
133
1,041
223
33
1,462
107
954
202
17
26
744
70
236
276
217
11
1
6
1,023
286
3
157
1,027
1,647
1,035
1
7
3
4,841
371
66
348
421
87
1,071
2,397
9,368
1,756
92
3
87
1.728
640
11,019
4,493
3,889
16,013
2,511
7.95*
353
219
242
3,744
2,953
3,058
6,295
3,054
3,341
328
108
548
5,941
1,673
834
1,664
6,90*
22,308
60*
8,291
17
635
640
S29
65,259
1,042
177
322
7«2
3,512
942
5.644
20,748
6,630
18,252
218,604
382
-------
Petroleum pipelining activity in Che United States,
January 1, 1974 to January 1, 1977
(Miles)
Size
(inches)
3
4 I/
6 11 .-
3
10
12
14.
16 ;
18
20......
22
24
Over 24. .......................
total
Gathering Lines
Nev pip*
laid
793
571
1,550
321
486
120
130
4
7
2
2
6
4,492
Secondhand
pipe laid
416
551
1,250
529
414
192
252
" 5
5
42
3,656
Pipe
taken up
2,131
1,890
3,105
1.133
1,018
177
54
12
27
9,597
Crude-oil trunklines
New pipe
laid
6
4
153
435
381
773
257
198
17
386
14
723
1,063
4,910
Secondhand
pipe laid
6
13
265
462
699
123
207
7
250
34
58
414
51
96
2,690
Pipe
taken up
28
33
502
1,047
2,701
578
491
16
141
33
165
33
1
59
5,878
Size
(inches)
2.
3,....
4 l/
5 2/
g ..
10
14
16
18
20
22
24
Over 24
Total
Products trunklines
New pipe
laid
7
48
276
2,350
2,673
1,623
703
255
157
59
139
209
8,499
Secondhand
pipe laid
10
11
235
535
304
947
660
295
106
15 .
31
23
68
3,290
Pipe
taken up
48
117
436
2,156
3,113
1,224
672
519
140
49
4
3
8,531
Total all pipelines
New pipe
laid
806
623
1,979
3,606
4,040
2,516
1,090
259
362
76
527
16
723
1,278
17,901
Secondhand
pipe laid
432
580
1,750
1,576
1,417
1,262
1,119
307
361
49
39
414
74
206
9,636
Pipe
taken up
2,257
2,040
4,093
4,336
6,832
1,979
1,217
535
281
132
131
33
1
39
24,006
I/ Includes a small amount of 5-inch pipe in trunklines.
2/ Includes a small amount of 7-inch pipe in erunklines.
383
-------
-Petroleum pipelining activity in the United States,
January 1, 1971 to January 1, 1974
(Miles)
Size
(inches)
3 ......................
4 i/
6 1l
8 ......................
10 .....................
12
16
18 .....................
20 .....................
22 .....................
24 .....................
G
New pipe
laid
488
423
1 287
1,1.01
505
337
97
185
2
4
1
1
4
3 334
*thering line
Secondhand
pipe laid
1 024
1 028
2 305
336
910
66
• 10
1
.
^
5 680
*
Pipe
taken up
2 428
1 597
3 674
1 899
817
95
363
_
7
—
10 880
Cru
New pipe
laid
16
54
I T e
689
717
167
718
136
103
46
157
28
3
130
3 079
de-oil trunkl.
Secondhand
pipe laid
Q
290
•J^O
4O£
1 im
155
9f)A
120
44
6
SI
32 .
3 120
Lnes
Pipe
taken up
62
207
9AA
1 027
1 149
837
109
185
1
in«
57
26
59
5 092
Size
(inches)
4 i-/
6 2/
12
14
16 .....................
18 .....................
22
24 .....................
Over 24
Total
Products trunklines
New pipe
laid
8
298
136
1,133
2,539
1,034
509
358
263
260
172
334
7,544
Secondhand
pipe laid
16
63
290
553
474
660
30
156
269
110
905
3,526
Pipe
taken up
22
201
322
1,850
2,425
450
914
169
142
2
116
14
6,627
Total all pipelines
New pipe
laid
512
775
1,538
2,327
3,593
1,298
1,412
496
370
306
330
29
3
968
13,957
Secondhand
pipe laid
1,049
1,381
2,867
1,485
2,691
881
246
156
390
44
116
83
937
12,326
Pipe
taken up
2,512
2,005
4,240
4,776
4,063
1,694
2,114
278
327
3
431
57
26
73
22,599
_!/ Includes a small amount of 5-inch pipe in trunklines.
2j Includes a small amount of 7-inch pipe in trunklines.
384
-------
•Petroleum pipelining activity in the United States,
January 1, 1963 to January 1, 1971
(Miles)
Siza
(inches)
3 l/
4 f/ :::::::::::::
6 3/
12 ...............
16 ...............
20
Over 24
Total
G
Hew pip*
laid
696
580
1 664
573
99
38
9
37
.
3,746
•athering line
Secondhand
pipe laid
32
259
1 605
139
370
19
184
1
_
2,659
a
Pin.
taken up
1,633
1 653
3,394
1 448
595
31
50
43
^
— .
.
9,397
Cru
Hew pipe
laid
106
71
908
1 380
348
558
790
18
342
336
5
7
1.646
7,015
de-oil tnrafcl
Secondhand
pipe laid
53
40
451
513
445
113
225
22
17
11
36
2 '
3
9
1,990
ines
Pipe
taken up
12
48
344
1-355
1,549
1 002
197
17
70
1
147
2
3
17
4,764
Size
(Inches)
2 ................
3 I/ .............
4 2/
6 3/ .............
14 ...............
16
22
24 ...............
Total ..........
Products txunklines
New pipe
laid
42
31
553
2,012
2,791
1,622
1,590
211
465
95
91
434
9,987
Secondhand
pipe laid
128
121
237
183
714
118
59
106
110
23
4
178
1,981
Pipe
taken up
90
221
438
1,465
693
142
106
5
413
136
263
119
4,091
Total all pipelines
New pipe
laid
344
732
3,125
3,965
3,738
2,268
2,389
229
344
95
427
5
7
2.080
20,748
Secondhand
pipe laid
263
420
2,293
835
1,529
250
468
123
128
34
90
2
3
187
6,630
Pipe
taken up
1,735
1,922
4,676
4,268
2;837
1,225
353
22
525
137
410
2
3
136
18,252
II Includes a. small amount of 2-inch pipe in trunklines.
y Includes a small amount of 5-inch pipe in crunklines.
T/ Includes a small amount of 7-inch pipe in trunklines.
385
-------
•Mileage of gathering lime .In ch« United Statee M of January 1, 1971
(Hllee)
State
Total Jan. 1, 1971 ....
Total Jan. 1, 1968 ....
Size (Inches)
2
Z
239
78
22
653
224
748
923
393
33
121
91
12
67
310
11
991
2,639
2,039
2,368
2,187
72
14.463
13,318
3
90
180
19
1,134
331
1,377
318
89
174
39
177
10
268
113
91
391
2,197
632
2,672
I
330
91
11.166
11,980
4/
18
437
396
101
1,360
202
2,707
466
612
266
257
209
99
2,125
7
266
333
4,339
197
12,237
43
239
609
27.769
28,394
6 y
9
48
1,001
113
232
26
677
201
440
48
82
270
206
387
12
39
143
2,084
27
4,317
43
194
479
11.280
12,016
8 I/
11
10
486
96
67
495
34
305
11
52
16
61
229
7
1
668
1
1,884
22
3
157
4.716
4,842
10
1
204
29
35
2
135
2
•
12
110
325
9
8
872
846
12 3/
•
<*
27
1
•>
61
311
2
1
5
•
274
73
5
32
794
611
Over
12 -
2
3
I
45
13
8
72
117
Total
Jan. 1,
1971
40
843
2,574
334
3,496
783
6,100
2,145
2,330
532
388
764
388
3,293
19
432
407
1,363
12,331
2,916
24 ,086
125
3,173
1.548
71.132
-
Jan. 1,
1968
18
776
2,722
346
3 ,720
916
6,934
2,198
2,493
1,027
466
746
581
3,224
550
482
1.854
12,544
3,419
23,704
290
3,275
1.639
—
74,124
I/ Ineludaa a mall Mount of 3-lnch pip*.
2/ Ineludaa a null amount of 7-Inch plpa.
I/ Ineludaa a sail nount of 11-inch pip*.
386
-------
of lathartai llaaa U tha Daltad Itaeaa aa ot
OU.lt.)
I, 197*
staca
Wymlnt
Total January 1, U74 «...
Tocnl January I, 1971 ....
lUa (lath**)
2
2
•
284
131
6
164
164
751
319
306
35
102
7i
5
133
169
11
1,001
2,234
2.048
2.3S1
2,104
49
13,547
14 ,'>63
3 .
2
118
220
60
998
275
1,429
445
111
207
49
187
17
312
US
94
384
2,060
616
2,615
1
S40
165
11,020
11,166
: *v
28
4
588
545
155
1,146
180
2,713
423
788
400
221
260
100
2,063
6
278
323
1,644
193
12,660
48
218
703
27,687
27 ,7(9
«_a/
22
4
61
878
118
3
203
17
719
405
60
57
322
177
12
537
82
146
1,395
27
4,227
47
174
509
10,222
11,280
«2/
11
•
10
457
65
6
26
484
161
315
26
14
13
51
7
206
13
1
414
1
2,534
21
1
265
5,146
4,716
10 */
•
23
•
1
202
4
29
30
31
133
2
12
133
•
305
9
2
940
872
12
*
24
5
1
61
144
2
1
5
260
86
5
32
•26
754
Ov«r
12
•
•
3
1
1
41
12
•
16
78
112
Total
Jan. 1,
1974
65
23
8
1,062
2,460
434
13
2.967
636
6,207
1,880
2,247
728
459
861
350
19
3,288
490
478
1,855
10 ,.162
2,883
24,794
131
3,019
1,725
69,266
-
Jan. 1,
1971
40
•
*
943
2,574
154
3,496
783
6,100
2,145
2,330
332
588
764
368
19
1,291
412
407
1,861
12,331
2,916
24,066
123
3,173
1,548
—
71,132
IiicluUui t niMll umme of 5-lncli nlna.
2l Iiicluuim a ttwill amaunc of 7*inuh iiipo.
T/ Ineiuila> a null mount at 9-lnch pip*.
4/ Includai « nail uount of 11-laeh plpa.
387
-------
Mllaaia of lUhMloc llM* In MM Uoltad SUM* •• of January 1, 1977
(MllM)
Seat*
local January 1, 1977...
Tocal January I, 1*74...
Slia (lacuna)
2
342
2M
26
45*
126
561
749
300
33
111
49
5
111
277
11
810
l.MO
1.913
2,439
1,MS
56
12.S7S
13,347
3
127
230
73
902
226
1,337
410
134
316
40
129
17
312
46
63
241
1,906
461
2,637
1
364
170
10.232
11,020
4V
19
566
566
167
1,269
171
2,819
393
739
463
190
301
too
2,129
7
304
249
3,163
173
12,111
44
209
760
27,362
27 ,667
«!/
21
53
696
121
9
210
18
818
539
82
11
319
176
319
94
106
1,432
17
4,189
60
119
574
10,423
£'
10,210
«!/
21
474
66
6
31
630
118
468
47
50
46
32
221
13
42
497
1,877
21
'3
268
5.021
£5.11,
10 y
182
4
29
38
208
43
2
56
73
423
9
2
1,075
940
12
11
10
1
151
336
20
2
3
41
229
88
3
31
934
626
Over
12
3
2
2
10
25
•11
37
112
78
Total
Jan. 1,
1977
40
21
1,090
2,691
490
19
2,898
541
6,400
1,692
2,776
1,037
421
846
330
3,615
330
507
1,311
9,226
2,566
24,023
138
2,683
1.863
67,796
-
Jan. 1,
1974
65
23
. 8
1,062
2,460
414
13
2,967
636
6,207
1,880
2,247
728
439
861
330
,.JL
490
478
1,853
10,162
2,885
24,794
131
3,039
1.725
_
69.247
lacludna a Mall aanune ot 5-lnch pip*.
Incluoaa a anall aaount of 7-Lnch pip*.
I/ Include* a anall aaoune of 9-iaeh plpa.
4/ lacludaa a Mail anount of 11-Inch pipa.
11 taviaad.
388
-------
AUb-.
T><|>M .%»•••••>•»••«•«••••«•••••• •••
JoUnn** w« MMrtw *4 CaliiMi ...
""""""" •
•M VMk *••••*•••*•••*»•••*••••••*••
itorth CoMllM **..*..*...•*.«.•.....
MMffc Dvtoc* •••*•••••••••••*•••••*••
OM* • ••on ••••••••• •••••••• ••
foMwvlmiU •••••••••••**•••*••••*••
SflMh talWW ••••••••••••••••••••••••
t»«il J«M«*y 1. I9tt
i> t^u^ » ~u .•MM .* i.uar
I/ lMl*«M • Mil II !•»! W 7-tMk
4
-
-
-
.
»
*
•
1
*
It
I
>
u
•
M
Ul
Hn>
m*
»V
•
Ul
-
u
9
1
Ml
M
1*
•
I
11
I
11
Ul
w
•
1
1
- ...
1,4 It
*v
1*
m
31
*M
rt
«
i«
1.4H
MT
jSfc
It
IM
ts«
i»
Itt
' u
in
Mt
Ul
11
•M
,.,«
5*4
It
54
It?
U Til
It. Ml
•
M
Ml
Jf
Ml
U
34*
1.009
I. IS*
m
»
t»
471
Mi
M
X.UJ
nt
»
in
141
131
l,J«3
4tff
l.M*
54
ISS
„ 4)^
il.ui
I*
u
IM
US
4
3?
Itt
T*
454
III
|
Ul
•
m
•u
ti
in
IN
1H
TM
Wl
as
i«
tu
l.M*
I
1]
U
tot
, ,„
1,111
U
»«I
a*
9U
IN
K
1«
100
IM
tn
M
u
IM
III
tu
1
41
71
U
)St
4»
IM
10
11
t.JTt
1
10*
H
41
1 712
1.109
Mo (
U
-
•
ITS
•
140
Ml
.
^
-
.
•
»
M
IM
M
1
no
lot
JOS
no
Ul
.
I Ul
1,1»
UOMt
u
1
1
it
•
•
.
IM
H
.
IM
U
uo
u
4«
1
U*
.
•
u
14
111
71
I 401
1.441
U
110
•
m
"
11
!
Ul
-
Ill
t
»
»
,
tu
.
•
H
»
.
.
„
.
fit
«7
»
•
m
•
•
.
m
Ul
•
"
70
-
H
.
77
.
•
ft
.
.
.
.
90?
1.9M
11
-
•
.
•
•
.
-
,
.
•
*
•
•
.
m
m
•
.
•
.
9
.
.
.
*j
9
H
(•«
•wr»
190
-
m
-
i
111
.
Ml
10
too
II
14?
17!
m
170
110
110
.
2.171
1.410
T.
M. 1,
im
l.Ul
»K
Ml
l.Ht
4M
11
1
17
1.710
440
1.J11
1.147
M
i.ni
IM
IM
Ml
1.410
l;m
1.110
IM
Ht
i.ai
•»
410
1.J1
1,41
M<
4,011
17
41
440
14i
ll.OH
11
»
74
«44
trll
72.404
Ul
J~. t.
I MO
1,041
4H
Ml
1,40)
SOI
(I
1
11
I.J47
411
2,111
1,101
4f
l.OM
in
in
in
1.271
111
1.021
m
Ml
t.uo
I1J
401
1.J11
1,100
401
4,017
17
147
17J
US
10.400
400
Ml
4M
7
444
I.Mt
w.m
389
-------
Mta«M •> infcil rtnllftm to H» MUM IMm. M •€
(llllMl
IU» (IllHil)
4V
•V
m, I,
19M
Maw* .
rteu* ..
iMlilim ............
MlM .....•.•...••...
l»l|ll«« •* HKMM
HtnlMtffl .
miMM ....
