5                                          Hazard to Life

5.1                                   Introduction

This Section presents the findings of the Hazard to Life Quantitative Risk Assessment (QRA) undertaken for the Project.  The assessment includes an evaluation of the risks associated with storage, transfer, handling and use of LNG and natural gas and other dangerous goods, marine transport and activities of LNG carriers and FSRU Vessel and natural gas subsea pipelines within Hong Kong waters in normal and adverse weather or tidal situations, and accidental spillage or leakage of LNG or natural gas.  The risks during the construction and operations phases of the Project are also assessed in consideration of other potential contributors to risk in the vicinity of the Project’s facilities.

5.2                                   Legislative Requirements and Evaluation Criteria

The key legislation and guidelines which are considered relevant to the QRA Study for the Project are as follows:

·             Environmental Impact Assessment Ordinance (EIAO), Cap. 499;

·             Technical Memorandum on EIA Process (EIAO-TM);

·             Gas Safety Ordinance, Cap. 51;

·             Dangerous Goods Ordinance, Cap. 295;

·             Merchant Shipping (Safety) Ordinance (Cap. 369) and its subsidiary regulations such as Merchant Shipping (Safety) (Gas Carriers) Regulations (Cap. 369Z);

·             Merchant Shipping (Local Vessels) Ordinance (Cap. 548) and its subsidiary regulations.

5.2.1                           Risk Criteria

Section 2 of Annex 4 of the EIAO-TM specifies the individual risk and societal risk guidelines that will apply to the Project.

Individual risk is the predicted increase in the chance of fatality per year to a hypothetical individual who remains at a given stationary point for 100% of the time.  The individual risk guidelines specify that the maximum level of off-site individual risk associated with a hazardous installation should not exceed 1 in 100,000 per year, i.e. 1 × 10-5 per year.

Societal risk expresses the risks to the surrounding off-site population in the vicinity of a hazardous installation.  The societal risk guidelines for acceptable risk levels are presented graphically in Figure 5.2.  The societal risk is expressed in terms of frequency (F) of fatalities against number of fatalities (N) in the population from incidents at a hazardous installation.  Two F-N risk lines are used to demark “Acceptable” or “Unacceptable” regions.  The region between the two F-N risk lines indicates the acceptability of the societal risk is borderline and should be reduced to As Low As Reasonably Practicable (ALARP) level.  This seeks to ensure that all practicable and cost-effective mitigation measures which can reduce societal risk are considered.  In order to avoid major incidents resulting in more than 1,000 fatalities, there is a vertical cut-off line at the 1,000 fatalities level extending down to a frequency of 1 in a billion years.

5.3                                   Assessment Methodology

5.3.1                           Components for QRA Study

With due consideration of the Project key components and activities and to address requirements of the EIA Study Brief, the remainder of this Section has been divided into four components of the QRA Study. Sections 5.4 to Section 5.7 detail the four components.  The key hazardous materials assessed in each section are illustrated in Table 5.1.

Table 5.1        QRA Study Components and Associated Hazard Evaluations

QRA Study Components:

Description of Hazard Evaluation

Associated Hazards

Section

LNG

Natural Gas

Other Dangerous Goods

Evaluation of risks associated with LNGC and FSRU Vessel transit routes within Hong Kong waters to the LNG Terminal (including transits for temporary sheltering under adverse weather condition), as well as emergency transits of LNGC and FSRU Vessel

ü

ü

ü

Section 5.4: LNGC / FSRU Vessel Transits to the LNG Terminal

Evaluation of risks of LNG / natural gas / other dangerous goods associated with the LNG Terminal operation

ü

ü

ü

Section 5.5: LNG Terminal Operation

Evaluation of risks of natural gas associated with subsea pipelines connecting the Jetty and the proposed GRSs at the BPPS and the LPS

N/A

ü

N/A

Section 5.6 : Subsea Pipelines

Evaluation of risks of natural gas associated with the proposed GRSs at the BPPS and the LPS

N/A

ü

ü

Section 5.7 : GRS Facilities

Notes: ü: Applicable, N/A: Not Applicable

 

Other risk factors which could induce potential risk on the Project components, as specified in Section 3.4.5 and Appendix B of the EIA Study Brief, are also assessed in the corresponding sections illustrated in Table 5.2.

Table 5.2        Other Assessed Risk Factors

Section Description

Risk of Collision from High Speed Ferries

Adverse Weather (e.g. typhoon, storm surge, extreme tide)

Risk from Operation of Helicopters / Aircrafts

Risk Associated with Existing GRSs

Section 5.4:

Marine Transits of LNGC / FSRU Vessel to the LNG Terminal

ü

ü

ü

N/A

Section 5.5:

LNG Terminal Operation

ü

ü

ü

N/A

Section 5.6:

Subsea Pipelines

N/A

N/A

N/A

N/A

Section 5.7:

GRS Facilities

N/A

ü

ü

ü

Notes: ü: Applicable, N/A: Not Applicable.

 

It should be noted that during the construction of the Jetty and both subsea pipelines, LNG, natural gas and other dangerous goods will not be present other than for commissioning purposes.  Therefore, construction phase associated risk has not been further assessed.

During the construction of the proposed GRSs at the BPPS and LPS, the hazards arising from the associated construction works may impact the existing neighbouring GRS facilities, leading to natural gas potential loss of containment.  The associated risk has been assessed in Section 5.7.6.

5.3.2                           General Approach

The overall assessment approach is illustrated in Figure 5.1.  The methodology adopted for the QRA Study is consistent with other studies that have been approved by the EPD and other relevant authorities, including both EIA and safety case studies, such as:

·             ERM, EIA for 1,800 MW Gas-fired Power Station at Lamma Extension (Register No.: AEIAR-010/1999), February 1999;

·             ERM, EIA for Liquefied Natural Gas (LNG) Receiving Terminal and Associated Facilities (Register No.: AEIAR-106/2007), December 2006;

·             ERM, EIA for Black Point Gas Supply Project, Revision 3 (Register No.: AEIAR-150/2010), February 2010; and

·             ERM, EIA for Additional Gas-fired Generation Units Project (Register No.: AEIAR-197/2016), June 2016.

The assumptions and methodology used for the QRA Study are in line with the Method Statement that was approved by EPD on 28 September 2017.

The general approach taken for each step of the QRA Study is provided below.

Information Collection and Review

Relevant information on the Jetty, Subsea Pipelines and GRS such as layout drawings, design basis, weather data, surrounding off-site population data, etc. were collected, reviewed and incorporated in the QRA Study.

Hazard Identification

A Hazard Identification (HAZID) analysis including a HAZID workshop for the Project components was conducted to identify all potential hazards.  A review of literature and incident / accident database was also conducted to identify all potential hazardous scenarios for consideration in the QRA Study.

Frequency Analysis

The failure frequencies / likelihood of the various hazardous scenarios outcomes were derived from historical failure databases and by using event tree analysis.  Where necessary and applicable, fault tree analysis was also conducted to take into account Project specific factors.

Consequence Analysis

All identified hazardous materials (LNG, natural gas, and other dangerous goods) were assessed in the QRA Study.  The consequence modelling for these hazardous materials was performed using an internationally recognized consequence modelling package, PHAST.

Cumulative Risk Assessment

The consequence and frequency data, together with surrounding off-site population data, was subsequently combined using approved software packages, SAFETITM and RiskplotTM, which is in line with previous EIA studies that have been approved by EPD and other relevant authorities ([1]) ([2]) and as stated in the Method Statement for the QRA Study.  The results from the cumulative risk assessment were compared against the risk criteria stipulated in Section 2 of Annex 4 of the EIAO-TM.

Risk Mitigation

Practicable and cost-effective risk mitigation measures were proposed, as required based on the findings of the QRA Study.

5.3.3                           Assessment Basis

Proposed Assessment Years for QRA Study

The proposed assessment years adopted for the QRA Study are as follows:

·      Year 2020: Target year for the Project to start commercial operations; and

·      Year 2030: Future scenario for the Project.

In addition, the proposed GRS facilities at the BPPS and LPS are expected to be under peak construction time in the beginning of 2020.  The GRS construction works may induce additional risks to the neighbouring operating GRS facilities.  Therefore, Year 2020 has been also adopted as the assessment year for the construction activities at both GRS locations.

Surrounding Population

Surrounding Marine Vessel Population Estimation

The marine traffic data in the vicinity of the Project components was collected and used to estimate the marine vessel population for consideration in the QRA Study.

The marine traffic that was identified in the vicinity of the Project components included fishing vessels, river trade, ocean-going vessels, fast ferries, and other types of smaller vessels.  A forecasting exercise for marine traffic was also conducted and the increment in marine traffic volume was derived by trend analysis for each vessel class so that a representative pattern was developed for the 2020 and 2030 timeframes.

The population of all marine vessels was treated as an area averaged density, except for fast ferries which were treated as point receptors.  The approach to estimate the marine vessel population for the proposed assessment years was consistent with the previous EIA studies that have been approved by the EPD and other relevant authorities ([3]) ([4]) and as included in the Method Statement for the QRA Study.

The detailed marine traffic data and marine vessel population estimation assumptions and methodology are summarised in Annex 5A.

Surrounding Land Population Estimation

Based on a review of the GeoInfo Map ([5]), there is no land based population in the vicinity of any of the Project components, and the detailed discussion on the land population is summarised in Annex 5A.

Surrounding Road Traffic Population Estimation

There is no road traffic population in the vicinity of the GRS at the LPS.  Road traffic population was identified in the vicinity of the GRS at the BPPS.  The detailed discussion on the road traffic population is summarised in Annex 5A.

Land On-site Population

Safety management measures will be put in place to minimize the potential risk exposure to all personnel at the BPPS and LPS facilities, including operational staff during the operation phase and construction workers during the construction phase. 

The key safety management measures for all personnel at the BPPS and LPS will include:

·      All personnel within the BPPS shall comply with CLP safety policy and requirements;

·      All personnel within the LPS shall comply with HK Electric safety policy and requirements;

·      All operation work procedures shall be complied with the operating plant procedures or guidelines and regulatory requirements;

·      All personnel shall be equipped with appropriate personal protective equipment (PPE) when working at the BPPS and LPS facilities;

·      Safety training and briefings shall be provided to all personnel; and

·      Regular site safety inspections/ audits shall be conducted. 

The key safety management measures to manage the risk associated with construction workers during the construction phase will include:

·      Method statements and risk assessments shall be prepared and safety control measures shall be in place before the commencement of construction works;

·      Work permit system, on-site pre-work risk assessment and emergency response procedure shall be in place before commencement of construction works; and

·      All construction workers shall be under close site supervision during the construction phase of the GRSs.

With the implementation of the above safety management measures at the BPPS and LPS, the potential risks to all personnel on-site, including operational staff during the operation phase and construction workers during the construction phase in the BPPS and LPS, are expected to be insignificant; hence they were not considered as off-site population and has not be taken into account in the QRA Study.

LNG Terminal On-site Population

Safety management measures will also be put in place to minimize the potential risk exposure to all personnel at the LNG Terminal, including operational staff during the operation phase and construction workers during the construction phase. 

The key safety management measures for all personnel at the LNG Terminal will include:

·      All personnel within the LNG Terminal shall comply with relevant safety policy and requirements;

·      All operation work procedures shall be complied with relevant codes and standards (e.g. SIGTTO) and regulatory requirements;

·      Work permit system and emergency response procedure shall be in place;

·      Robust and extended process control system, safety control system, fire-fighting system and security system shall be provided;

·      Sufficient and trained / competent staff shall be provided to operate the LNG Terminal; and

·      Regular safety inspections/audits shall be conducted.

The key safety management measures to manage the risk associated with construction workers during the construction phase of the LNG Terminal will include:

·      Method statements and risk assessments shall be prepared and safety control measures should be in place before the commencement of construction works;

·      Work permit system, on-site pre-work risk assessment and emergency response procedure shall be in place before commencement of construction works; and

·      All construction workers shall be under close site supervision during the construction phase of the LNG Terminal.

With the implementation of the above safety management measures at the LNG Terminal, the potential risks to all personnel on-site, including operational staff during the operation phase and construction workers during the construction phase in the LNG Terminal, are expected to be insignificant; hence they were not considered as off-site population and has not be taken into account in the QRA Study.

Meteorological Data

The 5-year meteorological data from Year 2012 to Year 2016 from the Hong Kong Observatory (HKO) has been selected to represent local meteorological conditions including wind speed, wind direction, atmospheric stability class, temperature, and relative humidity.

The weather stations in the vicinity of the Project components were reviewed, and the following weather stations listed in Table 5.3 were selected for the QRA Study.

Table 5.3        Selected Weather Stations for the QRA Study

Component / Weather Station

Cheung Chau

Sha Chau

Lamma Island

LNGC / FSRU Vessel Transits to the LNG Terminal

ü

N/A

N/A

FSRU Vessel, the Jetty and LNGC Unloading at the LNG Terminal

ü

N/A

N/A

Subsea BPPS Pipeline

N/A

ü

N/A

Subsea LPS Pipeline

ü

N/A

N/A

GRS Facility at the BPPS

N/A

ü

N/A

GRS Facility at the LPS

N/A

N/A

ü

Notes: ü: Applicable, N/A: Not Applicable.

The meteorological data of these selected weather stations are summarised in Annex 5A.

5.4                                   QRA for Marine Transits of LNGC and FSRU Vessel to the LNG Terminal

This Section presents the QRA Study for the evaluation of the risks associated with the LNGC and FSRU Vessel along the transit routes to the LNG Terminal during normal operations (including transits for temporary sheltering under adverse weather condition), as well as emergency transits of the LNGC and FSRU Vessel.

