.MEMO
To
Norman Collier, James Kaye
Cc
Graham O’Connel
From
Safety and System Performance, System Design
Date
18 February 2022
Subject
Auckland Harbour Bridge shared path safety assessment
Purpose
The project involves repurposing traffic lanes on the Auckland Harbour Bridge (AHB) for vulnerable
road users.
Recap
Mid last year the infrastructure management consultancy, Resolve Group, was tasked with
exploring possible options to utilise the existing structure to accommodate a shared path and
create a pedestrian and cycle link. Subsequently Resolve Group in total have explored 11 options
and provided a high-level input on indicative costing.
System Performance team and traffic modelling team at ASM have also carried out network
analysis and options were provided.
Learnings
•
Whilst the report identified 11 options, the options on the eastern side of AHB were
preferred due to impact on the wider network resulting from ramp closures that occur with
the western options.
•
Structural capacity is an issue for some options and refinement of the analysis is needed to
determine how the heavy vehicles will be managed for each option.
•
Further refined analysis of user safety systems should be undertaken before implementing
any option. These systems should work together to provide maximum protection for
shared path users.
Safe System Assessment (SSA)
under the Official Information Act 1982
•
A safe system assessment is an assessment of relative Death and Serious injury (DSi) risk
(between options) for commonly occurring crash types. As the SSA uses a relative scoring
system, it is used to compare options to each other, rather that used to meet a minimum
standard.
•
Using the SSA framework two options were assessed, Option 1: consisted of 1x south
bound lane, & Option 2: consisted of 2x southbound lane. The above two options were
compared against existing scenario.
•
Existing scenario acted as a baseline scoring (33/448) over 1 lane option (214/448) and 2
Released
lane option (202/448). Please refer to appendix for a detailed analysis carried out using
Austroads recommended SSAF.
•
Reasons for the deteriorated scores include:
1. Northern Express (NEX) bus services wil be forced to use bul run lanes, which are
far narrower than clip-on lanes in 2 lane repurposing option. Buses are prone to
yaw-steer (a well-known characteristic of single-box vehicles with a rear weight
bias), which increases their chances of colliding with the barrier or a bridge truss
member.
2. NEX bus services will have to operate under bull run lane condition between a
barrier and bridge truss even with the 1 lane repurposing option. Clip-on widths
were better compared to bridge lanes, however NEX will still have to operate in a
bull run conditions.
3. Cyclist vs cyclist head-on collision with impact speeds exceeding 60km/hr is highly
likely. According to Austroads standards bicycle operating speeds for a 5% downhill,
commuting traffic (85th percentile) could travel at 60km/hr downhil and 20km/hr
uphil , while recreational traffic would travel at 15km/hr downhil and 5km/hr
uphill.
4. Cyclist vs pedestrian high speed collision is highly likely due to the exposure
(increase in volume).
5. Cyclist vs cyclist sideswipe or Cyclist vs e-scooters/skateboarders sideswipe is
likely.
Exclusions from the SSA
1. All options involve severe weather risk, which cannot be assessed in the
rapid SSA. Severe weather conditions also present dynamic risks for vehicles,
cyclists and pedestrians. In this report cyclist aerodynamics is not
considered in detail, therefore factors like headwind, tailwind and
crosswinds are not discussed here in greater detail. Standards from UK
shows that it becomes unfavourable conditions for cyclists when winds are
over 20mph(32km/hr) and cycling becomes impossible at wind speeds of
40mph (64km/hr). As per Weather Watch website, Auckland City often have
windy weather lingering longer all year round simply because of the Roaring
Forties. Which is why Auckland has westerly or south-westerly winds/gusts
all around the year that poses constant threat of closing the AHB for cyclists.
Wild wing gust Incident on AHB in the month of September 2020 occurred
without a warning. Recurrence of such incidents wil cause greater
inconvenience to the vulnerable users who are already present on the AHB at
that instance. Barriers may help reduce the wind effects to some extent,
since winds get to 80km/hr speeds on AHB all around the year wil pose a
constant threat to operational safety of users.