1.1»
i,«i
1,177
47
IM
M
171
m
197
100
It
i
IN
4T>
n
701
000
II
1.14,
I
lit
1,111
n
n
in
i
71
n
t
M
M
Mt
7U
m
1.J00
i.itt
1,491
1.441
M
n
140
I.M
Ul
174
1
U0
IM
79
4U
l»
111
1.117
441
IM
710
409
•11
U
U
,.».
I.4U
It
Mt
1.077
1.9M
919
It
I
m
1.0H
ID
4,199
1,171
1,191
1.J17
100
t.m
in
no
»7
i.iw
i.m
}'.m
MO
7,791
IM
III
1,101
1.140
900
477
l.JOl
4.011
4n
1.M9
17
MO
Ml
Ul
11.100
Ml
IM
m
»
1.141
I
17
1,710
1.444
1,310
1.M7
5.970
m
lit
74t
1,410
1.771
1,111
1,110
l.Otl
t.l»
in
Mt
1.391
1,1*0
IM
410
1.111
1,491
4,0»
17
111
440
Ml
11.0M
111
in
711
•
J"
J™
1,090
1/141
u,;n
7I.IM I/ 9.791
I.7W
l,4tl
f 1,171
17, JM
tnteM • <«ll i
390
-------
of product pip«Un*s in the Unit** State*. •• of January L, 197?
Sc»ta
Maryland ad District of Colu^ia..
2*~v*
UT°-lnt
Total January 1. 1977
Total January 1, 1974
Sin (Inchai)
2
12
1
2
8
1
1
1
Z
2
41
54
123
lit
}
44
272
11
2
40
1
2
7
]
24}
2
1
930
«•.
4 I/
5
34
103
12
6
•0
22
133
3*
267
21
14
4
17
t
1
107
17
14
41
J21
2
63
1.275
2
I
60
11
2,895
EZ.»70
6 2/
'3*
t
173
38
403
447
22
112
197
277
396
1,331
1,4*2
41
516
123
60
17«
407
337
t
lit
82
1,044
90
tw
3M
3M
143
82
S19
7M
61
1,326
14
»
329
3.649
9
55
156
249
71
527
17,565
r/
nt,7«»
t
95
2
242
56
945
87
1
712
55*
1,370
746
1,643
2,931
10
571
24
9
459
•77
1,431
399
1 ,03»
141
41
7»4
390
237
194
1,377
1.894
343
1,223
240
91
249
4,441
299
135
174
lit
172
692
27,193
«|7.»
10
30
74
403
470
162
7
123
320
79
646
617
314
1,306
1
493
167
109
1,526
443
39
13
163
197
109
220
70S
516
329
104
19;
1,789
5
13
14
109
171
11,939
r/
~10,612
12
402
2
292
249
323
64
254
667
236
923
713
315
37
24
229
291
522
299
152
2
37
69
19
609
901
213
10
22
32
1,521
2
109
23
40
9,394
£8.703
14
19
263
149
370
110
295
I
44
79
94
169
94
1
6
299
102
395
209
141
2,817
2,796
It
2
3
1
78
4
200
302
84
249
1
111
24t
77
62
34
12
55
to
149
76
54
126
72
2,077
£(.,54
18
310
260
12
39
13
110
312
2
114
10
1,202
1,177
20
1
297
log
195
44
4
104
70
51
34
103
131
76
1.239
r/
1,072
22
-
9
9
9
24
(and
ovar)
593
16
3
365
233
19
405
99
455
294
96
194
333
155
203
521
220
4,193
3,896
Total
Jan. 1,
1977
1,463
90
967
1,095
2,859
969
94
3
235
2.024
633
3.917
2.696
4,539
6.734
99
2,822
123
219
242
1,309
1,447
1,496
3,927
930
2,330
275
494
1,699
1,117
896
496
3.739
4.706
414
4,012
17
669
642
475
14,130
306
924
713
426
4t4
1.403
81,296
-
Jan. 1,
1974
1,462
19
967
1,077
2,909
839
92
6
232
1,886
633
4.159
2.671
3,892
r/ 6,013
109
2,804
125
220
272
1,490
£/ 1,475
1,579
3,711
618
2,293
199
453
1,193
1,148
900
477
3,581
I/ 4,250
673
3,949
17
669
642
443
I/ 312
834
696
338
466
1,212
.
!/
78.039
I/
I/ tneludaa a
11 Includaa a
.all aaount of 5-Inch pip*.
«11 amount of 7-inch pipa.
391
-------
•Mtluc* •< «•<• cmkllm u en* OUM4 tueu «• «f Juury 1, 1*71
CMllM)
Jt«t
•ntk tekm .....
Jm»ry I, 1971
JM«T 1, !»»»
D»fcr
4
10
I
U
1
3
1
12
1
30
10
14
131
130
7
40
471
11*
«y
3
31
31
115
11
10*
20
111
1*7
M
93
4
14
17
24
114
172
1,314
7
1
1*
I.OIO
2,075
•V
171
4*1
2*0
21*
127
(IS
11
41)
lit
»7
M
n
37
73
IM
71*
1*1
" 31
4U
71*
VN
I,NT
-.,
30
21
155
1,711
93
241
1*5
(1*
1
4H
111
241
1
1*7
34
3*2
371
2,OO
121
' 11
11
1.4*1
W.26)
1I.S2S
10
4
)
(M
242
904
310
(41
117
1
333
117
117
9*
13
1,101
11
1,3*1
414
til
11.1*7
ll.lt*
11
1
21
517
(7
947
207
25
1)5
311
7]
1
11
1,14)
4*0
11
1*
*4
2*
171
311
3*4
1,311
2
3*
(17
10.10*
*.nf
14
14
Ul
41
14
It
(4
M
2
2
t)
62
371
541
IU* (tMt
1*
147
30*
12
(1
to
1
14
411
1*1
115
1
11
too
41*
404
1*
3*1
1
1,(*0
411
5.630
1.J41
U
Ul
11
23*
It
217
(4
71
2*1
35
21
1
4*
(5)
3*
11
2.001
l.ttl
M)
10
41
2t*
200
10
421
205
I*
401
577
It
4
451
S
Ul
10
1,1(3
t
2M
3.040
4,7*5
21
(5
3*5
Ul
in
340
7
IM
311
37
t
131
141
114
174
•
. 2.M4
2,tl)
24
4
13
2
30
7*
114
34
1
24*
111
3)
1.304
1,4)7
2*
20*
1
211
21
3
301
U
t.Ol)
110
21
V
-
.
.
30
intf _
(
4
227
(3
31
T4
1
It*
211
21*
21
110
2
430
2.370
Ml
Tttll
J««. 1,
1*71
34
7*
377
631
4,431
• 904
4,07*
1,134
41
3,tl7
31*
3,3*7
21*
1,102
1,1*1
1,347
3,177
1,2*7
640
too
1,441
11)
1,313
(,*«*
1,3(1
2(4
21.11*
604
177
411
47*
75.0tt
.
1*1*'
34
74
3H
(37
4,411
01
3.11*
1,25*
66
3,962
271
3,011
21*
1,4»)
1.150
*M
3,413
tst
(17
10)
1,47)
Ul
7*3
1,911
(,***
1.4*7
154
2t,004
314
177
(4
412
130
3.233
.
70,123
£/ tIMfttfd** • KMll OTalMt *f 7-lM* piM*
392
-------
•MtlM(« «f
ennkllMi l« tta O.lt.4 Ititn u •«
0«Ui>
I, MM
5t««
HldttfM ••••••••••*•
Total JMU I, 1974.
fetal J«. I, 1971.
2
.
j
7
2
-
1
25
I
MM
v
3
.
ft
14
17
1
31
7
7
1
479
4lf
*V
1
28
72
fti
tu
42
303
174
99
u
1 t§4
7
ft
71
3,344
X/ 5,»l
«y
37
371
3W
237
234
11
•94
273
9)
27
74
180
3 414
3J
44)
7M
l,«4|
I/ »,*"
«i/
}X
I*
9
111
1.253
93
I
491
307
1
3tl
1(1
M
I.S37
20.424
i/ 19,221
10
5
19
4H
247
121
1
ISt
17^
4 MS
431
IH
U,OM
11.M7
. IX
102
t
21
11
412
(1
937
374
t
11
1 14f
451
3(
91
111
184
621
2
jf
621
10, l«
£/ 10,094
14
74
•
mi
41
If
"
a
94
2
f
f?
S9t
571
Sit
u
37
347
321
12
18
39C
lil
31
100
404
39
20
m
421
3,H8
3,430
I (iMfa
18
-
•
191
to
143
72
283
2fl
3f
u
1,091
2,001
•)
20
45
298
200
30
18
(49
978
1
44
m
288
4;f72
I/ 5,117
22
-
•
(5
7
381
271
,
_
.
l,M4
I/ 2, *10
24
.
•
134
^
.
1,483
$/ 1,508
28
.
•
208
2|
U
1,034
1,039
30
IM'
it tncl«4«« • Mall Marat o( t-tMh f If*.
«/ frl« » ItT* li ' '
393
-------
ut* »«
tt» o«ui< lout •• «c
t, un
SUM
IMll J~. I, 1M7...
t«t»l Jw. I, V17»...
i/ i£££! I MI! -
x
4
W
12
I
«4
100
me «
wtic »
iwiC oi
.
2J
1
B
t
i»
m
i
tu
4?»
S-iMll
1-Ud
1
I
47
34
144
91
144
u
19
194
191
1,434
M
3.240
3,344
1
37
.
302
204
147
394
9)
91
2g
93
74
132
904
*02
3,274
SO
*n
9.31i
1,444
22
19
20
123
2 122
93
131
410
234
309
31
324
14
491
2.293
94
1.911
91
1.441
19.303
20,424
10
ft
3
19
545
244
290
327
344
167
M
127
99
*i
1,002
310
4.743
447
27t
11.391
11,040
1
21
11
343
(•
204
29
679
339
73
..
1 246
ill
jg
84
91
170
24S
624
13
3,200
39
613
10.154
10,103
24
.
^
LS4
B
14
9
f
4
94
_
2
-
90
41
309
590
StM (I*
37
347
328
12
11
441
100
409
329
12
630
8
1.707
39
27
421
3.973
5,644
eh«t)
so
227
203
39
*•
m
44
407
13
2.039
2,09t
20
m
100
279
89*
2n
5.251
4,972
22
207
116
143
317
3,059
1,644
24
m
34
^
249
1.014
2,254
1,411
24
-
a
.
^
21
j
-
474
t.M*
I.IW
10
».c)
a
20
7
647
3.367
t.SQZ
T0M
M. 1,
1977
190
1 647
770
I 275
* 77S
1.319
27.811
674
3.M*
77.972
.
1
M. 1,
1974
232
907
IS
4 123
1 394
640
100
1.456
301
786
1,329
6,390
1.204
2*4
27.490
605
177
64
3.960
.
»,zv>
394
-------
ri*«UjM mi In
i. in*
u a» tmu*t seam.
i, mi
Jim, I,
in*
Jafe I,
un
te. t,
1*74
Hit
'«•. t,
an
10
u
I
71
•30*
74
21*
34
tn
114
4«J
5
in
n
i, tit
3.12J
a
120
271
1U
131
M
343
U3
5M
3
410
34
M
1,114
in
jl 2*1
5,955
1.541
St
1.470
»34
4,0>7
401
J.HI
4,7>I
1.120
3,127
70]
1.047
200
i,oa
us
M3
1,175
3.M1
17.7(4
30t
10)
M
2,440
2.354
4*2
2.31*
r/ S.n?
1.411
50
l.tU
l;Ht
S/ 1,377
l.lli
1.00*
371
1,047
170
1,030
21
*I3
1,177
4,032
1.239
110
217
13
2,440
2.121
in
141
447
773
3,47*
471
4,171
1.373
I*
4.341
1;OM
i,in
403
4,017
4,7*1
1.171
1,10
101
1,107
s
1,401
m
1.2M
3,017
20,n3
137
2*7
100
204
2.440
2,434
40
in
711
1,03*
30
4.27*
1.0*4
4.3*3
37*
I/ 4,010
S/ 3.377
3,171
1.10*
43*
1,134
171
3
1.440
1.017
1,171
3.14*
1.213
ff n.477
331
2*7
ai
2,440
2.31*
4.24*
11
71?
MS
1.37*
U*
31
21
tl
2,41]
1,00*
1,417
1.J11
2,1*3
U
I.M4
14
311
7*
US
2,3*7
2,41*
Ul
n
41*
m
44*
i.m
111
1.4M
Z.03*
240
2,120
1
111
147
4,171
10*
1.311
3
247
303
rl 4,231
311
1*7
if 1,10*
107
S3
14
31
2,2*0
40
1,94*
1,303
1.3M
2,402
10
3,411
73
311
31
;40
401
1.1U
1,310
430
74*
413
401
1.10*
121
1,4(1
1.M7
217
2,22*
1
1,131
144
144
5,414
10*
1.137
30*
2
247
n*
7,113
7,M4
7t,U»
71 .Ml
77.114
41.773
£/ 42,272
395
-------
tlfflimm till tor p»ml«i» »t»*Ua» *» to* 0»lt»« iucu. liaurr L, 1977
rr.1.)
Jnury 1, 1>7*
sun
T«*l
Gutari
JM. 1,
1977
117
m
15*
56
107
^
421
|
92
1,042
7$
299
«,OI1
kat !!•••
Jm. 1,
1174
1A4
^
(4
52
M
^
r/
u
4f
1 lift
12
271
X/ 7 .SIS
Cl»<» »11 t
Jm. 1.
1*77
50
1 725
421
4 Mt
4 M2
3 111
3 Iff
791
2M
1 4fS
77
OT4
4,354
479
21 9f9
113
2 434
2.3U
71,514
XMUiMC
JM. I.
1974
*409
4 7*2
3 120
)'ll7
703
20g
1 029
132
MS
3 Ml
234
LOB
Z 4*0
2,35*
71, IS*
Tvul en
J-. 1.
1977
1 142
4 MX
3 1*7
3 999
IM
2*0
1 91*
as
94*
$,394
597
IIS
2 434
1,443
M.M9
•*• 1UM
JM. 1,
1974
4 344
HI
*/
145
991
5 097
101
2 4*0
2,U4
?*,977
m*
JM. 1.
1977
* fi
I*
310
297
423
55,4*4
«C limtt
JM. I.
1974
* 11
7$
£/ 543
224
£/'lO*
500
72
247
343
r/ 50,121
396
-------
K»>Ua« fill tor »•«•!*•• vif»llj>*t la Urn I*lu4 SCUM.
l. 1971 •** J—««rr 1. 194*
tamli)
Stan
KttMMky •••*••••••»*•»•••••*•»••**•••
mrylMrf ••* QUtttac *< C*!M*U ....
IttMlniBpi ••••••••••••*•**••*••*•••*
MMCMift ••••*•••••»•»•*•••••••••••••••
MfcVUltt ***•**••••*>•••*«••••••••••••*
NMff MVHliW •••••••••••••••••••••••••o*
ftocefc Canliw ...*......*....••......
Mwdi OfliMn ••••*••*«•••**••••••*•••*
GidMrt
JM. 1.
1*71
7
—
SO
543
47
—
.
—
2(2
42
643
133
33*
434
64
IS
4t
]
410
12
34
M
1,114
n
.
m
l.JOO
a
.
121
717
6,904
•( liM«
J«. 1.
1*M
1
^
54
647
101
f
^
^
us
ss
741
133
323
too
42
77
M
—
424
17
32
H
1,41*
112
.
^
2,744
41
.
IM
2«4
«,303
e*mit HI
JM. 1,
im
33
12*
444
442
2,314
604
_
f
4,310
1,4(1
50
3,413
313
3,»27
374
3,»44
2.3M
3,112
3, (04
371
l.OM
17*
1,030
22
N3
1,177
4,032
273
.
—
1,233
II.3M
310
2*7
>3
2,440
2.328
64,714
umkllM
J~- 1,
1*41
33
122
443
Ml
2,737
34*
^
—
9
3,441
1,1(0
61
3.3*2
343
2,M*
371
1,M3
3.1W
123
3.41*
443
1.033
1*0
1.032
42
1*3
1,330
4,040
330
^
—
337
17,944
300
2*7
in
S3
337
2.03V
94,603
T»Ul a
3tm. I.