5.4.1                           Description of the Marine Transits for LNGC and FSRU Vessel

Type of LNGC

Two (2) types of LNGC with double hull are typically used in the market to deliver LNG cargoes, namely:

·      Membrane type; and

·      MOSS type (spherical LNG storage tank).

More than 90% LNGCs are membrane type at the current LNGC market, as such, the LNGC of membrane type was selected as the representative case for the QRA Study.

Size of LNGC

The QRA Study was conducted based on two (2) sizes of LNGC (each with five (5) membrane-type LNG Cargo Tanks):

·      Small LNGC (170,000 m3 capacity, with each LNG storage tank capacity of about 34,000 m3); and

·      Large LNGC (270,000 m3, capacity, with each LNG storage tank capacity of about 54,000 m3).

The typical safeguards for LNGC design and operations have been summarised in Section 3 of the EIA Report.

Description of LNGC Process System

Typically, an LNGC has the following main process systems:

·      LNG Storage System;

·      LNG Unloading Arms;

·      Diesel Storage System;

·      Lubricating Oil System; and

·      Fuel Oil Storage System.

The detailed description of the above process systems is summarised in Annex 5B, while the key description of FSRU Vessel is summarised in Section 5.5.1.

LNGC Transit Routes

Figure 5.3 presents the indicative LNGC transit routes to the LNG Terminal, and the FSRU Vessel will also use the same LNGC transit routes as the initial marine transit to the LNG Terminal.  The length of the LNGC transit route is about 30 km, and the description of the route segments is presented in Table 5.4. 

Table 5.4        LNGC Transit Route Segment

Segment Code

Segment Description

Length of Segment

Segment ‘a’

Transit

27.1 km

Segment ‘b’

Approaching the LNG Terminal

3.1 km

 

According to the Marine Traffic Impact Assessment (MTIA) Report for the LNG Terminal ([6]), the support of a tug fleet for access to/from the LNG Terminal ensures that even with engine or control system failure on the LNGC or FSRU Vessel during the approaching the LNG Terminal, there will be adequate control capability to mitigate such events.  A total of four (4) tugs, of 80T bollard pull or higher are anticipated to support all LNGC’s scheduled arrivals and departures, and FSRU Vessel arrival and departures due to typhoon.  In addition, tugs will also be required to assist departures prior to the onset of a typhoon.  These tugs will have the necessary electrical system compliance and gas detection to be safe to work in close proximity with the LNG Terminal.

LNGC Transit Frequency

Based on the estimated LNG Terminal throughput, it is envisaged that the frequency of LNGC visits on average will be one LNGC arriving every five to eight days.  As a conservative approach, the QRA Study was conducted based on the following maximum annual visit frequency:

·      75 visits per year (equivalent to every 4.8 days) for Small LNGC; and

·      50 visits per year (equivalent to every 7.3 days) for Large LNGC.

With regard to the LNGC transit to the LNG Terminal, the following transit conditions were considered in the QRA Study:

·      The LNGC transit is conducted by the Small LNGC as the worst-case scenario since the number of LNGC transit and associated transit risk is higher; and

·      The maximum annual visit frequency of the Small LNGC to the LNG Terminal is seventy five (75) visits per year.

Initial Transit of FSRU Vessel to the LNG Terminal

For the initial transit to the LNG Terminal, the FSRU Vessel will transit and approach the LNG Terminal on the same route as the LNGC normal transit route shown in Figure 5.3.

Transit of LNGC under Adverse Weather Condition

Prior to the transit of an LNGC to the LNG Terminal for LNG unloading operation, the transit route and the weather forecast for the transit area will be reviewed and analyzed to determine the suitability and safety of the LNGC transit.  It is expected that the LNGC will only be allowed to transit and enter Hong Kong waters if the forecasted weather condition is within an agreed weather envelope.  Therefore, it is highly unlikely that an LNGC will be at berth at the Jetty when a typhoon is predicted.  Nevertheless, a frequency of once per five (5) years was conservatively assumed to be adopted in the QRA Study.

In case the on-set of a typhoon occurs during the LNG unloading operation at the Jetty, the LNGC will, depending on weather conditions and at the discretion of the Master head, depart the berth to an area of open sea outside HKSAR waters.  Once the weather conditions have returned to acceptable operating limits for berthing, the LNGC will return to the LNG Terminal using the same LNGC normal transit route as presented in Figure 5.3.

Transit of the FSRU Vessel under Adverse Weather Condition

In case of adverse weather condition (e.g. typhoon, monsoon), the FSRU Vessel berthed at the Jetty will also, depending on weather conditions and at the discretion of the Master head, depart the berth to an area of open sea outside HKSAR waters.

Although it was identified from the prior mooring capability assessment that the FSRU Vessel could maintain at the LNG Terminal in winds associated with Typhoon Signal 3 (sustained speeds of 41-62 km/hr) ([7]), it was conservatively assumed that departure of the FSRU Vessel would be required upon Typhoon Signal No. 3 or higher for the QRA Study.

According to the HKO, the average number of days per year with Typhoon Signal No. 3 or higher in Hong Kong between 1961 and 2010 is 9.56 days ([8]), and the average annual frequency of Typhoon Signal No. 3 or higher in Hong Kong between 1956 and 2014 is 8.7 times (2).  With the aim to build conservatism in this study, the FSRU Vessel’s departure frequency from the Jetty under adverse weather condition was conservatively selected as ten (10) times per year. 

Once the weather conditions have returned to acceptable operating limits for berthing, the FSRU Vessel will return to the LNG Terminal using the same LNGC normal transit route as presented in Figure 5.3.

A Typhoon Departure Plan will be put in place (and agreed with Marine Department). The plan will fully document the procedure to be followed in the event of a typhoon affecting LNG Terminal operation.

Transit of the LNGC and FSRU Vessel under Emergency Situation

In the case of an emergency situation (e.g. uncontrolled fire event at the Jetty), the FSRU Vessel berthed at the Jetty and any LNGC that may be on berth at the time of the emergency will be required to depart the berth to an area of open sea outside HKSAR waters.  In addition, a standby vessel is available to provide an emergency response and will have the capability to assist the FSRU Vessel and LNGC depart the berth.  The frequency of this scenario was conservatively assumed as once every five (5) years for the QRA Study.

Once the emergency situation is over and the Jetty is made safe, the LNGC and FSRU Vessel will return to the LNG Terminal using the same LNGC transit route as presented in Figure 5.3.

An emergency response plan will be put in place which fully documents the procedures to be followed in the event of an emergency.

5.4.2                           Hazard Identification

The hazardous scenarios associated with the marine transits of the LNGC and FSRU Vessel to the LNG Terminal were identified though the following tasks:

·      Review of hazardous materials;

·      Review of potential Major Accident Events (MAEs);

·      Review of relevant industry incidents;

·      Review of potential initiating events leading to MAEs; and

·      HAZID Workshop.

Review of Hazardous Materials

LNG on board the LNGC and FSRU Vessel was the major hazardous material considered in the QRA Study, while other dangerous goods including diesel, marine diesel oil, and lubricating oil were also considered.  The details of the storage of LNG and other dangerous goods on board the LNGC and FSRU Vessel during marine transit are summarised in Table 5.5 and Table 5.6 respectively.

Table 5.5        LNG and Other Dangerous Goods Associated with LNGC during Marine Transit

Chemicals

 

Dangerous Goods Classification*

Maximum Storage Quantity

Temperature

(°C)

Pressure (barg)

LNG for Large LNGC

 

-

270,000 m3

-156

0.7

LNG for Small LNGC

 

-

170,000 m3

-156

0.7

Diesel (Heavy Fuel Oil)

 

Category 5

~6,000 m3

25

ATM

Marine Diesel Oil

 

Category 5

≤ 800 m3

25

ATM

Lubricating Oil

 

-

≤ 100 m3

25

ATM

Calibration Gas^ 

 

Category 2

1 cylinder

25

137

Notes:

*: The dangerous goods category is classified based on “Fire Protection Notice No. 4, Dangerous Goods General” by Fire Services Department ([9]).

^: The key composition of the calibration gas for Gas Chromatograph is methane (90 vol%), ethane (5 vol%), Nitrogen (2.5 vol%), and carbon dioxide (1 vol%) and propane (1 vol%).

 

Table 5.6        LNG and Other Dangerous Goods Associated with FSRU Vessel during Marine Transit

Chemical

Dangerous Goods Classification*

Maximum Storage Quantity

Temperature (°C)

Pressure (barg)

LNG

-

270,000 m3

-156

0.7

Diesel (Heavy Fuel Oil)

Category 5

~6,000 m3

25

ATM

Marine Diesel Oil

Category 5

≤ 800 m3

25

ATM

Lubricating Oil

-

≤ 100 m3

25

ATM

Calibration Gas^ 

Category 2

1 cylinder

25

137

Notes:

*: The dangerous goods category is classified based on “Fire Protection Notice No. 4, Dangerous Goods General” by Fire Services Department. (1)

^: The key composition of the calibration gas for Gas Chromatograph is methane (90 vol%), ethane (5 vol%), Nitrogen (2.5 vol%), and carbon dioxide (1 vol%) and propane (1 vol%).

The detailed description of each identified hazardous material is provided below.

LNG

LNG is an extremely cold, non-toxic, non-corrosive and flammable substance. 

If LNG is accidentally released from a temperature-controlled container, it is likely to come in contact with relatively warmer surfaces and air that will transfer heat to the LNG.  The heat will begin to vapourise some of the LNG, returning it to its gaseous state. 

The relative proportions of liquid LNG and gaseous phases immediately following an accidental release depends on the release conditions.  The released LNG will form a LNG pool on the surface of the sea in the vicinity of the FSRU Vessel/ LNGC/ Jetty which will begin to “boil” and vapourise due to heat input from the surrounding environment.  The vapour cloud may only ignite if it encounters an ignition source while its concentration is within its flammability range.

Any person coming into contact with LNG in its cryogenic condition will be subjected to cryogenic burns. 

Diesel (Heavy Fuel Oil), Marine Diesel Oil and Lubricating Oil

Diesel, marine diesel oil and lubricating oil have a relatively higher flash point (greater than 66 °C), which is above ambient temperature, and with a high boiling point.  Thus, evaporation from a liquid pool is expected to be minimal.

Calibration Gas

The volume of the compressed gas inside the cylinders is limited and the associated inventory available is small, and those compressed gas cylinders are located at machinery room.  Should loss of containment occur for the compressed gas cylinders, there is no off-site impact on surrounding marine population.  Hence, it is not further assessed in the QRA Study.

Review of Potential MAEs

LNG

The possible hazardous scenarios considered in the QRA Study, upon the ignition of any released LNG during the marine transits of the LNGC or FSRU Vessel with consideration of operating conditions, are:

·      Pool fire; and

·      Flash fire.

Diesel (Heavy Fuel Oil), Marine Diesel Oil and Lubricating Oil

Considering the high flash point temperature of the other dangerous goods such as marine diesel oil present in the LNGC and FSRU Vessel, the possible hazardous scenarios considered in the QRA Study are pool fire and flash fire.

Detailed characteristics of the above hazardous scenarios (i.e. pool fire and flash fire) are described in Annex 5G.

Review of Relevant Industry Incidents

To further investigate possible hazardous scenarios from the LNGC and FSRU Vessel, review of the applicable past industry incidents at similar facilities worldwide was conducted based on the following incident/ accident database:

·      Institution of Chemical Engineers (IChemE) accident database;

·       eMARS ([10]);

·       ERNS ([11]);

·      Major Hazard Incident Data Service (MHIDAS) database ([12]); and

·      Society of International Gas Tanker and Terminal Operators (SIGTTO) ([13]).

Details of the past industry incident analysis are presented in Annex 5C.

Review of Potential Initiating Events Leading to MAEs

The key potential hazardous scenarios arising from marine transits of the LNGC and FSRU Vessel were identified as loss of containment of LNG.  The potential initiating events which could result in the loss of containment of LNG are listed below:

·      Ship Collision;

·      Groundings;

·      Sinking or foundering;

·      General equipment/piping failure (due to corrosion, construction defects etc.);

·      LNG containment system failure; and

·      External effects - adverse weather (typhoon, poor visibility, storm surge, extreme tide), tsunami, and lightning.

Descriptions of these potential initiating events are presented in Annex 5D.

HAZID Workshop

A HAZID workshop was conducted to confirm and further identify the potential initiating events which may lead to MAEs along the LNGC and FSRU Vessel transit route based on the HAZID team representatives’ experience, past industry accidents, lessons learnt and guideword checklists.  The HAZID workshop worksheet is presented in Annex 5E.  The HAZID workshop output served as a basis for the identification of potential initiating events and hazardous scenarios for the QRA Study.

Development of Hazardous Sections

A number of hazardous sections for detailed analysis in the QRA Study based on location of emergency shutdown valves and process conditions were developed.  The details of each hazardous section are presented in Annex 5D.

5.4.3                           Frequency Analysis

Ship Collision Frequency Analysis

A ship collision frequency analysis was conducted following the approach adopted in the previous EIA Report that was approved by the EPD ([14]).  DYMTRI (Dynamic Marine Traffic simulation) model ([15]) was adopted as the platform for the marine traffic simulation to predict the collision frequencies along the LNGC and FSRU Vessel transit route.

The key steps of the ship collision frequency analysis included:

·      Identification of Modelled Marine Traffic;

·      Hazard Identification;

·      Model Validation;

·      Marine Traffic Forecasts; 

·      Scenario Development ;

·      Collision Frequency Assessment; and

·      Collision Energy Distribution.

The description of these key steps is described in Annex 5F.

The total collision frequencies leading to the loss of containment of LNG are provided in Table 5.7 and Table 5.8.

