Rail Accident Report
Collision between a passenger train and a fallen
tree at Broughty Ferry, Dundee
27 December 2023
Report 13/2024
December 2024
This investigation was carried out in accordance with:
• the Railway Safety Directive 2004/49/E C
• the Railways and Transport Safety Act 2003
• the Railways (Accident Investigation and Reporting) Regulations 2005.
© Crown copyright 2024
You may reuse this document/publication (not including departmental or agency logos) free of
charge in any format or medium. You must reuse it accurately and not in a misleading context.
The material must be acknowledged as Crown copyright and you must give the title of the
source publication. Where we have identified any third party copyright material you will need
to obtain permission from the copyright holders concerned. This document/publication is also
available at www.gov.uk/raib.
Any enquiries about this publication should be sent to:
RAIB Email: enquiries@raib.gov.uk
The Wharf Telephone: 01332 253 300
Stores Road Website: www.raib.gov.uk
Derby UK
DE21 4BA
This report is published by the Rail Accident Investigation Branch, Department for Transport.
Preface
Preface
The purpose of a Rail Accident Investigation Branch (RAIB) investigation is to
improve railway safety by preventing future railway accidents or by mitigating their
consequences. It is not the purpose of such an investigation to establish blame or
liability. Accordingly, it is inappropriate that RAIB reports should be used to assign
fault or blame, or determine liability, since neither the investigation nor the reporting
process has been undertaken for that purpose.
RAIB’s findings are based on its own evaluation of the evidence that was available at
the time of the investigation and are intended to explain what happened, and why, in a
fair and unbiased manner.
Where RAIB has described a factor as being linked to cause and the term is
unqualified, this means that RAIB has satisfied itself that the evidence supports both
the presence of the factor and its direct relevance to the causation of the accident or
incident that is being investigated. However, where RAIB is less confident about the
existence of a factor, or its role in the causation of the accident or incident, RAIB will
qualify its findings by use of words such as ‘probable’ or ‘possible’, as appropriate.
Where there is more than one potential explanation RAIB may describe one factor as
being ‘more’ or ‘less’ likely than the other.
In some cases factors are described as ‘underlying’. Such factors are also relevant
to the causation of the accident or incident but are associated with the underlying
management arrangements or organisational issues (such as working culture).
Where necessary, words such as ‘probable’ or ‘possible’ can also be used to qualify
‘underlying factor’.
Use of the word ‘probable’ means that, although it is considered highly likely that the
factor applied, some small element of uncertainty remains. Use of the word ‘possible’
means that, although there is some evidence that supports this factor, there remains a
more significant degree of uncertainty.
An ‘observation’ is a safety issue discovered as part of the investigation that is not
considered to be causal or underlying to the accident or incident being investigated,
but does deserve scrutiny because of a perceived potential for safety learning.
The above terms are intended to assist readers’ interpretation of the report, and to
provide suitable explanations where uncertainty remains. The report should therefore
be interpreted as the view of RAIB, expressed with the sole purpose of improving
railway safety.
Any information about casualties is based on figures provided to RAIB from various
sources. Considerations of personal privacy may mean that not all of the actual effects
of the event are recorded in the report. RAIB recognises that sudden unexpected
events can have both short- and long-term consequences for the physical and/ or
mental health of people who were involved, both directly and indirectly, in what
happened.
RAIB’s investigation (including its scope, methods, conclusions and recommendations)
is independent of any inquest or fatal accident inquiry, and all other investigations,
including those carried out by the safety authority, police or railway industry.
Report 13/2024 December 2024
Broughty Ferry
This page is intentionally left blank
Report 13/2024 4 December 2024
Broughty Ferry
Collision between a passenger train and a fallen
tree at Broughty Ferry, Dundee, 27 December
2023
Contents
Preface3
Summary7
Introduction8
Definitions 8
The accident9
Summary of the accident 9
Context10
The sequence of events14
Events preceding the accident 14
Events during the accident 15
Events following the accident 16
Analysis17
Identification of the immediate cause 17
Identification of causal factors 17
Identification of underlying factor 32
Factors affecting the severity of consequences 35
Observation37
Previous occurrences of a similar character 38
Summary of conclusions 39
Immediate cause 39
Causal factors 39
Underlying factors 39
Factors affecting the severity of consequences 39
Observation40
Previous RAIB recommendation relevant to this investigation41
Actions reported as already taken or in progress relevant to this report43
Actions reported that address factors which otherwise would have resulted in
an RAIB recommendation 43
Other reported actions 43
Report 13/2024 5 December 2024
Broughty Ferry
Recommendations44
Appendices46
Appendix A - Glossary of abbreviations and acronyms 46
Appendix B - Investigation details 47
Report 13/2024 6 December 2024
Broughty Ferry
Summary
Summary
At around 13:09 on 27 December 2023, the 10:46 Perth to Aberdeen passenger
service collided with a fallen tree approximately 1 mile (1.6 km) east of Broughty Ferry,
Dundee. The train was travelling at around 84 mph (135 km/h) when the collision
occurred. The train suffered significant damage to the leading driving cab. There were
no physical injuries to the 37 passengers and three staff members on board the train.
The tree had fallen from Barnhill Rock Gardens, a public park owned by Dundee City
Council, and was brought down by winds during Storm Gerrit. This storm had been
subjecting the area to high winds and heavy rain for several hours preceding the
accident. RAIB’s investigation found that the soil in which the tree was rooted had
characteristics which limited the tree’s ability to resist the wind forces acting on it. In
addition, three other trees at this location had been felled before May 2023, increasing
the exposure of the tree which fell to winds from the Firth of Tay.
Around 12 minutes before the collision, a member of the public became aware that
a tree had fallen across the railway and contacted Network Rail using the public
helpline. The helpline call handler attempted to pass this information on to Network
Rail’s Scotland route control on a number of occasions, but the call from the helpline
call handler was not answered until after the accident. This meant that a warning
about the fallen tree did not reach the driver of the train in time to prevent the accident.
The risk of trees in Barnhill Rock Gardens falling onto the railway not being
effectively controlled was the factor underlying the accident. Network Rail is reliant
on neighbouring landowners controlling the risk associated with visually healthy trees
falling onto the railway lines from outside of the railway boundary. However, Dundee
City Council did not effectively manage the risk of trees falling from its land onto the
adjacent railway lines.
As a consequence of the accident, the survival space in the cab was considerably
reduced. The driver only escaped serious injury by crouching behind the driving
seat once they had made an emergency brake application on realising the collision
was inevitable. RAIB also observed that the telephone equipment used at Scotland
integrated control centre did not display missed call information.
Since this accident, Network Rail has provided helpline staff with an additional contact
telephone number for use in emergencies.
RAIB has made three recommendations as a result of its investigation. The first of
these is to Network Rail to consider how technology could assist in the detection
of trees subject to altered exposure, including those trees on third-party land. The
second recommendation is that Dundee City Council should review its management of
the trees for which it is responsible to ensure that it is effectively controlling the risk of
them falling onto the railway.
RAIB has also recommended that the Rail Safety and Standards Board’s Carmont
recommendations steering group should review its response to recommendation 19
made within RAIB report 02/2022, following the investigation into the derailment of a
passenger train at Carmont, Aberdeenshire on 12 August 2020.
Report 13/2024 7 December 2024
Broughty Ferry
Introduction
Introduction
Definitions
1 Metric units are used in this report, except when it is normal railway practice to
give speeds and locations in imperial units. Where appropriate the equivalent
metric value is also given. Left and right relate to the train’s direction of travel.
2 The report contains abbreviations and acronyms, which are explained in
appendix A. Sources of evidence used in the investigation are listed in
appendix B.
Report 13/2024 8 December 2024
Broughty Ferry
The accident
The accident
Summary of the accident
3 At around 13:09 on 27 December 2023, the 10:46 Perth to Aberdeen passenger
service collided with a tree which had fallen onto the line from outside the railway
boundary approximately 1 mile (1.6 km) east of Broughty Ferry, Dundee. The
train, reporting number 1A37, was returning south to Dundee because flooding
had closed the line further north.
4 A member of the public had contacted Network Rail about the fallen tree
12 minutes before the collision, but this warning did not reach the driver in time to
stop the train and prevent the accident.
5 The train was travelling at around 84 mph (135 km/h) when the collision occurred,
and the driver only escaped serious injury by crouching behind the driving seat.
The train suffered significant damage to the leading driving cab. This disabled
the train and prevented access to the driver’s control desk and communications
equipment.
6 Shortly after the collision, the signaller at Dundee signalling centre received an
automated alarm from the train. The signaller attempted to contact the driver but,
when the call connected, the signaller heard only engine noise. By this time, the
driver had walked back along the train to ask the conductor to report the accident.
After meeting partway along the train, the driver asked the conductor to make an
emergency call. The driver also reported the accident to the Arbroath signaller by
mobile telephone.
7 There were no injuries to the 37 passengers and three staff members on board,
but it was necessary for the fire service to assist in the evacuation of the train.