1*71
40
12S
444
722
3,03*
673
^
—
4,772
1.323
30
4,274
71«
4 343
374
4,402
2.JM
3,171
3.804
634
1, 113
ITS
5
1.440
34
1,017
1,271
5,1*4
344
7r
_
1.235
H .8**
331
297
212
2,440
3.045
1
73,642
•4* ll»u
JIB. 1.
1*4*
34
122
443
717
3,444
(72
3,744
1.2U
(1
4,333
67*
3 3*4
371
3,013
3.19*
8*1
3,8**
340
1,12*
1*0
1,434
7*
M3
1.424
3.43*
442
.
w
337
20,70*
3d
2*7
10*
217
537
2.333
64,910
Fndu
J«u 1,
1971
2,13*
2
321
8*7
81*
107
53
14
32
2.240
(0
1,944
1,303
1,34*
2,402
10
2,411
n
311
32
7M
401
1,11*
1,310
430
(44
»7
744
415
401
1.10*
' 122
1,443
1.147
227
2,22*
3
1,131
144
144
5,434
101
1,337
30*
2
247
29«
39,904
M llou
J-. t.
194*
2 120
322
8*7
834
10*
54
1A
XI
1 433
227
t (*(
1,333
1,179
1,772
10
2 3*0
61
313
32
M*
44*
r/ 1,004
1.4U
3*2
3(1
N
944
330
430
1,230
174
1,23*
1,147
130
2,201
3
til
12*
144
4,31*
422
1,321
3*4
1
244
289
r/34,701
397
-------
APPENDIX C
SUMMARY OF LIQUID PIPELINE ACCIDENT REPORTS ON DOT FORM 1000-1
COMPILED BY THE OFFICE OF PIPELINE SAFETY FOR YEARS 1969-1976
i n, i**t
±fc
"••'"
In lirtitri
•""*"»«;
(1) IMlMM Ml Mtt«M »l«
(» I*clo44> MttMM* Ota Ft.
(J) IMlMM MtMMM «tU
•f m.OOO, MJ.OM, IM.OM M.OO*.
(«) mtrtu «M MctiMC ifltk IM.OM ,nn*n I»M«»
f4| lAcl^aa MB MBU«lt With I1S.OM »IM**CV 4
II) iMtMM MUMMHttt ,t4m *-M.T.< (M.«M, W..OW ^ 1100,00..
(I) UUMM M« MOMM OlCk >IM1H> MMM •< dLOM.
(I) IMlMM M> MtMM OtU UM •( If.ltO Mmtl.
(10) IMlMM M> «KUM< .lit UM .C •.«! MmU.
(11) lMta4M M> MtUMt otck UM •< 7.UI t«ml..
1. INI. »raM> OMMMl 11. IfM
en* Offln «f PtMilM S*(Kr
11, if»
398
-------
- It COMMPBm UVOtYta
CO
lO
IO
UUIMUUlIt
)PEHATIOMS ACCIDENTS ,'
:t«4. on v
"...oil,,, v
1C k
'ucl Oil
I1...I Fu.1 /
tat Fu.1 f
Uhyl.t*
tahydroii* AMunU*
?ond*n**t* /
..ro.ln.
Natural Caaolln*
Tran* Hln
TOTALS - BUllMC OtEIA-
TIOHS
TEST ACCIDENTS
Crud* Oil
|u«l Oil
fOTALS - DUIIHC TESTS
NO. or
ACCIDENTS
««
62
* "
1
1 •
1
40)
I
7
tor
TOTAL
»
1
1
i
i
t
I
1
100
M
14
100
LOSS
(IAUELS)
174.»40
100
" rBrtrrt DAHACE <»
CAMin
811,5
(,{
:»,i
I,
tk>
H
100.
-1
(4)
1
(*)
S
»
4
w
UM.1J7
M5
72)
1,401
OTHEM
144 7JO
r
(
i
1 C
(
i.
i I
(( »)
J(»)
1
-ff-
»5».621
ISO
TOTAL
1. 176. 240
1 •!!'
ii ,fir
" i.j*
i 9
2. 1
100.0
-0-
»1.7»5,7»0
17)
1 2.1M
X OF
TOTAL
66
(
2
0
1
-5-
100
17
100
DEA
CAUin
EMPLOYEES
1
2
)
0
RS
NON
2 '
2
0
INJUtlES
CAUIU
EMTLOTEES
f
1
0
NON
EMTLOtEES
1
1
0
(>) U.« than 1.0 parcant.
(1) Includ.. 5 .ccld.ot. whar* loaa waa J.OOO harrala or o»«r.
(2) Inclu4.a accUanta with lo.aaa at M12, 7641, 7227, an* 6750 harrala.
(1) IncluiUa ona acc|il«nt at (70* harrala.
(4) Inclu4aa on* accUant with 4*M|* at 1750,000.
(1) Includ.. on* accUant with 4a«|* at 190.000.
(«) Includ.* ana with 151,000 uul on* with IM.OOO •*•*!*.
(7) Includaa aecldanta wtth4»M|*a of 1225,000, 125,000 and »19,«5J.
(I) )^tlud*a accldrata with daugaa of IJ7.000 and (20,000.
(9) loclud.. ona with d».|. of $65.000.
PtfAHTMENT Or TMNSrOJITATIQN
NptltiM >ccU»t> turn DOT fan 7000-1
January 1, 1969. through 0*»«bir It, !>»»
CoBpllnl hy th« Offle* of rtp«lU< tiltiy
Fcbruir; U, K70
-------
tumour - K oomooiw INVOLVED
O
O
GOHMOOm
QPBtuYTIGMS KCCIOEHTfl
Crude Oil
3*
S21.S4*
t or
TOTAL
7
4
n
f
i
0
100. 0
PROPERTV DA IAGE (f)
CARRIE*
111 71»
{•,212
44.402
{,4tt
j, ?•'
I,»!
" " if
\.iH
J.4oi
20
4»J,OU
OTHERS
7* »07
7,
321 ,<
•i*
2 .
1,
20
J
H
T
n'i
01
46
Ji
n
4K.212
TOTAL
411, 7S«
;
2^ *i
JiJ
«
V
ii
,50'
i
1
2j
t!
ii
Si
4,SS(
20
(91.22I
1 OP
TOTAL
4C.O
M
14.
",
«,
0,
,
too.o
PEA
CARRIER
1
1
HS
NON
EMPLOYEES
1
2
'
IHJ
CARRIER
EMPLOYEES
1
1
' 2
4
BUS
BoH
EMPLOYEE!
5
12
17
at Tc»t>«patt»tloo
pat
frc
Flpclln* •ccden* frctt DOT Pom 7000-1
January 1. If 70, through tecaabar 11, 1*70
CwplUd by U»« otflc* of »l(«Xln» Sktaty
April 1, 1171
-------
nUMI MXIMUt MMUM - Jwuiy I. UN, TWOKV MttMU II, ItN
P«B«Ct»«llt at
IIr*U** MclWb Koi terror. ]••*-!
i, HJ«, i*««i9» BtcMlxc II. ttn
GOWlt»4 fc» MM OlflM Of
l«rll 1.
-------
uquu rim.ii> ICCIMIK MMUII . JM»MT I. ••'». TCMUOI MCINM* 11, IIM
O
ro
P«c«rt»«nt
Mpiil.. iccu ^'t&T-roo. 7000-}
0«u.ty I. 1*1*. thromh McwbM II. U10
. .
k« UM olllc* •! HP»UA« >•(•(>
I. itll
-------
- n COKMMT* IMVM,VIP
o
co
COMMODITY
OPERATIONS ACCIMNTS
Crude Oil
UlolilM
£&
1l*l Oil
klcicl Fuel
1*1 fuel
Uthydroui AIUKKII*
:aro«tn«
utural Gaioltn*
Propyi en*
rOTAlS - DURING
OPERATIONS
HO. or
ACCIOCNTS
21*
• I
*
i
147
toru.
(2.2
I
!
i
.00.0
i.3»u,
111 KO
t.
. 1
1 ]
, 1
J 1
!
, 1
r
1
II
S21.04*
t 01
TOtl
71
!
a
&
.0
LOO.O
tKfinv DI *« in
CMUUM
ni,
(1
11
j
!
(
71»
IV
41
i Si
' 1
! )
1
i
7!
44
2
I
44J.OU
OTHERS
7».»07
^'
V '
J '
. '
, '
1
1
L '
i
ii
i
A
41*. 212
TOTM.
411.7SC
7t,152
ill f4S
1' |^
{ ,! i
'
-------
1 riKllKC MXIMtt IWMUI - MMIUH I, H7I. IMOUCII UCtHltl II. l»l
-p.
O
CMISC Of ACC1UCNT
op«r»(ia*t 4ccldffi(l
:..rfa»tun - C«i«^tfl
E•*•{ lueturlRA Llfif
LffrcHYt fttf tin — ~r~
lacorctct OBI rat la* »y
Cffflff f«r«xiiut
HI •(•) |AA«OI||
Rupmrt* «r U«fc|»|
LUiktf
|tu«(wrt4 *r l*tMf| !*•!
InknaiA
tuftwr*4. l*4fclM|, Of
MalliMCfloa at V»lv«
iiipttirt of Frtvloutly
DM>M*<| tlo«
rulfiMctloft of Control or
K*tt«f t*MtM*p|
:«ld u.*th*r
>*f«c«i«* cittk u«f4
[»r**4f »ifl.».4 or Icoht*
PIMW pfcfcui r»iiwr*
V*ml»Ilw
l»Kl'!«l«B
FUTAL
or
ucmeni
Itl
f)
11
Jl
11
i
t
1
4
1
>
1
1
Ml
or
OtAl
)>•)
II •'
n.i
1,1
LI
} 0
1-0
l-i
!•}
i.i
IP
|.i
1,9
100.0
^!
C1IIIII
«
1)
1
a
4
>
«
0
i)
0
I
0
»
a
Hrunui
«
<
• '
a
t
•
I
«
•
V
i
0
*
1
CURl4A
uirutlli
I
*
«
^
|
0
«
1
t
1
1
*
«
1
sm2i£is
•
t
•
i
t
<
•
•
•
•
i
(
«
CAIIICI
11.10
Sl.«lf(»
1.141
u.4>e<»
i.iif
u
KO
».«?!
).U«
I.HM
Ma
If
l.»t
»?•
»
w
!0».«»l*>
HO. »4
OWf
M.IM
ttl!lOj"»
»».)« .
4. DM
l.tlf
W
10
t'tM,ti '
l,?°9
».ooo<"
Iff
US
fOO
IH
t.OH
J.OOO
0
»4.i)l
I«T«l
M.H?
Ml. Ill
ll.^t
11. WO
10 IM
in
IM
ll.D?
I.UO
».ioa
»w
lit
1.4U
I7»
!.«»
».0)0
IOJ.JOO
u>.»»
— — TJ~M — i
comMiin
(tAPICLI)
n.fii
M.HI
H.m"?
!•.«•"«
IB.JU«">
t Hi
I.MI
!!.»>"'•
'•?•»
».«•<»>
ll.lll«'*>
f.lM
II*
JJO
HJ
M«
ll«
VOW
14J.OJ?
>i
1»JO-
l»5l
II
?>•
»
.
1
1
I
_
1
,
.
.
i
1
42
IMO- lit:- • ::::
in* :;•-. i.,:i
_ 1
il : - '
•
i - :
i -. . i
1 :
I
t
1
I
.
. j
1
II : . :
(I) luluibi 4CC|4»» uf 1)0,000 ml 111.000.
(1) l«clu«i> MU Mclfau «l IM.OOa.
(1) l«lti<«i m. Mcll»t el IISO.OOO.
(4) l«cluJi< l HO.000 tut til.too.
(i) lucludti ucl44 110.000.
(I) Uclufei DM ICE I Jot al US,000.
(t)
(I)
(10)
(II)
Ill)
(II)
(U)
ulck lo» at t.«>< k«rc«l«.
« lc<> •! I.SM k.cr.l..
U»«l >l 1.100 «J 1.1)1 k.n.l..
lk !••• >< I.WO k.,c.l4.
luliitet >uti«i> »Uk !»«••• •! II, MO «M i.ito kuitli
IxlirfM !••• •< I.MO ktn.li.
Uclu4ii «M «cU»l "Ilk Un nl i,IM k.rnli.
«u »Mut>
lKli>4« OM «cl«
U«»«n»«af »f T|yH|X>f|f^afl
tlt«ll*« »cclA«Bl» CtM lor yon 1000-t
j«iuuir i. Itii. u|k mcMMi it. tin
Co^lUd ky lk> olflci of Mfdliu l.l.iy
-------
SUMMARY - BY COMMODITY INVOLVED
COMMODITY
I
'OPERATIONS ACCIDENTS
|Crudu Oil
Gasoline
iL.P.fi.
•Fuel Oil
lleael Fuel
Conden«ate
Jet Fuel
Natural Haioline
Anhydrous Ammonia
Keroslne
Alkylate
TOTALS - DURING
OPERATIONS
NO. Of
ACCIDENTS
172
SI
3*
21
5
5
4
4
3
2
2
108
\ or
TOTAL
55.9
16.6
12.7
«•
1.
1.
1.
1.
1.
.
.
100.0
LOSS
(BARRELS)
11S,7«0U>
42,001
39.1.7'"
11,724
4,953
3,<5«
2.2U
• ,741
».»10»>
700
I,5i5
245,057
» OF
TOTAL
47.2
17.
1C.
5.
2.
1.
1.
4.
.1
.7
100.0
PROPER1
CARRIER
44.1«l">
I71.»4?««»
70,U4<«»
11,072
540
10,510
4,liS
10S.024<»1>
0
400
72
$420,174
» DAMAGE 1
OTHERS
73.770«S>
M.iJjUl
7*,000<»>
».»OS
250
U.lSflOO
1,000
2J
2S.000112
1,000
200
$264,935
1
TOTAL
117, »!•
21S,S«4
14t,li4
20,t77
7*0
2«,730
S.liS
105,049
25,000
1,400
272
>6I5,309
« OF
TOTAL
17.2
J4.4
21.1
1.1
.1
1.1
r!
15.1
1.7
.2
.0
100.0
DEATHS
CARRIER
EMPLOYEES
0
0
0
0
0
0
0
0
0
0
0
0
NON
HPLOYEES
0
0
0
0
0
1
0
0
0
0
0
1
JNJUR
CARRIER
EMPLOYEES
2
. 0
0
0
0
0
0
0
0
0
0
2
es
NON
EMPLOYEES
0
4
0
0
0
1
0
0
1
0
0
c
(1) Include* accident I with lo**e* of 12,100, «,<*0, 7,3*0, 5,300, and 5,111 barrel*.
• (2) Include* accident* with loiee* of «,47», 5,500, and 5,100 barrel*.
(3) Include* one accident of 6,BOO barrel*.
(4) Include*,one accident with property damage of $11,000.
(5) Include* one accident with property damage of $10,000.
(6) Include* one accident with property damage of $150,000.
(7| Include* one accident with property daatag* of $52,100.
(I) Include* accident* with property damage* of $10,000 and $50,000.
(9) Include* one accident with property damage of $t0,000.
(10) Include* one accident with property damage of $15,000.
(11) Include* on* accident with property damage of $105,024.
(12) One accident of $25,000.
Department ot Tcansportation
Pipeline accident* Iroo DOT Form 7000-1
January 1, 1871, through December 11, 1*71
Compiled by the Office of Pipeline Safety
-------
O
CT>
'-"" •* '"•""•"
-i^l^i^jl
...lf«l»H • I.M.IMI
i""1.;';.^!"""
tts£F"*-
..„,... .1 n..>~.i>
Ik...... ii.in.4 .. ..-»..
-!£?:!:^ _.