2. Structural resilience is excluded from the rapid SSA – the reduction in the
under the Official Information Act 1982
number of trafficable lanes will increase the usage (live load) on the truss
structure and western side clip-on. The “two lanes” option will place the
clip-ons under an eccentric load.
Operational Safety Risks of the Proposal
Vehicle safety –
•
Less capacity to recover from congestion if there is a crash and less room for traffic
to pass a site if there is a crash, leading to increased queue lengths which will
necessitate more thorough and widespread delays. Likelihood of increase in
Released weaving crashes.
•
Two bull runs (in “one lane” option) can lead to reduced capacity and the onset of
sudden braking and increased rear-end crash risk.
Active mode safety –
•
Speed differentials between commuter traffic and e-cyclists downhills and
recreational traffic uphill increases the severity of crashes. There is a high chance of
a collision between opposing users in the one-lane scenario given the mixed
2
abilities of uphill users leading to passing in the uphill directions. This has already
led to a serious crash on the north western shared path and is one of the most
serious concerns here.
• 1xlane southbound option increases the risk of hitting roadside object at a speed
greater than 25km/hr.
• The high operating speeds likely with the active modes create a serious risk to
elderly and kids in collisions.
• Increased usage of micro-mobility (electric scooters and e-bikes) are capable of
high speeds which create behaviour and stability issues. There is a surge in e-
1982
scooters usage in Auckland, which currently is not quantified and assessed due to
lack of data. However, recent trends in pedestrian and e-scooter crashes are on the
rise.
• Any crash wil require containment and a 0.9m barrier will not contain a cyclist (due
Act
to a high centre of gravity) from overtopping the parapet or the dividing barrier
between traffic active modes. These barriers usual y required a minimum of 1.4m,
but given the extreme danger and height, 1.6m is the minimum recommended.
• Emergency service needs access to attend to crash victims or medical incidents.
• Barriers should be high enough s 9(2)(g)(i)
Infr
cture Safety –
• Structural resilience – the reduction in the number of trafficable lanes wil increase
the live load on the truss structure and the western side clip-on.
• There will be less lateral space for vehicle trajectory correction in the event of high
winds.
Information
• NEX running in bull run lane increase the likelihood of a collision with the barrier or
part of the truss structure.
• The 2x lane option will place the clip-ons under an eccentric load. With a higher
density of traffic on the north bound clip-on and next to no load on the south
bound clip-on, the fatigue resilience of the structures and their connections may
require careful assessment.
• Bridge needs to be structural y assessed to ensure the additional load of either
Official
concrete barrier or steel barrier wil be able to be supported by the bridge.
Network Impact
the
s 9(2)(g)(i)
to reduce vehicle volumes across the Harbour Bridge to a level
where the congestion caused by a 12.5% capacity loss (the one lane option) or a 25% capacity loss
(two lane option) would have minimal effect. Long term traffic volume reduction is more likely to
occur with a viable travel alternative for the displaced traffic, which the active mode facility is
unlikely to provide. The result is likely to be greater redistribution of traffic onto the SH18/16
corridor and onto less suitable local streets. There is therefore likely to be a consequential
under
reduction in safety through crashes.
Recommendation
This memo from Safety and System Performance team supports the Waka Board’s decision given
the complexities and consequences involved in repurposing the traffic lane. Major risks noted are
in structural resilience, network impact, and a significant risk of a death or serious injury of active
users. There are also unresolved operational risks of traffic management and incident
Released
management.
s 9(2)(g)(i)
3
Appendix A
Safe System Assessment Framework (SSAF)
The rapid SSAF process is based on Austroads Safe System Assessment Framework (Austroads
2016, Research Report AP-R509-16,
Safe System Assessment Framework). For this assessment
southbound (SBD) one lane vs SBD two lane option is compared.
