Rail Accident Report
Collision between passenger trains at Salisbury
Tunnel Junction, Wiltshire
31 October 2021
Report 12/2023
October 2023
This investigation was carried out in accordance with:
• the Railway Safety Directive 2004/49/EC
• the Railways and Transport Safety Act 2003
• the Railways (Accident Investigation and Reporting) Regulations 2005.
© Crown copyright 2023
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 253300
Stores Road Website: www.raib.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 12/2023 October 2023
Salisbury Tunnel Junction
This page is intentionally left blank
Report 12/2023 4 October 2023
Salisbury Tunnel Junction
Collision between passenger trains at Salisbury
Tunnel Junction, Wiltshire, 31 October 2021
Contents
Preface3
Summary7
Introduction9
Definitions9
The accident 10
Summary of the accident 10
Context13
The sequence of events20
Events preceding the accident20
Events during the accident 22
Events following the accident 27
Analysis29
Background information 29
Identification of the immediate cause 39
Identification of causal factors 40
Identification of underlying factors 67
Observations 79
Factors affecting the severity of consequences 81
The role of the safety authority 85
Previous occurrences of a similar character 87
Summary of conclusions 90
Immediate cause 90
Causal factors 90
Underlying factors 90
Additional observations 91
Previous recommendations that had the potential to address one or more
factors identified in this report 92
Actions reported as already taken or in progress relevant to this report 95
Recommendations 105
Report 12/2023 5 October 2023
Salisbury Tunnel Junction
Appendices110
Appendix A - Glossary of abbreviations and acronyms 110
Appendix B - Sources of evidence 112
Report 12/2023 6 October 2023
Salisbury Tunnel Junction
Summary
Summary
At around 18:43 hrs on 31 October 2021, train reporting number 1L53, the 17:20 hrs
South Western Railway passenger service from London Waterloo to Honiton, passed
a red signal and collided with the side of train 1F30, the 17:08 hrs Great Western
Railway passenger service from Portsmouth Harbour to Bristol Temple Meads. At the
point of collision, train 1L53 was travelling at approximately 52 mph (84 km/h) and
train 1F30 at 20 mph (32 km/h). The collision took place at Salisbury Tunnel Junction,
which is on the immediate approach to Fisherton Tunnel, near Salisbury in Wiltshire.
The impact of the collision caused the front two carriages of train 1L53 and the
rear two carriages of train 1F30 to derail. Both trains continued some distance into
Fisherton Tunnel before they came to a stop. Thirteen passengers and one member
of railway staff required treatment in hospital as a result of the accident, which
also caused significant damage to the trains and railway infrastructure involved. A
potentially far more serious collision between train 1L53 and an earlier train travelling
in the opposite direction was avoided by less than a minute.
The causes of the accident were that wheel/rail adhesion was very low in the area
where the driver of train 1L53 applied the train’s brakes, that the driver did not apply
the train’s brakes sufficiently early on approach to the signal protecting the junction to
avoid running on to it, given the prevailing low level of adhesion, and that the braking
systems of train 1L53 were unable to mitigate this very low adhesion.
The level of wheel/rail adhesion was very low due to leaf contamination on the
railhead, and had been made worse by a band of drizzle that occurred immediately
before the passage of train 1L53. This leaf contamination resulted from the weather
conditions on the day of the accident, coupled with an increased density of vegetation
in the area which had not been effectively managed by Network Rail’s Wessex route.
Network Rail’s Wessex route had also not effectively managed the contamination on
the railhead with either proactive or reactive measures.
RAIB’s investigation found that a probable underlying factor was that Network Rail’s
Wessex route did not effectively manage the risks of low adhesion associated with the
leaf fall season. RAIB also found that South Western Railway not effectively preparing
its drivers for assessing and reporting low adhesion conditions was a possible
underlying factor.
RAIB has also made two safety observations. These relate to the application
of revised design criteria for the Train Protection and Warning System and the
assessment of signal overrun risk and how this accounts for high risk of low adhesion
sites. Two issues were found relating to the severity of the consequences. These were
a loss of survival space in the driver’s cab of train 1L53, and the jamming of internal
sliding doors, which obstructed passenger evacuation routes.
Since the accident, Network Rail has reviewed its training and competence framework
for off track staff at network level, and is also reviewing its adhesion management
standards. Network Rail’s Wessex route is reviewing its arrangements for proactively
responding to reports of low adhesion, including how it undertakes railhead treatment.
South Western Railway has made changes relating to training and briefing of its
drivers to ensure information on autumn arrangements has been effectively briefed
and understood.
Report 12/2023 7 October 2023
Salisbury Tunnel Junction
Network Rail and South Western Railway have also jointly updated their annual
Summary
autumn leaf fall working arrangements to ensure that sites at high risk of low adhesion
are identified, reassessed, managed and monitored.
The Rail Safety and Standards Board has revised the rail industry standard that
provides guidance for the rail industry regarding the management of low adhesion.
Cross-industry working groups have also issued revised guidance regarding low
adhesion.
In December 2021, the safety authority for the mainline railway in Great Britain, the
Office of Rail and Road, issued an improvement notice to Network Rail’s Wessex route
requiring it to improve its vegetation management and its assessment and control of
low adhesion risks.
As a result of the investigation, and accounting for the work done by the industry since
the accident, RAIB has made ten recommendations. Seven of these recommendations
are made to Network Rail. These relate to: a review of the processes, standards and
guidance documents relating to the management of leaf fall low adhesion risk; the
training and competence of staff dealing with vegetation management and seasonal
delivery; responses to emerging and potential railhead low adhesion conditions;
management of railhead treatment regimes; assessment of the risk of overrun at
signals which have a site at high risk of low adhesion on their approach; and a review
of the retrospective application of design criteria for the Train Protection and Warning
System.
One recommendation is made to South Western Railway to review and improve its
arrangements for training and briefing drivers to ensure that they are able to effectively
identify areas of low adhesion and report them as appropriate.
One recommendation is made to the Rail Delivery Group in consultation with
train operators and the Rail Safety and Standards Board regarding the review of
technologies other than sanding systems and wheel slide protection to improve
braking in low adhesion conditions.
One recommendation is made to Porterbrook, Eversholt and Angel Trains regarding
the design of the internal sliding doors on class 158 and 159 carriages.
Report 12/2023 8 October 2023
Salisbury Tunnel Junction
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.
2 The terms ‘up’ and ‘down’ refer to the lines heading towards and away from
London Waterloo station, respectively. For ease of reference, some distances in
this report are also measured from the accident site. Mileages given in the report
are measured from a datum at London Waterloo.
3 The report contains abbreviations. These are explained in appendix A. Sources of
evidence used in the investigation are listed in appendix B.
Report 12/2023 9 October 2023
Salisbury Tunnel Junction
The accident
The accident
Summary of the accident
4 At around 18:43 hrs1 on Sunday 31 October 2021, train reporting number 1L53,
the 17:20 hrs South Western Railway passenger service from London Waterloo to
Honiton, collided with the side of train 1F30, the 17:08 hrs Great Western Railway
passenger service from Portsmouth Harbour to Bristol Temple Meads. Train 1L53
was travelling at approximately 52 mph (84 km/h)2 and train 1F30 at 20 mph
(32 km/h). The collision occurred at Salisbury Tunnel Junction, which is on the
approach to Fisherton Tunnel, near Salisbury in Wiltshire (figures 1 and 2).