........ ,,.l. B...II..1
V !:'"-• t!« -
"..«.....;"" • ••'
Ifcl.vll*! MtU
"«•'-
' """I'll,
Ill Ittfllllfc-l 4C |4>*l ||
m iMiu4.t .c I«.M* i
1*1 tncluA.1 «c ltf.«i |l
III lMlo*,l *t I4>M 1
t*t |M:|W|>I «tilA.M |
1*1 1*1*4.. *ttt«.«l. |
IIWI |wl««.« M.CI4.M »|
tilt lib|«4~i •ttlfeM »l
lilt l^l*4>« MGl*IM I
u*
to_
»
11
*
»
1 * 1
k
1
i
..._ * -
•- *•"•
1
1
1
1
I
1
1
„
i^i. ....
l.wn.
!»!ii.. ~.
:£
Of
IHlfcl
**
V.I
i.i
i.t
*.*
...
i.*
l.fe
i.*
I.i
1.*
I.U
-•*
.;•
.»
.4
.*
IUU.U
W>. M44 114.
kM.«M.
u
CMME*
.
«
•
.
H
•
•
•
•
•
•
a
.
•
d
•
•
MMt
,
*
•
• •• " '
*
*
•
«
•
— - —
•
*
*
.
•
•
II
'
IM1
CtfMKft
tMlOIKl
1
•
*
.
•
•
•
— ; —
•
•
•
. ...
•
H
•
*
HltS
MM
uMimt*
1 —
t
»
V
.
•
•
•
— r~
•
•
•
.
•
*
•
11
rwr
UMUB
,,-/»
n.in
».H»
»»*
w.W*'
H.tM0'
»»*
I.M*
•
•M
»!
*.*»•
!.•»
*W
II. >ll. XI
111
111
111
»*l
*•» |wl-;| ||,
.,«.
,,...."'
1,*»1
-»"*
*t*
I.N*
t.VM
M*
I.I**
I.IM
H*
I.IM
IM
Ml
I.OW
11*
*
•
«..*»
£l± «:!J
!"!!Jn n«!t
£!±:«:iJ
|Ul4i
..,.,
KM
1.V.1.W*
MI.1M
I.M*
It.tlJ
IH.IBt
i.m
I.IM
l.ll*
— rsr
a. it*
I.IM
"-
ail.Bd*
»,*«•
l.*H
fcM
..»>..»
•i •! fftw.e
M •! »»I,W>
MI wtifc IM
M» «lifc IM
IMS Of
UMMMritl
IMMCltl
-W«
— !r"!"^
M.m11"
II. At*
I.IM
...I.,'"1'
'•*"»
HI
I.IM
!.*»>
I, III
IM
I.Mft
Jl*
41*
1,1*8
MU
«..»
•M «f i.*W. •>*
M* *l ».?»» >M
*.* •! II. IM. t.
••• «f II. M*. t.
li-
ft
— r~
a
1
•
•
»
-_
1
'
1
•
it
M. 1,111
.Aw* bti
•a.-i*
Lur*cc
?f
1
1
t
1
•
i
1
—r
i
•
.
•
t.
• !•.
"IX.
l|uii-i
i!»
„
i
t
B
1 - 1
i
•
•
*
•
i
i
•
••»
;:;;
^ mn (i
1ST
„
—r
i
•
.
i
•
i
i
„„•.
14
MMl *,««•
y I^HI
ItM
,
li
1
,
1
•
1
•
1
* !M«««|
latHN,]
IH«
1
1
~l~
1
1
1
1
1
•
'
till
.
_
i
1
a
1 — I — 1
1
i
~7~
•
!
IM
M
~7~
-7-
i
•
•
i
•
•
•
•
-------
- B» COMMOOin INvn..».
OPERATIONS ACCIDENTS
Crude Oil
Gasoline
fuel Oil
I. P.O.
Diesel fuel
Jet Fuel
Condensate
Kerosene
Natural Gasoline
Alkylale
TOTALS - DURING
OPERATIONS
NO. OF
ACCIDENTS
IBO
CO
25
22
a
5
4
)
)
1
309
I OF
T01AL
58.3
19.4
a.i
7.1
2.6
1.6
1.3
1.0
0.3
0.3
100.0
LOSS
(BARRELS)
233 .690*"
43.6S4<2)
17.924
30.40»
5 .814
5.869
io.2oo(s>
1.000
1.212
360.654
f OF
TOTAL
64.8
12.1
5.0
8.S
1.0
1.6
1.6
2.8
0.3
0.3
100.0
PROPERTY DAMME It)
CARRIE*
364.268'"
I.366.6I51"
38.404""
116,840<»)
1.715
2.010
MO
4I.300
55
100
$1.931. 847
onin
396.530""
40..001"1
9,750
I6.7IO<»>
2,825
1.850
2,400
800
0
0
TOlAl
760.798
1,406.715
48.154
113.550
4.540
3.860
2,940
42.100
55
100
(470.965 (2.402.812
PHMtfiWII!^LV"*VPPVmH
kof
TOTAL
31.7
58.4
2.0
6.6
0.2
0.2
0.1
1.8
0
0
100.0
BOB
oe THS
CARRIER
EMPLOYEES
0
0
0
1
0
0
0
0
0
0
1
HBBBE
NON
EMPLOYEES
2
0
0
4
0
0
0
0
0
1
7
kB^B^B^Rei!
INJ
CARRIER
EMPLOYEES
»
5
0
0
0
0
0
0
0
0
6
DRIES
NON
EMPLOYEES
6
4
0
3
0
0
0
0
0
0
13
•Boat
Includes accidents with losses of 19.965, 17,000. 15.100. 11.598. 10.260. 8.000. 8.000. 8.000. 8.000. 7.211, 6.800. 6.797.6.000. and 5.200 barrels.
Includes one accident of 6.915 barrels.
Includes accidents with lasses of 5.474. and 5.000 barrels.
Includes one accident of 5.010 barrels.
Includes one accident of 6,875 barrels.
Includes accidents with property damages of 1118.000, (100,000. (37.473, (33,000. (10.000, and (10,000.
Includes accidents tilth property danagei of it.250.000. and (100.000.
Includes accidents with property daugef of il2.500, and (12.500.
Includes accidents with property dwoes of 160.000. (20.000. (20.000. and (14.000.
Includes one accident with property damage of (40.000.
Includes accidents with property damages of (200.000. (67.900. (40,000, (14,000, and (13,000.
Includes one accident with property damage of (30,000.
Includes one accident with property dauge of (10,000.
-------
IIWII MMIIM MCIU.I
• WM..I I, lilt. IMMU licimu ||, |,n
CD
00
CAUll Of UtlOfll
rtMtIO* UflOtfllS
•rr«ila» • filar**!
•••••••I OMBlwrlBj tl«a
• rrltr Pirll.ill
:.rr.tl.* . l.lim.l
'l>tlll.....l
»>Hl ll,»"" *"' '
• M.r. 1. ll«ir Crinllill'H
l.llif l«.lt*t.l
• II ...(Mr
>»(.cii«i link viu
.IMM"."'.'""""'
i.llM.tll.. .f !•!*•
... l.ll...
upl*"!' *f l.llllf l.ll
I.. C...IIM llll.il
.1..,!,. M,.
IIM.IM
out
M. W
Ulllllll
III-
M
II *
14
14 V
II
•
1 '
1
1
1
>
1
1
»
1
4
1
1
t
1
I
III
1 W
nm
H.I
II. 1
1.1
I.I
I.I
I.I
I.I
II
I.I
14
I.I
I.I
I.I
I.I
I.I
I.I
.1
.1
.1
.t
.1
IM 1
......IM.
Cllllfl
Miltlfl
•
•
•
1
•
I
•
- ,-
I
1
I
1
1
1
1
I
I
•
I
•
I
I
1
1
urllllll
•
•
I
1
•
I
1
I
•
1
1
1
•'
.Ill
1
•
I
•
I
•
I
*
'
MNtllfl
•
•
•
1
•
1
'
•
I
'
1
I
•
•
•
t
I
I
•
•
1
IT
IMtlllll
•
•
1
1
1
1
'
•
•
'
' > '
1
•
•
1
•
I
•
•
•
1
IMIIfl
1 1.114
II .Ml
II. IM
41. Ill'"
l.lll
441. IM!"
• MI.WJ
1,111
III
MM
1,111
I.IM
1,111
»«.IM<>>
II.IH""
II.Mlf"^
1
IM
II.IM"!
Ill
III.M.I'I
II .III. Ml
"MSH"-"]
1 14. m""
...41,"'*
ii.ni
I.IM
IM
I.IM
I.IM
1
I.IM
I.MI
1
I.MI
Ml
II.IM<">
II.Ml""
IM
I.MI
•
I
•
II.I4I<"'
t
im.in
i«u
1 ii.ni
III. 114
II.Ml
ll.lll
I.IM
III.IM
II, IM
1
1,111
i.m
I.MI
I.IM
I.MI
ll.lll
III.IM
11,111
II. IM
1
IM
II. IM
11,111
III.IM
tl.M4.lll
IMS M
Ti.iTil'H
11,111
IM.MI""
41.114"'
11.411
4.111
I.IM
l§,M4"ll
1...I1"1
4,114
1. „.''«>
I.MI
411
1,111
III
lit
II. M?'1
111,111
ijniq-uii^
11
i
i
i : i
•
i
i
•
' •
i
i
~~|
W
H
•
i
i
i
t
i
i
i
t :. • | •
i iii
t ' i '"*
• i
• • t
MKK4
mi
i
ii
i
4
1
1
1
I
,
I
• t i
r-H
,
I
fir
i
ii
• i
,
,
i
_i_
,
,
•
wi
Illl
1
1
1
1
1
1
IF
•
i
•
i
0
1 I
• I
t 1
t
1
_. — , — . —
• i
1
0
__
0
iii
i i .: i ' • j • 1 i • "
* i- * \
~L
it
.
t H
§
, t i 1 1
h ..-
1
0
ii t
I
1
1
0
,, ;„
4t
"
it
It
RUHIH
0
1
0
1
0
0
1
1
1
•
0
§
0
0
0
I
0
•
1
All 4l*tkl («.*«4 fev Ml .Id...!.
Utl.411 KCI4..I •! III.IM.
Uil.4.1 .(CI4..II •* ItM.MIi Illl
Uil.4.1 I1II4..I .1 III.IM.
l.i I.J.I >.il»l.l leilll.l .1 111.Ml
I.ll.lll liclll.l II III.Ml
Ucl.MI icclll.1 .1 ill.MI.
.el.lll .C.U..I .1 llM.IM.
Ucl*«t« 4C
Uti»*u »
UlU«tl «<
t.llll «•
Utili •! III. til •*« IIO.
i«»u «r IM.MO <«4 iio.
M«l •* III.OH.
i .1 tit.uo
10
II
10
II
11
II
14
>S
1*
•«!«*•* Ml ttt
Htlmltt Ml »Ct
ftCl«4tl Ml ICC
•ClH^II Ml ICC
•Clu'tt IM ICC
Kl«Mt i\«Uti>
Bduilai MI tec
•clkdii MI acc
•cl*«*l IMF ra
t«a *l l*ll|ya 1
turn
<
• i.
••* IM
-------
SUHMAM . »t COHHOBH;
1 rff „
COHHOOIII
•JrCDAIION ACCIDCNU
Cruda Oil
GitollB*
I P.O.
funl Oil
Jcl fuel
Uilid fuel
Coiiden&«ttt
/trbieAe
Aiiliyilroui AMMOIIU
Tout tiumiiG
OPCBATKWS
HO. Of
ACCIOENtS
155
SI
. 25
20
9
t
5
1
)
213
S Of
TOIAt
St.;
IS.?
S.2
).l
3.1
2.1
I.B
.4
.4
100.0
IUSS
(BA«mS)
20»,173("
40.2071^
SI.SM")
51.505<«>
IJ.12l'**
S.»«
6.404
229
3.210
379.36S
1 Of
TOTAL
64.1
10. C
13.6
U.t
3.S
1.5
I.S
.1
.1
100.0
f«
(AAlUk
i 2»«,>
315. 055*"
25>.3«?l«l
14». !»><**
9.938
too
48
0
80
11. Oil. 909
UftdlJ OAMAGf
AIHCR 1
»m.«2«"«>
It .112
IS.«0<'M
»>.»I*'UI
2*.I001»3)
1.950
600
0
0
U86.930
(\\
lOIAl
1 III. OW
114. 23»
2)3,24)
24). 059
Jt.038
Z.550
64«
0
w
J1.304.1I3S
S OF
roiAi
31.6
25. t
21. »
IB. 9
2.t
.2
.0
.0
.0
100.0
Lift
CA««U«
LKFLOIEES
0
0
1
0
0
U
0
0
0
'
HIS
NOk
EHflOntS
0
0
»(T4F
0
0
U
0
0
0
6
KJMI
ENPLOKES
0
0
I
0
0
0
«
0
0
1
us
HO*
EMPLOYEE:
a
0
2
0
0
0
i
a
2
i
52 101*1 OUK KG
*£> OPCBATIO«S
2)3
100.0
379.36S
100.0
tl. 018. 90S
1 Includes incident* xltk losses al 10.185; 18,700; 11.000; 13.000;
i
i
4
5
(
(7
.15
10
U
12
13
14
Include i out Accident at 0.640 btrrels.
Includes iccla
Include! ant *
tnls wltk It
:ci
U85.930 |l. 104. Hit 100.0 1 6
10.500; 6.000; 5.»I2; 5.63). iud iwu it 5,00b IxrreU
jrrels.
Include ant «ccidc»t af 6,04» birrtls
Includes accidents xltk property dcctd«*t nit
Includes ccld
enti uitk »
deAjgt o
dakege a
di«*ge o
h prwtttrty 4«««ft i
roperty
209 .000
2 SO .000
200.000
; 110.400; mil
; 120.000; »I5.
.nd 150.000.
•«d J42.500.
ilO.IOO.
000; <>d $13.000.
35.J2S; IU.443; «nd til. 220.
C ill .000.
dinefte ft 160.000:
Includes a< iccldeat Hllk properly timtft al 120.0
12K.OOO; <«d |
00.
10.000.
-------
***tr~.i iwtrtwi,* ilM
CM*M|M • *•«•«*•!
CftrilM tMMMil*
•^iS/j/'****"11'
tilW^iST"* "
I^.U^c. M U-.ll-
Mfactlv* «|«tb (*•!•
V~4.1I..
to.ll.MiiM •! «*!..
fflu»4« *lt|f**4 M feet.*
014 •**<*>•*
• *I(M« Hi flM|«M4«l|
tl«M.I^
to M*fc
•*—»»«—"•
VbtM.
II
II
1 at
i*
It
4
a
i
4
1
I
I
>
a
i
i
n
m
U.ft
14.1
i.i
i.i
1.1
1.4
1.4
1.4
1.4
I.I
».l
.1
.,
.1
..
».*
'"»
-car.-***
urutlit
•
l
•
__! —
,
•
•
t
•
•
4
,111
1
t
1
4
u „_
ciwioni*
•
•
•
!
,
•
i
•
•
«
4
,'"
4
4
,"'
4
-Ha**
•
,i»
i
• -
.
,
•
•
•
4
i
.
•
.
•
4
Ki~
4
i
,«•
4
1
.
I
4
•
•
•
•
,
t
4
•
I
UMIH
1 41. §14*''
11.141
• >,.»>>«
I.U4
lt.444*1*
444.4«.»l
I.Mt
I.U»
1.411
4I.444*1*
U4
I.«M
.
4.M4
»4«
41.. >.!»•'
I1II,IU
,.,»
*'***
4M
IU
1.144
IM
•
1.144
•
Mt
4
.
•«.»•»«">
KIU
1 Xl.444
11.444
44. 4M
1.144
141. 4M
1.144
1.414
1.111
41. 4M
1.144
!.««
•M
ft. 4*4
...
LOtfi Uff
IMMHJI
141, III HH
II. III
"1 mi
i.iii
a.Di
i.tit
i.iii
i>t
*.4l|tlll
in
iJi
"^
141
1.1 tl
«ro^-
IIJ*
i
i
4
1
4
4
*
t
' 4
HIS-
111*
*
It
(
«
•
•
•
•
.
