8 HAZARD TO LIFE ASSESSMENT
8.1 INTRODUCTION
This section of the EIA Study Report presents the risk assessment for the proposed Gas Receiver Station at the Tai Po Gas Production Plant and the proposed twin gas pipelines from the proposed LNG receiving terminal at Cheng Tou Jiao, Shenzhen to the production plant.
The objective of the risk assessment is to determine whether there are any insurmountable risk issues associated with all aspects of the construction and operation of the proposed facilities. It should be noted that the risk assessment is concerned with the risk of fatal injury to the public, i.e. any incident that could have fatal effects at and beyond the site boundary. The risk to workers at the Tai Po Gas Production Plant is excluded from the scope of the assessment and is addressed under other statutory requirements (e.g. Factories and Industrial Undertakings Ordinance 1955 and Occupational Safety and Health Ordinance 1997).
The scope of the risk assessment is set down in the EIA Study Brief as follows:
* Description and review of the design of the Gas Receiver Station and the gas pipelines;
* Identification of all the possible hazards associated with these installations and evaluation of the potential safety impacts to the public and the environment, in particular the potential risks to the users of the Tai Po Waterfront Park, the other uses in the Tai Po Industrial Estate and the marine traffic activities within Tolo Harbour, Tolo Channel and Mirs Bay. Specific consideration shall be taken at locations under the dangerous goods anchorage area as indicated in the Project Profile;
* Recommendations on appropriate risk mitigation measures.
8.2 APPLICABLE REGULATIONS
The project is governed by the following key legislation:
* Gas Safety Ordinance, Gas Safety (Registration of Gas Supply Companies) Regulations, Gas Safety (Supply) Regulations 1991
* Factories and Industrial Undertakings Ordinance 1955
* Occupational Safety and Health Ordinance 1997
* Hong Kong Planning Standards and Guidelines, Chapter 11
8.3 APPROACH AND METHODOLOGY
The methodology for the risk assessment comprised the following key steps:
* Hazard Identification: The first step in the risk assessment involves identifying the hazards associated with the project that may lead to a major accident such as fire, explosion or the release of a dangerous substance;
* Frequency Analysis: The next step involves assessing the likelihood of occurrence of the hazards identified above;
* Consequence Analysis: This stage of the risk assessment involves determining the potential impact of hazardous events on the public;
* Risk Evaluation: The results of the frequency and consequence analyses are qualitatively integrated at this stage to determine the risk to the public from the proposed facilities; and
* Risk Mitigation: Based on the results of the risk evaluation, risk mitigation measures are recommended, as necessary.
8.4 DESIGN DESCRIPTION AND REVIEW
8.4.1 Gas Receiver Station
The Gas Receiver Station (GRS) is located within the existing Towngas Gas Plant compound. The main function is to provide pigging facilities for the offshore pipeline and a metering /pressure reduction terminal for the gas. The GRS will be designed to receive pigs for cleaning, testing and routine maintenance of the pipelines.
The facilities in the GRS comprise the following:
Pig Receivers: A pig receiver installed on the pipelines allows pigging operations to be carried out. A valve interlock system will be provided to ensure that the opening of pig trap during the pigging operation is undertaken in a safe manner. During operations of the GRS, pigging will be carried out approximately once in 10 years; otherwise, the pig receiver will be isolated from the pipelines in standby mode.
Metering and Pressure Reduction Facilities: The main facilities of the GRS are the metering and pressure reduction facilities. Filters are provided upstream of the metering and pressure reduction facilities to ensure the removal of any solid particles.
Safeguarding Facilities: Safeguarding facilities are strictly implemented to protect the GRS facilities and personnel against hazard conditions initiated by fires or explosions due to process malfunctions. The GRS is provided with Emergency Shut Down Valve (ESD), for isolating the GRS in case hazardous conditions arise. Control system related to ESD and metering field instruments, etc., is centrally monitored in a dedicated and 24-hr manned control room.
Overpressure Protection: Overpressure protection by means of slam shut valves is provided to safeguard the pressure reduction station against the failure of the pressure control valves. The quick acting shut off of the gas supply has the advantage of not requiring flaring of a large quantity of gas.
8.4.2 Onshore Gas Pipelines
The design details for the onshore section of the pipelines are not finalised. It is, however, assumed that the pipeline will be buried with a minimum 1.1m soil cover and will have a design factor compliant with IGE/TD/1 and not exceeding 0.5. The pipelines will be protected from corrosion using an impressed current system or magnesium-based sacrificial anodes. The design of the system will be subject to prior approval from the GSO before construction is commenced.