The evacuation was completed 2 hours after the train had come to a stand.
100%
Location of accident
Contains Ordnance Survey data: @Crown Copyright and database right 2024.
OS license number: AC0000833184. Source: Department for Transport, RAIB 2024
Figure 1: Extract from Ordnance Survey map showing the location of the accident near Broughty Ferry,
a suburb of Dundee, Scotland.
Report 13/2024 9 December 2024
Broughty Ferry
Context
The accident
Location
8 The collision occurred approximately 1 mile (1.6 km) east of Broughty Ferry
station on the Dundee to Aberdeen line. Travelling south from Aberdeen, the
railway broadly follows the eastern coastline of Scotland before turning inland
through Arbroath and following the northern shoreline of the Firth of Tay passing
through Carnoustie, Broughty Ferry and onwards to Dundee (figure 2).
Inverness
Aberdeen
Broughty Ferry
Dundee
Arbroath
Location of
Glasgow accident
Edinburgh
Figure 2: Railway context.
9 At the location the railway is made up of two lines. The up line is used by trains
heading towards Dundee and the down line is used by trains heading towards
Arbroath, Aberdeen and beyond to Inverness.
10 Train 1A37 was travelling on the up line from Arbroath. On the approach to
the fallen tree, the line passes under a road bridge and curves to the right.
These features limit the view of the location where the tree fell to a maximum of
240 metres, although at the time of the accident this was probably further reduced
by adverse weather conditions. The maximum permitted train speed at the
location was 90 mph (145 km/h).
11 The tree had fallen from within Barnhill Rock Gardens, part of a public park
situated to the left of the railway between the railway boundary and the shore of
the Firth of Tay. The gardens include tree species which are native and non‑native
to the United Kingdom. The tree that fell onto the railway was a Monterey cypress
tree which grows naturally in North America (figure 3).
Report 13/2024 10 December 2024
Broughty Ferry
The accident
Figure 3: Google Street View image of Barnhill Rock Gardens (courtesy of Google).
Organisations involved
12 ScotRail Trains Limited (ScotRail) was the operator of the train and is the
employer of the driver and conductor who formed the train crew on board train
1A37.
13 Angel Trains Limited (Angel Trains), a rolling stock leasing company, leased the
rail vehicles which formed train 1A37 to ScotRail.
14 Network Rail is the owner and maintainer of the infrastructure at this location,
which forms part of its Scotland route, and Scotland’s Railway region. Network
Rail also employs the signallers and the Scotland integrated control centre (SICC)
staff who were on duty at the time of the accident.
15 Journeycall Limited (Journeycall) is contracted by Network Rail to provide
customer service resources including handling of calls to the Network Rail
public helpline number. Journeycall also handles enquires for other non-railway
companies and therefore its staff are not railway experts.
16 Dundee City Council (DCC) owns the public Barnhill Rock Gardens from where
the tree fell onto the railway. The forestry office, part of the DCC Environmental
Department, is responsible for maintenance of the trees within the gardens.
17 ScotRail, Angel Trains, Network Rail, Journeycall and DCC all freely co-operated
with the investigation.
Report 13/2024 11 December 2024
Broughty Ferry
Train involved
The accident
18 Train 1A37 was a high speed train (HST) set. This comprised four mark 3
passenger coaches with a class 43 diesel-electric power car at each end
(figure 4). Power car 43129 was leading and power car 43133 was at the rear of
the train when the accident occurred. These vehicles were originally constructed
by British Rail Engineering Ltd and entered service between 1976 and 1980. The
construction and approval into service of HSTs pre-dates a number of modern
railway standards relevant to crashworthiness and, because rolling stock does
not require ongoing permission to remain in operation, HSTs remain in passenger
service on the mainline rail network.
19 The power cars are each fitted with forward-facing closed-circuit television
(FFCCTV) and an on-train data recorder (OTDR). The FFCCTV recorders
store short periods of video in a short-term memory before saving it. During the
collision, the FFCCTV equipment fitted to the leading power car lost power and
did not save the contents of its short-term memory. As a consequence, the last
seconds of video before the collision were not retrievable. The train was also fitted
with a remote monitoring system which recorded the location of the train using
satellite positioning technology.
Figure 4: A typical ScotRail HST with power cars and mark 3 coaches (courtesy of ScotRail).
Staff involved
20 The driver of train 1A37 started as a conductor in 2011 and began training as a
driver in 2018. They qualified as a driver in December 2019.
21 The call handler who dealt with the call from the member of the public joined
Journeycall in February 2023 and had been trained to handle Network Rail public
helpline enquires from August 2023.
Report 13/2024 12 December 2024
Broughty Ferry
22 The Network Rail route control manager (RCM), who was on duty at SICC during
The accident
the period immediately leading up to the accident, had 33 years’ operational
railway experience and 6 years’ experience in the RCM role.
23 The DCC senior tree officer had 45 years’ experience as a forestry officer and
holds a higher national diploma qualification from the Scottish School of Forestry.
External circumstances
24 At the time of the accident, the Met Office had issued a weather warning
associated with named Storm Gerrit. This warning was for high winds and heavy
rain which extended across the country. The subsequent Met Office report
stated that ‘Storm Gerrit brought damaging winds and heavy rain to the United
Kingdom from 27 to 28 December with Wales, north-west England and Scotland
worst affected. In the most exposed locations, winds gusted at over 70 Knots
(130 km/h) while heavy rain led to increased flooding concerns.’ The maximum
gusts recorded during the storm are shown in figure 5.
Accident location
Figure 5: Met Office chart indicating 53 Knots (98 km/h)
maximum wind gusts associated with Storm Gerrit near
accident location (courtesy of the Met Office).
Report 13/2024 13 December 2024
Broughty Ferry
The sequence of events
The sequence of events
Events preceding the accident
25 Network Rail had received a weather forecast from the Met Desk predicting
extreme weather on 27 and 28 December 2023 associated with Storm Gerrit.
On the morning of the accident, the forecast included red (extreme) warnings for
heavy rain and strong winds effecting much of Scotland. To mitigate the hazards
associated with these weather conditions, a decision was made to apply blanket
speed restrictions (BSRs) in the worst affected areas. A BSR had been applied
between Carnoustie and Aberdeen in anticipation of the heavy rain causing
flooding along the coastal section of the route to be taken by train 1A37. No BSR
was in place on the route between Dundee and Aberdeen, on which the accident
occurred (see paragraph 81).
26 Train 1A37 departed Perth on time at 10:46 on 27 December and began its
journey northwards towards Aberdeen. The train passed through Broughty
Ferry station on the down line at 11:24 and, a few minutes afterwards, passed
the location where the tree would later fall. At that time nothing untoward was
reported by the driver of the train. Train 1A37 was the last train to pass through
the area in either direction before the accident.
27 As the train approached Carnoustie station, around 6 miles (9.6 km) from
Arbroath, the driver reduced the train’s speed in accordance with the BSR. The
driver complied with the 40 mph (64 km/h) BSR until coming to a stand at a red
(stop) signal at Arbroath station around 11:39. The Arbroath signaller informed
the driver via the train’s GSM-R (Global System for Mobile Communications
– Railway) radio system that it was necessary to return the train to Dundee
following reports of flooding closing the line ahead. After a short wait, the train
was driven into a siding north of Arbroath and signalled back into the station on
the up line for the return journey. Because no other hazards had been identified
on the inland portion of the route, the driver was not given any special instructions
by the signaller as to how to proceed (see paragraph 81). The train departed
Arbroath and headed back towards Dundee on the up line at 12:54.
28 At 12:57, shortly after the train had departed Arbroath, a member of the public
contacted the Network Rail public helpline. The member of the public reported
that a tree had fallen across the tracks from Barnhill Rock Gardens located on the
opposite side of the railway to their property. During this telephone conversation,
the call handler attempted to contact the SICC, but the call was not answered.
The call handler attempted to contact the SICC again after completing the call
with the member of the public, but each time the call remained unanswered.
29 After departing Arbroath, the train travelled at 40 mph (64 km/h) until reaching
the end of the BSR where the driver increased the speed of the train. The train
reached 96 mph (155 km/h) and travelled around 3.5 miles (5.6 km) before the
driver shut off traction power and began to slow down to meet a reduction in
permissible speed from 100 mph (161 km/h) to 90 mph (145 km/h) which started
around 1 mile (1.6 km) before the accident location.
Report 13/2024 14 December 2024
Broughty Ferry
Events during the accident
The sequence of events
30 The train was negotiating a right-hand curve and passing under a road bridge
when the driver became aware of the tree across the track ahead. The driver
made an emergency brake application around 80 metres before the collision.
Realising that an accident was unavoidable, the driver then crouched on the floor
behind the driving seat.
31 According to OTDR data, the collision occurred at 13:09 while the train was
travelling at 84 mph (135 km/h). The train did not derail and continued to travel
on the up line until it came to a stand approximately 850 metres beyond the fallen
tree.