^
:•>.
»
»
1 — i — 1
t
1
4
•
•
1
4
ItSS1.
1141
" 1
• !
t
•
•
i
•
i
—
M.U'.'-'UISI
lt:l .HI
11 } 14
' i '
l
l ; 4
: i
> « l
1 ' 4
• ••
I t
: . .
T|»
1!"
4
4
'
1
4
4
1
*
4
.
- -v« -
MvMtll
•
1 '
'
t
1
4
•
4
•
4
•
•
1
1
<**• MC.4«MI -I
li.j« <.\4 t.'i.n
Oil IM
Illl IM
Illl 1(14
Illl Itu
Illl I*.
I 111 I*
Illl IT.
Illl tr,
Illl »«
r.i« cl IUC.CM* l)«4.4*4i IIM.Mt. IM.Mfi »4 I1*.4I|.
>•• •! llt.CMi llt,4M| ll«.M4i fU,44li *M 119.444.
•I *4 lll'ue'.
•«• vl*C !«*•«• •« *4.444| i.lMt »J I.It: l-.«..U
•*• with Uit*« at l.rMi >.»Mi I.IMi «*J I.III »•!(•:•.
«i* C;.* !»••» trl 3ft*l>4* M>« »-•»! *M«I*U.
-------
»BHM»M • »¥ COMHOOm HIVO1.VUI
COHHOOITI
OKUTIOU ACCIMNTI
CruiU Oil
L.r.s.
O..OIU.
ru«i on
itt nui
OU»1 fu.l
tahrdrou. »— onU
Co".......
TOTU. MWING
ortKATioM
HO. or
ACCIOiNT*
144
"
11
14
1
•
1
1
,S,
i or
TOTU
44.1
11.1
11. 1
1.1
1.1
1.1
.1
• •«
111.*
III Include iccldin wltk lo><>» of
111 Includo >ccld«n xltk !»•••• of
111 Includd ant >cc »t al 1,411 k»i
141 IneludM ccciton of I20*.tl*i 11
111 Includti iccid«n o| till, III! ||
1(1 Inclwltt >ccid«r. al 441,111.
Ill Include iccid.n al 111. Ill «n4
III I>ctud« «ccld«n al l!M,M*i It
111 Includai i.-cid.n ol 111. lit ccla«i>t.
Illl Ml 4*«kht CAu««J fcy two «ccld*nt«
Illl raur la|uri*fl c^ustJ by ena jcctd*
UM1
IUMIUI
I4l.ll.(»>
l4.Hll»
14,111
11,111
ll,!ll<»
1,11*
>.14I
1.141
It). 14) .
t or
tatu.
11. i
n.i
t.l
4.1
1.1
.4
1.1
.4
111.*
CMU1C*
1 114,111^^
II. til '*!
4,0 JO
t.tlt
4.11*
100
II
11,111,11)
stmytjaa
1 1*, mil)
ill.lOlWt
1,11*
H.lllWI
ll.JOOlHI
III
1,1*1
no
liii.ii)
^-^T-
1 4U.444
1.11*. ill
11,111
it, 111
11, lit
4,111
1,1*1
111
ll.Ctl.41!
I or
TOTU.
11.4
II. S
1.4
1.1
1.1
.1
.1
.1
III.*
DEI
CUMM
nwu»i«
4IUI
I
*
t
I
I
*
•
4
«.5m.
(llll
•
1
*
t
0
1
1
1
amf.nu
t
,-.»:
:
;
:
.
:
:
4
HON
twiomc
0
i
i
i
i
0
JII4I
i
i
40,llt| l.llli l.iMi l.llli l.tlll 4.J»»| ••« I. III. txci«U.
11,111) 4.40I| Mid ,1,111 b*ff«l4.
oil.
11,11*1 III. Illl 111. Illl 111. 11*1 «~» 111.***.
41*. *•«.
>,*!*< lll.MIl «rd 114,000.
III. III.
-------
iimiia nm mi Kcitta tumor • J«IMU i. i««. inou* «.a..t. 11. nn
ro
liAtItt OF ALillltMl
Oft ft* 1 ION KCItiNfi
tqit*JM«M lu^lnd*! *•*"•(!>
lH«lfM"M f»U»r*(it
l*.»ri«ti OfMMlM kir
CtflUt ntlMMiUlUk
CitMMlM • Ul«(Ml
»*f«cii** HM S*M(|»
»»l«cil»« Cltlk *•«
r»ii»(* •( r***i*utir
•»•!•< fi»«(«t(t)
tt»4»IU»
fciluf* •! ff •*!*«» Mil*
••Ml*
U«kl«IMll4)
NI*callMMM«(lti
IMU
NO. Of
ACCtMNU
71
»
<«tl
U(l) .
IMU '
I
I
1
1
Jf
Ul
fkBCtllT
or
101 At.
ti. »
i*.*
4.4
I.I
l.»
t.t
l.»
1.*
k.i
• i
U.I
!••.•
«*
CAKHliK
JiMTiUlltf
•
>
1
•
*
•
•
•
1
1
1
H»
*Mfi»nts
•
•
. • /
»
•
«
t
•
t
•
t
!««
wnoiiii
i
i
•
•
•
•
•
•
•
t
t
!« .
wruikU
i j .,
•
•
•
•
•
•
i
u
| iTf.llft
U4.IU
!.«•
4.IM
It.IAt
>.WI
»l.l»«
1M.U»
M.llt.lH
r^sSn
IU4.III
• .••1
I.IH
UI.M*
•
I.IM
t*M.M>
I "'mu
1 441. til
141 .Ml
ail.kM
JM.ftM
»!.!»
ll.ltl.MI
lOtl M
COMHOttlff
(MMILII
M.I14
».«ll
tl.Ut
II. J»*
1.1)4
111.41k
•MM
urow
111!
>
•
4
1
1
1
!•»•
IIII
1
•
4
It
* Of MCIWMTI 1
Ml**
1114
I
I
1
11
::;:
u
T~
4
14
IT !**» Of IMfTAl
lt*4*
IIII
14
1
1
1
11
ia:
n
,14
1
1
'
II
LAI JIM
fsr
i
t
i
»
i
4
14
Mil
u. ratlin
i
4
1
4
4
•
14
(II•
I
<
1*1 I
pr*»*'i|T *•*•!».
lilt to* •tcl4«*t--M I
t !••• •! D.II4 fc»»r»U (
-------
SUMMAIIt • i» COMMOPm INVOLVED
-p.
I—«
GJ
COtOUDlTI
CluJc Olid)
lit
41
it
2*
>
»
4
i
2
1
2Si
fwcBNT
Of
TOTAL
S2.I
U.I
11.4
U.I
1.1
1.1
l.t
1.2
D.I
».4
10*. o
LOSS
IN
IAIULS
105,171
10.041
24.11*
101.221
4.241
2I.41S
2S.010
1,511
IS*
110
11*. 421
MKCiMT
OP
TOTAL
11.14
«.;
J.H
12.14
1.11
».M
7.14
(.41
i.17
«.«!
!»•.«
rtoruTir IMMCI (t)
CAUUI
Il,2tl.t21
771.221
11.411
2H.I70
4t2
• It
*,»1»
10
47S
•
12. 112. lit
OTHM
1147.141
114.2*1
14. 80S
4I*.«*»
S.*M
»4.2U
I. IS*
l.eo*
u*
la*
t*14.t*I
TOTAL
tl.441.;«4
•«f.(«4
>2,4»1
»»l.«7*
i.4»i
•i.Ut
t.*tt
l.ll*
57 i
It*
11,117. til
NMINT
OF
TOTAL
41.1
21.1
l.t
tl.i
*.2
l.»t
(.21
t.*l
«.*2
«.tl
!»«.«
MATHS
CAUIH
tMPLOIEES
1
«
*
i
*
«
t
*
•
(
1
NON
tMcioms
• g
0
0
4
a
t
t
«
0
t
4
INJURIES
CA»BU«
tWLOTEBS
«
«
•
1
I
•
*
II
0
0
1
NON
KMTLOUES
1
1
*
10
0
0
0
0
0
•
12
(I) ccld«ti--|l«o.too properly du>|« •ml am c.rrl.r ••»>»„• d.tik; Ilil.tM Pro|i«r>r
«•'!•; 1291.01* praftrlr J>.||. IAJ II.lit klrrtli •( cuuodlty Uiti 1S,*M k»»l> of
(2)
I2SI.IM
u|<; «4 IliS, 10* pro
clJinii- litl.tll proiu
<•••>*•.
4.122 k>rr«li of couwjlly Ion lid f7>S.*tt
(1) Accld»ll--|S*,*oa Biopirif d»>|>, four «OM«olor<*
dllthl, M4 21,SO* kirrtll «f COMOdlty Ion: 11)1.ISO
property du>|« •••! o«« noauplaxM Injury: I12S.OSS
vratircy d>«|« >»d •!«• ooooployM Ujutlis; «nd
111*.MO propiny d»>|i, tun cirrl«r InjurUs.
-------
uauio urtiiHl Acciimrr MMMAIK - JJUBULM I. it^t. TUKOUCH ntcman 11.
CAII9K ur fcCCIDCNT
CrEMTIOK kCCIDUHI
lwt ky
Coiroalon - lataraal
ralluta of rravloucly
Daatagad fl|>a
nalfunctlon of Control or
Cald Hft«ih«(
tutor VMblcl*
HO. or
ACCIDENT*
U
41
a*
14
it
I
4
i"
4
a
i
i
"i"
s
i
ii
III All daatfaa cauaad ky ona accldaoc.
Ill All loiurUi ciuaad ky oat tccldaat.
• or
TOTAL
11. •
ii.t
t.t
4.1
l.t
1.1
*.»
1.1
i .7
~i'.i'
i~cT
a~«
3.4
" •".»"
J.<
4.S
».l
Itt.t
MAIM
CAMIE*
KWUnn*
•
"•
•
*
•
•
•
•
*
i"
*
•
•
• "
•
•
a
t
HUN
UWUTUI
•
"• "" "
t
,111
•
0
•
•
....
•
•
•
•
~» '
•
•
•
1
INJIM
CAMIMI
IWUHtli
1
•
1
•
,111
t
I
•
1
t
•
t
1
(
(
1
- - .
It*
MM
uirunu
«
•
•
• "
•
•
1
t
t
•
*
1
*
•
•
•
t
-. .. .
nonm umoc 1
CAUIU
1 ll.iat*"
ir.ti*
ii. an
m. 14*1*1
4I,U1«»>
*
""l.ttl
11. lit. Ill
. .
OTHU
111*. lit*"
a».a**<*>
l*.l**">
I.tM
ia*,*l|l>ll
«
l.tl*
1.1M
I.MB
il.M*l»'l
4»»
lot
•
*
t.lt*
1411. til
II
TOTAL
1 I»4.t4»
'">>,*>•
11,114
11,111
111,11*
111.1*1
•
I.llt
l.llt
• _ l.Mt
It, Ml
l.llt
Jl*
l.t«*
IM
Itl.lti
11, lit, 1*1
u>«* or
COHHOOITI
IIAMIUI
14,llt'"l
" ii, mini
i*.ui»»
U.IM'"'
1,141
M.tl.<1"
111
'"iV.iiii'W"
'">.«•»«•»
i.iii
in
4.144
1,11*
1.141
III
I».«14<"'
lll.OI
•crmi
111*
)
1
1
•
a
t
t
" •
•
«
•
i
•
•
•
....
*
11
«
iiit-
llii
i
<
>
•
*
*
•
a "
•
t
•
t
»
•
•
i*
NJNM* Of
Hit-
Mil
i
•"
i
•
i
•
•
i "
i
i '
•
I
I
•
•
'
J
ai
HCCIBOI
114*-
1141
II
•
1
1
a
i .
i
"«
•
a
i
•
i
*
i
•
41
T* •¥ VK
IIM-
Illl
ai
>
i
i
i
4
t
,'"
•
•
•
1
•
|
t
1
41
M or m
nit-
IMI
11
1
1
t
1
a
•
i
a
•
•
0
a
*
i
•
41
CTM.LAT1
1111-
im
i
"i
i
t
*
•
•
"~«"~
•
i
i
a
i
*~
i
-.__,
it
OH
HOT
mroiTtc
•
a
•
t
1
•
•
•
«
»
«
•
*
i
*
4
Ill
III
ACludt «CCId«»l of ili.MI.
nclud* ucldut of lit*.Ml.
nclud* iccldoic of Ill.Otl.
f>cluj. ACCld»l« Of lltl.tttl ilt*,0llti .
nclud* «ccld.nlt of lll.tt* m4 111,*.**.
ncUa. >ccld»t of 111,***.
111!
1141
IH)
(III
cU«lt> ut |l*,**( »»4 111,Ml.
uludx uci«Ml «l III.M*.
MllldM MCldMM 01 lilt,*** Utt lit.MO.
oldMtl «llk |0««M Of l.llll l.aiil 4,4SJ| and
,111 ktriolt.
Mludai locldaolt "Itk loiaoo of t.llJ aod >.*•• karrola.
acludai AooloaoU Hltk lom.ai ol II.1*1 aad 4,1(1 katlalo.
•cludaa 4cold«ti vltk lotoai ol l.llli I.41II I.JIIi «ad
.111 karrall.
U>l Inoludaa
111! I«o!
Ill) lac
111) IM
vltk !<>•••• of i.ilti S,t«*i «iwl 4,ill
.cclauti >ltk lo»«> of 4.115 >»4 !,*«• b«ir«l«.
iccldrat t>ltk lo» ol I,SI* k»r«U.
•ccld.nt ultk |0» Of 1,111 b«II4l4.
-------
.» COMHOBIW
cn
COHHOOm
orCRATIOH ACCIDENT*
Cruda Oil
0..01U.
N.a.L.
riMi on
i.r.B.
j.t r».i
oieui riui
Anhydroua AMMmla
Tc.na.1.
«"""»•
TOTAL
NO. or
ACCIDENT*-
'
111
M
5
11
14
1
7
1
1
1
201
> or
TOTAL
M.I
10.7
1.4
(.1
(.7
1.4
1.1
1.4
4.S
O.S
100.0
LOU
(kAMXUI
94tg)a(l|
29.474(21
2f .ol$(l|
41,4S7(^)
10, t lad
l.JOi
Il,«t4
1*,14.«« '
111
»"
155.017
« or
TOTAL
17.7
11.1
11.4
17.1
ia.o
0.7
S.I
4.0
0.1
0.4
100.0
(11 "^U4" MjUanta with loaeae of »,17li 4,5)1, 5,150, s
(11 ncladaa an* accident o( 4,»7 barnle.
()l ncludei aecldanti ulth loeaee of l.llli «,m, and t 419
141 nclwUt .ccU.ot. ulth loaaee at 14. Itt >^1 4.S17 Mrr.i
ISI HOlud«« Aecldanto uith lo*>«* ol C $04i S Stt7i and 4 OS!
lil ncludfts OIM acoid«AK ol 4,iSl barrel* ' '
171 neludaa accidutt o( (15,040.
Ill neludai accldant ol I1SO.OOO.
r»mn« BANAO* lit
CAM1SA
J «»,«.">
17..U7»«
1.1IS
a.tif
IK.MS'"
1,1*0
l.ltO
7,140
0
21
1,11*. Ill
llai S.OOOi and
IOTJUU
U?,
• ,710
n.oio'»>
1>.*J«I»»
U5.I50'1)»
•
1.500
sa,M*'">
54,04*
1.
411,
-------
APPENDIX D
A STATISTICAL ESTIMATE OF PIPELINE LEAKAGE
D.I BACKGROUND
A statistical methodology for estimating the frequency and magnitude of
pipeline leaks utilizing reported data is presented. The particular data
base used consists of a computerized compilation prepared by SAI personnel
containing key data on all oil spills reported to the Office of Pipeline
Safety for the years 1971 through 1975 inclusive.
