BSPL-1520-LCL1-SS-REP-2014 (C1) Quantitative Risk Assessment (QRA) Study Report
BSPL-1520-LCL1-SS-REP-2014 (C1) Quantitative Risk Assessment (QRA) Study Report
19th Floor, Sunway Tower 2, No 86, Jalan Ampang, 50450 Kuala Lumpur, Malaysia
Telephone: +60 3 2032 2488
Facsimile: +60 3 2032 2488
www.worleyparsons.com
CONTENTS
1. EXECUTIVE SUMMARY ..................................................................................................6
1.1 ABBREVIATIONS.............................................................................................................7
2. INTRODUCTION ..............................................................................................................8
3. OBJECTIVES .................................................................................................................10
7. HAZARD IDENTIFICATION............................................................................................19
8. FREQUENCY ASSESSMENT........................................................................................24
8.3 Leak Frequency for Operating Pipelines and Flexible Hose ...........................................25
9.2 Consequences................................................................................................................31
Appendices
APPENDIX 1 - ASSUMPTIONS REGISTER
1. EXECUTIVE SUMMARY
Cairn Energy India Pty. Ltd (CEIL) is planning to install twin 24-inch downstream pipelines as part of
the marine loading facilities. The proposed facility will include pipelines from onshore Bhogat COT,
offshore pipelines, a Pipeline End Manifold (PLEM), connecting subsea hoses, a Single Point
Mooring (SPM), floating hoses and reject water pipeline. The pipelines will be used to load tankers
with RJ Crude oil from COT once a week. As part of this project, a Quantitative Risk Assessment
(QRA) has been conducted for the subsea pipelines including the marine facilities by
WorleyParsons Services, Kuala Lumpur.
The study analyses threats for the facility damage, frequency assessment based on the historic
databases and consequence modelling based on the material handled. The frequency assessment
utilizes PARLOC (Ref: 2) and Offshore Hydrocarbon Release Statistics (Ref: 3) databases to
complete a generic estimate of the leak frequency for the proposed pipeline and marine facilities
design.
The frequency analysis based on data provided in PARLOC (Ref: 2) finds that the dominant cause
contributing to pipeline damages are internal and external corrosion. This is followed by other
impacts from dropped objects / third party damages. The estimated failure frequencies due to
corrosion and impact are 3.85 x 10-3 per year and 2.82 x 10-3 per year respectively.
Pool fire is basically a fire burning above a pool of vaporizing hydrocarbon fuel. Since the RJ Crude
mainly consists of heavy hydrocarbons, potential spillage is not expected to result in pool fire if
release offshore (on water) and onshore leak from buried pipeline. Based on the RJ Crude
properties, on accidental release to atmosphere it will not result in to pool fire. Part of the loading
operation involves handling of LFO. In absence of Company specified LFO composition, a typical
LFO was used for the study purpose Based on this composition of LFO, accidental LFO release
may cause flash / pool fire. For small leaks flash/pool fire may be unlikely.
Although leaks due to fullbore release might result in a large sea pool and potential sea pool fire of
-7
LFO, the frequency of sea pool fire occurring is very low of the order of 1.73 x 10 per year. The
-6
total frequency for flash fire is also very low occurring at 1.56 x 10 per year.
Referring to the Risk Assessment Matrix (Section 6.2) and low probability of habitation along the
onshore and offshore facility, the overall fatality risk from the proposed pipeline facilities is very low
and considered tolerable.
In order to maintain the current level of risk and further improve the pipeline systems safety levels,
the following recommendations are made:
1. The risk of flash fire and subsequent pool fire can be minimised by using diesel or equivalent
Light Flushing Oil composition with very low light ends. This will help lowering the risk further
and make operation inherently safer.
2. The exclusion zone around the SPM should be conspicuous by installing buoys around SPM.
3. Emergency response plan should be developed based on the consequences determined in this
report clearly identifying resources, personnel and communication requirements.
4. Hold back tugs which may required for the loading operations can be provided with means of
fire fighting depending on the quality of LFO being utilised for the operation. The fire fighting
provision should be in line with NFPA 1925: Standard on Marine Fire-Fighting Vessels.
1.1 ABBREVIATIONS
API American Petroleum Institute
CEIL Cairn Energy India Pty. Ltd
COT Crude Oil Tankfarm
CWC Concrete Weight Coating
FB Fullbore
FBE Fusion Bonded Epoxy
HAZOP Hazard Operability
2. INTRODUCTION
Cairn Energy India Pty. Ltd. (‘CEIL’) is the operator of block RJ-ON-90/1 (the ‘Block’) in the State of
Rajasthan, India, and operates the Block on behalf of itself and its Joint Venture (JV) Partners Cairn
Energy Hydrocarbons Limited and Oil and Natural Gas Corporation. The Block contains a number
of major oil discoveries, including the Mangala field, the largest onshore oil discovery in India since
1985. CEIL and its JV Partners are currently in the process of implementing a pipeline to transport
crude oil from the Block at Barmer, Rajasthan to a costal terminal facility.
The pipeline project involves an Oil Evacuation Pipeline and Gas Pipeline of approximately 675 km,
pumping / export terminals and related facilities (such as pumping stations, blending facilities,
pigging stations, metering skids, etc.) for transportation of at least 150, 000 barrels per day of crude
oil from the upstream process terminal at the Mangala field to the downstream terminal point as
designated.
The Marine Facilities for Bhogat Terminal shall comprise of pigging facilities at Crude Oil Tankfarm
(COT) located off Bhogat, 24” twin loading buried pipelines (onshore 8.0km and offshore 6km),
Pipeline End Manifold (PLEM), Single Point Mooring (SPM) System and other associated
components for crude oil loading from Bhogat Terminal to the export oil tankers;
recirculation/flushing system using Light Flushing Oil (LFO); and LFO import through SPM system
from marine tankers to Bhogat Terminal.
