10.1.1. The proposed Project comprises the construction of a new 6 storeys vehicle depot-cum-office building at Chai Wan for the Hong Kong Police Force (HKPF), Food and Environmental Hygiene Department (FEHD), Electrical and Mechanical Services Department (EMSD) and Government Laboratory (GL). Construction of the proposed Project will tentatively be commenced in period from Mid 2016 to Mid 2017, depending on the design process.
10.1.2. As stated in Clause 3.4.4 of the EIA Study Brief (No. ESB-267/2014), hazard assessments are required for the construction stage and operation stage of the proposed Project that associated with DG processes, storage facilities and any potentially hazardous sources in the neighbourhood of the proposed Project, including but not limited to the Sinopec HK (Ex-CRC) Oil Terminal Chai Wan, two petrol-cum-LPG filling stations, a LPG wagon parking site, NWFB Depot, Citybus Depot and the future DG stores in the proposed Project. The locations of the Project site and the hazardous sources are presented in Figure 10.1.
10.1.3. The Hazard to Life Assessment is prepared in accordance with the technical requirements as set out in Appendix B of the EIA Study Brief. The technical requirements are replicated below:
Construction Phase
10.1.4. The Applicant shall carry out hazard assessment to evaluate the risk to construction workers of the proposed Project due to the neighbouring DGs processing and storage facilities (including but not limited to Sinopec HK (Ex-CRC) Oil Terminal Chai Wan, two petrol-cum-LPG filling stations operated by ExxonMobil at Sheung Mau Road and Sinopec at Chong Fu Road respectively, LPG wagon parking site at junction of Sheung On Street and Sheung Ping Street, as well as any nearby potentially hazardous sources). The hazard assessment shall include the following:
(a) Identify hazardous scenarios associated with the neighbouring DG processing and storage facilities with a view to determining a set of relevant scenarios to be included in a Quantitative Risk Assessment (QRA);
(b) Execute a QRA of the set of hazardous scenarios determined in item (a), expressing population risks in both individual and societal terms;
(c) Compare individual and societal risks with the criteria for evaluating hazard to life stipulated in Annex 4 of the TM; and
(d) Identify and assess practicable and cost-effective risk mitigation measures.
Operation Phase
10.1.5. The Applicant shall carry out hazard assessment to evaluate the off-site population risk due to the operation of the proposed Project, and the on-site risk to project workers from neighbouring DG processing and storage facilities (including but not limited to Sinopec HK (Ex-CRC) Oil Terminal Chai Wan, two petrol-cum-LPG filling stations operated by ExxonMobil at Sheung Mau Road and Sinopec at Chong Fu Road respectively, LPG wagon parking site at junction of Sheung On Street and Sheung Ping Street, as well as any nearby potentially hazardous source). The hazard assessment shall include the following :
(a) Identify hazardous scenarios associated with the operation of the proposed Project, and the neighbouring DG processing and storage facilities (including but not limited to Sinopec HK (Ex-CRC) Oil Terminal Chai Wan, two petrol-cum-LPG filling stations operated by ExxonMobil at Sheung Mau Road and Sinopec at Chong Fu Road respectively, LPG wagon parking site at junction of Sheung On Street and Sheung Ping Street, as well as any nearby potentially hazardous source), with a view to determining a set of relevant scenarios to be included in a QRA;
(b) Execute a QRA of the set of hazardous scenarios determined in item (a), expressing population risks in both individual and societal terms;
(c) Compare individual and societal risks with the criteria for evaluating hazard to life stipulated in Annex 4 of the TM; and
(d) Identify and assess practicable and cost-effective risk mitigation measures.
10.1.6. The Applicant shall conduct cumulative risk assessment of DG to evaluate the risk due to operation of the proposed Project and the neighbouring DG processing and storage facilities (including but not limited to Sinopec HK (Ex-CRC) Oil Terminal Chai Wan, two petrol-cum-LPG filling stations operated by ExxonMobil at Sheung Mau Road and Sinopec at Chong Fu Road respectively, LPG wagon parking site at junction of Sheung On Street and Sheung Ping Street, as well as any nearby potentially hazardous source).
10.1.7. The methodology to be used in the hazard assessment shall be consistent with previous studies having similar issues. (e.g. NWFB Depot at Chai Wan, ESB-034/1999 and Proposed Headquarters and Bus Maintenance Depot in Chai Wan ESB-065-2001).
10.1.8. A Hazard Assessment shall be conducted for projects if, and only if, risk to life is a key issue with respect to Hong Kong Government Risk Guidelines as specified in Section 12 of the Environmental Impact Assessment Ordinance Technical Memorandum (EIAO-TM). The estimated risk levels of hazardous sources are compared with the risk guidelines stipulated in the EIAO-TM Annex 4 to determine the acceptability. As set out in the EIAO-TM Annex 4, the risk guidelines comprise the following two components:
10.1.9. 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.
10.1.10. Societal Risk: it can be presented graphically as in Figure 10.2. The Societal Risk Guideline is expressed in terms of lines plotting the frequency (F) of N or more fatalities in the population from accidents at the facility of concern. In the figure, ALARP means As Low As Reasonably Practicable. Risk in this region should be mitigated to As Low As Reasonably Practicable.
10.2.1. A hazard assessment is conducted to evaluate:
· the risk to construction workers associated with the Sinopec HK (Ex-CRC) Oil Terminal Chai Wan, two petrol-cum-LPG filling stations, a LPG wagon parking site, as well as any nearby potentially hazardous sources;
· the risk to project workers in the proposed Project due to the Sinopec HK (Ex-CRC) Oil Terminal Chai Wan, two petrol-cum-LPG filling stations, a LPG wagon parking site, the DGs within the proposed Project, as well as any nearby potentially hazardous sources; and
· the risk to off-site population due to the operation of the proposed Project.
10.2.2. The methodology of the hazard assessment consists of the following tasks that are presented in Figure 10.3:
· Data / Information Collection: collects relevant data / information which is necessary for the hazard assessment.
· Hazard Identification: identifies hazardous scenarios associated with the operation of the potentially hazardous sources and its associated facilities by conducting HAZID workshops with facilities operators and reviewing historical accident database, such as MHIDAS, and relevant similar studies. A set of relevant scenarios are then determined to be included in a hazard to life assessment.
· Frequency Assessment: assesses the likelihood of occurrence of the identified hazardous scenarios by reviewing historical accident data, previous studies or using Fault Tree Analysis. Event Tree Analysis is adopted to determine the possible outcome from the identified hazardous events and to estimate the frequencies.
· Consequence Assessment: establishes consequence assessment for every outcome developed from initial event by using internationally well recognised consequence model – PHAST in the SAFETI Package, to assess the impacts from gas leaks, fires, explosions, toxicity and other process hazards.
· Risk Assessment: evaluates the risks level, in terms of individual risk and societal risk, associated with the identified hazardous scenarios. The overall risk level is compared with the criteria as stipulated in Annex 4 of the TM to determine their acceptability. Mitigation measures will be identified where the risk is considered in the ALARP (As Low As Reasonably Practicable) region or above. The reduction in risk achievable by these means will then be quantified. The cost-effectiveness and the practicable of these measures will also be assessed.
10.2.3. The construction of the proposed Project will tentatively be commenced in period from Mid 2016 to Mid 2017 depending on the design process and will last for about 29 months. The hazard assessment covers five scenarios as listed below:
· Case 1 – Construction Stage in Year 2016 without construction workers. This case aims to assess the risk to the surrounding population in Year 2016 that acts as the baseline to Case 2;
· Case 2 – Construction Stage in Year 2016 with construction workers. This case studies the risk impact on the construction workers of the proposed Project;
· Case 3 – Operation Stage in Year 2018 without Project workers. To study the risk to the future population in Year 2018 due to potential hazardous sources;
· Case 4 – Operation Stage in Year 2018 with Project workers only. To study the risk to the proposed Project population only in Year 2018 due to the potential hazardous sources; and
· Case 5 – Operation Stage in Year 2018 with Project workers and all off-site population. It determines the accumulated risk to the future population due to all potential hazardous sources and the DG stores in the proposed Project.
10.2.4. The Project site is located at Chai Wan and is currently allocated as a works and staging area by the Drainage Services Department (DSD). The neighbouring lands of the Project site are mainly government institutes, outdoor car parks and industrial buildings. The nearest residential building is around 160 m from the Project site.
10.2.5. Besides the potential hazardous sources mentioned in Section 3.4.4.2 of the EIA Study Brief, two bus depots adjacent to the Project site are also identified due to the possible hazards arose from handling and transferring of diesel for bus maintenance works. Table 10.1 below lists out the potential hazardous sources to be assessed.
Table 10.1 List of Potential Hazardous Source
|
Potential Hazardous Source |
Location |
|
Sinopec HK Oil Terminal Chai Wan |
Chong Fu Road |
|
Sinopec Petrol-cum-LPG Filling Station |
Chong Fu Road |
|
ExxonMobil Petrol-cum-LPG Filling Station |
Sheung Mau Road |
|
LPG Wagon Parking Site |
At the junction of Sheung On Street and Sheung Ping Street |
|
Diesel Storage in NWFB Depot |
At the junction of Chong Fu Road and Sheung On Street |
|
Diesel Storage in Citybus Depot |
Shing Tai Road |
|
DGs Storage in the Proposed Project |
Sheung On Street |
10.2.6. The following data / information is collected:
· Details of the potential hazardous sources:
· Population
· Meteorological data
· Source of ignition
· Construction and operation activities
Sinopec HK Oil Terminal Chai Wan
10.2.7. The Sinopec HK Oil Terminal Chai Wan is about 170m to the north of the Project Site. According to the site surveys conducted by the Consultants in July 2014, September 2014 and February 2015, the oil terminal was under maintenance and there was temporarily no schedule to resume operation. Nevertheless, it is assumed that the oil terminal will be in full operation during the construction and operation of the proposed Project in this study.
10.2.8. The oil terminal consists of the storage of diesel oil, kerosene and LPG cylinder storage as well as bulk filling of diesel oil according to the previous EIA report New World First Bus Permanent Depot at Chai Wan (AEIAR-029/2000) (hereinafter referred as the New World Depot EIA), the following table summarizes the details of the facility according to the report:
Table 10.2 Details of Sinopec HK Oil Terminal Chai Wan
|
Item |
Description |
|
|
Fuel Storage |
Automotive diesel oil (ADO) |
Max. Capacity: 1000m3 per
tank (2 tanks) |
|
|
Industrial diesel oil (IDO) |
Max. Capacity: 1000m3 per tank (2 tanks) |
|
|
Kerosene |
Max. Capacity: 500m3 (1 tank) |
|
|
LPG (in cylinder) |
Max. Capacity: 12 tonnes |
|
Site Dimension |
Approximate 55m × 54m |
|
|
Bund Wall Dimension |
Approximate 25m × 25m × 2m |
|
|
Fuel Transport |
ADO, IDO and kerosene by barges |
|
|
|
LPG cylinders by trucks |
|
|
Fire Wall |
150mm thick and 2.5m high R.C wall |
|
|
Fire Fighting System |
Manual operated foam injection and water spray at top of tanks |
|
|
Bulk Filling Operation |
ADO / IDO filling for road tanker, 6000 visits per year |
|
|
|
IDO and kerosene filling for drum truck, 60 visits per day |
|
10.2.9. The information listed in Table 10.2 above refers to the situations in 1999, further updates are therefore required in order to reflect the current status of the oil terminal. An information request package has been issued to Sinopec (Hong Kong) Ltd. to collect up-to-date information, which includes any change in the fuel storage type and capacity, as well as the size and frequency of fuel barge for unloading.
10.2.10. The most obvious change to the oil terminal when comparing the current layout to that in 1999, is the inclusion of a petrol-cum-LPG filling station and the demolition of one side wall for access from the Chong Fu Road. An underground LPG vessel is presumed to be installed in the oil terminal area according to the site surveys.
10.2.11. Fire Services Department (FSD) has also been consulted to acquire the information about the DG licence holding by the oil terminal.
10.2.12. It is however due to the sensitivity of the data, information requests from both Sinopec (Hong Kong) Ltd. and FSD were turned down. The data adopted for the hazard assessment on the Sinopec HK Oil Terminal Chai Wan in this study is therefore based on the New World Depot EIA with justifications.
Sinopec and ExxonMobil Petrol-cum-LPG Filling Station
10.2.13. The Sinopec and ExxonMobil Petrol-cum-LPG Filling stations are around 170 m to the north and 20m to the southwest of the Project site respectively.
10.2.14. The following information is necessary for the study:
· Layout plan of the LPG filling station;
· Number of underground LPG vessel and its capacities;
· Number of road tanker per day for unloading;
· Number of LPG vehicles per day; and
· Safety provision and firefighting system.
10.2.15. Information requests have been made to Sinopec (Hong Kong) Ltd. and ExxonMobil Hong Kong Limited to collect the aforementioned information for this study. Given the sensitivity of the operation data, both Sinopec and ExxonMobil did not provide the information required.
10.2.16. Three site surveys were conducted in July 2014, September 2014 and February 2015 to collect operation information of the two filling stations. For information not able to obtain directly through visual inspection, such as size and quantity of underground vessel, assumptions are made based on the best available information.
10.2.17. It is assumed that the underground LPG vessel to be filled up to a maximum level of approximately 85% (12 tonnes × 0.85 = 10.2 tonnes) and LPG will be delivered by a 9 tonne road tanker.
10.2.18. A summary of the data collected and the assumptions made are presented in Table 10.3 below.
Table 10.3 Details of Sinopec and ExxonMobil Petrol-cum-LPG Filling Station
|
Item |
Data Collected / Assumption |
|
|
Sinopec Filling Station |
ExxonMobil Filling Station |
|
|
LPG vessels |
Assume 1 × 12 tonnes vessel |
Assume 2 × 12 tonnes vessel |
|
LPG dispensers |
2 × LPG dispensers |
4 × LPG dispensers |
|
Vehicles refuelling |
44 LPG vehicles and 5 non-LPG vehicles per hour during peak hour |
86 LPG vehicles and 34 non-LPG vehicles per hour during peak hour |
|
LPG road tanker delivery [1] |
1.7 tankers per day deduced from number of LPG vehicles refuelling |
3.2 tankers per day deduced from number of LPG vehicles refuelling |
|
Petrol road tanker delivery [2] |
0.5 tankers per day deduced from number of non-LPG vehicles refuelling |
1.7 tankers per day deduced from number of non-LPG vehicles refuelling |
|
Petrol/diesel vessels |
Assume 2 × 22,750L vessels |
4 × 22,750L vessels[3] |
|
Petrol/diesel dispensers |
2 x petrol/diesel dispensers |
4 x petrol/diesel dispensers |
Note [1]: Assume each LPG taxi is refueled with 75% of a 109L fuel tank. And there are 2 peak hours during day and night time per day. The LPG vehicle handling volume at non-peak hour is 25% of that of peak hour. Each road tanker is fully loaded with 9 tonnes of LPG.
Note [2]: Assume each petrol vehicle consumes 40L[1]. And there are 12 peak hours in day time. The petrol vehicle handling volume at non-peak hour is 25% of that of peak hour (i.e. night time). Each petrol road tanker is fully loaded with 9 tonnes of petrol. Density of petrol is 0.75kg/L.
Note [3]: Information according to FSD
LPG Wagon Parking Site
10.2.19. The following information is collected:
· Number of LPG cylinder stored at the site; and
· Size of each LPG cylinder stored at the site.
10.2.20. The outdoor LPG wagon parking site at the junction of Sheung On Street and Sheung Ping Street is one of the designated parking spaces available for parking cylinder wagons[2]. The parking site is approximately 300 m from the Project site.
10.2.21. The site is under a short term tenancy issued by the Lands Department (LandsD) and is at the same time overseen by the Gas Authority of EMSD. According to phone conversations with EMSD and Lands Department, there is no limit on the number of vehicles allowed in the site.
10.2.22. According to a site survey conducted in July 2014, there were 3 LPG cylinder wagons at the site and the area should be able to hold approximate 12 cylinder wagons at maximum. Each cylinder wagon was stored with approximate 150 to 250 number of 10.5kg domestic use LPG cylinders inside the cargo compartment. There were no indications on whether the LPG cylinders were filled or empty.
10.2.23. A fuel road tanker was also found parking at the site together with the LPG cylinder wagons. In a site survey conducted in February 2015, it was confirmed that the road tanker contained kerosene and diesel products for bulk transport purpose.
10.2.24. The parking site is an open space that is segregated from an outdoor car park and streets by corrugated steel sheets. There is no formal firefighting system except a fire extinguisher which was placed within the parking site according to the site survey.
Diesel Storage in NWFB Depot and Citybus Depot
10.2.25. The following information is collected:
· Number and storage capacities of diesel storage tanks;
· Number of road tankers per day;
· Number of vehicles refuelled per day; and
· Length of diesel distribution pipework.
10.2.26. The New World Depot EIA[3] has assumed 6 number of 22,500 L underground tanks for diesel storage and a total of 72,000L diesel delivery per day by 4 road tankers.
10.2.27. An information request was made to the NWFB / Citybus to collect the operation data for this study in July 2014. Table 10.4 below listed out the data of the bus depots according to the reply from both depot operators.
Table 10.4 Operation Data of NWFB and Citybus Depots
|
Item |
Data Received |
|
|
Citybus Depot |
NWFB Depot |
|
|
Size of underground tank |
45,460L |
22,500L |
|
Tank max. filling percentage |
93% |
50% |
|
No. of underground tank |
4 |
10 |
|
Size of road tanker |
16,000L |
16,000L |
|
Max. no. of road tanker per day |
4 |
5 |
|
Duration of unloading |
30 minutes |
60 minutes |
|
Tanker filling hose size |
3 inches |
3 inches |
|
Bus refuelling rate per day |
300 |
470 |
|
Duration of refuelling |
70 seconds |
80 seconds |
|
Bus filling hose size |
1 inch |
1.5 inch |
|
Equipped with firefighting system, isolation valves and ESD |
Yes |
Yes (Fire drencher, fire extinguisher, ESD, hose reel and sand bucket) |
|
Any planned modification or development of the proposed Project |
No |
No |
Table 10.5 Dangerous Goods License of NWFB Depot
|
Item |
Quantity |
Storage Method |
|
Methyl Ethyl Ketone Peroxide (Cat. 10) |
176 L |
DG store on 1/F |
|
Lacquer Thinner (Cat. 5) |
216 L |
DG store on 1/F |
|
Acetone (Cat. 5) |
251 L |
DG store on 1/F |
|
Paint (Cat.5) |
704 L |
DG store on 1/F |
|
Sulphuric Acid (Cat. 3) |
360 L |
DG store on 1/F |
|
Oxygen (Cat.2) |
6.8 m3 × 12 cylinders |
DG store on 1/F |
|
Acetylene |
6.2 m3 × 12 cylinders |
DG store on 1/F |
|
Refrigerant |
1.75 m3 × 8 cylinders |
DG store on 1/F |
Dangerous Goods Storage in the Proposed Project
10.2.29. The proposed Project is located at the junction of Sheung On Street and Sheung Mau Street. The following data regarding the DG storage are provided based on the estimation provided by the Project Team:
· There is no planned DG storage for the HKPF and FEHD; and
· DG stores are planned for the EMSD and GL that will have chemical waste, Cat. 2, Cat. 3 and Cat. 5 materials etc.
10.2.30. Table 10.6 below presents the planned quantity of DGs in the future DG stores estimated by the Project Team:
Table 10.6 Dangerous Goods and Planned Quantities
|
Item |
Planned Quantity |
Planned Area |
|
Waste Oil |
6 × 200L |
NA |
|
DGs Cat. 2 |
10 cylinders |
4m2 |
|
DGs Cat. 3 |
20kg and 100L |
4m2 |
|
DGs Cat. 5 (Division 1) |
160L |
4m2 |
|
DGs Cat. 5 (Division 2) |
210L |
4m2 |
|
Battery |
2000kg (Approx. 200 units) |
9m2 |
10.2.31. Other than the type and quantity of DGs in the DG stores, the following information are also collected and considered in the study:
· The configuration and operation of ventilation / exhaust gas system such as the presence of scrubber system and emergency shut down;
· Firefighting system; and
· Emergency response plan.
10.2.32. Hazard identification (HAZID) workshops were carried out in August 2014 with the ArchSD, HKPF, EMSD and GL to understand the standard handling procedure of DGs and normal operation activities in the existing facilities to identify the potential hazards involved. The meeting minutes and the identified hazards identified from the HAZID workshops are included in Appendix 10.3.
10.2.33. The Project site is located at Chai Wan. The nearest residential building is 160 m away from the boundary of the Project site. Most of the neighbouring developments are industrial buildings and outdoor car parks that the lands are either classified as Governmental, Institution or Community (GIC), Open Space (O) or Other Specified Uses (OU).
10.2.34. The New World Depot EIA that studied the risks associated with the Sinopec HK Oil Terminal Chai Wan did not specify the study area. On the other hand, the Citybus Depot EIA that studied the ExxonMobil Petrol-cum-LPG Filling Station has specified a 150 m radius study area. In this assessment, a study area of 225 m radius will be adopted for petrol-cum-LPG filling stations according to the worst consequence distance. The study areas considered are illustrated in Figure 10.4.
10.2.35. For oil terminal and wagon parking site, the study areas will be determined based on the hazard distance of the worst case scenarios. The determination of hazard distances is further explained in Section 10.4.
10.2.36. Population to be considered in this study includes residential population, employment in commercial / industrial buildings and population using car park and playground, etc., whenever the area falls within the study areas.
10.2.37. Population data is obtained from statistic published by the Government as well as information request to the premises operators whenever applicable. Where official information is unavailable, population will be estimated based on site survey undertaken by the consultants.
· For residential buildings, figures are based on 2011 Population Census from the Census and Statistic Department. For those areas not covered by Population Census, figures are estimated from average household size using data of 2011 Territorial Population and Employment Data Matrices (TPEDM);
· Information on future development in the vicinity of the Project site is collected from Planning Department (PlanD);
· For schools, kindergartens and elderly homes, information is collected from the Education Department, Social Welfare Department and corresponding websites;
· For outdoor car parks and playgrounds, populations are estimated by surveys undertaken by the consultants; and
· For number of construction workers involved in the construction of the proposed Project, a maximum of 100 workers is assumed to work simultaneously during day time as estimated by the Project Team using a similar scale project.
10.2.38. This Hazard to Life Assessment assesses the risks in the construction (Year 2016) and operation stage (Year 2018) of the proposed Project. An average annual growth rate of -0.2% is obtained from the proposed Projection of Population Distribution 2013 – 2021[4], a zero growth rate in residential population between 2014 and 2018 will be assumed as a conservative approach. For employment population, the growth rate is calculated by considering the employment in the PDZ No. 34 of the 2011 TPEDM in between 2011 and 2021. The growth rate in between 2014 and 2016 is 0.3% and that in between 2016 and 2018 is 2.5%.
10.2.39. Traffic population in Year 2018 is estimated from the traffic flow in accordance with the traffic forecast conducted by Traffic Consultant and endorsed by Transport Department (TD). The traffic population is calculated using the following equation:
10.2.40. The traffic data is extracted from the traffic assessment and presented in Appendix 10.1. The traffic flows of the road in “am” and “pm” are taken as the sum of maximum flows in day time and night respectively in both bounds.
10.2.41. The number of people per vehicle is estimated from the Coverage (B) Station No. 2215 Island Eastern Corridor in accordance with the Annual Traffic Census 2013.
10.2.42. Since the traffic flow is expected to increase from 2016 to 2018 due to the proposed Project, the traffic population in Year 2016 is conservatively taken as the same as that of Year 2018. For the details of calculation, they are shown in Appendix 10.1.
10.2.43. For pedestrian population on the roads in the vicinity of the potentially hazardous sources, a conservative number of 50 people are assumed for each road based on the estimation from a site survey that there were around 10 people on the road.
10.2.44. For the section of MTR train falls into the study area, the population is calculated according to the information provided by the MTRC. The information provided and the calculations of MTR population are shown in Appendix 10.2.
10.2.45. A presence factor is considered for different types of buildings to account for occupancies during different times of the day. References are made to the HATS2A EIA[5], Kai Tak Development EIA[6] and other EIAs approved by the EIAO. For open space areas, population are estimated based on site observation.