In the absence of a related fire, explosion or pollution, the OPSO data
does not include data on spills under 50 barrels. It is the objective of
the methodology discussed here to supply a reasonable estimate for the fre-
quency of occurrence and magnitude of spills falling below this 50 barrel
reporting threshold. The study estimates that about 70 percent of all spills
from crude pipelines fall below 50 barrels. In terms of oil lost, this
amounts to about two percent of the total spillage. It should be noted, how-
ever, that 70 percent is based on the types and causes of spills reported.
Thus, there is an absence in OPSO spill data of causes of small leaks, such
as at small fittings, sensor taps into the line, etc. Thus, this 70 percent
figure is a minimal number and it is estimated that if all line spill causes
are included the 70 percent figure would be increased upward to the 80 to 85
percent range.
D.2 ASSUMPTIONS
Two primary assumptions underlie the method. First, it is assumed that
the OPSO data base is a fair representation of actual fact. It is assumed
that all leaks exceeding the 50 barrel reporting threshold are reported, and
that their magnitudes are determined without bias. Lack of bias is more im-
portant than simple error of estimation. For example, the data does not in-
dicate groupings or clusterings of spill volumes about convenient numbers.
A great many spills are reported as "50,", "100," or "200" barrels, for ex-
ample, but very few or none as "52" or "97," etc. By assuming no bias, it
is meant that these groupings are a result of honest estimation to the near-
est reasonable number rather than a result of an attempt to consistently
exaggerate or play down the size of spills. The correctness of this assump-
tion weighs more heavily on the accuracy of the derived estimate than on the
validity of the approach in general.
The second assumption is that the nature of the distribution of re-
ported spills over 50 barrels is shared by spills below the 50 barrel
416
-------
reporting threshold. This assumption will be supported by the degree to
which the observed data (reported spills) fit a meaningful probability den-
sity function and by observing that the same contributing physical laws
which control the magnitude of spills over 50 barrels are operable in con-
trolling the magnitude of spills under 50 barrels.
D.3 STATISTICAL ESTIMATION (CRUDE SPILLS)
Table D-l is a summary of all reported spills from U.S. crude pipelines
over the five years 1971 through 1975 inclusive. This summary was" extracted
from OPS data which includes spills from all parts of a pipeline system:
the pipes, pumping stations, delivery points, tanks, valves, etc. The data
presented in Table D-l is limited to spills from the"pipes themselves, ex-
cluding all other sources.
Attempts were made to fit alternative probability density functions to
the five-year spill data. The primary density functions considered were the
gamma and the log-normal. These two were selected because they share the
property of having a fixed lower bound (there are no negative valued spills)
and because they result "naturally" from certain intuitive and reasonable
bases.
The gamma distribution is a generalization of the exponential distribu-
tion. If all spills were generated by the same size leaks (same leak rates)
and the difference in magnitudes between spills were solely attributable to
the time from inception to system shutdown, and if the likelihood of detec-
tion and shutdown remained constant with time, then it would follow that the
exponential distribution would underlie the distribution of spill sizes.
The gamma distribution, being the parent distribution of the exponential
distribution, was thus considered as a meaningful and "natural" candidate
for trial.
The log-normal distribution results when a random variable is itself
the product of many other random variables, just as a normal (or Gaussian)
random variable results from the sum of many other random variables. It is
seen immediately that the two major factors determining the magnitude of a
pipeline spill, namely leak rate and time to system shutdown, contribute
multiplicatively to spill volume. That is: magnitude = rate x time. Ad-
ditionally, leak rate is proportional to the product of the opening of the
pipe (area) and the square root of the pipeline gage pressure. Time till
detection of the leak (and hence system shutdown) is also conceivably the
result of the product of several factors such as leak detection resolution
(through the use of line variance measurements, etc.), total line flow rate,
and, of course, leak rate.
In attempting to fit gamma and log normal distributions to the histori-
cal data, it was necessary to estimate the total number of spills (reported
as well as unreported) while simultaneously determining the controlling
parameters of the candidate distribution. The results of our investigation
found an extremely close fit of the historical data is achieved with the
log-normal distribution. We will therefore discuss our method of parameter
417
-------
TABLE D-l. CRUDE SPILLS FROM LINE PIPE (NO. OF SPILLS)
Range
(Barrels)
50-100
100-200
201-400
401-800
801-1600
1600-25000
1971
36
35
26
21
14
14
146
1972
37
28
19
22
11
24
141
1973
27
25
18
19
13
20
122
1974
44
28
24
16
12
19
143
1975
23
27
13
11
6
8
' 88
5-Year
Total
167
143
100
89
56
85
640
418
-------
estimation utilizing this distribution. The method was applied in similar
fashion with the gamma distribution. The log-normal distribution is defined:
f(x;y,a) = — - — exp |- -^- (Inx - u)
ax/2¥ L 2a J
where
x > 0
a > 0
_ oo < y < ao
Now let a., i = 1,2,..., n+1 denote the end points of n intervals of spill
volume
b- , i = 1,2,..., n denote the historical data values (spills)
for each interval .
The intervals are defined in terms of spill magnitude. For example,
we used a^ = 0, 82 = 50, or the magnitudes 0 to 50 barrels for the first
interval, 32 = 50 to 33 = 100 for the second interval, and so on.
The determination of best possible fit was measured using x2 values as
follows (x2 denotes the chi -square variable as used to test statistical hypo-
theses. See Section 6.2 for additional development.) For given values of
u and a (the log-normal parameters) together with T, an assumed number of
total spills, define F.. as:
F. = I f(x;u,a)dx
ai
or the probability that a spill falls in the ith interval, and
1=2
Again, b-j is the actual spill frequency in the ith interval, while
is the estimated frequency based upon the assumed values of u, a and T.
419
-------
A program was developed which found the optimal values of y, a and T,
minimizing x2 for the historical data. This program utilized a three level
search or optimization scheme.
• Values for y were determined by direct trial, stepping through
a range of values.
• Values for a were determined by utilizing a Fibonacci search
to seek the best a for each y.
• Values of T were determined analytically for each pair: y,a.
The values of y and a determine the values of F-j in equation (1).
The optimal value of T is then found by differentiating (1).
n n n
X2 = T 2 F. - 2 2 b, + T 2 b,2/F. (from (1))
1-2 n 1-2 1=2 n n
1-2
and setting
yields
The optimal values determined for the five-year data in Table D-l were:
U = 2.31
a = 2.94
T « 2217.48
The x2 value was 1.65. A comparison of the actual and predicted values is
given in Table D-2.
Utilizing this log-normal distribution we generated Table D-3 which
provides actual versus predicted spill frequencies for each of the five
years, together with an estimate of barrels spilled for each interval.
Table D-3 also provides the resulting estimates for spill frequencies and
losses for the 0 to 50 barrel range. In generating these estimates the
basic parameters (y and a) of the log-normal distribution remained unchanged.
420
-------
TABLE D-2. LOG-NORMAL COMPARISON WITH FIVE-YEAR DATA
Number of Spills
Index Range (barrels) Prediction Actual Difference
(1) a, - a1+l F.T b.
2 50 - 100 167.22 167 0.22
3 100 - 200 139.23 143 -3.77
4 200 - 400 109.68 100 9.68
5 . 400 - 800 81.75 89 -7.25
6 800 - 1600 57.66 56 1.66
7 1600 - 25000 85.28 85 0.28
421
-------
TABLE D-3. PREDICTED AND ACTUAL SPILL FREQUENCIES
1971 I9M 1973 1974
Ningt Actutl trttSi »»rr«ll Actual trti Itrrtli Actvtl trtt limit Acluit trtt lirrcli
0-10 - 239 (29 - 212 III -- 201 S2I -- 234 (IS
10 - 20 - 47 (84 •- 46 664 - 40 S74 - 46 6(9
20 - 30 — 27 (59 - 26 640 -- 22 S53 — 26 644
10 - 40 - 18 632 - 18 614 -- IS S3I -- II 611
!
40 - SO - 14 607 -- 13 590 — U Sl» -- U 494
50 - 100 36 3* 2,746 37 37 2.666 27 32 2.306 44 38 2,665
j^ 100 - 200 35 32 4.557 28 31 4.426 25 27 3.827 28 11 4.454
ro
200 - 400 26 25 7,158 19 24 (.951 18 21 6.011 24 25 6.999
400 - «00 21 19 10.636 22 IB 10.330 II 16 8.932 16 II 10.401
«00 - 1600 14 13 I4.J55 II 13 14.524 1) II I2.SS9 12 13 14.623
It00 - 2SOOO 14 20 101. 1(3 24 19 104,082 20 16 90.002 19 19 104,798
1975 fl»e-»««r Jouli
Actutl trtt Itrnll Actvt) trtt Itrrelt
147 317 - 1042 2.739
29 421 -- 207 2.977
16 40S -- 116 2.869
II 319 - 79 2.75.1
1 374 - S9 1.644
23 24 1,689 167 167 11.955
27 20 2.804 143 139 19.843
13 IS 4.404 100 110 31.164
II 12 6,544 89 82 46.310
6 8 9.201 56 58 65.113
8 12 (5,937 85 IS 4(6.630
^og-normal estimate (prediciton)
-------
Expression (2) was utilized to generate the total number of spills for each
reporting period.
D.4 DISCUSSION OF RESULTS
In review and summary, a survey of alternative candidate distributions
to model the pipeline leak process showed a remarkable ability of the log-
normal distribution to fit the observed historical data. Fitting the his-
torical data alone was not considered adequate, however. There are, for
example, many so-called empirical distributions available to a researcher
interested in modeling a given process. These empirical distributions can
be manipulated so as to fit virtually any observed data. The degree of fit
is limited only by the ingenuity and industry of the researcher. Empirical
distributions are particularly useful for summarizing data when no clear-cut
theoretical explanation is available. Since the historical data on pipeline
spills is incomplete, and since the objective in this case is to estimate
the magnitude of the missing data, it follows that the probability density
function used to generate the estimate should possess a reasonable theoreti-
cal basis. Degree of fit is secondary and serves to validate the use of
the suggested distribution.
Our approach, as discussed earlier, was to try several likely distribu-
tions without excessive regard for theoretical explanations. The aim was
to first see if any of the "natural" distributions adequately fit the data.
The two best candidates in terms of fit, the gamma and the log-normal, also
had the strongest underlying justifications. Of the two, only the log-
normal exhibited a "good" fit by common statistical standards.
Having discovered the ability of the log-normal distribution to fit the
observed data we now reverse our discussion and hypothesize that the magni-
tude of crude pipeline spills is distributed log-normally. With some un-
avoidable repetition we will defend this hypothesis.
D.4.1 Characteristics of the Log-Normal Distribution
Since a primary objective of modeling the pipeline leak process is to
estimate the frequency of events for which direct data is not available, it
is necessary that the hypothesis (of log-normality) exhibit a reasonable
theoretical foundation. Not only must the log-normal distribution fit the
observed data over reported ranges, but there must also be a reasonable ex-
planation to justify extension of the distribution into ranges void of data.
The log-normal distribution can be derived as the model for a process
whose value results from the multiplication of contributing factors just as
the normal distribution is derived as the limiting form of a process whose
value results from the addition of contributing factors. In the case of
the normal distribution, the well known Central Limit Theorem shows that
the individual distributions of the contributing factors (random variates)
does not affect the resulting normal tendency of the sum. A simple example
is the sum of heads achieved in a sequence of flips of a "fair" coin. Each
individual trial (flip) is a binary distribution with only two possible and
equally likely outcomes. The sum of "heads" is a binomial distribution
423
-------
which approaches the normal distribution as the number of trials increases.
An equivalent "central limit theorem" exists for the log-normal distribution
(ref., The Log-Normal Distribution. J. Aitchison and J.A.C. Brown, Cambridge
University Press, Cambridge, 1957).
The log-normal distribution is the model of a random process variate
whose logarithm follows the normal distribution with parameters u (mean) and
a (standard deviation). Thus, if y = Inx is normally distributed, the
probability density function g(y) for y is:
-V (y-y)2l ; - <
[2a2 J
y <
Since y = Inx is equivalent to x = e^ which is a strictly increasing function
of y, we can transform g(y) using the Jacobian of y(x), yielding:
-L (lnx-u)2 |I|
2a x
f(x) = g[y(x)l ||£| • -L- exp
L J UA ov^F
Since for all values of y, x = e^ is positive, it follows that:
f(x) =
which is the log-normal distribution.
; 0 < x < <»
Figure D-l displays several log-normal curves with various values of u
and a. In the normal distribution, u specifies a location (the mean value)
and a specifies a scale (the standard deviation). In the log-normal distri-
bution u becomes a scale factor and a becomes a shape factor.
The log-normal distribution represents many familiar processes. Exam-
ples are:
• Distribution of personal incomes
• Size of an organism (simple life forms) whose growth rate
is subject to many small impulses (food, light, temperature,
cell division, etc.)
• Distribution of particle sizes obtained from breakage (ref.,
The Mathematical Description of Certain Breakage Mechanisms
Leading to the Logarithmic-Normal Distribution. B. Epstein,
Journal of the Franklin Institute, 1947).
424
-------
Figure 0-1. Log-normal distributions with values values
of u and a.
425
-------
The prospect of relating the sizes of oil spills from pipelines to the sizes
of particles resulting from dropping a china cup on a tile counter may or may
not appeal to one's reason. Certainly, in both cases, it seems evident that
there are more small spills than large spills as there are more small china
particles (even dust) than large particles. Also obvious: There are no
negative spills nor are there negatively sized particles.
The fundamental support of the hypothesis is that the key factors which
determine spill size contribute multiplicatively. It is clear that spill
magnitude is directly proportional to leak rate and correction shutdown re-
sponse time. Leak rate is further proportional to the area of the opening
in the pipe and the square root of the internal line pressure. Correction
time is either proportional to the ratio of leak rate and line flow (if leak
detection monitoring is sensitive to leakage as a percent of flow) or simply
proportional to leak rate (if detection is based on visual search dependent
on leak volume alone). The key factor which underlies the mechanics of the
spill process is proportionality:
• The larger the line the larger the spill
• The higher the pressure the greater the leak rate
• The greater the split or opening the greater the leak rate
• The larger the time till detection the greater the spill.
and so on. The exact functional form relating spill size to these factors is
hot so important. Clearly, these major factors contribute in a multiplica-
tive as opposed to additive fashion. This observation strongly supports the
log-normal hypothesis.
D.4.2 Goodness of Fit
Table D-3 (see Section D.3) gives a comparison of actual versus esti-
mated spill frequencies for various spill magnitude ranges. Figure D-2 pre-
sents a graphical comparison of the same data. Additionally, Figure D-2
gives the x2 values as a measure of goodness of fit for each of the five
years (1971-1975) as well as for the five-year totaled values.
The chi-squared (x2) goodness-of-fit test is the most commonly used pro-
cedure for evaluating distributional assumptions. The use of the test in-
volves grouping observed data into frequency cells. The resulting cell fre-
quencies are compared with the expected frequencies from a proposed distribu-
tion. This comparison generates a test statistic which approximates a chi-
square variate. The test of the proposed distribution is in the form of a
statistical test of the hypothesis that the observed data come from the pro-
posed distribution. As the value of x2 (the test statistic) increases the
probability that the observed data came from the proposed distribution dimin-
ishes. A level of significance is usually established to determine acceptance
or rejection of the hypothesis. The level typically adopted is five percent
(or .05), meaning that x2 values exceeding the .05 level could have occurred,
426
-------
•a o
•a o
§o o o LO
§O O vo evi
c\j «r co •—
•— o
o o o o **
o *•• *J *^ •*-*
Arf O
o o o o o us
tf) '^ CM ^ - 00 ••
APAPAPAPAPAP
FIVE YEARS
X2 « 1-55
A = ACTUAL
P * LOG-NORMAL PREDICTION
APAPAPAPAPAP
1977
X2 * 2.36
APAPAPAPAPAP
1972
X2 ' 3.87
APAPAPAPAPAP
1974
X2 = 1.84
APAPAPAP APAP
1973
X2 ' 3.20
APAPAPAPAPAP
1975
X2 - 5.10
Figure D-2. Comparison of predicted and observed spill-frequencies (crude),
427
-------
were the hypothesis true, with a chance of less than one in twenty. The x2
test statistics are calculated using expression (1), Section D.3.