Location of accident
© Crown Copyright. All rights reserved. Department for Transport 100039241. RAIB 2023
Figure 1: Extract from Ordnance Survey map showing the location of the accident at Salisbury Tunnel
Junction.
5 The movement of train 1F30 across the junction was being protected from trains
approaching on the Down Main line by signal SY31, which was displaying a red
(danger) aspect. The signal had previously been protecting another train, train
1F27, which passed over the junction in the opposite direction less than a minute
before the accident. Train 1L53 passed this signal by around 200 metres before
colliding with train 1F30 (figures 3 and 4).
6 The impact of the collision caused the front two carriages of train 1L53 and the
rear two carriages of train 1F30 to derail. Both trains continued some distance
into Fisherton Tunnel following the collision before they came to a stop.
1
The times shown in the report have been adjusted to synchronise with the signalling system time.
2
On a class 159, the speed signal recorded by the on-train data recorder (OTDR) is derived from the rotational
speed of the second wheelset. When this wheelset slides, the recorded speed will be less than the actual speed of
the carriage. Therefore, it is possible that at times, train1L53 was travelling at a higher speed than that recorded by
the OTDR during the accident.
Report 12/2023 10 October 2023
Salisbury Tunnel Junction
Basingstoke
The accident
Andover
and London
Grateley
Salisbury Tunnel
Junction
0 5 miles
Bristol
0 8 km
Route of 1L53
Exeter
Salisbury
station
Dean
Route of 1F30 Mottisfont &
Dunbridge
Eastleigh
Romsey
Southampton
Figure 2: Overview of the location of the accident and geographical relationship of the main features.
N To / from Basingstoke
Direction of travel
of 1L53
Salisbury Tunnel Signal
SY31
30)
Junction
Road (A
line
Fisherton M ain
w n
Tunnel Do
London
Up D
ean
line
Direction of travel To / from
To / from Dow Eastleigh
of 1F30 n Dea
Salisbury n l in
e
Direction of
travel of 1F27
Signal SY37
Not to scale
Figure 3: Diagram of the location showing the direction of train 1F27 that had just passed over Salisbury
Tunnel Junction and the approach of trains 1L53 and 1F30 to Fisherton Tunnel.
Report 12/2023 11 October 2023
Salisbury Tunnel Junction
The accident
Direction of
travel
Direction of
travel
Direction of
travel
Figure 4: The aftermath of the accident.
Report 12/2023 12 October 2023
Salisbury Tunnel Junction
7 The accident caused substantial damage to the track, the tunnel and the
The accident
surrounding railway infrastructure. The leading carriage of train 1L53 and the rear
two carriages of train 1F30 were damaged beyond economic repair. The rear
two carriages of train 1L53 were less seriously damaged and the leading two
carriages of train 1F30 were undamaged.
8 The driver of train 1L53 was seriously injured in the accident and spent three
weeks in hospital. Thirteen passengers were also taken to hospital, with one
suffering serious injuries. A further ten passengers were treated at the scene for
cuts and bruises.
9 In total, there were 197 passengers and five members of staff on the two trains.
Each train had a driver and a guard on board. There was also an on-duty driver
travelling as a passenger on train 1F30 when the accident occurred.
Context
Location
10 At Salisbury Tunnel Junction, the Up and Down Main lines, which run to and from
London Waterloo via Basingstoke, meet the Up and Down Dean lines, which run
to and from Southampton via Romsey. Beyond the junction is the 405 metres
(443 yards) long Fisherton Tunnel and, just over 1 mile (1.6 km) further west,
Salisbury station. The collision occurred around 200 metres after train 1L53 had
passed signal SY31, at 82 miles 35 chains from London Waterloo.
11 The maximum permitted speed for trains approaching the junction on the Down
Main line is 90 mph (145 km/h), reducing to 50 mph (80 km/h) approximately
700 metres before the junction. Drivers are informed of the 50 mph (80 km/h)
speed restriction by a warning board about 1,400 metres on approach to its start,
and the start of the speed restriction itself is indicated by a commencement board.
12 Due to a sharp curve, the maximum permitted speed on the Down Dean line
approaching the junction is 20 mph (32 km/h).
Organisations involved
13 The railway infrastructure at Salisbury is owned, operated and maintained by
Network Rail. Salisbury Tunnel Junction forms part of Network Rail’s Wessex
route which, along with the Kent and Sussex routes, forms part of the Southern
region. Each Network Rail region acts as a devolved management organisation
within the company.
14 Due to the size of the Wessex route, maintenance and inspection of the
infrastructure is split between outer and inner maintenance delivery units (MDUs).
The Salisbury line falls under the responsibility of the Wessex outer MDU.
15 Network Rail employs the signallers at Salisbury signal box, autumn control room
staff at the Wessex integrated control centre (WICC) in Basingstoke, the seasons
delivery specialist (SDS) and the lineside and drainage senior asset engineer
(SAE). It also employs the MDU engineering and inspection off track staff, who
manage the control and inspection of vegetation.
Report 12/2023 13 October 2023
Salisbury Tunnel Junction
16 Network Rail Supply Chain Operations, part of Network Rail Route Services,
The accident
organises the contractual arrangements and route schedules for the Wessex
route railhead treatment multi-purpose vehicle (MPV) fleet. Although Network
Rail’s Wessex route is the fleet’s owner and entity in charge of maintenance,
operation of the MPV fleet and the actual undertaking of the maintenance is
contracted out.
17 Train 1L53 was operated by First MTR South Western Trains Ltd, trading as
South Western Railway (SWR). SWR employed the train driver and the guard of
this train.
18 Train 1F30 was operated by First Greater Western Ltd, trading as Great Western
Railway (GWR). GWR employed the train driver and guard of this train.
19 Porterbrook Leasing owned the carriages which formed trains 1L53 and 1F30 and
leased them to the train operators concerned.
20 Network Rail and SWR together undertook an internal railway industry
investigation into the accident (‘the joint industry investigation’, see
paragraph 334).
21 All parties freely co-operated with RAIB’s investigation.
Staff involved
22 The driver of train 1L53 started his railway career in 1962 with British Railways,
working as a fireman on steam locomotives and later as a train driver’s assistant.
He started driving trains for British Rail in 1982 and, following privatisation of
the industry, was employed by South West Trains and latterly SWR as a driver
and driver instructor. The driver had been based at Salisbury depot for his entire
career and had moved to a part-time contract in 2019.
23 The driver had been deemed competent to drive trains by SWR and all his
competence assessments were in date in accordance with the company’s train
driver competency management process. These assessments included defensive
driving techniques relating to low adhesion, against which the driver demonstrated
his competence between August 2018 and July 2021, as part of a three-year
competence assessment cycle, and was due to be reassessed in 2024.
24 At the time of the accident, the driver was working two days a week, driving class
158 and 159 trains between London Waterloo and Exeter, Southampton and
Basingstoke, as well as on the Salisbury to Bristol route. For the week starting
25 October 2021, the driver worked on Tuesday 26 and Saturday 30 October. He
was then scheduled to work nine-hour shifts on Sunday 31 October, the evening
of the accident, and Monday 1 November 2021.
25 The SDS joined Network Rail in late 2019 on the graduate entry scheme. They
were initially seconded as an assistant supporting the then incumbent Wessex
route SDS, during which time they also gained experience as a track patrol
assistant. They later transferred to London Waterloo, returning to their SDS
support role in early 2020. After the incumbent SDS was transferred to another
role, they took up the position of Wessex Route SDS on a permanent basis.
