4.0
Hazard Assessment
4.1
Introduction
4.1.1.1 The New Acute Hospital (NAH) in Kai Tak Development Area will be the main acute hospital services provider in the central Kowloon region. As part of the acute hospital with trauma facilities, a helipad will be built at the rooftop of the Acute Block of the NAH, where the Accident & Emergency Department is located.
4.1.1.2
The helipad will be
installed and operated at the west corner on the roof of the proposed Acute
Block of NAH, there will be no fuelling facilities provided in the proposed
rooftop helipad. The rooftop helipad design will meet the ‘ICAO Standards for
Heliport Design’ and ‘Government Flying Service Helicopter Landing Site
Specification Guidelines’.
4.1.1.3
The EIA requirements
for the proposed helipad are detailed in EIA Study Brief No. ESB-311/2019[1]. As stated in Section 3.2.1 of the brief [1], the EIA study shall address potential hazard to
life impact during the construction and operation of the helipad due to the
potentially hazardous facilities, including in particular the Kerry Dangerous
Goods Warehouse (Kowloon Bay) (KDGW) at 7 Kai Hing Road and the LPG Filling
Station at Cheung Yip Street.
4.1.1.4 This chapter presents the methodology, results, and
findings of the Hazard Assessment, including Quantitative Risk Assessment
(QRA), as part of the EIA study for the proposed helipad project.
4.2
Assessment Objectives
4.2.1.1 The
main objective of the Hazard Assessment has been to demonstrate that the risk
criteria set out in Technical Memorandum on Environmental Impact Assessment
Process (TM-EIAO)[2]
will be met during the construction and operational periods of the helipad, and
to identify practical mitigation measures to reduce the risk, where required.
4.2.1.2 The
detailed scope of hazard assessment includes the following:
·
Undertaking a review of the risks from the KDGW and
the LPG Filling Station to the Project to determine if risk to life is a key
issue with respect to the Risk Guidelines given in Annex 4 of the TM-EIAO [2];
·
Identification of hazardous scenarios associated
with the LPG Filling Station and the KDGW during the construction and operational
phases of the helipad;
·
Prepare a QRA to estimate the risks to the
surrounding population in both individual and societal terms;
·
Compare Individual Risks (IR) and Societal Risks
(SR) with the TM-EIAO Criteria to determine the acceptability of the Project in
terms of risk; and
·
Identify and assess practicable and cost effective
risk mitigation measures, to demonstrate compliance with the Risk Criteria (Figure
4.1), if the risk is determined to be not ‘Acceptable’.
4.3
Risk Criteria
4.3.1.1 The
risk guidelines are stipulated in the TM-EIAO Annex 4 to determine the risk acceptability.
The risk guidelines and criteria comprise the following two components:
1. Individual Risk: the maximum level of off-site individual risk should not exceed 1 × 10-5 / year, i.e. 1 in 100,000 per year; and
2. Societal Risk: Societal risk is expressed in the form of an
F-N curve, which represents the cumulative frequency (F) of all event outcomes
leading to N or more fatalities. The F-N curve criteria are shown in Figure
4.1. The three different regions are defined as:
·
Unacceptable region:
where risk is so high that they should usually be reduced regardless of the
cost or else the hazardous activity should not proceed;
·
ALARP region: where risk
is tolerable, provided that it has been reduced to a level As Low As Reasonably
Practicable (ALARP); and
·
Acceptable region: where
risk is broadly acceptable and does not require further risk reduction.
4.4
Project Description
4.4.1.1 The proposed helipad will be located on the rooftop
of the Acute Block of the NAH, at the south of Kai Fuk Road and Kwun Tong
Bypass, as shown in Figure
4.2. The LPG Filling Station, a dedicated filling station, is located
at Cheung Yip Street, about 211m from the proposed helipad. The KDGW is located
at 7 Kai Hing Road, about 265m from the proposed helipad. Both facilities are
within the Kai Tak Development Area and their locations relative to the Project
site shown as region 1 and 2 respectively in Figure
4.3.
4.4.1.2 The current zoning of the KDGW site is
"Commercial" under the approved Kai Tak Outline Zoning Plan No.
S/K22/6[3]. According to Planning
Permission Case No. A/K22/27, a proposal for minor relaxation of plot ratio
restriction has been applied and is under review at the time of the preparation
of this EIA report, and the information or development programme of the
existing KDGW site is also not confirmed at the
time of preparing this EIA report. As
detailed in Section 4.6, the tentative date for the commencement of
operations of the helipad would be in 2025. However, as the timing for the KDGW
redevelopment has not been confirmed to be before this date, this facility has
been included in this assessment.
4.5
Overview of Hazard Assessment Methodology
4.5.1.1 According to the Project EIA Study Brief [1], a review of the risks from the KDGW and the LPG
Filling Station shall be conducted to examine whether the risk to life is a key
issue with respect to Risk Guidelines given in Annex 4 of the TM. A QRA is required if, and only if, risk
to life is considered to be a key issue.
4.5.1.2 A hazard review study has been performed to
determine if the risks associated with the proposed helipad are significant. Consideration has been given to both the
risks due to impact from the KDGW and the LPG Filling station on the proposed
helipad, and vice versa. If the
increase in frequency of the hazardous events is anticipated to be above 1 ×10-9
per year, which is the lower limit of the societal criteria of Hong Kong, then
it is considered to be significant requiring further assessment using a
detailed QRA approach (Section 4.7). With the frequency screening criterion of 10-9
per year, the individual risk would also be covered since the individual risk criterion
of Hong Kong concerns only the risk level at 10-5 per year, which is
in principle, not affected if the frequency increase, is below 10-9
per year.
4.5.1.3 If a QRA is considered to be necessary, it will be
performed following the traditional QRA methodology, as outlined in Figure 4.5,
which includes the following key steps:
·
Hazard Identification;
·
Consequence Analysis;
·
Frequency Analysis; and
·
Risk Summation and Assessment.
4.5.1.4 The QRA also presents an update of the previous QRA
Study for the KDGW and the LPG Filling station carried out under the approved
EIA report for Kai Tak Development (Kai Tak QRA)[4], including the additional consideration of the impact
due to the proposed helipad. Where appropriate, the latest available
information of the area including the population data (Section 4.8) has been considered in the QRA update.
4.5.1.5 The methodology and assumptions used in the QRA have
made reference to the following three previously approved EIA studies:
·
Kai Tak Development Environmental Impact Assessment
Report. Hong Kong, 2006 (AEIAR-130/2009)[4];
·
A Rooftop Helipad at the Proposed New Block at Queen
Mary Hospital Environmental Impact Assessment. Hong Kong, 2017 (AEIAR-208/2017)[5]; and
·
Engineering Study for Police Facilities in Kong Nga
Po – Feasibility Study, Hong Kong, 2016 (AEIAR-201/2016)[6].
4.6
Hazard Review Study
4.6.1
Review
of Impact from Proposed Helipad
·
Construction Phase: the construction phase of
the helipad will introduce additional worker population to the area, which,
albeit temporary, may lead to an increase in the societal risk levels
associated with the operations of KDGW and the LPG filling station. Therefore,
an assessment has been carried out to examine whether such an increase in risk
is considered to be significant with respect to the societal risk criteria
adopted in Section 4.3.