The x2-values for the five years data on spills from crude pipelines,
using the log-normal distribution f(x;y,a), y = 2.31, a = 2.94, are:
Year x2
1971 2.36
1972 3.87
1973 3.20
1974 1.84
1975 5.10
All five years 1.65
The x2 value at the .05 level (five degrees of freedom) is 11.07. It is
clear therefore that the proposed distribution provides an exceptional fit of
the observed data.
D.4.3 Conclusions
The distribution of spill sizes from U.S. crude pipelines over the five
years 1971 through 1975 is reasonably modeled by a log-normal probability
density function, (f(x;y,a), u = 2.31, a = 2.94. This function is displayed
in Figure D-3. The average yearly distribution of spills and barrels lost
is given in Table D-4 and depicted by Figure D-4.
Regarding the question of spill frequency below the reporting threshold,
it is seen that the frequency of spills increases radically as the spill mag-
nitude decreases. The mode of the log-normal function (y = 2.31, a = 2.94)
occurs at about the .0018 barrel (9.7 oz or 286 cc) level. The median occurs
at close to ten barrels (almost half of all spills are ten barrels or less)
and the average (mean) spill is about 300 barrels.
From Table D-4 it appears that about 70 percent of all spills which
occur are not reported, being below the 50 barrel OPS threshold. In terms
of volume lost, however, spills in the 0-50 barrel range, while numerous,
account for only 2.1 percent of the total spillage from crude pipelines.
D.5 PRODUCT SPILLS
The estimation methodology was applied to product spills from pipeline
leaks with similarly favorable results. Table D-5 contains a summary of
product spill frequencies for the five-year period covered by the data base.
Products included were:
428
-------
sfiu tin limn n
l « • Ml. • • >-M
»iu sut (MMUI
•UUMCHL MM (IIFKItl
Figure D-3. log-normal density function.
429
-------
TABLE D-4. ANNUAL DISTRIBUTION OF SPILL FREQUENCIES, BARRELS LOST (CRUDE)
Range
(Barrels)
0 -
10 -
50 -
100 -
200 -
400 -
800 -
1600 -
10
50
100
200
400
800
1600
25000
Frequency
208
92
33
28
22
16
12
17
Percent
49
21
8
7
5
4
3
4
Barrels Lost
548
2,248
2,391
3,969
6,233
9,262
13,023
93,326
Percent
Oa
2
2
3
5
7
10
71
TOTAL 428 131,000
aLess than 1/2 of- one percent.
430
-------
OOOS3
- OOS21
(U
u
c
o
•r—
U
(/I
(U
•o
(U
(O
3
-------
TABLE D-5. PRODUCT SPILLS FROM LINE PIPE (NO. OF SPILLS)
Range
(Barrels)
50 -
100 -
201 -
400 -
800 -
1600 -
100
200
400
800
1600
25000
1971
7
7
15
21
9
15
1972
8
14
14
18
17
13
1973
7
13
14
9
6
15
1974
7
6
8
6
11
11
1975
9
11
17
9
15
15
5-Year
Total
38
53
68
63
58
69
74 84 64 49 76 347
432
-------
• Heating/fuel oil
• Kerosene
• .Gasoline
• Diesel
• Natural gas liquid/condensate gas
• Ammonia.
As with the estimate for crude, only leaks from the line pipes were considered.
Spills or leaks from valves, pumps, tanks, etc. were excluded.
The best log-normal fit to the five-year data was obtained with u = 6,
a = 1.55. The goodness-of-fit for the five-year data was even better than
the "five-year fit" on the crude data, with x2 = -70. Figure D-5 displays
the fit of the log-normal distribution to each of the five years. The x
values for each of the years data are:
Year Y
2
1971 5.71
1972 2.64
1973 3.95
1974 3.88
1975 3.48
All five years 0.70
Considering again that the x2 value (5 D.F.) at the .05 significant level
is 11.07 shows the remarkable fit of the log-normal distribution to the ob-
served data.
Table D-6 provides a tabulated comparison of actual versus recorded
spill frequencies. Table D-7 gives the average annual distribution of spill
frequencies and barrels lost by spill magnitude ranges.
Figure D-6 depicts the log-normal probability density function, f(x;y;o),
u = 6, a = 1.55. Figure D-7 gives cumulative density functions for spill
frequencies and barrels lost in relation to spill size.
A comparison of the log-normal distributions of crude and product spills
is noteworthy. The mode (or peak) of the crude distribution occurs at the
.0018 barrel level, while the mode for product spills occurs at 37 barrels.
This indicates that crude spills increase in frequency as spill size decreases,
almost to zero, while product spills "peak out" at around the 37 barrel level.
The median of crude spills is at about ten barrels, while for product '
spills the median is at about 400 barrels. While half of all crude spills
are ten barrels or less, half of all product spills are 400 barrels or less
(or conversely - larger than 400 barrels).
433
-------
8
ca
§ §
§ §
a
o
CM
LO
CM
t
O
o
VO
A * Actual
P » Log Normal Prediction
° =17",
i& Ci
*•
K\\\\VI
f7
x ,
y^X
1
A P A
*
X
x
x
X
x
x
x
X
X
X
X
X
^^ ^
x
•v
X
X
^
—
X
X
X
/
x
X
X
x
—
PAPAPAPAP
FIVE YEARS
X2 • -70
.UJ,J
*
^^•a
J
x^
X
|7
X
x
x
x
X
X
x
- — " -
X
X
;^~j
X
X
x
APAPAPAPAP AP
1971
X2 « 5.77
7~
z.
•IBM
X
X
x
x
__
7
/
/
—
/
7
/
/
/
/
X
^M*«
_
X
x
x
APAPAPAPAPAP
X
X
X
X
X
—
~^^m
X
/
L.
7]
X
X
7—
M^HH
,\\\\\N
—
APA PAPAPAPAP
1972
2.64
XI
x
x
X
X
X
x
X
x
x
APAPAPAPAPAP
1974
2
1973
« 3.95
\\
P
•-^
/.
^
3.33
A P APAPAPAPAP
1975
\ * 3.48
Figure D-5. Observed and predicted product spill-frequencies,
434
-------
co
CJ1
TABLE D-6. PREDICTED AND ACTUAL SPILL FREQUENCIES (PRODUCTS)
Interval
fUn,
0 -
10 -
20 -
10 -
40 -
SO-
100 -
200 -
400 -
800 -
1600 -
It Actutl
10
20
10
40
50
100 7
200 7
400 IS
100 ' 21
1600 9
25000 IS
1*71
frttCI Itrrtli Actuil
.7 i
l.i 21
I.I 44
I.I tl
I.I 80
8 598 8
12 1.757 14
IS 4.240 14
IS 8.40S 18
12 11.685 17
16 69.645 U
1972
frtdO Itrrtlt
.1 i
1.7 25
1.9 48
2.0 70
2.0 8*
» 664
U 1.950
16 4.706
16 9.128
11 IS. 187
17 77.218
117)
Actiul »rrt° limit
.( 4
1.) 20
l.i 17
I.S 54
l.i tl
7 7 ill
U 10 1.508
14 U 1.619
9 11 7.211
t 10 11.744
li 1) SI.7U
1174
AcUil trt*O
.i
1.0
l.i
1.2
I.I
7 i
t t
8 10
t 10
II 8
II 10
1 tori-til
>
IS
21
42
il
19t
1.164
2.109
s.st*
9.067
46.142
I»7S
Actiul fredO
.-- •»
I.S
1.8
I.I
1.8
9 8
II 12
17 IS
9 IS
IS 12
IS It
Ftvt-Itir 1»Ult
ItrrtU
i
2)
44
64
81
60S
1.777
4.288
8.499
11.818
70.422
Actiul frtdO
1.1
6.8
7.9
I.I
1.0
18 16
SI 54
68 it
. 61 66
S8 54
69 70
1 hrrtlt •
21
104
197
284
162
2.702
7.919
19.158
17.975
61.829
1I4.6SI
aLog-normal estimate.
-------
TABLE D-7. ANNUAL DISTRIBUTION OF SPILL FREQUENCIES
BARRELS LOST (PRODUCTS)
Range
(Barrels)
0
50
100
200
400
800
1600
- 50
- 100
- 200
- 400
- 300
- 1600
- 25000
Frequency
7
7
11
13
13
11
14
76
Percent
9
9
14
17
17
14
18
Barrels Lost
194
541
1,588
3,832
7,595
12,366
62,930
89,046
Percent
0
1
2
4
9
14
71
436
-------
f(x;p,o) \i = 6, o = 1.55
1 I
,0020
.0015
co
.0010 •
.0005 '
I i
HISTORICAL DATA (TYPICAL)
50 100 200
400
SPILL SIZE (BARRELS)
800
Figure D-6. Log-normal distribution, product spills,
-------
co
00
CUMULATIVE DENSITY OF SPILLS
CUMULATIVE DENSITY OF BARRELS LOST
00 O
o o o
CO 10
SPILL SIZE (BARRELS)
Figure D-7. Cumulative density functions (products).
-------
The average crude .spill is about 300 barrels, the average product spill
about 1166 barrels. In short, product spills are, on the average, somewhat
larger than crude spills. This is an expected result since products are
generally lighter and less viscous than crude.
439
-------
APPENDIX E
METHODOLOGY
E.I VISUAL AND AIDED VISUAL LINE OBSERVATION
E.I.I Visual Inspection by Air or Ground Level
(See Table 46 methods (a) through (f) and Sections 5.3.2.1 and 7.3.1.3.)
E.I.2 Visual Inspections by Ground Patrol with Hydrocarbon Detector or
Other Comparable Device
(See Table 46 method (h) and Sections 5.3.2.1, 5.3.2.4 and 7.3.1.1.)
E.2 OIL SPILL DETECTORS ON OR NEAR THE WATER
E.2.1 Oil Spill Detectors
(See Table 47 (c) and Sections 5.3.2.2 and 7.3.1.2.)
E.3 INTERNAL FLUID VARIATIONS DURING TRANSFER
E.3.1 Pressure Deviations
(See also Table 48 (a) and Sections 5.3.2.3 and 7.3.1.3.)
Principle of Operation
The pressure deviation of leak detection methods basically involve con-
tinuous monitoring by computer of line pressure at various locations along
the pipeline is carried out during oil transfer. If pressure deviations
(i.e., pressure drops, etc.) are excessive and exceed a set point value,
system alarms are generated.
Sensitivity
Pressure deviation inspection is most sensitive for large leaks close
to the discharge pumps and close to the downstream end of the pumping sta-
tions. At locations where the line pressures are low, such as upstream of
the pumps, large leaks cause small pressure drops that are difficult to de-
tect.
440
-------
Manufacturer and Costs
All necessary equipment and components for typical pressure deviation
inspection systems are commercially available. Total system costs are gen-
erally much lower than other continuous leak monitoring methods used during
petroleum transfer.
Advantages
This is a commonly used inspection method for detecting large leaks. It
is simple to implement.
Disadvantages and Limitations
It is extremely difficult to detect small leaks using this inspection
method. Because of the principle used and equipment limitations, the follow-
ing items (and others) contribute to the poor sensitivity of this method:
• Variations in input or output tank heads
• Changes in fluid properties
» Changes in temperature
• Flow variations affect line pressure drop because pressure
drop varies as the flow rate squared
• Leaks are inadvertently compensated for by pressure control
valves operating slightly when a system is operating at
maximum capacity under pressure control
• Setpoints are usually set very high to prevent false shut-
downs. In these instances/ only very large leaks or rup-
tures can be detected.
E.3.2 Flow Rate
E.3.2.1 Comparison—
(See also Table 49(c) and Sections 5.3.2.3 and 7.3.1.3.)
Principle of Operation
Flow rate comparison inspection detects pipeline leaks by measuring the
difference in the rate of flow at two locations. Computer systems are avail-
able that can continuously compare flow rates every few seconds and generate
an alarm when deviations exceed a setpoint value (one percent to five per-
cent of normal flow rate). Alarm levels are usually set to take into ac-
count changes in pumping rate, temperature or density of the oil, etc. A
variety of transducers that provide signals proportional to rate of flow are
commercially available. The most commonly used types include ultrasonic,
turbine, and orifice transducers.
441
-------
Capability
This leak detection equipment is generally used to inspect for large
leaks in the pipeline. The method generally works best on lines where the
flow is relatively stable.
Manufacturer and Costs
Flow rate comparison systems are commercially available from a number
of manufacturers. Costs vary widely depending upon line location and dimen-
sions, transducers, and the required accuracy and automation. Typical manu-
facturers of systems include Daniel Industries, Adec, Inc., and Waugh Con-
trols, etc.
Advantages
This inspection method is widely used and provides rapid detection of
large leaks.
Disadvantages and Limitations
The method can only be used to detect major oil leaks. Also, there is
a tendency by operators to raise setpoints to reduce the possibility of
alarms and thus decrease leak sensitivity.
E.3.2.2 Flow Rate Deviation—
(See Table 48(d) and Sections 5.3.2.3 and 7.3.1.3.)
The flow rate deviation method provides continuous leak detection of
the pipeline by measuring the deviations in flow rates at specific flow
stations. If the A flow rate exceeds a certain value (for example, unex-
plained changes of normal flow rate (one to five percent)), a leak alarm is
generated.
This leak detection is similar to the flow rate comparison except that
a comparison of flow rate at another flow station is not required.
E.3.3 Volume Comparisons (Balance)
(See also Table 48(b) and Sections 5.3.2.3 and 7.3.1.3.)
Principle of Operation
A variety of volume comparison techniques for detection of leakage is
commonly used in the pipeline oil transportation industry. The basic opera-
tion is to measure the input volume, output volume and line pack to check
metered barrels into the pipeline against barrels measured out. These mea-
surements are generally based on the following equation:
- VQ
442
-------
where
V, = leak volume during time t
V.j = volume of liquid put into system during time t
V = volume of liquid taken out of system during time t
AVS = change in volume of liquid in pipe and tanks in system.
Corrected flowmeter readings are used to provide measurements of Vi and VQ
at the input and output. Vs is usually computer corrected for line pack
effects by measurements of temperature and pressure at various locations on
the pipeline.
If the line is leak-free, no crude is lost. Volumes can be measured
by meters (i.e., turbine flowmeters, etc.) and by tank gages. The metering
system provides the required volume data with the computerized supervisory
control system automatically gathering, comparing and correcting for various
parameters (temperature, pressure, density of crude, etc.). Computerized
control systems are used that make continuous volume comparisons over typi-
cal time intervals ranging from every few minutes to every one to two hours.
An alarm is generated when a volume comparison difference exceeds a prede-
termined setpoint. Settings are made taking into account corrected volumes
and system tolerances in transducers, electronics, power variations, etc.
Capability
This leak detection equipment is used primarily to inspect for large
leaks in a short period of time and small leaks over a long period of time.
Manufacturer and Costs
Volume comparison systems are commercially available from a number of
manufacturers. Costs vary widely depending upon line locations and dimen-
sions, transducers, and the required accuracy and automation. Typical
manufacturers of'systems include Adec, Daniel, Siemens, and Waugh.
Advantages
Equipment can be used on an almost continuous basis. Systems can be
automated to provide simple operation. Major leaks can be detected in a
short period of time.
Disadvantages and Limitations
Some of the main disadvantages and limitations of this method for in-
spection purposes are:
443
-------
• Detects leaks after they occur
• Difficult to detect slow leaks that over a period of time may
result in a major oil spill
• Cannot detect a catastrophic failure in sufficient time to
prevent a major oil spill
• Tendency by operators to raise setpoints to reduce alarms be-
cause of line pack and other considerations
• Detects leaks only about once per hour for most commonly used
systems.
E.3.4 Mathematical Modeling
(See also Table 48(f) and Sections 5.3.2.3 and 7.3.1.3.)
Principle of Operation
Mathematical modeling is a real-time computerized pipeline monitoring
method for detection of small amounts of oil leakage; only losses in the
inventory are of interest and the inspection is a form of dynamic inventory-
ing of the pipeline product. The method is affected by solving momentum and
continuity equations for a specified pipeline and/or hose string network.
