8 Hazard To Life.. 1

8.1 Introduction.. 1

8.2 Environmental Legislation, Standards And Criteria.. 3

8.3 Project Construction Arrangement and Schedule.. 5

8.4 QRA Methodology.. 5

8.5 Meteorological Data and Population Data.. 12

8.6 QRA For Biogas Facilities. 16

8.7 QRA For Tai Po Gas Production Plant. 31

8.8 QRA For Liquefied Petroleum Gas (LPG) Storage Facilities. 48

8.9 Residual Impacts. 60

8.10 Environmental Monitoring and Audit Requirements. 60

8.11 Conclusion.. 60

8.12 Recommendations. 61

8.13 Reference.. 62

List of TABLES

Table 8.1 EIA Study Brief (Appendix E) – Hazard to Life Requirements. 1

Table 8.2 Summary of Ignition Sources assumed in QRA.. 9

Table 8.3 Ignition Probabilities of ignition Source. 9

Table 8.4 Overpressure Effects. 11

Table 8.5 Population Data for Concurrent Projects. 12

Table 8.6 Day Time Meteorological Data. 12

Table 8.7 Night Time Meteorological Data. 13

Table 8.8 Population for Various Project Development Phases. 14

Table 8.9 Temporary Change in Population.. 15

Table 8.10 Operating Conditions of Proposed Biogas Facilities. 16

Table 8.11 Operating Conditions of Existing Biogas Facilities. 17

Table 8.12 Composition and Properties of Biogas from Anaerobic Digestion Process. 17

Table 8.13 Storage Quantities of Dangerous Goods. 18

Table 8.14 Storage Quantities of Other Chemicals. 18

Table 8.15 Hazardous Systems Identified at Biogas Facilities within TPSTW... 19

Table 8.16 Failure Frequency Data. 20

Table 8.17 Accident Involvements of Medium / Heavy Goods Vehicles in Hong Kong. 21

Table 8.18 Event Frequencies. 22

Table 8.19 Properties of Naphtha. 34

Table 8.20 Composition of Town Gas [18]. 34

Table 8.21 Physical Properties of Town Gas [18]. 34

Table 8.22 Hazardous Systems Identified at TPGPP.. 35

Table 8.23 Failure Frequency Data for Tank. 38

Table 8.24 Failure Frequency Data for Equipment. 38

Table 8.25 Failure frequency Estimated for Hazardous Sections. 39

Table 8.26 Information of LPG storage tank in Apex Print Limited and Zama Print Limited.. 48

Table 8.27 Physical Properties of LPG.. 50

Table 8.28 Hazardous Scenarios. 51

Table 8.29 Event Frequencies. 52

List of EXHIBITS

Exhibit 8.1 .......... Biogas Facility and other Fuel Gas DG facilities covered in QRA.. 7

Exhibit 8.2 .......... Hong Kong Societal Risk Criteria. 9

Exhibit 8.3 .......... QRA Methodology. 12

Exhibit 8.4 .......... Event Tree Extracted from MPACT (for Gaseous Release). 15

Exhibit 8.5 .......... Event Tree Extracted from MPACT (for Liquid Release). 16

Exhibit 8.6 .......... Event Tree Extracted from MPACT (for Catastrophic Rupture). 17

Exhibit 8.7 .......... Individual Risk Contours – Construction of New West Plant in Proposed Expansion Site. 36

Exhibit 8.8 .......... Individual Risk Contours – Construction in Existing West Plant. 37

Exhibit 8.9 .......... Individual Risk Contours – Operational Phase. 38

Exhibit 8.10 ....... F-N Curve – Biogas Facilities within the Project Site (Base Case). 2

Exhibit 8.11 ....... F-N Curve – Biogas Facilities within the Project Site (Sensitivity Case for Concurrent Project). 4

Exhibit 8.12 ....... TPGPP and its CZ.. 5

Exhibit 8.13 ....... Layout of TPGPP.. 6

Exhibit 8.14 ....... CRG Process Flow Diagram [18]. 7

Exhibit 8.15 ....... Individual Risk Contours – Tai Po Gas Production Plant. 13

Exhibit 8.16 ....... F-N Curve – Tai Po Gas Production Plant. 16

Exhibit 8.17 ....... F-N Curve – Tai Po Gas Production Plant (Sensitivity Case for Concurrent Project). 18

Exhibit 8.18 ....... Location of LPG Storage and Vaporizer at Apex Print Limited.. 19

Exhibit 8.19 ....... Plot Plan of LPG Storage and Vaporizer at Zama Print Limited.. 19

Exhibit 8.20 ....... Location of LPG Storage vessel at Zama Print Limited.. 20

Exhibit 8.21 ....... Individual Risk Contours – LPG Storage Facility at Apex Print Limited.. 24

Exhibit 8.22 ....... Individual Risk Contours – LPG Storage Facility at Zama Industries Limited.. 25

Exhibit 8.23 ....... F-N Curve - LPG Storage Facility at Apex Print Limited.. 26

Exhibit 8.24 ....... F-N Curve - LPG Storage Facility at Zama Industries Limited.. 27

Exhibit 8.25 F-N Curve - LPG Storage Facility at Apex Print Limited (Sensitivity Case for Concurrent Project). 28

Exhibit 8.26 F-N Curve - LPG Storage Facility at Zama Industries Limited (Sensitivity Case for Concurrent Project). 29

APPENDICES

Appendix 8.1 Population Data

Appendix 8.2 Calculation of Transient Population

Appendix 8.3 Preliminary Layout of Proposed Biogas Facilities and Tentative Locations of Chemical Storage

Appendix 8.4 Existing Biogas Facilities and Tentative Construction Plan

Appendix 8.5 External Hazards Review

Appendix 8.6 PHAST RISK Parameters

Appendix 8.7 Review of Historical Accident

Appendix 8.8 Consequence Data

Appendix 8.9 Failure Frequency for Scenarios Associated with Naphtha Storage Tanks

Appendix 8.10 Projectile Risk Assessment

Appendix 8.11 Detailed FTA

Appendix 8.12 Top Risk Contributors

8 Hazard to Life

8.1 Introduction

8.1.1 This Project would involve co-digestion of indigenous sewage sludge from Tai Po Sewage Treatment Works (TPSTW), imported sewage sludge from other Sewage Treatment Works (STWs) in East New Territories and imported organic / food waste from the proposed Organic Waste Pre-treatment Centre (New Territories East) (OWPC) co-located adjacent to TPSTW. The sewage sludge and organic / pre-treated food waste will be anaerobically co-digested to generate biogas as an energy source for power supply.

8.1.2 Under Section 5(7)(a) of the Environmental Impact Assessment (EIA) Ordinance (Cap. 499) (EIAO), the Director of Environmental Protection (the Director) has issued a EIA Study Brief No.ESB-321/2019 for this Project (EIA Study Brief), which specifies the technical requirements of Hazard to Life assessments to be conducted for the Project. As stated in Section 2.1 of the EIA Study Brief, the objective of the assessment is to identify any potential hazard to life due to generation, storage, utilization, processing and transmission (if applicable) of biogas within the Project site and due to neighbouring fuel gas dangerous goods (DGs) facilities during the construction and operation of the Project and to propose measures to mitigate these impacts if required. For reference, the detailed requirements as given in Appendix E of the EIA brief are repeated in Table 8.1.

Table 8.1 EIA Study Brief (Appendix E) – Hazard to Life Requirements

Biogas

1. The Applicant shall investigate methods to avoid and/or minimize risk from biogas during the construction and operation stages of the Project. The Applicant shall carry out hazard assessment to evaluate potential hazard to life due to biogas.

2. The hazard assessment shall include the following:

i. Identify hazardous scenarios associated with the generation, storage, utilization, processing and transmission (if applicable) of biogas due to the Project and hazardous scenarios due to neighbouring dangerous goods (DGs) facilities which may cause impact to the biogas facilities of the Project (including but not limited to DGs stores at Linde HKO Limited) and then determine a set of relevant scenarios to be included in a Quantitative Risk Assessment (QRA);

ii. Execute a QRA of the set of hazardous scenarios determined in sub-section (i) above, expressing population risks in both individual and societal terms;

iii. Compare individual and societal risks with the criteria for evaluating hazard to life stipulated in Annex 4 of the TM; and

iv. Identify and assess practicable and cost-effective risk mitigation measures.

3. The methodology to be used in the hazard assessment shall be consistent with previous studies having similar issues.

Fuel Gas Dangerous Goods including Liquefied Petroleum Gas and Town Gas

4. The Applicant shall carry out hazard assessment to evaluate the risk due to fuel gas dangerous goods (DGs) facilities in the vicinity, including Zama Industries Limited and Apex Print Limited during construction and operation stages of the Project. The Applicant shall provide the estimated numbers of construction workers and staff of the Project during construction and operation stages of the Project respectively, and seek the Director’s agreement on whether the hazard assessment has to cover risk due to Tai Po Gas Production Plant. The hazard assessment shall include the following:

i. Identify hazardous scenarios associated with the fuel gas DG facilities with a view to determining a set of relevant scenarios to be included in a QRA;

ii. Execute a QRA of the set of hazardous scenarios determined in sub-section (i) above, expressing population risks in both individual and societal terms;

iii. Compare individual and societal risks with the criteria for evaluating hazard to life stipulated in Annex 4 of the TM; and

iv. Identify and assess practicable and cost-effective risk mitigation measures.

5. The methodology to be used in the hazard assessment shall be consistent with previous studies having similar issues.

8.1.3 On this basis, a Quantitative Risk Assessment (QRA) has been conducted to evaluate hazardous scenarios associated with the proposed Project and other fuel gas DG facilities in the vicinity of the proposed Project. The facilities covered in the QRA include the followings:

1. Biogas Facilities within the proposed Project Site

2. Tai Po Gas Production Plant (TPGPP)

3. Liquefied Petroleum Gas (LPG) storage facility at Apex Print Limited

4. Liquefied Petroleum Gas (LPG) storage facility at Zama Industries Limited, and

5. Dangerous Goods (DGs) storage at Linde HKO Limited

8.1.4 The Project site location in relation to the above facilities is shown in Exhibit 8.1. Notably, TPGPP is a Potentially Hazardous Installation (PHI).

Exhibit 8.1 Biogas Facility and other Fuel Gas DG facilities covered in QRA

8.2 Environmental Legislation, Standards and Criteria

8.2.1 The assessed risk levels of hazardous sources shall be compared with the risk guidelines stipulated in the EIAO-TM Annex 4 [2] to determine the acceptability. As set out in the EIAO-TM Annex 4, the risk guidelines comprise the following two components:

1. Individual Risk: The maximum level of off-site individual risk should not exceed 1 × 10^-5 / year, i.e. 1 in 100,000 per year; and

2. Societal Risk: Societal risk is expressed in the form of an F-N curve (Exhibit 8.2), which represents the cumulative frequency (F) of all event outcomes leading to N or more fatalities. The F-N curve consists of three different regions defined as follows:

n Unacceptable region: where risk is so high that they should usually be reduced regardless of the cost or else the hazardous activity should not proceed;

n ALARP region: where risk is tolerable, provided that it has been reduced to a level As Low As Reasonably Practicable (ALARP); and

n Acceptable region: where risk is broadly acceptable and does not require further risk reduction.

Exhibit 8.2 Hong Kong Societal Risk Criteria

8.3 Project Construction Arrangement and Schedule

8.3.1 The existing TPSTW comprises two independent plants, namely West Plant and East Plant. In order to maintain normal sewage treatment services of the existing TPSTW during the construction phase, a New West Plant would be built in the proposed expansion site. The New West Plant would be a relatively compact Sewage Treatment Works (STW) and be able to provide adequate sewage treatment capacity to meet the projected sewage flow buildup before the normal treatment services of the existing West Plant is decommissioned. After the New West Plant is fully commissioned, the existing West Plant would be partially demolished to make room for the construction of new facilities such as the sludge treatment and co-digestion works.

8.3.2 The construction works of this Project are tentatively scheduled to commence in 2025 for completion in 2036. Majority of the demolition works in the existing West Plant of TPSTW would be carried out from 2027 to 2033 after the New West Plant in the proposed expansion site is in operation.

8.4 QRA Methodology

8.4.1 The key elements of a QRA study are outlined in Exhibit 8.3.

Exhibit 8.3 QRA Methodology

8.4.2 While the following paragraphs of this section describe the general QRA methodology employed, the facility-specific approaches and assumptions are detailed in the respective facility QRA sections of this hazard to life assessment.

8.4.3 In general, the following previously approved QRA studies have been referenced when developing the QRA for this Project:

n SEKD EIA Report [11]

n Yuen Long Biogas [33]

n Chai Wan Government Complex and Vehicle Depot EIA Report [9]

n HATS Stage 2A EIA Report [10]

n LPG Shell Depot [27]

n LPG Storage at Choi Po Court [35]

Hazard Identification

8.4.4 The hazard identification involved a review of the hazardous material properties and a review of the past accidents, with the objective of identifying potential hazards and scenarios to be modelled in the subsequent frequency and consequence analysis. The summary of review of historical accidents is presented in Appendix 8.7. In addition, external hazards review was also carried out to assess potential hazards raised due to external sources such as aircraft crash, earthquake, etc. and the summary is presented in Appendix 8.5.