32 The stem of the tree, commonly known as the trunk, was broken into two lengths
during the collision. The lower portion was around 9 metres in length and after
being uprooted it remained where it fell. The upper portion, which was around
6 metres long, was thrown through the boundary fence and landed approximately
20 metres forwards in the direction of travel of the train and to the left of the line.
33 During the collision, the tree stem entered the cab area around driver’s eye level,
passing through the cab pillars either side of the windscreen, before cutting
through the body of the cab, the quarter light window and the driver’s door. The
tree stem’s progression through the driving cab stopped just above the driver as
it reached the bulkhead structure separating the driving cab from the power car
equipment compartment behind the cab (figure 6).
Figure 6: Damage sustained to left‑hand side of driving cab.
Report 13/2024 15 December 2024
Broughty Ferry
34 The driver was showered in glass and other debris by the impact but escaped
The sequence of events
serious injury. The windscreen had shattered and was lying across the cab
controls and, after clearing enough of the debris to be able to stand, the driver
realised it was not possible to reach the GSM-R radio to make an emergency call.
The driver left the cab and began to walk back along the train to understand the
extent of the damage and to liaise with the train’s conductor.
35 Around 1 minute after the train had come to a stand, the signaller at Dundee
signalling centre received an automated emergency alarm generated by
equipment on board the damaged train. The Dundee signaller attempted to
contact the driver on the GSM-R radio system but, when the call connected,
the signaller was only able to hear engine noise because the driver had left the
driving cab.
Events following the accident
36 After meeting partway along the train, the driver asked the conductor to make an
emergency call. The driver also reported the accident to the Arbroath signaller by
mobile telephone. Around this time, a second train driver, who had been travelling
as a passenger on the train, also joined the driver and conductor. The second
driver agreed to assist the driver to apply the parking brake to secure the train.
The two drivers then made their way forwards and together managed to move
enough debris to enable them to shut down the engine in the front power car. The
engine in the rear power car was kept running to maintain an air supply to the
train’s braking system and power to the communications, heating, lighting and
other facilities for the passengers.
37 After ensuring the lines were blocked to other trains, the two drivers returned to
the conductor to assist looking after the 37 passengers reported to be on board. A
rail incident officer appointed by Network Rail arrived on site at 14:29 to facilitate
the evacuation of the passengers with assistance from the fire and rescue
service, as well as prepare for the recovery of the stranded train.
38 To evacuate the passengers, the fire and rescue service provided a ladder to
reach the ground from one of the carriage doors and cut an access route through
the railway boundary fence into an adjacent public park. The evacuation took
around 17 minutes, with the last passengers leaving the train around 2 hours
after the collision. Around 6 hours after the evacuation, the damaged train was
recovered to sidings near to Dundee station by an assisting HST.
Report 13/2024 16 December 2024
Broughty Ferry
Analysis
Analysis
Identification of the immediate cause
39 Train 1A37 was unable to stop before colliding with a fallen tree that was
obstructing the railway.
40 When driving back to Dundee, having been turned back at Arbroath because of
flooding closing the line ahead, the driver of train 1A37 complied with the 40 mph
(64 km/h) BSR which was in place on the journey to Carnoustie (paragraph 25).
After passing through the limits of the BSR, and with no additional instruction to
drive at a reduced speed, the driver obeyed the permanent speed restrictions for
the route (see paragraph 81).
41 The train passed under a road bridge on the approach to Barnhill Rock
Gardens and entered a right-hand bend. At this point, the train was travelling at
approximately 84 mph (135 km/h) and the driver was allowing it to coast to keep
within the permitted line speed of 90 mph (paragraph 29). Analysis undertaken
by RAIB shows that the driver would have had a maximum of around 6 seconds
sighting of the fallen tree in good conditions. However, on the day of the accident,
heavy rain had been falling throughout the journey from Arbroath. This would
have reduced the driver’s visibility of obstructions ahead and the time available
for them to perceive any hazards and react to them. This is discussed further in
paragraphs 91 to 93.
42 OTDR data shows that the driver commanded an emergency brake application
around 2 seconds before the collision occurred, when the train was 80 metres
from the tree. This left insufficient time and distance for the braking application to
reduce the speed of the train.
Identification of causal factors
43 The accident occurred due to a combination of the following causal factors:
a. A tree on adjacent land and in close proximity to the railway boundary fell
across the lines (paragraph 44).
b. Notification of the obstruction did not reach the driver following a call from a
member of the public telling the railway that the tree had fallen across the lines
(paragraph 72).
c. On sighting the fallen tree, the driver was unable to stop the train before
colliding with it (paragraph 81).
Each of these factors is now considered in turn.
Fallen tree on the railway lines
44 A tree on adjacent land and in close proximity to the railway boundary fell
across the lines.
45 During the time between train 1A37 travelling northwards past Barnhill Rock
Gardens, and its return journey south, a tree had fallen across both lines. The tree
had fallen from within the gardens and was lying horizontally across both lines at
around windscreen height of the approaching train (paragraph 33).
Report 13/2024 17 December 2024
Broughty Ferry
46 The tree was a Monterey cypress (Cupressus macrocarpa), a species only
Analysis
native to the coastal areas of California, which was growing in a small group
of trees of the same species. Historic imagery from the National Collection of
Aerial Photography shows no trees in the location of the fallen tree in 1941 and a
collection of established trees in 1969. Barnhill Rock Gardens is the responsibility
of DCC and its website states that the garden was started in 1955 on the site of
a former golf course. Therefore, it is likely the tree was planted in the gardens
sometime between 1955 and 1969.
Figure 7: Site of the future Barnhill Rock Gardens in 1941 (courtesy of NCAP/ncap.org.uk).
Figure 8: Barnhill Rock Gardens in 1969 with RAIB annotation (courtesy of NCAP/ncap.org.uk).
Report 13/2024 18 December 2024
Broughty Ferry
47 The tree was located about 15.5 metres away from the railway boundary. The
Analysis
impact occurred around 17.5 metres from where the lower section of the tree’s
stem would have entered the ground and where the stem was approximately
300 mm in diameter. The tree was severely damaged in the collision, with the
stem of the tree being broken into two parts (paragraph 32). The two stem
sections together measured approximately 21 metres in length, but it is likely that
with the leaf canopy intact the tree would have stood taller than this. An exact
measurement was not possible as almost all the branches were stripped from
the stem either by the impact or work undertaken to clean up the site after the
collision.
48 RAIB engaged the services of an expert arboriculturist to examine the remaining
parts of the tree after the accident. The expert concluded that the tree was healthy
with no signs of disease or decay that would account for it falling onto the railway.
Further examination of the roots indicated that the tree had suffered windthrow.
This is where a tree has been uprooted by wind forces acting on it which are
greater in magnitude than the tree can withstand.
49 In this case, the tree fell due to windthrow because of a combination of the
following factors:
a. The soil at this location had poor mechanical adhesion characteristics which
limited the ability of the tree to resist the wind forces acting on it (paragraph
50).
b. The tree had been subjected to increased wind loading due to altered
exposure (paragraph 58).
c. The wind loading acting on the tree was elevated by the high winds associated
with Storm Gerrit (paragraph 65).
Each of these factors is now considered in turn.
Soil adhesion
50 The soil at this location had poor mechanical adhesion characteristics
which limited the ability of the tree to resist the wind forces acting on it.
51 The Monterey cypress is a coastal species adapted to thrive in its natural habitat,
including withstanding high winds and sandy soil. When wind is stopped by the
surface of an object, a pressure is generated. This creates a force which acts on
the object. Anchor roots provide most of a tree’s stability to resist the forces acting
upon it, including those generated by wind. Anchor roots typically only extend
1 to 2 metres beyond the stem and form a mass known as the mechanically
active rootplate (MAR) (figure 9).
52 Roots broadly perform three functions. These are anchorage, harvesting of soil
water and dissolved nutrients, and the transportation of those nutrients and water.
Harvester roots are very fine and are mainly without bark. Transport roots are
much thicker and extend beyond the MAR into the surrounding soil and provide
a means of transporting the nutrients and water absorbed by the harvester roots
back to the tree. The area of transition between the MAR and the transport roots
is known as the ‘zone of rapid taper’.
Report 13/2024 19 December 2024
Broughty Ferry
Analysis
Mechanically active
rootplate (MAR)
Tree stem
Figure 9: Fallen tree before work to clear the site.
53 Failure of a tree by windthrow generally occurs if the roots are unable to resist the
forces acting on the tree by one or a combination of the following modes:
1. a loss of anchor root integrity inside the radius of the MAR (usually due to
decay or by root severing, for example, by cutting during excavation works)
2. root breakage at the zone of rapid taper caused by wind loads exceeding the
shear strength of the roots causing them to fracture, together with a loss of soil
shear strength, resulting in rootplate rotation
3. a loss of soil adhesion, with a number of the transport roots pulling out of the
soil, rather than most or all roots fracturing (as would be seen in mode 2).
54 Breakage of some roots at the edge of the tree’s MAR indicated a wind load
which had exceeded the shear strength of those roots and the surrounding soil.