AHB is an iconic structure that provides a critical function connecting Northshore to the City
Centre yet performing a role of national significant corridor. AHB in its current operational 1982
conditions is providing excellent safety benefits due to the disciplined operations achieved via
segregated direction of travel, barrier systems, straight alignment, ITS, signage and road
markings, and monitored 24/7.
Act
Austroads SSAF assessment needs to be modified slightly to meet the bespoke nature of the
proposal. Therefore, following assumptions or amendments made:
1. Runoff – no change in runoff road type of crashes for existing situation, option 1 and
option 2.
2. Head-on –
a. For existing, this is not applicable,
b. For option 1 and option 2 – cyclists head-on is considered. Under SSAF
Information
classification if impact speeds exceed 70 km/hr then a DSi is certain. Referring to
Austroads bicycle operating speeds chart for a 6% gradient it is noted that
commuter cyclists could reach up to 60km/hr speed on downhil . Recreational
cyclist or physically not so fit cyclist speeds going uphill assumed to be 15-
20km/hr. Narrow the SUP, risk of a head-on collision increases; wider the SUP
operational speed increases but the likelihood would be somewhat reduced due to
Official
the increased sightlines.
the
under
Released
3. Intersection – for existing arrangement, this is not applicable.
4
4. Other – weaving and lane changing is considered. Modifications to lanes be made northern
side of AHB to make the entry to the bicycle path safe. This includes tapered barriers and
termination of bus lane from its existing termination point.
5. Cyclist – two additional scenarios envisaged,
a. Side swiping – cyclist vs cyclist.
b. Impact with roadside barrier (25km/hr operational speed is assumed to cause a
serious injury to the cyclist), or cyclist vs e-scooter or skateboard.
1982
6. Pedestrian – pedestrian vs cyclists vs e-scooters.
7. Motorcycle – mostly remain unchanged for existing condition, which is straight alignment.
But in option 1 and option 2 scenarios, exposure and likelihood would increase due to the
Act
tapers added northern side of AHB.
The scoring exercise was carried out and the results are plotted in the chart below. For detailed
scoring refer to tables provided at the end of this report.
Overall, option 1 – converting one lane SBD into an SUP shows worst results due to head-on
likelihood and severity, and pedestrian vs cyclist crash severity.
Option 2- converting two lane SBD into a SUP still shows DSi risks associated with lane
changing, cyclists sideswiping, pedestrians and cyclists, cyclists and e-
scooters/skateboarders.
Information
Point to be noted here that the scoring is likely to increase due to the addition of tie-ins safety
scores.
Assessment:
The SSAF assessment is general y carried out on concepts to understand the nature of risks,
likelihood of a DSi crash occurring and severity of the crash. This would give early indication to
Official
the decision makers indication whether the solution is fit-for-purpose or not. A high scoring
defeats the purpose of the facility and a low scoring aligns to the Vision Zero strategy. In a
bespoke design these scores provides an early opportunity to focus on eliminating those safety
risks which has a potential to cause a DSi, and thus bringing the total score as close as
the
possible to Zero from a score of 448.
Safe System Assessment Framework
under
70
60
50
40
30
20
Existing
10
Opt1-1LSBD
0
Opt2-2LSBD
Released
5
Currently AHB is operating closer to that Zero value. By removing two lane SBD and repurposing it
to SUP would worsen the safety score. This assessment only looks at the safety aspects of the
options. Please note deduction of two lanes removes capacity of 1800veh/lane and adds that back
into the queue. It is hard to measure the impact of losing those two lanes and any incident on
WRR would add additional congestion onto the network.
The point highlighted here is that the temporary or interim solution that involves repurposing the
existing facility is not build for vulnerable users purpose and it cannot be tailored to a great
extent to make it safe for shared use path operations. In other words, there’ll be crashes involving
vulnerable users resulting in high severity outcome.