·
Operational Phase: a frequency assessment has
been carried out to examine whether the operation of the proposed helipad will
lead to any significant increase in risk, with respect to the Hong Kong Risk
Criteria, to the hazardous facilities in the vicinity due to potential
helicopter crash accidents. The frequency assessment for helicopter crashes
makes reference to the methodology given in the report prepared by UK HSE[7] and is consistent with the previous EIA study
in Hong Kong (AEIAR-130/2009)[4]. The
individual risk has been evaluated based on the helicopter crashes rate. Besides that, the
impact on societal risk associated with the additional population associated
with the operational phase of the helipad has also been considered in the
assessment.
4.6.1.2
The impact from the
helipad construction activities would be related to the increased population
(i.e. construction workers) in the vicinity of the two sites of concern, which
may in turn lead to an increase in the societal risk. As per the construction
plan, the construction phase of the helipad will last over a period of 12
months, commencing towards the end of the construction for the NAH. The maximum
number of workers involved in the helipad construction phase is anticipated to
be about 20. However, given that the construction of the helipad will be
undertaken partly in parallel to the construction of the NAH, this number of
workers for helipad construction phase is not considered to be significant
compared to the overall workforce that will be required. In addition, the
construction of helipad would only involve line painting and supporting frame,
landing platform, lighting system, fire services facilities, covered safety
walkway and equipment in the final construction stage of the hospital, as
majority of the helipad deck would be comprised of prefabricated steel work
conducted at off-site premises. Hence, the increase in
risk due to the helipad construction alone is not considered to be significant
compared to the expected background risk level during the period of the overall
hospital construction. Besides, the number of personnel during the construction
phase is considered much less than the population during the operation phase. Nevertheless,
the risk assessment for the KDGW and the LPG Filling station during the helipad
construction phase is presented in Section
4.9 for completeness.
Operational Phase Assessment
Impact on Overall
Exposed Population
4.6.1.3 During the operational phase, the hospital staff
will be responsible for the helipad operations with no additional dedicated staff
solely responsible for the helipad. As advised by the Government Flying Service
(GFS), the maximum capacity of the helicopter would be about 20 people. For the
typical casualty evacuation (CASEVAC) event, there would be approximately 10
people including aircrew (i.e. Captain, Co-pilot, winch operator and winchman,
air medical officer and air medical nurse officer and escort) and patient(s) in
the helicopter.
4.6.1.4 In addition for a single CASEVAC event, one patient
is expected normal, but in some rare instances up to several patients may be
possible.
4.6.1.5 These staff members are regarded as transient
populations, who will be presented only during helipad operations. Also, this size of helicopter crew would not
be significant as compared to the total population of the hospital. Therefore, there is no significant
increase in population during the operational phase of the proposed helipad, in
addition to the hospital personnel.
Helicopter Crashes Scenarios
4.6.1.6 Considering the nature of the helipad, the key
potential hazard has been identified as the potential impact on the KDGW and
the LPG filling station in the case of a helicopter crash.
4.6.1.7 The probability of a helicopter crash for a given
target near the helipad, and its associated flight paths, can be approximated
using the methodology given by UK HSE[7], which is consistent with the previous Kai Tak QRA
(AEIAR-130/2009)[4]. Based on this methodology, the total helicopter
crash rate for a specific site comprises of 3 components, which are:
·
the background crash rate[7];
·
the helipad related crash rate, that is, landings
and take-offs; and
·
the crash rate below flight paths, that is the
flight paths.
4.6.1.8 As specific routes are known to be used on a
regular basis, it is then possible to treat these routes as airways and crash
rates for the routes can be calculated.(see Section 4.5 of AEA paper[7]). The methodology given by UK HSE [7] suggests that any facility lying under the
flight paths associated with the helipad, or situated within a 200 m radius
from the helipad, could be subjected to a potential increase in risk due to
accidental helicopter crash. Beyond this range,
the risk imposed by the helipad to other locations is generally considered to
be not significant and can be regarded as the same as the baseline risk level
elsewhere due to the background air traffic [7], that is, the background crash rate.
4.6.1.9 In Figure
4.3, both the KGDW and the LPG filling station are located beyond 200m
from the centre of the helipad; hence, the risks associated with helipad
related crashes are also considered to be negligible[7]. Therefore,
only the risk relating to helicopter crashes below flight paths are relevant for
assessment.
Helicopter Crash
Frequency below Flight Paths
4.6.1.10 Based on the project information, the helicopter
model to be used will be the Airbus H175, which is being operated by the GFS.
The annual number of helicopter movements depends upon the requirement for
emergency services which may vary from year to year, and the project also does
not have any estimation on this figure. Currently, public hospitals under the
Hospital Authority (HA) provided with helipad facilities include Pamela Youde
Nethersole Eastern Hospital (PYNEH) and Tuen Mun Hospital (TMH). According to
the usage data from PYNEH and TMH between 2007 to 2019, PYNEH has the highest
number of helicopter operations; hence, the PYNEH figure has been used to
provide a reasonably conservative estimation for the new helipad usage. It
should also be noted that the data for PYNEH does not show any downward or
upward trend in the number of helicopter movements over the years.
4.6.1.11 The emergency helicopter operation records at the PYNEH
indicate that the maximum number of helipad operations has been 295 per year
between 2007 and 2019. In order to adopt a conservative approach, the annual
number of helicopter movements for the proposed helipad has been assumed to be
330 for the purposes of this assessment. Note that the annual number of helicopter
movements of 330 is applied to the base and future cases in Section 4.10.
4.6.1.12
Figure 4.4 presents the helicopter flight paths associated
with the proposed helipad, as agreed with the GFS, and the distances from the
concerned hazardous facilities to each flight path are summarised in Table
4.1. The Flight Path W is not considered relevant here,
since the two facility sites are on the eastern side of the helipad while the
flight path is on the western side, as indicated in Figure 4.4, and, thus, this flight path does not cross both the
sites.
Table 4.1 Distance between Hazardous Facilities and Helicopter Flight Paths
|
Site |
Distance/m |
||
|
Flight Path NE |
Flight Path SE |
Flight path W |
|
|
KDGW |
116 |
116 |
N/A |
|
LPG filling station |
0.00 |
211 |
N/A |
4.6.1.13
The rate of aircraft
crashes below flight paths is given by the following equation [7]:
Where:
CA is in km-2yr-1,
NA
is the annual number of helicopter movements on the flight path,
RA is the aircraft inflight reliability. The
in-flight reliability is given according to the aircraft category[7]. For helicopters, the in-flight reliability is
1×10-7 crashes per flight km [7] .
The area factor, afac, takes into account the mean flight
path attitude and the minimum distance from flight path to the concerned sites.
For the purpose of
this analysis, the altitude of the flight path, alt, has been assumed to be the same as the height of the helipad
which is 119.15 mPD [1] or 117.85 m from Mean Sea Level[16]. This is considered to be a conservative
assumption to account for the lower altitude operations of helicopter when it
is approaching (or leaving) the helipad in the area.
4.6.1.15
Based on the above
equation, the crash rate below flight paths for each hazardous site has been
calculated, as presented in Table 4.2 Since the estimated crash rates are not insignificant (>10-9
per year), the scenarios have been further assessed in the QRA (Section 4.7).