Report 12/2023 14 October 2023
Salisbury Tunnel Junction
Railway systems and infrastructure involved
The accident
26 Signalling in the area is controlled from a signal box at Salisbury station. On
the Down Main line, Salisbury Tunnel Junction is protected by signal SY31, a
three- aspect colour light signal. Signal SY31 is located at 82 miles 25 chains,
around 200 metres from the point where the collision occurred. Although signal
SY31 had been passed at danger twice before, in May 2001 and May 2006,
neither of the signals passed at danger (SPADs) were documented as being
caused by low adhesion.
27 Before reaching signal SY31, trains approaching the junction on the Down Main
line will first pass signal SY29R (80 miles 30 chains) and then signal SY29
(81 miles 66 chains), as shown in figure 5 and further discussed in paragraph
186. These are approximately 3,320 metres and 980 metres from the point of
collision, respectively. If signal SY31 is displaying a red aspect, signal SY29R will
display a double yellow (preliminary caution) aspect and signal SY29 will display
a single yellow (caution) aspect.
28 The gradient beyond signal SY29R is 1 in 169 descending, before it levels out
again on approach to signal SY29. Beyond signal SY29 the gradient is 1 in 733
descending (from 81 miles 67 chains), increasing to 1 in 610 descending (from
82 miles 6 chains). This latter gradient encompasses the line on approach to and
beyond signal SY31 and towards the point of the collision and Fisherton Tunnel
(82 miles 37 chains).
SY29R
80 m 30 ch Broken Cross bridge Direction of
80 m 58 ch train 1L53
Tree debris
80 m 70 ch
To Salisbury
50 mph warning board
81 m 11 ch
50 mph commencement board
1i 82 m 0 ch
50
n1
69 SY29
From Andover 81 m 66 ch SY31
82 m 25 ch
50 Fisherton
Tunnel
Level 1 in 733 82 m 37 ch
1 in 610
81 m 81 m
82 m
50 ch 67 ch
06 ch
Miles from London Waterloo
Not to scale Point of collision
Figure 5: Signalling layout, gradient profiles and distances on approach to the point of collision.
29 For trains approaching on the Down Dean line, the protecting signal for the
junction is signal SY37, located 120 metres from the point of the collision.
30 To reduce the risk from signals being passed at danger, Automatic Warning
System (AWS) and Train Protection and Warning System (TPWS) equipment is
provided in the area controlled by Salisbury signal box.
Report 12/2023 15 October 2023
Salisbury Tunnel Junction
31 AWS provides an audible and visual warning to a driver on the approach to
The accident
signals as well as some other infrastructure features, including certain changes
in permitted speeds. AWS uses track mounted magnets which are detected by
receivers fitted to trains. When a train passes over a magnet, the system on the
train sounds a bell or chime, if approaching a signal displaying a green aspect, or a
horn if approaching a signal displaying any other aspect. The system also sounds
a horn if approaching a specified reduction in permissible speed. The driver must
acknowledge the horn warning by pressing a button on the driving desk within a
specified time, otherwise the system on the train will make an emergency brake
application. The acknowledgement of a horn warning is shown on a visual indicator
which changes to a yellow and black display known as a ‘sunflower’.
32 TPWS was developed in the mid‑1990s and implemented in accordance with the
Railway Safety Regulations 1999 to automatically apply the train's brakes and
reduce the risk arising from trains passing signals at danger. It is fitted at signals
on passenger lines which protect certain conflicting movements. The system is
also used to enforce reductions in permissible speed on approach to stop signals
and speed restrictions and to intervene if trains approach buffer stops at too high a
speed. TPWS is not designed to prevent incidents of speeding or signals passed
at danger (SPADs); rather, it is designed to reduce the consequences of such
incidents by slowing trains or, in the case of a SPAD, by stopping a train before it
reaches the conflict point at a converging junction. The system is also not designed
to fully protect all types of train at all speeds.
33 TPWS uses radio frequency transmitters (known as ‘loops’) fitted between the rails
and receiving equipment on the train. At signals capable of showing a red aspect,
a pair of loops is used in a configuration called a train stop system (TSS).
34 If the features of the layout mean that a TSS alone would not stop a train before
the conflict point, then a second pair of loops can be placed at a specified distance
on approach to the signal. These loops are known as an overspeed sensor system
(OSS) and may also be fitted where there are significant reductions in permitted
speed (that is, where permissible speed warning indicators are used). The two
OSS loops in a pair are placed a set distance apart according to the speed at
which an intervention is required at that location.
35 The OSS and TSS loops for signals are energised when the signal is displaying
a red aspect, whereas loops for speed reductions are permanently energised. If
a train passes over TSS loops when they are energised, the TPWS equipment
on the train will demand an emergency brake application. For OSS loops, the
brake demand depends on the train reaching the second loop in the pair before
a specified time period has elapsed, thus indicating that the train’s speed is
higher than the ‘set speed’ for the TPWS installation. The driver receives a visual
indication that the brake intervention has occurred and must acknowledge it as part
of the process of resetting the system before continuing. Signals SY31 and SY29
are fitted with TSS loops; both signals and the 50 mph (80 km/h) permanent speed
indicator are fitted with OSS loops.
Trains involved
36 On the day of the accident, train 1L53 was formed of a single three-carriage class
159 diesel multiple unit (DMU), number 159102. Although shorter formations are
not unusual, this train service would normally have been formed of nine carriages.
On the day of the accident the train length was reduced due to service disruption.
Report 12/2023 16 October 2023
Salisbury Tunnel Junction
37 Train 1F30 was formed of two class 158 DMUs, numbers 158762 (leading)
The accident
and 158763 (trailing). Each of these units comprised two carriages, forming a
four‑carriage train.
38 Class 158 and 159 trains are very similar and are constructed by welding
aluminium extrusions together to form the bodyshells. The carriages which
formed trains 1L53 and 1F30 were built between 1989 and 1992 by British Rail
Engineering Limited in Derby.
39 The braking system fitted to class 158 and class 159 units decelerates the train
by supplying air to brake cylinders mounted on the train’s bogies. These brake
cylinders apply friction pads to brake discs mounted on the wheelsets. A driver
can apply three levels of braking in normal service. Step 1 brake provides the
lowest level of braking, while step 3 (known as ‘full service braking’) provides the
maximum braking effort. A driver can also make an emergency brake application.
This applies the same level of retardation3 as step 3 but uses a different control
system to normal service braking so the train can still be braked in the event of a
failure of that system.
40 Both trains involved in the accident were fitted with a wheel slide protection
(WSP) system. Analogous to the anti-lock braking system in a car, the WSP
system monitors the rotational speed of the train’s wheelsets to determine if any
have stopped rotating and are therefore sliding on the rail. If the WSP system
detects that wheel slide is occurring, the system automatically reduces the brake
force being applied to the sliding wheelsets until the system determines that they
are no longer sliding.
41 The WSP system is designed to optimise the train’s braking in conditions of low
friction between the wheel and the rail, while also minimising the potential for the
wheels to be damaged by sliding. The system can also increase the available
friction (and hence the available brake force) by injecting sand from nozzles into
the wheel/rail interface of some wheelsets. The sanding system is controlled
electrically but uses the train’s air system to blow sand into the wheel / rail
interface. Sand is discharged automatically when the WSP system detects wheel
slide in any brake step.