10.2.46. Site specific presence factors are adopted according to the population data given by the site premises operator such as the NWFB Depot and the proposed Project through information request.
10.2.47. Time period of the day will be divided into 4 sections, namely daytime and nighttime for both weekday and weekend to reflect temporal distribution of population. The temporal changes in population for different population categories are shown in Table 10.7 below.
Table 10.7 Temporal Changes 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[2] |
100% |
10% |
40% |
5% |
|
Industrial[2] |
100% |
10% |
40% |
5% |
|
Residential[2] |
25% |
100% |
70% |
100% |
|
Recreational[2][3] |
50% |
5% |
100% |
5% |
|
Car Park[1] |
100% |
10% |
50% |
10% |
|
School[4] |
100% |
1% |
100% |
1% |
|
Cargo Working Area[1] |
100% |
10% |
100% |
10% |
|
Bus Depot[5] |
100% |
10% |
100% |
10% |
|
Govt. Complex[5] |
100% |
19% |
40% |
18% |
|
Govt. Logistics Centre[5] |
100% |
13% |
50% |
13% |
|
EMSD Workshop[5] |
100% |
18% |
100% |
18% |
Note:
[1] Based on site survey and judgment.
[2] Reference to HATS Stage 2A EIA.
[3] Reference to SEKD CFS EIA.
[4] Reference to CKR EIA[7].
[5] Based on information request.
10.2.48. An indoor ratio of 95% will be applied to the population in residential, commercial buildings and in schools considering the presence of outdoor activities. Passengers in vehicles on the roads and in trains of railway will be considered as 100% outdoors although vehicles and trains may provide certain protection. Population in the car park and the open leisure space will be considered as 100% outdoors. The populations identified are shown in Table 10.8.
Table 10.8 Population Data
|
ID |
Description |
Category |
Population 2014 |
Population 2016 |
Population 2018 |
Indoor Fraction |
No. of Storey |
Remarks |
|
1 |
Heng Fa Chuen Block 48, 49, & 50 |
Residential |
1523 |
1523 |
1523 |
95% |
22 |
2011 Population Census as per Centamap. 2538 for Block 46-50. Assume population is evenly distributed between 5 blocks. |
|
2 |
Heng Fa Chuen Playground |
Recreational |
99 |
99 |
99 |
0% |
0 |
An open space with an area of 9815 sq.m according to Outline Zoning Plan. Assume a density of 0.01 sq.m as per EIA-059/2001. |
|
3 |
Seawater Pump House |
Industrial |
0 |
0 |
0 |
95% |
1 |
Assume unmanned operation |
|
4 |
Sinopec HK Oil Terminal Chai Wan |
Industrial |
20 |
21 |
23 |
95% |
0 |
A conservative assumption |
|
5 |
Sinopec Petrol-cum-LPG Filling Station |
Filling Station |
5 |
6 |
7 |
95% |
0 |
A conservative assumption |
|
6 |
Waste Recycling Workshop |
Workshop |
10 |
11 |
12 |
10% |
0 |
Less than 5 people as per site inspection |
|
7 |
Paper Recycling Workshop |
Workshop |
10 |
11 |
12 |
10% |
1 |
Less than 5 people as per site inspection |
|
8 |
Government Logistics Centre |
Commercial |
125 |
145 |
145 |
95% |
17 |
In 2015, 110 staff + 15 visitors during day time and 5 security guards during night time. The Printing Division of GLD will move in but number of workers is not known. Assume 20 staff during day time and 1 staff during night time coming from this division. In 2018, 130 staff + 15 visitors during day time, 6 staff during night time. Information obtained from Governmental Logistic Department. |
|
9 |
Car Park A |
Car Park |
37 |
37 |
37 |
0% |
0 |
Approximate 120 parking slots as per site inspection |
|
10 |
HKE Heng Fa Chuen Substation |
Industrial |
2 |
3 |
4 |
95% |
7 |
The station is normally unmanned as per information from HKE. Assume 2 maintenance staff for emergency works. |
|
11 |
Vacant Site |
Others |
0 |
0 |
0 |
10% |
0 |
Current population as per site inspection. There is no information on future development. |
|
12 |
NWFB Depot |
Bus Depot |
590 |
594 |
624 |
95% |
5 |
As per information from NWFB |
|
13 |
Construction Site / proposed Project |
Commercial / Govt. Complex |
50 |
100 |
240 |
10% / 95% |
1 / 6 |
Currently a construction site occupied by Contractors of DSD
Project. |
|
14 |
ExxonMobil Petrol-cum-LPG Filling Station |
Filling Station |
20 |
21 |
23 |
0% |
1 |
As per site inspection. |
|
15 |
Citybus Depot |
Bus Depot |
350 |
353 |
371 |
95% |
7 |
As per information from Citybus. |
|
16 |
Planned THEi New Campus |
Educational |
123 |
4960 |
4960 |
95% |
1 |
Currently a construction site. |
|
17 |
Chai Wan Public Cargo Working Area Marine Department |
Workshop |
50 |
51 |
54 |
10% |
0 |
Around 30 people working as per site survey |
|
18 |
Hong Kong Institute of Vocational Education (Chai Wan) |
Educational |
5794 |
6227 |
6701 |
95% |
8 |
According to VTC's website, there are 39622 full time students and
13291 part time students in 10 HKIVE schools in Year 2012. Assume number of
student is equal among 10 schools, there were 5291 students in day time and
1329 students in night time in Year 2012. |
|
19 |
Knight Court |
Residential |
300 |
300 |
300 |
95% |
25 |
Comprised 2 blocks with 25 floor each and there are 2 flats per
floor |
|
20 |
MTR Chai Wan Depot |
Industrial |
100 |
101 |
107 |
95% |
1 |
According to information received from MTRC |
|
21 |
EMSD Workshop |
Industrial |
85 |
86 |
0 |
95% |
1 |
Information according to approved EIA-202/2012: |
|
22 |
Car Park B |
Car Park |
37 |
37 |
37 |
0% |
0 |
Approximate 120 parking slot as per site inspection |
|
23 |
Car park / Planned LCSD Pet Garden at Sheung On Street |
Car Park / Recreational |
10 |
16 |
16 |
0% |
0 |
Currently a LPG wagon parking site and car park. |
|
24 |
Chai Wan Industrial City, Phase II |
Industrial |
1873 |
1885 |
1980 |
95% |
22 |
Floor area is approximate 1703 sq. meter each floor. http://www.sino-industry.com.hk/en/our-properties/chai-wan-industrial-city-phase-ii/ Population estimated according to HKPSG's Business Use worker density of 20 sq. meter/worker" |
|
25 |
Chai Wan Industrial City, Phase I |
Industrial |
1902 |
1914 |
2010 |
95% |
22 |
Floor area is approximate 1730 sq. meter each floor. http://www.sino-industry.com.hk/en/our-properties/chai-wan-industrial-city-phase-i/ Population estimated according to HKPSG's Business Use worker density of 20 sq. meter/worker" |
|
26 |
Cornell Centre |
Industrial |
1390 |
1399 |
1469 |
95% |
22 |
Floor area is approximate 1265 sq. meter each floor. http://www.sino-industry.com.hk/en/our-properties/cornell-centre/ Population estimated according to HKPSG's Business Use worker density of 20 sq. meter/worker" |
|
27 |
Tsui Hong House, Tsui Wan Estate |
Residential |
1821 |
1821 |
1821 |
95% |
31 |
2011 Population Census as per Centamap. |
|
28 |
Tsui Shou House, Tsui Wan Estate |
Residential |
1758 |
1758 |
1758 |
95% |
31 |
2011 Population Census as per Centamap. |
|
29 |
Tsui Fuk House, Tsui Wan Estate |
Residential |
1912 |
1912 |
1912 |
95% |
31 |
2712 people in Tsui Fuk House and Tsui Lok Estate as per 2011
Population Census |
|
30 |
Playground |
Recreational |
60 |
60 |
60 |
0% |
0 |
Less than 20 people as per site survey |
|
31 |
Hang Tsui Court Indoor Carpark |
Residential |
48 |
48 |
48 |
95% |
6 |
153 parking slot as per site survey |
|
32 |
Hang Tsui Court |
Residential |
2101 |
2101 |
2101 |
95% |
34 |
2011 Population Census as per Centamap |
|
33 |
TWGHs & LKWFSL Mrs Fung Yiu Hing Memorial Primary School |
School |
0 |
0 |
0 |
95% |
1 |
The school is currently abandoned as per site survey |
|
34 |
Chai Wan Faith Love Lutheran School |
School |
367 |
367 |
367 |
95% |
1 |
12 classes and 26 teachers as per Primary School Profile 2013 |
|
35 |
Chong Fu Road |
Road |
55 |
55 |
55 |
0% |
0 |
50 pedestrian + population estimated from Traffic Flow Forecast |
|
36 |
Sheung Tat Street |
Road |
52 |
52 |
52 |
0% |
0 |
50 pedestrian + population estimated from Traffic Flow Forecast |
|
37 |
Sheung Mau Street |
Road |
53 |
53 |
53 |
0% |
0 |
50 pedestrian + population estimated from Traffic Flow Forecast |
|
38 |
Shing Tai Road |
Road |
80 |
80 |
80 |
0% |
0 |
50 pedestrian + population estimated from Traffic Flow Forecast |
|
39 |
Sheung On Street |
Road |
58 |
58 |
58 |
0% |
0 |
50 pedestrian + population estimated from Traffic Flow Forecast |
|
40 |
Sheung Ping Street |
Road |
52 |
52 |
52 |
0% |
0 |
50 pedestrian + population estimated from Traffic Flow Forecast |
|
41 |
Wing Tai Road |
Road |
107 |
107 |
107 |
0% |
0 |
50 pedestrian + population estimated from Traffic Flow Forecast |
|
42 |
MTR Island Line |
Railway |
163 |
187 |
214 |
0% |
0 |
Based on the information provided by MTRC |
Notes:
[1] An annual growth rate of residential population is -0.2% for Eastern District between 2011 and 2021 referring to Table 5 of Projections of Population Distribution 2013-2021. As a conservative approach, it is assumed the population is identical between 2011 and 2018.
[2] For growth rate of primary schools, it shows a decreasing trend of the number of students in Year 2008 to 2014 according to Education Bureau website. As a conservative approach, it is assumed the number of students per class remains the same between 2014 and 2018.
[3] Employment population growth rates of 0.3% (2014-2016) and 2.5% (2016-2018) per year are estimated by interpolation of employment population data of 2011 and 2021 that are given in TPEDM 2011 Table 9.
[4] Annual growth rate of 6.9% is estimated from MTRC’s Annual Report 2013. Details in Appendix 10.2.
10.2.49. Meteorological data is required for consequence modelling and risk calculation. Consequence modelling (i.e. dispersion modelling) requires wind speed and stability class to determine the degree of turbulent mixing potential whereas risk calculation requires frequencies for each combination of wind speed and stability class. The meteorological data from the North Point Weather Station in 2013 has been adopted in this assessment. The data are rationalised into a set of weather classes in accordance with TNO purple book[8] for daytime and night-time, and can be expressed in combination of wind speed and Pasquill stability classes. Pasquill stability classes (A to F) represent the atmospheric turbulence with Class A being the most turbulent class while Class F being the least turbulent class. The 6 most dominant sets of wind speed-stability classes combination for both daytime and nighttime are identified and the occurrence probability of each weather class is summarised in Table 10.9 and Table 10.10. The average ambient temperature adopted in the analysis is 23°C and relative humidity is 78%.
Table 10.9 Daytime Weather Condition
|
Wind Direction |
3.5B |
1.5D |
4.0D |
7.0D |
2.5E |
1.5F |
Total |
|
0 |
2.46 |
0.32 |
1.07 |
0.07 |
0.57 |
2.81 |
7.30 |
|
30 |
1.07 |
0.17 |
0.70 |
0.05 |
0.25 |
0.17 |
2.41 |
|
60 |
4.13 |
0.45 |
2.14 |
0.57 |
0.52 |
0.47 |
8.28 |
|
90 |
31.41 |
1.57 |
15.27 |
7.21 |
2.61 |
1.54 |
59.61 |
|
120 |
1.12 |
0.50 |
0.30 |
0.00 |
0.22 |
0.70 |
2.84 |
|
150 |
0.12 |
0.05 |
0.05 |
0.00 |
0.00 |
0.00 |
0.22 |
|
180 |
0.12 |
0.07 |
0.00 |
0.00 |
0.00 |
0.02 |
0.21 |
|
210 |
0.67 |
0.10 |
0.00 |
0.00 |
0.00 |
0.02 |
0.79 |
|
240 |
4.53 |
0.37 |
1.14 |
0.05 |
0.27 |
0.37 |
6.73 |
|
270 |
6.39 |
0.52 |
1.37 |
0.25 |
0.45 |
0.72 |
9.70 |
|
300 |
0.27 |
0.07 |
0.05 |
0.00 |
0.02 |
0.17 |
0.58 |
|
330 |
0.52 |
0.17 |
0.22 |
0.00 |
0.12 |
0.22 |
1.25 |
|
Total |
52.81 |
4.36 |
22.31 |
8.2 |
5.03 |
7.21 |
99.92 |
Table 10.10 Nighttime Weather Condition
|
Wind Direction |
3.5B |
1.5D |
4.0D |
7.0D |
2.5E |
1.5F |
Total |
|
0 |
0.00 |
0.02 |
1.59 |
0.00 |
1.90 |
11.99 |
15.50 |
|
30 |
0.00 |
0.00 |
0.87 |
0.05 |
1.23 |
0.91 |
3.06 |
|
60 |
0.00 |
0.00 |
3.56 |
0.46 |
2.12 |
2.48 |
8.62 |
|
90 |
0.00 |
0.00 |
24.52 |
7.20 |
11.84 |
8.95 |
52.51 |
|
120 |
0.00 |
0.00 |
0.43 |
0.00 |
1.37 |
3.20 |
5.00 |
|
150 |
0.00 |
0.00 |
0.00 |
0.00 |
0.10 |
0.19 |
0.29 |
|
180 |
0.00 |
0.00 |
0.00 |
0.00 |
0.02 |
0.10 |
0.12 |
|
210 |
0.00 |
0.00 |
0.05 |
0.00 |
0.10 |
0.34 |
0.49 |
|
240 |
0.00 |
0.00 |
1.52 |
0.17 |
1.28 |
2.45 |
5.42 |
|
270 |
0.00 |
0.00 |
2.21 |
0.19 |
2.24 |
2.50 |
7.14 |
|
300 |
0.00 |
0.00 |
0.12 |
0.00 |
0.10 |
0.19 |
0.41 |
|
330 |
0.00 |
0.00 |
0.26 |
0.00 |
0.51 |
0.67 |
1.44 |
|
Total |
0.00 |
0.02 |
35.13 |
8.07 |
22.81 |
33.97 |
100.00 |
10.2.50. The presence of ignition sources in the study area is a primary concern in case of flammable gas release. Ignition sources other than on-site one, such as dwellings and vehicles along carriageways, contribute to delayed ignition in Vapour Cloud Explosion (VCE) and flash fire. The energy level, timing, location of ignition sources in the vicinity of the hazardous sources and hence the probability of ignition of gas cloud are reviewed and assessed.
10.2.51. Major ignition source is the surrounding road network. Traffic volume, travelling speed (speed limit) and length of a road are used to calculate the presence time.
10.3.1. Potential hazardous scenarios associated with the transfer, storage and use of LPG, diesel oil and DGs in the potential hazardous sources are identified. Historical incidents and relevant studies are reviewed to identify possible hazardous scenarios in similar facilities. The consequences of the identified scenarios are then evaluated to confirm whether they will induce on-site risks of the proposed Project.
10.3.2. Historical incidents related to the storage, transport and use of DGs and petroleum products in oil terminals, filling station, wagon parking site and facilities of similar nature are reviewed from database such as the Major Hazard Incident Data Services (MHIDAS) database, eMARS[9] and FACTS[10] covering incidents in 1980 – 2012. Relevant incidents involved are grouped into different incident scenarios for further analysis.
10.3.3. Keywords “oil terminal” and “jetty” were searched to identify the incidents related to transfer and storage of fuel oil / petrol / crude oil. 21 out of 39 incidents related oil terminal resulted in fire and explosion that were caused by welding, lightning and over-pressurization during transfer, etc. Part of the incidents related to jetty were caused by misconnected loading arm, collision of vessel with jetty and overflow during unloading while only 22 out of 59 resulted in fire and explosion.
Table 10.11 Summary of Incidents Related to Oil Terminal and Jetty (1980-2012)
|
Incident
Origin |
No. of Cases |
Number of Fatality Involved |
|
Oil terminal |
39 |
77 |
|
Jetty |
59 |
15 |
10.3.4. 123 incidents have been identified by reviewing the database related to overfilling and over-pressurization of storage vessels and another 3 incidents are related to fire due to welding and heating. 116 people were killed in these incidents.
Table 10.12 Summary of Incidents Related to Overfilling Vessels (1980-2012)
|
Incident
Origin |
No. of Cases |
Number of Fatality
Involved |
|
Storage vessel |
126 |
116 |
10.3.5. Keywords “LPG cylinder” and “LPG wagon” were searched in the database to identify the incidents that are related to the loading / unloading and storage of LPG. 25 out of 30 incidents involving LPG cylinder resulted in fireball, flash fire and confined explosion. While 7 out of 12 of incidents involving LPG wagon resulted in fireball, BLEVE and explosion but all incidents happened during transportation instead of parking.
Table 10.13 Summary of Incidents Related to LPG Cylinder and LPG Wagon Parking Site (1980-2012)
|
Incident
Origin |
No. of Cases |
Number of Fatality
Involved |
|
LPG cylinder |
30 |
30 |
|
LPG wagon |
12 |
1 |
10.3.6. Keywords “diesel underground tank” and “diesel road tanker” were searched in the database to identify the incidents related to diesel storage. For underground diesel tank incidents, no fire was resulted due to leakage. For diesel road tanker incidents, 5 out of 12 incidents resulted in fire and explosion and only 1 related to unloading diesel to a depot. A summary of the incidents is shown in the table below.
Table 10.14 Summary of Incidents Related to Diesel Storage (1980-2012)
|
Incident
Origin |
No. of Cases |
Number of Fatality
Involved |
|
Diesel
underground tank |
6 |
0 |
|
Diesel road
tanker |
12 |
0 |
10.3.7. The keyword “warehouse fire” was used to identify those incidents that may be similar to the DG storage of the proposed Project. 228 incidents are identified in which 124 people were killed. 198 out of the 228 incidents induced fireball, Boiling Liquid Expanding Vapour Explosion (BLEVE) and explosion while the remaining involved formation of gas cloud and pool of chemical.
Table 10.15 Summary of Incidents Related to Warehouse (1980-2012)
|
Incident Scenario |
No. of Cases |
Number of Fatality
Involved |
|
Warehouse |
228 |
124 |
10.3.8. Failure events and hazardous scenarios have been identified in relevant studies such as the approved EIA reports. The failure frequencies and identified hazards in those reports have been reviewed and adopted in this study whenever applicable:
· Permanent Aviation Fuel Facility for Hong Kong International Airport (PAFF EIA)[11]
- This report evaluated the risks to the off-site population due to the storage of aviation fuel (Jet A1) in oil tanks and transportation of fuel through jetty. Hazardous scenarios and failure frequencies such as tank failure and rupture of loaded vessels have been reviewed;
· Providing Sufficient Water Depth for Kwai Tsing Container Basin and its Approach Channel (Kwai Tsing Channel EIA)[12]
- A Hazard to Life Assessment that studied the hazards associated with the storage and operation of LPG and oil depot. It reviews the hazard distances of the release scenarios identified in the PAFF EIA[13]. The worst consequences are selected and further evaluated in this study;
· New World First Bus Permanent Depot at Chai Wan (New World Depot EIA)[14]
- The diesel oil and kerosene storage in the former CRC (now Sinopec) HK Oil Terminal Chai Wan, as well as the diesel storage of a bus depot were assessed to determine the risks to the off-site population in the surrounding and to the on-site population in the bus depot. The hazardous scenarios adopted in this report have been reviewed;
· Proposed Headquarters and Bus Maintenance Depot in Chai Wan (Citybus Depot EIA)[15]
- The study assessed the risks on the bus depot associated with the operation of a planned Petrol-cum-LPG Filling Station;
· Kai Tak Development (Kai Tak Development EIA)[16]
- The DG warehouses with flammable, toxic and compressed gases were studied to identify the hazardous scenarios and to determine the risk level to the off-site population. The methodology adopted in the assessment have been reviewed ; and
· Harbour Area Treatment Scheme Stage 2A (HATS2A EIA )[17] and South Island Line (SIL EIA)[18]
- Both of the EIA reports evaluated the risk associated with the Shell LPG Depot. Hazardous scenarios, failure frequencies and event trees due to the storage of LPG cylinder and LPG vessels in wagons have been reviewed.
· Properties of Petrol, Diesel, Kerosene and Dye Marker
- Petrol, or gasoline, consists of hydrocarbons range from C4 to C12. Since the typical flash point of petrol is around – 40 oC, it forms an explosive air/vapour mixture at ambient temperature;
- Kerosene is a mixture of hydrocarbons with range C9 to C16. It has a boiling point in a range of 150 – 300 oC and a flash point typically at 38 oC. Kerosene is categorized as CAT5 Class 2 DGs;
- Diesel contains higher molecular weight compounds (C13 to C25) than kerosene and is less volatile. Its boiling point ranges from 220 – 350 oC. Due to its relatively higher flash point (around 76 oC), a release of diesel will only present a hazard if exposed to excessive temperatures such as those resulting from a fire. Diesel is categorized as CAT5 Class 3 DGs;
- Dye marker shall be 1,4-dihydroxyanthraquinone according to the Dutiable Commodities (Marking and Colouring of Hydrocarbon Oil) Regulations. Typical 1,4-dihydroxyanthraquinone has a melting point of around 200oC and a flash point in the range of 220 oC -250oC. Dye marker is categorized as CAT 5 Class 3 DG.
- Kerosene and diesel are ignited only being heated above the flash points or open flame. Incomplete combustion will generate black smoke and carbon monoxides[19],[20]; and
- Diesel, kerosene and dye marker do not pose flammable vapour hazard at ambient temperature in Hong Kong and therefore limits the hazards to liquid pool fire. Vapour cloud explosion and flash fire to off-site population can therefore be eliminated.
· Properties of LPG
- LPG supplied in Hong Kong is a pressurized mixture of propane and butane (3:7 in mole ratio). Upon release to the ambient environment it vaporises and mixes with air, forming a dense flammable gas cloud which tends to flow and disperse close to the ground. The gas cloud may extend over a long distance until it gets too diluted or encounters ignition sources.
10.4.2. Various hazardous scenarios related to the storage of aviation fuel in an oil terminal have been identified in the PAFF EIA[21]. The hazard distance of those scenarios have been further studied in the Kwai Tsing Channel EIA[22]. Having considered the similarity of kerosene in the Sinopec HK Oil Terminal Chai Wan to the aviation fuel (aviation kerosene) in PAFF, as well as the similar oil storage tank nature of oil terminal, the scenarios in the PAFF EIA will be used a basis for this Hazard to Life Assessment.
10.4.3. Those release scenarios with the worst hazardous distances are reviewed in order to evaluate whether they can cause potential risks to the proposed Project.