Table 4.2 Crash Rates for Hazardous Facilities
|
Site |
Annual Crash Rate (per unit area / km-2yr-1) |
Area of the Target
Site (km2) |
Crash Rate (per Year) |
|
KDGW |
4.56×10-5 |
4.18×10-3 |
1.91×10-7 |
|
LPG Filling Station |
4.52×10-5 |
1.12×10-3 |
5.05×10-8 |
Consequence Assessment for Helicopter Crash
4.6.1.16 In the event of a helicopter crashing onto either
the KDGW or the LPG Filling Station, the consequence is expected to be
catastrophic. Similar catastrophic events have been assessed in the previous
EIA study (AEIAR-130/2009)[4], and the worst-case scenarios are extracted as
follows:
·
Major Warehouse Fire
(building collapsed) for KDGW (Scenario ID D-WF-EX of Table 11.4.13 in
(AEIAR-130/2009)[4]);
·
LPG Fire Loss Area
(building collapsed) for KDGW (Scenario ID D-LF-EX of Table 11.4.13 in
(AEIAR-130/2009)[4]);
·
Pentane Fire Loss Area
(building collapsed) for KDGW (Scenario ID D-PF-EX of Table 11.4.13 in
(AEIAR-130/2009)[4]);
·
Chlorine toxic release
(building collapsed) for KDGW (Scenario ID DG-CT-W of Table 11.4.13 in
(AEIAR-130/2009)[4]);
·
Ammonia toxic release
(building collapsed) for KDGW (Scenario ID DG-AT-W of Table 11.4.13 in
(AEIAR-130/2009)[4]);and
·
Catastrophic Failure of
LPG Storage Vessel for LPG filling station (Scenario 1 of Table 11.5.6 in
(AEIAR-130/2009)[4]).
4.6.1.17
The potential
helicopter crash scenarios due to the proposal helipad are considered to
increase the event frequency of these catastrophic events. According to the previous
Kai Tak QRA[4], extreme events represent those accidents relating to extreme
initialising events such as earthquake, which would lead to building collapse.
The consequence of a helicopter crash on the KDGW would fall in this category. The
results of the frequency calculation can be found in Section 4.7.4.
4.6.2
Review
of Impact from KDGW and LPG Filling Station
4.6.2.1 As shown in Figure 4.4,
the flight paths of the helicopter are close to the airspace above the LPG
Filling station and the KDGW. Adopting the method used in the EIA
report for Operation of the Existing Tai Lam Explosives Magazine at Tai Shu Ha,
Yuen Long for Liantang / Heung Yuen Wai boundary Control Point Project (AEIAR-193/2015)[15], the potential impact on the passing helicopter
can be estimated.
4.6.2.2 With regard to the LPG Filling Station, according to
paragraph 11.5.61 of the previous Kai Tak QRA[4], the maximum LPG cloud height calculated is 10m
for all continuous release events using the dense gas dispersion models as per
the previous EIA[4]. Therefore, only instantaneous releases, namely
‘Catastrophic Failure of a Storage Vessel’ and ‘Catastrophic Failure of Road
Tanker’ are considered to be relevant, having the potential for impacting the
passing helicopters. The frequency
of impact on any point along the flight paths can be calculated as the sum of
the two event frequencies which is 2.36×10-6 per year.
4.6.2.3 The maximum consequence distance of the concerned
scenario is 259m based on the LPG filling station QRA. The total helicopter
movements are 220 annually along Flight Path SE and Flight Path NE (Section
4.6.1.14). The approaching speed of the helicopter (Airbus H175) is
conservatively assumed to be 60 km/h[a]. On
this basis, the transit time of a helicopter along the concerned section of
flight path has been approximated to be 31 seconds. The presence factor is,
therefore, calculated as 2 × 110 × 31 / (365 × 24 × 3600) = 2.17×10-4,
which results in a probability of potential impact of 2.36×10-6×2.17×10-4=
5.12×10-10 per year.
4.6.2.4 A significant warehouse fire scenario at KDGW can result in a smoke plume impacting the passing helicopter, if
the smoke plume encroaches upon the flight paths. A smoke plume generated from
a fire follows the tilt angle of the flame initially and rises upward due to
high buoyancy caused by high temperature. Plume Height (H) can be calculated as
L × tan(90° - tilt angle), where L is horizontal distance from the fire size based on Section 11.4.129 of the previous EIA[4]. The tilt angle should be set as 60° to represent
the worst case scenario. Using this equation, the smoke plume height is
estimated as 67m, considering a distance of 116 m between KDGW and the flight
paths. Since helicopters at this segment of the flight path are anticipated to
remain above an altitude of 117.85 m from Mean Sea Level, smoke plume
encroaching upon the flight paths is considered to be possible. However, in the
event of a fire accident at KDGW, the resulting fire and smoke plume will be
visible to the pilot, who can take necessary actions to avoid approaching near
the smoke plume. Therefore, the impact of the smoke plume has not been
considered further in this QRA. As per paragraph 11.4.132 of the Kai Tak EIA[4] ,a
comprehensive study was done to compute the indoor fatality probability
relative to the outdoor fatality probability. For the purpose of the assessment
it was assumed that all commercial sites would be centrally air conditioned
with fresh air change rates of 1 air change per hour. Furthermore, it is
assumed that population inside the building can escape in 30 minutes. Hence,
having taken these factors into account, it is concluded that the indoor
fatality probability is 5% of the outdoor fatality probability.
4.6.2.5 Also, based on consequence modelling, all other KDGW scenarios, such as
chlorine, pentane and ammonia loss of containment, were found to have an impact
height less the 117.85 m from Mean Sea Level (the maximum impact height was
found to be lower than 50m considering all weather classes); therefore, the
impact from these other scenarios on flight paths are also not considered
further in the study.
4.6.2.6 Since helicopter crash frequencies due to the scenarios by the LPG
Filling station and the KDGW are not considered to be significant, it is
concluded that the impact of the hazard events to the helicopter operation is
negligible.
4.7
Quantitative Risk Assessment
4.7.1
Introduction
4.7.1.1 Based on the hazard review study (Section 4.6), the risk associated with helicopter crashing
into the KDGW and the LPG filling station was found to be significant requiring
further assessment using the QRA approach. The QRA approach is outlined in Figure 4.5.
4.7.1.2 The NAH is one of the major projects in the Kai Tak
Development Area, for which an EIA [4], including a QRA for the KDGW and the LPG filling,
was approved in 2009. As such, some
of QRA aspects relating to the proposed helipad project, such as the construction
population during the project construction phase, have been already assessed in
the previous Kai Tak QRA[4]. Hence, this QRA has adopted the methodology,
assumptions, and data used in the previous Kai Tak QRA[4].
4.7.1.3 The QRA has considered three population scenarios
for societal risk assessment:
·
Base case in 2025 (the
year in which the proposed helipad will begin operation); and
·
Future case in 2036 (a
future case to account for population growth in the surrounding areas over the
years).
·
Future case in 2036 (future
case accounting for the population growth in the surrounding areas over the
years, without the Project i.e., without the Rooftop Helipad and its Operation)
4.7.2
Hazard Identification
4.7.2.1 The hazard identification has comprised a review of
the hazardous material scenarios studied in the previous Kai Tak QRA[4], with the objective of identifying potential
hazards involved during the construction and operation phase of the helipad,
and understanding the ways in which those hazards may be realised.
4.7.2.2 As noted in Section 4.6.1.2, the construction phase of the proposed helipad is not expected to pose
any significant risk to the operation of KDGW and LPG filling station.
4.7.2.3 In the event of a helicopter loss of control when operating in the
vicinity of the proposed helipad, the helicopter can potentially crash into the
LPG filling station or KDGW, leading to damage to the facilities with
subsequent fire, explosion or toxic release. This has been identified as the key
hazard associated with the proposed helipad during its operation phase.