The following route options are proposed for the onshore section of the pipelines, as shown in Figure 2.2b.
* Option 1, which lies along the waterfront. This has been developed as a water front promenade/park with a cycling track; and
* Option 2 lies along public roads and is located in an industrial area with buildings on either side. The site with the highest population is the Motorola factory with 1,700 persons. The vacant site opposite the Motorola factory is proposed to be developed and will have population of about 2,000. The two public roads have very little regular traffic and are about 15m wide. This option has been analysed because a final decision concerning which option is to be used has yet to be made by the relevant parties.
8.4.3 Subsea Gas Pipelines
The proposed route for the subsea pipelines is shown in Section 3. The subsea section will consist of twin 18'' OD (450 mm outer diameter), API 5L Grade X60 pipelines with a pipe wall thickness of 11.1 mm. The pipelines are designed as a bundle and will be installed in a common trench at a nominal separation of 857 mm centre to centre. Most of the pipelines will be jetted to 3m below the seabed. Full details of the pipeline design are presented in Section 3 of this EIA; the routing of the subsea pipelines is shown in Figure 2.4a.
8.5 HAZARD IDENTIFICATION
The potential major accident hazards arising from the proposed facilities are mainly associated with the loss of containment of natural gas at the GRS or pipelines. A loss of containment event is defined as a significant accidental release of hazardous material beyond its normal containment barriers. This could occur as a result of failure of major equipment or pipework. The hazardous properties of natural gas and the various causes of loss of containment are discussed below.
8.5.1 Hazardous Properties of Natural Gas
The natural gas to be transmitted by the pipelines predominantly contains methane (89.55 - 96.33 mol%). Other components of the gas include ethane (1.93 - 5.63%), propane (0.33 - 3.23%) and butane (0.15 - 1.41%). The gas is flammable and lighter than air (buoyant). Since the gas does not contain components that induce corrosion, such as water/moisture, carbon dioxide and hydrogen sulphide, no internal corrosion is expected. The chemical properties of natural gas are summarised in Table 8.1.
Table 8.1 Properties of Natural Gas
|
Property |
Value |
|
Synonyms |
Methane |
|
State |
Gas |
|
Molecular Wt |
16.6 – 18.4 |
|
Density (kg/m3) |
0.55 (at atmospheric conditions) |
|
Boiling Point (°C) |
-161.5 |
|
Flash Point (°C) |
Not pertinent |
|
Flammable Limits (%) |
5 - 15 |
|
Auto-ignition Temperature (°C) |
540 |
8.5.2 Gas Receiver Station
Loss of containment from the GRS may occur due to the following:
* Spontaneous pipework failure, i.e. failure of flange joints, gaskets, instrument connections or failure in the pipe body resulting from corrosion, wear and tear, misconnection, material or weld failure, etc;
* Operational/maintenance error during receiving operations, e.g. opening of the pig trap while it is still pressurised or failure to isolate the receiver prior to maintenance;
* External events, e.g. dropped object due to nearby construction activities or fire from the production plant; and
* Natural hazards, e.g. subsidence or lightning strike.
8.5.3 Onshore Gas Pipelines
The principal causes of a loss of containment from an onshore pipeline are as follows.
* Third party interference due to work on other underground utilities, drilling for ground sampling, construction work on adjoining areas, etc;
* Corrosion. Internal corrosion is not expected since the gas does not contain components that induce corrosion, such as water/moisture, carbon dioxide and hydrogen sulphide. However, external corrosion could occur due to failure or breakdown of the corrosion protection system;
* Material or construction defects;
* Improper operation of the pipeline system; and
* Natural hazards, e.g. flooding, subsidence or earthquake.
Failure in the pipeline could manifest as a pinhole leak (about 10mm in diameter), hole (about 25 to 75mm diameter) or a rupture (a complete severance of the line or a failure extending to several metres of the pipe). The nature of failure depends on the causative factor. Pinhole leaks usually result from corrosion, holes generally result from third party damage (TPD), while ruptures may be caused by weld failure, material failure or natural hazards such as subsidence.