10.4.4. The hazard distance of the release scenarios in PAFF EIA is reproduced in Table 10.16 below.
Table 10.16 Summary of Hazardous Scenarios in PAFF EIA
|
Release Scenario |
Hazard Distance (m) |
|
Jetty Transfer |
|
|
Fire due to
rupture/ leak of oil products from loaded vessel |
236 |
|
Fire due to
rupture/ leak of loading arm during unloading |
69 |
|
Fire due to
rupture/ leak of jetty equipment |
236 |
|
Fire due to
rupture/ leak of jetty riser |
69 |
|
Fire due to
rupture/ leak of submarine pipeline from jetty to tank farm ESD valve |
148 |
|
Tank Farm Storage |
|
|
Fire due to
discharge from tank vent |
Not significant |
|
Tank head fire
or explosion in tank head space |
Not significant |
|
Multiple tank
head fires |
Not significant |
|
Tank failure
due to overpressure |
Not significant |
|
Explosion in
empty tank (under maintenance) |
Not significant |
|
Bund fire |
18 |
|
Fire outside
bund due to rupture/ leak of pumps, pipework and fittings |
4 |
|
Fire on sea due
to release through drainage |
219 |
|
Fire due to
instantaneous tank wall failure, bottom seam failure |
< 399 |
|
Fire due to instantaneous
tank wall failure, unzipping |
< 399 |
|
Aircraft impact |
< 399 |
|
Fire due to
multiple tank failure |
399 |
|
Tank boilover |
Not significant |
|
Fire due to
release from top of tank due to overfilling |
Not significant |
|
Vapour cloud
explosion or flash fire |
Not significant |
|
Fire due to 10%
instantaneous release from the top of a tank |
39 |
10.4.5. Although the risks associated with LPG release are not considered in the New World Depot EIA, dispersion of LPG vapour cloud formed from the release can pose off-site risk. The failure of LPG cylinders in the Sinopec HK Oil Terminal Chai Wan is considered with reference to the failure cases in the SIL EIA.
Pool fire on seaon sea
where
D = pool diameter (m);
M = release mass (kg);
t = average pool thickness (m); and
ρ = density (797.5kg/m3 for kerosene / aviation fuel).
10.4.8. The average thickness of pool is assumed to be 10mm which is a minimum requirement for flame spread on sea and the thermal impact range for fatality for the releases on sea are taken as equivalent to the pool radius. The largest area covered in a multiple tank rupture is predicted to occur at the start of the release, an initial spill of 7% spread to a depth of 10mm is a conservative assumption for pool fire from multiple tank rupture[23].
10.4.9. Since information of the size of vessel visiting the Sinopec HK Oil Terminal Chai Wan is not available, the size is therefore predicted from the constraints of the jetty. According to the Berthing Guidelines of Marine Department[24], the maximum draft and LOA (overall length) of the Sinopec Chai Wan berth are 5 m and 65 m respectively. The jetty is able to accommodate a typical small port vessel of not more than 2000DWT with draft of 3.9 m and LOA of 75m[25]. Conservatively assume a 5000 DWT loaded vessel could be berthed and a multiple tank rupture could give a pool with 236m diameter. The proposed Project is around 250 m away from the jetty of the Sinopec HK Oil Terminal Chai Wan that hazards induced affecting the proposed Project are not significant. Nevertheless, the release scenario is included in the risk summation using a pool fire model on sea.
Instantaneous tank wall and multiple tank failure
10.4.10. The remaining release scenarios with the worst hazard distances come from the fire due to instantaneous tank wall failure and fire due to multiple tank failure, they are further analysed below.
10.4.11. An instantaneous tank wall failure considers a sudden unzipping or split open of a tank wall, releasing the entire content of a fuel tank. Considering an instantaneous release of 1000 m3 kerosene in the Sinopec HK Oil Terminal Chai Wan, it is possible that part of the liquid be retained in the bund and part of the liquid overtops the bund due to splashing or momentum.
10.4.12. For a pool of fuel contained in a 25 m × 25 m × 2 m bund, a pool fire model using PHAST to simulate a 25 m diameter pool gives an effective distance of 51 m at a radiation level of 4 kW/m2 (fatality is not anticipated for radiation level lower than 4 kW/m2). Since the north boundary of the Project site is approximately 200 m from the bund wall, the hazard from a bund pool fire has no effect to the fatality of the Project construction worker and Project worker.
10.4.13. For the fuel overtopping the bund, the liquid will be directed to the sea, to the grated u-channels and outside the oil terminal since the site is surrounded by two solid walls on two sides. There is a 100mm kerb at the jetty according to the New World Depot EIA that could contain part of the spilled fuel inside the site. The fuel would also flow into the sea through the grated u-channels inside the oil terminal and the remaining fuel would flow outside the oil terminal to the public road.
10.4.14. The quantity of fuel overtopping the bund can be estimated by the correlation by Thyer et al.13:
Where
h = bund wall height (m)
r = distance from the centre of the tank to the bund wall
H = tank liquid height including the tank foundation.
10.4.15. By the configuration of Sinopec HK Oil Terminal Chai Wan, h is 2 m, r is approximately 7.6 m (estimated from site layout) and H is approximately 14.2 m (100% fill level of 12.77 m plus assuming a 1.5 m tank foundation). The quantity of fuel overtopping the bund Q is calculated as 566 m3. (The volume calculated by Thyer et al. was concluded to be 50% overestimated in the PAFF EIA. However given that the configurations of bund and tanks are different between PAFF and Sinopec HK Oil Terminal Chai Wan, 566 m3 calculated from Thyer et al. is directly adopted)
10.4.16. Considering the kerb at the jetty, the area of oil terminal could hold up to about 235 m3 of overtopped fuel ((54 m × 55 m – 25 m × 25 m) × 0.1 m), which is about 41% of the overtopped fuel. The road surface outside the oil terminal has a fall towards the oil terminal that the fuel will be collected by the u-channels. See Appendix 10.6 for the illustration of falls and u-channels of the oil terminal. However, upon an instantaneous release of full inventory inside the fuel tank with a high momentum and large volume, the u-channels can only contain a small amount of spilled fuel. The u-channels are therefore ignored from the calculation and it is therefore conservatively assumed the remaining 59% of spilled fuel will flow outside the oil terminal and no spilled fuel is assumed to be directed to the sea by the u-channels.
10.4.17. By simulating an unbunded instantaneous release of 332 m3 kerosene by PHAST 6.7 gives a maximum pool radius of 145 m, which will cover the road outside the oil terminal. Fire event due to overtopping of bund from instantaneous tank failure and spreading outside the oil terminal is further considered.
10.4.18. Hypothetically it is possible for multiple tanks to release the contents into the bund. The bund wall at Sinopec HK Oil Terminal Chai Wan has a volume of 1250 m3 that is not designed to hold all contents (4500 m3) in the oil terminal. The excess spill will flow into the sea to form a pool either directly from the site or through the u-channels at the site and at the same time to form a pool covering the entire oil terminal (a 55 m × 54 m area as shown in the layout plan in Figure 10.5). A pool fire on the sea does not pose off-site risk to the proposed Project as discussed in previous section. While a pool fire formed within the oil terminal is further assessed.
10.4.19. By considering a confined pool fire within the oil terminal by modelling a 54m diameter pool fire, the result shows an effective distance of 89.8 m at a radiation level of 4 kW/m2. Since the north boundary of the Project site is approximately 190 m from the oil terminal site boundary, the hazard to the Project site is insignificant.
10.4.20. Multiple fuel tank failure due to instantaneous rupture and rupture of tank interconnection pipe are possible, In an event of release from multiple fuel tank, the fuel could overtop the bund and at the same time overflow to the public road as considered in instantaneous single tank failure, which would involve a larger amount of fuel and cover a larger extent of spillage area. The scenarios are therefore further considered in Section 10.5.12.
LPG Cylinder Failure
10.4.21. LPG cylinders of sizes 10.5 kg, 16 kg and 45 kg are stored in the oil terminal and the maximum storage capacity in the terminal is 12 tonnes according to the New World Depot EIA[26].
10.4.22. Cold catastrophic rupture leading to instantaneous release and partial failure leading to continuous release of LPG are deemed to be possible, in which 45 kg is selected to represent the worst case.
10.4.23. Multiple BLEVE of LPG cylinders is also probable if the LPG cylinders are engulfed by pool fire outside the bund wall resulted from a multiple oil tank failure. LPG cylinders are equipped with pressure relief valves to release internal pressure during emergency conditions that not all cylinders will result in BLEVE under fire engulfment. Take a conservative assumption that 10% of total cylinders result in BLEVE18, the consequence is shown below.
10.4.24. The failure cases and the corresponding consequences are summarised in Table 10.17. It shows that no release scenario will cause potential hazard to the proposed Project due to a large separation between the LPG cylinder storage shed and the Project site.
Table 10.17 Failure cases for LPG Cylinder in Sinopec HK Oil Terminal Chai Wan
|
LPG Cylinder Storage |
Release Quantity |
Outcome |
Hazard Distance |
|
Cold catastrophic rupture of LPG cylinder |
45 kg |
Fireball |
10.4 m (Fireball radius) |
|
|
|
Flash Fire |
36.2 m (100% LFL) |
|
|
|
VCE |
17.4 m (0.3 bar) |
|
Partial failure (1mm leak) of LPG cylinder |
45 kg |
Jet Fire |
28.1 m (4 kW/m2) |
|
|
|
Flash Fire |
15.9 m (100% LFL) |
|
|
|
BLEVE |
10.4 m (Fireball radius) |
|
Multiple BLEVE of LPG cylinders |
1,200 kg |
Fireball |
32.5 m (Fireball radius) |
Fuel Oil Filling Failure
10.4.25. According to the New World Depot EIA, two stations are used for distribution of fuel oil within the Sinopec HK Oil Terminal Chai Wan. One station is for the bulk filling of road tanker which is mainly used for Automotive Diesel Oil (ADO) and Industrial Diesel Oil (IDO); and the other one is for the filling of package products (drums) which is mainly used for IDO and kerosene. And as per the observation in the site survey, the two stations still exist though not in operation.
10.4.26. Fuel oils are pumped from the storage tanks to the filling stations. A 100 mm articulated arm and a 25 mm loading hose are used in the bulk filling station and drum filling station respectively. Following the approach of the New World Depot EIA, the road tanker is assumed to have a capacity of 16,000 litres and the pool fires are enclosed by the drainage channels (15.2 m × 6.2 m) and the site area (54 m × 41 m) of the oil terminal.
10.4.27. Dye marker is stored in a storage tank outside the bund. By making reference to the New World Depot EIA, a pool fire could be resulted due to failure of the dye marker tank. The volume of dye marker tank is 3.4 m3 as per a site survey conducted by the Consultant. By approximation of tank diameter from the site layout plan and the site survey conducted by the Consultant, the size of the dye marker tank is estimated to be 1.2 m in diameter and 3m high (approximate 3.4m3). An unconfined pool fire is considered as a conservative approach due to insufficient information of the bund size obtained from the operator, site layout plan and site visits.
10.4.28. The failure cases and the corresponding consequences simulated using PHAST are summarized in Table 10.18. As a conservative approach, kerosene is used for all scenarios instead of ADO and IDO. The failure cases listed are further considered in the frequency analysis in Section 10.5.
Table 10.18 Failure cases for Fuel Oil Filling in Sinopec HK Oil Terminal Chai Wan
|
Fuel Oil Filling Operation |
Release Quantity |
Outcome |
Hazard Distance |
|
Cold catastrophic rupture of fuel oil road tanker |
16,000 L |
Pool Fire (confined by site area) |
84.3 m (4 kW/m2) |
|
Partial failure (25mm leak) of fuel oil road tanker |
16,000 L |
Pool Fire (confined by site area) |
71.3 m (4 kW/m2) |
|
Rupture of loading arm during loading to fuel oil road tanker |
1,000 m3 |
Pool Fire (confined by site area) |
84.3 m (4 kW/m2) |
|
Rupture of flexible hose during loading to drum truck |
1,000 m3 |
Pool Fire (confined by drainage channels) |
126.9 m (4 kW/m2) |
|
Full bore rupture of pump connecting pipeline in pump farm |
1,000 m3 |
Pool Fire (confined by site area) |
85.4 m (4 kW/m2) |
|
Catastrophic rupture of dye marker tank |
3.4 m3 |
Pool Fire within site |
56.3 m (4 kW/m2) |
Jetty Unloading Failure
10.4.29. The pool fire at the jetty of the Sinopec HK Oil Terminal Chai Wan was not considered in the risk summation of in the previous New World Depot EIA since it concluded a fire in bund and pool fire with overtopping the bund were more onerous. A pool fire at the jetty is further assessed in this study.
10.4.30. Rupture of the loading hose at the jetty could be caused by a misconnection and by the marine vessel steered away from the jetty during unloading, which could result in a pool fire on the sea if ignited.
10.4.31. In a case of loading hose rupture, operators in the field will spot the failure and initiate a rapid isolation. Should a rapid isolation fails, it is assumed that the release will be isolated by an emergency shut off intervention (such as pump cut-off) within 10 minutes[27] [28].
10.4.32. Since there is no information from Sinopec (Hong Kong) Ltd or from the New World Depot EIA about the operation data of the jetty, it is assumed it takes 2 hours on average to unload a 1,000 m3 storage tank, resulted in a pumping rate of 500 m3/hr (this is conservative as the pumping rate in PAFF is 3,500 m3/hr for serving a 35,000 m3 storage tank but the size of storage tank in Sinopec HK Oil Terminal Chai Wan is 1/35 of that of PAFF). For a release of 10 minutes in the event of rapid isolation fails, the volume of release is 83 m3 (500 m3 × 10 / 60). For a 10 mm thickness of pool, this is equivalent to a pool diameter of 103m. The effect distance of pool fire is 141.6 m at a radiation level of 4 kW/m2.
10.4.33. There is also a potential rupture of the jetty fixed pipeline as a result of spontaneous rupture of the pipeline and striking by a marine vessel. Considering the same pumping rate and isolation mechanism as that of loading hose, the consequence is a pool fire with an effect distance of 141.6 m at a 4 kW/m3 radiation ellipse.
Smoke Generation from Kerosene Fire
10.4.34. Combustion gases from burning hydrocarbons consist of mainly nitrogen, carbon monoxide and carbon dioxide[29]. Engulfed by smoke from combustion of hydrocarbon can cause incapability to a person due to combined effect of CO2 and CO.
10.4.35. In this study, smoke ingression to the proposed Project will be the major concern.
10.4.37. PHAST 6.7 is used to model smoke dispersion and the results are summarised in Table 10.19 below. The hazard distances at ground level as a result of bund fire in different wind speeds are shown, the worst hazard distances of single tank and multiple tank failure are 24.7 m and 38.7 m at a concentration of 17% respectively under a high wind speed (7.5 m/s) condition. It is obvious that at such distance the smoke will not cause any fatality to the Project site. Smoke ingression is deemed to be insignificant and will not be considered further.
Table 10.19 Summary of Smoke Dispersion Results of Oil Terminal Bund Fire
|
Parameter |
Single Tank Failure |
Multiple Tank Failure |
||
|
Fuel pool diameter |
25 m |
54 m |
||
|
Burning rate |
0.06 kg/m2s |
0.06 kg/m2s |
||
|
Smoke production rate |
471.2 kg/s |
2198.6 kg/s |
||
|
Dispersion distance @ ground level & Conc. 17% |
20.5 m |
3.0B |
35.9 m |
3.0B |
|
16.5 m |
2.0D |
34.9 m |
2.0D |
|
|
18.1 m |
4.5D |
35.6 m |
4.5D |
|
|
24.7 m |
7.0D |
38.7 m |
7.0D |
|
|
20.4 m |
3.0E |
33.2 m |
3.0E |
|
|
18.5 m |
1.5F |
29.7 m |
1.5F |
|
10.4.38. Carbon monoxide component in smoke is known to cause the majority of deaths in fire and an exposure to carbon monoxide in a concentration of 30,000 ppm can cause immediate death[30]. For a liquid fire, the typical carbon monoxide concentration in smoke is 3.1%30. Table 10.20 below presents the hazard distance of carbon monoxide modelled using PHAST 6.7.
Table 10.20 Summary of Carbon Monoxide Effects of Oil Terminal Bund Fire
|
Parameter |
Single Tank Failure |
Multiple Tank Failure |
||
|
Fuel pool diameter |
25 m |
54 m |
||
|
Smoke production rate |
471.2 kg/s |
2,198.6 kg/s |
||
|
Carbon monoxide production rate |
14.6 kg/s |
68.2 kg/s |
||
|
Dispersion distance @ ground level & Conc. 30000ppm |
5.2 m |
3.0B |
8.2 m |
3.0B |
|
4.3 m |
2.0D |
7.0 m |
2.0D |
|
|
5.7 m |
4.5D |
8.8 m |
4.5D |
|
|
7.8 m |
7.0D |
12.0 m |
7.0D |
|
|
4.7 m |
3.0E |
6.7 m |
3.0E |
|
|
0.3 m |
1.5F |
14.7 m |
1.5F |
|
10.4.39. Taking into account the consequences induced by the hazardous scenarios of the Sinopec HK Oil Terminal Chai Wan, a 200m radius study area is adopted.
10.4.40. The properties of LPG is discussed in Section 10.4.1, the gas cloud formed due to release of LPG is flammable upon ignition and therefore is possible to induce off-site risk. The hazardous scenarios of petrol-cum-LPG filling station will follow that of Citybus Depot EIA[31] and Kai Tak Development EIA[32]. A study area of 225 m radius from the petrol-cum-LPG filling station will be adopted in this assessment according to the Table 10.21 below.
10.4.41. Table 10.21 below summarizes the hazardous events related to the storage and use of petrol / LPG in the filling station.
Table 10.21 Summary of Hazardous Scenarios for Petrol-cum-LPG Filling Station
|
Hazardous Scenarios |
|
LPG
Filling Station |
|
Cold catastrophic failure of LPG storage vessel |
|
Cold partial failure of LPG storage vessel |
|
Cold catastrophic failure of LPG road tanker |
|
Cold partial failure of LPG road tanker |
|
Guillotine failure of liquid-inlet pipework (rupture) |
|
Partial failure of liquid-inlet pipework (leak) |
|
Guillotine failure of liquid supply line to dispenser (rupture) |
|
Partial failure of liquid supply line to dispenser (leak) |
|
Failure of dispenser |
|
Guillotine failure of flexible hose to vessel (rupture) |
|
Partial failure of flexible hose to vessel (leak) |
|
Guillotine failure of flexible hose to vehicle (rupture) |
|
Submersible Pump Flange Leak |
|
BLEVE of LPG road tanker (fire escalation) |
|
Petrol
Filling Station |
|
Rupture / leak of petrol road tanker |
|
Rupture / leak of petrol storage tank |
|
Rupture / leak of delivery pipe |
|
Leak of dispenser |
10.4.42. Petrol vapour is heavier than air that it tends to sink through normal atmosphere. Since all petrol delivery pipeline and storage tanks in a Petrol-cum-LPG filling station are underground, their content is unlikely to contact with an ignition source. And a vent pipe is equipped with a flame arrestor that the released vapour is discharged at a high point.
10.4.43. Only a limited quantity of petrol in the hose of the petrol dispenser will be spilled in the event of loss of containment as petrol vapour is heavier than air. For a petrol dispenser with a typical size of 1.28 m × 0.63 m × 2.175 m = 1.75 m3 [33] and an oil interceptor with a typical size of 0.75 m × 0.75 m × 1.2 m × 3 = 2.0 m3 [34], the spill will be collected by the drainage system and the oil interceptor.
10.4.44. According to previous experience, the most hazardous scenarios are the failure of LPG storage vessel and road tanker, the hazard distances of these two scenarios are shown in Table 10.22 below.
Table 10.22 Consequences for Release Scenarios at Petrol-cum-LPG Filing Station
|
Hazardous Scenarios |
Release Quantity |
Outcome |
Hazard Distance |
|
Cold catastrophic failure of LPG storage
vessel |
10200kg |
Flash Fire |
224.9m (100% LFL) |
|
|
|
VCE |
105.8m (0.3 bar) |
|
Cold catastrophic failure of LPG road tanker |
9000kg |
Flash Fire |
216.8 (100% LFL) |
|
|
|
VCE |
100.0 (0.3 bar) |
|
|
|
Jet Fire |
44.5m (4kW/m2) |
|
|
|
BLEVE |
62.5m (Fireball radius) |
10.4.45. Since the Project site is within the hazardous distances resulted from the releases of petrol-cum-LPG filling stations (170 m from Sinopec and 20 m from ExxonMobil), a quantitative risk assessment will be conducted in this Hazard to Life study to evaluate the risks from both Sinopec and ExxonMobil Petrol-cum-LPG Filling Station.
10.4.46. Domestic LPG cylinders are stored inside the LPG wagons at the parking site, the general size of a domestic use LPG cylinder is 10.5 kg. The properties of LPG are discussed in Section 10.4.1. Instantaneous release due to cylinder rupture and continuous release due to cylinder leakage are considered in previous EIA reports that assess the effects of 50 kg LPG cylinders inside a LPG depot[35],[36]. Single cylinder rupture and leak with a hole size of 1mm were modelled to identify the hazard distance of each scenario.
10.4.47. BLEVE of multiple LPG cylinders in the wagon is possible if they are engulfed by the pool fire of a petrol tanker. Assume a pool fire of a road tanker could affect two LPG wagons involving 500 number of cylinders that 10% of the cylinders would undergo BLEVE.
10.4.48. In view of possible parking of kerosene and diesel road tanker in the LPG wagon parking site, pool fires due to the ignition of kerosene (which is more volatile than diesel) from the rupture and leakage of a 9 tonnes road tanker is also modelled to evaluate the effects.
10.4.49. The hazard distances of the release scenarios are shown in Table 10.23. The worst hazardous distance, which is a pool fire due to rupture of a petrol road tanker, can only reach 88.6 m from the release point. The study area is set to 150m radius from the LPG wagon parking site.
10.4.50. Since the LPG wagon parking site is around 300 m away from the Project site, it will therefore not pose risks to the proposed Project.
Table 10.23 Consequences for Release Scenarios at LPG Wagon Parking Site
|
Hazardous Scenarios |
Release Quantity |
Outcome |
Hazard Distance |
|
Cylinder rupture |
10.5kg |
Fireball |
6.5 m (Fireball radius) |
|
|
|
Flash
fire |
22.0 m (100% LFL) |
|
|
|
VCE |
10.7 m (0.3 bar) |
|
Cylinder leak (Hole size: 1 mm) |
10.5kg |
Jet
fire |
27.4 m (4kW/m2) |
|
|
|
Flash
fire |
15.9 m (100% LFL) |
|
|
|
BLEVE |
6.5 m (Fireball radius) |
|
Multiple
BLEVE of cylinders |
525kg |
BLEVE |
24.8 m (Fireball radius) |
|
Kerosene
road tanker rupture |
9,000kg |
Pool Fire |
88.6 m (4kW/m2) |
|
Kerosene
road tanker leak (25 mm) |
9,000kg |
Pool Fire |
71.3 m (4kW/m2) |
|
|
|
Jet Fire |
2.8 m (4kW/m2) |
10.4.51. Underground diesel storage tanks are installed in both NWFB and Citybus Depots where the tanks are replenished by 16,000L road tankers. Spillages from road tanker, storage tank, delivery pipe and dispenser are considered possible.
10.4.52. The spilled diesel from underground tank is however unlikely to come in contact with ignition source since it is buried. And diesel vapour does not easily disperse in still air. Release from storage tank will not cause off-site risk.
10.4.53. In the event that diesel is released from a dispenser during refuelling, the spilled quantity will limited to the volume in the hose. Oil interceptors are installed according to the layout plan received from the bus depot operators, which will be able to confine the spillage within the site.
10.4.54. Diesel is not easily ignited as mentioned in Section 10.4.1 but ignition of spilled diesel during unloading is deemed possible. It is assumed a spill from a diesel tanker will lead to a pool fire[37]. The hazard distance of each scenario is summarised in Table 10.24 below:
Table 10.24 Summary of Hazardous Scenarios for Diesel Storage in Bus Depot
|
Hazardous Scenarios |
Release Quantity |
Outcome |
Hazard Distance |
|
Tanker rupture |
16,000 L |
Pool Fire (Radius 31.9 m) |
98.0 m (4 kW/m2) |
|
Tanker medium liquid leak (Hole size: 25 mm) |
16,000 L |
Pool Fire (Radius 20.2 m) |
71.7 m (4 kW/m2) |
10.4.55. Since the Project site is within the hazard distance of the release scenarios of diesel road tanker, the scenarios will be taken into account in the risk assessment.
10.4.56. As mentioned in Section 10.4.34 that burning of hydrocarbons could cause smoke ingression to the proposed Project. Using the same approach mentioned in Section 10.4.36, the largest distance the smoke can disperse is 41.7 m at ground level.