In order to support the operations at Bhogat COMPANY proposes to abstract saline ground water
from the aquifer and treat the water for further consumption. The rejects of the water treatment
system (Desalination Plant) will be discharged through a marine reject water pipeline system into an
outfall diffuser to the Arabian Sea.
INTECSEA has been appointed by Leighton Contractors India Pvt. Ltd. (‘LCL1’) as the design
engineering consultants to execute detailed engineering design of the twin 24-inch pipelines and a
reject water pipeline for the Mangala Development Pipeline Project. As part of the scope of work,
Worley Parsons is required to perform a Pipeline Quantitative Risk Assessment (QRA).
3. OBJECTIVES
This report presents the methodology and results of the onshore pipeline, subsea pipeline and
PLEM/SPM Quantitative Risk Assessment (QRA).
The objective of the study is to assess the risks associated with the onshore, subsea pipeline
including PLEM and SPM facilities during the operational phase. The following task will be carried
out for the QRA Study:
• Identify the hazards associated with the pipeline facility based on historical data and project
specific hazards;
• Estimate the likelihood of identified scenarios based on historical statistical data available
for pipeline and associated facilities;
• Determine the effects of the consequences associated with loss of containment/ damage to
the pipeline facilities;
• Provide recommendations for control and mitigation measures, if required, based on the
highest risk contributors identified in the risk assessment.
4. SCOPE OF WORK
The scope of work for the QRA covers the onshore pipeline, Land Fall Point (LFP) to offshore
subsea pipeline, PLEM, SPM facilities including floating hoses.
The facilities within the COT are outside the scope of this study. The QRA study assesses Loss of
Containment (LOC) and related safety aspects; environmental aspects are addressed separately.
As part of the Mangala field development, twin 24-inch downstream pipelines are planned to be
installed to complete the proposed marine loading facilities in the Arabian Sea which shall comprise
of a Single Point Mooring (SPM) with a provision of a future SPM, a Pipeline End Manifold (PLEM),
connecting subsea hoses, floating hoses, the twin pipelines and a reject water pipeline. The
pipelines are divided into two (2) main sections i.e. onshore and offshore. The onshore section is
designated for pipelines traversing from the COT at Bhogat Terminal down to the LFP while
offshore section is designated for pipelines from LFP traversing to the subsea PLEM with estimated
distances of 8.0km and 6.0km respectively. The marine facilities scope of work covers from the
onshore hook-up point (at a distance of 600m upstream of the LFP) up to the floating hose system.
The twin pipelines will be electrically heated using Skin Effect Heat Management System (SEHMS)
and fully insulated to maintain the flow integrity of the crude oil by reducing its high viscosity. The
transported product will be sweet crude.
AFRAMAX tanker class with an assumed capacity of 700,000 bbl of oil is envisaged to be
evacuated in 24 hours period including the berthing/de-berthing time.
CEIL also plans to install water treatment plant to support its future project clusters. The plant with a
water intake capacity of approximately 1680 m3/day will discharge approximately 840 m3/day of
reject water offshore off Bhogat via reject water pipeline system. A diffuser system will disperse the
reject water into the surrounding offshore water and be at ambient at some distant away from the
discharge point.
A summary of pipeline characteristics and fluids properties are provided in Table 5.1.
Table 5.1: Pipeline Characteristics [Ref: 1]
The onshore and offshore pipelines are designed in accordance with DNV-OS-F101 standard
whereby the Location Class for Submarine Pipelines are classified as Zone 1. Near Shore and
PLEM area are classified as Zone 2 while Onshore Pipelines are classified as Zone 3.
Pipelines design takes into account earthquake loads as appropriate based on the relevant zone
specified in IS Code IS 1893:2002 Criteria for Earthquake resistant Design of Structure.
The onshore and offshore pipeline data are presented in Table 5.2.
Table 5.2 Pipelines Data [Ref: 1]
Offshore pipelines are provided with weight coating for on bottom stability of the pipeline and
mechanical protection due to dropped objects. Fusion Bonded Epoxy is used as corrosion coating
while Polyurethane Foam is used for insulating the pipeline. External coatings details of the
pipelines are presented in Table 5.3.
Table 5.3 Typical Pipelines External Coating [Ref: 1]
Table 5.4 and Table 5.5 present the composition and characteristics of the RJ Crude.
As can be seen from the crude properties, the crude is highly viscous and is expected to form gel at
ambient conditions due to high pour point. There are no components within the crude oil which may
pose toxic hazard.
6. STUDY METHODOLOGY
6.1 Overview
This section describes the methodology used in the pipeline risk assessment. The traditional
quantitative risk assessment methodology was employed to establish the risks. The risk
assessment seeks to address the following questions:
The risk assessment methodology thus generally involves the following steps.
• System definition – this stage involves setting the scope and boundary limits of the study
and is defining the facilities layout, protection systems such that they can be used to
develop the risk model;
• Hazard identification – this is a key step in risk analysis. It is an important step in safety
studies. It involves identifying the various causes or events that have the potential to cause
a Major Accident Event e.g. gas release, fire, explosion, etc. Each cause/ event is then
developed into various possible outcomes;
• Consequence modelling – once the hazards have been identified, consequence modelling
involves the determination of the effects of such hazards happening. Estimation of
consequences is typically based on software modelling of the outcome events of the Major
Accident Event. However, consequences may be estimated qualitatively based on the
events identified.