4.7.2.4 For the LPG Filling Station, the LPG facilities
have been verified and confirmed by site visit and interview that no significant
modification has been undertaken as compared to those assumed in the previous
Kai Tak QRA[4], while operational information such as LPG tanker
delivery frequency has been updated. In terms of the risk, the flammability of
LPG remains to be the key hazard of concern. In line with similar LPG QRA studies in
Hong Kong[4] [22], the following types of scenarios have been
identified for assessment:
·
Failure of tank and road
tanker;
·
Failure of pipe;
·
Failure of storage tank
valve; and
·
Boiling Liquid Expanding
Vapour Explosion (BLEVE) of road tanker.
4.7.2.5 For the KDGW, the following scenarios have been identified[4]:
·
Major warehouse fire;
·
Fire in single store and
escalating toxic release;
·
LPG release;
·
Pentane fire;
·
Chlorine release; and
·
Ammonia release.
4.7.3
Hazard Scenarios
4.7.3.1
Based on the hazards
described above, Table 4.3 and Table 4.4 present the hazard scenarios
considered for the KDGW and the LPG filling station based on the previous Kai
Tak QRA[4]. It is reasonable to assume that the consequence
resulting from a helicopter crash would be catastrophic. For the LPG station,
such an event was identified as ‘Catastrophic Failure of a Storage Vessel’,
which is associated with the maximum consequence distance among all failure
scenarios based on the previous Kai Tak QRA[4]. For the KDGW, building collapse has been
considered in the event of a helicopter crash. As per the previous Kai Tak QRA[4], building collapse was regarded as an extreme
scenario, as a consequence of a major earthquake in Hong Kong.
4.7.3.2
Since the identified catastrophic
events already involve the entire hazardous inventory of the respective site
releasing in the most rigorous manner, these are considered to be the worst
case scenarios even in the event of a helicopter crash. Therefore, no
modification is, in principle, required for the consequence assessment, as given
in the previous Kai Tak QRA[4]; however, the associated event frequencies should
be updated to include the additional frequency contribution due to the
potential helicopter crash scenario.
4.7.3.3
Table 4.3 and Table 4.4 summarise the hazard
scenarios which are affected by the proposed helipad for the LPG Filling
Station and the KDGW, respectively.
Table 4.3 Hazard Scenarios Considered in the previous Kai Tak QRA[4] for LPG Filling Station
|
Hazard Scenarios |
Event Frequencies
to be Modified due to the Helipad |
|
Catastrophic Failure of a Storage Vessel |
Yes |
|
Catastrophic Failure of a Road Tanker |
No* |
|
Partial Failure of
a Storage Vessel |
No |
|
Partial Failure of
a Road Tanker |
No |
|
Guillotine Failure
of Liquid Filling Line to Storage Vessel |
No |
|
Guillotine Failure
of Liquid Filling Line to Dispenser |
No |
|
Failure of
Dispenser |
No |
|
Guillotine Failure
of Hose during Unloading from Tanker to Storage Vessel, LPG Released from
Tanker |
No |
|
Guillotine Failure
of Hose during Unloading from Tanker to Storage Vessel, LPG Released from
Vessel |
No |
|
Failure of Flexible
Hose during Loading to LPG Vehicles, LPG Released from Dispenser |
No |
|
Failure of Flexible
Hose during Loading to LPG Vehicles, LPG Released from Vehicle |
No |
|
Release from
Storage Vessel Pump Flange |
No |
|
Release from
Storage Vessel Drain Valve |
No |
|
Guillotine Failure of
Vapour Return Line |
No |
|
Guillotine Failure
of Liquid Line from Tanker to Loading Hose |
No |
|
BLEVE of Road
Tanker |
No |
Notes* Road tanker has
the same consequence distances as storage tank rupture. However, since road
tanker is present only intermittently at the filling station, the additional
failure frequency due to helicopter impact is assigned to storage tank only for
modelling purposes.
Table 4.4 Hazard Scenarios Considered in the previous Kai Tak QRA[4] for KDGW
|
Hazard Scenario |
Event Frequencies
to be Modified due to the Helipad |
|
Major Warehouse
Fire |
No |
|
Fire in Single
Store |
No |
|
LPG Fire Loss area
Internal (no offsite effect) |
No |
|
LPG Fire Loss
cylinder Internal (no offsite effect) |
No |
|
LPG Fire Loss area
External |
No |
|
LPG Fire Loss
cylinder External (no offsite effect) |
No |
|
Pentane fire Loss
Area Internal (no offsite effect) |
No |
|
Pentane fire Loss
drum Internal (no offsite effect) |
No |
|
Pentane fire Loss
Area External |
No |
|
Pentane fire Loss
drum External (no offsite effect) |
No |
|
Chlorine toxic
release (building collapsed) |
Yes |
|
Ammonia toxic
release (building collapsed) |
Yes |
|
Chlorine–Loss of
Area Internal |
No |
|
Chlorine–Loss of
Cylinder Internal |
No |
|
Ammonia–Loss of
Area Internal |
No |
|
Ammonia–Loss of
Cylinder Internal |
No |
|
Chlorine–Loss of
Area External |
No |
|
Chlorine–Loss of
Cylinder External |
No |
|
Ammonia–Loss of
Area External |
No |
|
Ammonia–Loss of
Cylinder External |
No |
|
Major Warehouse
Fire (building collapsed) |
Yes |
|
LPG Fire Loss Area
(building collapsed) |
Yes |
|
Pentane fire Loss
Area (building collapsed) |
Yes |
4.7.4
Frequency Assessment
4.7.4.1 For the LPG filling station, although most of the scenarios
are not affected by the proposed project, the event frequency and consequence
have been re-assessed for all scenarios in this study. The failure frequencies of LPG facilities
have been derived based on literature [20]. The current operation data
of the LPG filling station has also been obtained [18]; it is noted that LPG tankers
will generally remain in the LPG station for about 90 minutes per delivery, in
which the initial and final 5 minutes are spent on preparation, connection and
disconnection of the transfer. As such, inventories of the road
tankers are assumed to be full for 5.6% of time, 50% for 88.8% of time and 0% for
5.6% of time. Based
on the above, Fault Tree Analysis (FTA) has been performed in this QRA to estimate
the various failure event frequencies. The latest operation data obtained and
the FTA are detailed in Appendix 4C. However, the data request sent to KDGW
was not responded; therefore, all data and assumptions used in the Kai Tak QRA[4], which are deemed conservative, have been retained
in this QRA.
4.7.5.1 As assessed in Section
4.6.1, the frequency of helicopter crashing into the LPG
filling station and KDGW has been found as
5.05×10-8 per year and 1.91×10-7 per year,
respectively. These crash frequencies have been added to the catastrophic
scenarios, as identified in Table 4.3 and Table 4.4. A summary of the updated frequency
data used in this QRA is provided in Table 4.5 and Table 4.6.
The
calculation details can be found in Appendix 4E Table 4E1 and 4E2.