8.4.5 The following hazardous events are generally considered in the event of a loss of containment scenario:

n Fireball: In the case of immediate ignition of an instantaneous gas release, this will result in a fireball. Typically, a fireball occurs due to catastrophic ruptures of process equipment/vessels. A fireball is of hemispherical shape emitting thermal radiation. Thermal radiation from a fireball is independent of weather and wind direction.

n Pool Fire: A pool fire occurs upon the ignition of flammable liquid which has been spilled on the ground.

n Jet Fire: A jet fire occurs as a consequence of ignition of pressurized gas releases. A jet fire emits thermal radiation and hence can cause harm to population in the vicinity.

n Flash Fire: A flash fire occurs when a flammable gas release occurs, however ignition is delayed. If the ignition source is within the Upper Flammability Limit and Lower Flammability Limit, it shall result in ignition of the flammable release. If the release does not come in contact with an ignition source, prior to being diluted below its Lower Flammability Limit, no harm is anticipated as the gas is too lean for ignition.

n VCE: When a flammable vapour cloud forms, disperses and accumulates in areas with high congestion and / or confinement, and is then ignited, a Vapour Cloud Explosion (VCE) may occur, leading to damaging overpressures to buildings and resulting in building damage / collapse.

n Toxic Release: An unignited release may pose a hazard if the release stream contains significant amount of toxic material.

n Boiling-Liquid Expanding-Vapor Explosion (BLEVE): An explosion caused by a liquid which is boiling and continuing to produce a flammable vapor.

Frequency Analysis

8.4.6 Frequency analysis is used to derive the final event outcome frequencies. By using historical failure frequency data, the failure event outcome frequency is determined. The likelihood of each identified scenario was quantified taking into account the site-specific features and project activities. In general, initial hazardous event frequencies, i.e. loss of containment, were derived from historical databases and then modified, as required, to factor in the site-specific conditions. Where required, Fault Tree Analysis (FTA) was also carried out to quantify the various possibilities of event combinations as well as the typical safety systems in place.

8.4.7 Following a leak or rupture, various hazardous events may arise depending on the release conditions (e.g. instantaneous or continuous release, rainout and vaporization of the released material) as well as the type of ignition (e.g. immediate or delayed ignition). The frequencies of these undesired outcome events such as flash fire, pool fire, jet fire, explosion, etc. were derived using Event Tree Analysis (ETA).

8.4.8 ETA is an analysis technique which identifies different possible outcomes following an initiating event and estimates the probabilities for each of these outcomes. Event Tree starts with initiating event and proceeds by examining each contributing factor in chronological order to identify all possible outcomes. The frequencies of event outcomes are calculated by multiplying the initiating event frequency and probabilities of all contributing factors leading to the specific hazardous event. In this study, PHASTRISK Event Tree was used to generate the outcome events. The detailed parameters used in PHASTRISK are presented in Appendix 8.6. The figures below present the event trees for various scenarios for MPACT used in the QRA, including gaseous release, liquid release, and vessel catastrophic rupture.

Exhibit 8.4 Event Tree Extracted from MPACT (for Gaseous Release)

Exhibit 8.5 Event Tree Extracted from MPACT (for Liquid Release)

Exhibit 8.6 Event Tree Extracted from MPACT (for Catastrophic Rupture)

8.4.9 The immediate ignition probability was determined based on TNO material reactivity method [6] in PHASTRISK, which takes into account the size of release and the flammability of released substance. The delayed ignition probability was estimated based on the specific ignition sources identified in the area. Based on the guidelines by Purple Book [6], the probability of ignition by an ignition source is dependent on its ignition efficiency and the presence factor within the source. Noted that the ignition efficiency for Industrial Building category is based on "Heavy Industry" and "Light Industry". For Light Industry, the efficiency is a function of population size and it is conservatively capped at 0.4 same as Road Vehicles. The main ignition sources in the area and the ignition efficiency assumed have been summarized in Table 8.2 and Table 8.3.

Table 8.2 Summary of Ignition Sources assumed in QRA

Ignition Sources	Description	Ignition Efficiency*
Flare and furnace	Open flame or very hot surfaces	1
Combined Heat and Power Generation System (CHP) in TPSTW	Hot surface and combustion	1
TPGPP Process Trains	Heating and rotating equipment	0.45
Factories	Potential hot works and heavy machine operations	0.7
Industrial buildings	Smoking, cooking and use of electrical appliances	0.4
Road Vehicles	Vehicle engines and hot exhaust gases	0.4
TPSTW Process Equipment	High power consumption or high speed pumps and motors	0.45

*ignition efficiency in a period of 60 seconds (derived from Purple Book based on the ignition source type).

Table 8.3 Ignition Probabilities of ignition Source

Ignition Sources	Type	Ignition Probability in 60s	Presence Factor
Tai Po Towngas Production Train (PT1 – 8)	Area	0.45	1
Tai Po Towngas Furnace Phase 1	Line	1	1
Tai Po Towngas Furnace Phase 2	Line	1	1
South China Morning Post	Area	0.4	1
Tao Heung Group	Area	0.4	1
Lee Kee Group	Area	0.4	1
Hong Kong Note Printing Ltd.	Area	0.4	1
Hong Kong Yamazaki Baking Co.Ltd.	Area	0.4	1
Nissin Foods *	Area	0.4	1
Hitachi Chemical	Area	0.4	1
Linde HKO Ltd.	Area	0.7	1
Golik Metal	Area	0.7	1
Yuen Shin Road	Transportation line	0.4	Refer to hourly vehicle flow in Appendix 8.2
Ting Kok Road
Dai Fuk Road
Dai Wah Road
Dai Fat Road
Dai Shing Road
Dai Cheong Road
Dai Wang Road
Dai Fu Road
Dai Kwai Road
Dai Hei Road
Dai King Road
Dai Shun Road

*Refers to Population D in Appendix 8.1.

Consequence Analysis

8.4.10 Consequence assessment was performed to predict the size, shape and orientation of the hazard zones resulting from releases of flammable and toxic substances. DNV PHAST V6.7 was used to perform the consequence modelling, which comprises the following elements:

n Source term/ discharge modelling

n Dispersion modelling

n Fire and explosion modelling

n Effects modelling

8.4.11 Source term or discharge modelling involved the determination of maximum discharge rate, release duration, and other physical properties (such as discharge velocity and temperature resulting from gas expansion) of the released material following a leak or rupture. These estimated parameters were then set as the initial conditions for the subsequent dispersion or fire effects modelling.

8.4.12 In the event of a release or rupture of piping or equipment, all releases were modelled as continuous release at the maximum flow rate. For catastrophic rupture of equipment, the entire volume of the process equipment was taken for modelling.

8.4.13 Dispersion modelling involved mathematical simulation of how the released materials disperse in the ambient atmosphere. Downwind and crosswind concentrations were determined to find the gas cloud hazard footprint. Vapor dispersion modelling was conducted using PHAST’s Unified Dispersion Model (UDM). The model considers the following aspects of vapor cloud behavior in dispersion modelling:

n Continuous, instantaneous or time-varying releases;

n Jet, heavy-gas and passive dispersion;

n Elevated, touchdown and ground level dispersion;

n Droplet dispersion, rainout and droplet vaporization; and,

n Dispersion over land or water surfaces.

8.4.14 Physical effect modelling determined the magnitude of damage caused by exposure to fire, heat radiation, toxic, or explosion overpressure.

8.4.15 Probit functions were used to estimate the probability of fatality due to a physical effect, e.g. thermal radiation, etc. For flammable hazards, such as jet fire and fireball, the probability of fatality due to the exposure to high thermal radiation can be calculated with the probit equation from Purple Book [6] in the following form:

Where,

is the probit;

is the heat radiation (Wm^-2); and

is the exposure time (s).

8.4.16 For toxic gas dispersion, probit functions were also used to relate the probability of human fatality with the gas concentration of the toxic substance. As an example, the following probit equation from Phast can be used to estimate the likelihood of fatality due to exposure to toxic H₂S gas:

Where,

is the probit;

is the H₂S concentration (ppm); and

is the exposure time (minutes).

8.4.17 For overpressure effects, people indoor have a higher harm probability compared to people outdoor due to the risk of building collapse and flying debris. Referring to Purple Book [6], the overpressure levels used in the assessment are presented in Table 8.4.

Table 8.4 Overpressure Effects

Explosion Overpressure (bar)	Fraction of Fatality
	Indoor	Outdoor
> 0.3	1.000	1.000
> 0.1 to 0.3	0.025	0

8.4.18 For flash fire, all persons outdoor within the flash fire envelope (LFL contour) were assumed to be fatally injured i.e. fatality rate of 100%.

8.4.19 Buildings are assumed to offer protection to occupants against fire, and thus indoor protection factor was assumed to be 90% following previous QRA study [9]. A 100% exposure was assumed for open space.

Risk Summation

8.4.20 Risk summation then combined the estimates of likelihood and consequence for the identified hazardous events to produce the risk results, which are expressed in terms of individual risk and societal risk as per EIAO-TM [2]. Risk mitigation measures are recommended, where required to reduce the risk to As low As Reasonable Practicable (ALARP). DNV PHASTRISK v6.7 was used for modelling and risk summation.

8.4.21 To study the effects on different facilities onsite and change in construction workforce in the various phases of the Project, three representative study cases were considered for societal risk assessment:

n Construction of New West Plant (2025)

n Construction in Existing West Plant (2030), and

n Operational Phase (2036).

Sensitivity Case for Concurrent Project Construction

8.4.22 In addition to the proposed Project, it is noted that there are two related projects which are planned to have construction works within the proposed Project location, or in adjacent plot. Therefore, a sensitivity analysis of societal risk considering these concurrent projects is also carried out. The identified concurrent projects are described below and their locations are shown in Figure 2.2. Note that the population data for concurrent project was estimated by the project contractor based on past actual data for similar projects.

Proposed Organic Waste Pre-treatment Centre

8.4.23 Development of Organic Waste Pre-treatment Centre (New Territories East) is an interfacing project proposed by EPD under “Agreement No. CE 5/2021 (EP)”. The proposed Organic Waste Pre-treatment Centre (OWPC) is planned to receive and pre-treat the source-separated food waste for transferring to the Tai Po Sewage Treatment Works (TPSTW) and / or off-site anaerobic digesters in other Sewage Treatment Works (STW) for co-digestion with sewage sludge. It is tentatively scheduled to commence construction in 2025 for completion by 2029.

Proposed THEES Upgrading

8.4.24 This is a separate interfacing project proposed by EPD under “Agreement No. CE 13/2015 (DS)”. It involves upgrading of the effluent conveyance capacity of Tolo Harbour Effluent Export Scheme (THEES) in order to cater for the projected effluent flow of this Project. The proposed THEES upgrading works involve expansion of the Tai Po Effluent Pumping Station (TPEPS) within the Project site, as well as construction of a new effluent rising mains in Tai Po Industrial Estate (TPIE) and a new submarine pipeline (across inner Tolo Harbour) to handle the Project flow. The proposed TPEPS expansion located within the Project site will be constructed and operated under this Project and incorporated into the construction programme and design of this Project. Construction of the new effluent rising mains and submarine pipeline will be undertaken separately and is tentatively scheduled to commence in 2025 for completion 2031. All the proposed THEES upgrading works are committed to match the implementation programme of this Project.

Table 8.5 Population Data for Concurrent Projects

Concurrent Project	Max. No. of Worker in TPIE During Construction Phase	Max. No. of Worker in TPIE During Operational Phase
OWPC	300 *	35 #
THEES Upgrading Works	100 *	0 #

* Size of construction workforce was estimated based on experiences in other construction contracts of similar scale.

# Data was confirmed by the respective project proponent to be acceptable for use in this assessment.

8.5 Meteorological Data and Population Data

Meteorological Data

8.5.1 The meteorological conditions affect the consequence of gas release in particular the wind direction, speed and stability, which influence the direction and degree of turbulence of gas dispersion. The latest meteorological data (as of January2022) was collected from Tai Po Kau Weather Station [3] considering the past 6-year data (2015 – 2020). Twelve weather directions were considered, and two different sets of Meteorological data were used for representing the Day time and Night time weather condition. Ambient temperature and relative humidity were taken as 25 ^oC and 80%, respectively [53].