• Risk integration – the frequencies and consequences of the various outcomes of the
identified hazard scenarios are integrated at this stage to determine the risk levels;
• Risk evaluation – risk criteria are used to assess the significance of the risks estimated.
Generally for onshore and offshore fixed hazardous installations tolerable risk levels are between
1x10-6 per year to upper limit of 1x10-3 per year, risk level above 1x10-3 per year is not acceptable to
the public outside the site boundary [Ref: 9]. These risk guidelines cannot be applied directly for
transport operations, in this case, pipeline transport. As transport operations extend over several
kilometres, they cannot be compared with risks from fixed installations which have a defined impact
zone.
Further for offshore pipelines, there is no specific individual risk criterion defined and used
worldwide. For the purpose of this assessment, in absence of Company Risk Assessment Matrix
(RAM), the matrix as defined below in Figure 6.1 is utilised to assess the risk.
Probability
A B C D E
Frequent/
SEVERITY People Asset Environment Improbable Unlikely Infrequent Occasional
Likely
1 in 100,000 1 in 10,000 1 in 1000 1 in 100
1 in 10 years
years years years years
Medium Medium High High High Risk
Multiple Widespread damage - Massive
CATASTROPHIC 5 fatalities total loss of production environmental
(10+) (>50MMUSD) damage in large area.
Single Low Medium Medium High High
Major effect to
fatality (1+)
Major damage/ Partial environment,
or
CRITICAL 4 shutdown (10 to extensive measures
permanent
50MMUSD) required for
total
restoration
disability.
Environmental Low Medium Medium Medium High
damage,
Major injury/
contamination
irreversible Major damage in one
MODERATE 3 affecting nearby
health area (2 to 10MMUSD)
facilities,
damage
spontaneous
recovery in long term
Serious damage to Low Low Medium Medium Medium
major component of Local environmental
MARGINAL 2 Minor injury.
equipment (0.5 to 2 damage
MMUSD)
Minor damage to Low Risk Low Low Low Medium
Illness/ First Minor effect to
NEGLIGIBLE 1 equipment
aid. environment
(<0.5MMUSD)
7. HAZARD IDENTIFICATION
Main hazard for the facilities under discussion is pressurised crude oil contained in the pipeline.
This section discusses possible threats / causes which may lead to accidental events i.e. Loss of
Containment and subsequent consequences.
The possible threats or causes leading to a loss of containment event in the pipeline are identified
based on the historical database and summarised below.
The most comprehensive database on pipeline failure data is PARLOC 2001 (Ref: 2) published by
the UK Health and Safety Executive which compiles failure data by incident causes, types, etc. The
latest update covers data up to year 2000, with a total pipeline length of 24,837 km and operating
experience of 328,858 km-years. Information recorded in this database has been obtained from
regulatory authorities concerned with the North Sea pipelines and operators in the UK, Dutch,
Norwegian and Danish sectors. Incidents to operating steel pipelines, flexible steel lines as well as
associated fittings are recorded.
Appendix 2 presents the causes of incidents to operating steel pipelines resulting in loss of
containment as recorded in the PARLOC data.
Though some details about spills in the Indian waters are available (Appendix 3), the type of
incidents, failure causes, leak sizes etc are not known which are required for the risk assessment.
Considering immense operating experience in the PARLOC database it is selected for the current
study. It may be noted that in absence of any site or region specific database, PARLOC provides
acceptable basis for the failure frequency and is utilized in the studies worldwide. For SPM piping
components, Offshore Hydrocarbon Release Statistics database is utilized which provides failure
frequency for various pipework components encountered in offshore oil and gas industry.
7.1 Threats
The generic causes for loss of containment from the pipelines facilities are as follows:
• Other Impacts – Trawl, dropped object, wreck and construction, excavation/third party
damage, potential impact to SPM by tanker collision;
• Others – Design errors whereby incorrect material of construction are used, inadequate
engineering analysis and inadequate specification of operating parameters.
2
Page 19 of 70 BSPL-1520-LCL1-SS-REP-2014 (Rev C1)
CAIRN ENERGY INDIA PTY. LTD.
BHOGAT TERMINAL MARINE FACILITIES QUANTITATIVE RISK ASSESSMENT (QRA) STUDY REPORT
SINGLE POINT MOORING (SPM) LOADING, ONSHORE AND OFFSHORE PIPELINES
MANGALA DEVELOPMENT PIPELINE PROJECT
It is noted that pipeline failures due to design errors or mal-operation will be discussed and
designed out during the HAZOP study. Except dropped object, trawl and dragged anchor, the
threats are also applicable to onshore pipeline section and SPM. These causes for loss of
containment are discussed in the following section:
The pipeline and SPM facilities under study are not located near the port. Only the vessels
approaching SPM for loading / unloading will be moored to the SPM. Furthermore, the entire
offshore pipeline is buried and backfilled. Besides that, 500m around SPM is also treated as a
safety exclusion zone. Thus frequency of anchor drop which can potentially lead to pipeline damage
is considered very low.
CEIL has conducted route analysis studies considering environmental factors. Based on the field
study carried out by NIO, Goa in January 2008, it is reported that the proposed land fall and
terminal location does not contain environmentally sensitive areas such as national parks / marine
parks, sanctuaries, wildlife habitats and mangroves. It also does not include breeding and spawning
grounds of fish and other marine life, areas of outstanding natural beauty/ historically/ heritage
areas. The coastal area of Bhogat to Dwarka is a traditional fishing zone and gill-netting is fairly
active during the fishing season from September to May. [Ref: 4].
Based on this, the project area is not considered as active site using trawl gears for fishing.
Furthermore, the entire offshore pipeline is buried and backfilled. Therefore, frequency of trawl
impact is considered low.