Table 10.25 Summary of Smoke Dispersion Results of Diesel Tanker Release
|
Parameter |
Tanker Rupture |
Tanker Medium Leak |
||
|
Fuel pool diameter |
31.9 m |
20.2 m |
||
|
Burning rate |
0.06 kg/m2s |
0.06 kg/m2s |
||
|
Smoke production rate |
3069.0 kg/s |
1225.8 kg/s |
||
|
Dispersion distance @ ground level & Conc. 17% |
38.8 m |
3.0 B |
28.6 m |
3.0 B |
|
39.1 m |
2.0 D |
24.7 m |
2.0 D |
|
|
40.7 m |
4.5 D |
25.0 m |
4.5 D |
|
|
41.7 m |
7.0 D |
32.4 m |
7.0 D |
|
|
37.8 m |
3.0 E |
27.6 m |
3.0 E |
|
|
36.2 m |
1.5 F |
24.8 m |
1.5 F |
|
10.4.57. The Citybus Depot is approximately 51m from the Project site. At such distance, the plume height is 29 m as calculated from [L x tan(90o – tilt angle)] where L is the horizontal distance from the fire site boundary and tilt angle taken as 60o to represent the worst case1. Which the 6-storey high (40 m) proposed Project can be affected by the smoke plume. And since the NWFB Depot is adjacent to the Project site, the smoke plume can affect the proposed Project. Smoke plume from the bus depots will be further considered in the study.
10.4.59. The hazard distances of smoke in terms of carbon monoxide dispersion are summarized in Table 10.26 below.
Table 10.26 Summary of Carbon Monoxide Effects of Diesel Tanker Release
|
Parameter |
Tanker Rupture |
Tanker Medium Leak |
||
|
Fuel pool diameter |
31.9 m |
20.2 m |
||
|
Smoke production rate |
3069.0 kg/s |
1225.8 kg/s |
||
|
Carbon monoxide production rate |
95.1 kg/s |
38.0 kg/s |
||
|
Dispersion distance @ ground level & 30000ppm |
8.8 m |
3.0 B |
6.1 m |
3.0 B |
|
9.0 m |
2.0 D |
7.7 m |
2.0 D |
|
|
7.8 m |
4.5 D |
6.3 m |
4.5 D |
|
|
11.0 m |
7.0 D |
9.0 m |
7.0 D |
|
|
8.4 m |
3.0 E |
6.0 m |
3.0 E |
|
|
21.3 m |
1.5 F |
No hazard |
1.5 F |
|
10.4.60. As mentioned in Section 10.2.28, small amount of Category 2, 3, 5 and 10 DGs were licensed in the NWFB Depot.
10.4.63. Lacquer thinner, Acetone and Paint are classified as Category 5 DG for their ability to give off inflammable vapour. Lacquer thinner, acetone and paint fall into Class 1, Division 1 of Category 5 DG, which have a flash point below 23 oC. If released, pool fire could occur from ignition of the vapour above the pool and the heat in turn vaporizes the liquid beneath to support the fire. Assume all Category 5 substances in the DG store with a quantity of 1171 L were released to form an uncontained pool, the hazardous distance of the pool fire is 49.6 m (4kW/m2) and flash fire is 44.8 m (LFL).
10.4.64. Methyl ethyl ketone peroxide is classified as Category 10 DG. It is combustible and can cause skin, throat and lung irritation if inhaled. Methyl ethyl ketone peroxide has a flash point in between 52oC and 93oC[38]. The hazardous distances from an uncontained pool fire and flash fire from the rupture of 171L of methyl ethyl ketone peroxide vessel are 26.2m (4kW/m2) and 2.57m (LFL) respectively.
10.4.65. Given that there is a lack of ignition source within a DG store and with a firefighting system, the fire arose from the DGs would be confined inside the building even if ignited. The Category 5 and Category 10 substances are therefore believed not to pose off-site risks and are not further considered in the analysis.
10.4.66. The facilities in the existing vehicle depots of the HKPF, EMSD and FEHD and the laboratory of GL will be relocated to the future Project once commissioned. HAZID workshops have been carried out with participants from HKPF, EMSD and GL to identify potential hazards in the existing facilities. Given that the future Project is of similar nature as the existing facilities, the hazards in the future complex and depot can be deduced.
10.4.67. The HAZID workshop involved interviewing with the operators of the existing facilities to learn the DG handling procedures, reviewing the existing safeguards and emergency response process and carrying out a brainstorming exercise to identify all hazards due to the operation of the proposed Project. The hazards were further evaluated and rated with risk ranking judging from their consequences and possibilities.
10.4.68. Details of the hazards identified in the workshops are provided in Appendix 10.3. The following table gives a short summary of the conclusions based on the outcome of the HAZID workshops:
Table 10.27 Summary of HAZID Workshops
|
Facility |
Main Functions |
Conclusion of Hazards |
|
HKPF Hong Kong Island PVP&EC |
To examine defective vehicles identified by the HKPF and damaged vehicles from car accidents. |
No hazard has been identified that could cause off-site risk. Hazards will be on-site only. |
|
EMSD
Depot |
To provide mechanical repairing and general maintenance of
government vehicles. |
No hazard has been identified that could cause off-site risk. Hazards will be on-site only. |
|
GL
Specialist Laboratory |
To provide chemical testing services for other government departments. |
No hazard has been identified that could cause off-site risk. Hazards will be on-site only. |
10.4.69. Considering the proposed Project of FEHD only serves as a car washing facility, the proposed Project Team did not anticipate any activities that could cause off-site risk to the public.
10.4.70. The hazardous scenario identified in HAZID workshop is further evaluated in Appendix 10.3A. Hazardous scenarios such as fire of flammable chemical in government laboratory and fire of petrol in the proposed Project were modelled to ensure the hazardous effect does not affect offsite population.
10.4.71.
Waste oil and lubricant having a high flash point of around 177oC[39]
are unlikely to be ignited. Typical automotive battery
contains 20–44 wt.% of sulphuric acid as electrolyte[40],
the hazard of sulphuric acid is however concluded to be localized if released
as discussed in Section 10.4.62. Moreover,
automotive battery is stored in a bunded DG store that is separated from the
storage of waste oil / lubricant, sulphuric acid will therefore be contained
within the bund when released and will not be heated to cause any off-site
risk. For other DGs including Cat 2, Cat 3 and Cat 5 of Table 10.6Table 10.6 that
are planned to be stored in the proposed Project, the quantities are less than
that stored in the NWFB Depot. The hazards induced will be bounded within the
site similar to those described in Section
10.4.61 to Section 10.4.64,
offsite risk is therefore not anticipated.
10.4.72. An incident review is also performed on WiseNews[41] in between 2000 and 2015 to identify worldwide news reporting incidents similar to vehicle depots and laboratories.
10.4.73. The keywords “vehicle maintenance”(汽車維修), “fire”(火警) and “death“(死亡) are searched, there were fire reported in small scale vehicle workshops but fatalities were on-site only.
10.4.74. The keywords “laboratory” (實驗室), “leakage”(洩漏) and “death“(死亡) are searched in the database. Three representable incidents were chlorine gas leakage from a university of Nanjing in 2001, Boron Tribromide gas leakage from a university of South Korea in 2013 and release of Chloroacetyl chloride from a university of Hong Kong in 2014, both involved evacuation of on-site staff and students but no off-site fatality was resulted. Those laboratories identified are chemical testing in nature and the Government Laboratory in this Study is for food and environmental testing purpose. Given that no acute fatality toxic chemicals are used in the Government Laboratory, similar incident is not anticipated in the proposed Project.
10.4.75. The keywords “laboratory” (實驗室), “fire”(火警) and “death“(死亡) are searched in the database. There were laboratory fires reported including ignition of chemicals in a university of Hong Kong in 2002, radioactive substance release from a medical laboratory in 2008 and machine overheat in a university of Taiwan in 2009 but none of the cases resulted in fatality.
10.4.76. Since all facilities are concluded not to pose off-site risk, no hazard scenarios from the proposed Project are considered in the risk modelling.
Natural Hazards and External Events
10.4.77. Loss of containment may occur due to external events and the consequences could be catastrophic failure or leak. The related natural hazards and external events are listed as follows, they will be further analysed in Section 10.5:
· Earthquake;
· Landslide;
· Lightning strike;
· Severe environmental events;
· Aircraft crash;
· Car crash;
· Dropped object;
· Subsidence; and
· External fire
10.4.78. It is noted that some of the consequences induced by the hazardous scenarios do not cause offsite risk to the Project site, they are included in the risk summation to take into account as the background risk as required by the Authority.
10.5.1. Based on the hazard identification in Section 10.4 above, the release scenarios that will be considered in this assessment are summarised in Table 10.28.
Table 10.28 Release Scenarios to Be Considered
|
Potential Hazardous Source |
Equipment |
Failure Type |
Release Type |
Event Outcome |
|
Sinopec/ ExxonMobil Petrol-cum-LPG Filling Station |
LPG storage vessel |
Catastrophic failure |
Instantaneous |
Flash fire, VCE |
|
Partial failure (leak) |
Continuous |
Flash fire, VCE, jet fire |
||
|
LPG road tanker |
Catastrophic failure |
Instantaneous |
Flash fire, VCE, fireball, BLEVE |
|
|
Partial failure (leak) |
Continuous |
Flash fire, VCE, jet fire |
||
|
Liquid-inlet pipework |
Guillotine failure |
Continuous |
Flash fire, VCE, jet fire |
|
|
Leak |
Continuous |
Flash fire, jet fire |
||
|
Liquid supply line to dispenser |
Guillotine failure |
Continuous |
Flash fire, VCE, jet fire |
|
|
Leak |
Continuous |
Flash fire, jet fire |
||
|
Dispenser |
Guillotine failure |
Continuous |
Flash fire, jet fire |
|
|
Flexible hose to vessel |
Guillotine failure |
Continuous |
Flash fire, VCE, jet fire |
|
|
Leak |
Continuous |
Flash fire, jet fire |
||
|
Flexible hose to vehicle |
Guillotine failure |
Continuous |
Flash fire, jet fire |
|
|
Submersible pump flange |
Leak |
Continuous |
Flash fire, VCE, jet fire |
|
|
Petrol road tanker (with knock-on effect) |
Catastrophic failure |
Instantaneous |
Fireball, VCE, flash fire |
|
|
Partial failure (leak) |
Continuous |
Pool fire followed by BLEVE, VCE, flash fire |
||
|
Sinopec HK Oil Terminal Chai Wan |
Fuel oil tank (Single tank) |
Catastrophic failure |
Instantaneous |
Bund pool fire, overtop pool fire, sea pool fire, site pool fire |
|
LPG cylinder |
Catastrophic failure |
Instantaneous |
Flash fire, fireball |
|
|
Partial failure (leak) |
Continuous |
Flash fire, jet fire, BLEVE |
||
|
Marine vessel fuel tank |
Catastrophic failure |
Instantaneous |
Sea pool fire |
|
|
Fuel oil road tanker |
Catastrophic failure |
Instantaneous |
Site pool fire |
|
|
Partial failure (leak) |
Continuous |
Site pool fire |
||
|
Loading arm to fuel oil road tanker |
Guillotine failure |
Continuous |
Site pool fire |
|
|
Flexible hose to drum truck |
Guillotine failure |
Continuous |
Site pool fire |
|
|
Pump connecting pipeline |
Guillotine failure |
Continuous |
Site pool fire |
|
|
Loading hose at jetty for unloading |
Guillotine failure |
Continuous |
Sea pool fire |
|
|
Fixed pipeline at jetty |
Guillotine failure |
Continuous |
Sea pool fire |
|
|
Dye marker tank |
Catastrophic failure |
Instantaneous |
Site pool fire |
|
|
LPG Wagon Parking Site |
LPG cylinder |
Catastrophic failure |
Instantaneous |
Flash fire, fireball |
|
Partial failure (leak) |
Continuous |
Flash fire, jet fire, BLEVE |
||
|
Kerosene tanker |
Catastrophic failure |
Instantaneous |
Pool fire |
|
|
Partial failure (leak) |
Continuous |
Pool fire followed by BLEVE |
||
|
Bus Depot |
Diesel tanker |
Catastrophic failure |
Instantaneous |
Pool Fire |
|
Partial failure |
Continuous |
Pool Fire |
10.5.2. The frequency for each of the identified hazardous scenario is estimated using the best available failure data or historical accident data. The frequencies documented in the relevant sources are reviewed and justified, to reflect the specific operation and risk reduction practices evident at hazardous facilities.
10.5.3. When the historic data on failure frequency is not fully available, failure frequencies of similar installations or events are adopted with suitable modifications. If there is no failure rate data for similar installations or events, the historical accident databases are further reviewed to find out the number of incidents happened in similar installations.
10.5.4. Fault Tree Analysis is adopted to estimate event frequency by considering initial events (e.g. failure probability of a valve) and site specific operation data (e.g. number of LPG tanker visiting the site) through a combination of simple “AND” and “OR” logic gates.
10.5.5. For LPG releases, the immediate ignition probabilities are determined in accordance with Cox, Lees and Ang[42], as listed in Table 10.29 below. Immediate ignition probability of 0.3 will be adopted for rupture cases and 0.07 will be adopted for leakage cases. A delay ignition probability of 0.5 is assigned and given a fairly open nature of surrounding, an explosion probability of 0.2 is assumed[43],[44]. The estimates by Cox are for general areas where density of ignition source is lower than industrial area, which is appropriate to this assessment.
Table 10.29 Ignition Probabilities from Cox, Less and Ang
|
Release Rate |
Ignition Probability |
Release Rate |
|
Gas Release |
Liquid Release |
|
|
Minor (<1 kg/s) |
0.01 |
0.01 |
|
Major (1-50 kg/s) |
0.07 |
0.03 |
|
Massive (>50 kg/s) |
0.3 |
0.08 |
10.5.6. For diesel oil / kerosene release on land, an ignition probability of 0.004 will be adopted in this assessment following the “cautious best estimation” of the PAFF EIA[45]. The ignition probability was assigned to the bund area within PAFF where there was limited vehicle access, limited hot work and no significant source of heating, which are applicable to the Sinopec HK Oil Terminal Chai Wan and bus depots with similar environment.
10.5.7. For diesel oil / kerosene release on sea, same approach as the PAFF EIA will be followed by applying a factor of 10 to the liquid ignition probabilities of Cox, Lee and Ang due to a lower flammability and more difficult to ignite on sea.
10.5.8.
An immediate ignition of 0.005
and a delay ignition of 0.005 are adopted for the failure events of LPG
cylinder by assuming even split18 between
the small leakleaks42 in Table
10.29.
Sinopec HK Oil Terminal Chai Wan
10.5.9. The rupture of fuel tanks in a loaded marine vessel considers striking the jetty while it is berthing. Although the marine vessels visiting the jetty of Sinopec HK Oil Terminal Chai Wan are smaller in size and less frequent when compared to the PAFF jetty, the frequency of striking is 8 × 10-6 per movement of marine vessel13 is directly adopted from PAFF EIA considering a similar nature of operation. Similarly, the spill probability of 0.015 is also adopted as not all strikes will lead to spill from the marine vessel’s fuel tank. The storage tanks are assumed to be full 40% of the time that the number of visits by the marine vessels is taken as 365 × 40% = 146 per year. The frequency of spill from the fuel tank of marine vessel is therefore 1.75 × 10-5 per year.
10.5.10. It is mentioned in Section 10.4 that spills from the rupture of storage tanks and the overtopping of bund from instantaneous failure of storage tank are to be further considered.
10.5.11. The instantaneous release of the complete inventory from an atmospheric tank directly to the atmosphere is 5 × 10-6 per year is adopted according to the TNO purple book[46]. PAFF EIA estimated a cautious best instantaneous release frequency of 5 × 10-9 per tank year. The estimation has however considered that the tanks were new and designed and test to modern standard. For the storage tanks in the Sinopec HK Oil Terminal Chai Wan and an assumption of the tanks are full 40% of the time, the frequency of overtopping the bund is 5 × 5 × 10-6 × 0.4 = 1 × 10-5 per year.
10.5.12. For off-site spreading of fuel due to overtopping of the bund from multiple tank rupture, the frequency of simultaneous release of the tank’s full content is assessed. An instantaneous release of the complete inventory from one single containment atmospheric tank directly to the atmosphere is 5 × 10-6/year46. For a release from 2 tanks would give a failure frequency of (5 × 10-6)2 = 2.5 × 10-11/year. For a large release from 3 tanks, the failure frequency would reduce to 1.25 × 10-16/year. The probability of an instantaneous release of more than one tank of the full inventory is much below 1 × 10-9/year.
10.5.13. Making reference to the failure scenario of the PAFF EIA. Considering a rupture of the interconnection pipe between two automotive diesel oil tanks or two industrial diesel oil tanks while the valves are fail to close, which would cause a release of the full inventory from two tanks simultaneously. Since official information is not available, take an assumption that the diameter of the interconnection pipe is less than 75mm and estimate the length of the interconnection pipe to be 20m from the layout plan (Figure 10.5). The frequency of a full bore rupture of an aboveground <75mm pipeline is 1 × 10-6/m/year46 and a valve failure is 3 × 10-4/year42. The failure frequency of the release is 1 × 10-6 × 40 × 3 × 10-4 × 3 × 10-4 = 3.6 × 10-12/year. Release from multiple tanks through the pipeline is below 1 × 10-9/year.
10.5.14. The failure frequencies of rupture and leak of a LPG cylinder are 1 × 10-6 per cylinder year46 and 2.6 × 10-6 per cylinder year respectively18. Assume the size of LPG cylinders is evenly distributed between 10.5kg, 16kg and 45kg, that there are around 503 cylinders with a maximum storage capacity of 12 tonnes LPG cylinder. The frequencies of LPG cylinder rupture and LPG cylinder leak are estimated at 5.03 × 10-4 per year and 1.31 × 10-3 per year respectively.
10.5.15. For escalation of LPG cylinder BLEVE, a flame impingement probability of 0.5 is adopted following the South Island Land EIA.
10.5.16. Firefighting facilities are available in the oil terminal according to the New World Depot EIA. A foam spraying network is operated manually for oil tanks, a failure frequency of 0.015 per demand is assumed.
10.5.17. The estimated base event frequencies of hazardous events are summaries in Table 10.30 and the event trees are shown in Figure 10.7 to Figure 10.9.
Table 10.30 Base Failure Frequencies of Hazardous Events for LPG Cylinder Releases (Sinopec HK Oil Terminal Chai Wan)
|
Hazardous Event |
Failure Frequency (per year) |
|
Cold catastrophic failure of marine vessel fuel tank due to striking |
5.03E-04 |
|
Instantaneous single fuel tank failure |
1.00E-05 |
|
Cold catastrophic failure of LPG cylinder |
1.50E-06 |
|
Cold partial failure of LPG cylinder |
1.31E-03 |
|
BLEVE of LPG Cylinder |
4.90E-08 |
10.5.18. The frequencies of rupture and leak (25mm) of a road tanker are 2.0 × 10-6 per tanker year and 5.0 × 10-6 per tanker year respectively[47]. While the guillotine failure frequency of a loading hose is 9.0 × 10-8 per hour47 and frequency of full bore rupture of a pump connecting pipeline is 1.0 × 10-4 per year[48].
10.5.19. It is assumed that a transfer operation at the jetty takes 2 hours on average. There are 146 visits as described in Section 10.5.9. The frequencies of hose misconnection, disconnection and vessel steered away are 3 × 10-5, 2 × 10-6 and 4 × 10-6 per operation respectively47.
10.5.20. For the rupture of the fixed pipeline at the jetty, the frequency of full bore rupture of a pipeline greater than 150mm is 1 × 10-7 per meter-year[49] (assume the jetty pipeline is 250mm). The length of the fixed pipeline from the storage tank to the jetty is approximately 100m estimated from the site layout plan. The frequency of striking and probability of impact enough to cause a pipeline rupture are 8 × 10-6 per movement of marine vessel and 0.015 respectively as described in Section 10.5.9.
10.5.21. The instantaneous release of the complete inventory from an atmospheric tank of 5 × 10-6 per year is adopted for the dye marker tank. The number of filling is also assumed to be 146 per year (0.4 × 365).
10.5.22. For the rocketing of dye marker tank, considering a pool fire due to overtopping of the bund that engulfs the dye marker tank; frequency of overtopping the bund is 1 × 10-5 per year (Section 10.5.11) with an ignition probability of 0.004. Following that, the fire service fails to put off the fire, a failure probability of 0.5 for an engulfment fire is assumed51. The pressure relief valve of the tank then fails to depressurize the tank with a failure probability of 0.02 according to Lee’s Loss Prevention in the Process Industries[50] (an open vent is installed on the dye marker tank as confirmed in a site survey that will have a much lower failure probability than a relief valve, 0.02 is adopted as a conservative approach). Finally, the bottom seam of the tank must then be fail instead of the top seam, a probability of 0.5 is taken. Therefore, an unbunded pool fire escalating to dye marker tank rocketing is estimated to be: 1 × 10-5 × 0.004 × 0.5 × 0.02 × 0.5 = 2.0 × 10-10 per year. The frequency calculated is therefore below 1 × 10-9 per year and is not considered in the study.
10.5.23. It should also note that dye marker has a typical boiling point of 450oC and a flash point of 220-250oC, taking into account (1) a storage temperature is much below the flash point, (2) time is required to boil off the liquid to form vapour causing a tank failure (fire services is reasonably reliable in Hong Kong), (3) liquid will be released at the bottom of the tank instead of pressurized vapour during failure and (4) the stands of the tank are mechanically mounted on a concrete base, the actual risk would be even lower. It can conclude that rocketing of the dye marker tank is very remote.
10.5.24. Typical loading line at a jetty is incorporated with an emergency release coupler, the failure frequencies of the coupler is taken as 0.013[51]. For unloading hose failure, there is personnel who would initiate an emergency shutdown, the probability of isolation failure is taken as 0.1 for human error that failure to start an emergency shutdown system.
10.5.25. There is no updated information from the operator of the Sinopec HK Oil Terminal Chai Wan in relation to the number of road tanker and drum truck for fuel oil filling. The number of visit of road tanker and drum truck are therefore directly quoted from the NWFB Depot EIA as 6000 road tanker filling operations per year and 60 drum trucks per day; bulk filling operation is assumed to have a duration of 2 hours. Where the pump in the pump farm is in service for 5 hours a day and 300 days per year.
10.5.26. The fault tree analysis is given in Appendix 10.4 and the event tree analysis is illustrated in Figure 10.10 and Figure 10.12. The base event frequencies are summarized in Table 10.31 below.
Table 10.31 Base Failure Frequencies of Hazardous Events for Bulk Filling Operations (Sinopec HK Oil Terminal Chai Wan)
|
Hazardous Event |
Failure Frequency (per year) |
|
Cold catastrophic failure of fuel oil road tanker |
3.08E-06 |
|
Cold partial failure (25mm leak) of fuel oil road tanker |
1.28E-05 |
|
Rupture of loading arm during loading to fuel oil road tanker |
2.44E-02 |
|
Rupture of flexible hose during loading to drum truck |
2.08E-01 |
|
Full bore rupture of pump connecting pipeline in pump farm |
1.71E-06 |
|
Rupture of jetty loading hose during marine vessel unloading |
3.54E-04 |
|
Full bore rupture of fixed pipeline from fuel oil tank to jetty |
1.76E-05 |
|
Cold catastrophic failure of dye marker tank |
3.41E-05 |
Sinopec and ExxonMobil Petrol-cum-LPG Filling Station
10.5.27. Release frequencies of LPG events are derived from “Quantitative Risk Assessment Methodology for LPG Installations”[52]. The methodology for frequency estimation will also be consistent with the approach of previous approved QRA[53] and EIA reports[54],[55].
10.5.28. The inventories of underground vessel and road tanker are assumed to vary within a day as a result of consumption and refilling, the assumptions on inventory variation will follow that in Kai Tak Development EIA with justification. For underground LPG vessel, it is assumed that the vessel is nominally full (85% maximum capacity of vessel) in 20% of time and nominally 60% full in 80% of time. For road tanker, it is assumed that the inventory is full in 20% of time and 50% full in 80% of time.
10.5.29. For escalation event, the scenarios involving a jet fire impinging on the road tanker and a pool fire engulf the road tanker are considered.