42 On class 159 units, such as that which formed train 1L53, sand is injected under
both wheels of the third wheelset from the front of the train. A sander system is
also provided on the rear carriage of the three-carriage train, but this is inhibited
when the carriage is trailing. This means that only the nozzles on the front
carriage of a three-carriage Class 159 unit will discharge sand. Drivers of class
159 units do not have a facility to manually operate the sanding system when
the train is braking; a manual sand button is provided for use when the driver is
applying traction power.
3
More recent multiple unit passenger trains are fitted with enhanced emergency brakes, which apply a
higher level of retardation than full service braking. The relevant standard does not require class 158
and class 159 trains to have this capability.
Report 12/2023 17 October 2023
Salisbury Tunnel Junction
External circumstances
The accident
43 On the day of the accident, an unusual type of weather front passed through
the Salisbury area, resulting in a decrease of air pressure concentrated on a
particularly small region of around 30 km in diameter. This type of weather front,
known as a mesoscale low-pressure event, is a relatively rare occurrence in the
United Kingdom, with only ten similar events reported between 2009 and the date
of the accident.
44 The mesoscale event in the Salisbury area on the morning of 31 October 2021
contributed to the worst recorded weather (in terms of combined wind and rainfall)
data recorded by MetDesk and the Met Office so far that year (see paragraph 91
and figure 6).
45 The weather front brought localised strong winds and heavy rain to the area,
causing major disruption to the Wessex route. Weather stations located near to
Salisbury recorded maximum wind gusts of approximately 73 mph (117 km/h),
peaking between 08:00 and 10:00 hrs. In the 12 hours preceding the accident,
around 21 mm of rainfall was recorded, some 36% of the average monthly rainfall
for the area.
46 After the weather front had passed, the area remained windy for the rest of the
day with peak wind gusts of approximately 23 to 37 mph (37 to 60 km/h). Between
18:35 and 18:40 hrs, radar imagery showed a band of drizzle passing over the
Up and Down Main lines in the area of Salisbury Tunnel Junction, just before the
passage of train 1L53 (figure 6).
Report 12/2023 18 October 2023
Salisbury Tunnel Junction
The accident
Salisbury Tunnel Junction
Time 05:00 06:00 07:00 08:00 09:00 10:00 11:00 12:00 13:00 14:00 15:00 16:00 17:00 18:00 19:00
Wind gust (mph) 30 33 34 35 49 55 36 34 35 34 - 32 - - -
Andover
Grateley
SY29R
SY29
SY31
Salisbury
Figure 6: Met Office radar image (top) showing weather front for 31 October 2021 with wind gust
speeds (middle) and macro radar image (bottom) showing the weather front passing over the Salisbury
area and railway line (moving south-eastwards) between 18:35 and 18:40 hrs, in relation to the location
of signals SY29R to SY31 (© Crown copyright 2023, the Met Office).
Report 12/2023 19 October 2023
Salisbury Tunnel Junction
The sequence of events
The sequence of events
Events preceding the accident
47 At 08:44 hrs on 31 October 2021, a train not involved in the accident, train 1L13,
struck a tree4 that had fallen onto the Down Main line at 80 miles 70 chains,
around 250 metres west of Broken Cross bridge (80 miles 58 chains) (figure 5).
The collision damaged the cab of the train, but no injuries were reported. The
driver of 1L13 reported that they came to a stop in the vicinity of the AWS magnet
for signal SY29, which was approximately 1350 metres beyond the fallen tree.
On reporting this incident to the signaller, the driver stated that the train’s braking
was affected by "slipping" before colliding with the tree, but the conversation did
not identify railhead contamination as a factor. There was no mobile operations
manager (MOM)5 available from Network Rail to attend the incident, so the train’s
driver and guard cut back the branches that were foul of the railway line, leaving
the tree in the cess of the Down Main line (figure 7). The line was reopened at
10:44 hrs.
48 The incident near Broken Cross bridge was one of fifteen weather-related railway
incidents that occurred on Network Rail’s Wessex route that day. During the
morning other incidents involving fallen trees and flooding caused significant
disruption, including a tree blocking the Down Main line at the Salisbury (west)
portal of Fisherton Tunnel.
Figure 7: The tree struck by train 1L13 (post-collision) on the Down Main line looking east towards
London with Broken Cross bridge in the background.
4
Another fallen tree was identified a short distance beyond signal SY29R, but this was examined and discounted
as the tree involved in the collision with train 1L13.
5
MOMs act as Network Rail’s front-line response to any incidents affecting the safe and effective operation of the
railway.
Report 12/2023 20 October 2023
Salisbury Tunnel Junction
49 The driver of train 1L53 booked on for duty at Salisbury depot at 10:30 hrs. He
The sequence of events
was originally due to travel to Honiton to start his driving duties, but the effect
of the severe weather conditions on the timetable led to several trains being
cancelled. The driver therefore stayed in the mess room in Salisbury depot for the
first part of his shift.
50 At around 14:50 hrs, the operational resource manager for SWR requested
that the driver take train 1L52, the 15:27 hrs service from Salisbury to London
Waterloo. Train 1L52 was formed of two class 159 DMUs, each of three carriages
forming a six-carriage train.
51 The driver walked to the platform at Salisbury to meet the incoming train which
would form train 1L52. Although there is witness evidence that he met another
driver on the way who warned him of the adverse weather conditions, the driver
stated that such a meeting did not occur. Later, while the driver was in the train’s
cab preparing train 1L52 for service to London, another driver contacted him from
the train’s rear cab using the cab-to-cab telephone. This was to warn him about
poor rail adhesion on the Down Main line approaching Salisbury, which had been
experienced by the driver of the incoming train from London Waterloo. The driver
acknowledged the warning and thanked his colleague.
52 Train 1L52 departed from Salisbury at 15:27 hrs. Approaching Broken Cross
bridge on the Up Main line towards London Waterloo, the driver, having been
informed by his colleagues that there was a fallen tree in the vicinity, was looking
for and noticed the remains of the fallen tree struck by train 1L13 on the adjacent
Down Main line.
53 During the journey to London, the bad weather subsided, and the sun set
at 16:42 hrs. The driver reported no conditions of low adhesion or any other
difficulties with the train’s braking or acceleration during this journey.
54 On arrival at London Waterloo, the driver changed trains, leaving train 1L52
and taking train 1L53 which departed from London Waterloo one minute late at
17:21 hrs. Soon after departure, the driver performed a routine running brake test6
in the Vauxhall area, from a speed of 53 mph (85 km/h). The driver reported no
issues with the braking capability of the train during the test. No further running
brake tests were carried out during the journey and none were required by the
SWR professional driving policy (see paragraph 269), unless the driver judged
there was a problem with the train’s braking.
55 The train called at Clapham Junction and Woking, where it departed six minutes
late due to being held at a signal. The train then stopped at Basingstoke and
Andover, departing Andover six minutes late at 18:30 hrs.
56 It was dark during the journey from London Waterloo and weather conditions
were dry. The driver reported that he did not experience any WSP activity nor
any instances of low adhesion that affected the train’s ability to slow, stop or
accelerate at station stops during the journey. Between 18:35 and 18:40 hrs a
band of drizzle moved south‑eastwards over the Up and Down Main lines in the
Middle Wallop area, near Salisbury (paragraph 46). The drizzle ceased in this
area before train 1L53 approached it.