8.5.2 Table 8.6 and Table 8.7 present the day time and night time meteorological data, respectively.

Table 8.6 Day Time Meteorological Data

Direction	Weather Class						Total
	3B	1D	4D	6D	1F	3E
0 - 30	0.14	0.71	0.11	0.07	1.41	0.01	2.45
30 - 60	0.29	0.96	0.26	0.10	0.78	0.15	2.53
60 - 90	4.72	2.53	2.37	1.08	1.30	0.51	12.5
90 - 120	10.4	4.20	4.94	6.94	3.04	1.04	30.5
120 - 150	2.33	2.88	3.07	1.05	3.97	0.94	14.2
150 - 180	0.88	1.62	0.52	0.00	3.59	0.11	6.71
180 - 210	1.33	1.14	0.22	0.03	1.97	0.03	4.71
210 - 240	1.23	0.74	0.19	0.00	0.89	0.03	3.08
240 - 270	1.07	1.40	0.97	0.08	2.42	0.77	6.71
270 - 300	1.94	1.25	3.05	0.47	2.77	0.70	10.2
300 - 330	1.00	0.71	0.75	0.21	1.38	0.22	4.27
330 - 360	0.44	0.53	0.31	0.03	0.66	0.08	2.05
All	25.7	18.7	16.8	10.1	24.2	4.59	100

Table 8.7 Night Time Meteorological Data

Direction	Weather Class					Total
	1D	4D	6D	1F	3E
0 - 30	0.06	0.07	0.03	4.96	0.02	5.14
30 - 60	0.03	0.17	0.02	1.30	0.26	1.78
60 - 90	0.06	1.84	0.48	2.28	1.08	5.75
90 - 120	0.09	3.76	3.16	5.07	1.75	13.8
120 - 150	0.21	2.38	0.53	12.78	1.92	17.8
150 - 180	0.17	0.09	0.00	14.79	0.19	15.2
180 - 210	0.09	0.04	0.00	9.68	0.07	9.87
210 - 240	0.08	0.09	0.00	4.86	0.09	5.12
240 - 270	0.21	0.88	0.01	7.85	1.35	10.3
270 - 300	0.09	2.18	0.12	6.97	1.17	10.5
300 - 330	0.02	0.46	0.08	2.35	0.19	3.11
330 - 360	0.03	0.16	0.06	1.19	0.07	1.50
All	1.13	12.13	4.49	74.09	8.16	100

Population Data

8.5.3 As noted previously, three assessment years, namely 2025, 2030, and 2036, were considered for the societal risk assessment. The latest population data (i.e. 2021) has been collected and derived from various sources, and then projection was made to give the estimate of the population in future years. The population data used in the QRA is summarized in Appendix 8.1.

Proposed Upgrading of TPSTW

8.5.4 Currently, the TPSTW has 124 personnel on site. The adjoining plot for the proposed expansion is currently used by external parties, and the number of occupiers is approximately 168 based on questionnaire survey conducted in year 2020. The maximum construction work force has been estimated by the project engineers based on past projects of similar scale. Estimates were made based on monthly anticipated work force on site, the maximum number of construction workers (in the outdoor area) has been estimated to be about 300 for the New West Plant between 2025 and 2029, and about 400 for construction in the Existing West Plant Area between 2029 and 2036. During the operational phase of the Project, there will be about 240 workers onsite, which is still less than the existing level with consideration of removing the existing tenants in the proposed expansion site. Existing and future population within the Project site is presented in the table below.

Table 8.8 Population for Various Project Development Phases

Phase	DSD Personnel	Existing Occupiers	Project Construction Workers (in Outdoor Area)
Current Situation / Baseline (Year 2021)	124	168	0
Construction of New West Plant (Year 2025)	124	0	300
Construction in Existing West Plant (Year 2030)	124	0	400
Operational Phase (Year 2036)	240	0	0

Note: Population of this Project is excluded in the QRA for biogas facilities but has been considered in the remaining QRAs of this study.

Building Population

8.5.5 Population surveys were conducted in December 2020 by interviewing owners or security guards in those sites and Tai Po Industrial Estate is occupied by industrial buildings and factories. In support of the QRA, a number of industrial sites provided the actual manning data, which was used directly in the QRA. From the received actual manning data, an average industrial population density was also derived, based on which the number of people for other industrial sites was estimated after factoring in the size of plot and the scale of facility/ industrial building. The estimated average density is 0.0045 people per square metre floor area

8.5.6 For other population, estimates were based on information collected from Census 2021, the Planning Department (PlanD), and site survey (dated August 2020). The following information and assumptions were adopted to estimate the population:

n Average household size of 3.0 for residential population with the study area according to the average of Tertiary Planning Unit (TPU) 7.2.6, 7.2.2 and 7.2.7 of 2021 Population Census [4];

n Based on Projections of Population Distribution 2021-2029 [5], the average annual growth for residential population in TPU 7.2.6 is found to be negative growth. To be conservative, residential population was assumed to be constant. It may be noted that the 1km CZ of TPGPP covers primarily TPU 7.2.7 and, to a lesser extent, also covers some parts of TPU 7.2.2 and 7.2.6. TPU 7.2.7 is fully developed industrial area and any increase in population is strictly controlled by CCPHI CZ; therefore, significant increase in population is not expected. Focusing on the surrounding area of the proposed Project, TPU 7.2.6 is mainly high rise public housing area (residential area) while TPU 7.2.2 is mainly vacant plots with low rise buildings up to a few stories. It may be further noted that the published data is provided only for combined TPU 7.2.2 and 7.2.7, of which TPU 7.2.7 represents the Tai Po industrial Estate which is the main area of concern, and TPU 7.2.2 represents a greater area to the north of Ting Kok Road covering several residential areas and vacant plots for future development. A;

n Number of students and teachers in kindergartens and schools obtained from the School Profiles; and

n Observations from site survey dated August 2020 with conservative judgement.

8.5.7 In general, it is observed that most buildings in Tai Po Industrial Estate are low to medium rises. The cloud dispersion height results from consequence analysis were used to determine exposed population inside building for each study. Further detail on this protection factor will be discussed in each assessment as applicable.

Road Population

8.5.8 Annual Average Daily Traffic (AADT) for road traffic data was acquired from the latest Annual Traffic Census (ATC 2020) [8]. For some roads with no information on the ATC, the road traffic data was estimated based on-site survey.

8.5.9 A vehicle speed of 50 km/hr has been considered in the assessment. Although for main roads such as Ting Kok and Yuen Shin Road, vehicles can reach up to 70 km/hr, a lower vehicle speed of 50 km/hr was conservatively assumed for estimating the population to account for possible congestions and traffic jams.

8.5.10 Road population can be estimated using the following equation:

Road population = No of vehicles per hr * No. of persons per vehicle * Length of Road within Study Zone / Speed

8.5.11 The road population data from the ATC has been further projected to the studied case year using the calculated average road population growth rate which was found to be 2.6% per year. The detailed calculation and tabulation of projected road population data, including the population between day and night is provided in Appendix 8.2.

Time Period and Occupancy

8.5.12 In order to reflect the transient change of population, 4 time periods, namely weekday day, weekday night, weekend day, and weekend night were considered. Accordingly, the percentage of occupancy assumed for each population category is presented in Table 8.9. It may be noted that the occupancy and outdoor distribution assumed are consistent with previous QRAs [8][9][10][11][12].

Table 8.9 Temporary Change in Population

Category	Time period
	Weekday Day (Mon-Fri 0700-1900 hrs)	Weekday Night (Mon-Fri 1900-0700 hrs)	Weekend Day (Sat-Sun 0700-1900 hrs)	Weekend Night (Sat-Sun 1900-0700 hrs)
Commercial	100%	10%	40%	5%
Industrial	100%	10%	40%	5%
Residential	25%	100%	70%	100%
Recreational	50%	5%	100%	5%
Car Park	100%	10%	50%	10%
School	100%	1%	100%	1%
Construction	100%	1%	100%	1%
Road*	69%	31%	69%	31%

Note:
*Day night distribution calculation for road population is provided in Appendix 8.2.

8.5.13 An indoor ratio of 95% was applied to the population in residential, commercial buildings and in schools in line with the typical assumption in Hong Kong [9]. Passengers in vehicles were considered to be 100% outdoors, although vehicles may provide certain protection. Population in car park, the open leisure spaces, and the construction sites were considered to be 100% outdoors.

8.6 QRA for Biogas Facilities

Proposed Biogas Facilities

8.6.1 Biogas is utilized for providing the heat needed to maintain digester operating temperature and for providing heat to the paddle dryers. In the event of an emergency or equipment outage, digester gas may be flared. The purpose of gas storage is to provide greater flexibility to manage the digester gas pressure. The type of storage assumed operates within the operating pressure of the digesters and does not rely on any kind of pressure boosting or compression. Gas storage is configured as a branch off of the main gas pipeline and can be isolated if maintenance is required.

8.6.2 Gas treatment is required for removing hydrogen sulfide and siloxane from the biogas prior to combustion. Gas pressure is boosted to send first through dehumidification and then siloxane removal. The pressure boosting is required to push the gas through dehumidification and siloxane removal as well as to meet the downstream system operating requirements.

8.6.3 The facility is designed to handle the total biogas production rate of 82,000 m³/ day. Biogas will be processed through the following equipment sequentially:

· Digester

· Hydrogen Sulfide Treatment Media

· Pressure Boosting (blowers)

· Dehumidification

· Siloxane Removal

8.6.4 The table below summarizes the key operating parameters of the biogas storage and treatment facilities, based on process design information available:

Table 8.10 Operating Conditions of Proposed Biogas Facilities

Equipment	No. of item (No. of working + No. of standby)	Volume (m^3)	Pressure	Temperature
Digester	3 batteries (4 in each battery); overall 9+3	8958	32 mbarg	35 degC
Gas holders	4 (3+1)	3207	32 mbarg	35 degC
Hydrogen Sulfide Treatment Media	5 (4+1)	48.3	32 mbarg	35 degC
Pressure Boosting (blowers)	3	-	32 mbarg	35 degC
Dehumidification	2 trains (1+1)	75	32 mbarg	35 degC
Siloxane Removal	3 trains (2+1)	75	32 mbarg	35 degC

8.6.5 The preliminary layout of the biogas related facilities of the Project is shown in Appendix 8.3.

Existing Biogas Facilities of TPSTW

8.6.6 There are two existing Biogas facilities on the east and west side of the existing TPSTW. The table below summarizes the key operating parameters of the existing biogas storage and treatment facilities.

Table 8.11 Operating Conditions of Existing Biogas Facilities

Equipment	No. of item	Volume	Pressure	Temperature
West Plant Digester	2	1,829 m^3	32 mbarg	35 degC
West Plant Gas holders	1	850m^3	32 mbarg	35 degC
East Plant Digester	3	2,771 m^3	32 mbarg	35 degC
East Plant Gas holders	2	850m^3	32 mbarg	35 degC

8.6.7 Note that under the Project, the existing biogas system in the West Plant will be decommissioned in 2030 and the space will be used to construct new facilities.

8.6.8 The existing biogas related facilities of TPSTW are shown in Appendix 8.4. The tentative construction plans, which illustrate the anticipated schedule for demolishing existing facilities and commissioning new facilities at different construction stages are also included in Appendix 8.4.

Hazard Identification

Biogas

8.6.9 Biogas is a colourless flammable a combustible mixture of gases at atmospheric conditions that comprises mainly methane (CH4) and CO₂. Generally, biogas from anaerobic digestion process has a methane content of 55% to 70% by volume. The exact composition of biogas depends on the substance that is being decomposed. If the material consists of mainly carbohydrates, such as glucose and other simple sugars and high-molecular compounds (polymers) such as cellulose and hemicellulose, the methane production is low. However, if the fat content is high, the methane production is likely to be high. In general, the physical properties of biogas are also similar to those of natural gas, except up to 2,000 ppm of H₂S is anticipated and thus the biogas can also exhibit some degree of toxicity.

8.6.10 A loss of containment can lead to jet fire since the system is operated slightly above atmospheric pressure. The released gas, if not ignited immediately, could form a flammable gas plume. The properties of biogas from Anaerobic Digestion (AD) process are summarized in Table 8.12, based on a previous QRA study for Biogas facilities.

Table 8.12 Composition and Properties of Biogas from Anaerobic Digestion Process

Property	Biogas from Anaerobic Digestion
Methane Content	55% - 70%
Carbon Dioxide Content	30% - 45%
Density	1.2 Kg/Nm^3
Lower Caloric Value	23 MJ/Nm^3
Flammability^#	Extremely Flammable
Auto-Ignition Temperature^#	580°C
Flash Points^#	-188°C
Melting Point^#	-182.5°C
Boiling Point^#	-161.4°C
Flammable Limits^#	5% (Lower) – 15% (Upper)
Vapour Density^#	0.59 – 0.72 (air = 1)

Remark: # Physical properties of biogas that are similar to natural gas

8.6.11 Given that the flammability increases with increase of methane content, and the exact composition of biogas varies with the substance that is being decomposed, biogas was conservatively modelled as pure methane gas with 2,000 ppm of H₂S in consequence analysis. It is highlighted that biogas storage area is a fully open area with no major congestion, and thus the risk of vapour cloud explosion (VCE) is considered to be low(note that congestion and confinement are key conditions for VCE formation). Therefore, all delayed ignition events were modelled as flash fire in QRA.

Other Chemicals Stored Onsite

8.6.12 The storage quantities of all chemicals stored or handled onsite within TPSTW are listed in Table 8.13 and Table 8.14 with storage location in Appendix 8.3. These chemicals are not anticipated to present any major hazard which could potentially impact offsite area.