10.5.30. Jet fire impingement from dispenser and flexible hose and engulfment by pool fires from a petrol tanker spillage, which could result in BLEVE of LPG road tanker. The probability of jet flame impingement is taken as 1/6 for aboveground pipework as in SIL EIA[56] and 1/12 for dispenser and vehicle filling hose considering a larger separation of LPG road tanker from dispensers. While a conservative factor of 0.5 is assumed for pool fire engulfment due to petrol road tanker release, which is a relative high factor by assuming an even distribution of opportunity for an LPG road tanker to be engulfed by the pool fire. Catastrophic rupture of a road tanker will be resulted that is already a worst case that no other knock on events have been identified.
10.5.31. To result in a petrol fire, an ignition source must be presence. Ignition probabilities of 0.08 and 0.03 are assigned for rupture and larger / medium leak of petrol road tanker respectively in accordance with the ignition probability of liquid release by Cox, Lees and Ang in Table 10.29.
10.5.32. Hazards from an accidental LPG release can be prevented or mitigated by the safety provisions at the LPG filling station. The following failure probabilities in Table 10.32 are assumed based on the previous “QRA methodology for LPG Installations” and Lees.
Table 10.32 Failure of Safety Provisions
|
Item |
Failure probability |
Remark |
|
Excess Flow Valve (LPG vessel) |
0.13 per demand |
|
|
Excess Flow Valve (LPG road tanker) |
0.013 per demand |
|
|
Excess Flow Valve (LPG dispenser) |
0.013 per demand |
Same one-year test interval as the LPG road tanker |
|
Non-Return Valve |
0.013 per demand |
|
|
ESD Trip System Fails |
1×10-4 per demand |
|
|
Breakaway Coupling |
0.013 per demand |
|
|
Double-Check Filler Valve |
2.6×10-3 per demand |
|
|
Water Spray System |
0.015 per demand |
|
|
Chartek Coating under Jet Fire Attack |
0.1 per demand |
|
|
Fire Service to Prevent BLEVE (Jet Fire
Impingement on the Road Tanker) |
0.5 per demand |
|
|
Pressure Relief Valve |
0.01 per demand |
PFD ranges from 1x10-2 to 1x10-4
per demand from Lees |
|
Truck Pump Over-pressure Protection System (LPG
Road Tanker) |
1×10-4 per demand |
Emergency protection |
10.5.33. The fault trees are shown in Appendix 10.4 and the event trees are shown in Figure 10.13 to Figure 10.17. The estimated base event frequencies of hazardous events are summaries in Table 10.33.
Table 10.33 Base Failure Frequencies of Hazardous Events (Petrol-cum-LPG Filling Stations)
|
Hazardous Event |
Inventory |
Failure Frequency (per year) |
|
|
Sinopec Station |
ExxonMobil Station |
||
|
Cold catastrophic failure of LPG storage vessel |
100% |
4.82E-08 |
9.53E-08 |
|
Cold catastrophic failure of LPG storage vessel |
60% |
1.93E-07 |
3.81E-07 |
|
Cold partial failure of LPG storage vessel |
100% |
1.50E-06 |
2.93E-06 |
|
Cold partial failure of LPG storage vessel |
60% |
5.99E-06 |
1.17E-05 |
|
Cold catastrophic failure of LPG road tanker |
100% |
6.20E-08 |
1.20E-07 |
|
Cold catastrophic failure of LPG road tanker |
50% |
2.48E-07 |
4.81E-07 |
|
Cold partial failure of LPG road tanker |
100% |
2.62E-07 |
4.99E-07 |
|
Cold partial failure of LPG road tanker |
50% |
1.05E-06 |
2.00E-06 |
|
Guillotine failure of liquid-inlet pipework
(rupture) |
100% |
3.26E-09 |
6.23E-09 |
|
Guillotine failure of liquid-inlet pipework
(rupture) |
60% |
1.31E-08 |
2.49E-08 |
|
Partial failure of liquid-inlet pipework (leak) |
100% |
4.18E-07 |
7.98E-07 |
|
Partial failure of liquid-inlet pipework (leak) |
60% |
1.67E-06 |
3.19E-06 |
|
Guillotine failure of liquid supply line to
dispenser (rupture) |
100% |
8.51E-08 |
1.37E-07 |
|
Guillotine failure of liquid supply line to
dispenser (rupture) |
60% |
3.40E-07 |
5.48E-07 |
|
Partial failure of liquid supply line to dispenser
(leak) |
100% |
1.65E-06 |
2.97E-06 |
|
Partial failure of liquid supply line to dispenser
(leak) |
60% |
6.61E-06 |
1.19E-05 |
|
Failure of dispenser |
100% |
7.27E-06 |
1.71E-05 |
|
Failure of dispenser |
60% |
2.91E-05 |
6.84E-05 |
|
Guillotine failure of flexible hose to vessel
(rupture) |
100% |
2.59E-06 |
4.94E-06 |
|
Guillotine failure of flexible hose to vessel
(rupture) |
50% |
1.04E-05 |
1.98E-05 |
|
Partial failure of flexible hose to vessel (leak) |
100% |
7.27E-06 |
1.39E-05 |
|
Partial failure of flexible hose to vessel (leak) |
50% |
2.91E-05 |
5.55E-05 |
|
Guillotine failure of flexible hose to vehicle
(rupture) |
100% |
1.69E-04 |
5.92E-04 |
|
Guillotine failure of flexible hose to vehicle
(rupture) |
60% |
6.77E-04 |
2.37E-03 |
|
Submersible Pump Flange Leak |
100% |
1.00E-06 |
1.00E-06 |
|
Submersible Pump Flange Leak |
60% |
4.00E-06 |
4.00E-06 |
|
BLEVE of LPG road tanker (fire escalation) |
100% |
4.69E-11 |
1.78E-10 |
|
BLEVE of LPG road tanker (fire escalation) |
50% |
1.88E-10 |
7.11E-10 |
10.5.34. For petrol road tanker failure in petrol-cum-LPG filling station, the failure frequencies are shown in the fault tree in Appendix 10.4 and event tree in Figure 10.18. The failure frequencies of catastrophic rupture and partial failure of road tanker are further modified to consider the escalation effect due to vapour cloud explosion from LPG failure releases. The calculation of failure frequency is shown in Table 10.34 below.
Table 10.34 Escalation Failure Frequencies of Petrol Road Tanker Due to LPG Explosion
|
Hazardous Event |
Inventory |
Failure Frequency without knock-on Effect (per year) |
Fraction of Time Onsite [1] |
Escalation Frequency [2] |
Failure Frequency with knock-on Effect (per year) |
Failure Frequency without knock-on Effect (per year) |
Fraction of Time Onsite [1] |
Escalation Frequency [2] |
Failure Frequency with knock-on Effect (per year) |
|
|
|
Sinopec |
Sinopec |
Sinopec |
Sinopec |
ExxonMobil |
ExxonMobil |
ExxonMobil |
ExxonMobil |
|
Cold catastrophic failure of petrol road tanker |
100% |
1.85E-8 |
0.04 |
1.13E-8 |
1.90E-08 |
6.40E-08 |
0.14 |
1.96E-8 |
6.68E-08 |
|
Cold catastrophic failure of petrol road tanker |
50% |
7.41E-8 |
0.04 |
4.53E-8 |
7.60E-08 |
2.56E-07 |
0.14 |
7.83E-08 |
2.67E-07 |
|
Cold partial failure of LPG petrol road tanker |
100% |
7.82E-08 |
0.04 |
1.13E-8 |
7.86E-08 |
2.66E-07 |
0.14 |
1.96E-8 |
2.69E-07 |
|
Cold partial failure of LPG petrol road tanker |
50% |
3.13E-07 |
0.04 |
4.53E-8 |
3.15E-07 |
1.06E-06 |
0.14 |
7.83E-08 |
1.07E-06 |
Note [1]: Fraction of time onsite = No. of road tanker per year × Time of presence / 24 / 365. Time of presence is assumed to be 2 hours per delivery.
Sinopec: Fraction of time onsite = 183 × 2 / 24 / 365 = 0.04
ExxonMobil: Fraction of time onsite = 621 × 2 / 24 / 365 = 0.14
Note [2]: Escalation frequency = LPG BLEVE frequency + VCE frequency from cold catastrophic failure of LPG storage vessel + VCE frequency from LPG guillotine failure of liquid supply line to dispenser. (i) LPG BLEVE frequency due to petrol road tanker jet fire is excluded from the escalation frequency. (ii) VCE frequencies due to cold catastrophic failure of LPG road tanker, guillotine failure of LPG liquid-inlet pipework and guillotine failure of flexible hose to LPG vessel are excluded as LPG road tanker unloading is not carried out simultaneously with petrol road tanker unloading.
Sinopec: Escalation frequency = 1.90E-10 + 1.69E-8 + 3.96E-8 = 5.66E-8 per year
ExxonMobil: Escalation frequency = 7.23E-10 + 3.34E-8 + 6.38E-8 = 9.79E-8 per year
Note [3]: Following the same practice as that of LPG road tanker, petrol road tanker is assumed that the inventory is full in 20% of time and 50% full in 80% of time.
LPG Wagon Parking Site
10.5.35. The frequency of rupture of LPG cylinder on wagon is 6.8 × 10-6 per vehicle year while that of leak is 2.6 × 10-6 per cylinder year18. Assume 3 number of LPG wagons onsite that each contains 250 number of 10.5kg LPG cylinders, the frequencies of LPG cylinder rupture and leak are 2.4 × 10-5 per year and 1.95 × 10-3 per year.
10.5.36. Failure frequencies for rupture and leak of kerosene tanker are 2.0 × 10-6 per tanker year and 5.0 × 10-6 per tanker year respectively[57]. The LPG wagons and kerosene road tanker are assumed to be parked onsite for 24 hours, which is already conservative as the wagons and road tankers will be offsite for commercial activities.
10.5.37. The estimated base event frequencies of hazardous events are summaries in Table 10.35. The fault trees are shown in Appendix 10.4 and the event trees are shown in Figure 10.19 to Figure 10.20.
Table 10.35 Base Failure Frequencies of Hazardous Events (LPG Wagon Site)
|
Hazardous Event |
Failure Frequency (per year) |
|
Cold catastrophic failure of LPG cylinder |
2.04E-05 |
|
Cold partial failure of LPG cylinder |
1.95E-03 |
|
Cold catastrophic failure of kerosene tanker |
2.02E-06 |
|
Cold partial failure of kerosene tanker |
5.00E-06 |
|
BLEVE of LPG cylinder |
4.98E-07 |
Diesel Storage in NWFB and Citybus Depots
10.5.38. The base failure frequency for tanker rupture and for tanker leak are 2.0×10-6 and 5.0×10-6 per tanker year respectively by taking reference from petrol/diesel tanker failure in the SIL EIA18, 57.
10.5.39. The number of road tanker and the corresponding unloading time in the NWFB Depot and Citybus Depot are provided by the proposed Project operators as shown in Table 10.4. The frequencies for release of diesel from road tanker are summarised in Table 10.36 below and the fault trees and event trees are shown in Appendix 10.5 and Figure 10.21 respectively.
10.5.40. Taking into account the ignition probability, firefighting and water spray system of the depot, the frequencies of pool fire as a result of diesel tanker release are above 1×10-9 /year. Pool fire due to release from diesel tanker is therefore further considered in the study.
10.5.41. Since smoke plume is generated from the pool fire from ignition of diesel tanker releases, the frequencies of smoke plume is the same as that of the pool fire frequencies.
Table 10.36 Frequencies of Release from Diesel Road Tanker
|
Hazardous Event |
Frequency (per tanker year) |
Frequency of release (per year) |
Frequency of Pool Fire / Smoke Plume (per year) |
||
|
NWFB |
Citybus |
NWFB |
Citybus |
||
|
Tanker rupture |
2.0E-6 |
5.34E-07 |
2.50E-07 |
2.14E-09 |
1.00E-09 |
|
Tanker medium liquid leak (Hole size: 25mm) |
5.0E-6 |
2.87E-06 |
1.88E-06 |
1.15E-08 |
7.51E-09 |
External Events: Aircraft Crash
10.5.42. The Project site is located around 32 km from the Hong Kong International Airport. The frequency of aircraft crash is estimated using the HSE methodology[58].
10.5.43. The model takes into account specific factors such as the target area of the proposed hazard site and its longitudinal (x) and perpendicular (y) distances from the runway threshold. The crash frequency per unit ground area (per km2) is calculated as:
10.5.44. Where N is the number of runway movements per year and R is the probability of an accident per movement (landing or take-off). FL(x,y) gives the spatial distribution of crashes for landing and is given by:
for x >-3.275 km
10.5.45. For aircraft take-off,
for x >-0.6 km
10.5.46. The 10-year moving average of the NTSB fatal accidents[59] suggests a downward trend in recent years of a failure rate of about 1×10-7 per flight. However, only 18.7% of accidents are associated with the approach to landing, 14% are associated with take-off and 4.7% are related to the climb phase of the flight according to the Annual Review of Aircraft Accident Data. The accident frequency for the approach to landings hence becomes 1.87×10-8 per flight and for take-off/ climb 1.87×10-8 per flight.
10.5.47. The number of flights from 2001 to 2013 is extracted from the Civil Aviation Department[60], and extrapolated to year 2016 and 2018 by adopting an annual growth rate of 5% for passenger and cargo traffic based on data from Hong Kong International Airport[61]. The number of flights at Chek Lap Kok for year 2016 and 2018 are estimated at 430,729 and 474,879 respectively.
10.5.48. The distance x and y are estimated to be 29 km and 15 km respectively. By applying the equations, the calculated crash frequencies and impact frequencies in 2016 and 2018 are given in Table 10.37 and Table 10.38 below.
Table 10.37 Aircraft Crash Frequency
|
Year |
Estimated
no. of flight |
Distance from runway (km) |
Crash Frequency (/km2/yr) |
|||||||
|
07L/25R |
07R/25L |
07L |
25R |
07R |
25R |
Total |
||||
|
x |
y |
x |
y |
Take off |
Land-ing |
Take off |
Land-ing |
|
||
|
2016 |
430,729 |
29 |
15 |
29 |
15 |
9.05E-21 |
4.35E-14 |
9.05E-21 |
4.35E-14 |
8.70E-14 |
|
2018 |
478,879 |
29 |
15 |
29 |
15 |
9.97E-21 |
4.79E-14 |
9.97E-21 |
4.79E-14 |
9.59E-14 |
Table 10.38 Aircraft Impact Frequency
|
Year |
Estimated
no. of flight |
Area of ExxonMobil
Petrol-cum-LPG Filling Station1 (km2) |
Impact
Frequency (/yr) |
|
2016 |
430,729 |
0.0126 |
1.10E-15 |
|
2018 |
478,879 |
0.0126 |
1.21E-15 |
Note:
Area is conservatively taken as the area of NWFB Depot since its area is the largest when compared to other identified potential hazardous sources such as the Sinopec and ExxonMobil Petrol-cum-LPG Filling Stations.
The calculated impact frequency is around 1×10-15 per year, which is much less than 1×10-9 per year. The impact of aircraft crash will therefore not be further considered in this assessment.
External Events: Earthquake
10.5.49. Hong Kong is situated on the southern coast of China. Hong Kong is not located within the seismic belt and according to Hong Kong Observatory, earthquakes occurring in the circum-Pacific seismic belt which passes through Taiwan and Philippines are too far away to affect Hong Kong significantly. Buildings and infrastructures in Hong Kong are designed to withstand earthquakes up to Modified Mercali Intensity (MMI) VII.
10.5.50. It is estimated that MMI VIII is of sufficient intensity to cause damage to specially designed structures. In this analysis it is assumed that such earthquake may result in storage vessel leakage and pipework rupture at a probability of 0.01[62]. The probability of earthquake occurrence at MMI VIII and higher in Hong Kong is very low comparing with other regions and is estimated to be 1.0 x 10-5 per year[63].
External Events: Helicopter Crash
10.5.51. Helicopter accidents during take-off and landings are confined to a small area around the helipad, extending up to 200 m only from the centre of the helipad. 93% of accidents occur within 100m of the helipad. The remaining 7% occur between 100 and 200 m of the helipad[64].
10.5.52. The nearest helicopter landing pad is located at the Pamela Youde Nethersole Eastern Hospital. The distance of the helicopter landing pad to the potential hazardous sources are listed in Table 10.39 below. Since the distance to nearest helicopter landing pad is about 585 m away, only the background crash rate for helicopters is considered in this report.
10.5.53. From a study in the previous approved EIA report, the background crash rate for helicopter is 1.0×10-5/km2/year[65] and hence, by accounting for the target area of each potential hazardous source, the background crash frequencies for helicopter are calculated and summarised as in Table 10.39.
Table 10.39 Background Helicopter Crash Frequencies
|
Potential Hazardous Sources |
Approximate Distance to The Nearest Helicopter Landing Pad (m) |
Site Area (m2) |
Estimated Helicopter Crash Frequency (per year) |
|
Sinopec HK Oil Terminal Chai
Wan |
850 |
2,970 |
2.97E-08 |
|
Sinopec Petrol-cum-LPG Filling
Station |
825 |
486 |
4.86E-09 |
|
ExxonMobil Petrol-cum-LPG
Filling Station |
600 |
3,930 |
3.93E-08 |
|
LPG Wagon Parking Site |
585 |
1,870 |
1.87E-08 |
|
Diesel Storage in NWFB Depot |
700 |
1,264 |
1.26E-08 |
|
Diesel Storage in Citybus Depot |
485 |
11,116 |
1.11E-07 |
|
DGs Storage in the Proposed
Project |
630 |
6,987 |
6.99E-08 |
External Events: Vehicle Crash
10.5.54. Road traffic accident statistics in Table 10.40 below as from the Transport Department shows 85% of all road accidents in Hong Kong are slight collision, 14% (take 20% in the aftermentioned calculation) are serious collision and 1% are fatal collision. Most of the road accidents are related to speeding, crossing the road, drunk / drug drive, poor road condition, bad weather, etc. In this assessment, it is assumed fatal accidents have the potential to cause catastrophic rupture of the tanker or guillotine failure of pipework, and serious accidents have the potential to cause leakage of the tanker / pipework.
10.5.55. Speed limit, sufficient lighting, well maintained concrete floor and warning signs are assumed to be provided in bus depot, oil terminal and LPG filling stations. To account for the aforementioned provisions, a modification factor of 0.5 is conservatively applied, i.e. the probability of fatal and serious damage in an impact accident is taken as 1% × 0.5 = 0.5% and 20% × 0.5 = 10%, respectively. For the liquid-inlet pipework at the LPG filling point, a modification factor of 0.1 is applied considering the extra protection from the crash barrier (Metal protection poles, Figure 10.6), i.e., the probability of 0.1% and 2% is adopted for fatal and serious damage in an impact accident.
Table 10.40 Road Traffic Accidents by Severity (2008 - 2012)
|
Severity |
Year |
||||||
|
2008 |
2009 |
2010 |
2011 |
2012 |
Sum |
% |
|
|
Fatal |
143 |
126 |
114 |
128 |
116 |
627 |
0.8 |
|
Serious |
2,096 |
1,943 |
2,052 |
2,190 |
2,385 |
10,666 |
14.2 |
|
Slight |
12,337 |
12,247 |
12,777 |
13,223 |
13,393 |
63,977 |
85.0 |
|
Total |
14,576 |
14,316 |
14,943 |
15,541 |
15,894 |
75,270 |
100.0 |
External Events: Landslide
10.5.56. The nearest slope in the vicinity is adjacent the Island Eastern Corridor that across the Shing Tai Road and MTR rail track, risk due to landslide on the potential hazardous sources and the Project site is considered negligible.
External Events: Lightning Strike
10.5.57. Lightning sparks could ignite the combustible gas in air in case of leakage. Most tank fire accidents caused by lightning are due the ignition of hydrocarbon leak around the seal of the floating roof oil storage tanks. Petrol-cum-LPG filling stations and oil terminals are protected with lighting conductors to safety earth direct lightning strikes. The grounding will be inspected regularly. Plus the frequency of lightning strike on a properly protected building structure is extremely low in Hong Kong. The chance of lightning strike is therefore very remote and ignition due to lightning strike is taken to be covered by the ignition probability.
External Events: Subsidence
10.5.58. Excessive subsidence may lead to failure of the structure and ultimately loss of containment scenario. However, subsidence is usually slow in movement and such movement can be observed and remedial action can be taken in time. Risk from subsidence is therefore deemed remote and not further considered.
External Events: Typhoon / Tsunami
10.5.59. LPG filling stations, oil storage tank and building structures are designed safe to withstand the wind load for typhoon. The site is not threatened by tsunami with Hong Kong Island located between Victoria Harbour and South China Sea. Therefore the risk is deemed unlikely and not further considered in the analysis.
External Events: External fire
10.5.60. External fire refers to the occurrence of a fire event outside the potential hazardous sources which may lead to the failure of the facilities. This might be expected from road accidents on the public road, probably involving car crash or engine failures (e.g. overheating during hot summer). The resulting fire is usually small, only affecting a few meters around the car, and could be quickly extinguished using fire extinguishers or by the fire brigade. The key facilities inside are further protected by concrete building structures (e.g. the LPG vessel compartment, bus depot and fence wall of oil terminal) and activation of emergency shutdown system for potential external fire threat. The risk of escalation of external fire to the potential hazardous sources is deemed remote and not further considered in the analysis.
External Events: Dropped Object
10.5.61. A Petrol-cum-LPG filling station is sheltered by a roof and the bus depots are located within a building. Where there is no building in the close vicinity of the Sinopec HK Oil Terminal Chai Wan that is high enough to drop an object. And the nearest building to the LPG Wagon Parking Site is around 30m away. Thus, the threat from dropped objects to the potential hazardous installation is insignificant and not further assessed in the analysis.
10.6.1. The consequence assessment is conducted in two steps: (1) Source term modelling to determine the release rate, duration and quantity; (2) Physical effects modelling to determine the gas dispersion, fire and explosion effects zone based on the output of source term modelling. The impact of the hazardous outcomes on the surrounding population is analyzed.
10.6.2. In this study, the simulation software SAFETI 6.7 by Det Norske Veritas (DNV) is employed to calculate the hazardous release and the effects zones.
10.6.3. The DGs involved in this study are modelled as mixtures of hydrocarbons. To enable the assessment, simple mixtures or representative component will be used for the risk model building. Assumptions are made as follows:
· LPG is modelled as mixture of propane and n-butane in 30:70 ratio by mass.
· Petrol, diesel, kerosene and dye marker– petrol with a flash point lower than 23 oC is modelled as n-pentane and diesel/kerosene/dye marker with a flash point higher than 23 oC is modelled as nonane[66].
· The toxic effect of smoke is modelled using carbon monoxide as its toxicity causes the majority of deaths in fire[67].
10.6.4. For continuous release, release duration is based on time to empty the whole content of tank/cylinder.
10.6.5. For instantaneous failure of container, whole content of a tank is used. In case of continuous release, release parameters such as release rate and exit velocity are calculated by discharge model according to storage conditions. Release parameters together with release duration are then fed into dispersion model to calculate the effect.
10.6.6. The following section briefly describes major mathematical models applied to various fire and dense gas dispersion in the consequence model.
Gas Dispersion
10.6.7. The Unified Dispersion Model (UMD), which is the dispersion model used in PHAST of SAFETI software to assess the dispersion, without rainout effect will be used for the dispersion in non-immediate ignition scenarios. The model takes into account various transition phases, from dense cloud dispersion to buoyant passive gas dispersion, in both instantaneous and continuous releases. Besides, toxic effect will be evaluated using the UDM dispersion model when the cloud reaches population sites for release of gas without ignition.
10.6.8. Upon release of flammable gas, a number of possible outcomes may be occurred depended on whether the gas is ignited immediately or ignited after a period of time. The dispersion characteristics will be influenced by the meteorological conditions and the material properties, such as density, of the released gas.
10.6.9. Fire scenarios of different kinds may be developed in the presence of ignition source in the proximity of gas release. If no ignition source exist, the gas cloud may disperse downwind and be diluted to the concentration below its Lower Flammable Limit (LFL). In this case, no harm effect is anticipated since the gas would become too lean to ignite.