Key items excluded from the assessment:
1. In Option 2, North-western Express (NEX) buses will have to operate on narrow bull run lane.
This would significantly increase the risk of a double decker bus hitting the main truss under
the windy conditions. This will have a major implication to Auckland State Highway/Motorway
2. s 9(2)(g)(i)
3.
t carbon – greenhouse gas emissions (GHG) due to shockwaves created over the
network is hard to measure. SSAF don’t take into consideration benefits of active modes vs
motor vehicles.
4. Tie-in connections at Sulphur Beach, Tennyson underpass, Shel y Beach offramp, and
Westhaven Drive, are not assessed.
under the Official Information Act 1982
Released
6
SSA Matrix – Existing Conditions
Run- off
Head- on
Intersectio
Other (lane
Pedestrian
Cyclist
Motorcyclis
road
n
changing/w
ts
eaving)
Exposure
Chances are Chances of
There are
Either end of Peds and
Peds and
Motorcyclist
Comments:
low, there
a head-on
no
the AHB
cyclists
cyclists
s exposure
could be a
is removed
intersection
prohibited
restricted
possibility
due to
s that’d
weaving is
on AHB
on AHB
is slightly
of run-off.
controlled
enable a t-
High traffic
access.
boning
minimised
high
1982
volume.
crashes.
considering
severe
weather or
mechanical
Act
failure
Exposure
3/4
0/4
0/4
1/4
0/4
0/4
3/4
Score:
Although,
Likelihood
Likelihood
Likelihood of Peds and
Peds and
Observation
Likelihood
exposure
is nil.
is nil.
weaving on
cyclists
cyclists
shows
Comments:
from a
AHB is
prohibited
restricted
motorcyclist
runoff is
unlikely,
on AHB
on AHB
s impacting
high due to
vehicle lane
rear end of
speed but
discipline is
other
the
increased
vehicles,
likelihood
much before
but
of that
they enter
likelihood
happening
AHB
are low
is unlikely.
Information
Likelihood
2/4
0/4
0/4
1/4
0/4
0/4
2/4
Score:
Severity is
Not
Not
Severity in a
Not
Not
Severity is
Severity
medium
applicable
applicable
side swipe
applicable
applicable
considered
Comments:
high due to
crash is
low due to
impact
moderate.
insignificant
angles
data around
assumed in
severity.
worst case
Official
scenario
Severity
3/4
4/4
0/4
3/4
0/4
0/4
2/4
Score:
Product
18/64
0/64
0/64
3/64
0/64
0/64
12/64
the
(multiply
scores above
for crash
type)
TOTAL
33/448
under
Released
7
SSA Matrix –Proposed shared use path - one lane southbound clip-on
Run- off
Head- on
Other (lane
Side
Cyclist vs
Pedestrian
Motorcycli
road
Cyclist vs
changing/weavi
sweeping -
roadside
vs Cyclist
sts
Cyclist
ng) due to sup
Cyclist vs
hazard/barr
arrangement
Cyclist
ier
Exposure
High
High due
Either end of
Narrow lane
Barriers
Highly
Exposure
Comments:
traffic
to the
AHB exposure
increases the segregating
likely due
remains
450-850
volume
gradient
for lane
sup creates a to the
cyclists/day
and wind
changing is
exposure.
risk of
tourists/pe
same.
factor.
increased.
impact due
ds vs
100
1982
to narrow
commuter
peds/day
width
cyclists.