Table 5.7        Total Ship Collision Frequency Leading to Loss of Containment of LNG (Year 2020)

Type of LNGC

Release Frequency in

Sub-Segment “a” (/m/year)

Release Frequency in

Sub-Segment “b” (/m/year)

Small LNGC

1.6 × 10-8

1.5 × 10-9

Large LNGC

1.6 × 10-8

1.5 × 10-9

 

Table 5.8        Total Ship Collision Frequency Leading to Loss of Containment of LNG (Year 2030)

Type of LNGC

Release Frequency in

Sub-Segment “a” (/m/year)

Release Frequency in

Sub-Segment “b” (/m/year)

Small LNGC

1.7 × 10-8

4.9 × 10-10

Large LNGC

1.8 × 10-8

5.2 × 10-10

 

Grounding Frequency Analysis

The anticipated grounding frequency for the LNGC and FSRU Vessel during their transits to and from the LNG Terminal has been developed from a review of historical incidents in Hong Kong waters associated with vessels over 200 m Length Overall (LOA).  Considering the number of marine transits per year and the probability of loss of LNG containment due to grounding events, the grounding release frequency adopted in the QRA Study was 1.2 × 10-9 per m per year.  The derivation of this grounding frequency is provided in Annex 5F.

Release Hole Sizes

The release hole sizes and associated penetration energy selected are as per the previous EIA Report ([16]) that was approved by the EPD, are presented in Table 5.9.

Table 5.9        Release Hole Sizes and Penetration Energy

Release Hole Size

Penetration Energy (MJ)

250 mm

100 to 110 MJ

750 mm

111 to 150 MJ

1,500 mm

>150 MJ

 

Ignition Probability

As per the previous EPD (1) approved EIA Report, the immediate ignition probability for the collision scenarios was selected as 0.8; and the immediate ignition probability for the grounding scenarios was selected as 0.2 for the QRA Study.

Event Tree Analysis

An event tree analysis was performed to model the development of each hazardous scenario outcomes (pool fire and flash fire) from an initial release scenario.  The event tree analysis considered whether there is immediate ignition or delayed ignition, with consideration of the associated ignition probability as discussed above.  The development of the event tree is presented in Annex 5F.

5.4.4                           Consequence Analysis

Physical Effects Modelling

PHAST was used to perform the physical effects modelling to assess the effects zones for the following hazardous scenarios:

·      Pool fire; and

·      Flash fire.

Detailed description of the physical effects modelling is presented in Annex 5G.

Consequence End-point Criteria

For thermal radiation impact, the associated fatality/ injury from a pool fire was estimated based on the following probit equation ([17]):

Y = -36.38 + 2.56 ln (t I 4/3)

where:

Y   is the probit

I    is the radiant thermal flux (W m-2)

t    is duration of exposure (s)

The exposure time, t, is limited to a maximum of twenty (20) seconds. 

With regard to a flash fire, the criterion chosen is that a 100% fatality was adopted for any person outdoor within the flash fire envelope, which was conservatively selected as 0.85 of the Lower Flammable Limit (LFL).

Details of the consequence modelling results are presented in Annex 5G.

5.4.5                           Risk Summation 

The risk summation for the LNGC and FSRU Vessel transits was modelled using SAFETI, which is in line with the previous EIA Report that was approved by EPD ([18]).

Individual Risk Results

The individual risk contours associated with the LNGC and FSRU Vessel transits are shown in Figure 5.4 and Figure 5.5. 

The individual risk contour of 1 × 10-5 per year was not reached for the LNGC and FSRU Vessel transit route in the Operational Year in 2020 and Future Scenario Year in 2030, as such the individual risk criterion stipulated in Section 2 of Annex 4 of the EIAO-TM is met.

Societal Risk Results

The societal risk for the LNGC and FSRU Vessel transits, in terms of F-N curve, was calculated based on the surrounding off-site marine vessel populations in the vicinity of the transit route.  The societal risks in terms of F-N curves for the Operational Year in 2020 and Future Scenario Year in 2030, as shown in Figure 5.6, lie within the Acceptable Region, as such the societal risk criteria stipulated in Section 2 of Annex 4 of the EIAO-TM are met.

5.4.6                           Conclusions of QRA Study for Marine Transits of LNGC and FSRU Vessel

Both individual risk and societal risk associated with the transits of the LNGC and FSRU Vessel are in compliance with the risk criteria stipulated in Section 2 of Annex 4 of the EIAO-TM.

5.5                                   QRA Study for the LNG Terminal

This Section presents the QRA Study for the risks evaluation associated with the LNG Terminal, including FSRU Vessel, the Jetty, and the LNGC unloading operation at the Jetty.

5.5.1                           Description of the LNG Terminal and its Unloading Operations

The Jetty

The Jetty is designed for simultaneous mooring of both the FSRU Vessel and an LNGC, and is a typical double berth marine structure that uses mooring / fender facilities to safely moor the FSRU Vessel and LNGC.

The LNGC at the Jetty transfers LNG at 5 barg and -156 °C via unloading arms across the Jetty to the FSRU Vessel, and the LNG unloading rate is a maximum of 12,000 m3/hr.  For the LNG unloading operation, the unloading arm configuration between the Jetty and the LNGC as well as between the Jetty and the FSRU Vessel consist of two (2) unloading arms dedicated for LNG service, one (1) hybrid arm normally transferring LNG but also capable of transferring LNG vapour, and one (1) dedicated vapour return arm.  Upon completion of the LNG unloading operation, all unloading arms will be isolated, de-inventorised and purged with nitrogen inert gas before being disconnected.

The LNG transferred from the LNGC is stored in the FSRU Vessel’s storage tanks.

FSRU Vessel

The LNG in the FSRU Vessel’s storage tanks is pumped to the regasification unit by LNG Storage Tank Pumps and LNG Booster Pumps.  The regasification unit comprises of regasification trains, with a maximum installed capacity of 1,000 mmscfd.

The natural gas at 5 °C and 88 barg is then sent from the regasification trains, via the metering system, to the Jetty at a maximum flow rate of 800 mmscfd to two (2) high pressure (HP) Gas Send-out Arms (1 duty and 1 standby) that supply the natural gas to the GRS at the BPPS via the 30” BPPS Pipeline, and to the GRS at the LPS via the 20” LPS Pipeline.

The LNG Terminal has the capability to receive LNG from LNGC while simultaneously sending out natural gas, and the FSRU Vessel also has the capability to reload LNG onto an LNGC or, in future, onto a LNG bunker vessel or barge.

Type of FSRU Vessel

Two (2) types of FSRU Vessel with double hull are typically used in the market to receive, store, regasify and send out natural gas, namely:

·      Membrane type; and

·      MOSS type (spherical LNG storage tank).

Membrane type storage tanks are favoured for new build FSRU Vessels, as their configuration provides a higher storage capacity for a given ship size due to no space between the storage tanks, as well as flat deck providing a better platform for the regasification facilities.

As such, membrane-type FSRU Vessel with LNG storage capacity of 270,000 m3 and five (5) LNG Cargo Tanks was selected as the representative case for the QRA Study.

FSRU Vessel Process Systems

Typically, an FSRU Vessel has the following main process systems:

·      LNG Regasification: LNG Send-out Booster Pump System;

·      LNG Regasification: LNG Vapourisation System;

·      BOG Handling and Recovery System;

·      Seawater Intake System;

·      Sodium Hypochlorite System;

·      Diesel Storage System;

·      Lubricating Oil System;

·      Fuel Oil Storage System; and

·      Nitrogen Generation System.

Detailed process description of the above process systems is summarised in Annex 5B.

Key Safety Systems for the FSRU Vessel and the Jetty

The following safety systems are typically provided on the FSRU Vessel, the Jetty and on the visiting LNGC:

·      Emergency Release Coupling System for Unloading Arms;

·      Process Overpressure Protection System;

·      Emergency Shutdown System;

·      Fire and Gas Detection System;

·      LNG Spillage Protection System;

·      Escape Routes / Paths and Escape System; and

·      Security Control System.

Detailed description of the above safety systems is summarised in Annex 5B.

LNGC Unloading Operations at the LNG Terminal

With regard to the LNGC unloading operations at the LNG Terminal, the following operating conditions were considered in the QRA Study:

·      The LNG unloading operation is conducted by the Small LNGC as the worst-case scenario since the unloading operation frequency and associated process risk is higher;

·      The maximum annual visit frequency of the Small LNGC to the LNG Terminal is seventy five (75) visits per year;

·      The maximum unloading time for the Small LNGC at the LNG Terminal is twenty four (24) hours; and

·      The maximum staying time of the Small LNGC at the LNG Terminal is forty eight (48) hours.

Marine Diesel Oil Bunkering Operations at the LNG Terminal

As per previous project experience for similar facilities, the number of bunkering operations for marine diesel oil using marine service vessels is typically three (3) times per year.  As such, it was conservatively assumed that the bunkering operation of marine diesel oil for the LNG Terminal will be performed three (3) times per year and that each operation is up to six (6) hours duration.  The risks associated with marine diesel bunkering operation as well as the associated escalation effect have been considered in the QRA hazard to life assessment.  The associated risk impacts to the off-site population is insignificant.

5.5.2                           Hazard Identification

Hazardous scenarios associated with the operation of the LNG Terminal, including an LNGC unloading at the LNG Terminal and sending out natural gas were identified through the following tasks:

·      Review of hazardous materials;

·      Review of potential MAEs;

·      Review of relevant industry incidents;

·      Review of potential initiating events leading to MAEs; and

·      HAZID Workshop.

Review of Hazardous Materials

LNG on board the LNGC and FSRU Vessel, and natural gas associated with the LNG Terminal were the major hazardous material considered in the QRA Study, while the other dangerous goods including diesel, marine diesel oil, lubricating oil, sodium hypochlorite, hydrogen and nitrogen were also taken into account in the QRA Study.  

The details of the storage of LNG and other dangerous goods associated with the LNG Terminal are summarised in Table 5.10. 

Table 5.10      LNG and Other Dangerous Goods Associated with the LNG Terminal

Chemical

Location

Dangerous Goods Classification*

Maximum Storage Quantity

Temperature (°C)

Pressure (barg)

LNG

FSRU Vessel

-

270,000 m3

-163

5

Natural gas

FSRU Vessel

-

On-site generation

5

88

Diesel (Heavy Fuel Oil)

FSRU Vessel

Category 5

~6,000 m3

25

ATM

Marine Diesel Oil

FSRU Vessel

Category 5

≤ 800 m3

25

ATM

Diesel Oil

Jetty

Category 5

~50 m3

25

ATM

Lubricating Oil

FSRU Vessel

-

≤ 100 m3

25

ATM

Sodium Hypochlorite

FSRU Vessel

Category 4

On-site generation

-

-

Hydrogen

FSRU

Vessel

Category 2

By-product of

Electrochlorination System

-

-

Nitrogen

FSRU

Vessel

Category 2

On-site generation

-

-

Calibration Gas^ 

FSRU

Vessel

Category 2

1 cylinder

25

137

Notes:

*: The dangerous goods category is classified based on “Fire Protection Notice No. 4, Dangerous Goods General” by Fire Services Department (1).

^: The key composition of the calibration gas for Gas Chromatograph is methane (90 vol%), ethane (5 vol%), Nitrogen (2.5 vol%), and carbon dioxide (1 vol%) and propane (1 vol%).

A detailed description of the LNG, diesel, marine diesel oil and lubricating oil hazards is provided in Section 5.4.2, while natural gas, sodium hypochlorite, hydrogen, and nitrogen hazards are discussed in the following section.

Natural Gas

Upon the regasification of LNG, natural gas is formed.  Natural gas is composed of primary methane gas with other fossil fuels such as ethane, propane, butane and pentane, etc.  Natural gas is extremely flammable when mixed with appropriate concentration of air or oxygen in the presence of an ignition source.

Not only is the maximum surface emissive power of pure methane higher, but the consequence distances for both flash fire and jet fire hazardous scenarios associated with pure methane is larger than that of natural gas.  Therefore, pure methane has been conservatively selected as representative material for natural gas in the consequence modelling conducted using PHAST.

The major hazards arising from loss of containment of natural gas may lead to hazardous scenarios including jet fire, flash fire, and vapour cloud explosion (VCE).

Sodium Hypochlorite (NaOCl)

Chemical Abstracts Service (CAS) number of NaOCl is 7681-52-9, and NaOCl solution is a corrosive liquid with the appearance of colourless to yellowish, and with a chlorine-like odour.  NaOCl is not flammable, but it can decompose and release corrosive chorine gas if in contact with acids.  NaOCl is produced by an electrochlorination system on board the FSRU Vessel which is a continuous process and does not rely on any stored chlorine gas or hypochlorite brought from off-site.

The expected off-site impact associated with decomposition of the solution is limited.  Also, once generated on board the FSRU Vessel, NaOCl is consumed immediately for treatment of seawater.  Therefore, NaOCl was not further assessed in the QRA Study.

Hydrogen

CAS number of hydrogen is 1333-74-0, and hydrogen is a colourless and odourless gas at ambient temperature and pressure.  It has a boiling point of -253 °C at 1 bara, critical temperature of -240 °C and critical pressure of 13 bara.

Hydrogen gas, produced as by-product during the sodium hypochlorite generation process, flows through the outlet header to the hydrocyclones.  Hydrogen degassing happens in the hydrocyclones, and hydrogen is diluted by an air blower before venting to atmosphere.  Considering no heat source in the vicinity of the vent stack, the likelihood for small amount of hydrogen to be ignited is limited and any risk impact will only be localized.  As such, the risks associated with sodium hypochlorite generation process have not been modelled in the QRA Study. 

Hydrogen gas is extremely flammable in oxygen and air, and has the widest range of flammable concentrations in air among all common gaseous hydrocarbons.  A limited amount of hydrogen is generated on-board and hence not foreseen to have risk impact on off-site population.  Therefore, hydrogen gas was not further assessed in the QRA Study.