Fireball and BLEVE
10.6.10. Immediate ignition of an instantaneous release of the whole inventory inside a pressurized vessel will result in a fireball. A fireball is characterized by its high thermal radiation intensity and short duration time. The principal hazard of fireball arises from thermal radiation, which is not significantly influenced by weather, wind direction or source of ignition. A BLEVE is similar to a fireball except that it is caused by integrity failure from fire impingement and therefore occurs as fire escalation events. The physical effects are calculated in the same way as fireballs. A 100% fatality is assumed for anyone within the fireball radius. For persons indoor within the fireball radius, a 50% fatality is assumed.
Jet Fire
10.6.11. When a pressurised flammable gas is released and ignited immediately, a jet fire will occur. The momentum of the release carries the flammable substance forward in a long plume, giving a flammable mixture by entraining air. Combustion in a jet fire occurs in the form of a strong turbulent diffusion flame, which is heavily influenced by the momentum of the release. The major concern regarding jet fire is the heat radiation effect generated from the fire. The thermal effect to adjacent population will be quantified in the consequence model.
Flash Fire
10.6.12. Following a hazardous gas release, it may form a flammable gas cloud initially located around the release point. If this cloud does not get ignited immediately, it will move in the downwind direction and be diluted as a result of air entrainment. Flash fire is the consequence of combustion of gas cloud resulting from delayed ignition. The flammable gas cloud can be ignited at its edge and causes a flash fire of the cloud within the Lower Flammable Limit (LFL) and Upper Flammable Limit (UFL) boundaries.
10.6.13. Major hazards from flash fire are thermal radiation and direct flame contact. Since the flash combustion of a gas cloud normally lasts for a short duration, the thermal radiation effect on people near a flash fire is limited. Humans who are encompassed outdoors by the flash fire will be fatally injured. A fatality rate of unity is assumed for outdoor population, and 90% protection factor is assumed for indoor.
Vapour Cloud Explosion
10.6.14. When there is a large amount of pressurised gas rapidly releasing to the atmosphere from a pressurised tank, a vapour cloud will be formed, dispersed and mixed with the surrounding air. If the vapour cloud is passing through a confined/ semi-confined environment and gets ignited, the confinement will limit the degree of expansion of the burning cloud and create an overpressure and explosion. This type of explosion is called a VCE.
Thermal Radiation
10.6.15. Hazardous consequences, such as jet fire, flash fire etc. will be assessed using PHAST’s consequence models. Fatality probability of various hazardous event outcomes will be evaluated at a number of end-point criteria in each type of hazard outcome. The estimation of the fatality / injury caused by a physical effect such as thermal radiation requires the use of Probit equations, which describe the probability of fatality as a function of some physical effect. The probability of fatality, Pr, due to exposure to heat radiation, i.e. jet fire and fireball is given by the following probit relationship by Eisenberg et al. which provides one of the more conservative estimates[68]:
Where,
Pr is the probit associated with the probability of fatality;
Q is the heat radiation intensity (kW/m2); and
t is the exposure time (s).
Smoke (Carbon Monoxide)
10.6.16. The toxicity of carbon monoxide is due to the formation of blood carboxyhaemoglobin. This results in a reduction of the supply of oxygen to critical body organs and is referred to as anemic anoxia. The probit function for death due to toxic exposure to carbon monoxide is given by TNO Purple Book[69]:
Where,
Pr is the probit associated with the probability of fatality;
C is the concentration (mg/m3); and
t is the exposure time (s).
Height Protection Factor
10.6.17. Following the methodology for LPG installations[70], a protection factor is used to factor down those population below the dispersed LPG cloud in the risk summation for flash fire events.
10.6.18. A cloud height of 36m is resulted from the catastrophic rupture of a LPG road tanker, which is highest cloud height amongst the LPG failure scenarios. Buildings are assumed to be 3 m for each storey.
10.6.19. The protection factor is determined by considering the height of a building relative to the dispersed cloud height, i.e., 36m divided by the height of a building.
Shielding factor
10.6.20. Shielding factors are used to allow for the shielding of buildings by other buildings from fireball effects. A shielding factor of 0.5 is assigned to those buildings within the fireball diameter, outside the fireball and partly inside and partly outside the fireball. The shielding factor is applied to the following populations:
|
ID |
Buildings
Assigned with Fireball Shielding Factor |
|
1 |
Heng Fa Chuen Block 48, 49, & 50 |
|
3 |
Seawater Pump House |
|
8 |
Government Logistics Centre |
|
10 |
HKE Heng Fa Chuen Substation |
|
12 |
NWFB Depot |
|
13 |
Proposed Project |
|
15 |
Citybus Depot |
|
16 |
Planned THEi New Campus in Chai Wan |
|
18 |
Hong Kong Institute of Vocational Education (Chai Wan) |
|
19 |
Knight Court |
|
20 |
MTR Chai Wan Depot |
|
21 |
EMSD Workshop |
|
24 |
Chai Wan Industrial City, Phase II |
|
25 |
Chai Wan Industrial City, Phase I |
|
26 |
Cornell Centre |
|
27 |
Tsui Hong House, Tsui Wan Estate |
|
28 |
Tsui Shou House, Tsui Wan Estate |
|
29 |
Tsui Fuk House, Tsui Wan Estate |
|
31 |
Hang Tsui Court Indoor Carpark |
|
32 |
Hang Tsui Court |
|
33 |
TWGHs & LKWFSL Mrs Fung Yiu Hing Memorial Primary School |
10.7.1. By summation of the results of meteorological data, frequency estimation and consequence analysis, risk levels of the assessed scenarios will be characterised in terms of individual risk (presented by individual risk contours).
10.7.2. By combining the results of population data, meteorological data, frequency estimation and consequence analysis, societal risk of the assessed scenarios will be characterised in terms of F-N curves and Potential Loss of Life (PLL).
10.7.3. The above steps will be done by using MPACT in the SAFETI software suite version 6.7.
10.7.4. The individual risk contours of the Sinopec HK Oil Terminal Chai Wan, Sinopec Petrol-cum-LPG Filling Station, ExxonMobil Petrol-cum-LPG Filling Station, LPG Wagon Parking Site, NWFB Depot and Citybus Depot are illustrated in Figure 10.22 to Figure 10.27 respectively. The overall individual risk contours are presented in Figure 10.28.
10.7.5. For Sinopec HK Oil Terminal Chai Wan, Sinopec Petrol-cum-LPG Filling Station, ExxonMobil Petrol-cum-LPG Filling Station, LPG Wagon Parking Site, NWFB Depot and Citybus Depot, the corresponding individual risks to offsite population are lower than 1×10-5/year and decrease at distances further away from the installations. Therefore in terms of individual risk, the criteria set in Section 2 of Annex 4 of the EIAO-TM are satisfied and no residual impact is anticipated.
10.7.6. It should be noted that the individual risk contours are independent of the population sets and are therefore consistent in both construction and operation phases.
Construction Phase
10.7.7. The F-N curves of Case 1 and Case 2 compare the societal risks in the construction phase without and with construction workers of the proposed Project. The results are presented in Figure 10.29 to Figure 10.34 for Sinopec HK Oil Terminal Chai Wan, ExxonMobil Petrol-cum-LPG Filling Station, Sinopec Petrol-cum-LPG Filling Station, NWFB Depot, Citybus Depot as well as LPG Wagon Parking Site.
10.7.8. With an increased population by the construction workers of the proposed Project, the risk in terms of societal risk posed by the ExxonMobil Petrol-cum-LPG Filling Station is increased most obviously in the N = 40 to 100 region. For the NWFB Depot, the risk level is increased in the region of N = 5 to 30. The change in risk level is considered as temporary due to the temporary increase in population for construction works.
10.7.9. On the other hand, the effect of additional construction workers to the risks posed by the Sinopec HK Oil Terminal Chai Wan, Sinopec Petrol-cum-LPG Filling Station, Citybus Depot and LPG Wagon Parking Site are negligible as shown in Figure 10.29, Figure 10.31, Figure 10.33 and Figure 10.34.
10.7.10. The societal risks for Sinopec HK Oil Terminal Chai Wan, Sinopec Petrol-cum-LPG Filling Station, ExxonMobil Petrol-cum-LPG Filling Station, NWFB Depot, Citybus Depot as well as LPG Wagon Parking Site in the construction phase are within the ACCEPTABLE region and therefore satisfy Section 2 of Annex 4 of the EIAO-TM.
10.7.11. The F-N data for construction phase is presented in Table 10.41 and top 5 most significant PLL contributors are shown in Table 10.43. It can be observed that the cold catastrophic and partial failure of LPG road tanker with 50% of inventory and guillotine failure of liquid-inlet pipework contribute the most in the PLL. It can be seen that the increase in construction workers in the proposed Project does not contribute change to the F-N and PLL data of the Sinopec HK Oil Terminal Chai Wan and LPG Wagon Parking Site.
Operation Phase
10.7.12. The F-N curves of Case 3 and Case 5 compare the societal risks in the operation phase without and with the proposed Project while the F-N curves of Case 4 represent the risk to the proposed Project workers only. The results are presented in Figure 10.35 to Figure 10.40.
10.7.13. The risk in terms of societal risk induced by the ExxonMobil Petrol-cum-LPG Filling Station is slightly increased by the increase of population of the proposed Project. As shown in Figure 10.36, the F-N curve remains inside the ACCEPTABLE region. And with an increase in the population of the proposed Project, the risk level is increased slightly but again remains in the ACCEPTABLE region. The change in risk level is considered as permanent and irreversible due to the increase in population.
10.7.14. The societal risks of the New World First Bus Permanent Depot without and with the Project workers are shown in Figure 10.38. The risk level is increased by an increase in population of the Project workers, the F-N curve however still remains in the ACCEPTABLE region.
10.7.15. The effect of Project workers to the risks posed by the Sinopec HK Oil Terminal Chai Wan, Sinopec Petrol-cum-LPG Filling Station, Citybus Depot and LPG Wagon Parking Site are negligible as shown in Figure 10.35, Figure 10.37, Figure 10.39 and Figure 10.40. No residual impact is anticipated.
10.7.16. When evaluating the risk induced by the ExxonMobil Petrol-cum-LPG Filling Station to the proposed Project workers alone (Case 4), the F-N curve is in the ACCEPTABLE region as shown in Figure 10.36. No residual impact is anticipated. Sinopec HK Oil Terminal Chai Wan, Sinopec Petrol-cum-LPG Filling Station and LPG Wagon Parking Site do not show an F-N data when considering Project workers alone.
10.7.17. The cumulative risk taking into account the proposed Project workers and all the offsite populations posed by all hazardous installations, including the Sinopec HK Oil Terminal Chai Wan, Sinopec Petrol-cum-LPG Filling Station, ExxonMobil Petrol-cum-LPG Filling Station, NWFB Depot, Citybus Depot and LPG Wagon Parking Site, is illustrated in Figure 10.41. With the additional population of the proposed Project, the cumulative risk in terms of the F-N curve is slightly increased and within the ACCEPTABLE region. The change in risk level is considered as permanent and irreversible due to the increase in population. No residual impact is anticipated.
10.7.18. The F-N data of the operation phase for individual potential hazardous installation is presented in Table 10.42 and top 5 most significant PLL contributors among all hazardous installations are shown in Table 10.44. The most significant events are the cold catastrophic and partial failure of LPG road tanker with 50% of inventory and guillotine failure of liquid-inlet pipework.
10.7.19. Further analyse the result from Table 10.44 can observe the ExxonMobil Petrol-cum-LPG Filling Station is the main PLL contributor (overall around 50%) in the cumulative risk case, in which the most significant event is the failure of LPG road tanker.
Table 10.41 F-N Data for Construction Phase
|
ExxonMobil Petrol-cum-LPG Filling Station |
Sinopec Petrol-cum-LPG Filling Station |
||||||
|
Case 1 – 2016 without construction workers |
Case 2 – 2016 with construction workers |
Case 1 – 2016 without construction workers |
Case 2 – 2016 with construction workers |
||||
|
No. of Fatalities |
Frequency
(/year) |
No. of Fatalities |
Frequency
(/year) |
No. of Fatalities |
Frequency
(/year) |
No. of Fatalities |
Frequency
(/year) |
|
1 |
2.10E-06 |
2.13E-06 |
2.15E-06 |
1 |
2.52E-06 |
1 |
2.52E-06 |
|
2 |
1.56E-06 |
1.59E-06 |
1.61E-06 |
2 |
1.19E-06 |
2 |
1.19E-06 |
|
3 |
1.34E-06 |
1.40E-06 |
1.42E-06 |
3 |
8.44E-07 |
3 |
8.44E-07 |
|
4 |
1.15E-06 |
1.19E-06 |
1.20E-06 |
4 |
5.52E-07 |
4 |
5.52E-07 |
|
5 |
9.77E-07 |
1.02E-06 |
1.04E-06 |
5 |
4.36E-07 |
5 |
4.36E-07 |
|
6 |
8.05E-07 |
8.21E-07 |
8.30E-07 |
6 |
3.78E-07 |
6 |
3.78E-07 |
|
8 |
5.82E-07 |
5.93E-07 |
6.00E-07 |
8 |
1.62E-07 |
8 |
1.62E-07 |
|
10 |
4.85E-07 |
5.09E-07 |
5.15E-07 |
10 |
1.00E-07 |
10 |
1.00E-07 |
|
12 |
3.75E-07 |
4.00E-07 |
4.05E-07 |
12 |
9.60E-08 |
12 |
9.60E-08 |
|
15 |
3.37E-07 |
3.44E-07 |
3.48E-07 |
15 |
8.37E-08 |
15 |
8.37E-08 |
|
20 |
2.57E-07 |
2.78E-07 |
2.81E-07 |
20 |
7.89E-08 |
20 |
7.89E-08 |
|
25 |
2.49E-07 |
2.55E-07 |
2.57E-07 |
25 |
3.82E-08 |
25 |
3.82E-08 |
|
30 |
2.20E-07 |
2.27E-07 |
2.29E-07 |
30 |
3.46E-08 |
30 |
3.46E-08 |
|
40 |
2.03E-07 |
2.09E-07 |
2.12E-07 |
40 |
1.95E-08 |
40 |
1.95E-08 |
|
50 |
1.15E-07 |
1.73E-07 |
1.75E-07 |
50 |
1.29E-08 |
50 |
1.29E-08 |
|
60 |
9.28E-08 |
9.89E-08 |
9.97E-08 |
60 |
7.50E-09 |
60 |
7.50E-09 |
|
80 |
4.72E-08 |
6.46E-08 |
6.51E-08 |
80 |
1.98E-09 |
80 |
1.98E-09 |
|
100 |
2.46E-08 |
2.74E-08 |
2.76E-08 |
100 |
6.43E-10 |
100 |
6.43E-10 |
|
120 |
1.50E-08 |
1.64E-08 |
1.65E-08 |
120 |
1.76E-10 |
120 |
1.76E-10 |
|
150 |
8.67E-09 |
9.17E-09 |
9.23E-09 |
|
|
|
|
|
200 |
4.85E-09 |
5.03E-09 |
5.07E-09 |
|
|
|
|
|
250 |
3.01E-09 |
3.09E-09 |
3.11E-09 |
|
|
|
|
|
300 |
1.91E-09 |
1.96E-09 |
1.97E-09 |
|
|
|
|
|
400 |
1.14E-09 |
1.16E-09 |
1.17E-09 |
|
|
|
|
|
500 |
8.47E-10 |
8.48E-10 |
8.53E-10 |
|
|
|
|
|
600 |
5.53E-10 |
5.55E-10 |
5.58E-10 |
|
|
|
|
|
800 |
3.70E-10 |
3.68E-10 |
3.70E-10 |
|
|
|
|
|
1000 |
1.83E-10 |
1.82E-10 |
1.83E-10 |
|
|
|
|
|
1200 |
1.52E-10 |
1.51E-10 |
1.52E-10 |
|
|
|
|
Note: The F-N data below 1×10-9/year is not shown in the figures of F-N curves
|
Sinopec HK Oil Terminal Chai Wan |
LPG Wagon Parking Site |
||||||
|
Case 1 – 2016 without construction workers |
Case 2 – 2016 with construction workers |
Case 1 – 2016 without construction workers |
Case 2 – 2016 with construction workers |
||||
|
No. of Fatalities |
Frequency
(/year) |
No. of Fatalities |
Frequency
(/year) |
No. of Fatalities |
Frequency
(/year) |
No. of Fatalities |
Frequency
(/year) |
|
1 |
9.88E-07 |
1 |
9.88E-07 |
1 |
2.50E-07 |
1 |
2.50E-07 |
|
2 |
2.20E-07 |
2 |
2.20E-07 |
2 |
2.58E-08 |
2 |
2.58E-08 |
|
3 |
2.15E-07 |
3 |
2.15E-07 |
3 |
4.04E-09 |
3 |
4.04E-09 |
|
4 |
1.50E-07 |
4 |
1.50E-07 |
4 |
1.40E-09 |
4 |
1.40E-09 |
|
5 |
1.45E-07 |
5 |
1.45E-07 |
5 |
1.15E-09 |
5 |
1.15E-09 |
|
6 |
1.20E-07 |
6 |
1.20E-07 |
6 |
6.00E-10 |
6 |
6.00E-10 |
|
8 |
1.10E-07 |
8 |
1.10E-07 |
|
|
|
|
|
10 |
1.05E-07 |
10 |
1.05E-07 |
|
|
|
|
|
12 |
8.57E-08 |
12 |
8.57E-08 |
|
|
|
|
|
15 |
7.00E-08 |
15 |
7.00E-08 |
|
|
|
|
|
20 |
7.00E-08 |
20 |
7.00E-08 |
|
|
|
|
|
25 |
2.00E-08 |
25 |
2.00E-08 |
|
|
|
|
|
30 |
2.00E-08 |
30 |
2.00E-08 |
|
|
|
|
Note: The F-N data below 1×10-9/year is not shown in the figures of F-N curves
|
NWFB Depot |
Citybus Depot |
||||||
|
Case 1 – 2016 without construction workers |
Case 2 – 2016 with construction workers |
Case 1 – 2016 without construction workers |
Case 2 – 2016 with construction workers |
||||
|
No. of Fatalities |
Frequency
(/year) |
No. of Fatalities |
Frequency
(/year) |
No. of Fatalities |
Frequency
(/year) |
No. of Fatalities |
Frequency
(/year) |
|
1 |
1.18E-08 |
1 |
1.21E-08 |
1 |
7.20E-09 |
1 |
7.20E-09 |
|
2 |
1.05E-08 |
2 |
1.09E-08 |
2 |
6.22E-09 |
2 |
6.22E-09 |
|
3 |
9.10E-09 |
3 |
9.59E-09 |
3 |
5.20E-09 |
3 |
5.20E-09 |
|
4 |
7.42E-09 |
4 |
7.98E-09 |
4 |
4.73E-09 |
4 |
4.73E-09 |
|
5 |
6.38E-09 |
5 |
6.99E-09 |
5 |
4.56E-09 |
5 |
4.56E-09 |
|
6 |
5.78E-09 |
6 |
6.52E-09 |
6 |
4.46E-09 |
6 |
4.46E-09 |
|
8 |
4.44E-09 |
8 |
5.44E-09 |
8 |
4.28E-09 |
8 |
4.28E-09 |
|
10 |
3.34E-09 |
10 |
4.25E-09 |
10 |
4.09E-09 |
10 |
4.09E-09 |
|
12 |
2.92E-09 |
12 |
3.44E-09 |
12 |
3.89E-09 |
12 |
3.89E-09 |
|
15 |
2.70E-09 |
15 |
3.05E-09 |
15 |
3.64E-09 |
15 |
3.65E-09 |
|
20 |
2.33E-09 |
20 |
2.62E-09 |
20 |
3.12E-09 |
20 |
3.12E-09 |
|
25 |
2.06E-09 |
25 |
2.32E-09 |
25 |
2.09E-09 |
25 |
2.09E-09 |
|
30 |
1.96E-09 |
30 |
2.10E-09 |
30 |
1.78E-09 |
30 |
1.78E-09 |
|
40 |
8.66E-10 |
40 |
9.28E-10 |
40 |
1.44E-09 |
40 |
1.44E-09 |
|
50 |
6.02E-10 |
50 |
6.33E-10 |
50 |
1.01E-09 |
50 |
1.01E-09 |
|
60 |
1.80E-10 |
60 |
2.07E-10 |
60 |
6.93E-10 |
60 |
6.93E-10 |
|
80 |
1.59E-10 |
80 |
1.63E-10 |
80 |
4.92E-10 |
80 |
4.92E-10 |
|
100 |
1.58E-10 |
100 |
1.59E-10 |
100 |
4.75E-10 |
100 |
4.75E-10 |
|
120 |
1.53E-10 |
120 |
1.54E-10 |
120 |
4.30E-10 |
120 |
4.30E-10 |
|
150 |
1.27E-10 |
150 |
1.27E-10 |
150 |
3.96E-10 |
150 |
3.96E-10 |
|
|
|
|
|
200 |
3.95E-10 |
200 |
3.95E-10 |
|
|
|
|
|
250 |
3.85E-10 |
250 |
3.85E-10 |
|
|
|
|
|
300 |
2.51E-10 |
300 |
2.51E-10 |
|
|
|
|
|
400 |
1.73E-10 |
400 |
1.73E-10 |
Note: The F-N data below 1×10-9/year is not shown in the figures of F-N curves
Table 10.42 F-N Data for Operation Phase
ExxonMobil Petrol-cum-LPG Filling Station |
|||||
|
Case 3 – 2018 without Project workers |
Case 4 – 2018 with Project workers only |
Case 5 – 2018 with Project workers and all offsite population |
|||
|
No.
of Fatalities |
Frequency (/year) |
No.
of Fatalities |
Frequency (/year) |
No.
of Fatalities |
Frequency (/year) |
|
No. Fatalities |
Frequency (/year) |
1 |
2.42E-07 |
1 |
2.13E-06 |
|
1 |
2.12E-06 |
2 |
1.00E-07 |
2 |
1.59E-06 |
|
2 |
1.57E-06 |
3 |
7.59E-08 |
3 |
1.38E-06 |
|
3 |
1.35E-06 |
4 |
2.20E-08 |
4 |
1.18E-06 |
|
4 |
1.16E-06 |
5 |
1.68E-08 |
5 |
1.02E-06 |
|
5 |
9.83E-07 |
6 |
1.60E-08 |
6 |
8.42E-07 |
|
6 |
8.11E-07 |
8 |
1.55E-08 |
8 |
5.95E-07 |
|
8 |
5.90E-07 |
10 |
2.25E-09 |
10 |
5.07E-07 |
|
10 |
4.97E-07 |
12 |
2.00E-09 |
12 |
3.98E-07 |
|
12 |
3.80E-07 |
15 |
1.67E-09 |
15 |
3.43E-07 |
|
15 |
3.40E-07 |
20 |
1.15E-09 |
20 |
2.96E-07 |
|
20 |
2.61E-07 |
25 |
8.74E-10 |
25 |
2.56E-07 |
|
25 |
2.54E-07 |
30 |
7.00E-10 |
30 |
2.31E-07 |
|
30 |
2.27E-07 |
40 |
4.72E-10 |
40 |
2.16E-07 |
|
40 |
2.13E-07 |
50 |
4.28E-10 |
50 |
1.38E-07 |
|
50 |
1.26E-07 |
60 |
4.03E-10 |
60 |
1.13E-07 |
|
60 |
1.10E-07 |
80 |
3.27E-10 |
80 |
6.33E-08 |
|
80 |
6.06E-08 |
100 |
2.69E-10 |
100 |
3.53E-08 |
|
100 |
3.41E-08 |
120 |
1.81E-10 |
120 |
2.24E-08 |
|
120 |
2.14E-08 |
150 |
1.14E-10 |
150 |
1.21E-08 |
|
150 |
1.12E-08 |
|
|
200 |
6.39E-09 |
|
200 |
5.73E-09 |
|
|
250 |
3.76E-09 |
|
250 |
3.37E-09 |
|
|
300 |
2.64E-09 |
|
300 |
2.40E-09 |
|
|
400 |
1.39E-09 |
|
400 |
1.30E-09 |
|
|
500 |
9.69E-10 |
|
500 |
9.35E-10 |
|
|
600 |
7.29E-10 |
|
600 |
7.03E-10 |
|
|
800 |
3.82E-10 |
|
800 |
3.82E-10 |
|
|
1000 |
1.89E-10 |
|
1000 |
1.89E-10 |
|
|
1200 |
1.57E-10 |
|
Sinopec Petrol-cum-LPG Filling Station |
|||||
|
Case 3 – 2018 without Project workers |
Case 4 – 2018 with Project workers only |
Case 5 – 2018 with Project workers and all offsite population |
|||
|
No.