This loss of the soil shear strength had allowed the rootplate to rotate and apply a
greater force on the transport roots extending beyond the MAR. These transport
roots remained intact and were pulled through the soil as the rootplate rotated.
This is an indication of poor mechanical adhesion within the soil (figure 10).
55 Examination of the soil depression left by the uprooted tree showed that the
tree was located on soil made predominantly of sand. Sand is an inherently
non‑cohesive soil with poor mechanical properties for tree root adhesion. Water
within the soil can also reduce the mechanical strength of a non-cohesive soil and
the ability of a tree root to adhere to it.
Report 13/2024 20 December 2024
Broughty Ferry
Analysis
Unbroken transport
roots pulled out of soil
Mechanically active
rootplate (MAR)
Sand back filling
depression left by MAR
Figure 10: Uprooted tree showing MAR, transport roots and sandy soil back fill.
56 The fallen tree had been established in a small group of trees of the same
species. It was apparent from tree stumps remaining in the ground that four other
trees from the group had been lost, including one which had suffered windthrow.
Although this work was not documented or recorded by DCC, the council stated
that two of these removed trees had suffered storm damage, one had been
removed due to its proximity to an adjacent glasshouse and it was necessary
to remove the fourth tree due to suppressed growth. The expert arboriculturist
assessed the rest of Barnhill Rock Gardens and found three other trees (two
pine trees and a eucalyptus tree) which had suffered complete windthrow and
estimated that this had occurred in the last 5 years. The expert arboriculturist
concluded this was indicative of an area where poor soil adhesion exists.
57 When exposed to extremes of wind loading, a tree will follow a strategy of
progressive collapse. A tree will first shed leaves, twigs and small branches
ahead of catastrophic stem breakage or uprooting for as long as the soil provides
sufficient root adhesion. Several trees suffering complete windthrow in a localised
area can be an indication of root disease or decay. Where no signs of ill health
exist, as found at Barnhill Rock Gardens, this can be an indication that the trees
have insufficient support from the soil for the wind loading conditions.
Report 13/2024 21 December 2024
Broughty Ferry
Altered exposure
Analysis
58 The tree had been subjected to increased wind loading due to altered
exposure.
59 The fallen tree was located within a group of trees and for most of its life it had
been in a position nearest to the railway and furthest from the Firth of Tay. As the
tree developed, three of the four companion trees, which had been recently lost
(paragraph 56), would have provided shelter from winds from the Firth of Tay.
60 Google Earth and Street View images show how this group of trees has changed
over a 12-year period. It can be seen in the satellite image taken of the group in
June 2018 that the group of trees was well established. By May 2023, Google
Earth images show that the group had been much reduced (figures 11 and 12).
The Street View images capture this change from ground level and show how
the trees on the southern side of the group provide shelter from the ground up to
those trees behind (figures 13 and 14).
Direction of train 1A37
Accident tree
Figure 11: Google Earth image showing the group of trees in June 2018 (courtesy of Google with RAIB
annotations).
Report 13/2024 22 December 2024
Broughty Ferry
Analysis
Direction of train 1A37
Accident tree
Figure 12: Google Earth image showing the group of trees May 2023 (courtesy of Google with RAIB
annotations).
Accident tree
Figure 13: Google Street View image showing the group of trees in July 2012 (courtesy of Google with
RAIB annotations).
Report 13/2024 23 December 2024
Broughty Ferry
Analysis
Accident tree
Direction of
southeasterly wind
Figure 14: Google Street View image showing the group of trees in May 2023 (courtesy of Google with
RAIB annotations).
61 These historical images also show that the fallen tree had thicker growth in the
upper foliage, but thinner growth further down the stem. This is characteristic of
a tree growing in the shelter of companion trees. These images also show how
the fallen tree had been left standing apart from the rest of the group following
the loss of the companion trees. This loss of shelter meant the tree was no longer
receiving shelter from winds blowing from a southeasterly direction over the Firth
of Tay.
62 Network Rail uses aerial images of the railway to assist in its management of the
infrastructure. These images extend outside of the railway boundary and, in this
case, include images of the tree before it fell. The aerial images show that the
tree had been growing with a natural inclination towards the railway, but that the
angle of inclination had remained unchanged between April 2022 and May 2023
(figures 15 and 16). A tree to the north of a group such as this will naturally grow
away from trees to its south and towards the better light.
Report 13/2024 24 December 2024
Broughty Ferry
Analysis
Fallen
companion tree
leaning against
rest of group
Area of
Companion tree Remains of ground where Angle of
broken 6-8m third tree companion accident tree
above ground trees have not changed
been removed
Figure 15: Network Rail aerial image showing the Figure 16: Network Rail aerial image showing the
trees in April 2022 (courtesy of Network Rail with trees in May 2023 (courtesy of Network Rail with
RAIB annotations). RAIB annotations).
63 The Network Rail aerial image dated 19 April 2022 (figure 15) shows two of the
companion trees and the storm damage they had suffered (paragraph 59). One
of the trees had suffered windthrow and fallen into another tree in the group. The
other tree suffered stem breakage approximately 6 to 8 metres above ground
level. The image taken in May 2023 (figure 16) shows that both trees have
been removed and a stump remnant from the third tree providing shelter, which
could be seen in the previous year, is no longer visible. DCC stated that it was
necessary to fell these trees following damage sustained during Storm Arwen,
which occurred in 2021.
64 A tree and its roots will grow to be as strong as is necessary to survive the loads
present in its microenvironment. Its strength will also incorporate an additional
safety factor to allow the tree to withstand exceptional events. This safety factor
is in the region of 3 to 3.5 times the normal loadings the tree has experienced up
to that point. If this microenvironment subsequently changes, a tree must adapt to
the change, or it will perish. An example of such a change is an increase in wind
loading because of a loss of sheltering companion trees, a situation known as
altered exposure. This will necessitate the subject tree to increase the strength
of its stem and root anchorage over the years following the altered exposure.
During the time a tree will take to adapt to this altered exposure, it can withstand
increases in wind loading for as long as this increase remains less than the
ingrown safety factor. If the additional wind loading forces caused by altered
exposure exceed the safety factor, then the tree will fail, even if it is otherwise
healthy and structurally sound.
Report 13/2024 25 December 2024
Broughty Ferry
Timing of tree failure
Analysis
65 The wind loading acting on the tree was elevated by the high winds
associated with Storm Gerrit.
66 On 27 December 2023, much of the United Kingdom was subjected to extreme
weather brought by Storm Gerrit. High winds and heavy rainfall accompanied this
storm before and after the tree fell. This weather had been preceded by two other
wet and windy named storms (Storm Elin on 9 December 2023 and Storm Fergus
on 10 December 2023).
67 Dundee City Airport (located 7 miles (11 km) to the west of Broughty Ferry)
reported average wind speeds of 47 km/h to 70 km/h and gusts of up to 95 km/h
throughout the morning leading up to the accident. In the hour between 12:00 and
13:00 during which the tree fell, the weather station reported gusts up to 87 km/h
from an east to southeasterly direction.
68 Although originally developed for use at sea, the Beaufort wind force scale was
adapted in 1906 for use by ‘land‐based observers’ by the then director of the
Met Office. The Beaufort scale assigns a number of ‘forces’ to approximate a
range of wind speeds. For example, force 8 represents a ‘gale’ with a wind speed
from 62 km/h to 74 km/h. Rather than using sea state observations, the Met Office
describes how the wind acted on trees and other land-based observations to
provide visual indication of wind force.
69 The description for a force 8 gale includes twigs breaking from trees, rising to a
loss of branches at force 9, a strong gale. For a force 10 storm, the description
includes trees being uprooted. These descriptions were based on observations
made at Royal Botanic Gardens, Kew, which has a free‐draining loamy soil. While
this description for force 10 includes whole tree failure, it is the view of the expert
arboriculturist engaged by RAIB that this is the onset of a wind force which might
see failure in some trees, rather than the complete failure of all trees subjected to
the gale.
70 The precise wind speed which the failed tree was subject to is unknown.
Observations taken from Dundee City Airport saw gusts up to 87 km/h for the
hour in which the tree failed, which is just below the wind speed range for force 10
of 88 km/h to 101 km/h. Although the weather station is located further along the
Firth of Tay shoreline from the tree, RAIB considers that the speeds recorded are
likely to be similar to the wind experienced by the failed tree (figure 17).
71 The tree had been grown in an area of sandy soil which provided poor adhesion
for the roots (paragraph 50). It is also possible that the soil had been softened by
an increase in the ground water content following the three storms. This might
have reduced the ability of the root anchorage system to continue to resist the
overturning forces acting on the tree. In addition to this, the primary structure
of the tree (the stem and roots) was subject to an increase in loading from
southeasterly winds resulting from the altered exposure. Although the fallen tree
had weathered several named storms since the loss of the companion trees, it
was still adapting to the change in its microenvironment (paragraph 58). This
predisposing factor would have made the tree more susceptible to windthrow,
providing an explanation for its isolated failure during Storm Gerrit, adjacent to a
group of similar trees which remained standing.