Nitrogen

CAS number of nitrogen is 7727-37-9, nitrogen is a nontoxic, odourless, colorless, non-flammable compressed gas generated on board the FSRU Vessel.  However, it can cause rapid suffocation when concentrations are sufficient to reduce oxygen levels below 19.5%.

The expected off-site impact associated with nitrogen is limited as nitrogen is generated for the purpose of inert gas purging after LNG unloading operation.  Therefore, nitrogen was not further assessed in the QRA Study.

Calibration Gas

The volume of the compressed gas inside the cylinders is limited and the associated inventory available is small, and also those compressed gas cylinders are located at machinery room.  Should loss of containment occur for compressed gas cylinders, there is no off-site impact on the surrounding marine population.  Hence, it is not further assessed in the QRA Study.

Review of Potential MAEs

LNG

The possible hazardous scenarios considered in the QRA Study upon the release of LNG with consideration of operating conditions are:

·      Jet fire;

·      Pool fire;

·      Flash fire; and

·      VCE.

Natural Gas

The possible hazardous scenarios considered in the QRA Study upon the release of high pressure natural gas with consideration of operating conditions are:

·      Jet fire;

·      Flash fire;

·      Fireball; and

·      VCE.

Considering that the Jetty and the regasification unit on board the FSRU Vessel are relatively congested, a VCE may potentially occur if flammable gas cloud accumulate in these congested areas and is ignited, leading to damaging overpressure.
 
Other Dangerous Goods

Considering the high flash point temperature of other dangerous goods such as marine diesel oil present in the FSRU Vessel, the possible hazardous scenarios considered in the QRA Study are a pool fire and flash fire. 

Detailed characteristics and modelling of the above hazardous scenarios are described in Annex 5G.

Review of Relevant Industry Incidents

To investigate further the possible hazardous scenarios from the FSRU Vessel, the Jetty and the LNGC unloading operation, a review of the applicable past industry incidents at similar facilities worldwide was conducted based on the following incident/ accident database:

·      Institution of Chemical Engineers (IChemE) accident database;

·       eMARS ([19]);

·       ERNS ([20]);

·      MHIDAS database ([21]); and

·      SIGTTO ([22]).

Details of the past industry incident analysis are presented in Annex 5C.

Review of Potential Initiating Events Leading to MAEs

The potential hazardous scenarios arising from the LNG Terminal were identified as loss of containment of LNG, natural gas and other dangerous goods.  The potential initiating events which could result in the loss of containment of flammable material including LNG, natural gas and diesel are listed below:

·      Collision with other passing / visiting marine vessels;

·      Mooring line failure;

·      Dropped objects from crane operations on FSRU Vessel;

·      General equipment/piping failure (due to corrosion, construction defects etc.);

·      Sloshing;

·      LNG containment system failure; and

·      External effects - adverse weather (typhoon, poor visibility, storm surge, extreme tide), tsunami, lightning, aircraft crash and helicopter crash.

Descriptions of the potential initiating events are presented in Annex 5D.

HAZID Workshop

A HAZID workshop was conducted to confirm and further identify the potential initiating events which may lead to MAEs at the LNG Terminal based on the HAZID team representative’s experience, past industry accidents, lessons learnt and guideword checklists.  The HAZID workshop worksheet is summarised in Annex 5E.  The HAZID workshop output served as a basis for the identification of potential initiating events and hazardous scenarios for the QRA Study.

Jetty Collision Analysis

The collision frequency at the Jetty was estimated based on the frequency of marine vessels that are likely to be in the vicinity of the LNG Terminal.  As a conservative approach, the ship collision frequency in Segment “b” in Table 5.4 (Approaching the LNG Terminal) was adopted as the Jetty collision frequency in the QRA Study.

Identification of Hazardous Sections

A total of twenty five (25) hazardous sections were identified from the LNG Terminal, with consideration of the location of emergency shutdown valves and process conditions (e.g. operating temperature and pressure).  The details of each hazardous section (including temperature, pressure, flow rate, etc.) are summarised in Annex 5D.  These hazardous sections formed the basis for the development of loss of containment scenarios.

5.5.3                           Frequency Analysis

Release Frequency Database

The historical database from the International Association of Oil and Gas Producers (OGP) ([23]) was adopted in the QRA Study for estimating the release frequency of hazardous scenarios associated with the LNG Terminal.  The release frequency in OGP is based on the analysis of the HSE hydrocarbon release database (HCRD) which collected all offshore releases of hydrocarbon in the UK (including the North Sea) reported to the HSE Offshore Division from 1992-2006.  Considering that the LNG Terminal is located in an offshore environment in HKSAR waters, this database was considered adequate for purpose of this QRA Study.

The release frequencies of various equipment items are summarised in Table 5.11, and the detailed discussion on the failure frequency is presented in Annex 5F.

Table 5.11      Release Frequency

Equipment

Release Scenario

Release Phase

Release Frequency

Unit

Reference

Piping 2” to 6”

10 mm hole

Liquid/ Gas

3.45E-05

per metre per year

OGP

 

25 mm hole

Liquid/ Gas

2.70E-06

per metre per year

OGP

 

50 mm hole

Liquid/ Gas

6.00E-07

per metre per year

OGP

Piping 8” to 12”

10 mm hole

Liquid/ Gas

3.06E-05

per metre per year

OGP

 

25 mm hole

Liquid/ Gas

2.40E-06

per metre per year

OGP

 

50 mm hole

Liquid/ Gas

3.70E-07

per metre per year

OGP

 

>150 mm hole

Liquid/ Gas

1.70E-07

per metre per year

OGP

Piping 14” to 18”

10 mm hole

Liquid/ Gas

3.05E-05

per metre per year

OGP

 

25 mm hole

Liquid/ Gas

2.40E-06

per metre per year

OGP

 

50 mm hole

Liquid/ Gas

3.60E-07

per metre per year

OGP

 

>150 mm hole

Liquid/ Gas

1.70E-07

per metre per year

OGP

Piping 20” to 24”

10 mm hole

Liquid/ Gas

3.04E-05

per metre per year

OGP

 

25 mm hole

Liquid/ Gas

2.40E-06

per metre per year

OGP

 

50 mm hole

Liquid/ Gas

3.60E-07

per metre per year

OGP

 

>150 mm hole

Liquid/ Gas

1.60E-07

per metre per year

OGP

Piping 26” to 48”

10 mm hole

Liquid/ Gas

3.04E-05

per metre per year

OGP

 

25 mm hole

Liquid/ Gas

2.30E-06

per metre per year

OGP

 

50 mm hole

Liquid/ Gas

3.60E-07

per metre per year

OGP

 

>150 mm hole

Liquid/ Gas

1.60E-07

per metre per year

OGP

Pressure Vessel - Large Connection (> 6”)

10 mm hole

Liquid/ Gas

5.90E-04

per year

OGP

25 mm hole

Liquid/ Gas

1.00E-04

per year

OGP

50 mm hole

Liquid/ Gas

2.70E-05

per year

OGP

>150 mm hole

Liquid/ Gas

2.40E-05

per year

OGP

Pump Centrifugal - Small Connection (up to 6”)

 

10 mm hole

Liquid

4.40E-03

per year

OGP

25 mm hole

Liquid

2.90E-04

per year

OGP

50 mm hole

Liquid

5.40E-05

per year

OGP

Pump Centrifugal - Large Connection (> 6”)

 

10 mm hole

Liquid

4.40E-03

per year

OGP

25 mm hole

Liquid

2.90E-04

per year

OGP

50 mm hole

Liquid

3.90E-05

per year

OGP

>150 mm hole

Liquid

1.50E-05

per year

OGP

Compressor Reciprocating - Large Connection (> 6”)

 

10 mm hole

Gas

3.22E-02

per year

OGP

25 mm hole

Gas

2.60E-03

per year

OGP

50 mm hole

Gas

4.00E-04

per year

OGP

>150 mm hole

Gas

4.08E-04

per year

OGP

Shell and Tube Heat Exchanger - Large Connection (> 6”)

 

10 mm hole

Liquid/Gas

1.20E-03

per year

OGP

25 mm hole

Liquid/Gas

1.80E-04

per year

OGP

50 mm hole

Liquid/Gas

4.30E-05

per year

OGP

>150 mm hole

Liquid/Gas

3.30E-05

per year

OGP

Unloading Arm

10 mm hole

Liquefied Gas

4.00E-06*

per transfer operation

UK HSE ([24])

 

25 mm hole

Liquefied Gas

4.00E-06*

per transfer operation

UK HSE (1)

 

>150 mm hole

Liquefied Gas

7.00E-06

per transfer operation

UK HSE (1)

Riser

10 mm hole

Gas

7.2E-05

per year

OGP

 

25 mm hole

Gas

1.8E-05

per year

OGP

 

>150 mm hole

Gas

3.0E-05

per year

OGP

Diesel Storage Tank

10 mm hole

Liquid

1.6E-03

per year

OGP

25 mm hole

Liquid

4.6E-04

per year

OGP

50 mm hole

Liquid

2.3E-04

per year

OGP

Rupture

Liquid

3.0E-05

per year

OGP

Unloading Hose

10 mm hole

Liquid

1.3E-05#

per hour

Purple  Book ([25])

 

25 mm hole

Liquid

1.3E-05

per hour

Purple Book

 

50 mm hole

Liquid

1.3E-05

per hour

Purple Book

 

Rupture

Liquid

4.0E-06

per hour

Purple Book

LNG Storage Tank

10 mm hole

Liquid

3.3E-06!

per year

OGP

 

25 mm hole

Liquid

3.3E-06!

per year

OGP

 

50 mm hole

Liquid

3.3E-06!

per year

OGP

 

Rupture

Liquid

2.5E-08

per year

OGP

*Notes: The leak frequency of unloading arm, presented in the UK HSE (1), has been evenly distributed into 10 mm and 25 mm hole sizes.

#Notes: The leak frequency of unloading hose, presented in the Purple Book (2), has been evenly distributed into 10 mm, 25 mm and 50 mm hole sizes.

!Notes: The leak frequency of LNG storage tank, presented in OGP, has been evenly distributed into 10 mm, 25 mm and 50 mm hole sizes.

Release Hole Sizes

The release hole sizes presented in Table 5.12, which are consistent with the OGP ([26]) database, were adopted in the QRA Study.

Table 5.12      Hole Sizes Considered in the QRA Study

Leak Description

Hole Size

Very Small Leak

10 mm

Small Leak

25 mm

Medium Leak

50 mm

Rupture

>150 mm

 

Flammable Gas Detection and Emergency Shutdown Probability

With reference to the Purple Book ([27]), the effect of block valve system is determined by various factors, such as the position of gas detection monitors and the distribution thereof over the various wind directions, the direction limit of the detection system, the system reaction time and the intervention time of an operator.  The probability of failure on demand of the block valve system as a whole is 0.01 per demand.

Considering that the LNG Terminal is provided with gas detection systems and automatic emergency shutdown systems, the probability of executing the isolation successfully when required was selected as 99% in the QRA Study.

Ignition Probability

The immediate ignition probability was estimated based on offshore ignition scenarios No. 24 from the OGP Ignition Probability Database (1).  For flammable liquids with flash point of 55°C or higher (e.g. diesel, fuel oil etc.), a modification factor of 0.1 was applied to reduce the ignition probability as suggested in OGP (1).

The delayed ignition for various ignition sources was referred to Appendix 4.A of the Purple Book (2).

Probability of Vapour Cloud Explosion

The probability of explosion given an ignition was taken from the Cox, Lees and Ang model ([28]), as shown in Table 5.13.  VCE occurs upon a delayed ignition from a flammable gas release at a congested area.  Details of the identified congested area and congestion volume are provided in Annex 5G.

Table 5.13      Probability of Explosion

Leak Size (Release Rate)

Explosion Probability

Minor (< 1 kg s-1)

0.04

Major (1 – 50 kg s-1)

0.12

Massive (> 50 kg s-1)

0.30

 

Escalation

An initially small release may escalate into a larger, more serious event if a jet fire or pool fire impinges on neighbouring equipment/ piping for an extended time.  This is taken into account in the modelling for isolation fail branch of the event tree, depicted in Figure 5F.3. 

If neighbouring equipment/ piping is within range of the jet fire event flame zone, an escalation probability of 1/6 ([29]) ([30]) has been taken to conservatively estimate the directional probability and chance of impingement.  In case pool fire events, the escalation probability was conservatively estimated without considering any directional probability.

Escalation has been assumed to cause only a full bore rupture of the affected equipment and piping, leading to fireball event as the worst-case scenario.

Event Tree Analysis

An event tree analysis was performed to model the development of each hazardous scenario outcome (jet fire, pool fire, flash fire, fireball and VCE) from an initial release scenario.  The event tree analysis considered whether there is immediate ignition, delayed ignition or no ignition, with consideration of the associated ignition probability as discussed above.  The development of the event tree is presented in Annex 5F.

5.5.4                           Consequence Analysis

Source Term Modelling

PHAST was used to estimate the release rates, which were used to determine the ignition probability.  Source term modelling was carried out to determine the maximum (e.g. initial) release rate that may be expected should a loss of containment occur.

Release Duration

With reference to the Purple Book ([31]), the closing time of an automatic block valve system is two (2) minutes; hence a release duration of two (2) minutes was adopted for isolation success case in the QRA Study.  Detection and shutdown system may however fail due to some reasons.  As per the Purple Book (3), the release duration is limited to a maximum of thirty (30) minutes, therefore this was conservatively adopted for the isolation failure case in the QRA Study.

Release Direction

The orientation of a release can have some effects on the hazard footprint calculated by PHAST.  The models take into account the momentum of the release, air entrainment, vapourisation rate and liquid rainout fraction.