Table 8.13 Storage Quantities of Dangerous Goods

Substance	Class	Usage	Estimated Storage Quantity
Thinners	Class 9	Plant operation	2,000L
Hydrochloric acid (HCl) @38% commercial concentration	Class 8	Cleaning of primary filter	10,500L (10.5 m^3)
		Backwashing of ultrafiltration of reclaimed water	5,600L (5.6 m^3)
Sodium Hypochlorite (NaOCl) @10% commercial concentration	Class 2	Cleaning of primary filter	10,500L (10.5m^3)
		Pre-chlorination of reclaimed water	28,000L (28 m^3)
		Cleaning of ultrafiltration of reclaimed water	22,400L (22.4 m^3)
		Residual chlorine dosage	1,400L (14 m^3)
Sodium hydroxide (NaOH) @50% commercial concentration	Class 8	Cleaning of ultrafiltration of reclaimed water	4,200L (4.2 m^3)

Table 8.14 Storage Quantities of Other Chemicals

Substance	Usage	Estimated Storage Quantity
Sodium Bisulphite (NaHSO3) @10% diluted concentration	Pre-chlorination of reclaimed water	2.7m^3
Carbon dosing (sugar)	Denitrification	16 tons
Ferric chloride (FeCl3), diluted	Coagulation / anaerobic digester sulphide suppression	526 m^3
Polymer	Flocculation / centrifuge thickening and pre-dewatering, or for hydraulic filter press	58 tons
Anti-scalant	Prevention of scaling and fouling at water reuse facility	2.7m^3

Hazardous Scenarios

Digester Failure

8.6.13 Failure of the digesters could be caused by undetected corrosion, fatigue material or construction defect. Release of biogas could be from various parts of the digesters as well as the associated piping and devices. Possible hazardous outcomes include fireball, jet fire, flash fire and toxic release.

Gas Holder Failure

8.6.14 Dry seal (Wiggins) type biogas holders will be used for the proposed facility. A dry seal (Wiggins) type gas holder is different from column guided water-sealed gas holder that do not have a gas holder crown. A seal is installed between the piston and the inside of the shell to maintain gas tightness inside the holder and prevent rotation or side movement of the piston. A levelling system consists of wire ropes and balance weights are equipped to prevent tilting of the piston. The seal and the levelling system will be inspected regularly.

8.6.15 Failure of the gas holders could be caused by undetected corrosion, fatigue, material or construction defect. Release of biogas could be from various parts of the gas holders or associated piping and devices. Possible hazardous outcomes include fireball, jet fire, flash fire and toxic release.

Sulphur Absorption Vessel Failure

8.6.16 The absorbents used for removal of H₂S in the sulphur absorption vessels are neither flammable nor explosive that the major hazard will be from the release of biogas. Failure of sulphur absorption vessels could be caused by undetected corrosion, fatigue, material or construction defect. Release of biogas could be from various parts of the process vessels as well as associated piping and devices. Possible hazardous outcomes include fireball, jet fire, flash fire and toxic release.

Aboveground Inlet or Outlet Piping Failure

8.6.17 Piping will be used to connect process vessels to the gas holder, compressor, and further purification unit and Combined Heat and Power Generation System (CHP). Failure along the on-site piping may be caused by undetected corrosion, fatigue, material or construction defect, or associated with flange gasket / valve leakage resulting in continuous gas release to the atmosphere. Failures of gaskets and valve leak only tend to give relatively small scale leakage and will not contribute to any off-site risk. Nonetheless, gasket and valve leak failure were considered and included into pipework failure in this hazard assessment with reference to previous similar study. Possible hazardous outcomes from aboveground piping jet fire, flash fire and toxic release.

Construction Activities

8.6.18 During the construction phase of the Project, all major construction activities should be arranged with either adequate setback or physical barrier from the existing gas installations. Implementation of those major construction works and movement of plants and vehicles would be controlled to have suitable setback clearance, or physical barrier with an empty digester / gas holder from the digesters / gas holders in operation. As such, no additional potential biogas hazard during the construction phase was considered in the QRA.

Vehicle Impact

8.6.19 Only authorized vehicles will be permitted to enter the proposed TPSTW with restriction of speed for vehicle movements in the site. Safety markings and marked crash barriers will be provided to the above ground piping, digesters and gas holders near the access road. Vehicle impact could cause leakage and rupture to digesters and gas holders [39].

8.6.20 The hazardous events considered are summarized in Table 8.15.

Table 8.15 Hazardous Systems Identified at Biogas Facilities within TPSTW

Hazardous Systems in TPSTW	Main Equipment	Hazardous Material*	Hazard Events
Digester	Vessel and Piping	Biogas with H2S	Fire ball, Jet fire, Flash fire, Toxic
Biogas Holder	Vessel and Piping	Biogas with H2S	Fire ball, Jet fire, Flash fire
H2S Removal	Vessel and Piping	Biogas	Fire Ball, Jet fire, Flash fire,
Booster	Compressor	Biogas	Fire Ball, Flash fire, Jet fire
Dehumidifier	Vessel, Heat Exchanger, and Piping	Biogas	Fire Ball, Flash fire, Jet fire
Siloxane Removal	Vessel and Piping	Biogas	Fire Ball, Flash fire, Jet fire

8.6.21 For each of the hazardous system, a range of leak sizes have been modelled to represent the potential failure scenarios following previous QRA study [33]:

n Process Equipment, Biogas Holders and Piping

· 25 mm hole

· Full bore rupture of piping

· Catastrophic failure of pressure vessel

Frequency Analysis

Failure Data

8.6.22 With reference to the similar QRA study [33], the base equipment failure frequencies applicable for the biogas facilities are summarized in Table 8.16.

Table 8.16 Failure Frequency Data

Component	Unit	Hole Size (mm)		Reference
		25mm	Rupture
Vessel	Per vessel per year	1.00E-04	1.00E-05	RIVM [7]
Compressor	per compressor per year	5.00E-05	1.00E-05	RIVM [7]
Piping (per meter)	per m-year	5.00E-07	1.00E-07	RIVM [7]

Vehicle Impact

8.6.23 In addition to the base failure rate, the accident rate due to vehicle impact was estimated based on statistical data for Vehicle/ Object Crash accident involving medium and heavy goods vehicles in recent years. The overall numbers of accidents involvements of Medium / Heavy Goods Vehicles (M/HGVs) [47] in Hong Kong are shown in Table 8.17. From 2006 to 2020, the overall high and medium impact accident involvement rate per million vehicle kilometer is 0.16, and therefore the vehicle crash frequency can be estimated as 1.6 x 10-7 per vehicle kilometer per year. This frequency has been carried forward to derive the failure frequency in biogas facilities, as detailed in the fault tree analysis in Appendix 8.11.

Table 8.17 Accident Involvements of Medium / Heavy Goods Vehicles in Hong Kong

Serious and Fatal Vehicle Involvements of M/HGVs	2006	2007	2008	2009	2010	2011	2012	2013	2014	2015	2016	2017	2018	2019	2020	Average
Involvement rate: per million vehicle-km	0.86	0.82	0.8	0.76	0.83	0.91	0.89	0.87	0.93	0.86	0.96	0.94	0.90	0.95	0.91	0.88
Total Involvements	1155	1081	1045	907	1031	1141	1105	1085	1125	1063	1167	1162	1083	1093	1008	1083
Fatal Involvements	25	21	17	27	16	21	17	25	23	23	18	26	19	22	17	21
Serious Injury Involvements	212	188	176	147	163	196	175	193	170	250	171	146	134	137	120	171.87
Fatal Vehicle Involvements ratio	2.2%	1.9%	1.6%	3.0%	1.6%	1.8%	1.5%	2.3%	2.0%	2.2%	1.5%	2.2%	1.8%	2.0%	1.7%	2.0%
Serious Injury Involvements ratio	18.4%	17.4%	16.8%	16.2%	15.8%	17.2%	15.8%	17.8%	15.1%	23.5%	14.7%	12.6%	12.4%	12.5%	11.9%	15.9%
High Impact Accident Involvement rate per million vehicle km	0.02	0.02	0.01	0.02	0.01	0.02	0.01	0.02	0.02	0.02	0.01	0.02	0.02	0.02	0.02	0.02
Medium Impact Accident Involvement Rate per million vehicle km	0.16	0.14	0.13	0.12	0.13	0.16	0.14	0.15	0.14	0.20	0.14	0.12	0.11	0.12	0.11	0.14

Projectile Risk from Linde HKO Site

8.6.24 There are significant numbers of cylinders stored in Linde HKO site and cylinder projectiles have potential to damage biogas holder leading to catastrophic failure. Noted that biogas holder in West Plant and new biogas holder are within 300m – 468m range from the Linde HKO site. Biogas holder in East Plant is outside the projectile impact range. The frequency calculation is included in Appendix 8.10. The probability of projectile hitting each biogas holder is found to be 9.5E-08/year, which is based on 3000 cylinders assumed in the Linde HKO Site. Although other facilities (i.e. LPG storage in Apex Print Limited and Zama Print Limited) may have a projectile risk, due to the limited numbers of vessels as compared to 3000 cylinders in HKO, the projectile impact is expected to be insignificant (i.e. below 1E-09/year). Therefore projectile impact only considered risk from Linde HKO Site.

8.6.25 The summary of event frequencies used for biogas facility assessment along with the study case year is presented in Table 8.18.

Table 8.18 Event Frequencies

Biogas Concerned Facility (Applicable Year)	Equipment	Scenario	Failure Rate (per year)
Existing - West (Applicable for 2025)	Digester	25mm leakage	2.68E-04
		Full bore rupture of piping	1.37E-05
		Catastrophic failure of pressure vessel	2.00E-05
	Biogas Holder	25mm leakage	2.52E-04
		Full bore rupture of piping	3.04E-05
		Catastrophic failure of pressure vessel	1.01E-05
Existing - East (Applicable for 2025, 2030, and 2036)	Digester	25mm leakage	3.81E-04
		Full bore rupture of piping	1.62E-05
		Catastrophic failure of pressure vessel	3.00E-05
	Biogas Holder	25mm leakage	2.91E-04
		Full bore rupture of piping	1.83E-05
		Catastrophic failure of pressure vessel	2.00E-05
New - West (Applicable for 2036)	Digester	25mm leakage	1.48E-03
		Full bore rupture of piping	8.61E-05
		Catastrophic failure of pressure vessel	9.00E-05
	Biogas Holder	25mm leakage	7.40E-04
		Full bore rupture of piping	7.80E-05
		Catastrophic failure of pressure vessel	3.04E-05
	H2S Removal	25mm leakage	3.30E-04
		Full bore rupture of piping	6.00E-05
		Catastrophic failure of pressure vessel	3.00E-05
	Booster	25mm leakage	1.00E-04
		Full bore rupture of piping	2.00E-05
	Dehumidifier	25mm leakage	1.24E-04
		Full bore rupture of piping	4.80E-06
		Catastrophic failure of pressure vessel	1.00E-05
	Siloxane Removal	25mm leakage	4.21E-04
		Full bore rupture of piping	4.20E-06
		Catastrophic failure of pressure vessel	4.00E-05

Consequence Analysis

8.6.26 Consequence modelling has been performed using PHAST based on approach described in Section 8.4 . It should be noted that both flammable and toxic (H₂S) effects were modelled in consequence analysis. The consequence analysis results are presented in Appendix 8.8.

Risk Summation

8.6.27 During construction of the New West Plant from year 2025 to year 2029, the existing biogas facilities in the West and East side of TPSTW would be in operation. Demolition of existing facilities and redevelopment of new facilities in the existing West Plant would occur from year 2029 to year 2033, and during this period, only the existing biogas facilities on the East side of TPSTW would be in operation. Operational Phase would start in year 2036, covering full operation of new and existing facilities within the Project site.

Individual Risk Contours

8.6.28 The individual risk contours for the various study cases are presented in Exhibit 8.7, Exhibit 8.8, and Exhibit 8.9.

8.6.29 A salient finding is that the 1 x 10^-5 per year contour is highly localized within the immediate area around the plant equipment for both the 2025 and 2030 cases. During operational phase in 2036, the 1 x 10^-5 per year contour is found to be present but is well contained within the TPSTW site. As such, it is concluded that the individual risk criterion of HKRG has been met.

Exhibit 8.7 Individual Risk Contours – Construction of New West Plant in Proposed Expansion Site

Exhibit 8.8 Individual Risk Contours – Construction in Existing West Plant

Exhibit 8.9 Individual Risk Contours – Operational Phase

Societal Risk

8.6.30 The societal risk results have been expressed in the form of Frequency and Fatalities (F-N) curves, overlaid on the societal risk criteria for comparison. The FN curves for the three assessed Project phases are presented in Exhibit 8.10. It is observed that the harmful effects (e.g. fire and toxic dispersion) as a result of biogas loss of containment are highly localized and mostly contained within the Project premises. Therefore, the potential fatalities reflected in the FN curves are all related to the Project construction workers, who will be located inside the TPSTW. As such, no FN curve has been generated (only 1 point at N =1 generated which corresponds to potential pedestrian near the site boundary considering presence factor) for existing risk in 2025 before construction and operational phase of the Project, as offsite population is not envisaged to be impacted.

8.6.31 Since the FN curves for all studies cases fall inside the acceptable region of the risk criteria, the societal risk associated with the biogas facilities is considered to be acceptable.

8.6.32 The Potential Loss of Life (PLL) and the top contributor scenarios are presented in Appendix 8.12.

Exhibit 8.10 F-N Curve – Biogas Facilities within the Project Site (Base Case)

Sensitivity Case for Concurrent Project Construction

8.6.33 The FN curve for the sensitivity case including the concurrent project population is presented in the figure below. As can be seen, the FN curve has increased compared to the base case. The FN curve for 2025 year case, is found to be in the lower ALARP region due to the additional construction workers from other concurrent projects. However, this is only temporary during the construction in 2025; the risk in 2030 during operational phase will be reduced to the acceptable region following the completion of the construction work of the projects.