Report 13/2024 26 December 2024
Broughty Ferry
Analysis
Dundee City Airport Location of accident
Contains Ordnance Survey data: @Crown Copyright and database right 2024.
OS license number: AC0000833184. Source: Department for Transport, RAIB 2024
Figure 17: Map showing Dundee City Airport and accident location relative to Firth of Tay.
Notification of the fallen tree
72 Notification of the obstruction did not reach the driver following a call from
a member of the public telling the railway that the tree had fallen across the
lines.
73 During the storm, a member of the public heard a loud noise coming from the
railway which ran along the bottom of their garden. Concerned about what might
have happened, they looked out of their window and saw the tree across the
railway lines. At 12:57, around 12 minutes before the collision, the member of the
public called the Network Rail public helpline to report the fallen tree.
74 The Network Rail public helpline deals with all calls from members of the public
relating to matters ranging from general enquiries to incidents which might affect
the safety of the railway. The helpline is operated by Journeycall, a third-party
organisation on behalf of Network Rail, which employs call handlers who triage
incoming calls to determine what course of action to take.
75 The call handler established the location of the fallen tree and that it would be
necessary to advise the SICC of the obstruction. Around 9 minutes before the
collision, and while the member of the public was put on hold, the call handler
attempted to contact the SICC.
76 The call handler did not have access to a priority telephone number for the SICC,
so rang the same number which would be used for non-urgent enquiries. This
initial call to the SICC was not answered, and because the line had gone silent
while on hold, the member of the public had hung up. Following the first attempt to
contact the SICC, the call handler completed the incident report before attempting
to contact the SICC again. This second call was made at 13:07, around 2 minutes
before the collision, and was again unanswered.
Report 13/2024 27 December 2024
Broughty Ferry
77 Calls from the helpline to the SICC would normally be answered by incident
Analysis
support controllers who assist incident controllers in the management of ongoing
and out-of-the-ordinary operational incidents. Network Rail stated that calls to the
SICC from the helpline rarely relate to safety of the line incidents, and as such
they are given a lower priority when SICC staff are busy dealing with ongoing
incidents.
78 In the hour before the incoming helpline call, there had been five railway line
closures in Scotland due to flooding. This included the closure of the Dundee
to Aberdeen line which had caused train 1A37 to turn back (paragraph 27). It is
probable that the two calls made before the collision had been unanswered due
to the incident support controllers giving the helpline call a lower priority while
dealing with these flood closures.
79 RAIB considers that there was sufficient time available from the initial helpline
call for SICC staff to prevent the collision. This could have been achieved either
by sending a railway emergency call via GSM-R to stop all trains within the area,
using radio equipment at the SICC, or by advising the signaller at Dundee about
the obstruction so that the train could be stopped. It is possible, but not certain,
that there was also sufficient time for this to have also occurred after the second
helpline call.
80 After the two calls made before the collision, the call handler called the SICC a
further five times. One of these unanswered calls occurred around the time of the
collision and a further three unanswered calls were made afterwards. The seventh
call made by the call handler was answered at 13:18, around 9 minutes after the
collision had occurred.
Distance to stop
81 On sighting the fallen tree, the driver was unable to stop the train before
colliding with it.
82 Several BSRs had been put in place across Scotland in response to the adverse
weather brought by Storm Gerrit (paragraph 24). This included a BSR which had
been applied due to the risk of flooding on the coastal part of the route between
Carnoustie and Aberdeen, a distance of around 61 miles (98 km). No BSR was
in place on the route between Dundee and Aberdeen, on which the accident
occurred.
83 Route control centre staff respond to forecast extreme weather in accordance with
Network Rail standard NR/L2/OPS/021, ‘Weather – Managing the Operational
Risks’. The version in place at the time of this accident was issue 8 dated
June 2019. Notification of extreme weather comes from 5-day weather forecasts
which are issued daily by the Met Office to each of the Network Rail route control
centres, such as the SICC. They consist of a detailed forecast for that day
together with an outlook for the following 4 days. These forecasts are broken
down into railway routes which are given a colour code based on the severity of
the weather it is likely to experience. These codes range from ‘red’ for extreme
weather, through ‘yellow’ and ‘amber’, to ‘green’ for normal conditions.
Report 13/2024 28 December 2024
Broughty Ferry
84 Each morning, a designated manager will review the weather report and
Analysis
determine if it is necessary to convene a meeting with representatives of the
affected infrastructure to manage the effects of the incoming weather. In the case
of extreme weather, this meeting is referred to as an extreme weather action
teleconference (EWAT).
85 On the morning of 27 December 2023, an EWAT was held and chaired by the
route control manager in the SICC. Attendees included representatives from the
infrastructure delivery units and train operators. The updated weather forecast
included red warnings for predicted wind gusts of 65 mph (105 km/h) on the line
between Dundee and Arbroath, and 70 mph (112 km/h) for the coastal route
between Arbroath and Aberdeen. The forecast also included red warnings for rain
throughout these sections.
86 The EWAT attendees use their historical knowledge of the route, any hazards
present on the route, and actions previously taken when deciding what mitigating
responses are required. These responses can include closing the line, applying
BSRs or, where appropriate, continued monitoring of the weather or affected
infrastructure. When deciding on what action to take and on what routes, the
EWAT attendees balance the need to continue train operations and avoid
unnecessary delays against the need to operate services safely.
87 Network Rail expects route control centre managers to follow a framework for
the decisions made during the EWAT conference. This framework is documented
in Network Rail National Operating Procedure 3.17, ‘Weather Arrangements’
(issue 3 dated June 2020, in force at the time of the accident). This procedure
states that BSRs should be considered to reduce the likelihood or consequence
(or both) of a train striking obstructions blown onto the line.
88 When considering mitigations during forecast high winds, section 13 of this
procedure provides guidance in the form of a ‘weather trigger table’ (table 1).
The procedure states that structured expert judgement can take precedence
over the guidance provided in this table. This requires those considering the
implementation of BSRs to account for information such as local features
(including lineside trees), darkness, wind speed and other weather conditions.
Any decisions taken and the reasons for them are required to be recorded and
should be revised as required, as weather conditions change.
Report 13/2024 29 December 2024
Broughty Ferry
Wind speed Action Element
Analysis
Forecast of gusts up to 39 mph [63 km/h]
No action Normal
Forecast of mean wind speeds of up to
29 mph [47 km/h]
Forecast of gusts from 40 to 49 mph
[64 to 79 km/h]
Be aware of the possibility of
Aware
higher speeds being reached
Forecast of mean wind speeds of 30 to 39 mph
[48 to 63 km/h]
Forecast of gusts from 50 to 59 mph
[80 to 95 km/h]
Be aware of the possibility of
Adverse
higher speeds being reached
Forecast of mean wind speeds of 40 to 49 mph
[64 to 79 km/h]
Forecast of gusts 60 mph [96 km/h] or over
50 mph speed restriction for all
trains in the affected Weather Extreme
Forecast of mean wind speeds of 50 mph
Forecast Area
[80 km/h] or over
All services suspended in the
Forecast of gusts 90 mph [145 km/h] or over affected Weather Forecast Extreme
Area
Table 1: Weather trigger table taken from Network Rail procedure.
89 Given the forecast gusts of wind in excess of 65 mph (105 km/h) between Dundee
and Aberdeen the applicable action from the guidance in table 1 would have been
the imposition of a 50 mph (80 km/h) BSR. However, Storm Gerrit was forecast
to bring severe weather across Scotland requiring restrictions on most routes.
To limit the impact of these restrictions, the EWAT conference considered where
known hazardous trees (for example, those trees which have been identified as at
risk of falling across the railway) were located between Dundee and Aberdeen.
90 Network Rail had identified trees at risk of falling along the coastal section north
of Arbroath but found that there were no trees of concern to the south (figure 18).
With this information, the EWAT conference attendees concluded it was not
necessary to recommend a BSR for high winds on the section alongside the
Firth of Tay on which the accident occurred.
91 In the absence of any weather-related speed restriction, the driver was observing
the permitted speeds on the approach to Broughty Ferry. Around 1.25 miles
(2 km) from the point of collision, the train was travelling at 90 mph (145 km/h).
At that location, the railway is on an exposed section of track with the shoreline
immediately to the left and no shelter from the incoming wind and rain. The
railway line then passes under a road bridge where it curves to the right, with
Barnhill Rock Gardens to the left and residential properties to the right.
92 The curvature of the railway means it is just possible to see the location where
the tree fell, some 240 metres (around 6 seconds at the train’s speed) beyond the
road bridge, but the visibility would have been reduced by the heavy rain and poor
sunlight conditions on the day of the accident. The conspicuity of the fallen tree
would also have been affected by the motion of the train and a lack of contrast
between the tree and lineside vegetation (figure 19).
Report 13/2024 30 December 2024
Broughty Ferry
Analysis
Accident
route
Figure 18: Network Rail database of hazardous trees
north of Dundee.