of Fatalities |
Frequency (/year) |
No.
of Fatalities |
Frequency (/year) |
No.
of Fatalities |
Frequency (/year) |
|
1 |
2.52E-06 |
|
|
1 |
2.52E-06 |
|
2 |
1.20E-06 |
|
|
2 |
1.20E-06 |
|
3 |
8.47E-07 |
|
|
3 |
8.47E-07 |
|
4 |
5.56E-07 |
|
|
4 |
5.56E-07 |
|
5 |
4.39E-07 |
|
|
5 |
4.39E-07 |
|
6 |
3.79E-07 |
|
|
6 |
3.79E-07 |
|
8 |
1.63E-07 |
|
|
8 |
1.63E-07 |
|
10 |
1.01E-07 |
|
|
10 |
1.01E-07 |
|
12 |
9.67E-08 |
|
|
12 |
9.67E-08 |
|
15 |
8.42E-08 |
|
|
15 |
8.42E-08 |
|
20 |
7.94E-08 |
|
|
20 |
7.94E-08 |
|
25 |
4.24E-08 |
|
|
25 |
4.24E-08 |
|
30 |
3.51E-08 |
|
|
30 |
3.51E-08 |
|
40 |
2.13E-08 |
|
|
40 |
2.13E-08 |
|
50 |
1.65E-08 |
|
|
50 |
1.65E-08 |
|
60 |
9.12E-09 |
|
|
60 |
9.12E-09 |
|
80 |
3.56E-09 |
|
|
80 |
3.56E-09 |
|
100 |
9.57E-10 |
|
|
100 |
9.57E-10 |
|
120 |
2.16E-10 |
|
|
120 |
2.16E-10 |
Note: Case 4 F-N data is not available for Sinopec Petrol-cum-LPG Filling Station and F-N data below 1×10-9/year is not shown in the figures of F-N curves
|
Sinopec HK Oil Terminal Chai Wan |
|||||
|
Case 3 – 2018 without Project workers |
Case 4 – 2018 with Project workers only |
Case 5 – 2018 with Project workers and all offsite population |
|||
|
No.
of Fatalities |
Frequency (/year) |
No.
of Fatalities |
Frequency (/year) |
No.
of Fatalities |
Frequency (/year) |
|
1 |
9.88E-07 |
|
|
1 |
9.88E-07 |
|
2 |
2.20E-07 |
|
|
2 |
2.20E-07 |
|
3 |
2.15E-07 |
|
|
3 |
2.15E-07 |
|
4 |
1.50E-07 |
|
|
4 |
1.50E-07 |
|
5 |
1.45E-07 |
|
|
5 |
1.45E-07 |
|
6 |
1.20E-07 |
|
|
6 |
1.20E-07 |
|
8 |
1.10E-07 |
|
|
8 |
1.10E-07 |
|
10 |
1.05E-07 |
|
|
10 |
1.05E-07 |
|
12 |
8.60E-08 |
|
|
12 |
8.60E-08 |
|
15 |
7.00E-08 |
|
|
15 |
7.00E-08 |
|
20 |
7.00E-08 |
|
|
20 |
7.00E-08 |
|
25 |
2.00E-08 |
|
|
25 |
2.00E-08 |
|
30 |
2.00E-08 |
|
|
30 |
2.00E-08 |
Note: Case 4 F-N data is not available for Sinopec HK Oil Terminal Chai Wan
|
LPG Wagon Parking Site |
|||||
|
Case 3 – 2018 without Project workers |
Case 4 – 2018 with Project workers only |
Case 5 – 2018 with Project workers and all offsite population |
|||
|
No.
of Fatalities |
Frequency (/year) |
No.
of Fatalities |
Frequency (/year) |
No.
of Fatalities |
Frequency (/year) |
|
1 |
2.50E-07 |
|
|
1 |
2.50E-07 |
|
2 |
2.58E-08 |
|
|
2 |
2.58E-08 |
|
3 |
4.04E-09 |
|
|
3 |
4.04E-09 |
|
4 |
1.40E-09 |
|
|
4 |
1.40E-09 |
|
5 |
1.15E-09 |
|
|
5 |
1.15E-09 |
|
6 |
6.00E-10 |
|
|
6 |
6.00E-10 |
Note: Case 4 F-N data is not available for LPG Wagon Parking Site and F-N data below 1×10-9/year is not shown in the figures of F-N curves
|
NWFB Depot |
|||||
|
Case 3 – 2018 without Project workers |
Case 4 – 2018 with Project workers only |
Case 5 – 2018 with Project workers and all offsite population |
|||
|
No.
of Fatalities |
Frequency (/year) |
No.
of Fatalities |
Frequency (/year) |
No.
of Fatalities |
Frequency (/year) |
|
1 |
1.18E-08 |
1 |
6.25E-09 |
1 |
1.23E-08 |
|
2 |
1.05E-08 |
2 |
4.22E-09 |
2 |
1.14E-08 |
|
3 |
9.16E-09 |
3 |
3.01E-09 |
3 |
1.02E-08 |
|
4 |
7.65E-09 |
4 |
2.22E-09 |
4 |
8.65E-09 |
|
5 |
6.41E-09 |
5 |
1.71E-09 |
5 |
7.40E-09 |
|
6 |
5.81E-09 |
6 |
1.43E-09 |
6 |
6.96E-09 |
|
8 |
4.49E-09 |
8 |
1.24E-09 |
8 |
6.20E-09 |
|
10 |
3.39E-09 |
10 |
1.05E-09 |
10 |
5.29E-09 |
|
12 |
2.95E-09 |
12 |
8.45E-10 |
12 |
4.38E-09 |
|
15 |
2.73E-09 |
15 |
6.06E-10 |
15 |
3.50E-09 |
|
20 |
2.37E-09 |
20 |
3.62E-10 |
20 |
3.12E-09 |
|
25 |
2.10E-09 |
25 |
2.66E-10 |
25 |
2.78E-09 |
|
30 |
1.97E-09 |
30 |
2.59E-10 |
30 |
2.49E-09 |
|
40 |
8.69E-10 |
40 |
1.26E-10 |
40 |
1.06E-09 |
|
50 |
7.23E-10 |
|
|
50 |
8.26E-10 |
|
60 |
2.38E-10 |
|
|
60 |
3.25E-10 |
|
80 |
1.59E-10 |
|
|
80 |
2.07E-10 |
|
100 |
1.59E-10 |
|
|
100 |
1.95E-10 |
|
120 |
1.57E-10 |
|
|
120 |
1.87E-10 |
|
150 |
1.27E-10 |
|
|
150 |
1.30E-10 |
Note: F-N data below 1×10-9/year is not shown in the figures of F-N curves
|
Citybus Depot |
|||||
|
Case 3 – 2018 without Project workers |
Case 4 – 2018 with Project workers only |
Case 5 – 2018 with Project workers and all offsite population |
|||
|
No.
of Fatalities |
Frequency (/year) |
No.
of Fatalities |
Frequency (/year) |
No.
of Fatalities |
Frequency (/year) |
|
1 |
7.23E-09 |
1 |
1.12E-10 |
1 |
7.23E-09 |
|
2 |
6.29E-09 |
|
|
2 |
6.30E-09 |
|
3 |
5.33E-09 |
|
|
3 |
5.34E-09 |
|
4 |
4.78E-09 |
|
|
4 |
4.78E-09 |
|
5 |
4.59E-09 |
|
|
5 |
4.59E-09 |
|
6 |
4.49E-09 |
|
|
6 |
4.49E-09 |
|
8 |
4.31E-09 |
|
|
8 |
4.31E-09 |
|
10 |
4.13E-09 |
|
|
10 |
4.14E-09 |
|
12 |
3.94E-09 |
|
|
12 |
3.94E-09 |
|
15 |
3.69E-09 |
|
|
15 |
3.70E-09 |
|
20 |
3.22E-09 |
|
|
20 |
3.22E-09 |
|
25 |
2.19E-09 |
|
|
25 |
2.20E-09 |
|
30 |
1.84E-09 |
|
|
30 |
1.84E-09 |
|
40 |
1.48E-09 |
|
|
40 |
1.48E-09 |
|
50 |
1.06E-09 |
|
|
50 |
1.06E-09 |
|
60 |
7.21E-10 |
|
|
60 |
7.26E-10 |
|
80 |
4.94E-10 |
|
|
80 |
4.94E-10 |
|
100 |
4.76E-10 |
|
|
100 |
4.76E-10 |
|
120 |
4.32E-10 |
|
|
120 |
4.32E-10 |
|
150 |
3.96E-10 |
|
|
150 |
3.96E-10 |
|
200 |
3.95E-10 |
|
|
200 |
3.95E-10 |
|
250 |
3.86E-10 |
|
|
250 |
3.86E-10 |
|
300 |
2.54E-10 |
|
|
300 |
2.54E-10 |
|
400 |
1.75E-10 |
|
|
400 |
1.75E-10 |
Note: F-N data below 1×10-9/year is not shown in the figures of F-N curves
|
Cumulative Risk |
|||
|
Case 3 – 2018 without Project workers |
Case 5 – 2018 with Project workers and all offsite population |
||
|
No.
of Fatalities |
Frequency (/year) |
No. of Fatalities |
Frequency (/year) |
|
1 |
5.90E-06 |
1 |
5.91E-06 |
|
2 |
3.03E-06 |
2 |
3.05E-06 |
|
3 |
2.43E-06 |
3 |
2.47E-06 |
|
4 |
1.88E-06 |
4 |
1.90E-06 |
|
5 |
1.58E-06 |
5 |
1.62E-06 |
|
6 |
1.32E-06 |
6 |
1.35E-06 |
|
8 |
8.71E-07 |
8 |
8.78E-07 |
|
10 |
7.10E-07 |
10 |
7.22E-07 |
|
12 |
5.69E-07 |
12 |
5.89E-07 |
|
15 |
5.01E-07 |
15 |
5.04E-07 |
|
20 |
4.16E-07 |
20 |
4.52E-07 |
|
25 |
3.20E-07 |
25 |
3.24E-07 |
|
30 |
2.86E-07 |
30 |
2.90E-07 |
|
40 |
2.36E-07 |
40 |
2.40E-07 |
|
50 |
1.44E-07 |
50 |
1.56E-07 |
|
60 |
1.20E-07 |
60 |
1.23E-07 |
|
80 |
6.48E-08 |
80 |
6.75E-08 |
|
100 |
3.57E-08 |
100 |
3.69E-08 |
|
120 |
2.22E-08 |
120 |
2.33E-08 |
|
150 |
1.18E-08 |
150 |
1.27E-08 |
|
200 |
6.18E-09 |
200 |
6.83E-09 |
|
250 |
3.76E-09 |
250 |
4.15E-09 |
|
300 |
2.66E-09 |
300 |
2.90E-09 |
|
400 |
1.47E-09 |
400 |
1.56E-09 |
|
500 |
9.95E-10 |
500 |
1.03E-09 |
|
600 |
7.43E-10 |
600 |
7.69E-10 |
|
800 |
4.21E-10 |
800 |
4.21E-10 |
|
1000 |
2.13E-10 |
1000 |
2.13E-10 |
|
1200 |
1.76E-10 |
1200 |
1.76E-10 |
Note: The F-N data below 1×10-9/year is not shown in the figures of F-N curves
Table 10.43 PLL
Values for Construction Phase
|
ExxonMobil Petrol-cum-LPG Filling Station |
|||||
|
Case 1 – 2016 without construction workers |
Case 2 – 2016 with construction workers |
||||
|
Event |
PLL
(/year) |
% of total PLL |
Event |
PLL
(/year) |
% of total PLL |
|
Cold catastrophic and partial failure of
LPG road tanker (50% inventory) |
7.56E-06 |
30.5% |
Cold catastrophic and partial failure of LPG road tanker (50% inventory) |
8.11E-06 |
30.6% |
|
Cold catastrophic and partial failure of LPG road tanker (full inventory) |
4.47E-06 |
18.0% |
Cold catastrophic and partial failure of LPG road tanker (full inventory) |
4.84E-06 |
18.3% |
|
Guillotine failure of liquid-inlet
pipework (50% inventory) |
4.46E-06 |
18.0% |
Guillotine failure of liquid-inlet pipework (50% inventory) |
4.77E-06 |
18.0% |
|
Cold catastrophic and partial failure of
LPG storage vessel (60% inventory) |
2.78E-06 |
11.2% |
Cold catastrophic and partial failure of LPG storage vessel (60% inventory) |
2.94E-06 |
11.1% |
|
Guillotine failure of liquid-inlet
pipework (full inventory) |
2.10E-06 |
8.5% |
Guillotine failure of liquid-inlet pipework (full inventory) |
2.27E-06 |
8.6% |
|
Others |
3.40E-06 |
13.7% |
Others |
3.57E-06 |
13.5% |
|
Total |
2.48E-05 |
100% |
Total |
2.65E-05 |
100% |
|
Sinopec Petrol-cum-LPG Filling Station |
|||||
|
Case 1 – 2016 without construction workers |
Case 2 – 2016 with construction workers |
||||
|
Event |
PLL
(/year) |
% of total PLL |
Event |
PLL
(/year) |
% of total PLL |
|
Guillotine failure of liquid-inlet
pipework (50% inventory) |
2.97E-06 |
34.6% |
Guillotine failure of liquid-inlet pipework (50% inventory) |
2.97E-06 |
34.6% |
|
Cold catastrophic and partial failure of LPG road tanker (50% inventory) |
1.84E-06 |
21.4% |
Cold catastrophic and partial failure of LPG road tanker (50% inventory) |
1.84E-06 |
21.4% |
|
Cold catastrophic and partial failure of
LPG road tanker (full inventory) |
1.01E-06 |
11.7% |
Cold catastrophic and partial failure of LPG road tanker (full inventory) |
1.01E-06 |
11.7% |
|
Guillotine failure of liquid-inlet
pipework (full inventory) |
9.57E-07 |
11.1% |
Guillotine failure of liquid-inlet pipework (full inventory) |
9.57E-07 |
11.1% |
|
Cold catastrophic and partial failure of
LPG storage vessel (60% inventory) |
5.46E-07 |
6.4% |
Cold catastrophic and partial failure of LPG storage vessel (60% inventory) |
5.46E-07 |
6.4% |
|
Others |
1.26E-06 |
14.7% |
Others |
1.26E-06 |
14.3% |
|
Total |
8.59E-06 |
100% |
Total |
8.59E-06 |
100% |
|
Sinopec HK Oil Terminal Chai Wan |
|||||
|
Case 1 – 2016 without construction workers |
Case 2 – 2016 with construction workers |
||||
|
Event |
PLL
(/year) |
% of total PLL |
Event |
PLL
(/year) |
% of total PLL |
|
Cold catastrophic failure of marine
vessel fuel tank due to striking |
2.06E-06 |
59.5% |
Cold catastrophic failure of marine vessel fuel tank due to striking |
2.06E-06 |
59.5% |
|
Instantaneous single fuel tank failure |
6.36E-07 |
18.4% |
Instantaneous single fuel tank failure |
6.36E-07 |
18.4% |
|
Rupture of jetty loading hose during marine vessel unloading |
5.39E-07 |
15.6% |
Rupture of jetty loading hose during marine vessel unloading |
5.39E-07 |
15.6% |
|
Cold catastrophic failure, cold partial
failure and BLEVE of LPG cylinder |
2.00E-07 |
5.8% |
Cold catastrophic failure, cold partial failure and BLEVE of LPG cylinder |
2.00E-07 |
5.8% |
|
Full bore rupture of fixed pipeline from fuel oil tank to jetty |
2.68E-08 |
0.8% |
Full bore rupture of fixed pipeline from fuel oil tank to jetty |
2.68E-08 |
0.8% |
|
Others |
2.47E-09 |
0.1% |
Others |
2.47E-09 |
0.1% |
|
Total |
3.47E-06 |
100% |
Total |
3.47E-06 |
100% |
|
LPG Wagon Parking Site |
|||||
|
Case 1 – 2016 without construction workers |
Case 2 – 2016 with construction workers |
||||
|
Event |
PLL
(/year) |
% of total PLL |
Event |
PLL
(/year) |
% of total PLL |
|
Cold catastrophic failure, cold partial failure
and BLEVE of LPG cylinder |
2.65E-07 |
93.7% |
Cold catastrophic failure, cold partial failure and BLEVE of LPG cylinder |
2.65E-07 |
93.7% |
|
Cold catastrophic failure and cold partial failure of kerosene tanker |
1.78E-08 |
6.3% |
Cold catastrophic failure and cold partial failure of kerosene tanker |
1.78E-08 |
6.3% |
|
Total |
6.25E-07 |
100% |
Total |
6.25E-07 |
100% |
|
NWFB Depot |
|||||
|
Case 1 – 2016 without construction workers |
Case 2 – 2016 with construction workers |
||||
|
Event |
PLL
(/year) |
% of total PLL |
Event |
PLL
(/year) |
% of total PLL |
|
Smoke from pool fire of diesel tanker
rupture and leak |
1.17E-07 |
75.2% |
Smoke from pool fire of diesel tanker rupture and leak |
1.32E-07 |
77.4% |
|
Diesel tanker rupture and leak |
3.85E-08 |
24.8% |
Diesel tanker rupture and leak |
3.85E-08 |
22.6% |
|
Total |
1.55E-07 |
100% |
Total |
1.70E-07 |
100% |
|
Citybus Depot |
|||||
|
Case 1 – 2016 without construction workers |
Case 2 – 2016 with construction workers |
||||
|
Event |
PLL
(/year) |
% of total PLL |
Event |
PLL
(/year) |
% of total PLL |
|
Smoke from pool fire of diesel tanker
rupture and leak |
2.90E-07 |
95.9% |
Smoke from pool fire of diesel tanker rupture and leak |
2.90E-07 |
95.9% |
|
Diesel tanker rupture and leak |
1.23E-08 |
4.1% |
Diesel tanker rupture and leak |
0.00E+00 |
0.0% |
|
Total |
3.02E-07 |
100% |
Total |
3.02E-07 |
100% |
Table 10.44 PLL Values for Operation phase
|
ExxonMobil Petrol-cum-LPG Filling Station |
||||||||
|
Case 3 – 2018 without Project workers |
Case 4 – 2018 with Project workers only |
Case 5 – 2018 with Project workers and all offsite population |
||||||
|
Event |
PLL
(/year) |
% of total PLL |
Event |
PLL
(/year) |
% of total PLL |
Event |
PLL
(/year) |
% of total PLL |
|
Cold catastrophic and partial failure of
LPG road tanker (50% inventory) |
7.87E-06 |
29.9% |
Cold catastrophic and partial failure of LPG road tanker (50% inventory) |
2.55E-07 |
43.4% |
Cold catastrophic and partial failure of LPG road tanker (50% inventory) |
8.20E-06 |
30.1% |
|
Cold catastrophic and partial failure of
LPG road tanker (full inventory) |
4.80E-06 |
18.3% |
Cold catastrophic and partial failure of LPG road tanker (full inventory) |
1.88E-07 |
32.1% |
Cold catastrophic and partial failure of LPG road tanker (full inventory) |
5.03E-06 |
18.5% |
|
Guillotine failure of liquid-inlet
pipework (50% inventory) |
4.58E-06 |
17.4% |
Guillotine failure of liquid-inlet pipework (50% inventory) |
6.22E-08 |
10.6% |
Guillotine failure of liquid-inlet pipework (50% inventory) |
4.75E-06 |
17.4% |
|
Cold catastrophic and partial failure of
LPG storage vessel (60% inventory) |
3.12E-06 |
11.9% |
Guillotine failure of liquid-inlet pipework (full inventory) |
3.33E-08 |
5.7% |
Cold catastrophic and partial failure of LPG storage vessel (60% inventory) |
3.20E-06 |
11.7% |
|
Guillotine failure of liquid-inlet
pipework (full inventory) |
2.12E-06 |
8.1% |
Cold catastrophic and partial failure of LPG storage vessel (60% inventory) |
2.25E-08 |
3.8% |
Guillotine failure of liquid-inlet pipework (full inventory) |
2.21E-06 |
8.1% |
|
Others |
3.78E-06 |
14.4% |
Others |
2.57E-08 |
4.4% |
Others |
3.86E-06 |
14.2% |
|
Total |
2.63E-05 |
100% |
|
5.86E-07 |
100% |
|
2.73E-05 |
100% |
|
Sinopec Petrol-cum-LPG Filling Station |
||||||||
|
Case 3 – 2018 without Project workers |
Case 4 – 2018 with Project workers only |
Case 5 – 2018 with Project workers and all offsite population |
||||||
|
Event |
PLL
(/year) |
% of total PLL |
Event |
PLL
(/year) |
% of total PLL |
Event |
PLL
(/year) |
% of total PLL |
|
Guillotine failure of liquid-inlet
pipework (50% inventory) |
2.99E-06 |
34.2% |
Cold catastrophic and partial failure of LPG storage vessel (full inventory) |
3.18E-12 |
52.1% |
Guillotine failure of liquid-inlet pipework (50% inventory) |
2.99E-06 |
34.2% |
|
Cold catastrophic and partial failure of
LPG road tanker (50% inventory) |
1.86E-06 |
21.3% |
Cold catastrophic and partial failure of LPG road tanker (full inventory) |
1.63E-12 |
26.7% |
Cold catastrophic and partial failure of LPG road tanker (50% inventory) |
1.86E-06 |
21.3% |
|
Cold catastrophic and partial failure of
LPG road tanker (full inventory) |
1.05E-06 |
12.0% |
Cold catastrophic and partial failure of LPG storage vessel during unloading (full inventory) |
1.08E-12 |
17.7% |
Cold catastrophic and partial failure of LPG road tanker (full inventory) |
1.05E-06 |
12.0% |
|
Guillotine failure of liquid-inlet
pipework (full inventory) |
9.61E-07 |
11.0% |
Cold catastrophic and partial failure of LPG storage vessel (60% inventory) |
1.61E-13 |
2.6% |
Guillotine failure of liquid-inlet pipework (full inventory) |
9.61E-07 |
11.0% |
|
Cold catastrophic and partial failure of
LPG storage vessel (60% inventory) |
5.69E-07 |
6.5% |
Cold catastrophic and partial failure of LPG storage vessel during unloading (60% inventory) |
5.49E-14 |
0.9% |
Cold catastrophic and partial failure of LPG storage vessel (60% inventory) |
5.69E-07 |
6.5% |
|
Others |
1.31E-06 |
14.6% |
Others |
4.26E-15 |
0.1% |
Others |
1.31E-06 |
14.6% |
|
Total |
8.74E-06 |
100% |
|
6.10E-12 |
100% |
|
8.74E-06 |
100% |
|
Sinopec HK Oil Terminal Chai Wan |
||||||||
|
Case 3 – 2018 without Project workers |
Case 4 – 2018 with Project workers only |
Case 5 – 2018 with Project workers and all offsite population |
||||||
|
Event |
PLL
(/year) |
% of total PLL |
Event |
PLL
(/year) |
% of total PLL |
Event |
PLL
(/year) |
% of total PLL |
|
Cold catastrophic failure of marine
vessel fuel tank due to striking |
2.11E-06 |
60.1% |
|
|
|
Cold catastrophic failure of marine vessel fuel tank due to striking |
2.11E-06 |
60.1% |
|
Instantaneous single fuel tank failure |
6.37E-07 |
18.1% |
|
|
|
Instantaneous single fuel tank failure |
6.37E-07 |
18.1% |
|
Rupture of jetty loading hose during marine vessel unloading |
5.39E-07 |
15.3% |
|
|
|
Rupture of jetty loading hose during marine vessel unloading |
5.39E-07 |
15.3% |
|
Cold catastrophic failure, cold partial
failure and BLEVE of LPG cylinder |
2.00E-07 |
5.7% |
|
|
|
Cold catastrophic failure, cold partial failure and BLEVE of LPG cylinder |
2.00E-07 |
5.7% |
|
Full bore rupture of fixed pipeline from fuel oil tank to jetty |
2.68E-08 |
0.8% |
|
|
|
Full bore rupture of fixed pipeline from fuel oil to jetty |
2.68E-08 |
0.8% |
|
Others |
2.48E-09 |
0.1% |
|
|
|
Others |
2.48E-09 |
0.1% |
|
Total |
3.52E-06 |
100% |
|
|
|
Total |
3.52E-06 |
100% |
Note: PLL data is not available for Case 4 of Sinopec HK Oil Terminal Chai Wan
|
LPG Wagon Parking Site |
||||||||
|
Case 3 – 2018 without Project workers |
Case 4 – 2018 with Project workers only |
Case 5 – 2018 with Project workers and all offsite population |
||||||
|
Event |
PLL
(/year) |
% of total PLL |
Event |
PLL
(/year) |
% of total PLL |
Event |
PLL
(/year) |
% of total PLL |
|
Cold catastrophic failure, cold partial
failure and BLEVE of LPG cylinder |
2.65E-07 |
93.7% |
|
|
|
Cold catastrophic failure, cold partial failure and BLEVE of LPG cylinder |
2.65E-07 |
93.7% |
|
Cold catastrophic failure and cold
partial failure of petrol tanker |
1.78E-08 |
6.3% |
|
|
|
Cold catastrophic failure and cold partial failure of petrol tanker |
1.78E-08 |
6.3% |
|
Total |
2.83E-07 |
100% |
|
|
|
Total |
2.83E-07 |
100% |
Note: PLL data is not available for Case 4 of LPG Wagon Parking Site
|
NWFB Depot |
||||||||
|
Case 3 – 2018 without Project workers |
Case 4 – 2018 with Project workers only |
Case 5 – 2018 with Project workers and all offsite population |
||||||
|
Event |
PLL
(/year) |
% of total PLL |
Event |
PLL
(/year) |
% of total PLL |
Event |
PLL
(/year) |
% of total PLL |
|
Smoke from pool fire of diesel tanker
rupture and leak |
1.21E-07 |
75.0% |
Smoke from pool fire of diesel tanker rupture and leak |
3.77E-08 |
100.0% |
Smoke from pool fire of diesel tanker rupture and leak |
1.59E-07 |
79.7% |
|
Diesel tanker rupture and leak |
4.04E-08 |
25.0% |
Diesel tanker rupture and leak |
4.15E-13 |
0.0% |
Diesel tanker rupture and leak |
4.04E-08 |
20.3% |
|
Total |
1.61E-07 |
100% |
Total |
3.77E-08 |
100% |
Total |
1.99E-07 |
100% |
|
Citybus Depot |
||||||||
|
Case 3 – 2018 without Project workers |
Case 4 – 2018 with Project workers only |
Case 5 – 2018 with Project workers and all offsite population |
||||||
|
Event |
PLL
(/year) |
% of total PLL |
Event |
PLL
(/year) |
% of total PLL |
Event |
PLL
(/year) |
% of total PLL |
|
Smoke from pool fire of diesel tanker
rupture and leak |
2.94E-07 |
95.8% |
Smoke from pool fire of diesel tanker rupture and leak |
1.87E-10 |
100.0% |
Smoke from pool fire of diesel tanker rupture and leak |
2.94E-07 |
95.8% |
|
Diesel tanker rupture and leak |
1.29E-08 |
4.2% |
Diesel tanker rupture and leak |
0.00E+00 |
0.0% |
Diesel tanker rupture and leak |
1.29E-08 |
4.2% |
|
Total |
3.07E-07 |
100% |
Total |
1.87E-10 |
100% |
Total |
3.07E-07 |
100% |
|
Cumulative Risk |
||||||||
|
Case 3 – 2018 without Project workers |
Case 4 – 2018 with Project workers only |
Case 5 – 2018 with Project workers and all offsite population |
||||||
|
Event |
PLL
(/year) |
% of total PLL |
Event |
PLL
(/year) |
% of total PLL |
Event |
PLL
(/year) |
% of total PLL |
|
ExxonMobil – Cold catastrophic and
partial failure of LPG road tanker (50% inventory) |
7.87E-06 |
20.0% |
ExxonMobil – Cold catastrophic and partial failure of LPG road tanker (50% inventory) |
2.55E-07 |
40.8% |
ExxonMobil – Cold catastrophic and partial failure of LPG road tanker (50% inventory) |
8.20E-06 |
20.3% |
|
ExxonMobil – Cold catastrophic and
partial failure of LPG road tanker (full inventory) |
4.80E-06 |
12.2% |
ExxonMobil – Cold catastrophic and partial failure of LPG road tanker (full inventory) |
1.88E-07 |
30.1% |
ExxonMobil – Cold catastrophic and partial failure of LPG road tanker (full inventory) |
5.03E-06 |
12.5% |
|
ExxonMobil – Guillotine failure of liquid-inlet pipework (50% inventory) |
4.58E-06 |
11.7% |
ExxonMobil – Guillotine failure of liquid-inlet pipework (50% inventory) |
6.22E-08 |
10.0% |
ExxonMobil – Guillotine failure of liquid-inlet pipework (50% inventory) |
4.75E-06 |
11.8% |
|
ExxonMobil – Cold catastrophic and partial failure of LPG storage vessel (60% inventory) |
3.12E-06 |
7.9% |
ExxonMobil – Guillotine failure of liquid-inlet pipework (full inventory) |
3.77E-08 |
6.0% |
ExxonMobil – Cold catastrophic and partial failure of LPG storage vessel (60% inventory) |
3.20E-06 |
7.9% |
|
Sinopec – Guillotine failure of
liquid-inlet pipework (50% inventory) |
2.99E-06 |
7.6% |
ExxonMobil – Cold catastrophic and partial failure of LPG storage vessel during unloading (60% inventory) |
3.33E-08 |
5.3% |
Sinopec – Guillotine failure of liquid-inlet pipework (50% inventory) |
2.99E-06 |
7.4% |
|
Others |
1.59E-05 |
40.5% |
Others |
4.84E-08 |
7.7% |
Others |
1.61E-05 |
40.0% |
|
Total |
3.93E-05 |
100% |
|
6.24E-07 |
100% |
|
4.03E-05 |
100% |
10.8.1. The Hazard to Life study is based on a number of assumptions that are mentioned in other sections of this report. The uncertainties of the results are discussed below.