Mitigation Measures

8.6.34 It is noted that the total amount of the onsite biogas storage is significant, and may potentially exceed the threshold quantities as specified in the Hong Kong Planning Standards and Guidelines (HKPSG) Chapter 12 for flammable substance [1], i.e. 15 tonnes for Gas or any mixture of gases which is flammable in air and is held in the installation as a gas. Should the threshold quantity be exceeded, the proposed Biogas facilities may be regarded as a Potentially Hazardous Installation (PHI) requiring further application for the related permit.

8.6.35 Flammable gas and H₂S detectors with alarms shall be provided in the biogas area to alert people to initiate the appropriate emergency actions including suspension of construction work and machine shutdown which may act as ignition sources when there is leak detection.

8.6.36 Emergency response plan with details of on-site emergency procedures shall be developed for both construction phase and operational phase of the Project to cover the potential accident due to biogas releases. This plan should be extended to cover the concurrent construction project on site, and regular drills should also be performed.

8.6.37 Only authorized vehicles shall be permitted to enter the proposed TPSTW with restriction of speed for vehicle movements in the site. Safety markings and marked crash barriers should be provided to the above ground piping, digesters and gas holders near the access road.

8.6.38 Prior to the Project construction, suitable concurrent construction and operations risk and safety assessment shall be carried out to identify the potential hazards arising from the simultaneous Project construction and operation of biogas facilities onsite. Suitable risk mitigation measure shall be implemented for any significant risk activity identified. For examples, all major construction activities should be arranged with either adequate setback or physical barrier from the existing gas installations; implementation of major construction works and movement of plants and vehicles would be stringently controlled to have suitable setback clearance, or physical barrier with an empty digester / gas holder from the digesters / gas holders in operation. This hazard control assessment should be extended to cover the concurrent construction project on site.

Exhibit 8.11 F-N Curve – Biogas Facilities within the Project Site (Sensitivity Case for Concurrent Project)

8.7 QRA for Tai Po Gas Production Plant

Introduction

8.7.1 Tai Po Gas Production Plant (TPGPP) is in the Tai Po Industrial Estate, covering an area of 11.71 hectares. TPGPP is classified as a Potentially Hazardous Installation (PHI), and a Consultation Zone (CZ) has been defined around the site extending 1000m from centre of TPGPP as presented in Exhibit 8.12. The proposed Project is located within the CZ and may result in an increase in the number of persons working within the CZ. Therefore, such development plans are also required to be submitted to the Co-ordinating Committee on Land-use Planning and Control relating to Potentially Hazardous Installations (CCPHI) for consideration.

Exhibit 8.12 TPGPP and its CZ

8.7.2 The Tai Po Gas Production Plant and the Ma Tau Kok Gas Works North Plant (MTKGWNP). TPGPP is responsible for producing over 95% of the entire requirement for Hong Kong, while the remaining 5% is produced at MTKGWNP which acts to satisfy demand during peak consumption periods [13]. Exhibit 8.13 presents the layout of TPGPP inclusive of Phase I and II Production trains, the Naphtha Storage area, and the Natural Gas Receiving Station [14], which are the main production facilities of the plant.

Exhibit 8.13 Layout of TPGPP

Naphtha Storage Tanks

8.7.3 The feedstock for town gas production is naphtha, which is stored in 3 naphtha storage tanks. As per the Hong Kong PHI register [15], the TPGPP contains 80,280 tonnes of Volatile Organic Compound (VOC), which is primarily Naphtha. This is equivalent to a total of about 120,000 m³ in 3 storage tanks. The naphtha storage tanks are provided with a bund, which has been designed to contain any spillage or release of naphtha from escaping into the environment if the storage tanks leak. The bund is sized to hold 110% of the volume of one tank as per typical design standard for hydrocarbon storage tanks.

Gas Production Trains

8.7.4 Town gas is produced in eight production trains using the Catalytic Rich Gas (CRG) Process, which is a continuous, high-pressure process. A schematic of the process is provided in Exhibit 8.14[18]. Also, based on literature in the public domain [19], a functional description of each main system in the production train is provided as below:

n Naphtha from Storage Tanks is fed through the Naphtha Pre-heated and then to the Naphtha Vaporizer, where it is mixed with hydrogenation gas. Following the mixing, it is superheated to 355-380 degC in the Naphtha Vaporizer.

n The hot vapor then passed to the Sulphur Hydrogenator where organic sulphur is converted to H2S by reacting with hydrogen over a nickel/molybdenum catalyst.

n The H2S and Naphtha is fed into an H2S Absorber where the H2S is absorbed by zinc oxide.

n The Naphtha vapor and hydrogenation gas leaving the Absorber are mixed with steam to be further superheated in the Catalytic Rich Gas Heater of 450-500 degC before being passed to the Rich Gas Reactor. In the reactor, the naphtha vapor and steam react over high nickel CRG-F catalyst to produce, after multiple intermediate reactors, a mixture of CH4, CO2, CO and H2. The reaction is slightly exothermic and the hot rich gas (60-65%) leaves the reactor at a temperature of 517-548 degC.

n In order to produce town gas meeting the required consumer specification, part of the rich gas coming out of the CRG Reactor is further reacted at high temperature with additional steam in the Reformer Furnace. The resulting lean gas is mixed with the gas that bypassed the Reformer Furnace and cooled in the mixed gas.

n The CO2 content of the mixed gas also needs to be reduced. Hence a portion of the mixed gas is passed through a CO2 Absorber. In the Absorber Potassium carbonate solution reacts with the CO2 in the gas forming potassium bicarbonate, which is regenerated in the Carbonate Regenerator, using waste heat from the carbonate Reboiler.

n Gas leaving the CO2 absorber is cooled and knocked out in a KO Drum and then joins the converted gas which bypassed the CO2 Absorber. The product gas is dried and sent for export into a common town gas export header.

Exhibit 8.14 CRG Process Flow Diagram [18]

8.7.5 Town gas’s production rate at each month varies significantly depending on the consumption. The Hong Kong Monthly Digest of Statistics [17] provided the monthly town gas consumption of Hong Kong from August 2019 to August 2021 in Terajoules (TJ). Using this data, the average consumption rate of town gas in Hong Kong can be calculated based on calorific value of town gas of 17.27 MJ/m³ [18]. Assuming that 95% of town gas in Hong Kong is produced at TPGPP [13], the average production rate at TPGPP has been estimated to be 173,000 m³/hr.

Hazard Identification

8.7.6 The main hazardous materials in TPGPP include:

n Naphtha

n Natural Gas

n Process Gas

n Product Town Gas

n Landfill Gas

n Diesel

n Other Dangerous Goods

8.7.7 Naphtha is a colorless flammable hydrocarbon liquid in normal conditions. It is a mixture of various hydrocarbons. For this study, naphtha has been modelled as 33 wt% n-hexane, 50 wt% n-pentane and 17 wt% cyclohexane. The properties of Naphtha [16] are presented in Table 8.19. The hazard events applicable to Naphtha loss of containment include pool fire and flash fire.

Table 8.19 Properties of Naphtha

Property	Details
Flammability	Flammable
Auto-ignition Temperature	>220 degC
Flash Points	-20 degC
Lower Flammability Limit	1.1%
Upper Flammability Limit	5.9%
Specific Gravity	0.67 (water = 1)
Vapor Pressure	0.6 kPa (@ 20degC)
Vapor Density	4 (air = 1)

8.7.8 Natural Gas is essentially methane in composition, and is a relatively clean fuel. Natural Gas was introduced as feedstock to the TPGPP as feedstock, in addition to naphtha to produce town gas.

8.7.9 The intermediate process gases which are involved in the town gas production process at TPGPP are Hydrogenation Gas, Rich Gas coming out of the Rich Gas Reactor Outlet, Reformed Gas coming out of the Reformer outlet, Mixed Gas upon mixing of Reformed Gas and Rich Gas, Convert Gas coming out of the CO converter. As such, some of the equipment involves processing of gas at very high temperature. So, process gases are all considered to be highly flammable due to composition of methane and hydrogen. For equipment containing significant process gas inventory, such as reactor and process vessels, catastrophic rupture.

8.7.10 The final product of the process at TPGPP is town gas. TPGPP produces town gas which meets a required consumer specification. It is a colorless and odorless gas which is buoyant under ambient conditions. In the town gas production train, the gas coming out of the CO₂ Absorber is mixed with the converted gas which bypassed the CO₂ Absorber. This mixed gas is odorized to make the final product town gas. The gas is odorized to for easy detection upon leakage.

Table 8.20 Composition of Town Gas [18]

Composition	Mol %
CO2	18.5
CO	2.6
H2	48
CH4	30.6
Average Molecular Weight	14.8

Table 8.21 Physical Properties of Town Gas [18]

Property	Details
Calorific Value	17.27 MJ/m3
Specific Gravity	0.52
Wobbe Index	24
Weaver Flame Speed	35

8.7.11 Landfill Gas is imported from NENT and Shuen Wan Landfill Site to TPGPP as fuel gas for Reformer Furnaces. It presents similar fire hazards as with natural gas and process area above. For modelling purpose, Landfill Gas is assumed to be 100% methane conservatively

8.7.12 Diesel is a mixture of hydrocarbons, with an average chemical formula in the form of C12H23, ranging approximately from C10H20 to C15H28. Diesel is likely used as a backup fuel in TPGPP. It is a combustible liquid and is not readily ignitable unless it comes in contact with strong ignition sources such as open flames, sparks and heat. The major hazard associated with diesel leakage is pool fire upon ignition.

8.7.13 It is expected that other chemicals are also stored on site to support the daily operation of the plant. These chemicals may include gas cylinders for welding, refrigerants, acid/ base, paints, etc. However, it is expected that all DGs are storage in small quality and in dedicated storage area with adequate leak detection and spill containment/ clean up facilities. As such, it is not expected to contribute significantly towards offsite risk.

Hazardous Scenarios

8.7.14 In the event of a release of natural gas, towngas, and process gas, the hazard events that could occur include jet fire, flash fire, and VCE. As the process train area is congested with various equipment and vessels, VCE was considered to be credible. The hazard events applicable to Naphtha loss of containment include pool fire and flash fire; VCE was not considered in naphtha storage area, as the area has no significant congestion and confinement which are typical conditions required for VCE to occur.

8.7.15 Based on the review above, the hazardous systems identified in Table 8.22 have been carried on for further assessment. Details of facility systems are documented in a separate data and methodology report [58].

Table 8.22 Hazardous Systems Identified at TPGPP

Hazardous Systems in TPGPP	Main Equipment	Hazardous Material*	Hazard Events
Naphtha Importing and Tank Farm	Piping and Large storage tanks	Naphtha (L)	Pool fire, Flash Fire
Naphtha Pumping and Distribution Headers to PTs	Piping, pumps	Naphtha (L)	Pool fire, Flash Fire
Natural Gas Receiving Station and Distribution Headers to PTs	Piping, and pumps	Natural gas (G)	Flash Fire, Jet Fire, Vapor Cloud Explosion
Phase I Production Units - PT1	Piping, pressure vessel, pumps. heat exchangers	Naphtha (L), process gas (G) and town gas (G)	Pool fire, Flash Fire, Jet Fire, Vapor Cloud Explosion, fireball
Phase I Production Units – PT2	Piping, pressure vessel, pumps. heat exchangers	Naphtha (L), process gas (G) and town gas (G)	Pool fire, Flash Fire, Jet Fire, Vapor Cloud Explosion, fireball
Phase I Production Units – PT3	Piping, pressure vessel, pumps. heat exchangers	Naphtha (L), process gas (G) and town gas (G)	Pool fire, Flash Fire, Jet Fire, Vapor Cloud Explosion, fireball
Phase I Production Units – PT4	Piping, pressure vessel, pumps. heat exchangers	Naphtha (L), process gas (G) and town gas (G)	Pool fire, Flash Fire, Jet Fire, Vapor Cloud Explosion, fireball
Phase II Production Units – PT5	Piping, pressure vessel, pumps. heat exchangers	Naphtha (L), process gas (G) and town gas (G)	Pool fire, Flash Fire, Jet Fire, Vapor Cloud Explosion, fireball
Phase II Production Units – PT6	Piping, pressure vessel, pumps. heat exchangers	Naphtha (L), process gas (G) and town gas (G)	Pool fire, Flash Fire, Jet Fire, Vapor Cloud Explosion, fireball
Phase II Production Units – PT7	Piping, pressure vessel, pumps. heat exchangers	Naphtha (L), process gas (G) and town gas (G)	Pool fire, Flash Fire, Jet Fire, Vapor Cloud Explosion, fireball
Phase II Production Units – PT8	Piping, pressure vessel, pumps. heat exchangers	Naphtha (L), process gas (G) and town gas (G)	Pool fire, Flash Fire, Jet Fire, Vapor Cloud Explosion, fireball
Towngas Export and Others	Piping	Town gas (G), Landfill Gas (G)	Flash Fire, Jet Fire, Vapor Cloud Explosion

*Note: G is gas phase, and L is liquid phase

8.7.16 For each of the hazardous system, a range of leak sizes have been modelled to represent the potential failure scenarios, as summarized below based on failure data [20][21]:

n Process Equipment and Piping

10 mm hole

25 mm hole

75 mm hole

Full bore rupture of piping

Instantaneous release of pressure vessel

n Naphtha Storage Tanks

300 mm hole

1000 mm hole

Instantaneous failure of tank

8.7.17 For Naphtha storage area, the presence of the bund acts as an important safeguard against the spread of Naphtha to other process area and offsite area in the case of a release from the tanks or piping. Despite the bund being designed to handle 110% of one storage tank content, bund overfilling may occur in the case of 2 or more tanks failing simultaneously. In addition, bund overtopping can also result due to instantaneous release from a tank unzipping failure (i.e. incident capable of directly resulting in rapid loss of the tank wall), such that the released liquid can vault an inclined side or pile up rapidly at the face of a bund wall and then flow over the top or a major part of it.