Approximate location
of fallen tree
Direction
of travel
Up line Down line
Figure 19: Image taken from train 1A37 forward-facing CCTV as it passed under the road bridge
(courtesy of ScotRail with RAIB annotations).
Report 13/2024 31 December 2024
Broughty Ferry
93 The driver stated they made an emergency brake application immediately upon
Analysis
seeing the fallen tree across the track. Satellite location data from on-board train
equipment records the emergency brake being applied around 80 metres, or just
over 2 seconds, from the site of the collision. The train was travelling at 84 mph
(135 km/h) and at this speed it would not have been possible to stop the train in
that distance, which made the collision unavoidable.
94 Had a BSR had been applied to this part of the route, it would have reduced
the maximum permitted speed of the train from 90 mph (145 km/h) to 50 mph
(80 km/h). This would not have affected how far away from the train the tree
became visible (paragraph 92) but at lower speed this would have been a
longer time and reduced the distance covered during the driver’s reaction time.
However, the view of the tree would still have been limited to a maximum of
240 metres by the railway’s alignment and the road bridge. Even if an emergency
brake application had been made at 50 mph (80 km/h) at the road bridge, there
remained insufficient distance to stop the train and avoid the collision. Therefore,
the collision was unavoidable regardless of the decision not to recommend a BSR
made by the attendees of the EWAT conference (paragraph 90).
95 A lower speed would have reduced the energy of the collision. This is discussed
further in paragraph 121.
Identification of underlying factor
Management of risk
96 The risk of trees in Barnhill Rock Gardens falling onto the railway was not
effectively controlled.
97 Landowners in Scotland have a duty of care to prevent foreseeable harm to
members of the public or adjacent property including hazards arising from
falling trees (similar requirements apply in England and Wales, making this a
requirement which applies across Great Britain). The means of managing risk
arising from a falling tree is covered by several industry guidance publications.
A common requirement of these guidance documents is for the landowner to
inspect their trees for any indication of possible failure arising from the growing
conditions or the onset of disease or decay. The guidance also requires
assessment of what harm a falling tree might present based on its surroundings.
For example, a tree located within falling distance of a regularly used public space
has a higher potential for harm compared to a tree in a rural area with little or no
public access.
98 The effectiveness of this risk assessment process requires access to inspect the
tree and its surroundings, making and maintaining accurate records, tracking
changes in the microenvironment and an understanding of the tree species being
assessed.
99 The risk of a visually healthy tree falling onto the railway lines from outside the
railway boundary was not being effectively controlled in the case of the tree
involved in this accident because:
a. Network Rail is reliant on neighbouring landowners controlling the risk
associated with visually healthy trees falling onto the railway lines from outside
of the railway boundary (paragraph 100).
Report 13/2024 32 December 2024
Broughty Ferry
b. DCC did not effectively manage the risk of trees falling from its land onto the
Analysis
adjacent railway lines (paragraph 107).
Network Rail
100 Network Rail is reliant on neighbouring landowners controlling the risk
associated with visually healthy trees falling onto the railway lines from
outside of the railway boundary.
101 Network Rail inspects the vegetation within its boundary in accordance with
standard NR/L2/OTK/5201/Mod01, ‘Lineside vegetation inspection and risk
assessment’. At the time of the accident this standard was at issue 4 dated
December 2020. The standard required that inspections be carried out regularly
and that these should include the assessment of trees which could present a
hazard to the railway and its infrastructure or to property beyond the railway
boundary. The inspections were required to be undertaken by local maintenance
teams from the off track section as a minimum every 36 months, but no more
than 44 months. These inspections were supplemented with a second inspection
undertaken by an arboricultural specialist, at a similar interval.
102 The standard required the inspection to assess vegetation on neighbouring land
where it posed a risk to the railway. Such trees were described as those situated
within falling distance of the running line which could cause derailment or harm
and having a stem diameter more than 150 mm, when measured at chest height.
These conditions would have applied to the tree which fell onto the railway lines
from Barnhill Rock Gardens.
103 Inspections were to be undertaken on foot by Network Rail or specialist contract
staff trained to identify whether a tree is dead or suffering from either disease
or decay which might lead to failure. Where the tree was on neighbouring land,
the inspection was to be carried out from within the railway boundary. It would
not have been routine practice under the standard to have assessed the tree
further, even if it was in a publicly accessible location. The inspections reported
hazardous trees by exception, that is, a tree which was not identified as dead,
diseased or decayed would not be reported for further action.
104 Where a tree was found to be at risk of failure it would be scheduled for remedial
work. Where such a tree was identified outside of the railway boundary, then
Network Rail would notify the landowner. The contents of this notification would
include a reminder of the landowner’s obligations and the potential harm which
could arise to the railway if action were not taken.
105 There had been two inspections on the section of line which passes Barnhill
Rock Gardens, both undertaken in 2021. The first inspection was undertaken by
a member of the local maintenance team and the second was carried out by a
specialist arboricultural contractor. The tree was not dead, diseased or decayed
(paragraph 48) and would not be considered hazardous in line with the criteria set
out by Network Rail. Neither inspection, therefore, identified the tree as posing a
risk to the railway. Network Rail had a working relationship with the forestry office
at DCC and it is likely that, had the tree been identified as presenting a hazard,
Network Rail would have notified DCC, and suitable remedial action would have
been undertaken.
Report 13/2024 33 December 2024
Broughty Ferry
106 It might have been possible for the inspections to identify that the sheltering
Analysis
companion trees had been removed, exposing the tree to the risks associated
with altered exposure (paragraph 58). However, to undertake an inspection for
trees on neighbouring land would require a level of detailed assessment and
record keeping which Network Rail did not include in its inspection processes. It
instead relied on its neighbouring landowners to manage this.
Dundee City Council
107 Dundee City Council did not effectively manage the risk of trees falling from
its land onto the adjacent railway lines.
108 To meet its obligations, DCC stated that in part they rely on Network Rail to
advise them of trees presenting a hazard, but they also ‘strive towards industry
best practice principles’. To support this, DCC stated that it uses the following
reference documents:
• National Tree Safety Group publication ‘Common Sense Approach to Managing
the Risk of Falling Trees’ 1
• Health and Safety Executive publication SIM 01/2007/05 ‘Management of the
risk from falling trees or branches’ 2
• Arboricultural Association publication ‘Tree Surveys a Guide to Good Practice’.
109 In particular, ‘Common Sense Approach to Managing the Risk of Falling Trees’
describes a methodical process for assessment of the risk presented by falling
trees. The process begins with the identification of the species, age and condition
of the tree to be assessed, along with the location and potential for a tree to
cause harm. The method of recording this information is not specified in the
publication, but in January 2009, DCC published its document ‘Tree and Urban
Forestry Policy’.3 Within this policy document was a commitment to audit the trees
within the care of DCC using a geographic information system (GIS) computer-
based tree recording system.
110 DCC initially used a proprietary product to meet this 2009 policy commitment.
However, the ongoing costs associated with the use of this product were later
considered unacceptable, so DCC discontinued its use. This brought with it a
consequential loss of data. To replace the proprietary product, DCC developed a
new system internally but, following staff changes within DCC, it was not possible
to continue its use.
111 DCC stated that it follows a risk-based approach to tree inspections, prioritising
areas throughout the city where a falling tree is likely to cause greater harm.
DCC provided a copy of its ‘risk zoning matrix’ which provides guidance on the
frequency and standard of tree inspection to be adopted based on the tree’s
surroundings. Trees at risk of falling onto the railway were not included within the
risk zoning matrix and DCC stated that this was an oversight.
1
Available from https://ntsgroup.org.uk/publications/.
2
Available from https://www.hse.gov.uk/foi/internalops/sims/ag_food/010705.htm.
3
Available from https://www.dundeecity.gov.uk/sites/default/files/publications/urbanpolicy.pdf.
Report 13/2024 34 December 2024
Broughty Ferry
112 The risk zoning matrix used by DCC does not specifically categorise trees in
Analysis
public gardens. However, areas with public access are included in several
categories, based on the type of pedestrian use. DCC was not able to advise
RAIB in which zone Barnhill Rock Gardens had been categorised, or the
frequency at which tree inspections had been conducted. However, DCC stated
that the location was considered to be low risk.
113 In November 2021, Storm Arwen caused widespread damage across the
north‑east of the United Kingdom. DCC stated that following the work to remedy
the damage caused by this storm, the trees in Barnhill Rock Gardens were
visually inspected. This inspection did not raise any concerns related to the fallen
tree as it appeared to be in good health. DCC stated that no reports were raised
by DCC staff, Network Rail or volunteers from the gardens relating to concerns
about the tree between the inspection in 2021 and its failure during Storm Gerrit.
114 DCC stated that its forestry office resources had been primarily focused on
urgent high-priority storm damage recovery work since Storm Arwen and that
this recovery work was ongoing because of the continued red and amber severe
winter storms affecting the area. DCC also stated that staff resources had been
depleted over a number of years which had impacted the ability of the forestry
section to carry out routine tree inspections.