Exposure
3/4
4/4
¾
3/4
3/4
4/4
4/4
Score:
Remain
Highly
Onewa citybound Occurs
Moderate to
Highly
Remains
Act
Likelihood
same as
likely
onramp merger
during peak
low
likely
same as
Comments:
highlighte
based on
will have to be
hours and
occurrence - based on
above. It is
d in Table the crash
managed to
severe wind
commuter
the data
possible
1
trends
protect sup users weather
cyclists avoid and trends that
observed
using barriers
conditions/l
peds or in
on AKL’s
likelihood
elsewhere and lane
oss of ctrl.
severe
SUP
would
on AKL’s
modifications.
weather
increase
network
conditions
due to the
lane change
arrangemen
t upstream
and
downstrea
m of AHB
Information
Likelihood
2/4
3/4
2/4
2/4
2/4
4/4
3/4
Score:
Remain
Due to the Severity at inter-
Severity is
Any speed
Severity is
Severity is
Severity
same as
6%
peak or late
moderate if
over
high if an
moderate
Comments:
highlighte
downhill
night is
there are
25km/hr
elderly or a to high
d in Table gradient,
moderate to high multiple
would result
kid is hit
1
assuming
commuter
in serious
by a
a worst-
cyclists are
injury crash
commuter
case
involved.
cyclist
Official
scenario
travelling
of
at 60km/hr
60km/hr
downhill
cyclist
hitting
the
15km/hr
uphill
cyclist
would
increase
DSi risk
Severity
3/4
4/4
¾
2/4
3/4
4/4
3/4
Score:
under
Product
18/64
48/64
18/64
12/64
18/64
64/64
36/64
(multiply
scores
above for
crash type)
TOTAL
214/448
Released
8
SSA Matrix –Proposed shared use path - two lane southbound clip-on
Run- off
Head- on
Other (lane
Side
Cyclist vs
Pedestrian
Motorcycli
road
Cyclist vs
changing/weavi
swiping -
roadside
vs Cyclist
sts
Cyclist
ng) due to sup
Cyclist vs
hazard/barr
arrangement
Cyclist
ier
Exposure
Remain
Moderatel
Either end of
Wider lane
Wider path
Highly
Exposure
Comments:
same as
y high due AHB exposure
mitigates
reduces the
likely due
remains
450-850
highlighte
to the
for lane
exposure
to the
cyclists/day
d in Table gradient
changing is
exposures
given wider
tourists/pe
same.
1
and wind
increased
space
ds vs
100
however
factor
commuter
peds/day
increases the
cyclists
speed
Exposure
3/4
3/4
¾
2/4
2/4
4/4
4/4
Score:
Remain
Likely,
Onewa citybound Occurs
Moderate to
Highly
Remains
Likelihood
same as
based on
onramp merger
during peak
low
likely
same as
Comments:
highlighte
the crash
will have to be
hours and
occurrence - based on
above. It is
d in Table trends.
managed to
severe wind
commuter
the data
possible
1
protect sup users weather
cyclists avoid and trends that
using barriers
conditions/l
peds or in
on AKL’s
likelihood
and lane
oss of ctrl.
severe
SUP
would
modifications. PT Also
weather
increase
buses running in increases the conditions
due to the
bull run creates a likelihood of
lane change
likelihood
cyclist
arrangemen
scenario of
sideswiping
t upstream
hitting the truss
skateboarder
and
member in
s or e-
downstrea
severe weather.
scooters
m of AHB
Likelihood
2/4
2/4
4/4
2/4
2/4
4/4
3/4
Score:
Remain
Due to the Severity at inter-
Severity is
Any speed
Severity is
Severity is
Severity
same as
6%
peak or late
moderate to
over
high if an
moderate
Comments:
highlighte
downhill
night is
high due to
25km/hr
elderly or a to high
d in Table gradient,
moderate to low. the speeds
would result
kid is hit
1
assuming
in serious
by a
a worst-
injury crash
commuter
case
cyclist
scenario
travelling
of
at 60km/hr
60km/hr
downhill
cyclist
hitting
15km/hr
uphill
cyclist
would
increase
DSi risk
under the Official Information Act 1982
Severity
3/4
4/4
¾
3/4
3/4
4/4
3/4
Score:
Product
18/64
24/64
36/64
12/64
12/64
64/64
36/64
(multiply
scores
above for
crash type)
TOTAL
202/448
Released
9