The differential equation that results is solved by iterative methods with
suitable techniques so that the mathematical model is run in real time and
can be trimmed as required to fit the actual pipeline. Mathematical models
are available that fit the pipeline during start-up and compensate for.
transients such as pump start-up, shutdown, valve closures, water-hammer
effects, etc. that normally occur in the pipeline. In addition, models can
provide accurate means of compensating for line pack due to product compres-
sibility, in pipe-wall and hose-wall deformation.
Modeling methods require that a significant amount of information be
known and a variety of measurements made continuously. These include pro-
duct information (density, viscosity, etc.); pipeline dimensions and mate-
rials; valving; and product propagation information (flow and pressure at
both ends of pipeline, temperature gradient of the product, etc.).
Capability
Mathematical modeling can be used for continuous leak inspection of the
pipeline. The method also can be used during static or hydrostatic leak
tests.
Sensitivity
Estimated sensitivity is about 0.1 percent of the flow rate at the time
the leak occurs. For example, if 100,000 barrels per hour is offloaded, a
100 barrel per hour leak can be detected.
444
-------
Manufacturer and Costs
Mathematical modeling systems for leak detection inspection are commer-
cially available from a few companies. The leak detection equipment is in
current use on oil pipelines. Costs are typically less than one-half percent
of the cost of the pipeline. Costs are even lower if the equipment can be
used with existing supervisory control systems.
Advantages
Some of the" main advantages of this inspection method are:
• Computerized reduction
• Good failure detection
• Provides leak detection improvements over conventional
hydrostatic pressure tests
• Can be used in conjunction with supervisory control systems
• Requires only repeatable rather than high accuracy flow meters
• Limited continuous inspection.
Disadvantages and Limitations
One disadvantage is that the leak can only be detected after the flow
reaches a meter. Hence, for long distances between meters significant spill-
age can occur for large leaks or ruptures before detection.
This method is new and not totally proven on a large number of lines.
Although automatic, it may require trained personnel to properly interpret
results or maintain the system. Commercial availability of these systems
is limited because of the small number of companies currently involved in
this area.
E.3.5 Negative Pressure Surge
(See Table 48(j) and Sections 5.3.2.3 and 7.3.1.3.)
E.4 DETECTION AND LOCATION OF LEAKED OIL ON OR AT A SHORT DISTANCE FROM
PIPELINE
E.4.1 External Rods with Passive Acoustic Sensor
(See Table 48(n) and Sections 5.3.2.4 and 7.3.1.4.)
E.4.2 Passive Acoustic Array
(See also Tables 49(o) and 55(a) and Sections 5.3.2.4 and 7.3.1.4.)
445
-------
Principle of Operation
Experimental results have shown that acoustic signals generated in pipe-
line from external impacts, excessive internal stresses from material de-
fects and damage, precursor internal stresses just before a leak or material
failure, are all different; each event produces a characteristic signal that
can be differentiated from the other. These acoustic signals are commonly
called "acoustic emissions" and are excellent indicators of incipient fail-
ure. Generally, these acoustic emissions, except for external impacts, are
repetitive. Repetition rate usually increases to a peak value, then drops
off slightly, and then increases dramatically just before a critical mate-
rial failure or leak occurs. The acoustic emissions only occur when the
component is stressed - externally loaded or pressurized. Acoustic emission
signals are complex, dependent upon structure and fault type and the fre-
quency typically extends to the megahertz range.
The same acoustic system that is used to detect the continuous waves
generated at a leak source and which propagate along the pipeline can, with
additions to the signal processor, be used to detect the acoustic emission
signals. Using known wave attenuation characteristics of the pipeline, and
also using suitable signal enhancement, counting, and processing technique,
the location and condition of the flawed area may be determined.
Capabilities
Passive acoustic array inspections using acoustic emissions potentially
can be applied to effectively reduce oil spill risks by continuously moni-
toring the pipelines for actual failures (leaks) and impending failures
(internal defects that may lead to a failure).
Sensitivity
Defect location depends upon a variety of factors such as pipeline
material, size and length, acoustic transducer design and spacing, and
signal processing techniques. Additionally, hydrostatic tests can be used
advantageously to enhance the internal defects so that such defects can be
detected, whereas such defects it might not be detected at normal operating
and flow conditions.
Advantages
Some of the main advantages of this inspection method are:
• Excellent potential for incipient failure detection
• Computerized automatic system can be adapted to existing
supervisory control systems
• Commercial system currently available for similar applications
such as periodic proof testing of tanks, pressure vessels, etc.
• Permanent records
446
-------
• Continuous monitoring
• Locates defects for more detailed inspection by other means.
Disadvantages and Limitations
The main disadvantage and limitation of acoustic emission leak detection
is that incipient failure data is subject to interpretation as to the sever-
ity of the defect and how long before the defect grows to a critical size and
then causes a rupture, leak, etc. Also, system effectiveness and performance
specifications are uncertain.
Applications
Two slightly different passive acoustic array systems (see Section
7.3.1.4) are selected in this study for two main applications. The first
system is for installation on new lines or on lines located in high risk
areas such as underwater. Such systems are expected to be highly sensitive
and capable of detecting most impending failures and leaks including outside
force damage and ruptures. The second system is for retrofit installations.
These types of systems are of lower sensitivity but expected to be capable
of providing prevention of failures by detecting damage by outside forces
and minimizing spill size by detecting ruptures.
E.4.2.1 Passive Acoustic Array - Systems for New Lines--
(See also Sections 5.3.2.4 and 7.3.1.4.)
Installation
Acoustic sensors with signal conditioning equipment are expected to be
installed at approximately five locations each mile. A multiconductor cable,
running the length of the line provides both power and transmission of the
signal from each acoustic sensor. Master units supply the power, signal pro-
cessing and control. These units would typically be located at each pump
station and two locations between each pump station.
Capabilities
Systems are expected to be capable of both preventing most failures and
early detection of most leaks. Capabilities include detection of outside
forces, damage, internal defects, small leaks and ruptures.
E.4.2.2 Passive Acoustic Array - Retrofit Systems—
(See also Sections 5.3.2.4 and 7.3.1.4.)
Installation
Acoustic sensors would be installed at the pump station and at various
locations, typically at four stations (approximately ten miles apart), be-
tween each pump station. Signal conditioning and telemetry would be used
for remote monitoring of sensor signals.
447
-------
Capabilities
Systems are expected to be capable of detecting damage by outside
forces and rupture detection.
E.5 PERIODIC PRESSURE TESTS
E.5.1 Pressure Static
(See also Table 50(a) and Sections 5.3.2.5 and 7.3.1.5.)
Principle of Operation
The pipeline is operated in an intermittent manner at nominal operating
pressure; static pressure measuring techniques are used to detect a leak
when sections of the line are closed off and shut down. If the pressure
holds, the line is considered tight. High accuracy pressure gages are used
to monitor the line pressure. High leakage rates (i.e., greater than 500
1/hr) can be detected over short time intervals (about 15 minutes) by mea-
suring the static pressure drop. Detections for lower leakage rates require
that effects such as temperature be taken into account by waiting until the
temperature stabilizes.
Advantages
This is a commonly used inspection method that is of low cost, simple
to implement.
Disadvantages and Limitations
Some of the main disadvantages and limitations are:
• Requires leak detection method if leaks are located
• Variations in temperature, etc. limit sensitivity of leak
inspection
• Difficult to detect slow leaks
• Less sensitive than hydrostatic pressure difference method.
E.5.2 Hydrostatic - Pressure Drop and Pressure Difference
(See also Table 50(b) and 50(c) and Sections 5.2.2.5 and 7.3.1.5.)
Principle of Operation
The pressure drop method of inspection uses pressure difference gages
that are installed across a series of block valves that isolate sections of
an empty pipeline or hose string. The empty sections are then pressurized,
typically to 600 psi. Pressure difference gages will indicate if a leak
exists after a suitable period of time. Actual stabilization time depends
upon the required tightness (leak rate allowed) of the system and temperature
448
-------
effects. Usually, a derivative dAP/dt is used to show the deviation of dif-
ferential pressure as a function of time in order to simplify conclusions of
leakage. Numerous types of pressure difference gages are commercially avail-
able for this measurement.
Variations of this inspection method can be made. One variation is to
pressurize the line with gases such as helium and detect leaky areas with
helium leak detectors. Passive ultrasonic detectors and the passive acoustic
array can also be used to detect escaping gas. In another variation, the
line could be pressurized and sections of the line closed off. Then the
static pressure of each section could be accurately measured. This method
can give, to some extent, incipient fault detection. Stresses from proof
pressures slightly above normal may cause leaks to prematurely occur and be
detected; these leaks, however, would have eventually occurred because of
corrosion or other effects at some later time at normal operating pressures.
Advantages
This inspection method is simple, provides good incipient failure detec-
tion and is a widely used inspection technique. It is a high cost inspection
since the system must be shut down.
Disadvantages
Some of the main disadvantages of this inspection method are:
• Requires out-of-service operation
• Potentially can cause damage to pipeline components if test
pressure is excessive
• Requires leak detection method to locate leak
• Downtime required because of time for temperature stabiliza-
tion can be of long duration (24 to 72 hours)
• High cost to retrofit.
E.6 CORROSION INSPECTION
E.6.1 Change or Add Inhibitors as Needed
(See Sections 5.2.2.6 and 7.3.1.6.)
E.7 STANDARD NON-DESTRUCTIVE TESTING
E.7.1 Inspection of Sample of Line for Wall Thickness
Changes by ultrasonics or comparable technique (see Table 52(a) and
Sections 5.2.2.7 and 7.3.1.7).
449
-------
E.8 INSPECTION PIGS
E.8.1 Magnetic Flux Inspection Pigs
(See also Table 53(a) and Sections 5.2.2.8 and 7.3.1.8.)
Principle of Operation
A magnetic field is induced into a pipe wall around the circumference
and the field flows- in a longitudinal direction. In an undamaged pipe, a
smooth flow of magnetic lines of flux all remain within the pipe body wall.
A damaged or unnatural area of the pipe affects the flow of the lines of
flux and causes the flow to "bridge" across this area and create a magnetic
disturbance or flux leakage. This flux leakage is proportional to the size
and depth of the damaged area.
One of the most widely used magnetic flux inspection pigs is the AMF
Tuboscope Linalog. The instrumented pig uses electromagnets to induce the
magnetic field and it is sent through the pipelines, typically at a few
miles per hour, propelled by oil or water flowing through the pipeline.
This device is expected to be operated in salt water next year. A magnetic
tape recorder is installed in both inspection pigs and is used to store the
electromagnetic data. Data tapes are then reduced and analyzed after the
pigs are run through the pipeline.
Capability
Magnetic flux inspection pigs are the most widely used type of inspec-
tion pigs. The most important capabilities of the device is that it can be
used to measure the severity of corrosion. It can also be used to inspect
for a variety of pipeline defects including hardspots, manufacturing defects
and flaws, girth welds, gouges, pits, etc. Additionally, the device can be
used to help evaluate the effectiveness of the cathodic protection system.
Sensitivity
The sensitivity of the magnetic flux inspection pig is quite good for
corrosion or pitting. Typically it is graded in three ranges of corrosion
severity: 15-30 percent of nominal wall; 30-50 percent of nominal wall;
greater than 50 percent of nominal wall. Defects as small as 1/8 are claimed
as detectable by the manufacturers.
Manufacturer .and Costs
The two main manufacturers of these magnetic flux inspection pigs are
AMF Tuboscope, Inc. and Vetco Pipeline Services. A similar type of inspec-
tion pig is currently under development in Canada.
These devices are usually provided as an inspection service that in-
cludes inspection pig, operating personnel and data analysis. A typical
cost for inspection of 20 miles of 36" pipeline is about $20K.
450
-------
Advantages
Advantages include the following:
• High reliability
• Locates defects
• Permanent record
• Monitors integrity of line
• Locates potential failures before they reach catastrophic
failure
• Helps evaluate effectiveness of cathodic protection
• Commercially available.
Disadvantages and Limitations
For maximum effectiveness, frequent inspection, once or twice a year,
are desireable. Hence, high inspection costs normally limit the number of
inspections.
Data records are difficult to interpret and require human interpretation.
It is also possible to have anomalies that are difficult to interpret. For
example, a weld that penetrates into pipe may result in cavitation downstream
causing corrosion and erosion. The girth weld would be picked up but the
adjacent anomalies may not show up because the operator may be monitoring
only one pulse from the device in that area.
The electromagnetic type cannot determine if a defect is in the inside
or outisde of a pipe. The permanent magnet type can get stuck in a pipeline
and is difficult to remove without cutting out a section of the pipeline.
Both devices do not adequately detect thin cracks.
E.9 SURVEY-PIPELINE SYSTEM CHARTING AND DEPTH OF BURIAL
E.9.1 Depth of Cover by Sonar or Other Comparable Techniques
(See Table 54(b) and (c) and Sections 5.2.2.9 and 7.3.1.9.)
E.9.2 Charting of Line Pipe
(See Table 54(e) and (f) and Sections 5.2.2.9 and 7.3.1.9.)
E.10 MISCELLANEOUS
E.ll.l Preventive Program for Outside Forces
(See Reference 12.)
E.11.2 Passive Acoustic Array
(See Section 5.3.2.10.)
451
-------
TECHNICAL REPORT DATA
(Please read Inunctions on the reverse before completing)
1. REPORT NO.
-&c
-cyo
2.
3. RECIPIENT'S ACCESSIONING.
4. TITLH AND SUBTITLE
Petroleum Pipeline Leak Detection Study
5. REPORT DATE
January 1982
6. PERFORMING ORGANIZATION CODE
7. AUTHOR(S)
8. PERFORMING ORGANIZATION REPORT NO.
John R. Mastandrea
9. PERFORMING ORGANIZATION NAME AND ADDRESS
Science Applications, Inc.
101 Continental Blvd.
El Segundo, CA 90245
10. PROGRAM ELEMENT NO.
1NE 826
11. CONTRACT/GRANT NO.
68-03-2532
12. SPONSORING AGENCY NAME ANO ADDRESS
Municipal Environmental Research Laboratory- Cin., OH
Jffice of Research and Development
J.S. Environmental Protection Agency
Cincinnati, OH 45268
13. TYPE OF REPORT AND PERIOD COVERED
Final
14. SPONSORING AGENCY CODE
EPA/600/14
15. SUPPLEMENTARY NOTES
J. S. Dorrler, R. A. Griffiths: Project Officers (201-321-6629)
16. ABSTRACT
This study reviews and analyzes the U.S. petroleum pipeline system, accidental
spills, and spill prevention programs. It concludes that improved pipeline spill
prevention measures are needed, because pipeline systems are aging, population
densities near lines are increasing dramatically, and new lines are expected to be
larger and longer.
An approach to developing a spill prevention program is presented. Then,
recommended spill prevention programs for individual pipeline systems are
described. These programs consist of scheduled inspection and/or leak detection
methods and are shown to be capable of significantly reducing the risk of serious
spills. Practical considerations such as the cost of inspections and spillage are
also included.
This effort involved two tasks: (1) proper assessment of the risk, and (2)
selection of optimum prevention measures. Thrs report solves two problems that
formerly prevented the accomplishment of these tasks. First, the contribution of
pipeline age, dimension, etc., to the overall risk could not be assessed, primarily
because of limitations in reported pipeline leak data and lack of a comprehensive
analysis measures for reducing the risk (frequency, volume, and the combined
frequency and volume of spills) was not previously available.
7.
KEY WORDS ANO DOCUMENT ANALYSIS
DESCRIPTORS
b.!OENTIFIERS/OPEN ENDED TERMS C. COSATI Field/Group
'etroleum pipelines
'ipeline transportation
Leak detectors
Pipeline risk analysis
Leak detection
3. DISTRIBUTION STATEMENT
RELEASE TO PUBLIC
19. SECURITY CLASS (ThisReport/
UNCLASSIFIED
21. NO. OF PAGES
472
20. SECURITY CLASS (This pafe)
UNCLASSIFIED
22. PRICE
EPA Form 2220-1 (9-73)
452
------- |