115 DCC stated that the visual inspection of the fallen tree would have taken into
account wind coming from the Firth of Tay, although the loss of the companion
trees and increased wind loading (paragraph 58) was not recorded as presenting
an additional risk (paragraph 113). With a successive loss of records, it is unlikely
that an effective assessment could be made of the potential risk arising from
windthrow, specifically that caused by altered exposure.
Factors affecting the severity of consequences
116 Serious injury from the loss of survival space was avoided by the driver’s
actions.
117 The HST was first introduced into mainline service in the mid-1970s and there
is no restriction on these trains operating in passenger service on the mainline
network in Great Britain. HSTs pre-date a number of modern crashworthiness
standards, including those relating to the design of the driving cab. The structure
of the driver’s cab is made of glass fibre reinforced plastic and bolted directly
to the underframe and bulkhead separating the driving cab from the power car
equipment and engine compartments. This is unlike most other modern train
driving cabs, which have a steel or aluminium cab superstructure.
118 The driver described making an emergency brake application immediately
upon seeing the fallen tree across the track. This occurred around 80 metres or
2 seconds from the collision while the train was travelling at 84 mph (135 km/h)
(paragraph 93). This emergency brake application remained until the train came
to a stop and there was nothing more the driver could do to avoid the collision. In
the seconds before impact, the driver therefore left the driving seat and sheltered
on the floor between the seat and the rear of the cab. Doing this meant serious
physical injury was avoided.
Report 13/2024 35 December 2024
Broughty Ferry
119 The tree stem impacted the cab at windscreen height, across the A pillars that
Analysis
support the windscreen at either side of the cab front and the weakest point
of the HST cab structure. The energy was greater than the cab structure was
intended to withstand and so it did not prevent the tree stem penetrating the cab
area previously occupied by the driver. The tree broke through the A pillars and
continued to break through the left‑hand side of the cab, showering the inside with
glass and debris.
120 The rootplate of the tree did not rotate in the ground with the forward motion of
the train, with the tree instead beginning to bend until it finally broke into two
pieces. This bending motion limited the damage to the right-hand side of the cab,
but the tree continued to cut through the left-hand side adjacent to the driver’s
seat. The structural damage to the left-hand side of the cab extended from the
windscreen back to the much stronger bulkhead at the rear, stopping just above
where the driver was sheltering (figures 20 and 21).
Windscreen A pillars
Right-hand side of cab Left-hand side of cab
Figure 20: External damage sustained to leading power car of 1A37.
Left-hand side of cab Right-hand side of cab
Figure 21: Internal damage sustained to leading power car of 1A37.
Report 13/2024 36 December 2024
Broughty Ferry
121 The collision occurred at 84 mph (135 km/h) because no speed restriction had
Analysis
been applied for the adverse weather. However, the collision with the fallen tree
would still have occurred even if a 50 mph (80 km/h) BSR had been in place
(paragraph 94). RAIB has not quantified exactly how a 50 mph (80 km/h) collision
speed would have changed the degree of damage sustained by the driver’s cab
in this collision, but analysis suggests that the collision energy would have been
reduced by around 65% at this lower speed.
Observation
122 The telephone concentrator equipment at Network Rail’s Scotland
integrated control centre was not able to show the history of multiple
missed calls from the public helpline call centre.
123 The helpline call handler made several attempts to contact the SICC. During this
time, the route control centre staff were dealing with other incidents and did not
pick up the call (paragraph 76). The route control centre staff handle calls using
desktop equipment known as a telephone concentrator. These combine several
incoming telephone lines to a single handset with each line allocated a button to
use to answer an incoming call. An incoming call is indicated on the concentrator
display for the duration of the call. If the incoming caller terminates the call, the
indication stops, but with no history of the missed call provided.
Figure 22: Telephone concentrator at the SICC (courtesy of Network Rail).
Report 13/2024 37 December 2024
Broughty Ferry
124 It is probable that the two calls made before the collision had been unanswered
Analysis
due to the incident support controllers giving the helpline call a lower priority while
dealing with flood closures (paragraph 78). However, the absence of a missed call
reminder on the concentrator display removes any prompt to the SICC staff that
someone might have called and that they may need to be called back.
125 Network Rail stated that calls from the helpline do not normally concern safety
of the line issues, and so were generally considered by SICC staff to be a lower
priority. For this reason, it is unknown when or if the SICC staff would have
returned the helpline call had they been aware that such a call had been missed,
or if this would have been done in time to warn the driver of train 1A37 and
prevent the collision.
Previous occurrences of a similar character
126 On 5 October 1999, an HST was involved in a fatal accident at Ladbroke
Grove near Paddington station which claimed the lives of 31 people. Rolling
stock leasing companies reviewed crashworthiness of HSTs in response
to Recommendation 53 of the Ladbroke Grove Inquiry and concluded that
modifications to HSTs to improve driver protection would not be reasonably
practicable (Health and Safety Commission Report November 20054).
127 On 10 July 2010, an HST passenger train collided with a tree at Lavington,
Wiltshire at 90 mph (145 km/h) (RAIB report 08/2012). The tree involved had
fallen across the two railway lines from land outside the railway boundary. In
common with the accident at Broughty Ferry, the impact occurred at windscreen
level, with the tree breaking through the left-hand A pillar of the driving cab. The
tree caused substantial damage to the left-hand side of the cab, but the damage
stopped at the leading edge of the driver’s door. On the basis of the November
2005 Health and Safety Commission report findings, RAIB concluded that the
costs associated with retrospective HST cab modifications were likely to exceed
the benefits gained if continued use for another 15 years was assumed, so no
recommendation was made in this area.
128 On 12 August 2020, an HST passenger train derailed after it had collided
with debris washed from a drain onto the track near Carmont, Aberdeenshire,
following heavy rainfall (RAIB report 02/2022). This accident resulted in three
fatalities, including the train driver. The driving cab of the HST was subjected to
severe impact conditions and became detached from the power car. The impact
conditions were significantly beyond those in which even modern cabs are
designed to provide protection for occupants. A relevant RAIB recommendation
resulting from this investigation is discussed in paragraph 135.
4
https://www.railwaysarchive.co.uk/documents/HSE_Public2005.pdf.
Report 13/2024 38 December 2024
Broughty Ferry
Summary of conclusions
Summary of conclusions
Immediate cause
129 Train 1A37 was unable to stop before colliding with a fallen tree that was
obstructing the railway (paragraph 39).
Causal factors
130 The causal factors were:
a. A tree on adjacent land and in close proximity to the railway boundary fell
across the lines (paragraph 44).
This causal factor arose due to a combination of the following:
i. The soil at this location had poor mechanical adhesion characteristics
which limited the ability of the tree to resist the wind forces acting on it
(paragraph 50).
ii. The tree had been subjected to increased wind loading due to altered
exposure (paragraph 58).
iii. The wind loading acting on the tree was elevated by the high winds
associated with Storm Gerrit (paragraph 65).
b. Notification of the obstruction did not reach the driver following a call from a
member of the public telling the railway that the tree had fallen across the lines
(paragraph 72, actions taken paragraphs 140 and 141).
c. On sighting the fallen tree, the driver was unable to stop the train before
colliding with it (paragraph 81).
Underlying factors
131 The risk of trees in Barnhill Rock Gardens falling onto the railway was not
effectively controlled because:
a. Network Rail is reliant on neighbouring landowners controlling the risk
associated with visually healthy trees falling onto the railway lines from outside
of the railway boundary (paragraph 100, Recommendation 1).
b. Dundee City Council did not effectively manage the risk of trees falling from its
land onto the adjacent railway lines (paragraph 107, Recommendation 2).
Factors affecting the severity of consequences
132 Serious injury from the loss of survival space was avoided by the driver’s actions
(paragraph 116, Recommendation 3).
Report 13/2024 39 December 2024
Broughty Ferry
Observation
Summary of conclusions
133 Although not linked to the accident on 27 December 2023, RAIB observes that the
telephone concentrator equipment at the SICC was not able to show the history
of multiple missed calls from the public helpline call centre (paragraph 122, no
Recommendation).
Report 13/2024 40 December 2024
Broughty Ferry
Previous RAIB recommendation relevant to this
Previous RAIB recommendation relevant to this investigation
investigation
134 The following recommendation, which was made by RAIB as a result of a
previous investigation, has relevance to this investigation.
Derailment of a passenger train at Carmont, 12 August 2020, RAIB report 02/2022,
recommendation 19
135 This recommendation reads as follows:
Recommendation 19
The intent of this recommendation is to evaluate the additional risk to train
occupants associated with the continued operation of HSTs, which entered
service before modern crashworthiness standards were introduced in July
1994. This will enable the future planning of HST deployment to be informed by
a fuller understanding of any additional risk and the costs and safety benefits
of any potential mitigation measures. This learning should also inform thinking
about the mitigation of similar risks associated with the operation of other types
of main line rolling stock.