6
Running brake tests are required to test the effectiveness of the train’s brakes and are undertaken in accordance
with the Rule Book GERT8000 Module TW1, ‘Preparation and Movement of Trains’ (issue 16, March 2021).
Report 12/2023 21 October 2023
Salisbury Tunnel Junction
57 At 18:41:09 hrs, approximately 1 minute 47 seconds before the accident, the
The sequence of events
OTDR from train 1L53 recorded that the traction power had been shut off. The
train then coasted7 down the prevailing 1 in 169 gradient towards signal SY29R
(figures 34 and 35).
58 Ten seconds later, OTDR data showed that the driver acknowledged the AWS
warning horn for signal SY29R. Signalling data and forward-facing closed‑circuit
television (FFCCTV) showed that this signal was displaying a double yellow
aspect. This was because signal SY29 was displaying a single yellow aspect, and
signal SY31 was displaying a red aspect.
59 The driver approached Broken Cross bridge at about 18:41:36 hrs with the train
still coasting. The driver stated that Broken Cross bridge was the point where he
would normally start braking in anticipation of stopping at signal SY31. However,
because he felt that there was a potential for low adhesion conditions at the
location of the fallen tree, he decided to delay braking and use the fallen tree
and associated debris as the marker point to start braking (figure 8). The driver
believed that this would still leave sufficient time and distance to bring the train to
a stand before reaching signal SY31.
Figure 8: (left) Tree that was struck by train 1L13 (branches had been cut back by the train driver and
guard) and later observed by the driver of train 1L53 (when driving train 1L52 on the Up line to London
Waterloo) and (right) enhanced image taken from train 1L53 FFCCTV system.
Events during the accident
60 Around 18:41:42 hrs, train 1L53 coasted past the location of the fallen tree
and debris (figure 7) at a speed of 89 mph (143 km/h). Eight seconds later, at
18:41:50 hrs, the driver acknowledged the AWS warning horn associated with the
warning sign for the approaching 50 mph (80 km/h) speed restriction.
61 At 18:42:02 hrs, with the train travelling at 86 mph (138 km/h) and around
1,560 metres beyond signal SY29R, the driver, having not noticed the fallen
tree, applied the brakes. At this point, the train was on a descending gradient
of 1 in 169, approximately 780 metres on approach to signal SY29, and
approximately 1,560 metres on approach to signal SY31.
7
Coasting is a state of train movement in which the driver is not braking or applying traction power.
Report 12/2023 22 October 2023
Salisbury Tunnel Junction
62 Evidence from the train’s OTDR data shows that the driver initially selected brake
The sequence of events
step 1, followed almost immediately by brake step 2. When the driver applied
the brakes, he stated that he immediately felt and observed that something was
wrong. The driver stated that the feeling ‘under his seat’ indicated that the train’s
wheels were sliding.
63 The driver realised that the train’s speed was not reducing as he expected. Six
seconds after his initial brake application, at 18:42:08 hrs, the driver made a
step 3 (full service) brake application. This was followed a further six seconds
later by the driver making an emergency brake application.
64 Equipment fitted to the train recorded that the WSP system was active from
the start of braking and throughout the various brake applications, and this
would have demanded the sanding system to discharge sand onto the railhead.
However, this had a limited effect on reducing the speed of the train (figures 34 to
36).
65 At 18:42:24 hrs, train 1L53 passed signal SY29 (displaying a single yellow
aspect) at approximately 77 mph (124 km/h).
66 At 18:42:42 hrs, the train’s TPWS detected that the train was travelling above
the set speed of 34.5 mph (55.5 km/h) for the OSS of the approaching signal
SY31, which was showing a red aspect. The TPWS therefore commanded an
emergency brake application. However, this had no additional effect as the driver
had already made an emergency brake application (paragraphs 34 and 35).
67 At 18:42:49 hrs, train 1L53 passed signal SY31 and approached Salisbury Tunnel
Junction. About 40 seconds earlier, GWR train 1F27 had passed over the junction
on the Up Dean line on its journey from Salisbury to Southampton. Meanwhile,
train 1F30 was approaching the junction on the Down Dean line heading towards
Salisbury.
68 The driver of 1L53 saw train 1F30 appear from the left and move into the path of
his train. Believing there was little he could now do to prevent a collision, he got
out of his seat with the intention of exiting the cab into the saloon. However, in
doing so, he tripped over his bag, and fell onto the floor just before the collision
occurred (figure 9).
69 As train 1F30 was entering Fisherton Tunnel, train 1L53 collided with its
right‑hand side, near the front of the fourth carriage (figure 10). Evidence from
the OTDR on train 1L53 suggests that the collision occurred at 18:42:57 hrs, with
train 1L53 travelling at a speed of approximately 52 mph (84 km/h). The OTDR
on train 1F30 recorded its speed as being around 20 mph (32 km/h) at the time of
the collision.
70 The 32 mph (51 km/h) speed differential between the two trains meant that train
1L53 travelled forwards in heavy contact along the right-hand side of the fourth
carriage of train 1F30. It then continued along the right-hand side of the third
carriage, with the leading left-hand edge of the cab of train 1L53 embedding itself
in the front end of this carriage (figure 11).
Report 12/2023 23 October 2023
Salisbury Tunnel Junction
The sequence of events
18:42:53
B3
B2
B1
Salisbury Tunnel
Junction
n
Fisherto 1L53
Tunn e l
A2 A3 A4
A1
1F30
Figure 9. Diagram showing train 1L53 (carriages marked as B1 to B3) and train 1F30 (carriages marked
as A1 to A4) immediately before the collision.
18:42:57
Salisbury Tunnel
Junction B3
B2
B1
nnel
ton Tu
Fisher A3 A4
A2
A1
Figure 10: Diagram showing train 1L53 colliding with train 1F30.
18:43:01
Salisbury Tunnel
Junction
B2 B3
Tunn el B1
Fis herton
A4
A3
A2
A1
Figure 11: Diagram showing trains 1L53 and 1F30 derailing into Fisherton Tunnel.
Report 12/2023 24 October 2023
Salisbury Tunnel Junction
71 The fourth carriage of train 1F30 was displaced to the left by the impact, resulting
The sequence of events
in its leading left corner coming into heavy contact with the facing portal wall of
Fisherton Tunnel (figure 12). This caused a failure of the coupling between the
second and third carriages of train 1F30, opening a gap of 23.8 metres between
the two halves of the train. At this time, passengers intending to leave train 1F30
at Salisbury station were still seated, and no passengers were standing in the
vestibule area where the train parted.
Figure 12: Impact mark from carriage A4 on facing portal
wall of Fisherton Tunnel. Vestibule of A3 is visible and
carriage A4 has been removed.
72 Both trains travelled into Fisherton Tunnel, with the tunnel walls acting to keep
the carriages upright. The rear two carriages of train 1F30 and all carriages of
train 1L53 derailed, with the cab of train 1L53 being severely damaged. The front
carriage of train 1L53 came to rest adjacent to the leading end cab of the third
carriage of train 1F30 (figures 13 and 14).
Report 12/2023 25 October 2023
Salisbury Tunnel Junction
The sequence of events
18:43:05
Salisbury Tunnel
Junction
B3
nnel B2
ton Tu B1
Fisher
A4
A3
A2
A1
Figure 13: Diagram showing carriages A2 and A3 separated within Fisherton Tunnel.