2
Page 20 of 70 BSPL-1520-LCL1-SS-REP-2014 (Rev C1)
CAIRN ENERGY INDIA PTY. LTD.
BHOGAT TERMINAL MARINE FACILITIES QUANTITATIVE RISK ASSESSMENT (QRA) STUDY REPORT
SINGLE POINT MOORING (SPM) LOADING, ONSHORE AND OFFSHORE PIPELINES
MANGALA DEVELOPMENT PIPELINE PROJECT
Since the facility is not in proximity to the port or shipping lane, the frequency of impact by sinking
ship is considered low. The impact frequency in PARLOC includes sinking ship impacts.
The average water depth around the SPM is 30m. Loading of crude would be prohibited in typhoon/
storm conditions. It is expected that the tanker will sail away from the facility and shore in case of
weather not favourable to the operation. Furthermore, the offshore pipeline is buried. Thus, vessel
grounding affecting the pipeline is not considered a credible scenario.
The onshore segment of the pipeline passes through mainly wheat fields or barren land. There are
some road / river / creek crossings in the pipeline route. There are isolated dwellings / buildings
along the pipeline route. No rail crossings are observed across the pipeline. The pipeline will be
buried and protection as per the industry practices will be provided across various crossings.
Pipeline identification marker tape will also be provided to alert and prevent damage by third party.
Periodic line walk would also minimise potential third party activities which may harm health of the
buried pipeline. In addition, it is envisaged that Company will educate the nearby population on the
hazards of the pressurised hydrocarbon pipelines and incident reporting. Hence, although third
party damages are major percentage contributors towards pipeline damages, the absolute risk of
damage by third party to the onshore pipeline is considered low.
2
Page 21 of 70 BSPL-1520-LCL1-SS-REP-2014 (Rev C1)
CAIRN ENERGY INDIA PTY. LTD.
BHOGAT TERMINAL MARINE FACILITIES QUANTITATIVE RISK ASSESSMENT (QRA) STUDY REPORT
SINGLE POINT MOORING (SPM) LOADING, ONSHORE AND OFFSHORE PIPELINES
MANGALA DEVELOPMENT PIPELINE PROJECT
There are no specific frequency data / incidents reports available for such incidents which may also
be interpreted as case occurrences of such incidents.
Environmental loads such as currents and waves due to tropical storms on the pipeline during the
construction phase can compromise the lateral and vertical on-bottom stability of the pipeline on the
seabed. Tropical cyclone records have been inspected for the last 32 years and in general, there
are occasional tropical cyclones at tropical storm stage, however, none of these have passed
directly over Mangala Field [Ref: 10]. The pipeline and marine loading facilities have a design life of
25 years and will be designed to withstand these environmental loads [Ref: 1]. Hence, pipeline and
marine loading facilities damage due to natural hazards is considered negligible and is not
assessed further in this study.
Although the loading operation will not take place during stormy weather condition, there is a
possibility of change in wind and currents which may result into buoy/tanker excursion potentially
overstressing the flexible hoses and hose/MBC failure. For under-buoy hoses this may happen
2
Page 22 of 70 BSPL-1520-LCL1-SS-REP-2014 (Rev C1)
CAIRN ENERGY INDIA PTY. LTD.
BHOGAT TERMINAL MARINE FACILITIES QUANTITATIVE RISK ASSESSMENT (QRA) STUDY REPORT
SINGLE POINT MOORING (SPM) LOADING, ONSHORE AND OFFSHORE PIPELINES
MANGALA DEVELOPMENT PIPELINE PROJECT
irrespective of loading / unloading operations. MBC failure consequences which may be the
outcome of natural /operational hazard is analyzed in the study.
In the worst case such operational errors may result in hose/MBC failure. The consequences of
these events have been assessed in the study.
2
Page 23 of 70 BSPL-1520-LCL1-SS-REP-2014 (Rev C1)
CAIRN ENERGY INDIA PTY. LTD.
BHOGAT TERMINAL MARINE FACILITIES QUANTITATIVE RISK ASSESSMENT (QRA) STUDY REPORT
SINGLE POINT MOORING (SPM) LOADING, ONSHORE AND OFFSHORE PIPELINES
MANGALA DEVELOPMENT PIPELINE PROJECT
8. FREQUENCY ASSESSMENT
8.1 Overview
The generic frequency assessment is carried out by utilising the PARLOC database [Ref: 2] for
pipeline and Offshore Hydrocarbon Release Statistics database [Ref: 3] for SPM to estimate the
overall leak frequency of the pipeline and offshore loading facilities. The PARLOC database is
based on riser and pipeline incidents that have occurred in the North Sea. PARLOC is the most
extensive database for the pipeline incidents.
Based on PARLOC 2001, a total of 542 incidents to risers, pipelines and associated fittings have
been recorded up to the end of year 2000, of which 396 occurred to operating lines. These incidents
are summarised in Table 8.1 below:
Table 8.1: PARLOC 2001 Incidents to Operating Lines
Flexible lines 39 31
Fittings 148 92
In case of the pipelines, there have been 18 incidents involving loss of containment within the safety
zone. One third of the incidents within the safety zone are related to impacts from anchors while
50% of the incidents are related to corrosion and material defect issues. There was one incident
related to an impact whilst the remaining 2 incidents have been recorded as resulting from “other”
issues.
Whilst these historical data provide a basis for establishing the likelihood of a release from the
pipelines, clearly the potential for an incident is related to the activities which are performed close to
the pipelines which may result in damage to the lines. Based on the historical data, the major
concern is from corrosion and material defects. Potential presence of corrosive component in the
crude oil represents a concern from internal corrosion, while salt water environment presents a
threat in terms of external corrosion.