Operators of HSTs, in consultation with train owners, ORR, DfT, devolved
nations’ transport agencies and RSSB should do the following:
a) Assess the additional risk to train occupants associated with the lack of
certain modern crashworthiness features compared to trains compliant
with Railway Group Standard GM/RT2100 issue 1 (July 1994), also taking
account of age-related factors affecting condition (such as corrosion). This
assessment should include a review of previous crashworthiness research
(including driver safety), a review of previous accidents, consideration
of future train accident risk, the findings presented in this report and any
relevant engineering assessments.
b) Based on the outcome of a) and cost benefit analysis, identify reasonably
practicable measures to control any identified areas of additional risk for
HSTs, and develop a risk-based methodology for determining whether, and if
so when, HSTs should be modified, redeployed or withdrawn from service.
c) In consultation with operators of other pre-1994 passenger rolling stock,
develop and issue formalised industry guidance for assessing and mitigating
the risk associated with the continued operation of HSTs and other types of
main line passenger rolling stock designed before the introduction of modern
crashworthiness standards in 1994.
136 On 6 April 2022, in response to this and other recommendations, the Office of
Rail and Road (ORR), the safety authority for railways in Great Britain, hosted
a meeting with owners and operators of HSTs, together with government
bodies and the Rail Safety and Standards Board (RSSB), to consider how
this recommendation should be addressed. The initial consideration of the
recommendation by relevant parties was completed by the ‘Carmont Seniors
Group’ co‑ordinated by Angel Trains. Actions arising included commissioning a
consultancy to undertake an HST design review and including the co‑ordination
function of the Carmont Seniors Group in a group known as the ‘RSSB Carmont
Recommendations Steering Group.’
Report 13/2024 41 December 2024
Broughty Ferry
137 On 15 February 2023, RSSB reported to ORR that it had commissioned SNC
Previous RAIB recommendation relevant to this investigation
Lavalin to undertake a literature search of previous relevant accidents to inform
engineering analysis of HST trailer vehicles.
138 On 9 March 2023, ORR reported to RAIB that while RSSB had taken the
recommendation into consideration and is taking action to implement it, it
considered the recommendation to still be open.
139 As a result of the accident at Broughty Ferry, RAIB has made a recommendation
to the RSSB Carmont Recommendations Steering Group to review the
circumstances of this accident as part of developing its response to Carmont
recommendation 19 (see recommendation 3, paragraph 145).
Report 13/2024 42 December 2024
Broughty Ferry
Actions reported as already taken or in progress relevant to
Actions reported as already taken or in progress relevant to this report
this report
Actions reported that address factors which otherwise would have
resulted in an RAIB recommendation
140 Network Rail has provided the helpline call handlers with a priority call telephone
number for their route control centres. If a helpline call hander becomes aware of
an issue affecting the safety of a railway line, they can use these numbers to warn
control centre staff.
141 Network Rail also reported that incoming calls to a route control centre using
the priority call telephone number will be answered as a high priority and by
a larger pool within the existing route control centre staff. This should prevent
future delays in answering calls relating to safety of the line matters. Other, more
general calls from the helpline will use the previous telephone number to avoid
diluting the status of the high priority calls.
142 Network Rail’s internal investigation into the accident recommended a review
of the existing telephone system within the SICC to determine its suitability
for modern-day railway control operations. This review should address the
observation regarding the adequacy of the current system (paragraph 122).
Other reported actions
143 DCC stated that work required to manage the effects of ash dieback has led to
the adoption of a new GIS system which DCC plans to apply to all trees for which
the council is responsible. DCC plans to share information from the GIS system
with Network Rail to improve collaboration.
144 Network Rail’s Technical Authority is developing an aerial survey system to assist
with the lineside inspection of trees. The system known as digitised lineside
inspection (DLI) uses data produced from equipment on board aircraft to survey
the railway corridor. The data is supplemented by light detection and ranging
(LiDAR) scans and hyperspectral imaging which is processed by software to
identify dead, diseased or decayed trees. Network Rail states that use of aerial
survey techniques allows the survey to better incorporate trees on neighbouring
land which are within falling distance of the railway.
145 On 3 September 2024, the Scottish Government announced that the procurement
process will begin to replace the HST sets operated by ScotRail.
Report 13/2024 43 December 2024
Broughty Ferry
Recommendations
Recommendations
146 The following recommendations are made:5
1 The intent of this recommendation is to use emerging technology to
identify trees which have been subject to altered exposure and are
potentially at risk of falling onto the railway.
Network Rail, as part of its development of aerial surveying technology,
should consider how current and emerging technology of this nature
could assist in the detection of trees subject to altered exposure,
including those trees on third-party land, which could present a risk to
the railway (paragraph 131a).
2 The intent of this recommendation is for Dundee City Council to improve
its management of the trees which are its responsibility, and which are at
risk of falling onto the railway.
Dundee City Council should review its management of the trees for
which it is responsible to ensure that it is effectively controlling the risk of
trees falling onto the railway. This review should consider:
i. compliance with legal requirements and available good practice
related to tree management
ii. how trees within falling distance of the railway are identified
iii. how factors that could increase the risk of healthy trees falling onto
the railway such as tree species, growing requirements (including soil
condition and effects of windthrow) are understood and accounted for
iv. how a risk-based approach to tree inspections is to be established
v. how accurate records of tree inspections are to be maintained.
Dundee City Council should develop a timebound programme to make
any appropriate changes identified to their policies, procedures and
systems (paragraph 131b).
5
Those identified in the recommendations have a general and ongoing obligation to comply with health and safety
legislation, and need to take these recommendations into account in ensuring the safety of their employees and
others.
Additionally, for the purposes of regulation 12(1) of the Railways (Accident Investigation and Reporting) Regulations
2005, Recommendations 1 and 3 are addressed to the Office of Rail and Road (ORR) and Recommendation 2 is
addressed to Dundee City Council to enable them to carry out their duties under regulation 12(2) to:
(a) ensure that recommendations are duly considered and where appropriate acted upon; and
(b) report back to RAIB details of any implementation measures, or the reasons why no implementation measures
are being taken.
Copies of both the regulations and the accompanying guidance notes (paragraphs 200 to 203) can be found on
RAIB’s website www.gov.uk/raib
Report 13/2024 44 December 2024
Broughty Ferry
Recommendations
3 The intent of this recommendation is to provide additional information to
the rail industry group which is currently evaluating the additional risk to
train occupants associated with the continued operation of high speed
trains, which entered service before modern crashworthiness standards
were introduced in July 1994.
The Rail Safety and Standards Board Carmont Recommendations
Steering Group should review its response to recommendation 19 made
within RAIB report 02/2022, following its investigation into the derailment
of a passenger train at Carmont, Aberdeenshire on 12 August 2020 to
ensure that the circumstances of this accident have been addressed
(paragraph 132).
Report 13/2024 45 December 2024
Broughty Ferry
Appendices
Appendices
Appendix A - Glossary of abbreviations and acronyms
Abbreviation / Full term
acronym
BSR Blanket speed restriction
DCC Dundee City Council
DLI Digitised lineside inspection
EWAT Extreme weather action teleconference
FFCCTV Forward-facing closed-circuit television
FLAC Forbes‐Laird Arboricultural Consultancy Ltd
GIS Geographic information system
GSM-R Global System for Mobile Communications – railway
HST High speed train
LiDAR Light detection and ranging
MAR Mechanically active rootplate
ORR Office of Rail and Road
OTDR On-train data recorder
RAIB Rail Accident Investigation Branch
RCM Route control manager
RSSB Rail Safety and Standards Board
SICC Scotland integrated control centre
Report 13/2024 46 December 2024
Broughty Ferry
Appendix B - Investigation details
Appendices
RAIB used the following sources of evidence in this investigation:
• information provided by witnesses
• information taken from the train’s OTDR, remote monitoring and FFCCTV equipment
• site photographs and measurements
• weather reports from Dundee City Airport
• a report prepared by Forbes‐Laird Arboricultural Consultancy Ltd (FLAC),
commissioned by RAIB. The report included FLAC’s conclusions on the examination
of the tree and the site where it grew, and its view on the reasons for why the tree
fell. This work is documented in Forensic Accident Investigation (Arboriculture)
Expert’s Report 44-1005_JFL. Since 2009, FLAC has provided specialist advice
on tree risk matters to Network Rail Infrastructure Ltd and acted as a consultant
on arboricultural matters to the world body for railways, Union Internationale des
Chemins de Fer, Paris.
• a review of documentation and information provided by Dundee City Council
• a review of documentation and information provided by Network Rail
• a review of the rail industry investigation report into the accident, prepared by
Network Rail
• a review of previous RAIB investigations that had relevance to this accident.
Report 13/2024 47 December 2024
Broughty Ferry
This page is intentionally left blank
Report 13/2024 48 December 2024
Broughty Ferry
This report is published by the Rail Accident Investigation Branch,
Department for Transport.
© Crown copyright 2024
Any enquiries about this publication should be sent to:
RAIB Email: enquiries@raib.gov.uk
The Wharf Telephone: 01332 253 300
Stores Road Website: www.raib.gov.uk
Derby UK
DE21 4BA