A3
B1
A4
B2
Figure 14: View from London portal of Fisherton Tunnel showing the rear carriages of train 1F30 (A3
and A4) and front carriages of train 1L53 (B1 and B2).
Report 12/2023 26 October 2023
Salisbury Tunnel Junction
Events following the accident
The sequence of events
73 The driver of train 1L53 was injured in the collision and was trapped in the heavily
damaged cab (figure 15). At around 18:46 hrs, the guard on train 1L53 attempted
to contact the driver but was unsuccessful in doing so. The guard then tried to
make a railway emergency call to the signaller using the GSM-R8 train radio in
the rear cab, but this again was unsuccessful. They then used their mobile phone
to contact the emergency services and provided information from 18:45 until
19:15 hrs when they were met by responding fire and rescue personnel.
A3
B1
Figure 15: View from within Fisherton Tunnel showing
the front cab of train 1L53 (B1) and carriage A3, having
separated from carriage A2, of train 1F30. Parts of the
driving cab of train 1L53 have been cut away by the
emergency services to release the trapped driver.
74 The driver of train 1F30 felt the impact of the collision and immediately made a
GSM-R railway emergency call, reporting to the signaller at Salisbury that they
believed their train had derailed. In response, the signaller placed all signals in
the area to red.
75 On completion of the emergency call, the driver of train 1F30 left their cab to
investigate what had happened. They walked back through the train and found
the rear two carriages had become detached. Realising then that another train
may have collided with train 1F30, they returned to the cab and made another
railway emergency call to the signaller to update them with this information.
8
Global System for Mobile Communications – Railway, a dedicated telecommunications network for the national
rail network. The systems on both trains were found to be working correctly.
Report 12/2023 27 October 2023
Salisbury Tunnel Junction
76 From 18:46 hrs onwards, a number of calls were made to the WICC at
The sequence of events
Basingstoke. Due to the volume of calls and the chaotic nature of the scene in the
tunnel, around 17 minutes elapsed following the collision before staff at the WICC
made contact with the guard of train 1L53 and established that it had also been
involved in the accident.
77 Both GWR and SWR mobilised ‘on call’ teams to the accident site. As members of
train crew at Salisbury station and depot became aware of the accident, they also
went on foot to the scene to offer assistance to passengers. On arrival, they found
passengers were already trying to evacuate from train 1F30 via a window that
had been broken (see paragraph 291).
78 Network Rail and SWR staff from Salisbury station arrived at the scene and, at
19:06 hrs, Network Rail declared a major incident.9 Wiltshire Police and Dorset
and Wiltshire Fire and Rescue Service arrived at the scene at 19:10 hrs and
confirmed the declaration of a major incident with the other agencies attending
(such as the South Western Ambulance Service and British Transport Police). By
19:56 hrs the fire and rescue service had reported that, with the exception of the
train driver of train 1L53, all passengers and staff had been evacuated from the
trains and had exited from the tunnel. At 20:45 hrs, the fire and rescue service
reported that the evacuation was complete.
79 The damage to the rail carriages prevented the movement of passengers
between some carriages. Consequently, it was necessary to evacuate
passengers simultaneously from both the front and rear of the trains involved.
Some passengers, predominantly those in the leading three carriages of train
1F30, were evacuated to Salisbury station. Passengers in the remaining carriages
were evacuated from the rear of the trains and walked trackside to a railway
access gate.
80 RAIB was notified of the accident at 18:53 hrs and immediately deployed a team
of inspectors, who arrived on site at 21:20 hrs. RAIB handed the accident site
back to the railway in stages, with the final areas being released on 7 November
2021. Repairs and reinstatement of the track and signalling were completed on
15 November 2021, with services through Salisbury Tunnel Junction resuming on
16 November 2021.
9
An event or situation with a range of serious consequences which requires special arrangements to be
implemented by one or more emergency responder organisation.
Report 12/2023 28 October 2023
Salisbury Tunnel Junction
Analysis
Analysis
Background information
Wheel/rail adhesion
81 Because trains rely on friction between wheel and rail to stop, the level of
wheel / rail adhesion available is critical to the rate at which a train can decelerate.
Research10 indicates that the level of adhesion available is mostly affected by
railhead contamination and moisture on the railhead.
82 Railhead contamination can be caused by a number of things, but the most
common cause is leaves falling onto the railhead and then being compressed
by the wheels of trains rolling over the material, building different layers and an
increased thickness of contamination. The layers of leaf fall are created and
bonded as a result of chemical reactions occurring during the rolling and sliding
of wheels over the contamination. The level of adhesion between a train’s wheel
and the railhead is normally expressed as a coefficient of friction. The lower the
value of the coefficient of friction, the lower the adhesion between wheel and rail
(table 1). Adhesion levels can vary considerably over relatively short distances
and timescales.
Adhesion Typical coefficient Description
level of friction
High > 0.15 Clean rail wet or dry
Medium 0.10 to 0.15 Damp rails with some contamination
Low 0.05 to 0.10 Typical autumn mornings due to dew / dampness
often combined with light overnight rust
Very low <0.05 Severe rail contamination often due to leaves but
sometimes other pollution
Table 1: Range of railhead adhesion on the rail network as defined by the Adhesion Working Group.
83 Research11 undertaken on behalf of the industry’s Adhesion Working Group
(AWG) led to it publishing guidance for the industry in 2006, most recently
updated in 2018.12 This provides the following comment on low adhesion
conditions:
‘The rail surface and the wheel treads can become coated with a range of
contaminants. The worst of these are crushed leaves, which, when combined
with moisture particularly in the form of dew or condensation, reduces the
adhesion level. For a train on dry rails adhesion is typically around 0.25, on wet
rails it is around 0.15, but on damp leaf (layers) it can be as low as 0.015. Rails
with damp leaves (layers) significantly constrain the rate of braking.’
10
Rail Safety and Standards Board (RSSB) project T354 ‘The characteristics of railhead leaf contamination’
https://www.rssb.co.uk/research-catalogue/CatalogueItem/T354’.
11
Tests undertaken by ARUP on behalf of Network Rail in 2006 ‘Autumn 2006 measurement Trials’.
12
AWG Manual, Sixth Edition, January 2006 amended 2018. The Adhesion Working Group has since evolved into
the Seasonal Challenge Steering Group (SCSG).
Report 12/2023 29 October 2023
Salisbury Tunnel Junction
84 Guidance on the thickness and levels of contamination (figure 24) and action
Analysis
to be taken are included within Network Rail standard NR/L2/OPS/045-4.07,
‘National Operating Procedure, Section 07 Railhead contamination levels’
(issue 2, December 2019).
85 Leaf layer thicknesses will typically measure up to 0.2 mm but can be thicker
when the base layers become bonded to the rail. Train wheels under acceleration
and braking can result in the contamination ‘creeping’ (a series of very small
horizontal movements), which can further increase the thickness of any railhead
contamination present at certain points. Sliding train wheels, in contrast, may
have the opposite effect and result in layers being removed.
86 Post‑accident examination of the Up and Down Main lines between signal SY29R
and the accident site showed that, although the topography and tree species
were very similar in number and density either side of the railway, the levels of
railhead contamination varied, with contamination on the Down line being medium
to heavy at locations where it was only minimal on the adjacent Up line. This
demonstrates how localised changes in wheel/rail adhesion conditions can be.