Table 4.5 Failure Event Frequencies for LPG Filing Station[4]
|
Scenario |
Event Frequency (per Year) |
|
Catastrophic Failure of Storage
Vessel (full)* |
5.05 × 10-7 |
|
Catastrophic Failure of Road
Tanker (full) |
2.53 × 10-7 |
|
Catastrophic Failure of Road
Tanker (50%) |
4.02 × 10-6 |
|
Pump Flange Leak |
4.36 × 10-4 |
|
Partial failure of storage vessel |
1.02 × 10-5 |
|
Partial failure/leak road tanker |
3.33 × 10-4 |
|
Guillotine failure of filling line
to storage vessel |
1.72 × 10-10 |
|
BLEVE of Road Tanker |
9.02 × 10-9 |
|
Failure of Liquid Supply Line to
Dispenser |
6.64 × 10-7 |
|
Failure of Dispenser |
5.71 × 10-3 |
|
Failure of Flexible Hose during
Loading to Vessel |
1.06 × 10-5 |
|
Release from Storage Vessel Drain
Valve |
4.80 × 10-4 |
|
Failure of Vapour Return Line |
1.65 × 10-7 |
* Scenario affected
by helicopter crash
Table 4.6 Event Frequencies for KDGW[4]
|
Scenario |
Event Frequency (per Year) |
|
Major Warehouse Fire |
1.68 × 10-4 |
|
LPG Fire Loss area External |
1.16 × 10-9 |
|
Chlorine toxic release(building
collapsed)* |
3.26 × 10-10 |
|
Ammonia toxic release(building
collapsed)* |
3.26 × 10-10 |
|
Chlorine-Loss of Area Internal |
1.97 × 10-8 |
|
Ammonia-Loss of Area Internal |
1.97 × 10-7 |
|
Chlorine-Loss of Area External |
9.71 × 10-9 |
|
Ammonia-Loss of Area External |
9.71 × 10-8 |
|
Pentane fire Loss Area (building
collapsed)* |
1.19 × 10-6 |
|
Chlorine-Loss of Cylinder External |
8.74 × 10-8 |
|
Ammonia-Loss of Cylinder External |
9.32 × 10-9 |
|
LPG Fire Loss Area(building
collapsed)* |
<<1.00 × 10-9 |
|
Major Warehouse Fire(building
collapsed)* |
1.19 × 10-6 |
|
Fire in Single store |
1.12 × 10-4 |
|
Chlorine-Loss of Cylinder Internal |
3.50 × 10-6 |
|
Ammonia-Loss of Cylinder Internal |
3.73 × 10-6 |
|
Pentane fire-Loss of
Area External |
1.40 × 10-3 |
* Scenario affected
by helicopter crash
4.7.4.2
For the LPG filling
station, the main ignition sources have been identified as the traffic on the
adjacent roads. By using PHAST Risk
and its built-in risk / event tree modelling program, MPACT, the road traffic
was modelled as ignition sources, whereby the various fire event frequencies could
be determined. The detailed parameters used in PHAST Risk are attached in Appendix 4D. As for KDGW, no further ignition
probability or event tree modeling is necessary, as the derived frequencies as
given in Table 4.6 already represent
the ultimate undesired event (e.g. fires and toxic dispersion) frequencies.
4.7.5
Consequence Assessment
4.7.5.1 Consequence modeling has been performed using PHAST
for the LPG filling station, based on the LPG storage and road tanker
information obtained, as detailed in Table 4C1 and 4C2 in Appendix 4C. The following physical effects have been
considered in the event of LPG loss of containment.
·
Flash Fire: flash fire results from delayed ignition of a
flammable vapour cloud, generated either through vaporisation directly from the
release, or from vaporising pools. The main hazards of a flash fire being
direct flame contact. The area of possible direct flame contact effects is
determined as the distance to the lower flammability limit (LFL) of the vapour
cloud. Due to the extreme short duration of a flash fire, radiation effects are
negligible. Flash fire was considered for delayed ignition of flammable gases.
·
Jet Fire: A jet fire
results from immediate ignition of the flammable material (i.e. LPG) from a
pressurised release. In the context of the LPG station, jet fires are relevant
to continuous leak from pipework and storage vessel which are under
pressure. The main hazards from a
jet fire are direct flame contact and radiation, both of which are modelled
using default parameters in PHAST Risk.
·
Fireball: Spontaneous
rupture (e.g. cold rupture) of LPG vessels or road tankers can give rise
fireballs in case of immediate ignition. 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.
·
Vapour Cloud Explosion
(VCE): When a flammable vapour cloud forms,
disperses and accumulates in areas with high congestion or confinement, and is
then ignited, a Vapour Cloud Explosion (VCE) may result. The VCE effects were
modelled using TNO multi-energy model in PHAST.
4.7.5.2 For all the physical effects above, the associated fatality
probabilities was estimated using Probit equations, the Tsao and Perry
Correlation, as given in PHAST Risk. A protection factor of 0.1 was assumed for
indoor population.
4.7.5.3 The consequence results for KDGW have been
extracted directly from the previous Kai Tak QRA[4]. It is noted that consequence results are not
reported in the Kai Tak QRA for some of the scenarios, such as the Chlorine and
Ammonia releases (Building Collapse). It is also noted that the consequence
results reported in the Kai Tak QRA for KDGW include only 3E and 4C weather
conditions. Nevertheless all scenarios that are reported in Table 11.4.13 of
Kai Tak QRA, except for those with no offsite effect, have been modeled using
PHAST for all weather conditions. A check was done to identify hazard distances
from Appendix 11.4.4 that are greater than those obtained from modeling in
PHAST. These were used alongside those from modeling in PHAST to allow for a
conservative approach. The consequence distances for the representative cases
of the LPG filling station and the KDGW are summarised in Table 4.7 and Table 4.8, respectively.
Table 4.7 Consequences Data of Building Collapse Scenarios and other representative
cases for KDGW[4].
|
Scenario |
Consequence |
Maximum Hazard Distance from Release Source (m) |
|
Major Warehouse Fire |
Toxic cloud (3%
probability of fatality) |
223a |
|
Major Warehouse Fire (building collapsed) |
Toxic cloud (3%
probability of fatality) |
426a,b |
|
LPG Fire Loss Area (building collapsed) |
Fireball |
23 |
|
Flash fire (0.85 LFL) |
96a |
|
|
Pentane Fire Loss Area (building collapsed) |
Thermal radiation (3%
probability of fatality) |
86 |
|
Ammonia–Loss of Area External |
Toxic cloud (3%
probability of fatality) |
218a |
|
Chlorine–Loss of Area External |
Toxic cloud (3%
probability of fatality) |
344 |
|
Chlorine release – loss of warehouse |
Toxic cloud (3%
probability of fatality) |
900a,b |
|
Ammonia release – loss of warehouse |
Toxic cloud (3%
probability of fatality) |
630a,b |
Note a: Scenarios having larger hazard
distances upon remodeling using PHAST, compared to hazard distances reported in
Appendix 11.4.4 of EIA report for Kai Tak Development no. AEIAR-130/2009.
Note b: Frequencies associated with these hazard distances have been taken
into account the occurrence probabilities of the specific wind conditions.
Table 4.8 Consequences Data of Catastrophic Failure Scenario of a Storage Vessel
for LPG Filling Station[4]
|
Mass of LPG
(tonnes) |
Fireball Radius (m) |
Fireball Duration
(s) |
Max. Dispersion
Distance to LFL (m) |
|
6.58 [4] |
57 |
8 |
105 |
4.7.5.4 The latest weather data (recent five years) has been
obtained from the Hong Kong Observatory in the QRA update and presented in Table 4.9 below.