Frequency Analysis

8.7.18 The equipment failure frequencies are taken from published international failure database, such as UK HSE [20] and OGP database [21], applicable for process facilities as summarized below:

Table 8.23 Failure Frequency Data for Tank

Component	Unit	Hole Size (mm)			Reference
		300 mm	1000 mm	Cat Rupture
Single Wall Atmospheric Vessel >12000m3	Per vessel per year	2.50E-03	1.00E-04	5.00E-06	UK HSE

Table 8.24 Failure Frequency Data for Equipment

Component	Unit	Hole Size (mm)					Reference
		10 mm	25 mm	75 mm	Line Rupture	Cat Rupture
Pipework (0 - 49mm)	per m-year	1.10E-05	5.00E-06	-	1.00E-06	-	UK HSE
Pipework (50 - 149mm)	per m-year	3.00E-06	1.00E-06	-	5.00E-07	-	UK HSE
Pipework (150 - 299mm)	per m-year	2.00E-06	7.00E-07	4.00E-07	2.00E-07	-	UK HSE
Pipework (300 - 499mm)	per m-year	1.80E-06	5.00E-07	2.00E-07	7.00E-08	-	UK HSE
Pipework (500 - 1000mm)	per m-year	1.70E-06	4.00E-07	1.00E-07	4.00E-08	-	UK HSE
Manual Valve	per item per year	2.00E-04	-	-	-	-	UK HSE
Actuated Valve	per item per year	2.00E-04	-	-	-	-	UK HSE
Flange (smaller than 2")	per item per year	5.00E-06	5.00E-06	-	-	-	UK HSE
Flange (2" and above)	per item per year	5.00E-06	-	5.00E-06	-	-	UK HSE
Pressure Vessel	per item per year	5.00E-05	5.00E-06	5.00E-06	-	4.00E-06	UK HSE
General Reactors	per item per year	5.00E-05	5.00E-06	5.00E-06	-	1.00E-05	UK HSE
Single Seal Pump	per item per year	5.00E-04	-	-	-	3.00E-05	UK HSE
Centrifugal Compressor	per item per year	7.20E-03	4.93E-03	1.43E-04	-	2.90E-06	UK HSE
Shell and Tube Heat Exchanger, Shell Side (50<d<150mm)	per item per year	3.62E-03	5.36E-04	-	4.52E-04	-	OGP
Shell and Tube Heat Exchanger, Shell Side d>150mm)	per item per year	3.62E-03	5.36E-04	2.10E-04	2.42E-04	-	OGP
Shell and Tube Heat Exchanger, Tube Side (50<d<150mm)	per item per year	2.47E-03	3.66E-04	-	2.77E-04	-	OGP
Shell and Tube Heat Exchanger, Tube Side d>150mm)	per item per year	2.47E-03	3.66E-04	1.28E-04	1.53E-04	-	OGP
Pig Launcher and Pig Receiver (d > 150 mm)*	per item per year	1.45E-0	2.04E-08	8.92E-09	-	1.76E-08	OGP

*Note: Assume Pigging frequency is once per 10 years and 3 hours each time

8.7.19 Using the numbers provided in the tables above, the failure frequencies related to Tai Po Gas Production Plant assessment have been calculated as documented in methodology report [58]. The failure frequency calculation associated with Naphtha storage tanks such as simultaneous catastrophic rupture of tanks, bund overfilling and bund overtopping scenarios is provided in Appendix 8.9.

8.7.20 It is to be noted that the naphtha storage area, i.e. bund area, is provided with fire suppression system. In the event of a naphtha release, it will be contained in the bund where fire / vapour suppression agent will be applied. The agent will cover the surface and blanket the hydrocarbon pool, and thereby prevent air (i.e. oxygen) coming in contact with the hydrocarbon vapours. It is noted that TPGPP implements strict work rules and procedures to control various high-risk operations, with robust backup from automatic alarms and safety interlocks. A high degree of emergency response preparedness is achieved by a wide range of fire-fighting facilities, well trained response teams and frequent emergency drills. Therefore, the fire suppression system was assumed to have a probability of failure on demand of 0.1.

8.7.21 In addition an operational factor of 0.75 has been considered for the production trains to reflect the operation [58].

8.7.22 Table 8.25 presents the total frequencies inclusive of frequencies for leak and rupture, for the sections considered in the study.

Table 8.25 Failure frequency Estimated for Hazardous Sections

Hazardous Systems in TPGPP	Main Equipment	Hazardous Material*	Failure Frequency per year
Naphtha Importing and Tank Farm	Piping and Large storage tanks	Naphtha (L)	1.20E-03
Naphtha Pumping and Distribution Headers to PTs	Piping, pumps	Naphtha (L)	5.30E-03
Natural Gas Receiving Station and Distribution Headers to PTs	Piping, and pumps	Natural gas (G)	9.30E-04
Phase I Production Units - PT1	Piping, pressure vessel, pumps. heat exchangers	Naphtha (L), process gas (G) and town gas (G)	3.90E-02^Note1
Phase I Production Units – PT2	Piping, pressure vessel, pumps. heat exchangers	Naphtha (L), process gas (G) and town gas (G)	3.90E-02
Phase I Production Units – PT3	Piping, pressure vessel, pumps. heat exchangers	Naphtha (L), process gas (G) and town gas (G)	3.90E-02
Phase I Production Units – PT4	Piping, pressure vessel, pumps. heat exchangers	Naphtha (L), process gas (G) and town gas (G)	3.90E-02
Phase II Production Units – PT5	Piping, pressure vessel, pumps. heat exchangers	Naphtha (L), process gas (G) and town gas (G)	3.90E-02
Phase II Production Units – PT6	Piping, pressure vessel, pumps. heat exchangers	Naphtha (L), process gas (G) and town gas (G)	3.90E-02
Phase II Production Units – PT7	Piping, pressure vessel, pumps. heat exchangers	Naphtha (L), process gas (G) and town gas (G)	3.90E-02
Phase II Production Units – PT8	Piping, pressure vessel, pumps. heat exchangers	Naphtha (L), process gas (G) and town gas (G)	3.90E-02
Towngas Export and Others	Piping	Town gas (G), Landfill Gas (G)	1.40E-02

Note 1: The frequency of the entire Production train is cumulative of the different sections considered in the train based on confidential process information provided by HKGC [26]. As TPGPP has multiple production trains, the respective frequency for the train is repeated for the total number of production trains. Detail of data are provided in [58].

Consequence Analysis

8.7.23 It is typical in QRA to consider that the people at elevated locations are not subject to fire hazards, such as flash fire and jet fire, for which the hazardous zones usually do not extend to higher floors. Such height protection factor for building population will be based on the consequence result finding with an assumption that each floor is 3m high [9].

8.7.24 From consequence analysis results, it is found that the maximum cloud height as a result of leak scenario for all events, except naphtha storage tank, is around 6 m while full bore rupture of piping and catastrophic rupture to pressure vessel and tank can lead to around 30 m high gas cloud. Therefore, it is reasonable to assume that people located at above 6 m or 2 storey are not exposed to the hazards associated with leak scenarios. As for catastrophic rupture scenarios, no such height protection was assumed and the entire population of buildings was used in risk calculations.

8.7.25 In the case of leak of naphtha storage tank, the effects of bund containment has been modelled in analyzing the extent of naphtha pool spread as well as the flammable vapour evaporating from liquid pool. Since the bund is fully sized for 100% of the storage tank capacity, the naphtha liquid pool following a tank leakage will be contained within the bunded area. Therefore, bunded pool fire has been modelled for leak in case of immediate ignition. Further, as the bund has a 3 m wall, the vapour cloud with a height below this level was assumed to be confined within the bund in the case of delayed ignition resulting in flammable gas dispersion.

Risk Summation

Individual Risk Contours

8.7.26 The individual risk contours for the various study cases are presented in Exhibit 8.15. It can be seen that the 1E-05 per year risk contour extends slightly from the site boundary in the north, west, and south side. However, it is important to note that the individual risk results reflect the existing level of risk, irrespective of the existence of the proposed Project. Nevertheless, considering the expected presence level of the affected features within the 1E-5 per year (industrial facilities), the actual risk exposed would be in range of 4.6E-06 per year, assuming a presence factor of 0.46, which reflects the nature of facilities based on Table 8.9 as well as the applicable protection factor of 0.1 as noted in Section 8.4.19. It is also clear that the proposed Project area is exposed to an individual risk level of lower than 1E-06 per year.

Exhibit 8.15 Individual Risk Contours – Tai Po Gas Production Plant

Societal Risk

8.7.27 From the FN curve results, as shown in the figure below, it can be deduced that the increase in risk due to the proposed Project is negligible, as the FN curves for all project construction and operational cases appear to be largely the same as the existing situation (i.e. year 2025 case). It is worth mentioning that the FN curve for existing situation has a higher N value than some of the construction and operational cases in the region between 1E-08 and 1E-09 per year. This is because the proposed Project expansion area is presently occupied by about 168 people as reflected in the existing risk results, but as the site is returned for Project development, these occupants will no longer be present in the project construction and operational phases.

8.7.28 The FN curve in the ALARP region is deemed to be the existing situation, irrespective of the proposed Project. It is worth noting that the FN assessment produced is anticipated to be conservative, since much of information relating to engineering and safety features in place in the TPGPP has not been made available for this QRA due to confidentiality reasons. As such, a separate mitigated case is also presented for reference to approximate the mitigated risk due to the effects of the expected safety features and protection. Risk reduction factor of 0.5 is used to address the following aspects:

Plant Emergency Actions - Leak Detection, Automatic Isolation and Evacuation

8.7.29 It is typical for fuel or flammable gas processing facilities like TPGPP to install sophisticated leak detection and automatic safety systems, for instance, flame and flammable detectors should be placed at strategic plant locations whereby any leak and fire detection will trigger automatic emergency plant actions such as facility shutdown, system isolation and blowdown of inventory. The time required to activate these measures can be as short as 1 minute following a leak [57]. In addition, no allowance has been made in the QRA for evacuation even though the surrounding areas are denoted as CZ where coordinated emergency communication and response is expected to be in place.

Assumptions in relation to Naphtha and Process Train Systems

8.7.30 Towngas is produced by dual feedstock of Naphtha and Natural Gas (NG) in TPGPP. Not only NG is regarded as a cleaner feedstock than Naphtha, but it is also buoyant gas with a strong tendency to disperse and dissipate in case of a release as opposed to Naphtha, which will form a flammable liquid pool upon released with subsequent flammable vapours accumulating and persisting above the pool at near ground level. In the absence of Naphtha storage operational data, the QRA assumed all three large naphtha storage tanks are full for 100% of the time. In actual operations, the storage quantities will be lower than this level. It is also anticipated that the use of the cleaner Natural Gas will increase (i.e. naphtha will decrease) in future for more environmentally friendly Towngas production. Besides, it is typical that fire suppression foam will be provided for naphtha systems; the foam can provide a varying degree of flammable vapour suppression of naphtha pool, thus lowering the flammable effect zones. However, such consequence reduction effects were not modelled in the QRA , as it is specific to the foam type and system design.

8.7.31 In relation to the process trains, process conditions were estimated based on literatures available as well as limited information provided HKCG. The data has been approximated in such a way that it will tend to give conservative estimates in risk results [58].

Past Safety Records and Safety Management System of TPGPP

8.7.32 TPGPP Phase 1 has been in operation since 1986 and Phase 2 expansion was completed in 1992, so Phase 1 has been in operation for 36 years while Phase 2 has been for 30 years. During TPGPP has started operations, there is no recorded major loss of containment scenarios that leads to injury or fatality for both onsite operators and offsite public. Based on the frequency analysis performed, apart from piping, there are more than 5,000 components (accounting for vessels, pumps, flanges, valves, etc.) in the overall TPGPP. With operating year of 30 to 36 years, the overall operating experience is in the range of 150,000 component-year.

8.7.33 With statistical techniques, it is possible to estimate the frequency of events related to systems where ‘zero fatal accident’ have occurred. It is assumed that the accident obeys a Poisson distribution in the time interval while the intervals between the events follow an exponential distribution. The Poisson Distribution equation can be written as:

where,

x is a discrete random variable;

p(x) is the probability function of x; and

l is a positive constant

8.7.34 In the case of ‘zero fatal accident’ where x = 0, the equation will be simplified to only p(x) = e-l. For ‘zero fatal accident’, the failure rates associated with various confidence levels can then be calculated. Starting with 90% confidence level in the calculation, the 90% probability of getting ‘zero fatal accident’ would be equal to l90% = ln (1/0.1) = 2.3.