8.5.4 Subsea Gas Pipelines
The main hazard associated with a subsea pipeline is loss of containment resulting in a gas release, which could be ignited by a vessel in the vicinity. The resultant fire could engulf the vessel and cause fatalities to the crew and passengers on board, either directly due to the fire or due to the vessel sinking (in the worst case).
The principal causes for loss of containment from a subsea pipeline are as follows:
* TPD due to anchor drop/drag, vessel sinking/grounding, objects dropped by passing vessels (e.g. construction tubulars and shipping containers);
* Corrosion. Internal corrosion is not expected for the proposed pipelines; however, external corrosion could occur due to failure or breakdown of the corrosion protection system;
* Mechanical failure, including material defect, weld failure, etc;
* Construction related damage, e.g. from failure of equipment on the pipeline laying barge, dropped/dragged anchors, dropped objects, pipe buckle and soil breakdown leading to mispositioning of the pipeline; and
* Natural hazards, such as subsidence, earthquake and typhoon.
As for the onshore pipelines, failure in the subsea pipelines could manifest as a pinhole leak, hole or a rupture.
8.6 FREQUENCY ANALYSIS
8.6.1 Gas Receiver Station
The historical frequency of releases from various equipment items can be obtained from the Hydrocarbon Releases (HCR) Database (1) maintained by the UK Health and Safety Executive (HSE). The HCR Database contains data on offshore hydrocarbon release incidents that are reported to the Offshore Division (OSD), Hazardous Industries Directorate of the HSE and covers the period 1 October 1992 to 31 March 2001. The frequencies of releases from pig receivers recorded in the HCR Database are shown in Table 8.2.
Table 8.2 Frequencies of Releases from Pig Receivers from HCR Database
|
Diameter (in) |
Release Frequency
(per receiver year) |
|
>16 |
9.93 ´ 10-3 |
|
12 < D £ 16 |
8.37 ´ 10-3 |
|
8 < D £ 12 |
5.49 ´ 10-3 |
|
£ 8 |
- (a) |
|
(a) No leaks reported. |
|
8.6.2 Onshore Gas Pipelines
The European Gas Pipeline Incident Data Group (2) (EGIG) provides failure data collected by a group of eight major gas transmission system operators in Western Europe, for onshore natural gas pipelines with a design pressure of greater than 15 barg. Pipelines operated by natural gas transmission companies in the Netherlands, Belgium, France, Germany, Italy and the UK are included.
The EGIG database for the period 1970 to 1997 includes 945 incidents for a total exposure of 1.98 x 106 km-years. The major causes of incidents are given in Table 8.3. The EGIG data suggest that TPD followed by material or construction defects are the main contributors to onshore pipeline failures. It is considered that material or construction defects will likely be mitigated by quality control procedures during design and hydrotesting during construction. The likelihood of a loss of containment from the onshore pipelines can therefore be assessed primarily based on the potential for external or TPD along the pipeline route.
Table 8.3 Breakdown of EGIG Incidents by Cause (1970 – 1997)
|
Cause |
Proportion of
Failures |
|
External interference |
50% |
|
Corrosion |
15% |
|
Construction defects/material failure |
18% |
|
Ground movement |
6% |
|
Hot tap by error |
5% |
|
Others |
6% |
8.6.3 Subsea Gas Pipelines
There are a number of well-established international failure databases for gas pipelines. The most comprehensive database on offshore gas pipeline failures is available in the report published by UK HSE entitled 'PARLOC 96'(3). The most recent version of this database covers incidents from the 1960s until 1995. The information in this database is based on data obtained from regulatory authorities in the UK, Norway, the Netherlands, Denmark and Germany, operators in the UK, Dutch and Danish sectors of the North Sea and published sources.
The main causes of pipeline failure, as identified from a review of the PARLOC 96 data, are listed in Table 8.4. The PARLOC data suggest that anchor/impact followed by internal corrosion are the main contributors to subsea pipeline failures. Since internal corrosion is considered to be negligible for the pipelines, the likelihood of a loss of containment from the subsea pipelines can be assessed primarily based on the potential for external or TPD along the pipeline route.