87 The speed at which layers of contamination build up is dependent on a number
of factors, including the number of trains that have passed over the location
since the last treatment, train braking and the leaf fall within a given period. The
rate of leaf fall is in turn affected by the environment, meteorological conditions,
topography (such as cuttings and embankments, figure 16) and tree species.
However, the relationship between the rate of build-up of contamination and these
factors remains largely undefined.
88 For the purposes of managing wheel/rail adhesion, Network Rail considers
the autumn leaf fall season to run from 1 October to 13 December each year.
While the rate of leaf fall can vary depending on weather conditions throughout
the year, the rate of fall normally begins to accelerate in the middle of October
with the ‘peak leaf fall’ period occurring from around 22 October through to
around 31 October. A further acceleration of leaf fall can also occur from around
31 October through to around 24 November, when around 40 to 60% of all
remaining leaves normally fall from trees, depending on the species involved.
89 The cross-industry Adhesion Research Group (ARG) promotes understanding of
wheel / rail adhesion, train detection and related topics, and ensures a common
and system-wide approach to these issues. It acts as a sponsor for research
projects and co-operates with the Seasonal Challenge Steering Group and
Seasonal Challenge Communications Group to implement research on the
running railway.
90 RSSB in collaboration with the Adhesion Research Group developed an adhesion
research programme called ADHERE. The programme has been running
since 2018 and continues to provide research to improve the rail industry’s
knowledge, and management, of low adhesion conditions. The programme
includes workstreams investigating: the modelling of low adhesion and braking;
the effectiveness of rail cleaning activities and treatments; driver behaviours;
current train-borne technologies, including double variable rate sanding (DVRS,
see paragraph 220) and magnetic track brakes; and adhesion forecasting and
observational capabilities to improve real-time decision-making.
Report 12/2023 30 October 2023
Salisbury Tunnel Junction
SY29R
Analysis
Level
Broken Cross bridge
Location of tree debris Embankment
Cutting
Brake step 1
Brake step 2
Full service brake
SY29
Emergency brake
Embankment
Cutting
SY31
Figure 16: Topography data showing the embankment/cutting profiles, including the previously
classified high risk of low adhesion site (pink shaded area - see paragraph 143), the location of Broken
Cross bridge, tree/associated debris (green dots), the braking points of train 1L53 (red dots) and
signals.
Weather and forecasting
91 Network Rail has a contract with a national weather forecast provider, MetDesk
Ltd (MetDesk). MetDesk sends weekly and daily weather and adhesion forecasts
to each integrated control centre, including the WICC (figures 18). Additional
information is also made available via the Network Rail Weather Service website.
92 Network Rail’s Wessex route also receives a daily weather and low adhesion
forecast (figures 17 and 23) from the Meteorological Office (Met Office). These
provide a forecast of railhead adhesion and damp rail risk for sections of line
between individual stations within the Wessex route. They are intended to
assist Network Rail operations staff in planning any necessary low adhesion
risk mitigation, for instance, additional MPV railhead treatment runs or MOM
inspections. The adhesion forecast considers site-specific forecasts of leaf fall
(from the Met Office leaf fall model), rainfall, rail temperature and dew point
temperatures to forecast low adhesion. The low adhesion forecasts are also
shared with SWR during the leaf fall season.
Report 12/2023 31 October 2023
Salisbury Tunnel Junction
Analysis
Salisbury
Figure 17: Image of the mesoscale weather front for 31/10/2021 (© Crown copyright 2023 The Met
Office).
Figure 18: (top) Extract from the MetDesk forecast showing an amber, adverse (circled) warning for
wind and heavy rain on Sunday 31 October 2021; and (bottom) five-day adhesion forecast for Wessex
route (outer) showing adhesion risk score seven (circled – see also figure 19 Adhesion forecast showing
the categories, with a score of 7 being a ‘red day’). The leaf fall density percentage is also shown.13
93 Weather conditions are forecast using a scale from ‘normal’ through ’adverse’
to ’extreme’. This scale is applied to flooding, rain, wind and snow. Network
Rail’s Wessex route operations control uses this forecast to liaise with the MDU
and asset managers to consider appropriate mitigations and the preparation
and provision of resources to undertake any mitigation activities identified as
necessary. If thresholds for certain weather-related trigger points are passed,
Wessex route’s autumn working arrangements document (see paragraph 97)
requires an extreme weather action team/teleconference (EWAT) to be convened.
13
Leaf fall density indicates the number of leaves (area and species) that are still on the trees. The Met Desk data
shows 65.2% (green box - circled) of leaves were still present and the leaf fall ‘drop’ was around two weeks behind
schedule for data gathered since 2016.
Report 12/2023 32 October 2023
Salisbury Tunnel Junction
94 The purpose of the EWAT is to assess the potential impact of extreme weather on
Analysis
the infrastructure and to determine mitigation, contingency plans and how actions
and decisions will be communicated within Network Rail and the associated
train operating companies. The forecast predicted ‘adverse’ weather conditions
for 31 October 202114 (figure 18). Network Rail standard NR/L2/OPS/021,
'Weather - managing operational risks', appendix (A), shows the hazards and
consequences of high wind speeds to be fallen trees and leaf fall leading to
railhead contamination that could result in low adhesion, loss of track circuit
detection from wrong side failures and a category A SPAD. However, the weather
forecast and the actual conditions on the day did not reach the thresholds for an
EWAT to be convened, resulting in the potential hazards not being considered
(see paragraph 252).
95 MetDesk’s data sheets are usually issued between 02:00 and 03:00 hrs for the
day ahead. Railhead adhesion forecasts are based on rainfall, leaf fall, dew point
and rail temperatures. The adhesion forecast uses an adhesion risk score which
runs between 0 (good) and 10 (very poor). These scores are then categorised into
five colour-coded levels, from ‘good’ adhesion (green) progressing through ‘poor’
(red) where there is a high risk of leaf fall and disruption to the network if the
railhead is not treated. ‘Very poor’ adhesion’ (black) is the most severe category,
with very high contamination likely due to leaf fall and damp conditions with a very
high risk of disruption to the rail network (figure 19).
96 Network Rail shares information on the adhesion conditions with SWR. SWR in
turn posts notices at depots for its drivers indicating the expected weather and
adhesion levels, using the same colour-coded index (see paragraph 182).
Operations
97 Network Rail operations standard NR/L3/OPS/021/01, ‘Autumn Management’
(issue 1, June 2019) outlines how Network Rail, in collaboration with train
operators, prepares, manages and responds to the risks arising from autumn
weather. The responsibility for developing and implementing the ‘route specific’
process lies with the SDS for each Network Rail route. Wessex route introduced
the post in 2014 (although roles with similar functions existed from around
2004).15 From March each year, the SDS develops the Wessex route autumn
working arrangements (designated as NR/L3/OPS/021/01/Wessex). This is
based on NR/L3/OPS/021/01 and is prepared against the timeframe specified
in Network Rail standard NR/L3/OPS/021/11 Module 11, ‘Seasonal calendars’
(issue 1, September 2020) (figure 20).
14
Network Rail standard NR/L2/OPS/021 provides a definition for ‘adverse’ or amber status as weather conditions
which, while not extreme, are known to be challenging to reliable operations. However, with effective maintenance,
timely delivery of seasonal preparation and robust deployment of mitigation measures, the full timetable can be
delivered with acceptable reliability.