8.7.35 Applying the existing operating experience of 150,000 component-year, the ‘zero fatal accident’ at 90% confidence level would then be 2.3/150,000 = 1.53 x 10^-5 per year. Given that the IR contour result predicted individual risk of more than 1 x 10^-4 per year in TPGPP, it can be concluded with 90% confidence that currently assessed IR result based on industrial average statistics is on the conservative side.

8.7.36 It is noteworthy TPGPP has maintained a high standard of operations and safety which will reduce the chance of accident. For example, it is common for operating plants to implement Risk Based Inspection (RBI) programme, which is a systematic risk-based approach identifying the main mechanical damage mechanisms and their frequencies of occurrence for aspects such as thickness thinning, stress corrosion cracking and corrosion under insulation; the programme will assist in defining and optimising inspection method and interval for safety critical systems and equipment, thus lowering the risk of release and justifying a better safety performance over the industrial average figures. With a demonstrated RBI and / or plant Safety Regime, the effects of risk mitigation can also be quantified and incorporated into QRA; however, this has not been performed in this study in the absence of related information.

8.7.37 The Potential Loss of Life (PLL) and the top contributor scenarios are presented in Appendix 8.12.

Exhibit 8.16 F-N Curve – Tai Po Gas Production Plant

Sensitivity Case for Concurrent Project Construction

8.7.38 The FN curves for the sensitivity case including the concurrent project population is presented in Exhibit 8.17. As can be seen from the results, no significant change in the FN curves is observed which means the concurrent projects construction has no significant impact to the overall risk.

Mitigation Measures

8.7.39 Since TPGPP is an existing plant in operation for more than 20 years without any major incident, it is not feasible/practical to introduce additional safety measures inside the facility. However, to achieve ALARP, it is recommended to develop a joint emergency response plan between the building management team of development, HKCG, FSD and TPSTW in case of emergency in the TPGPP. It is of particular importance during construction phase of TPSTW due to significant increase in population contributed by construction workers. The joint emergency response plan should detail the communication protocol between TPGPP and emergency responders as well as between TPGPP and TPSTW (if required) and also review if drill is required periodically.

Exhibit 8.17 F-N Curve – Tai Po Gas Production Plant (Sensitivity Case for Concurrent Project)

8.8 QRA for Liquefied Petroleum Gas (LPG) Storage Facilities

Introduction

8.8.1 Two LPG storage facilities are identified to be in the vicinity of the Project located at Apex Print Limited and Zama Industries Limited. While Apex Print Limited is located at the south of the TPSTW, Zama Industries Limited is located at the south east of the TPSTW. The storage facilities involve underground storage of LPG and vaporizer facility.

8.8.2 The location of LPG storage and vaporizer at Apex Print Limited and Zama Print Limited are provided in Exhibit 8.18, Exhibit 8.19, and Exhibit 8.20. Table 8.26 presents the information of LPG storage tanker in Apex Print Limited and Zama Print Limited.

Table 8.26 Information of LPG storage tank in Apex Print Limited and Zama Print Limited

Information	Details
Number of LPG underground storage tank in each site	2
The volume of each LPG storage tank in ZAMA site	11.5 kL
The volume of each LPG storage tank in Apex site	5 tons
The frequency of LPG tanker visit for LPG storage re-filling to each site	Once every 3 days*

* Based on collected facility information, LPG tanker visit is expected approximately every 4- 10 days. To allow for future increase in consumption, a conservative visiting frequency of every 3 days was assumed in the QRA.

Exhibit 8.18 Location of LPG Storage and Vaporizer at Apex Print Limited

Exhibit 8.19 Plot Plan of LPG Storage and Vaporizer at Zama Print Limited

Exhibit 8.20 Location of LPG Storage vessel at Zama Print Limited

Hazard Identification

8.8.3 From historical review, accidents related to LPG storage facility involved uncontrolled release of LPG from road tanker, filling connection incidents, damaged pipe, etc. In particular, Boiling Liquid Expanding Vapour Explosion (BLEVE) has been identified as a major risk for LPG road tanker in case of an external fire.

8.8.4 LPG is a mixture of propane and butane with composition of 30:70 by weight. Due to its mass which is heavier than the air, LPG tends to stay closer to the ground in case of a leak. Table 8.27 summarizes the physical properties of LPG.

Table 8.27 Physical Properties of LPG

Parameter	Value
Composition (propane : butane)	30:70 (by weight)
Molecular Weight (average)	54.5
Liquid Density @25^oC	545.8 kg/m^3
Normal Boiling Point	-23.7^oC
Lower Flammability Limit	16488 ppm
Upper Flammability Limit	91743 ppm

8.8.5 LPG release could lead to these following hazardous events:

n Jet fire

n Flash fire

n Vapour Cloud Explosion (VCE)

n Fireball

n BLEVE

Hazardous Scenarios

8.8.6 The main hazard associated with an LPG facility is fire or explosion upon ignition of an uncontrolled release of LPG. Another main hazard is Boiling Liquid Expanding Vapour Explosion (BLEVE) of road tanker due to escalation of a fire event that engulfs the road tanker. Hose connection or disconnection errors in addition to tanker drive away during unloading from tanker to vessel is also considered as a hazard that can lead to leaks which eventually become fire or explosion upon ignition.

8.8.7 The following events could lead to an LPG release [27]:

n Spontaneous and partial failure of pressurized LPG equipment

Storage vessel failure

Road tanker failure

Pipework failure

Hose failure

n Loading failures:

Hose failure

Connection / disconnection error

Tanker drive away error

Tanker impact

Loading pipework overpressurisation

Storage tank overfilling or overpressurisation

8.8.8 In line with previous similar QRA [27], the failure scenarios considered in this QRA are summarised in the following table.

Table 8.28 Hazardous Scenarios

Equipment Description	Failure Type	Hole size	Hazard Events
LPG Storage Vessel	Cold catastrophic failure	Catastrophic Rupture	Fireball, Flash Fire, Explosion
	Cold partial failure	1” Leak	Jet Fire, Flash Fire, Explosion
Flexible Hose	Guillotine failure	Full Bore Rupture	Fireball, Flash Fire, Explosion
	Partial failure	1” Leak	Jet Fire, Flash Fire, Explosion
Filling Line to Storage Vessel	Guillotine failure	Full Bore Rupture	Jet Fire, Flash Fire, Explosion
	Partial failure	1” Leak	Jet Fire, Flash Fire, Explosion
Line from Storage Vessel to Vaporizer	Guillotine failure	Full Bore Rupture	Jet Fire, Flash Fire, Explosion
	Partial failure	1” Leak	Jet Fire, Flash Fire, Explosion
Vaporizer	Guillotine failure	Full Bore Rupture	Jet Fire, Flash Fire, Explosion
LPG Road Tanker (Unloading)	Cold catastrophic failure	Catastrophic Rupture	Jet Fire, Flash Fire, Explosion
	Cold partial failure	1” Liquid Leak	Fireball, Flash Fire, Explosion
		2” Liquid Leak	Jet Fire, Flash Fire, Explosion
LPG Road Tanker (Transit)	Cold catastrophic failure	Catastrophic Rupture	Fireball, Flash Fire, Explosion
	Cold partial failure	2” Liquid Leak	Jet Fire, Flash Fire, Explosion
		2” Vapor Leak	Jet Fire, Flash Fire, Explosion
		1” Liquid Leak	Jet Fire, Flash Fire, Explosion
	BLEVE	BLEVE	BLEVE

Frequency Analysis

8.8.9 In line with the previous QRAs [27][28][29], Fault Tree Analysis (FTA) was carried out to quantify the various possibilities of event combinations as well as the typical safety systems in place, such as operator intervention, check valves, and excess flow valves. The detailed FTA is presented in Appendix 8.11 , which was primarily based on the derivation from the Ap Lei Chau Depot QRA[27]. Two types of failure scenarios for LPG road tankers were considered in the analysis: Unloading case and Transit case. The unloading case relates to the failure of road tanker during unloading operations of LPG when tanker is stationary, for which the analysis has referenced the deviation and FTA performed in Kai Tak Development QRA [28]. While, the transit scenario addresses the potential accidents during LPG road tanker transport inside the facility premises, for which the analysis was performed based on the SEKD[11].

8.8.10 The summary of base event frequencies used in this QRA report is presented in Table 8.29.

Table 8.29 Event Frequencies

No.	Initiating Event Description	Failure Rate (per year)
1	Cold catastrophic failure of LPG Vessel	3.65E-07
2	Cold Partial Failure of an LPG Vessel	1.01E-05
3	Cold catastrophic failure of LPG road tanker	2.25E-07
4	Cold Partial Failure of Road Tanker	1.66E-05
5	Guillotine Failure of in-let filling pipework	8.91E-12
6	Partial Failure of in-let filling pipework	4.33E-09
7	Guillotine Failure of Flexible Hose during loading to vessel	5.23E-07
8	Guillotine Failure of liquid filling line to Vaporizers	6.68E-08
9	Partial Failure of liquid filling line to Vaporizers	1.66E-06
10	BLEVE of Road Tanker	4.61E-13
11	Rupture of Vaporizers	2.24E-09

8.8.11 The event trees analysis was performed to calculate the frequencies of the possible outcomes such as jet fire, flash fire, fireball etc. are presented in Section 8.4. In addition, BLEVE scenario was considered for LPG road tanker in case of fire escalation due to fire impingement on the road tanker. The BLEVE frequency was calculated by summing the fire frequencies associated with inlet filling pipework failure, liquid line from tanker to loading hose failure, flexible hose failure, and liquid filling line to vaporizer failure. The BLEVE frequency estimation has also taken into account the probability of jet fire impingement and fire protection offered by Chartek coating on the road tanker, which is a typical provision in Hong Kong.

Consequence Analysis

8.8.12 In the consequence modelling, the LPG road tanker is assumed to be in the full capacity of 9 tonnes. All fire and dispersion modelling was performed using PHAST.

Risk Summation

Individual Risk Contours

8.8.13 The individual risk contour for LPG storage facility at Apex Print Limited and Zama Industries Limited are presented in Exhibit 8.21 and Exhibit 8.22 below.

8.8.14 Based on the result, it is found that the maximum level of individual risk generated from both LPG storage facilities does not exceed 1 x 10^-5 per year at any location. Therefore, these two LPG storages are not expected to lead to any significant risk impact to the proposed Project site during the construction and operational phase.

Exhibit 8.21 Individual Risk Contours – LPG Storage Facility at Apex Print Limited

Exhibit 8.22 Individual Risk Contours – LPG Storage Facility at Zama Industries Limited

Societal Risk

8.8.15 The FN curves for the two assessed study cases for LPG storage at Apex and Zama are presented in Exhibit 8.23 and Exhibit 8.24, and the FN curves are found to be in the acceptable region of the risk criteria. The FN curves generated for existing risk in 2025 before the Project commencement and during the construction of New West Plant in 2025 are found to be similar for both LPG sites, which suggests that the proposed Project has insignificant contribution to the overall risk level. In addition, the shape of the curves along the study case year are highly similar, implying that the increase in the risk in the subsequent year is merely reflecting the assumed growth in the road population over the years.

8.8.16 Since the FN curves for all studies cases fall inside the acceptable region of the risk criteria, the societal risk associated with both of LPG storage facilities are considered to be acceptable.

8.8.17 The Potential Loss of Life (PLL) and the top contributor scenarios are presented in Appendix 8.12.

Exhibit 8.23 F-N Curve - LPG Storage Facility at Apex Print Limited

Exhibit 8.24 F-N Curve - LPG Storage Facility at Zama Industries Limited

Sensitivity Case for Concurrent Project Construction

8.8.18 The FN curves for the sensitivity case including the concurrent project population is presented in Exhibit 8.25 and Exhibit 8.26 for LPG storage facility at Apex and Zama respectively. From both the results, no significant change in the FN curves is observed.

Exhibit 8.25 F-N Curve - LPG Storage Facility at Apex Print Limited (Sensitivity Case for Concurrent Project)

Exhibit 8.26 F-N Curve - LPG Storage Facility at Zama Industries Limited (Sensitivity Case for Concurrent Project)

8.9 Residual Impacts

8.9.1 Mitigation measures recommended to further manage and minimize the potential risk during construction and operational phases of the Project are provided in Sections 8.6.33, 8.7.39 and 8.13. Provided with proper implementation of all the recommended mitigation measures, no unacceptable risk is expected from the Project construction and operation.

8.10 Environmental Monitoring and Audit Requirements

8.10.1 Implementation of the recommended mitigation measures should be regularly audited during the construction phase.

8.11 Conclusion

8.11.1 A Quantitative Risk Assessment (QRA) was carried out to assess the potential hazard to life risk due to generation, storage, utilization, processing and transmission (if applicable) of biogas within the Project site and due to neighbouring fuel gas dangerous goods (DGs) facilities during the construction and operation of the Project. The neighbouring fuel gas DGs facilities covered in the QRA included Tai Po Gas Production Plant (TPGPP), LPG storage facility at Apex Print Limited and Zama Industries Limited, and DGs storage at Linde HKO Limited.