Table 8.4 Causes of Subsea Pipeline Incidents from PARLOC 96
|
Cause |
Platform Safety
Zone (a) |
Subsea Well Safety
Zone (a) |
Mid-line (b) |
Total |
|
Anchor/Impact |
7 (47%) |
- |
7 (33%) |
14 (33%) |
|
Corrosion (internal) |
3 (20%) |
4 (67%) |
6 (29%) |
13 (31%) |
|
Corrosion (external & other causes) |
1 (7%) |
- |
2 (9.5%) |
3 (7%) |
|
Material defect |
2 (13%) |
1 (16.5%) |
2 (9.5%) |
5 (12%) |
|
Others |
2 (13%) |
1 (16.5%) |
4 (19%) |
7 (17%) |
|
Total |
15 |
6 |
21 |
42 |
|
(a) Platform and subsea well safety zones refer to pipelines located within 500m of an offshore platform and subsea well respectively. (b) Mid-line refers to pipelines located more than 500m from a platform or subsea well. |
||||
8.7 CONSEQUENCE ANALYSIS
8.7.1 Gas Receiver Station
A release from the GRS would take the form of a high pressure jet. If ignited immediately, an intense momentum jet fire (also called torch fire) would occur. Combustion in a jet fire occurs in the form of a strong turbulent diffusion flame that is strongly influenced by the initial momentum of the release. If oriented horizontally, the jet could cut through adjoining structures and equipment and cause escalation due to secondary failures.
If the release is not ignited immediately, it will disperse with the wind and be diluted as a result of air entrainment. The principal hazard arising from a cloud of dispersing flammable material is its subsequent (delayed) ignition, resulting in a flash fire or Vapour Cloud Explosion (VCE). The latter is only likely to occur in congested or confined areas.
8.7.2 Onshore Gas Pipelines
Immediate ignition of releases caused by a rupture in the onshore pipeline may give rise to a fireball upon ignition. The potential for a fireball in the event of a gas release from a pipeline may be similar to the effects from vessel ruptures. Immediate ignition of smaller releases from the onshore pipelines would result in jet fires. In the event that the release is not ignited immediately, delayed ignition would result in similar outcomes to that for the GRS, ie a flash fire or VCE.
8.7.3 Subsea Gas Pipelines
In the event of a release from the subsea pipeline, the gas jet is expected to lose momentum and bubble to the surface. Upon reaching the sea surface, the released gas will begin to disperse into the atmosphere. The nature of the dispersion, and consequently the hazard, will depend on the momentum of release at the sea surface. At lower depths and high release rates, the gas will have sufficient momentum at the sea surface to result in a momentum jet. In the case of small releases, the gas will lose all its momentum by the time it reaches the surface and will disperse as a buoyant gas.
In order for a fire to occur, the released gas must reach an ignition source while still at a flammable concentration. Unlike onshore releases where some kind of ignition source is always likely to be present, ignition of subsea releases is expected only from marine traffic - either from passing vessels approaching the scene of the incident or from the vessel that caused the damage, which may still be present in the vicinity.
When the dispersed gas encounters an ignition source while within its flammable limits, a jet fire (in the case of a momentum release at the sea surface) or flash fire (in the case of a buoyant gas release at the sea surface) will occur. In the case of a flash fire, the flame front will travel through the flammable region until it reaches the release point of the gas plume where upon a sea surface fire will result. At very low water depths, a fireball may even occur in the event of ignition of a full bore rupture of the pipeline.
8.8 RISK EVALUATION
By reference to the historical data presented in Section 8.6, the likelihood of occurrence of the identified hazards is assessed qualitatively in this section, using the following categories:
* 1 - low likelihood;
* 2 - moderate likelihood; and
* 3 - high likelihood.
The consequences of the hazard were assessed using the following categories:
* 1 - hazard likely to affect no more than a few individuals;
* 2 - hazard likely to affect more than a few individuals; and
* 3 - hazard likely to affect a community.
A simple 3x3 risk matrix, as shown in Figure 8.1, was used to assess the likely levels of risk for the identified hazards.
Figure 8.1 Risk
Assessment Matrix
|
Hazard Frequency |
Hazard Consequence |
||
|
1 |
2 |
3 |
|
|
1 |
Low |
Low |
Moderate |
|
2 |
Low |
Moderate |
High |
|
3 |
Moderate |
High |
High |
8.8.1 Gas Receiver Station
The risk assessment for the GRS is summarised in Table 8.5.
8.8.2 Onshore Gas Pipelines
The risk assessment of the two route options for the onshore section of the pipeline is summarised in Table 8.5. As detailed previously, the likelihood of a loss of containment from the onshore pipelines has been assessed based on the potential for TPD along the pipeline route.