Table 4.9 Wind Direction Frequencies
|
Day Time |
|||||||
|
Direction |
Weather Class |
Total |
|||||
|
1F |
3E |
1D |
3B |
4C |
7D |
||
|
0 - 30 |
0.42 |
0.39 |
0.32 |
0.57 |
0.97 |
0.05 |
2.72 |
|
30 - 60 |
0.23 |
0.70 |
0.18 |
0.74 |
1.99 |
0.14 |
3.98 |
|
60 - 90 |
0.16 |
0.40 |
0.14 |
0.73 |
3.15 |
1.18 |
5.76 |
|
90 - 120 |
0.26 |
1.47 |
0.26 |
1.39 |
15.57 |
8.04 |
26.99 |
|
120 - 150 |
1.11 |
2.30 |
0.97 |
4.46 |
16.73 |
1.23 |
26.80 |
|
150 - 180 |
0.60 |
0.32 |
0.59 |
1.92 |
1.78 |
0.05 |
5.26 |
|
180 - 210 |
0.19 |
0.16 |
0.38 |
1.31 |
1.50 |
0.04 |
3.58 |
|
210 - 240 |
0.29 |
0.29 |
0.44 |
2.92 |
3.03 |
0.17 |
7.14 |
|
240 - 270 |
0.30 |
0.27 |
0.34 |
1.70 |
3.45 |
0.67 |
6.73 |
|
270 - 300 |
0.42 |
0.33 |
0.34 |
0.70 |
1.41 |
0.19 |
3.39 |
|
300 - 330 |
0.37 |
0.48 |
0.26 |
0.69 |
2.24 |
0.34 |
4.38 |
|
330 - 360 |
0.28 |
0.38 |
0.25 |
0.73 |
1.51 |
0.12 |
3.27 |
|
All |
4.63 |
7.49 |
4.47 |
17.86 |
53.33 |
12.22 |
100.00 |
|
Night Time |
|||||||
|
Direction |
Weather Class |
Total |
|||||
|
1B |
3E |
1F |
4D |
7D |
1D |
||
|
0-30 |
0.00 |
1.48 |
2.12 |
0.38 |
0.01 |
0.04 |
4.03 |
|
30 - 60 |
0.00 |
2.86 |
1.04 |
0.85 |
0.03 |
0.09 |
4.87 |
|
60 - 90 |
0.00 |
2.47 |
0.78 |
1.73 |
0.94 |
0.05 |
5.97 |
|
90 - 120 |
0.00 |
10.26 |
1.51 |
11.98 |
6.05 |
0.05 |
29.85 |
|
120 - 150 |
0.00 |
11.64 |
6.54 |
4.27 |
0.35 |
0.05 |
22.85 |
|
150 - 180 |
0.00 |
2.08 |
3.91 |
0.36 |
0.00 |
0.08 |
6.43 |
|
180 - 210 |
0.06 |
1.40 |
1.91 |
0.20 |
0.01 |
0.02 |
3.60 |
|
210 - 240 |
0.00 |
2.51 |
1.95 |
0.52 |
0.06 |
0.03 |
5.07 |
|
240 - 270 |
0.00 |
2.39 |
2.07 |
0.95 |
0.15 |
0.05 |
5.61 |
|
270 - 300 |
0.00 |
1.34 |
1.62 |
0.53 |
0.13 |
0.07 |
3.69 |
|
300 - 330 |
0.00 |
1.96 |
1.10 |
1.08 |
0.18 |
0.09 |
4.41 |
|
330 - 360 |
0.00 |
1.50 |
1.25 |
0.63 |
0.17 |
0.07 |
3.62 |
|
All |
0.06 |
41.89 |
25.80 |
23.48 |
8.08 |
0.69 |
100.00 |
4.8
Population Estimation
4.8.1
Approach
4.8.1.1 Overall, the approach adopted in the previous Kai
Tak QRA[4] has been adopted to estimate the population to be
considered in the helipad QRA. Based on the Kai Tak QRA, the population that
can be potentially impacted can be classified as:
·
Building Population
(residential and employment), and
·
Traffic Population
(marine and road)
4.8.1.2 For the LPG Filling Station and KWDG, the study
areas have been determined based on the maximum hazard radius of the worst case
scenarios. The determination of the maximum hazard radius is further explained
in Section 4.7.5.3.
4.8.1.3 An average annual population growth rate of 1% for
the effective regions is calculated by considering the population data given by
the Planning Department [14].
4.8.2
Time Period and Occupancy
4.8.2.1
Four time periods have
been considered, namely weekday and weekend at day and night time. Following
the assumptions used in AEIAR-130/2009[4], the percentages of the population (occupancy) at
different times of the day/ night are presented in Table 4.10, and the indoor/ outdoor fraction for different
population groups is presented in Table 4.11.
Table 4.10 Population Category Distributions
|
Population Category |
Weekday |
Weekend |
||
|
Day |
Night |
Day |
Night |
|
|
Car Park |
100% |
10% |
40% |
5% |
|
Hospital |
100% |
100% |
100% |
100% |
|
Industrial Building |
100% |
10% |
40% |
5% |
|
Commercial Building |
100% |
10% |
40% |
5% |
|
Petrol Station |
50% |
1% |
50% |
1% |
|
Police Station |
100% |
30% |
65% |
30% |
|
Station |
100% |
10% |
55% |
10% |
|
Leisure |
100% |
10% |
100% |
10% |
Table 4.11 Indoor and Outdoor Ratios
|
Population Category |
Indoor (Outdoor) Ratio |
|
Residential |
0.95 (0.05) |
|
School |
0.95 (0.05) |
|
Park |
0.00 (1.00) |
|
Road |
0.00 (1.00) |
|
Railway/ Bus Station |
0.00 (1.00) |
|
Marine |
0.00 (1.00) |
4.8.3
Building Population
4.8.3.1 As stated in the previous Kai Tak QRA[4], the permanent population (i.e.
residential and commercial population) in the Kai Tak Development (KTD) was
estimated based on the Kai Tak Engineering Study [8]. The previous Kai Tak QRA has also made reference
to the Outline Zoning Plan (OZP) zoning and the Planning Vision and Strategy
(PVS)[4] zones for estimating the indoor / outdoor
population ratios and the areas of the zones.
4.8.3.2 For consistency, the building population data as
reported in the previous Kai Tak QRA[4] has been used as much as practicable. Nevertheless, an update of the population
has been performed for the followings:
·
Recently developed sites
where more detailed population data is available.
·
Sites reserved for
future development where the land use type as per latest PVS[3]
has been modified as compared to that assumed in the Kai Tak
QRA[4]. The modified sites are identified in Appendix
4A.
4.8.3.3 For building population update, a detailed
estimation method has been implemented using the following sources of
reference, for which the approach is similar to the previous EIA[9]:
·
Census Statistics Report
2016[10]
·
Unit floor plans from
Centadata[11];
and
·
Specific websites for
different building category.
Population for each
building has been estimated based on the type of building (i.e. residential,
commercial, etc.), number of floors and number of units per floor. Buildings are
assigned to the appropriate building usage categories using information from
property developers’ websites and site visits. The specific population
assumptions utilised are summarised in Table 4.12[9].