8.11.2 The QRA has considered the following study cases representing the various phases of the Project development:

n Construction of New West Plant (2025)

n Construction in Existing West Plant (2030), and

n Operational Phase (2036).

8.11.3 In addition, a sensitivity case has also been carried out to consider concurrent projects within the proposed Project location. The projects identified are Organic Waste Pre-treatment Centre (OWPC) which is tentatively scheduled to commence construction in 2025 for completion by 2029, and Tolo Harbour Effluent Export Scheme (THEES) that is tentatively scheduled to commence in 2025 for completion in 2031.

8.11.4 Overall, the QRA has confirmed that the construction and operations of the proposed Project would not cause significant increase to the existing risk levels of all TPSTW, TPGPP, LPG storage facilities, and Linde HKO sites.

8.11.5 The detailed assessment conclusion of each site is summarized as follows:

Biogas Facilities within Project Site

8.11.6 Both the individual risk and societal risk associated with the construction phase and operational phase of the proposed Biogas facilities were found to be within the acceptable region according to the risk criteria in Hong Kong. In addition, a projectile risk from area outside the Project site is performed to assess the probability of projectile hitting biogas holders in site. A probability of projectile hitting each biogas holders is found to be 9.5E-08/year, which is based on 3000 cylinders assumed in the Linde HKO Site. Although other facilities (i.e., LPG storage in Apex Print Limited and Zama Print Limited) may have a projectile risk, due to the limited numbers of vessels as compared to 3000 cylinders in HKO, the projectile impact is expected to be insignificant (i.e., below 1E-09/year). Therefore, projectile impact only considered risk from Linde HKO Site. A sensitivity case was performed to examine the potential impact on the risk level due to construction and operations of other concurrent projects which will be located within or in close proximity to the Project site; the risk of the sensitivity case is found to be higher compared to the base case, as expected, due to the additional onsite construction workers. In particular, the 2025 case (construction of new West plant) result was found to be slightly crossing into the ALARP region; however, this is only temporary during the concurrent construction period of new West plant. The risk during the operational phase after the construction works completed is in the acceptable region. Nevertheless, risk mitigation measures have been recommended to reduce the risk.

Tai Po Gas Production Plant (TPGPP)

8.11.7 With respect to individual risk, the Project site was found to be exposed to 1E-6 to 1E-9 per year risk level due to TPGPP. In terms of societal risk, the FN curve associated with TPGPP was found to be in the ALARP region. It should, however, be noted that this finding simply reflects the existing risk level, irrespective of the existence of the proposed Project. The risk associated with the Project is indeed not significant as compared to existing risk level, as shown in the comparison of FN curves among different assessment years.

LPG storage facility at Apex Print Limited and Zama Industries Limited

8.11.8 The proposed Project, during both construction phase and operational phase, was found to have no significant contribution to the overall risk level of both the LPG sites. The LPG sites were also assessed to comply with the individual risk and societal criteria in Hong Kong.

8.12 Recommendations

8.12.1 Based on QRA, the following recommendations have been proposed to further reduce the risk:

n It is noted that the total amount of the onsite biogas storage is significant, and may potentially exceed the threshold quantities as specified in the Hong Kong Planning Standards and Guidelines (HKPSG) Chapter 12 for flammable substance [1], i.e. 15 tonnes for Gas or any mixture of gases which is flammable in air and is held in the installation as a gas. Should the threshold quantity be exceeded, the proposed Biogas facilities may be regarded as a Potentially Hazardous Installation (PHI) requiring further application for the related permit.

n Flammable Gas and H₂S detectors shall be provided in the biogas area to alert people to initiate the appropriate emergency actions including suspension of construction work and machine shutdown which may act as ignition sources when there is leak detection.

n Emergency response plan with details of on-site emergency procedures shall be developed for both construction phase and operational phase of the Project to cover the potential accident due to biogas releases. This plan should be extended to cover the concurrent construction project on site, and regular drills should also be performed.

n Only authorized vehicles shall be permitted to enter the proposed TPSTW with restriction of speed for vehicle movements in the site. Safety markings and marked crash barriers will be provided to the above ground piping, digesters and gas holders near the access road.

n Prior to the Project construction, suitable concurrent construction and operations risk and safety assessment shall be carried out to identify the potential hazards arising from the simultaneous Project construction and operation of biogas facilities onsite. Suitable risk mitigation measure shall be implemented for any significant risk activity identified. For examples, all major construction activities should be arranged with either adequate setback or physical barrier from the existing gas installations; implementation of major construction works and movement of plants and vehicles would be stringently controlled to have suitable setback clearance, or physical barrier with an empty digester / gas holder from the digesters / gas holders in operation. This hazard control assessment should be extended to cover the concurrent construction project on site.

n It is recommended to develop a joint emergency response plan between the building management team of development, HKCG, FSD and TPSTW in case of emergency in the TPGPP. It is of particular importance during construction phase of TPSTW due to significant increase in population contributed by construction workers. The joint emergency response plan should detail the communication protocol between TPGPP and emergency responders as well as between TPGPP and TPSTW (if required) as well as review if drill is required periodically.

8.13 Reference

[1] The Planning Department of HKSAR, Hong Kong Planning Standards and Guidelines, 2021

[2] Technical Memorandum of Environmental Impact Assessment Process (EIAO-TM)

[3] Hong Kong Weather Observatory, Tai Po Kau Weather Station.

[4] The Census and Statistics Department of HKSAR, 2021 Population Census

[5] HKSAR Planning Department, Projections of Population Distribution 2021-2029.

[6] Committee for the Prevention of Disasters, Guidelines for Quantitative Risk Assessment “Purple Book”, CPR18E, 2005

[7] Reference Manual Bevi Risk Assessments, National Institute of Public Health and the Environment (RIVM), 2009

[8] HKSAR Transport Department, Annual Traffic Census, 2020

[9] AEC (July 2015), Environmental Impact Assessment for Chai Wan Government Complex and Vehicle Depot

[10] Harbour Area Treatment Stage 2A Environmental Impact Assessment, Drainage Services Department, Hong Kong, AEIAR-121/2008

[11] Comprehensive Feasibility Study for the Revised Scheme of South East Kowloon Development EIA, Territory Development Department, Hong Kong, AEIAR-044/2001

[12] Central Kowloon Route Environmental Impact Assessment, Highways Department, AEIAR-171/2013

[13] Kai Tak Development Engineering Study cum Design and Construction of Advance Works – Investigation, Design and Construction, Agreement No. CE 35/2006(CE)

[14] Natural Gas as Feedstock for Town Gas Production, HKIE/IGEM Joint Gas Seminar, 24 April 2013 , https://www.yumpu.com/en/document/read/18124356/natural-gas-as-feedstock-for-town-gas-production-near-you/7

[15] Hong Kong PHI Register 2017

[16] Appendix 11.2.2, Kai Tak Development Engineering Study cum Design and Construction of Advance Works – Investigation, Design and Construction, Agreement No. CE 35/2006(CE)

[17] Census and Statistics Department, HKSAR, Hong Kong Monthly Digest of Statistics, October 2021

[18] Towngas Website, https://www.towngas.com/en/About-Us/Hong-Kong-Gas-Business/Gas-Production

[19] Desmond Chow Kwok On: The Environmental Improvement Due to the relocation of base load Gas Production Plant from Ma Tau Kok to Tai Po Industrial Estate, September 1994

[20] UK Health and Safety Executive (UKHSE), Failure Rate and Event Data for use within Risk Assessments, 2012

[21] International Association of Oil & Gas Producers (OGP), Risk Assessment Data Directory – Process Release Frequencies, 2010.

[22] International Association of Oil & Gas Producers (OGP), Risk Assessment Data Directory – Storage Incidents Frequency, 2010.

[23] Ove Arup & Partners Hong Kong Ltd, Hazard to Life Assessment for the Airport Fuel Tank Farm (AFTF), July 2009, https://www.epd.gov.hk/eia/register/report/eiareport/eia_1722009/html/Section%2013%20(Hazard%20to%20Life)/Section%2013%20(Hazard%20to%20Life)%20.htm

[24] Hazard to Life Assessment for the Permanent Aviation Fuel Facility (PAFF), ESR Technology, February 2007, ESR/D1000190/001/Issue 2

[25] Bund overtopping – the consequence of catastrophic tank failure, A M Thyer, I L Hirst, S F Jagger, Journal of Loss Prevention in the Process Industries 15 (2002) 357-363

[26] Confidential Information provided by HKGC, 30^th August 2021

[27] ERM (2010). QRA for Shell LPG Depot

[28] Hazard to Life Assessment for Kai Tak Development (EIA-157/2008)

[29] Allied Environmental Consultants Ltd. (2002). EIA for the Proposed Joint User Complex and Wholesale Fish Market at Area 44 Tuen Mun

[30] Confidential Information provided by Linde, 2^nd September 2021

[31] Dangerous Goods Ordinance (Chapter 295)

[32] NFPA Flammability

[33] AECOM (2019), Hazard to Life Assessment for Yuen Long Effluent Polishing Plant – Investigation, Design and Construction, Agreement No. CE 3/2015(DS)

[34] Harbour Area Treatment Scheme (HATS) Stage 2A Environmental Impact Assessment Study – Investigation (Final EIA Report), Agreement No. CE 43/2005(EP)

[35] AECOM (2020), QRA for the Public Housing Development at Sheung Shui Areas 4 and 30 Site 1, Phase 1 and Phase 2 for Hong Kong Housing Authority

[36] ERM (2016), Additional Gas-fired Generation Units Project – EIA Study, EIA237/2016

[37] AWN Consulting Ltd. (2019), Comah Land Use Planning Assessment of Proposed Development at Intel Ireland Ltd.

[38] Tierney, K. J., & Anderson, R. (1990). Risk of hazardous materials release following an earthquake. Preliminary Paper #152. University of Delaware, Disaster Research Centre,

[39] Environmental Protection Department. (2013). Environmental Impact Assessment for Organic Waste Treatment Facilities, Phase 2 (Register No.: AEIAR-180/2013). Prepared by Mott MacDonald.

[40] Code of Practice for the Electricity (Wiring) Regulations, Electrical and Mechanical Services Department, HKSAR, 2020.

[41] Aviation Statistical Reports, US National Transportation Safety Board.

[42] “Air Traffic Statistics.” Civil Aviation Department, HKSAR. 29 November 2021. https://www.cad.gov.hk/english/statistics.html

[43] Geotechnical Engineering Office. (12002). Seismic hazard analysis of the Hong Kong region. GEO Report No. 65, Geotechnical Engineering Office, Government of the Hong Kong SAR.

[44] Geotechnical Control Office. (1991). Review of Earthquake Data for the Hong Kong Region. GCO Publication No. 1/91, Civil Engineering Services Dept., Hong Kong Government.

[45] Scott, D.N., Pappin, J.W., Kwok, M.K.Y., Seismic Design of Buildings in Hong Kong, Hong Kong Institution of Engineers, Transactions, Vol. 1, No. 2, p.37-50, 1994

[46] Byrne, J. P. (1997). The calculation of aircraft crash risk in the UK, HSE\R150. Health and Safety Executive.

[47] “Road Traffic Accident Statistics.” Transport Department, HKSAR. 10 July 2018. www.td.gov.hk/en/road_safety/road_traffic_accident_statistics/index.html.

[48] Baum M. R., 1988, Disruptive failure of pressure vessels – preliminary design guidelines for fragment velocity and the extent of the hazard zone, J. Press. Vess. Technol. – Trans ASME 1988:110:168-76.

[49] Holden, P.; Reeves, A., (1985). Fragmentation Hazards from Failures of Pressurized Liquefied Gas Vessels. Assessment and Control of Major Hazards. IChemE Symposium Series # 93 IChemE, Rugby, UK. 205-220.

[50] ENSR, 2008, Harbour Area Treatment Scheme (HATS) – Stage 2A EIA Study – Investigation, Agreement No. CE 43/2005.

[51] Linde, Composition for each chemical stored in the site

[52] CCPS, 1999, Guidelines for Evaluating the Characteristics of Vapor Cloud Explosions, Flash Fires, and BLEVEs, Center for Chemical Process Safety.

[53] Hong Kong Observatory, Climate of Hong Kong, https://www.hko.gov.hk/en/cis/climahk.htm

[54] Linde Group, cylinder information, https://www.lindedirect.com/resources/product-information/cylinders-gases-consumables/gas-cylinders-and-liquid-dewars/cylinder-size-information

[55] News.gov.hk ,HK less prone to tremors, tsunamis, expert says, https://www.news.gov.hk/isd/ebulletin/en/category/environment/html/60a6be49-12f3-4d66-bc86-ac4eca26ed14.htm

[56] Gist if LSPS Application No. LSPS/001, https://www.devb.gov.hk/filemanager/en/content_1154/LSPS_001_Gist.pdf

[57] Singapore Government, QRA Technical Guidance, 9 November 2016

[58] IRESC, QRA Methodology and Assumptions Report for TPGPP, 7 April 2022

[59] LPG QRA(https://www.epd.gov.hk/eia/register/report/eiareport/eia_1852010/EIA/PDF/Appendices/7%20Hazard/App7B.pdf)