8.8.3 Subsea Gas Pipelines
The risk assessment of various sections of the subsea section of the pipeline is summarised in Table 8.6. As detailed previously, the likelihood of loss of containment from the subsea pipelines has been assessed based on the potential for TPD along the pipeline route.
Table 8.5 Risk Assessment of Gas Receiver Station and Onshore Pipelines
|
Section |
Length (a)
(km) |
Likelihood of
Occurrence |
Consequence to
Public |
Assessed Risk
Level |
Notes |
|
Gas Receiver Station |
- |
2 |
2 |
Moderate |
Likelihood of loss of containment considered to be moderate. Release likely to affect more than a few individuals in the vicinity of
the plant. |
|
Route Option 1 |
1 |
1 |
2 |
Low |
Likelihood of TPD considered to be low due to location along
waterfront. Release likely to affect more than a few individuals at the water front
promenade/park, especially at weekends. |
|
Route Option 2 |
1.56 |
2 |
2 |
Moderate |
Likelihood of TPD considered to be moderate due to location along
public roads. Release likely to affect more than a few individuals at the industrial
buildings in the vicinity of the pipeline route. |
|
(a) Section length is approximate. |
|||||
Table 8.6 Risk Assessment of Subsea Pipelines
|
Section |
Length (a)
(km) |
Likelihood of
Occurrence |
Consequence to
Public |
Assessed Risk
Level |
Notes |
|
Tai Po shore approach |
0.8 |
2 |
2 |
Moderate |
The likelihood of TPD is considered to be moderate. Although the frequency
of anchor damage is likely to be low, the potential for pipeline impact due
to grounding of small vessels is considered to be higher, and thus a moderate
likelihood is given. Release likely to affect more than a few individuals on the shore. |
|
Tolo Harbour |
3.0 |
2 |
2 |
Moderate |
The level of vessel movement in Tolo Harbour and Tolo Channel is
moderate; the likelihood of TPD from anchor impact is therefore considered to
be moderate. Release likely to affect more than a few individuals on vessels in the
vicinity of the release. |
|
Tolo Harbour anchorage zone
|
2.6 |
2 |
2 |
Moderate |
The level of vessel movement in Tolo Harbour and Tolo Channel is
moderate; the likelihood of TPD from anchor impact is therefore considered to
be moderate. Release likely to affect more than a few individuals on vessels in the
vicinity of the release. |
|
|
|
|
|
|
|
|
West Tolo Channel |
4.9 |
2 |
2 |
Moderate |
The level of vessel movement in Tolo Harbour and Tolo Channel is
moderate; the likelihood of TPD from anchor impact is therefore considered to
be moderate. Release likely to affect more than a few individuals on vessels in the
vicinity of the release. |
|
East Tolo Channel |
8.3 |
2 |
2 |
Moderate |
The level of vessel movement in Tolo Harbour and Tolo Channel is
moderate; the likelihood of TPD from anchor impact is therefore considered to
be moderate. Release likely to affect more than a few individuals on vessels in the
vicinity of the release. |
|
Tolo mouth |
1.8 |
2 |
2 |
Moderate |
The level of vessel movement in the Tolo mouth section is moderate; the
likelihood of TPD from anchor impact is therefore considered to be moderate. Release likely to affect more than a few individuals on vessels in the
vicinity of the release. |
|
Mirs Bay W |
2.3 |
2 |
2 |
Moderate |
The likelihood of TPD is considered to be moderate due to potential for
anchor impact from DG vessels. Release likely to affect more than a few individuals on anchored DG
vessels. |
|
Mirs Bay E |
3.2 |
1 |
1 |
Low |
The level of vessel movement in this section of the pipeline is low;
the likelihood of TPD from anchor impact is therefore considered to be low. Since vessel movement in this section of the pipeline is low, it is
considered that the release is likely to affect no more than a few
individuals on vessels that may be present in the vicinity of the release. |
|
North Ping Chau |
2.7 |
1 |
1 |
Low |
The level of vessel movement in this section of the pipeline is low;
the likelihood of TPD from anchor impact is therefore considered to be low. Since vessel movement in this section of the pipeline is low, it is
considered that the release is likely to affect no more than a few
individuals on vessels that may be present in the vicinity of the release. |
|
Shenzhen shore approach |
1.9 |
2 |
2 |
Moderate |
The likelihood of TPD is considered to be moderate. Although the
frequency of anchor damage is likely to be lower, the potential for pipeline
impact due to grounding of small vessels is considered to be higher. Release likely to affect more than a few individuals on the shore. |
|
(a) Section length is approximate. |
|||||
8.9 RISK MITIGATION
For the purposes of this study, risk mitigation has been evaluated based on the risk levels assessed in the preceding section, as follows:
* High risk: Further risk mitigation measures are considered necessary to reduce the risk associated with the particular section;
* Moderate risk: Further risk mitigation measures should be considered to reduce the risk associated with the particular section to as low as reasonably practicable; and
* Low risk: Further risk mitigation measures are not considered necessary since the risk associated with the particular section is broadly acceptable.