10.8.2. There is currently no activity in the Sinopec HK Oil Terminal Chai Wan that the operation is stopped, a full operation of oil terminal was however assumed in the study. The hazardous scenarios associated with jetty unloading operation, fuel oil storage, dye marker storage, LPG cylinder storage and fuel oil bulk filling were taken into account in calculating the societal risks, which is a conservative assumption.
10.8.3. LPG wagons and kerosene road tanker that are parked inside the LPG Wagon Parking Site are normally driven in and out of the site for distributing LPG cylinder and kerosene. Presences of the LPG wagons and kerosene road tanker were assumed to be full time onsite, which gave conservative results.
10.8.4. Without information from Sinopec (Hong Kong) Ltd, the LPG cylinder inventory of the Sinopec HK Oil Terminal Chai Wan was set at its maximum capacity at all time. The inventory is expected to be less and varies from time to time.
10.8.5. A 5000DWT marine vessel was assumed when analysing the rupture of its loaded cargo tank. The size of marine vessel berthing the Sinopec HK Oil Terminal Chai Wan is however expected to be smaller as the size is restricted by the maximum draft and overall length (LOA) of the berth. The inventory that was involved in the loss of containment and therefore the pool diameter formed could be over-estimated.
10.8.6. In the event of the potential rupture of a fuel oil storage tank in the Sinopec HK Oil Terminal Chai Wan that results in an overflow of fuel oil outside the site, part of the spilled fuel oil could be contained and directed to the drainage system and oil interceptor when considering the surface gradient outside the oil terminal (which falls towards the oil terminal). In the study, the fuel oil was considered spilled out to the public road without taking in account the drainage system and oil interceptor. The estimated risk should be conservative and the actual risk is therefore lower.
10.8.7. In view of applying varies conservative assumptions in the study that has minimised the uncertainty, it is safe to assume the risk level assessed would comply with the risk guideline of Section 2 of Annex 4 of the EIAO-TM.
10.9.1. The effect of increasing 50 workers in both construction phase and operation phase are illustrated in Figure 10.42 and Figure 10.43 respectively. The comparison of F-N data is shown in Table 10.45 below
10.9.2. With an additional 50 construction workers (hence 150 construction workers, increased by 50%) in the construction phase, the risk level is slightly increased in the N = 20 to 300 region. The total PLL is increased by 2.6%
10.9.3. With an additional 50 Project workers (hence 290 Project workers, increased by 21%) in the operation phase, the risk level is slightly increased in the N = 40 to 500 region. The total PLL is increased by 0.56%.
Table 10.45 FN Data and PLL for Sensitive Analysis
|
Cumulative Risk |
|||||
|
|
Case 2 – 2016 with construction workers |
Case 2 with 50 more construction workers |
|
Case 5 – 2018 with Project workers and all offsite population |
Case 5 with 50 more Project workers |
|
No. of Fatalities |
Frequency (/year) |
Frequency (/year) |
No. of Fatalities |
Frequency (/year) |
Frequency (/year) |
|
1 |
5.92E-06 |
5.94E-06 |
1 |
5.91E-06 |
5.91E-06 |
|
2 |
3.07E-06 |
3.09E-06 |
2 |
3.05E-06 |
3.05E-06 |
|
3 |
2.50E-06 |
2.53E-06 |
3 |
2.47E-06 |
2.47E-06 |
|
4 |
1.92E-06 |
1.95E-06 |
4 |
1.90E-06 |
1.91E-06 |
|
5 |
1.63E-06 |
1.67E-06 |
5 |
1.62E-06 |
1.62E-06 |
|
6 |
1.34E-06 |
1.35E-06 |
6 |
1.35E-06 |
1.36E-06 |
|
8 |
8.81E-07 |
8.86E-07 |
8 |
8.78E-07 |
8.78E-07 |
|
10 |
7.28E-07 |
7.37E-07 |
10 |
7.22E-07 |
7.25E-07 |
|
12 |
5.94E-07 |
6.09E-07 |
12 |
5.89E-07 |
5.94E-07 |
|
15 |
5.08E-07 |
5.20E-07 |
15 |
5.04E-07 |
5.06E-07 |
|
20 |
4.35E-07 |
4.55E-07 |
20 |
4.52E-07 |
4.61E-07 |
|
25 |
3.20E-07 |
3.23E-07 |
25 |
3.24E-07 |
3.24E-07 |
|
30 |
2.88E-07 |
2.92E-07 |
30 |
2.90E-07 |
2.91E-07 |
|
40 |
2.33E-07 |
2.37E-07 |
40 |
2.40E-07 |
2.41E-07 |
|
50 |
1.89E-07 |
1.94E-07 |
50 |
1.56E-07 |
1.88E-07 |
|
60 |
1.08E-07 |
1.12E-07 |
60 |
1.23E-07 |
1.23E-07 |
|
80 |
6.77E-08 |
7.11E-08 |
80 |
6.75E-08 |
7.98E-08 |
|
100 |
2.88E-08 |
3.18E-08 |
100 |
3.69E-08 |
3.71E-08 |
|
120 |
1.72E-08 |
1.99E-08 |
120 |
2.33E-08 |
2.34E-08 |
|
150 |
9.81E-09 |
1.11E-08 |
150 |
1.27E-08 |
1.29E-08 |
|
200 |
5.49E-09 |
5.83E-09 |
200 |
6.83E-09 |
7.03E-09 |
|
250 |
3.50E-09 |
3.65E-09 |
250 |
4.15E-09 |
4.33E-09 |
|
300 |
2.22E-09 |
2.27E-09 |
300 |
2.90E-09 |
3.00E-09 |
|
400 |
1.34E-09 |
1.36E-09 |
400 |
1.56E-09 |
1.67E-09 |
|
500 |
9.09E-10 |
9.19E-10 |
500 |
1.03E-09 |
1.04E-09 |
|
600 |
5.98E-10 |
6.01E-10 |
600 |
7.69E-10 |
7.79E-10 |
|
800 |
4.10E-10 |
4.10E-10 |
800 |
4.21E-10 |
4.21E-10 |
|
1000 |
2.06E-10 |
2.06E-10 |
1000 |
2.13E-10 |
2.13E-10 |
|
1200 |
1.71E-10 |
1.71E-10 |
1200 |
1.76E-10 |
1.76E-10 |
|
Total PLL (/year) |
3.93E-05 |
4.03E-05 |
Total PLL (/year) |
4.03E-05 |
4.05E-05 |
10.10.1. As mentioned in Section 10.7, the individual and societal risks of the proposed Project and posed by hazardous installations are acceptable, while the increase in population from the proposed Project has negligible effect on the cumulative societal risk. Therefore mitigation measure is not required.
10.11.1. Although mitigation measure is not required, the following good site practices are suggested to be implemented during the construction phase:
· ignition of fire on site should be controlled throughout the construction programme;
· any temporary storage of fuel and flammable chemical should be minimised to reduce chance of causing explosion or escalation of fire in the case of emergency event at nearby potentially hazardous sources;
· fire extinguisher or other firefighting equipment should be made easily accessible to on-site workers; and
· establish communication channel and evacuation plan in the case of emergency event at nearby potentially hazardous sources.
10.11.2. The following good site practices are suggested to be implemented during the operation phase:
· arrangements and facilities for the storage of any flammable goods should be in strict compliance with relevant legislation and guidelines;
· the building should be carefully designed to allow for rapid evacuation of people in protected routes; and
· proper training on safety procedures and evacuation arrangement should be conducted to enhance building users’ capability to handle emergencies. An emergency response plan should be adopted during the operation phase of the depot. The plan should list out emergency procedures, identify members of emergency response teams and summarise contact information of nearby potentially hazardous sources.
10.12.1. As mentioned in Section 10.4 and 10.7, the proposed Project did not pose off-site risk and the individual and societal risks posed by the hazardous installations are acceptable, while the increase in population from the proposed Project has negligible effect on the cumulative societal risk. Therefore specific Environmental Monitoring for hazard to life is not required in general. However, good site practices recommended in Section 10.11 should be implemented and checked in routine site inspections and regular audits.
10.13.1. The risk to construction workers of the proposed Project complies with Section 2 of Annex 4 of the EIAO-TM in terms of both individual risk and societal risk.
10.13.2. It was concluded in the HAZID workshops that the operation of the proposed Project had no failure event that could cause off-site risk and therefore was not considered in the risk summation.
10.13.3. The risks to the on-site Project workers posed by the Sinopec HK Oil Terminal Chai Wan, ExxonMobil Petrol-cum-LPG Filling Station, Sinopec Petrol-cum-LPG Filling Station, NWFB Depot, Citybus Depot and LPG Wagon Parking Site comply with Section 2 of Annex 4 of the EIAO-TM as shown in the F-N curves.
10.13.4. The cumulative risk by taking into account the on-site Project workers together with all off-site population sets due to the proposed Project and all hazardous installations comply with the Section 2 of Annex 4 of the EIAO-TM.
10.13.5. Uncertainty analysis is conducted to conclude the assumptions adopted in the study are conservative. Sensitivity analysis is also performed on the increase in Project population in construction and operation phase. Good site practices are recommended in both construction phase and operation phase of the Project.
[1] Kai Tak Development, Civil Engineering and Development Department, AEIAR-130/2009
[2] Government News Archives, Parking spaces for vehicles carrying cylinder liquefied petroleum gas, 2010, http://www.info.gov.hk/gia/general/201004/28/P201004280178.htm
[3] New World First Bus Services Limited, New World First Bus Permanent Depot at Chai Wan, AEIAR-029/2000
[4] Planning Department, Projection of Population Distribution 2013 –
2021
[5] Harbour Area Treatment Scheme Stage 2A Environmental Impact Assessment, Drainage Services Department, Hong Kong, AEIAR-121/2008
[6] Comprehensive Feasibility Study for the Revised Scheme of South East Kowloon Development EIA, Territory Development Department, Hong Kong, AEIAR-044/ 2001
[7] Central Kowloon Route Environmental Impact Assessment, Highways Department, AEIAR-171/2013
[8] Guidelines for Quantitative Risk Assessment “Purple Book”, CPR18E, Committee for the Prevention of Disasters, 2005
[9] Major Accident Reporting System, eMARS, https://emars.jrc.ec.europa.eu/
[10] Failure and Accidents Technical information System, FACTS, http://www.factsonline.nl/
[11] Permanent Aviation Fuel Facility for Hong Kong International Airport, Airport Authority Hong Kong, AEIAR-107/2007
[12] Providing Sufficient Water Depth for Kwai Tsing Container Basin and its Approach Channel, Civil Engineering and Development Department, AEIAR-156/2010
[13] Permanent Aviation Fuel Facility for Hong Kong International Airport, Airport Authority Hong Kong, AEIAR-107/2007
[14] New World First Bus Services Limited, New World First Bus Permanent Depot at Chai Wan, AEIAR-029/2000
[15] Citybus Limited, Proposed Headquarters and Bus Maintenance Depot in Chai Wan, AEIAR-045/2001
[16] Kai Tak Development, Civil Engineering and Development Department, AEIAR-130/2009
[17] Harbour Area Treatment Scheme Stage 2A Environmental Impact Assessment, Drainage Services Department, Hong Kong, AEIAR-121/2008
[18] South Island Line, MTR Corporation Limited, AEIAR-155/2010
[19] New World First Bus Services Limited, New World First Bus Permanent Depot at Chai Wan, AEIAR-029/2000
[20] Permanent Aviation Fuel Facility for Hong Kong International Airport, Airport Authority Hong Kong, AEIAR-107/2007
[21] Permanent Aviation Fuel Facility for Hong Kong International Airport, Airport Authority Hong Kong, AEIAR-107/2007
[22] Providing Sufficient Water Depth for Kwai Tsing Container Basin and its Approach Channel, Civil Engineering and Development Department, AEIAR-156/2010
[23] Permanent Aviation Fuel Facility for Hong Kong International Airport, Airport Authority Hong Kong, AEIAR-107/2007
[24] Berthing Guidelines, Marine Department, HKSAR, 2014
[25] Manual for Deep Draft Navigation, National Economy Development, U.S. Army Engineer Institute for Water Resources, 2010
[26] New World First Bus Services Limited, New World First Bus Permanent Depot at Chai Wan, AEIAR-029/2000
[27] Permanent Aviation Fuel Facility for Hong Kong International Airport, Airport Authority Hong Kong, AEIAR-107/2007
[28] Liquefied Natural Gas (LNG) Receiving Terminal and Associated Facilities, AEIAR-106/2007
[29] A Guide To Quantitative Risk Assessment for Offshore Installations, DNV 7.0DTechnica, 1999
[30] HSE, SPC/Tech/OSD/30, Methods of approximation and determination of human vulnerability for offshore major accident hazard assessment. 2011
[31] Citybus Limited, Proposed Headquarters and Bus Maintenance Depot in Chai Wan, AEIAR-045/2001
[32] Kai Tak Development, Civil Engineering and Development Department, AEIAR-130/2009
[33] Kai Tak Development, Civil Engineering and Development Department, AEIAR-130/2009
[34] Typical Details of A Petrol Interceptor, ProPECC PN 5/93, Environmental Protection Department
[35] Harbour Area Treatment Scheme Stage 2A Environmental Impact Assessment, Drainage Services Department, Hong Kong, AEIAR-121/2008
[36] South Island Line, MTR Corporation Limited, AEIAR-155/2010
[37] South Island Line, MTR Corporation Limited, AEIAR-155/2010
[38] Hazardous Substance Fact Sheet - State of New Jersey, http://nj.gov/health/eoh/rtkweb/documents/fs/1259.pdf
[39] Material Safety Data Sheet, Gulfpride® Motor Oil 10W-40, Gulf Lubricants
[40] MATERIAL SAFETY DATA SHEET, Battery Fluid Acid, Century Yuasa and Lead Acid Battery Wet, East Penn Manufacturing Co., Inc
[41] WiseNews, http://libwisesearch.wisers.net.lib.ezproxy.ust.hk/
[42] Loss Prevention in the Process Industries, F. Lees, Butterworth-Heinemann, United Kingdom, 2005.
[43] Harbour Area Treatment Scheme Stage 2A Environmental Impact Assessment, Drainage Services Department, Hong Kong, AEIAR-121/2008
[44] South Island Line, MTR Corporation Limited, AEIAR-155/2010
[45] Permanent Aviation Fuel Facility for Hong Kong International Airport, Airport Authority Hong Kong, AEIAR-107/2007
[46] Guidelines for Quantitative Risk Assessment “Purple Book”, CPR18E, Committee for the Prevention of Disasters, 2005
[47] Quantitative Risk Assessment Methodology for LPG Installations, Conference on Risk & Safety Management in the Gas Industry, A. Reeves, V. Chow and F. Minah, Hong Kong, 1997
[48] Guidelines for Quantitative Risk Assessment “Purple Book”, CPR18E, Committee for the Prevention of Disasters, 2005
[49] Guidelines for Quantitative Risk Assessment “Purple Book”, CPR18E, Committee for the Prevention of Disasters, 2005
[50] Loss Prevention in the Process Industries, F. Lees, Butterworth-Heinemann, United Kingdom, 2005.
[51] Quantitative Risk Assessment Methodology for LPG Installations, Conference on Risk & Safety Management in the Gas Industry, A. Reeves, V. Chow and F. Minah, Hong Kong, 1997
[52] Quantitative Risk Assessment Methodology for LPG Installations, Conference on Risk & Safety Management in the Gas Industry, A. Reeves, V. Chow and F. Minah, Hong Kong, 1997
[53] Quantitative Risk Assessment of Proposed LPG Filling Station at Kwai Chung, DNV, 2002
[54] Harbour Area Treatment Scheme Stage 2A Environmental Impact Assessment, Drainage Services Department, Hong Kong, AEIAR-121/2008
[55] Kai Tak Development, Civil Engineering and Development Department, AEIAR-130/2009
[56] South Island Line, MTR Corporation Limited, AEIAR-155/2010
[57] Risk Assessment for the Siting And Developments Near Liquefied Petroleum Installations, Crossthwaite, P. J., Fitzpatrick, R. D., and Hurst, N. W., 1988
[58] The Calculation of Aircraft Crash Risk in the UK, Health and Safety Executive, United Kingdom, 1997.
[59] Aviation Statistical Reports, US National Transportation Safety Board, https://www.ntsb.gov/data/aviation_stats.html
[60] Hong Kong International Airport Hong Kong International Airport (2001-2004), Civil Aviation Department, http://www.cad.gov.hk/english/p-through.htm
[61] Demand Forecast, Hong Kong International airport, https://www.hongkongairport.com/eng/future/an-engine-for-growth/demand-forecast.html
[62] Risk of Hazardous Materials Release Following an Earthquake, K. J. Tierney, Disaster Research Centre, University of Delaware, 1990
[63] Quantitative Risk Assessment Methodology for LPG Installations, Conference on Risk & Safety Management in the Gas Industry, A. Reeves, V. Chow and F. Minah, Hong Kong, 1997
[64] The Calculation of Aircraft Crash Risk in the UK, Health and Safety Executive, United Kingdom, 1997
[65] Tsuen Wan Bypass, widening of Tsuen Wan Road between Tsuen Tsing Interchange and Kwai Tsing Interchange, and associated junction improvement works, Civil Engineering and Development Department, AEIAR-124/2008
[66] Guidelines for Quantitative Risk Assessment “Purple Book”, CPR18E, Committee for the Prevention of Disasters, 2005
[67] HSE, SPC/Tech/OSD/30, Methods of approximation and determination of human vulnerability for offshore major accident hazard assessment. 2011
[68] Vulnerability model. A simulation system for assessing damage resulting from marine spill , Eisenberg N A, et al., 1975
[69] Guidelines for Quantitative Risk Assessment “Purple Book”, CPR18E, Committee for the Prevention of Disasters, 2005
[70] Quantitative Risk Assessment Methodology for LPG Installations, Conference on Risk & Safety Management in the Gas Industry, A. Reeves, V. Chow and F. Minah, Hong Kong, 1997