“Horizontal non-impinging” was selected for modelling the jet fire and flash fire scenarios since the associated hazard footprint is more conservative.  “Downward impinging” was selected for modelling the pool fire scenario since the momentum of the release is reduced, thereby increasing the liquid rainout fraction.

Physical Effects Modelling

PHAST was used to perform the physical effects modelling to assess the effects zones for the following hazardous scenarios:

·      Jet fire;

·      Pool fire;

·      Flash fire;

·      Fireball; and

·      VCE.

Detailed description of the physical effects modelling is presented in Annex 5G.

Consequence End-Point Criteria

Similarly, as stated in Section 5.4.4, the probit equation was used for assessing the thermal radiation impact and the end point criteria for flash fire, were also adopted in the QRA Study for the LNG Terminal.

The fatality rate within the fireball diameter is assumed to be 100%.

In terms of overpressure, a relatively high overpressure is necessary to lead to significant fatalities for persons outdoor.  Indoor population tends to have a higher harm probability due to the risk of structural collapse and flying debris such as breaking windows.  Table 5.14 presents the explosion overpressure levels from the Purple Book ([32]), which were adopted in the QRA Study.

Table 5.14      Effect of Overpressure - Purple Book

Explosion Overpressure (barg)

Fraction of People Dying

 

Indoor

Outdoor

> 0.3

1.000

1.000

> 0.1 to 0.3

0.025

0.000

Details of the consequence modelling results are presented in Annex 5G.

5.5.5                           Risk Summation 

The risk summation for the LNG Terminal was modelled using SAFETI, which is in line with the previous EIA Report that was approved by EPD ([33]).  The hazardous scenarios, the associated frequencies, meteorological data, surrounding off-site population data, and suitable modelling parameters identified were input into SAFETI.

Individual Risk Results

Before commissioning, no LNG, natural gas and other dangerous goods are present at the LNG Terminal in the baseline condition (Year 2017); therefore, the baseline condition assessment was not considered in the QRA Study.  In addition, there is no other hazardous installation such as a Potentially Hazardous Installation (PHI) in the vicinity of the LNG Terminal which leads to an increase in cumulative risk.

The cumulative risk was calculated by summing various types of process risks from the LNG Terminal, including the FSRU Vessel, the Jetty and LNGC unloading operation as summarised in Table 5.15.

Table 5.15      Potential Risks Considered in the Cumulative Risk Assessment

Potential Risk

2020

2030

FSRU Vessel, covering LNG, natural gas, other dangerous goods, for all operating modes

Yes

Yes

Jetty with topsides equipment, for all operating modes

Yes

Yes

LNGC during LNG unloading operations

Yes

Yes

 

The individual risk contours for the LNG Terminal are shown in Figure 5.7 to Figure 5.8.  The individual risk contour of 1 × 10-5 per year in the Operational Year in 2020 and Future Scenario Year in 2030 is identified to be within the Safety Zone within the Project Site (Section 3.3).  All access into the Project Site must be authorized by the Terminal Operator such that no off-site population will be present within the Project Site.  Consequently, the maximum level of off-site individual risk (i.e. outside Safety Zone of the LNG Terminal) associated with the Hong Kong Offshore LNG Terminal Project would not exceed 1×10-5 per year and the individual risk criterion stipulated in Section 2 of Annex 4 of the EIAO-TM is met. 

Societal Risk Results

The societal risk for the LNG Terminal was calculated based on the associated process risks and the surrounding off-site marine traffic populations in the vicinity of the Project Site.  The societal risk, in terms of F-N curves, for Operational Year in 2020 and Future Scenario Year in 2030, as shown in Figure 5.9, lie within the Acceptable Region, as such the societal risk criteria stipulated in Section 2 of Annex 4 of the EIAO-TM are met. 

The key risk contributors, in terms of potential loss of life, associated with the LNG Terminal are summarised in Annex 5H.

5.5.6                           Conclusions of QRA Study for the LNG Terminal  

The individual risk associated with the LNG Terminal is in compliance with the individual risk criterion stipulated in Section 2 of Annex 4 of the EIAO-TM. 

The societal risks, in terms of F-N curves, are within the Acceptable Region and are in compliance with societal risk criteria stipulated in Section 2 of Annex 4 of the EIAO-TM.

5.6                                   QRA Study for the Subsea Pipelines

This Section presents the QRA Study for the evaluation of the risks associated with the subsea BPPS and LPS Pipelines from the Jetty to the proposed GRS facilities at the BPPS and the LPS.

5.6.1                           Description of Subsea Pipelines

Subsea BPPS Pipeline

The 30” diameter subsea BPPS Pipeline route commences at the tie-in point to the pig launcher on the Jetty and ends at the proposed GRS at the BPPS.   The length of the subsea BPPS pipeline is approximately 45 km.

Subsea LPS Pipeline

The 20” diameter subsea LPS Pipeline route commences at the tie-in point to the pig launcher on the Jetty and ends at the proposed GRS at the LPS.  The length of the subsea LPS pipeline is approximately 18 km.

Segmentation of Pipeline Route

In order to examine the risks along the BPPS and LPS Pipeline, it is first necessary to identify the types of vessel that will traverse across the pipeline then segregate the alignment accordingly for assessment and this has been defined as “segment”.  This has been done on the basis of water depth which is linked with marine traffic activity (summarised from Table 5A.7 to Table 5A.10 of Annex 5A), as water depth (indicated in Figure 1.2 of Annex 7B) is a constraint to navigation.  The water depth along both BPPS and LPS Pipeline route was reviewed within a GIS system based on Hong Kong Electronic Navigational Chart. 

The segmentation is also used for trench design of pipeline routes; the principle being that the busier pipeline sections with the larger vessels (hence larger anchor size) required a greater level of trench design and rock berm pipeline protection in these BPPS Pipeline and LPS Pipeline segments.

Based on above considerations, the BPPS and LPS Pipeline routes were divided into eleven (11) and four (4) segments respectively as shown in Figure 5.10 and are presented in Table 5.16 and Table 5.17 respectively.

Table 5.16      BPPS Pipeline Segmentation*

Segment

KP- From  (km)

KP- To  (km)

Water Depth#

(m)

Length

(km)

Trench type*

Protection Anchor Size (T) ([34])

X

Jetty Approach to South of Soko Islands

0.0

8.9

15

9.2

8.9

4

<24

A

Southwest of Soko Islands

8.9

12.1

9.2

12.0

3.2

5

<5

B

Southwest of Fan Lau

12.1

15.6

12.0

12.9

3.5

5

<5

C

Southwest Lantau

15.6

21.3

12.9

10.6

5.7

2

< 7.3

D

West of Tai O

21.3

26.2

10.6

7.7

4.9

5

<5

E

West of HKIA

26.2

31.5

7.7

4.6

5.3

5

<5

F

West of Sha Chau

31.5

36.0

4.6

6.0

4.5

5

<5

G

West of Lung Kwu Chau

36.0

37.5

6.0

3.1

1.5

3

< 10.6

H

Lung Kwu Chau to Urmston Anchorage

37.5

41.1

3.1

4.8

3.6

5

<5

I

Urmston Road

41.1

42.9

4.8

5.6

1.8

4

< 24

J

West of BPPS

42.9

45.0

5.6

1.6

2.1

5/1

<5

*Proposed pipeline construction methods and trench types are subject to further review with government departments. The lowest protection anchor size for each segment was used for the risk modelling.

#:Source: 2016, Charts for Local Vessels, The Hydrographic Office Marine Department; depths are measured at KP Point in meters and are reduced to Chart Datum

 

Table 5.17      Subsea LPS Pipeline Segmentation*

Segment

KP- From  (km)

KP- To  (km)

Water Depth#

(m)

Length

(km)

Trench type*

Protection Anchor Size (T) (1)

A

Jetty Approach to South of Shek Kwu Chau

0.0

5.0

15.0

14.4

5.0

4

<24

B

South of Cheung Chau

5.0

14.5

14.4

15.5

9.5

5

<5

C

West Lamma Channel

14.5

17.4

15.5

15.7

2.9

5

<5

D

Alternative Shore Approach

17.4

18.2

15.7

2.6

0.8

1

<5

*Proposed pipeline construction methods and trench types are subject to further review with government departments.

#:Source: 2016, Charts for Local Vessels, The Hydrographic Office Marine Department; depths are measured at KP Point in meters and are reduced to Chart Datum   

Pipeline Protection

Different levels of armour rock protection will be provided for each segment of the proposed BPPS and LPS Pipelines based on the identified potential anchor drag and drop hazards.  The cross sections of the trench designs and associated armour rock protection are illustrated in Section 3 of the EIA Report.

With consideration of the armour rock protection for the subsea pipelines, the following pipeline protection factors ([35]) were adopted in the QRA Study:

·      99.99% is applied, if the vessel anchor size is smaller than the intended design capacity of the pipeline protection; and

·      0% is applied, if the vessel anchor size is larger than the intended design capacity of the pipeline protection.

5.6.2                           Hazard Identification

The hazardous scenarios associated with the two subsea pipelines were identified through the following tasks:

·      Review of hazardous materials;

·      Review of potential MAEs;

·      Review of relevant industry incidents;

·      Review of potential initiating events leading to MAEs; and

·      HAZID Workshop.

Review of Hazardous Materials

The high pressure natural gas is sent out from the LNG Terminal through the BPPS and LPS subsea Pipelines to the GRSs at the BPPS and the LPS.  The only identified hazard arising from loss of containment of high pressure natural gas from these subsea pipelines is flash fire.  The properties of natural gas have been described in Section 5.5.2.

Review of Potential MAEs

In the event of any leakage or rupture of either the BPPS or LPS Pipelines leading to loss of containment, the flammable gas will bubble to the surface of the sea, and then disperse.  The only possible hazardous scenario associated with any leakage or rupture of either of the subsea pipelines is flash fire if the dispersing flammable gas cloud comes in contact with an ignition source, most likely from a passing marine vessel.

Detailed characteristics of the above hazardous effect are described in Annex 5G.

Review of Relevant Industry Incidents

To investigate further the possible hazardous scenarios related to BPPS and LPS pipelines, a review of the applicable past industry incidents at similar facilities worldwide was conducted based on following incident/ accident database:

·      MHIDAS database;

·      Institution of Chemical Engineers (IChemE) accident database;

·      US Gas Pipeline Incident Database;

·      Pipeline and Riser Loss of Containment (PARLOC); and

·      Incident Records for Subsea Pipelines in Hong Kong waters.

Details of the past industry incident analysis are presented in Annex 5C.

Review of Potential Initiating Events Leading to MAEs

The key potential hazardous scenarios arising from the subsea pipelines were identified as the loss of containment of natural gas.  The potential initiating events which could result in the loss of containment of natural gas are listed below:

·      Anchor drop and drag;

·      Grounding;

·      Vessel sinking;

·      Aircraft crash;

·      Fishing activity;

·      Dredging activity;

·      Subsea cable maintenance activity;

·      General pipeline failure (due to corrosion, construction defects etc.);

·      Pressure cycling; and

·      External effects - adverse weather (typhoon, storm surge, extreme tide), subsidence, and tsunami.

Descriptions of the potential initiating events are presented in Annex 5D.

HAZID Workshop

A HAZID workshop was conducted to confirm and further identify the potential initiating events which may lead to MAEs along the subsea pipelines, based on the HAZID team representative’s experience, past industry accidents, lessons learnt and guideword checklists.  The HAZID workshop worksheet is summarised in Annex 5E.  The HAZID workshop output was served as a basis for identification of potential initiating events and hazardous scenarios for the QRA Study.

Identification of Hazardous Sections

The whole BPPS and LPS pipelines were considered as the hazardous section with the consideration of emergency shut down valves available at the Jetty and the GRS area at the BPPS and the LPS respectively.

5.6.3                           Frequency Analysis

Release Frequency Database

The selected release frequency database is consistent with the previous EIA Reports which have been approved by the EPD ([36]) ([37]).  The frequency of the major causes such as corrosion, material defect, anchor drop and impact incidents leading to a loss of containment from the subsea pipelines was estimated based on the international database, including PARLOC 2012 ([38]) and PARLOC 2001 ([39]).  A local incident database was also reviewed and compared with the PARLOC database.  The detailed discussion and analysis of the release frequency estimation are presented in Annex 5F.  The release frequencies adopted in the QRA Study are summarized in Table 5.18 and Table 5.19.

Table 5.18      Summary of Release Frequency along BPPS Pipeline

Pipeline Section

Trench Type

Anchor Impact (/km/yr)

Corrosion/

Defects (/km/yr)

Others (/km/yr)

Total (/km/yr)*

Jetty Approach to South of Soko Islands (X)

4

1.25E-05

1.18E-06

7.90E-07

2.30E-06

Southwest of Soko Islands (A)

5

1.25E-05

1.18E-06

7.90E-07

2.69E-06

Southwest of Fan Lau (B)

5

1.77E-04

1.18E-06

7.90E-07

1.99E-06

Southwest Lantau (C)

2

9.49E-05

1.18E-06

7.90E-07

5.85E-06

West of Tai O (D)

5

9.49E-05

1.18E-06

7.90E-07

1.98E-06

West of HKIA (E)

5

9.49E-05

1.18E-06

7.90E-07

1.98E-06

West of Sha Chau (F)

5

9.49E-05

1.18E-06

7.90E-07

1.98E-06

West of Lung Kwu Chau (G)

3

1.77E-04

1.18E-06

7.90E-07

1.99E-06

Lung Kwu Chau to Urmston  Anchorage (H)

5

1.77E-04

1.18E-06

7.90E-07

1.99E-06

Urmston Road (I)

4

1.77E-04

1.18E-06

7.90E-07

1.99E-06

West of BPPS (J)

5/1

9.49E-05

1.18E-06

7.90E-07

1.98E-06

Note: The armour rock protection factor for the subsea pipeline has been taken into account in the total release frequency.