Table 4.12 Building
Population Assumptions
|
Category |
Building
Size* |
Assumptions |
Total |
||
|
Bus
Terminus |
Based
on the building area |
||||
|
Carpark |
|
The
car park population was adjusted based on the parking level. |
|
||
|
Parking Level |
Parking Spaces |
People/ Parking
Space |
|||
|
H |
5 |
40 |
0.2 |
40 |
|
|
L |
1 |
20 |
0.2 |
4 |
|
|
Clinic |
Clinics
with higher no. of floors were scaled up proportionally. |
||||
|
Floors |
Unit |
People/ Unit |
|||
|
H |
3 |
20 |
3 |
180 |
|
|
M |
2 |
10 |
2 |
40 |
|
|
L |
1 |
1 |
10 |
10 |
|
|
Commercial
Building |
Floors |
Unit |
People/ Unit |
||
|
H |
10 |
20 |
2 |
400 |
|
|
M |
5 |
20 |
2 |
200 |
|
|
L |
2 |
10 |
2 |
40 |
|
|
Fire
Station & Ambulance Depots |
About 10240
uniformed staff are employed in Hong Kong. It is assumed that the members of
fire and ambulance streams are evenly distributed in 81 fire stations and 39
ambulance depots respectively. It is also assumed that members fire stream
will roster on 24 hours (on-duty) and 48 hours (off-duty) and members of
ambulance stream will roster on 12 hours, 2 shifts each day. |
32 |
|||
|
Industrial
Building |
The
Industrial building population was scaled proportionally with the number of
floors. |
||||
|
Floors |
Unit |
People/ Unit |
|||
|
H |
25 |
8 |
8 |
1600 |
|
|
M |
15 |
6 |
8 |
720 |
|
|
L |
8 |
6 |
6 |
288 |
|
|
Leisure |
H |
200
people for leisure facility with large size. |
200 |
||
|
M |
100
people for leisure facility with medium size. |
100 |
|||
|
L |
50
people for leisure facility with small size. |
50 |
|||
|
LL |
10
people for leisure facility with very small size. |
10 |
|||
|
Petrol
Station |
It
is assumed that there are 2 staff stationed in the convenience shop, 4
stationed in the fuel area for filling and 4 vehicles each with 3 people,
parked into the Petrol Station for petrol filling. |
18 |
|||
|
Police
Station |
About
29400 policemen are employed in Hong Kong. It is assumed that they are
distributed evenly among 55 branches. It is also assumed that they will
roster on 2 shifts each day and about 50% will be out for patrol. |
149 |
|||
|
Station |
H |
5
people in Refuse disposal and Mortuaries |
5 |
||
|
M |
2
people in Traffic Control Stations |
2 |
|||
|
L |
No
people will stay in Sewage treatment works, Toilet, Electric substations or
pump house |
0 |
|||
|
Temple/
Church |
H |
100
people for temple/ church with large size |
100 |
||
|
M |
50
people for temple/ church with medium size |
50 |
|||
|
L |
10
people for temple/ church with small size |
10 |
|||
*Legend for Building Height/Size
-
H for Tall/Large,
-
M for Medium,
-
L for Low/Small
-
LL for Very Low/Very Small
4.8.3.4 Other than the above building types, the population
in residential buildings was estimated based on site survey and other publicly
available sources, e.g. floor plan of public housing estates from the Hong Kong
Housing Authority, properly developers’ websites, and property agencies
websites.
4.8.3.5
For the acute block of
the hospital, a total population of 10,000 was assumed in the QRA for 2019. The
assumed population has taken into consideration the number of hospital
employees and patients as given by the Sewerage Impact Assessment report
(Section 4.2.2)[13],
with additional people to account for the transient population, such as
visitors and contractors.
4.8.3.6 For areas reserved for future development, the data
available from the public domain has been referenced for estimation. The population
data used in the QRA for year 2025 and 2036 has been estimated by extrapolating
the data from Projection
of Population Distribution 2019-2028 [14].
4.8.3.7 Site visits have also been carried out with the
objective of confirming any latest development of commercial or residential
buildings over the past few years, which may not have been captured in the
published PVS (2016)[10]. The site visits also supplemented and verified
the estimate of the building information such as number of floors as given by
the various sources previously mentioned[4]. Appendix 4B
presents the buildings identified in each OZP and the population data.
4.8.4
Traffic Populations
4.8.4.1 For updating the marine population density, site
surveys have been conducted to estimate the latest marine traffic population in
the vicinity of the Kwun Tong Typhoon Shelter. The maximum survey radius has
been determined with respect to the maximum radius of the 1E-09 per year individual
risk contour of the Kai Tak QRA[4] and
the current QRA, as shown in Figure 4.6.
4.8.4.2 The survey used the same technique as adopted in the
previous Kai Tak QRA[4]. At the survey location, the surveyor took 3
snapshots at the survey location. The survey was conducted at different periods
of the day (i.e. day and night) covering both weekday and weekend. Records of
the surveys and estimation of the marine population are detailed in Appendix 4B.
The marine population was calculated by the following equation:
4.8.4.3 It was observed during the site surveys that the
majority of the marine vessels berthed within the Typhoon Shelter were unoccupied.
Therefore, a presence factor of 0.1 was applied in estimating the marine
population based on the number of marine vessels counted. On this basis, the
marine population density for the Kwun Tong Typhoon Shelter has been estimated as
1.44 x 10-5 people / m2 , assuming 10 occupants per boat,
as per the approach in the previous Kai Tak QRA [4]
4.8.4.4 Road population has been updated mainly based on the information from Annual Traffic Census (ATC) [12]. Specifically, the data from the following core stations and coverage stations, as detailed in Table 4.13 below, have been used in the analysis.
Table
4.13
Core Station and Coverage Stations near Kai Tak Area
|
Station No. |
Station Name |
From |
To |
|
Core Station |
|||
|
3012 |
Kwun Tong Road |
Ngau Tau Kok Road |
Kwun Tong Road |
|
3020 |
Wai Yip Street |
Lai Yip Street |
Hoi Yuen Road |
|
3023 |
Kwun Tong Bypass
<K77> |
Wai Yip Street |
Cheung Yip Street |
|
Coverage Stations |
|||
|
3833 |
Kwun Tong Road |
Kai Fuk Road FO
<K58> Eastern |
Hong Ning Road |
|
3870 |
Ngau Tau Kok Road |
Hong Ning Road |
Elegance Road |
|
3873 |
Lai Yip Street |
Wai Yip Street |
Kwun Tong Road |
|
4075 |
Sheung Yee Road |
Wai Yip Street |
Wang Chiu Road |
4.8.4.5
ATC provides traffic
density information in the form of Annual Average Daily Traffic (AADT)[12], which is considered a suitable representation for
normal traffic flows.
The parameters used in
the above equation are the same as those given in the previous Kai Tak QRA
AEIAR-130/2009[4].
4.8.4.6 By analysing the AADT trend, a population growth rate
can be derived for each road, assuming that the road population increases
proportionally each year. If the population is expected to decrease in the
future, then the population data reflecting the current status has been used,
without any adjustment for growth or reduction, as a conservative approach.
4.8.4.7 Population data for the following roads within the
maximum hazard zones of KDGW and LPG filling station[4] is, however, not provided by the published ATC[12]
·
Kai Hing Road
·
Hoi Bun Road
·
Wang Hoi Road
4.8.4.8 Therefore, site surveys have been conducted for the
above roads in order to estimate the road population. In the survey, a count of
the various types of vehicles has been made and recorded. The surveys were
conducted at different periods of the day/night, in line with the
representative periods considered in Table 4.14. Using the survey data, the number of vehicle per
day for each road was calculated, based on which the road population can be estimated
using the equation in Section 4.8.4.5.
Table 4.14 Representative Time Periods for Road Population Surveys
|
Representative
Time Period |
Start
Time |
End
Time |
|
1 |
8:00 am |
8:15 am |
|
2 |
2:00 pm |
2:15 pm |
|
3 |
6:00 pm |
6:15 pm |
|
4 |
10:00pm |
10:15 pm |
4.8.4.9 Appendix 4B shows the updated traffic population. For Kai Hing Road and Hoi Bun
Road, the population projection for year 2025 and 2036 has been made based on ATC[12].
4.9
Risk
Summation
4.9.1.1 Risk summation combines the estimates of likelihood
and consequence for the hazardous events to produce the risk results, which are
expressed in terms of individual risk and societal risk, as per the TM-EIAO. For the LPG filling station, consequence
analysis and risk modelling were performed using PHAST Risk to give the
individual risk and societal risk results.