With a total of 65 incidents which have been reported on steel lines resulting in loss of containment,
the total number of incidents has been further expressed in terms of damage based on equivalent
hole diameter as presented in Appendix 2.
2
Page 24 of 70 BSPL-1520-LCL1-SS-REP-2014 (Rev C1)
CAIRN ENERGY INDIA PTY. LTD.
BHOGAT TERMINAL MARINE FACILITIES QUANTITATIVE RISK ASSESSMENT (QRA) STUDY REPORT
SINGLE POINT MOORING (SPM) LOADING, ONSHORE AND OFFSHORE PIPELINES
MANGALA DEVELOPMENT PIPELINE PROJECT
The data for fittings in PARLOC, while substantial, does not include operating experience data to
enable an assessment of leak frequencies. Therefore the PARLOC fittings data has not been used
in this assessment. Instead the contribution of fittings (e.g. flange connection between the spool
piece and the pipelines, valves and piping) has been calculated separately based on data from the
Offshore Hydrocarbon Releases Statistics 2001 as presented in Appendix 2. The fittings failure
frequency is applied to SPM components.
The highest percentage of the leak is categorized under ‘small’ leaks. It can also be seen that the
percentage of rupture is higher then medium leaks.
2
Page 25 of 70 BSPL-1520-LCL1-SS-REP-2014 (Rev C1)
CAIRN ENERGY INDIA PTY. LTD.
BHOGAT TERMINAL MARINE FACILITIES QUANTITATIVE RISK ASSESSMENT (QRA) STUDY REPORT
SINGLE POINT MOORING (SPM) LOADING, ONSHORE AND OFFSHORE PIPELINES
MANGALA DEVELOPMENT PIPELINE PROJECT
1 and 1a Land Fall Point to Subsea Pipeline (up to 1st subsea Actuated Valve)
The isolatable sections are designated as sections in between two actuated valves along the
pipeline. There are a total of 5 actuated valves along the pipeline up to the SPM. Isolatable Section
1 and 2 are duplicated based on two 24 inch pipeline installed parallel.
Table 8.7 presents the leak frequencies based on small, medium, large and full bore leak sizes,
which is based on the historical leak size distribution.
2
Page 26 of 70 BSPL-1520-LCL1-SS-REP-2014 (Rev C1)
CAIRN ENERGY INDIA PTY. LTD.
BHOGAT TERMINAL MARINE FACILITIES QUANTITATIVE RISK ASSESSMENT (QRA) STUDY REPORT SINGLE POINT MOORING (SPM) LOADING, ONSHORE AND
OFFSHORE PIPELINES
MANGALA DEVELOPMENT PIPELINE PROJECT
Pipeline Flexible Hose Piping Flanges Actuated Valve Manual Valve Instruments
No Failure Cases
3" < D 3" < D 3" < D
D< D> D< D> D< D>
<= <= <= D<= 3" < D < D> D< 3" < D < D> D< 3" < D < D>
= 3" 11" = 3" 11" = 3" 11" All Sizes
11" 11" 11" 3" = 11" 11" = 3" = 11" 11" = 3" = 11" 11"
(km) (km) (km) (km) (m) (m)
(km) (km) (m)
Land Fall Point to
Subsea Pipeline (up
1 13.8 1
to 1st Actuated
Valve)
Land Fall Point to
Subsea Pipeline (up
1a 13.8 1
to 1st Actuated
Valve)
PLEM to SPM
2 0.05 3 5 2 2 1 1
(underbuoy hoses)
PLEM to SPM
2a 0.05 3 5 2 2 1 1
(underbuoy hoses)
SPM including
3 0.5 100 19 18 15 3 12
floating hoses
Anchor 2 (7%)
Impact 8 (30%)
Corrosion 11 (41%)
Other 4 (15%)
Total (Pipelines) 27
Based on Table 8.8 the pipeline failure frequency for each failure cause is derived and presented in
Table 8.9.
Table 8.9 Failure Frequency Based on Each Failure Cause
The frequency data in this chapter are utilized further to calculate event outcome frequencies in the
following chapters of the study.
9. CONSEQUENCE ANALYSIS
9.1 Overview
The consequence modelling of pressurised crude oil releases has been conducted by using DNV
software Phast version 6.54. The results from the modelling exercise are used to evaluate the
effects of loss of containment due to leaks in the pipeline.
9.2 Consequences
The nature of failure depends on the causative factor. Small / pinhole leaks usually result from
corrosion, medium size holes generally result from third party damage (TPD), while ruptures may be
caused by dropped anchor, ship impact, weld failure, material failure or natural hazards.
The types of hazardous consequences that could develop following a release are discussed in this
section.
A release of pressurised crude oil (stabilised) from the oil pipeline is expected to occur in one or two
phases:
• Onshore/Sub-sea - Initial release, maximum flow limited to the flow rate of the pipeline (i.e.
driven by the pumps) until the section is isolated or pump is shutdown;
• Sub-sea - Release to equalise the hydrostatic head of the pipeline and the sea; and
Subsequent release of crude oil due to natural buoyancy until “u-bend” section of the
pipeline is emptied. However the release rate due to natural buoyancy would be very low
compared to pumping rates for the liquid.
For onshore pipeline, flow from leak will cease once section is isolated or pump(s) is stopped.
The Light Flushing Oil on release from the buried pipeline may channelize and rise to surface
depending on the soil characteristics. This is possible only when the pumping is ON. Furthermore,
the released oil may form localised oil pool if the pipeline is exposed. The onshore section of the
pipeline will be buried and hence it is more likely to form pockets of oil soaked soil rather than large
oil pool. Large oil pool may form only if leak is undetected for a long period.