Table 5.19      Summary of Release Frequency along LPS Pipeline

Pipeline Section

Trench Type

Anchor Impact (/km/yr)

Corrosion/

Defects (/km/yr)

Others (/km/yr)

Total (/km/yr) *

Jetty Approach to South of Shek Kwu Chau (A)

4

1.32E-05

1.18E-06

7.90E-07

1.97E-06

South of Cheung Chau (B)

5

1.32E-05

1.18E-06

7.90E-07

1.97E-06

West Lamma Channel (C)

5

1.32E-05

1.18E-06

7.90E-07

1.97E-06

Alternative Shore Approach (D)

1

5.95E-05

1.18E-06

7.90E-07

1.98E-06

Note: The armour rock protection factor for the subsea pipeline has been taken into account in the total release frequency.

Hole Size Distribution

The international databases (such as PARLOC 2012 and PARLOC 2001) were reviewed.  The hole size distributions for anchor impact scenarios and corrosion/other scenarios are given in Table 5.20 and Table 5.21, are consistent with previous EIA Reports which have been approved by the EPD ([40]).  Detailed analysis of the derivation of the hole size distribution is provided in Annex 5F.

 

Table 5.20      Hole Size Distribution for Anchor Impact Cases

Category

Hole Size

(Subsea BPPS Pipeline)

Hole Size

(Subsea LPS Pipeline)

Proportion

Rupture (Full Bore)

Full Bore

Full Bore

10%

Major

15”or 381 mm (Half Bore)

10”or 254 mm (Half Bore)

20%

Minor

4”or 100 mm

4” or 100 mm

70%

 

Table 5.21      Hole Size Distribution for Corrosion and Other Failure Cases

Category

Hole Size

(Subsea BPPS Pipeline)

Hole Size

(Subsea LPS Pipeline)

Proportion

Rupture (Half Bore)

15” or 381 mm

10” or 254 mm

5%

Puncture

4”or 100 mm

4”or 100 mm

15%

Hole

2”or 50 mm

2”or 50 mm

30%

Leak

<25 mm

<25 mm

50%

 

Ignition Probability

The ignition of any release natural gas is expected only from passing vessels in the vicinity of either subsea pipeline.  The ignition probabilities consistent with previous EIA Reports which have been approved by the EPD ([41]) ([42]), were adopted in the QRA Study and are summarised in Table 5.22.

Table 5.22      Ignition Probability for Subsea Pipeline

Release Case

Ignition Probability

 

Passing Vessels*

Vessels in Vicinity#

<25 mm

0.01

n/a

50 mm

0.05

n/a

100 mm

0.10

0.15

Half bore

0.20

0.30

Full bore

0.30

0.40

Note:

*: Values applied to passing vessels for all types of incidents, i.e. corrosion, others and anchor impact.

#: Values applied only to scenarios where the vessel causing pipeline damage due to anchor impact is still in the vicinity.

Event Tree Analysis

An event tree analysis was performed to model the development of each hazardous scenario outcome (flash fire) from an initial release scenario.  The event tree analysis considered whether there is delayed ignition or no ignition, with consideration of the associated ignition probability as discussed above.  The development of the event tree analysis is presented in Annex 5F.

5.6.4                           Consequence Analysis

Source Term Modelling

The release rate was estimated based on standard equations for discharge through an orifice.  As per previous EIA Reports which have been approved by the EPD ([43]) ([44]), for large release with hole size greater than 100 mm, the empirical correlation developed by Bell and modified by Wilson ([45]) was adopted in the QRA Study.  Detailed explanation of source term modelling is provided in Annex 5G.

Dispersion Modelling For Subsea Pipeline Releases

In the event of a flammable gas release from the subsea pipelines, the flammable gas will bubble to the sea surface and disperse.  As per previous EIA Reports (1) (2) which have been approved by the EPD, a simple cone model was adopted to determine the release area on the sea surface.  For the deepest water depth (i.e. about 25 m around Southwest of Fan Lau) along the subsea pipelines, it was predicted by the cone model that the diameter of the release area was about 10 m.  Detailed explanation of the cone model is provided in Annex 5G.

Dispersion above Sea Surface

The flammable gas disperses into atmosphere upon reaching the sea surface.  The distance to which the flammable gas envelope extends depends on ambient conditions such as wind speed and atmospheric stability as well as source conditions.  As per previous EIA Reports which have been approved by the EPD (1) (2), the extent of the flammable area was taken as the distance to 0.85 LFL.  PHAST was used to model the plume dispersion as an area source on the sea surface.  Detailed explanation of the dispersion modelling above sea surface is provided in Annex 5G.

Consequence and Impact Assessment

The consequence and impact assessment, as described below, was conducted as per previous EIA Reports which have been approved by the EPD (1) (2).

Impact on Marine Vessel Population

A flash fire could cause injury to personnel on marine vessels.  It may also cause secondary fires on the marine vessel.  If a marine vessel passes close to the ‘release area’ (where bubbles of natural gas break through the sea surface), the consequences will be more severe and a 100% fatality probability was taken for this scenario.  Once a fire has ignited, it is presumed that no further marine vessels will be involved because the fire will be visible and other marine vessels can take action to avoid the area.  In other words, at most only one marine vessel may be affected.

The hazardous impact area of the flammable cloud was taken to be the distance to 0.85 LFL.  Taking into account the protection factors of various types of marine vessel, the fatalities adopted in the QRA Study are as given in Table 5.23.

Table 5.23      Fatality Probability for Subsea Pipelines’ MAEs

Marine Vessel Class

Fatality Probability

 

Release Area

Cloud Area

Fishing vessels

1

0.9

Rivertrade coastal vessel

1

0.3

Ocean-going vessels

1

0.1

Fast launches

1

0.9

Fast ferries

1

0.4

Others

1

0.3

Note: Release area indicates the area where gas bubbles break through the sea surface; and cloud area indicates the hazardous distance of 0.85 LFL of the flammable gas cloud.

In addition, the probability that a marine vessel will pass through the flammable plume was calculated based on the size of the plume (obtained from dispersion modelling) and the marine traffic density.

Detailed discussion on the estimation of the above probabilities is presented in Annex 5G.

Impact on Road Traffic Population on Hong Kong Link Road

The BPPS Pipeline will pass under the Hong Kong Link Road (HKLR) at a location within the West of Tai O Section.  The transient road traffic population on the bridge may be affected if a flammable gas cloud is ignited under / in the vicinity of the bridge area.  This hazardous scenario was considered in the consequence analysis for the West of Tai O Section of the BPPS Pipeline.  The associated risk impact did not make a significant contribution to the overall risk results.  Detailed assessment result is presented in Annex 5G.

Impact of Aircraft approaching Hong Kong International Airport

The West of HKIA Section of the BPPS Pipeline is located in the vicinity of the Hong Kong International Airport (HKIA).  Large gas releases from the BPPS Pipeline, such as those that occur from a full bore or half bore rupture, may have the potential to produce a flammable gas cloud that extends higher than 200 m.  It is therefore possible that an aircraft on its approach to landing may pass through a gas cloud within the flammability limits.  This scenario was considered in the consequence analysis and it was observed that the associated risk did not make a significant contribution to the overall risk results.  Detailed assessment result is presented in Annex 5G.
 
Impact on Macau Helicopters

Helicopters shuttling to and from Macau pass over the Southwest of Fan Lau Section of the BPPS Pipeline at about 500 feet (150 m) altitude.  Similarly, the above large gas releases may impact on the helicopters.  The hazard distance was taken to be the maximum width of the gas cloud above 150 m altitude.   The associated risk did not make a significant contribution to the overall risk results.  Detailed assessment result is presented in Annex 5G.

Consequence Results

The hazard distances that were used in the QRA Study were determined from the gas dispersion modelling.  The hazard distance for marine vessels was defined as the maximum width of the gas cloud below a height of 50 m above sea level.  Similarly, the hazard distance for aircraft was defined above as 200 m and for helicopters was defined as above 150 m from sea level.  The hazard distances obtained from dispersion modelling are summarised in Annex 5G.

5.6.5                           Risk Summation

The risk summation for the BPPS and LPS Pipelines combines the estimation of the consequences of an event with the event probabilities to give an estimate of the resulting frequency of varying levels of fatalities.  Risk summation was implemented in ERM’s proprietary risk integration package, which took into account input data for initiating event frequency, event frequency, event tree branch probabilities, number of exposed persons and fatality probability.

Individual Risk Results

The individual risk contours associated with the BPPS Pipeline and LPS Pipeline are shown in Table 5.24, Table 5.25, Table 5.26 and Table 5.27  respectively for Operational Year in 2020 and Future Scenario Year in 2030.

The individual risk contour of 1 × 10-5 per year was not reached for all sections of the subsea BPPS and LPS Pipelines in both assessment years, as such the individual risk criterion stipulated in Section 2 of Annex 4 of the EIAO-TM is met for the current proposed subsea pipeline design.


 

Table 5.24      Risk Results for BPPS Pipeline in 2020 – Operational Year

Segment

IR (/km/year)

IR (/year)

X

Jetty Approach to South of Soko Islands

3.53 × 10-8

3.14 × 10-7

A

Southwest of Soko Islands

4.20 × 10-8

1.34 × 10-7

B

Southwest of Fan Lau

7.80 × 10-9

2.73 × 10-8

C

Southwest Lantau

3.41 × 10-7

1.94 × 10-6

D

West of Tai O

3.10 × 10-8

1.52 × 10-7

E

West of HKIA

3.59 × 10-9

1.90 × 10-8

F

West of Sha Chau

1.18 × 10-9

5.31 × 10-9

G

West of Lung Kwu Chau

3.81 × 10-9

5.72 × 10-9

H

Lung Kwu Chau to Urmston Anchorage

3.54 × 10-9

1.27 × 10-8

I

Urmston Road

3.80 × 10-8

6.84 × 10-8

J

West of BPPS

5.69 × 10-9

1.19 × 10-8

 

Table 5.25      Risk Results for LPS Pipeline in 2020 – Operational Year

Segment

IR (/km/year)

IR (/year)

A

Jetty Approach to South of Shek Kwu Chau

5.63 × 10-9

2.82 × 10-8

B

South of Cheung Chau

2.01 × 10-8

1.91 × 10-7

C

West Lamma Channel

6.49 × 10-8

1.88 × 10-7

D

Alternative Shore Approach

1.41 × 10-7

1.13 × 10-7

 

Table 5.26      Risk Results for BPPS Pipeline in 2030 –Future Scenario Year

Segment

IR (/km/year)

IR (/year)

X

Jetty Approach to South of Soko Islands

3.56 × 10-8

3.17 × 10-7

A

Southwest of Soko Islands

4.23 × 10-8

1.35 × 10-7

B

Southwest of Fan Lau

8.78 × 10-9

3.07 × 10-8

C

Southwest Lantau

3.41 × 10-7

1.94 × 10-6

D

West of Tai O

3.12 × 10-8

1.53 × 10-7

E

West of HKIA

4.14 × 10-9

2.19 × 10-8

F

West of Sha Chau

1.48 × 10-9

6.66 × 10-9

G

West of Lung Kwu Chau

4.40 × 10-9

6.60 × 10-9

H

Lung Kwu Chau to Urmston Anchorage

4.06 × 10-9

1.46 × 10-8

I

Urmston Road

4.21 × 10-8

7.58 × 10-8

J

West of BPPS

6.51 × 10-9

1.37 × 10-8

 

 

Table 5.27      Risk Results for LPS Pipeline in 2030 – Future Scenario Year

Segment

IR (/km/year)

IR (/year)

A

Jetty Approach to South of Shek Kwu Chau

6.93 × 10-9

3.47 × 10-8

B

South of Cheung Chau

2.23 × 10-8

2.12 × 10-7

C

West Lamma Channel

6.62 × 10-8

1.92 × 10-7

D

Alternative Shore Approach

1.51 × 10-7

1.21 × 10-7

 

Societal Risk Results

The societal risk in terms of F-N curves for all sections of the BPPS and LPS Pipelines in Operational Year 2020 and Future Scenario Year 2030 lie within the Acceptable Region, as shown from Figure 5.11 to Figure 5.14.  Therefore, the societal risk criteria stipulated in Section 2 of Annex 4 of the EIAO-TM are met.

The societal risk, in terms of potential loss of life, associated with each segment of the BPPS and LPG subsea pipelines are summarised in Annex 5H.

5.6.6                           Uncertainty Analysis

An uncertainty analysis was conducted to assess the sensitivity of the subsea pipeline protection factors adopted in the QRA Study.  In the uncertainty analysis, the subsea pipeline protection factors adopted in previous EIA Report that has been approved by the EPD ([46]) were adopted as shown in Table 5.28.  This effectively considers a worst case protection factor for the subsea pipeline protection factor.

Table 5.28      Subsea Pipeline Protection Factors for Uncertainty Analysis

Anchor Size

Trench Type Design

Subsea Pipeline Protection Factor

<2 tonnes

Protect against 20 tonnes

99.9%

>2 tonnes

Protect against 20 tonnes

99.0%

<2 tonnes

Protect against 2 tonnes

99.0%

>2 tonnes

Protect against 2 tonnes

50.0%

 

BPPS Pipeline

Based on the uncertainty analysis results, even if a worst case protection factor is assumed, the individual risk remains less than 1 ×  10-5 per year for all segments of the BPPS Pipeline.  The societal risk remains within the Acceptable Region in Year 2020 (Figure 5.15) for the uncertainty analysis.  In Year 2030 the societal risk generally remains within the Acceptable Region (Figure 5.16) for the uncertainty analysis.  Therefore the risk criteria stipulated in Section 2 of Annex 4 of the EIAO-TM are met.