No high risk areas were identified from the risk evaluation undertaken in Section 8.8. The risks associated with Route Option 1 of the onshore pipelines and the Mirs Bay E and North Ping Chau sections of the subsea pipelines were assessed to be low. Therefore, it is considered that no further risk mitigation measures are required for these areas.
Further risk mitigation measures should be considered for the following sections, for which moderate risk levels were evaluated:
* Gas Receiver Station;
* Route Option 2 for the onshore pipelines; and
* All sections of the subsea pipelines, except for the Mirs Bay E and North Ping Chau sections.
Potential risk mitigation measures are summarised below:
* Gas Receiver Station:
* Implementation of well defined operational and maintenance procedures for receiving operations; and
* Liaison with external parties (e.g. Fire Services Department and adjoining facilities) to ensure that an off-site emergency plan is in place in the event of an incident at the Gas Receiver Station.
* Route Option 2 for the onshore pipelines:
* Ensure that the location of the pipelines is clearly marked on relevant drawings to limit the potential for damage due to excavation works; and
* Liaison with external parties (e.g. Fire Services Department and adjoining facilities) to ensure that an off-site emergency plan is in place in the event of a pipeline incident.
* Certain sections of the subsea pipelines:
* Ensure that the location of the pipelines is clearly marked on admiralty charts to limit the potential for damage due to anchor impact; and
* Provision of additional pipeline protection, i.e. rock armour. The practicability of additional pipeline protection should has been subjected to a detailed risk assessment.
Before construction works on the submarine pipelines, onshore pipelines and GRS can commence, approval to construct is required from the Gas Standards Office of the Electrical and Mechanical Services Department (EMSD). This approval will be given once EMSD are satisfied with the risk levels posed by the facilities and any necessary emergency response plans. This information will be provided to EMSD before construction commences in the form of Quantitative Risk Assessment reports.
8.10 CONCLUSIONS
A risk assessment study of the proposed gas receiver facility at the Tai Po Gas Production Plant and associated twin gas pipelines was conducted.
The study considered releases that may occur as a result of loss of containment from the proposed facilities, as follows:
* external impacts such as dropped objects and anchor drop/drag;
* spontaneous failures, e.g. due to corrosion or material defect; and
* natural hazards, e.g. earthquakes and subsidence.
With reference to historical data, the likelihood of loss of containment from the proposed facilities was qualitatively assessed. The consequences of the resultant leak include jet fires, sea surface 'pool' fires and flash fires. Based on the frequency and consequence assessments undertaken, the likely levels of risk from various sections of the proposed facilities were determined using a risk matrix. The results of the assessment are as follows:
* No high risk areas were identified;
* The risks associated with Route Option 1 of the onshore pipelines and the Mirs Bay E and North Ping Chau sections of the subsea pipelines were assessed to be low; and
* Moderate risk levels were determined for the following areas:
* Gas Receiver Station;
* Route Option 2 for the onshore pipelines; and
* All sections of the subsea pipelines, except for the Mirs Bay E and North Ping Chau sections.
From the qualitative risk assessment undertaken of the proposed gas receiver facility at the Tai Po Gas Production Plant and twin gas pipelines, it is concluded that there are no insurmountable risk issues associated with the project. Mitigation measures to further reduce the risk levels associated with the project have been recommended for consideration, as detailed in Section 8.9.
_____________________
(1) Hydrocarbon Releases Database, http://www.hse.gov.uk/hid/osd/hsr2001/contents.htm
(2) European Gas Pipeline Incident Data Group: Gas Pipeline Incidents. 4th Report 1970-1998, 1999.
(3) Health and Safety Executive, UK, PARLOC 96: The Update of Loss of Containment Data for Offshore Pipelines.