Release of RJ Crude from a sub-sea pipeline would result in the lumps of oil rising to the surface
due to waxy nature of the crude. No fire hazards are expected from the release of RJ Crude.
Release of LFO liquid from a sub-sea pipeline would result in the liquid rising to the surface in a
plume. The liquid would form an unconfined pool on the sea surface and the gas fraction would
flash and evaporate from the liquid pool with little or no momentum. The liquid pool is likely to drift
with the current and may start to form an oil slick; if the vapour above the pool were to subsequently
ignite the resulting sea fire may occur some distance from the initial leak point which under
unfavourable wind conditions may lead to a fire surrounding the SPM or tanker. The size of the pool
fire will depend on the amount of inventory released.
Equivalent Hole
< 20 mm 20-80 mm > 80 mm Rupture
Diameter (mm)
Representative
10 mm 50 mm 100 mm 596 mm (pipe ID)
Hole Size (mm)
generating a pool fire. The unconfined pool on sea surface will burn back to a smaller pool diameter
which represents steady state diameter corresponding to leak rate and burning rate. If liquid leak
ceases, the pool fire will diminish with time.
Sub-sea release of RJ Crude will not result in pool fire due to waxy nature of the crude. Heated RJ
Crude release at SPM is not generating vapours or flash fraction and is not expected to ignite.
Release of Light Flushing Oil will result in a flammable cloud being formed due to flashing of the
light ends. If the flammable cloud is ignited, a flash fire followed by pool fire can potentially be
developed.
The pool fires have been modelled for 14 barg and 7 barg release pressure to represent the release
from subsea pipelines and SPM respectively. The modelling also includes assessment on
breakaway couplings whereby the release is assumed to last for 20 seconds before the couplings
closes. The modelling results representing the pool diameter and thermal radiation contours for
LFO releases from the pipeline and SPM are presented in Table 9.3. Modelling results representing
breakaway couplings are presented in Table 9.4.
Based on the modelling result, the largest pool fire diameter of approximately 94m is expected to
form if there is a full bore pipeline failure. Since the breakaway couplings are expected to close in
approximately 20 seconds, the inventory released due to this scenario will be limited, hence not
expected to create a large pool and subsequent potential pool fire. The probability of pool fires
being formed due to small leaks is very low since the quantity of release is small.
Table 9.3 Pool Fire Consequence Modelling Results for SPM and Pipelines
The value of the cone angle is generally given as between 10-12°, with some sources quoting
values of up to 23°. Billeter and Fannelop suggested that the ‘release area’ (where bubbles break
through the surface) has approximately twice the diameter of the bubble plume. Hence, a cone
angle of 23° has been adopted for this assessment.
The water depth for the subsea pipeline is approximately 30m, giving a release area of 12m in
diameter.
When the gas reaches the sea surface, it will begin to disperse into the atmosphere. The gas
buoyancy with respect to air and the momentum of release at the sea surface will affect the nature
of the dispersion. At low release rates and deeper water depths, the release momentum will be
small and the gas dispersion will be governed by its buoyancy within air and the ambient wind
speed. At high release rates and shallower water depths, the momentum of the release becomes
significant compared to gas buoyancy and the dispersed plume will, initially, be more vertical than
the previous case.
LFO with the properties as indicated in the report (Assumption No. 6) would result in gas cloud for
subsea leak, leak at SPM and floating hose as well. The gas dispersion results for the release of
LFO at the SPM and pipeline are presented in Table 9.5. The largest gas cloud that could
potentially be developed due to fullbore release might cover a distance up to 105m. Flash fire is not
expected for small releases due to the low release rate whereby the flammable gas cloud is
expected to be diluted by air upon release.
Table 9.6 presents the gas dispersion results for the release if breakaway couplings fail. Based on
the result, the flammable gas plume is expected to disperse up to a maximum distance of
approximately 72m.
Table 9.5 Flash Fire Consequence Modelling Results for SPM and Pipeline
This section summarises qualitative assessment of the risk from the pipeline facilities followed by
quantitative account of the risk based on data and analysis in the previous chapter.
Table 10.1 catalogues various threats, top events, event outcomes with the existing safeguards.
6 Internal / All Spontaneous failure of Small leak Internal Corrosion Allowance (1.2mm for Likely / Marginal –
External pipeline For onshore: onshore pipeline, Low
corrosion RJ Crude – oil soaked soil 1.5 mm for offshore pipeline)
LFO – oil soaked soil. External Corrosion Coating Protection on the
For offshore: pipeline.
RJ Crude – release of RJ Crude Periodic maintenance / Corrosion monitoring.
lumps
LFO – oil pool
Based on the qualitative analysis presented in Table 10.1, all the risk identified are in LOW or
MEDIUM category whereby none of the risk are in HIGH category.
To determine event outcome frequency for risk quantification, event trees are constructed for a
range of consequence. The basic event tree structure is shown in Figure 10.1.
Immediate
Ignition/
Leak Frequency Ignition Outcome
Delayed
Ignition
Yes
Light Flushing Oil (LFO) or equivalent material will be utilised to flush out the pipeline during RJ
crude loading operations cycles. Hence the worst case consequences are determined based on
pumping of the Light Flushing Oil. In absence of Company supplied properties of the light flushing
oil, a typical Malaysian light crude oil composition (Appendix 1 Assumption no. 6) was utilised for
the study. Depending on the properties of the LFO, flash/pool fire hazards will be encountered.
Based on historic leak frequencies and consequence modelling, the event outcome frequencies as
determined in the preceding sections are very low. Highest contributors to leak frequencies are
corrosion and other impacts. Event frequencies of the flash fire are higher than the pool fire events.