The BPPS Pipeline segments, including Southwest of Fan Lau and Lung Kwu Chau to Urmston Anchorage, were identified as the highest risk because of those segments are with high marine traffic (high failure frequency) and high marine traffic population.

LPS Pipeline

Based on the uncertainty analysis results, even if a worst case protection factor is assumed, the individual risk remains less than 1 ×  10-5 per year for all segments of the LPS Pipeline.  The societal risk was within the Acceptable Region in Year 2020 and Year 2030, as shown in Figure 5.17 and Figure 5.18.   Therefore the risk criteria stipulated in Section 2 of Annex 4 of the EIAO-TM are met.

The LPS Pipeline segments, Alternative Shore Approach and West Lamma Channel, were identified as the relatively high risk because this segments are with relatively high failure frequency considering marine traffic and segment length.

5.6.7                           Conclusions of QRA Study for Subsea Pipelines

It is concluded that the risks associated with the BPPS and LPS Pipelines in terms of individual risk and societal risk are in compliance with risk criteria stipulated in Section 2 of Annex 4 of the EIAO-TM for the current proposed subsea pipeline design.

5.7                                   QRA Study for the Gas Receiving Stations at the BPPS and LPS

This section presents the QRA Study for the evaluation of the risks associated with the GRSs at the BPPS and the LPS.

5.7.1                           Description of the BPPS Gas Receiving Station

The GRS at the BPPS receives high pressure natural gas (at the maximum allowable operating pressure of 88 barg at 5 °C) transported through the 30” subsea BPPS Pipeline from the Jetty.  The maximum flow rate is 700 mmscfd, and the GRS controls the pressure that the natural gas enters the BPPS, prior to entering the gas turbines for power generation.  The section of the interconnecting onshore gas pipeline within the GRS site boundary was also considered in the QRA Study. 

The major equipment items associated with the GRS include (the detailed process description is provided in Annex 5B):

·      Pig Receiver;

·      Gas Filter;

·      Gas Metering;

·      Pipeline Gas Heater;

·      Pressure Reduction Skid; and

·      Mixing Station.

5.7.2                           Description of the LPS Gas Receiving Station

The GRS at the LPS receives high pressure natural gas (at the maximum allowable operating pressure of 88 barg at 5 °C) transported through the 20” subsea LPS Pipeline from the Jetty.  The maximum total flow rate is 254 mmscfd.  Four (4) natural gas conditioning trains are provided at the GRS, with maximum flow rate of 63.5 mmscfd for each train.  The GRS allows final natural gas conditioning prior to entering the gas turbines for power generation.  The section of the interconnecting onshore gas pipeline within the GRS site boundary was also considered in the QRA Study. 

The major equipment items associated with the GRS include the detailed process description is provided in Annex 5B.

·      Pig Receiver;

·      Gas Filter;

·      Gas Metering;

·      Mixer;

·      Water Bath Heater; and

·      Pressure Reduction Skid.

Key Safety Systems for GRSs at the BPPS and the LPS

The following safety systems are provided at the GRSs at the BPPS and LPS, and the detailed description of the safety system is provided in Annex 5B.

·      Emergency Shutdown System;

·      Blowdown System;

·      Overpressure Protection System; and

·      Fire and Gas Detection System.

5.7.3                           Hazard Identification

The hazardous scenarios associated with the operation of the GRSs at the BPPS and LPS were identified through the following tasks:

·      Review of hazardous materials;

·      Review of potential MAEs;

·      Review of relevant industry incidents;

·      Review of potential initiating events leading to MAEs; and

·      HAZID Workshop.

Review of Hazardous Materials

Natural gas is received at the GRSs at the BPPS and LPS before being sent out to the gas turbines for power generation.  The properties of natural gas have been described in Section 5.5.2.

Other Non-Fuel Gas Dangerous Goods

Calibration gas and carrier gas cylinders will be provided at the GRSs at the BPPS and LPS for Gas Chromatography (GC).  The types of gas cylinders provided in the proposed GRS at the BPPS and LPS are given in Table 5.29 and Table 5.30 respectively.

Table 5.29      Non-Fuel Gas Dangerous Goods Associated with the Proposed GRS at the BPPS

Chemical

Dangerous Goods Classification (1)

Maximum Cylinder Quantity

Cylinder Volume (m3)

Storage Pressure (barg)

Calibration Gas (2)

Category 2

2 cylinders

0.02 m3 per cylinder

137

Helium Gas

Category 2, Cl.1

4 cylinders

0.07 m3 per cylinder

137

Calibration Gas (3)

Category 2

1 cylinders

0.07 m3 per cylinder

137

Calibration Gas (4)

Category 2

1 cylinders

0.07 m3 per cylinder

137

Hydrogen Gas

Category 2, Cl.1

2 cylinders

0.07 m3 per cylinder

137

Note:

(1): The dangerous goods category is classified based on “Fire Protection Notice No. 4, Dangerous Goods General” by Fire Services Department. ([47])

(2): The key composition of the calibration gas for Gas Chromatograph is methane (90 vol%), ethane (5 vol%), Nitrogen (2.5 vol%), and carbon dioxide (1 vol%) and propane (1 vol%).

(3): The key composition of the calibration gas 1 for Sulfur analyzer is 27 ppm H2S, balance with Nitrogen.

(4): The key composition of the calibration gas 2 for Sulfur analyzer is 270 ppm H2S, balance with Nitrogen.

 

 

Table 5.30      Non-Fuel Gas Dangerous Goods Associated with the Proposed GRS at the LPS

Chemical

Dangerous Goods Classification (1)

Maximum Cylinder Quantity

Cylinder Volume (m3)

Storage Pressure (barg)

Hydrogen (4)

Category 2, Cl.1

1 cylinder

0.06 m3 per cylinder

150

Nitrogen (5)

Category 2, Cl.1

1 cylinder

0.06 m3 per cylinder

150

Synthetic Air

Category 2, Cl.1

1 cylinder

0.06 m3 per cylinder

150

Reference Gas for H2S Meter (2)

Category 2

1 cylinder

0.03 m3 per cylinder

150

Reference Gas for GC (3)

Category 2

1 cylinder

0.06 m3 per cylinder

150

Helium (5)

Category 2, Cl.1

1 cylinder

0.06 m3 per cylinder

150

Note:

(1): The dangerous goods category is classified based on “Fire Protection Notice No. 4, Dangerous Goods General” by Fire Services Department (1).

(2): The key composition of the reference gas for H2S meter is methane (90 vol%), ethane (10 vol%), H2S (4 ppm).

(3): The key composition of the reference gas for GC is methane (90 vol%), ethane (7 vol%) and propane (2.5 vol%).

(4). High Purity Grade

(5). Ultra High Purity Grade

The volume of the compressed gas inside the cylinders is limited and the associated inventory available is small, and also those compressed gas cylinders are located within control room building.  Considering the above, should loss of containment occur for the compressed gas cylinders, there is no off-site impact on the surrounding marine population.  Hence, it is not further assessed in the QRA Study.

Review of Potential MAEs

Leakage or rupture scenarios of process equipment, pipeline or pipework handling flammable natural gas can result in a flammable gas cloud, which may be ignited if it encounters an ignition source while its concentration lies within the flammable range.  In some cases, static discharge may also cause immediate ignition of flammable gas release.

The possible hazardous scenarios considered in the QRA Study upon the ignition of any released natural gas at the GRSs are:

·      Jet fire;

·      Flash fire;

·      Fireball; and

·      VCE.

Detailed characteristics of the above hazardous effect are described in Annex 5G.

Review of Relevant Industry Incidents

To investigate further the possible hazardous scenarios, a review of the applicable past industry incidents at similar facilities worldwide was conducted based on the following incident/ accident database:

·      IChemE accident database

·      eMARS;

·      ERNS; and

·      MHIDAS database.

Details of the past industry incident analysis are presented in Annex 5C.

Review of Potential Initiating Events leading to MAEs

The potential hazardous scenarios arising from the operation of the GRSs at the BPPS and LPS was identified as the loss of containment of natural gas.  The potential initiating events which could result in loss of containment of natural gas are listed below:

·      General equipment/piping failure (due to corrosion, construction defects etc.); and

·      External effects - earthquake, subsidence, tsunami, lightning, hill fire, storm surge and flooding, aircraft crash and helicopter crash.

Descriptions of the potential initiating events are presented in Annex 5D.

HAZID Workshop

A HAZID workshop was conducted to confirm and further identify the potential initiating events which may lead to MAEs at the GRSs, based on the HAZID team representatives’ experience, past industry accidents, lessons learnt and guideword checklists.  The HAZID workshop worksheet is summarised in Annex 5E.  The HAZID workshop output was served as a basis for the identification of potential initiating events and hazardous scenarios for the QRA Study.

External Hazards of Construction Activities

During the peak construction period of the GRSs in the beginning of 2020, the associated construction activities may cause potential external hazards on the existing GRSs facilities located in the nearby area.  The construction activities considered in the QRA Study are listed below:

·      Movement of large equipment/ construction vehicles in the vicinity of the existing GRS facilities area;

·      Dropped object from crane operation;

·      General construction hazards such as hot work, drilling, etc.; and

·      Tie-in works to existing facilities.

Detailed analysis of each identified construction activities are presented in Annex 5D.

Development of Hazardous Sections

The new GRSs and existing GRSs at the BPPS and LPS were divided into a number of hazardous sections for detailed analysis in the QRA Study based on location of emergency shutdown valves and process conditions (e.g. operating temperature and pressure).  The details of each hazardous section (including temperature, pressure, flow rate, inventory etc.) are presented in Annex 5D.

5.7.4                           Frequency Analysis

As per previous EIA Reports that have been approved by EPD ([48]) ([49]) ([50]), the release frequencies from Hawksley, as summarised in Table 5.31, were adopted in the QRA Study.

Table 5.31      Release Event Frequencies

Equipment

Release Scenario

Release Phase

Release Frequency

Unit

Reference

Pipe size 600 mm to 750 mm

i) 10 & 25 mm hole

Liquid/ Gas

1.00E-07

per metre-year

Hawksley ([51])

ii) 50 & 100 mm hole

Liquid/ Gas

7.00E-08

per metre-year

Hawksley

iii) Full bore rupture

Liquid/ Gas

3.00E-08

per metre-year

Hawksley

Pipe size 150 mm to 500 mm

i) 10 & 25 mm hole

Liquid/ Gas

3.00E-07

per metre-year

Hawksley

ii) 50 & 100 mm hole

Liquid/ Gas

1.00E-07

per metre-year

Hawksley

iii) Full bore rupture

Liquid/ Gas

5.00E-08

per metre-year

Hawksley

 

In addition, in accordance with the methodology used in previous EIA Reports that have been approved by EPD (2) a fault tree analysis was conducted to calculate the frequency of construction vehicles impacting the existing GRS facilities during the construction phase of the GRSs at the BPPS and LPS.  The frequency of construction vehicle impact on the existing BPPS GRS and LPS GRS was estimated as 1.53 × 10-6 per year and 9.20 × 10-7 per year respectively.  Detailed discussion on the above failure frequencies are presented in Annex 5F.

Release Hole Sizes

As per previous EIA Reports that have been approved by EPD ([52]) ([53]) ([54]), the hole sizes in Table 5.32 were considered in the QRA Study:

Table 5.32      Release Hole Sizes

Leak Description

Hole Size

Very Small Leak

10 mm

Small Leak

25 mm

Medium Leak

50 mm

Large Leak

100 mm

Line Rupture

Pipeline Diameter

 

Flammable Gas Detection and Emergency Shutdown Probability

As discussed in Section 5.5.3, the probability of executing the isolation successfully when required during emergency shutdown was adopted as 99%.  However, as a conservative approach, the probability of failure on demand for all detection and shutdown system was adopted as 100% in the QRA Study for GRSs, as per previous EIA Reports that have been approved by EPD (2).

Ignition Probability

Table 5.33 summarises the ignition probabilities adopted in the QRA Study as per previous EIA Reports that have been approved by EPD.  The total ignition probability is 0.32 for large leaks/ruptures, and 0.07 for other leaks.  These ignition probabilities are consistent with the model of Cox, Lees and Ang.  The ignition probabilities were distributed between immediate ignition and delayed ignition.  Delayed ignition was further divided between delayed ignition 1 and delayed ignition 2 to take into account that a dispersing gas cloud may be ignited at different points during dispersion.

Delayed ignition 1 results in a flash fire and takes into account the possibility that an ignition could occur within the GRS facilities area due to the presence of ignition sources on-site.  Delayed ignition 2 gives a flash fire after the gas cloud has expanded to its maximum (steady state) extent. If both delayed ignition 1 and 2 do not occur, the gas cloud disperses with no hazardous effect.

Table 5.33      Ignition Probably Adopted in the QRA Study for GRSs at the BPPS and LPS

Leak

Immediate Ignition

Delayed

Ignition 1

Delayed

Ignition 2

Delayed Ignition Probability

Total Ignition Probability

a) Large/ Rupture

0.10

0.200

0.020

0.22

0.32

b) Leaks other than Large/ Rupture

0.02

0.045

0.005

0.05

0.07

 

Vapour Cloud Explosion

It is noted that a VCE could explosion could potentially occur at the BPPS and LPS GRS areas where the flammable gas cloud could accumulate.  Nevertheless, based on the consequence modelling, the explosion effect is localized; hence flash fire with larger hazard footprint was conservatively modelled in the QRA Study.  Detailed comparison of the consequences is presented in Annex 5G.

Escalation Effects

An initially small release may escalate into a larger, more serious event if a jet fire impinges on neighbouring equipment/ piping for an extended time.  This is taken into account in the modelling for the isolation fail branch of the event tree, depicted in Figure 5F.8).  If neighbouring equipment and piping is within range of the flame zone of a jet fire, an escalation probability of 1/6 has been taken to conservatively estimate the directional probability and chance of impingement.  Escalation has been assumed to only cause a full bore rupture of the affected equipment and piping, leading to fireball event as the worst-case scenario.