For KDGW, risk summation of the remodeled consequence and updated frequency
data (updated with potential helicopter impact frequency) as given in the previous
Kai Tak QRA [4] was carried out using IRESC risk summation tool.
4.9.2
Individual
Risk
4.9.2.1 The individual risk contours for KDGW and LPG
Filling Station are presented in Figure 4.7 and Figure 4.8, respectively.
The cumulative individual risk contours for the two facilities are presented in
Figure
4.9.
4.9.2.2 Despite the additional impact due to the proposal
helipad, the individual risk associated with KDGW and the LPG filling station
has been found to be acceptable with respective to the Individual Risk Criterion
that the off-site individual
risk does not exceed 1 × 10-5 / year.
4.9.3
Societal
Risk
4.9.3.1 The societal risk modelling for the LPG filling station
was carried out using PHASTRISK (with default parameters, except otherwise
specified), with the following assumptions on people’s exposure to hazard: the
exposed building population was limited to the maximum height of the hazards[4]; a shielding factor of 0.5[17] was considered for assessing the risks to people
located inside buildings which are within or partly within the fireball
diameter; and the road population immediately next to the firewall of the LPG
filling station was assumed to be protected from the fire scenarios.
4.9.3.2 The societal risk calculation for KDGW simply
combined the frequency, consequence data (note that some scenarios were remodelled
using PHAST) and population data, as reported in the previous sections of this
report, without further adjustment or processing of data.
4.9.3.3 As discussed in Section 4.6.1.2, the construction of the helipad does not involve
any heavy duty construction activities, and
operations of the two concerned hazard sources are not expected to be affected. Thus, the risk impact associated with the
construction phase is related to only the additional population of about 20
workers who will be located outdoors and exposed to
the two hazard sources.
4.9.3.4 The total population at NAH site is assumed to be 15,547, of which 777
is considered to be outdoors, in the maximum NAH capacity case (refer to Section 4.10.1.3 and Appendix 4B
for details). This number of people outdoors is 185 more than that considered
in the year 2036 Case (Figure 4.10 and 4.11), and is much higher than
that expected during helipad construction phase. As the 20 workers during
construction phase is considered to be well within the assumed population for
the maximum NAH capacity case, the risk level of the maximum NAH capacity case
can be taken as representative for construction phase, though conservative.
Even with this increase in the outdoor population, Figure
4.14 and Figure 4.15 show that
the risk level is still within the acceptable limit of Risk Guidelines given in
Annex 4 of the TM-EIAO. Hence, the risk associated with the construction phase
is deemed acceptable.
4.9.3.5 The societal risk is presented in the form of FN
Curves for comparison against societal risk criteria. It can be seen in Figure 4.10 and Figure 4.11 that the FN curves
for both the assessed years, namely 2025 and 2036, lie within the Acceptable
Region of the risk criteria. Figure 4.10 and Figure 4.11 also include an FN curve for a case without the
Project (i.e. without the Helipad Operation), in order to assess the
incremental risk due to the Project. In terms of Potential Loss of Life (PLL),
the helicopter crash scenario contributes to only 1.5% of the total PLL of
2.01×10-5 per year for the LPG filling station. For KDGW, the
helicopter crash scenario contributes to about 6.1% of the total of 6.82 ×10-6
per year.
4.10
Sensitivity Analysis
4.10.1.1
As described in Section 4.6.1.14, there are three helicopter flight paths associated
with the helipad. The previous base
case assessment has assumed that the total annual number of helicopter movement
is equally distributed among the three paths. To ensure the robustness of the
base case assessment, a sensitivity check has been performed, assuming that all
helicopter movements will be via only the flight path closest to the LPG
filling station or KDGW, and accordingly the maximum crash rates for KDGW and
LPG Filling Station have been estimated as 2.60×10-7 per year and
1.14×10-7per year, respectively. Given that the crash frequency is still
within the order of 10-7 per year, the 1×10-5 per year
individual risk contour is not affected due to the change and hence not
assessed further.
4.10.1.2
The FN curves
comparing the base case and sensitivity case for year 2036 are illustrated in Figure 4.12 and Figure 4.13 for the LPG filling
station and KDGW, respectively.
While the risk is observed to have notably increased in the sensitivity
case for both the facilities, it still remains within the acceptable region of
the risk criteria.
4.10.1.3 To assess the effect on NAH population, another
sensitivity case was performed applying the maximum design population capacity within
hospital[1]. The risk associated with both facilities remains
within the acceptable region of the risk criteria, with an increased in
fatalities. The FN curves comparing the base case and the maximum population
case for year 2036 are shown in Figure 4.14 and Figure 4.15 for the LPG filling
station and KDGW, respectively.
4.11
Risk Mitigation Measure
4.11.1.1 Professional trainings and guidelines should be
provided to the helicopter pilots in order to ensure the pilots be familiar
with the procedures to avoid approaching near the smoke plume in event of a
major fire accident at KDGW.
4.11.1.2 Since both individual risk and societal risk posed
by the proposed helipad at Acute Hospital meet the criteria of Hong Kong Risk
Guidelines, no further mitigation measures are required.
4.12
Conclusion
4.12.1.1 A QRA was conducted to evaluate the risk associated
with the proposed helipad at the Acute Hospital. The risk, both in terms of
individual risk and societal risk, has been found to be in compliance with the
risk criteria stipulated in Section 2 of Annex 4 of the TM-EIAO.
4.13
References
[1]
EIA Study Brief No. ESB-311/2019, Feb 2019
[2]
Technical Memorandum of Environmental Impact
Assessment Process (TM-EIAO)
[3]
Town Planning Board, Kai Tak Outline Zoning Plan
(S/K22/6)
[4]
Kai Tak Development Environmental Impact Assessment Report. Hong Kong, AEIAR-130/2009, 2006
[5]
A Rooftop Helipad at the Proposed New Block at Queen Mary Hospital Environmental Impact Assessment
Report. Hong Kong, (AEIAR-208/2017), 2017
[6]
Engineering Study for Police Facilities in Kong Nga Po – Feasibility
Study. Hong Kong, (AEIAR-201/2016),
2016
[8]
Kai Tak Development Engineering Study cum Design and Construction of
Advance Works – Investigation, Design and Construction, Agreement No: CE 35/2006 (CE)
[9]
Kwun Tong Line Extension (KTE): Hazard to Life Assessment for the
Transport Storage and Use of Explosives, AEIAR-154/2010, 2010
[10]
Census and Statistics Department, Hong Kong 2016 Population By-Census
[11]
Centadata Website, http:// www.centamap.com/gc/home/aspx
[12]
Hong Kong Transport Department, Annual Traffic Census 2018
[13]
New Acute Hospital at Kai Tak Development Area Sewerage Impact
Assessment Report, Drainage Services Department, 2019
[14]
Planning Department, Projection of Population Distribution 2019-2028
[15]
Operation of the Existing Tai Lam Explosives Magazine at Tai Shu Ha,
Yuen Long for Liantang / Heung Yuen Wai Boundary Control Point Project, Hong Kong, AEIAR-193/2015, 2015
[18]
Phone Interview on 17th September 2019
[19]
Meinhardt. Schedules,
Site Coverage and Plot Ratio Calculations. Drawing No.
NAH-A/WTP/AR/GBP-125, Rev. 0, Jan 2019
[21]
Workbook on the dispersion of dense gases, Britter
and MacQuaid, 1988
[22]
Proposed Join User Complex and Wholesale Fish Market at Area 44 Tuen Mun, Hong Kong, AEIAR-070/2003,
2003