It may be noted that unlike fixed installation probability of habitation along the pipeline installation
onshore and offshore is very low. Therefore the risk which is combination of the frequency and
consequences to people from the proposed pipeline facilities is very low, of the order of 10-6/year
and lower (Table 10.4).
With facility design as per the industry standard codes / standards, compliance to local rules and
regulations, the risk reduction measures are considered As Low As Reasonably Practicable
(ALARP). Comparing the fatality risk against the RAM (Section 5.2), the risk levels from the
proposed Bhogat COT to SPM pipeline loading/unloading facilities are very low and tolerable.
11.1 Conclusion
A Quantitative Risk Assessment (QRA) study is conducted to for the RJ crude oil pipeline facilities
from Bhogat COT up to the PLEM and SPM of Cairn Energy India Pty. Ltd.
• Based on the historic data (PARLOC and Offshore Hydrocarbon Release Statistics), the
dominant causes contributing to pipeline damages are internal and external corrosion. This
is followed by other impacts from dropped objects / third party damages. The estimated
failure frequencies due to corrosion and impact are 3.85 x 10-3 per year and 2.82 x 10-3 per
year respectively.
• Based on the RJ Crude properties on accidental release to atmosphere, it will not result in to
pool fire. Part of the loading operation involves handling of LFO. Based on the LFO
composition utilised, accidental LFO release may cause flash / pool fire. For small leaks
flash/pool fire may be unlikely.
• Although leaks due to fullbore release might result in a large sea pool fire of LFO, the
frequency of sea pool fire occurring is very low of the order of 1.73 x 10-7 per year. The total
frequency for flash fire is also very low occurring at 1.56 x 10-6 per year.
• Referring to the Risk Assessment Matrix (Section 6.2) and low probability of habitation
along the onshore and offshore facility, the overall fatality risk from the proposed pipeline
facilities is very low and considered tolerable.
11.2 Recommendation
For all the Quantitative Risk Assessment studies, the underlying assumption is that the facilities are
designed as per the good engineering practices, operated as per the safe operating procedures by
trained Operating Personnel and maintained by following preventive maintenance practices.
Attention is drawn to the local rules / guideline and it is assumed that Company will follow applicable
rules for this facility especially those listed below.
2. OISD Standards:
In order to maintain the current level of risk and further improve the pipeline systems safety levels,
the following recommendations are made:
1. The study is carried out with assumed LFO composition. The risk of flash fire and subsequent
pool fire can be minimised by using diesel or equivalent Light Flushing Oil composition with
very low light ends. This will help lowering the risk further and make operation inherently
safer. It may be noted that the use of diesel will result in only pool fire hazards that too if
ignited by ignition sources in the close vicinity of the pool.
2. The exclusion zone around the SPM should be conspicuous by installing buoys around SPM.
4. Hold back tugs which may required for the loading operations can be provided with means of
fire fighting depending on the quality of LFO being utilised for the operation. The fire fighting
provision should be in line with NFPA 1925: Standard on Marine Fire-Fighting Vessels.
12. REFERENCES
1. Design Basis Memorandum – Onshore & Offshore Pipelines, Doc No. 402020-00123-MX-BOD-
0001, January 2010.
2. Pipeline and Riser Loss Of Containment Database (PARLOC) 2001, 5th Edition, July 2003.
4. Environmental Impact Assessment for the Proposed SPM and Associated Activities in Arabian
Sea off the Coast of Bhogat, Kalyanpur in Jamnagar District Gujarat, Volume B.
6. A Guide to Quantitative Risk Assessment for Offshore Installations (CMPT), John Spouge.
7. Piping and Instrumentation Diagram Turret Buoy System, Doc No. PE-B-840-DP-0008 001, Rev
B.
10. Mangala Development Pipeline Project Metocean Conditions Study, Report No. L24532, Rev 0,
December 2009.
ASSUMPTION SHEET
Assumption Number 1
Description
LFO and RJ Crude Pump Discharge Pressure
Assumptions
1. The worst case pump discharge pressure considered is approximately 22barg for LFO flushing
operation with pumping rate of 2300 m3/hr [1].
2. Pressure loss from pump discharge to Land Fall Point is assumed to be approximately 5 barg
thus pressure in the pipeline at LFP being 17 barg. With 30m water depth, 3bar hydrostatic
head is assumed to act on the sub-sea pipeline [2].
Reference
1. Steady State Hydraulics Study Report, Doc No. 402020-00123, Rev A1.
ASSUMPTION SHEET
Assumption Number 2
Description
Frequency of Loading Crude into Tanker
Assumptions
1. The frequency of loading the RJ Crude into the tanker is assumed to be once a week with each
loading lasting for one day [1].
2. At the end of loading operation, the pipeline will be flushed with Light Flushing Oil (LFO) to
expel RJ Crude; Pumping of LFO will last for approximately 2 - 4 hours. Pipeline will be filled
with LFO and in non-flowing condition (COT pump shut down) for approximately 6 days [2].
Reference
1. Information from INTECSEA.
ASSUMPTION SHEET
Assumption Number 3
Description
Leak Hole Size Distribution for Fittings
Assumptions
Reference
1. Offshore Hydrocarbon Releases 2001.
ASSUMPTION SHEET
Assumption Number 4
Description
Leak Hole Size Distribution for Pipelines
Assumptions
10mm <20mm
100mm >80mm
Reference
1. PARLOC 2001.
ASSUMPTION SHEET
Assumption Number 5
Description
RJ Crude and Light Flushing Oil Event Outcomes
Assumptions
1. RJ Crude leak at ambient temperature is not envisaged result in potential pool and
subsequent fire.