Event Tree Analysis

An event tree analysis was performed to model the development of each hazardous scenario outcome (jet fire, flash fire, fireball, and VCE) from an initial release scenario.  The event tree analysis considered whether there is immediate ignition, delayed ignition or no ignition, with consideration of the associated ignition probability as discussed above.  The development of the event tree is presented in Annex 5F.

5.7.5                           Consequence Analysis

Source Term Modelling

The modelling assumptions, as illustrated in Section 5.5.4, were also adopted in the QRA Study on the GRSs at the BPPS and the LPS.

Physical Effects Modelling

PHAST was used to perform the physical effects modelling to assess the effects zones for the following hazardous scenarios:

·      Jet fire;

·      Flash fire;

·      Fireball; and

·      VCE.

Detailed description of the physical effects modelling is presented in Annex 5G.

Consequence End-Point Criteria

The same consequence end-point criteria, as illustrated in Section 5.5.4, were also adopted in the QRA Study for the GRSs at the BPPS and the LPS.

5.7.6                           Risk Summation

The risk summation for the GRS facilities was modelled using ERM’s proprietary risk integration package Riskplot™, as per previous EIA Reports that have been approved by the EPD ([55]) ([56]).

Cumulative Risk Assessment

Since the existing GRSs at the BPPS and LPS are located in the vicinity of the new GRSs, the existing GRSs could be impacted by the hazardous events arising from the new GRSs.  

The construction activities of the GRSs could induce additional risks to the existing neighbouring GRS facilities.  Therefore, in the proposed assessment year of 2020 for the peak construction phase, additional risk arising from the construction of the GRSs was considered in the QRA Study.

It is noted that an additional CCGT unit at the BBPS and LPS may also be under construction concurrently.  However considering the locations of the additional CCGT units and GRS facilities, the risk of construction hazards arising from the new CCGT units impacting on the new GRS facilities is considered insignificant and hence not further assessed in the QRA Study.

The existing oil tanks with relative large inventory are separated from GRSs area by more than 300 m while the non-fuel gas dangerous storage areas are within buildings and separated from GRSs area by more than 50 m.  The individual risk impacts from GRSs area facilities to the existing dangerous goods facilities are in the order of magnitude from 1E-07 to 1E-06 per year without consideration any obstacle between them.  The likelihood of escalation effects from GRSs area facilities on those existing oil tanks and non-fuel gas dangerous storage areas is not considered as significant and indeed already included in the generic failure database.  As such, they are not required in the cumulative risk assessment for the QRA Study.

When in full operation, the associated process risk of the existing and new GRS facilities was assessed in the proposed assessment years of 2020 and 2030.

The details of the cumulative risk assessment considered in each of the proposed assessment years are summarised in Table 5.34.

Table 5.34      Details of Cumulative Risk Assessment

Potential Risk

2019

2020

2030

a)     Existing GRS facilities at the BPPS and the LPS (Baseline Condition)

Yes

Yes

Yes

b)     Construction activities for proposed GRS facilities leading to potential impact on nearby existing GRS facilities at the BPPS and the LPS, respectively

Yes

 

 

c)     Proposed GRS facilities at the BPPS and the LPS for natural gas intake from the LNG Terminal

 

Yes

Yes

 

Individual Risk Results for GRS at the BPPS

Construction Year (Beginning of 2020)

As shown in Figure 5.19, the individual risk contour of 1 × 10-5 per year was not identified, therefore the individual risk criteria stipulated in Section 2 of Annex 4 of the EIAO-TM is met.

Operational Year (Year 2020) and Future Scenario Year (Year 2030)

As shown in Figure 5.20 and Figure 5.21, the individual risk contour of        1 × 10-5 per year is mostly within the proposed GRS site boundary, with slight overlap in the sea area.  Nevertheless, when considering the exposure factor for the surrounding off-site population, the individual risk criteria stipulated in Section 2 of Annex 4 of the EIAO-TM is met.

Societal Risk Results for GRS at the BPPS

As shown in Figure 5.22, the societal risks in terms of F-N curves for Construction Year in the beginning of 2020, Operational Year in 2020 and Future Scenario Year in 2030 lie within the Acceptable Region; as such the societal risk criteria stipulated in Section 2 of Annex 4 of the EIAO-TM are met.

The key risk contributors, in terms of potential loss of life, associated with the GRS at the BPPS are summarised in Annex 5H.


 

Individual Risk Results for GRS at the LPS

Construction Year (Beginning of 2020)

As shown in Figure 5.23, the individual risk of 1 × 10-5 per year was not identified, therefore the individual risk criteria stipulated in Section 2 of Annex 4 of the EIAO-TM is met.

Operational Year (Year 2020) and Future Scenario Year (Year 2030)

As shown in Figure 5.24 and Figure 5.25, the individual risk contour of 1 × 10-5 per year is confined with the proposed GRS site boundary, therefore the individual risk criteria stipulated in Section 2 of Annex 4 of the EIAO-TM is met.

Societal Risk Results for GRS at the LPS

As shown in Figure 5.26, the societal risks in terms of F-N curves for Construction Year in the beginning of 2020, Operational Year in 2020 and Future Scenario Year in 2030 lie within the Acceptable Region; as such the societal risk criteria stipulated in Section 2 of Annex 4 of the EIAO-TM are met.

The key risk contributors, in terms of potential loss of life, associated with the GRS at the LPS are summarised in Annex 5H.

5.7.7                           Conclusion of GRS QRA Study

It is concluded that the risks associated with the GRSs at the BPPS and LPS, in terms of individual and societal risks, are in compliance with risk criteria stipulated in Section 2 of Annex 4 of the EIAO-TM.

5.8                                   Overall Conclusion

A QRA Study was conducted to evaluate the risk level associated with the following activities and facilities of the Project with consideration of the identified LNG, natural gas and other dangerous goods:

·      Marine Transits of LNGC and FSRU Vessel to The LNG Terminal;

·      The LNG Terminal, including the FSRU Vessel, the Jetty and LNGC Unloading Operations;

·      Subsea BPPS and LPS Pipelines; and

·      GRSs at the BPPS and LPS.

The assessment methodology and assumptions were based EIA Reports that have been approved by EPD ([57]) ([58]).

For marine transits of LNGC and FSRU Vessel, subsea BPPS and LPS Pipelines, the LNG Terminal, and the GRSs at the BPPS and LPS, the individual risk is in compliance with the risk criteria in Section 2 of Annex 4 of the EIAO-TM.

In terms of societal risk, the F-N curves for all Project components have been developed and shown to be in compliance with risk criteria stipulated in Section 2 of Annex 4 of the EIAO-TM.

 


([1])        ERM, EIA for Liquefied Natural Gas (LNG) Receiving Terminal and Associated Facilities (Register No.: AEIAR-106/2007), December 2006.

([2])        ERM, EIA for Black Point Gas Supply Project (Register No.: AEIAR-150/2010), February 2010.

([3])        ERM, EIA for Liquefied Natural Gas (LNG) Receiving Terminal and Associated Facilities (Register No.: AEIAR-106/2007), December 2006.

([4])        ERM, EIA for Black Point Gas Supply Project (Register No.: AEIAR-150/2010), February 2010.

([5])        GeoInfo Map, http://www.map.gov.hk (accessed 7th August 2017)

([6])        BMT Asia Pacific Limited, Hong Kong Offshore LNG Terminal FSRU Terminal MTIA Report, R9331.05

([7])        BMT Asia Pacific Limited, Hong Kong Offshore LNG Terminal FSRU Terminal MTIA Report, R9331.05

([8])        HK Observatory, http://www.hko.gov.hk

([9])        Fire Protection Notice No. 4, Dangerous Goods General, Fire Services Department.

([10])      Major Accident Hazards Bureau (MAHB), https://emars.jrc.ec.europa.eu/?id=4.

([11])      ERNS database: http://www.rtk.net/erns/search.php.

([12])      UK AEA, Major Hazard Incident Database (MHIDAS) Silver Platter.

([13])      http://www.sigtto.org .

([14])      ERM, EIA for Liquefied Natural Gas (LNG) Receiving Terminal and Associated Facilities (Register No.: AEIAR-106/2007), December 2006.

([15])      BMT Asia Pacific Limited, Hong Kong Offshore LNG Terminal Marine Impact Assessment

([16])      ERM, EIA for Liquefied Natural Gas (LNG) Receiving Terminal and Associated Facilities (Register No.: AEIAR-106/2007), December 2006.

([17])      TNO, Methods for the Determination of Possible Damage to People and Objects Resulting from Releases of Hazardous Materials (The Green Book), Report CPR 16E, The Netherlands Organisation of Applied Scientific Research, Voorburg, 1992.

([18])      ERM, EIA for Liquefied Natural Gas (LNG) Receiving Terminal and Associated Facilities (Register No.: AEIAR-106/2007), December 2006.

([19])      Major Accident Hazards Bureau (MAHB), https://emars.jrc.ec.europa.eu/?id=4.

([20])      ERNS database: http://www.rtk.net/erns/search.php.

([21])      UK AEA, Major Hazard Incident Database (MHIDAS) Silver Platter.

([22])      http://www.sigtto.org .

([23])      OGP, Risk Assessment Data Directly, Report No. 434-.1, March 2010.

([24])      UK HSE, Failure Rate and Event Data for use within Risk Assessment, 28 June 2012.

([25])      Guidelines for Quantitative Risk Assessment, “Purple Book”, 2005.

([26])      OGP, Risk Assessment Data Directly, Report No. 434-.1, March 2010.

([27])      Guidelines for Quantitative Risk Assessment, “Purple Book”, 2005.

([28])      Cox, Lees and Ang, Classification of Hazardous Locations, IChemE.

([29])       ERM, EIA for Liquefied Natural Gas (LNG) Receiving Terminal and Associated Facilities (Register No.: AEIAR-106/2007), December 2006.

([30])      ERM, EIA for Black Point Gas Supply Project (Register No.: AEIAR-150/2010), February 2010.

([31])      Guidelines for Quantitative Risk Assessment, “Purple Book”, 2005.

([32])      Guidelines for Quantitative Risk Assessment, “Purple Book”, 2005.

([33])      ERM, EIA for Liquefied Natural Gas (LNG) Receiving Terminal and Associated Facilities (Register No.: AEIAR-106/2007), December 2006.

([34])      Worley Parsons, Pipeline Construction Report.

([35])      Worley Parsons, Pipeline Construction Report

([36])       ERM, EIA for Liquefied Natural Gas (LNG) Receiving Terminal and Associated Facilities (Register No.: AEIAR-106/2007), December 2006.

([37])      ERM, EIA for Black Point Gas Supply Project (Register No.: AEIAR-150/2010), February 2010.

([38])      Energy Institute, London and Oil & Gas UK, Pipeline and Riser Loss of Containment 2010 – 2012 (PARLOC 2012), 6th Edition of PARLOC Report Series, March 2015.

([39])      Mott MacDonald Ltd., The Update of Loss of Containment Data for Offshore Pipelines (PARLOC 2001), Revision F, June 2003.

([40])       ERM, EIA for Liquefied Natural Gas (LNG) Receiving Terminal and Associated Facilities (Register No.: AEIAR-106/2007), December 2006.

([41])       ERM, EIA for Liquefied Natural Gas (LNG) Receiving Terminal and Associated Facilities (Register No.: AEIAR-106/2007), December 2006.

([42])      ERM, EIA for Black Point Gas Supply Project (Register No.: AEIAR-150/2010), February 2010.

([43])       ERM, EIA for Liquefied Natural Gas (LNG) Receiving Terminal and Associated Facilities (Register No.: AEIAR-106/2007), December 2006.

([44])      ERM, EIA for Black Point Gas Supply Project (Register No.: AEIAR-150/2010), February 2010.

([45])      P J Rew, P Gallagher, D M Deaves, Dispersion of Subsea Release: Review of Prediction Methodologies, Health and Safety Executive, 1995.

([46])      ERM, EIA for Liquefied Natural Gas (LNG) Receiving Terminal and Associated Facilities (Register No.: AEIAR-106/2007), December 2006.

([47])      Fire Protection Notice No. 4, Dangerous Goods General, Fire Services Department.

([48])      ERM, EIA for Additional Gas-fired Generation Units Project (Register No.: AEIAR-197/2016), June 2016.

([49])      ERM, EIA for Black Point Gas Supply Project (Register No.: AEIAR-150/2010), February 2010.

([50])      ERM, EIA for Liquefied Natural Gas (LNG) Receiving Terminal and Associated Facilities (Register No.: AEIAR-106/2007), December 2006.

([51])      Hawksley, J.L., Some Social, Technical and Economic Aspects of the Risks of Large Plants, CHEMRAWN III, 1984.

([52])      ERM, EIA for Additional Gas-fired Generation Units Project (Register No.: AEIAR-197/2016), June 2016.

([53])      ERM, EIA for Black Point Gas Supply Project (Register No.: AEIAR-150/2010), February 2010.

([54])      ERM, EIA for Liquefied Natural Gas (LNG) Receiving Terminal and Associated Facilities (Register No.: AEIAR-106/2007), December 2006.

([55])      ERM, EIA for Black Point Gas Supply Project (Register No.: AEIAR-150/2010), February 2010.

([56])      ERM, EIA for Liquefied Natural Gas (LNG) Receiving Terminal and Associated Facilities (Register No.: AEIAR-106/2007), December 2006.

([57])      ERM, EIA for Liquefied Natural Gas (LNG) Receiving Terminal and Associated Facilities (Register No.: AEIAR-106/2007), December 2006.

([58])      ERM, EIA for Black Point Gas Supply Project (Register No.: AEIAR-150/2010), February 2010.