2. LFO leak onshore is not envisaged to result in liquid pool.
3. For pool fire and flash fire scenarios, LFO is considered for consequence modelling
representing worst case scenario for the pipeline operation handling both RJ Crude
and LFO, though LFO pumping is at fraction of time compared to RJ Crude. LFO will
remain in the pipeline in the blocked condition longer than RJ Crude.
Reference
1. Assumption based on the properties and buried pipeline.
ASSUMPTION SHEET
Assumption Number 6
Description
Light Flushing Oil Composition
Assumptions
1. The composition for LFO is taken as below:
Ethane 0.0215
Propane 0.0314
i-Butane 0.0177
n-Butane 0.0219
i-Pentane 0.0203
n-Pentane 0.0148
n-Hexane 0.1208
n-Heptane 0.0486
n-Octane 0.0828
n-Nonane 0.0676
n-Decane 0.2401
n-C14 0.1423
n-C18 0.0365
n-C24 0.0818
CO2 0.0039
Nitrogen 0.000
H2O 0.006
ASSUMPTION SHEET
Assumption Number 6
C10+ 0.0258
Reference
1. Information from Worley Parsons (typical Malaysian sweet and light crude).
ASSUMPTION SHEET
Assumption Number 7
Description
Event Tree (Ignition Probabilities)
Assumptions
1. It is assumed that for subsea releases, 90% will be delayed ignition, as presence of
continuous ignition sources over the sea are not foreseen [1].
2. The subsea ignition probability based on different release rates are presented below [2].
Small <2 0
Medium 2 - 20 0.001
Reference
1. Assumption by WorleyParsons.
ASSUMPTION SHEET
Assumption Number 8
Description
Time to Detect Leak
Assumptions
The consequence modelling was performed based on pump discharge pressure and shutdown time
when the leak is detected.
Leakages due to small and medium leaks usually are difficult to detect and remains undetected for
long time therefore it is assumed that it can only be detected through visual check and patrolling
during loading / unloading of the crude.
Leakages due to large leaks and pipe rupture can be easily detected through difference in the pump
discharge pressure.
1. The time to detect leak based on different leak sizes are presented below [1]:
10 120
50 60
100 30
Full Bore 5
Reference
1. Assumption by Worley Parsons.
ASSUMPTION SHEET
Assumption Number 9
Description
Impact Criteria for Radiation Contour
Assumptions
The radiation impact criteria towards human are presented below [1]:
2
Radiation Level (kW/m ) Impact Criteria
Reference
1. A Guide to Quantitative Risk Assessment for Offshore Installation (CMPT).
Table 1 Size Of Damage Expressed as an Equivalent Hole Diameter to Operating Steel Pipelines
Pipeline Diameter Range
0 to 9 10 to 16 >16
Location of Incident
Equivalent Hole Diameter (mm) Equivalent Hole Diameter (mm) Equivalent Hole Diameter (mm)
0-20 Rupture 20-80 Rupture >80 Rupture? 0-20 Rupture 20-80 Rupture >80 Rupture 0-20 Rupture 20-80 Rupture >80 Rupture
Platform Total
Piping 1
Splash
1 1 1 1 1 1
Zone
Riser Subsea 1 1
Unknown 1 1 2 1 1 1
Total 2 0 0 0 2 2 4 0 1 0 2 2 0 0 1 0 0 0
Near 5 3 1 1
Safety Far 1 1 1 2 2 1
Zone Unknown 1 2 1
Total 6 3 1 2 1 1 4 3 1
Mid Line Total 14 4 2 1 1 1 3 2 2 2
Subsea
Total 4 1 1 1
Well
Shore Total
Land Total
SPM Total 1
Total 65 26 0 8 3 5 3 7 0 2 0 9 5 4 0 2 0 2 2
Table 2 Failure Rate and Hole Size Distribution for Flanges, Piping and Valves
Failure Rate / Hole Size Distribution (mm)
Equipment Diameter (in.)
Year < 10 10 – 25 25 – 50 50 – 75 75 – 100 > 100 N/A Note 1
-5
D≤3 3.95 x 10 78% 10% 8% 2% 1% 1% 0%
-5
Flanges 3 < D ≤ 11 5.56 x 10 84% 5% 4% 1% 0% 6% 0%
-5
> 11 9.85 x 10 85% 4% 0% 4% 0% 7% 0%
-3
D≤3 1.07 x 10 75% 4% 11% 4% 0% 0% 7%
-4
Valve, Actuated 3 < D ≤ 11 8.00 x 10 74% 0% 16% 0% 0% 11% 0%
-4
> 11 8.44 x 10 100% 0% 0% 0% 0% 0% 0%
-5
D≤3 6.99 x 10 65% 23% 4% 8% 0% 0% 0%
-4
Valve, Manual, Block 3 < D ≤ 11 1.08 x 10 77% 10% 6% 0% 0% 3% 3%
> 11 3.95 x 10-5 82% 0% 18% 0% 0% 0% 0%
-4
D≤3 2.00 x 10 79% 13% 6% 1% 1% 0% 1%
-5
Piping, Steel 3 < D ≤ 11 5.87 x 10 75% 5% 2% 2% 0% 10% 6%
-5
> 11 5.49 x 10 64% 8% 0% 0% 0% 17% 11%
-4
Instruments 5.85 x 10 79% 17% 2% 0% 0% 0% 1%
Unknown 17 11 1 29
Immediate
Leak Ignition/ Event
Ignition Outcome
Frequency Delayed Frequency
Ignition
0.001
SMALL
0.001
MEDIUM
0.001
Y