15
Witness evidence shows a role known as a seasonal manager/planner was introduced in 2004. In 2012 the
Wessex route created a new performance team, and the role of seasonal planning and strategy manager was
introduced. In 2014 the title was changed to performance improvement specialist and in 2018 the title of the role
became seasons delivery specialist (SDS).
Report 12/2023 33 October 2023
Salisbury Tunnel Junction
Analysis
Adhesion Index Description
Good adhesion conditions expected
0 to 2 Leaf contamination unlikely except in very prone locations. Rails generally
dry or briefly damp.
Wet railhead expected
3 Rails damp or wet, generally devoid of leaf contamination away from prone
spots, but sufficient to reduce adhesion between the wheel and rail,
potentially leading to wheel slippage.
Moderate adhesion conditions
4 to 5 Moderate leaf fall risk with dry conditions. Slight contamination with damp
rail. Some disruption to the network could be expected, especially in
cuttings or densely vegetated areas.
Poor adhesion conditions
6 to 8 High leaf fall risk with dry conditions. moderate leaf fall contamination with
damp rail conditions. Disruption to the network likely if treatment not
completed.
Very poor adhesion to extreme leaf fall conditions
Very high contamination of the railhead due to leaf fall.
9 to 10
High to very high contamination of the railhead due to leaf fall and damp rail
conditions. Very high risk of disruption to the network
Figure 19: Replication of Network Rail’s Adhesion Index forecast table (courtesy of Network Rail).
NR OPERATIONS NR OFF TRACK
NR TRACK National Operations TRAIN OPERATOR Lineside Vegetation manual
Procedures (NOP) NR/L2/OTK/5201
Managing
operationaL risk
NR/L3/OPS/021/01 Autumn working arrangements
Railhead contamination Professional driving Policy
levels (PDP) Module 4
NR/L2/OPS/045 Tree management
Track design NR/L3OPS/045 RECOVERY OF
Reports of low adhesion
NR/L2/TRK/2049 RAILHEAD SWABS
(ROLA) Module 3
Vegetation plan
Track inspection SEASONAL DELIVERY
Working on or near the line High risk sites for wrong side Wessex Autumn working Module 2
NR/L2/OHS/019 track circuit failure (WSTCF) arrangements Lineside vegetation
and low adhesion (HRLA) NR/L3/OPS/021/01Wessex management
NR/L2/OPS/095 and Seasonal Calendars
NR/L3/OPS/021/01
Traction gel apparatus Module 1
(TGA) Lineside vegetation
NR/L3/TRK/3501/C01 inspection and risk
assessment
Lineside Vegetation
Inspection and Risk
assessments
F-01 Adhesion risk matrix
MOM Inspections
F-02-HRLA removal plan (High risk for low F 3079 vegetation
inspection
F-03 Removal form adhesion (HRLA))
F-04 Adhesion template
F 3077 Tree hazard
5201- Lineside vegetation NR/L3/OPS/021/01/Wessex inspection
risk assessment F 3076 Leaf fall
inspection
F 3270 Cab ride
inspection
Remote desk top
Work Arising Work Arising Maintenance delivery inspection
ELLIPSE
Information Form Information Form unit
NR/L3/MTC/MG0176
(WAIF) (WAIF) (MDU)
Fallen tree report
Mitigation for Autumn
Figure 20: Diagram showing relationships between the various departments within Network Rail and the
standards applicable to managing, preparing and collaborating internally and externally with the train
operator in planning and responding to autumn and low adhesion incidents.
Report 12/2023 34 October 2023
Salisbury Tunnel Junction
98 In August 2021, the Wessex route SDS presented and discussed the proposed
Analysis
arrangements for autumn 2021 with other Network Rail departments in the route
and with SWR.16 SWR then incorporated some of the relevant details from the
proposed Wessex autumn working arrangements within its autumn train driver
briefing document, which was issued in September 2021.
99 Network Rail’s Wessex route published and circulated its finalised autumn
working arrangements document on 15 September 2021. This document
identified high risk of low adhesion (HRLA) sites and outlined how the route would
take action regarding low adhesion across its network. The options available in
the document for the SDS to mitigate the risk from railhead contamination were to
arrange for railhead treatment through deployment of MPVs, install static lineside
traction gel applicators (TGAs), or arrange for inspections and manual railhead
treatment by a MOM.
Railhead treatment
100 The objective of railhead treatment is to remove crushed leaf film or other
contaminants from the railhead using high‑pressure water jetting with the option,
once the railhead is clean, of applying a sand-based gel known as an adhesion
modifier. This is intended to help break up any remaining leaf film on the railhead
and to raise the adhesion level by introducing a friction improver at the wheel/rail
interface.
101 Railhead treatment is delivered across longer stretches of track by MPVs and / or
railhead treatment trains (RHTTs). MPVs have less capacity than RHTTs and
generally need to return to the depot after each eight-hour shift. Network Rail’s
Wessex route only uses MPVs. In 2021, the route had six MPVs which ran over
nine circuits, covering more than 750 route miles.
102 Network Rail’s Wessex route’s autumn working arrangements document sets
out an aspiration for all routes to be covered by railhead treatment once every
24 hours throughout the autumn. During autumn 2021, Wessex route scheduled
its MPVs over the circuits twice each day, once in the morning and once in the
afternoon / evening, from Monday to Friday. Due to track access restrictions at
weekends, caused, for example, by engineering possessions, only one run was
scheduled per circuit on a Saturday and Sunday. This scheduling practice is
common nationally across Network Rail during weekends.
103 As well as the scheduled circuits, additional runs of the MPV can be made as
needed following requests from the SDS, signaller or other staff at the WICC (see
paragraph 110). Such ‘work as required’ requests usually follow reports of low
adhesion by train drivers or MOMs.
104 The working arrangements document defines a ‘missed site’ as any location
which was planned to be treated within the base plan but was not treated within
a 24-hour period, due to engineering access restrictions, fleet failure, resourcing
issues, or any other factor preventing railhead treatment.
16
The Railways and Other Guided Transport Systems (Safety) Regulations 2006 (as amended) require
infrastructure managers and transport undertakings to co‑operate to keep the railway system safe by sharing
information on risks, jointly meeting and reviewing risk assessments, and keeping their staff fully briefed. Rail
Industry Standard RIS-8040-TOM ‘Low Adhesion between the Wheel and the Rail - Managing the Risk’ (issued
December 2016, and updated in 2022) provides specific guidance on collaboration with respect to managing low
adhesion.
Report 12/2023 35 October 2023
Salisbury Tunnel Junction
105 If a planned treatment was to be delayed such that a site would not be treated
Analysis
in a 24-hour period, the Wessex route autumn working arrangements document
stated that the network delivery manager should issue a ‘missed site’ form by
03:00 hrs on the day the treatment was planned to take place. Train operators
were asked to acknowledge receipt of this form by email.
106 Railhead treatment can also be delivered across shorter stretches of track by
TGAs. TGAs are attached to the rail at a fixed location and deposit adhesion
modifier to the railhead before the passage of a train. The installation
and maintenance of TGA equipment is set out in Network Rail standard
NR/ L3/ TRK/3510/C01, ‘Use of Traction Gel Applicators’ (issue 1, September
2011). At route level, TGAs are managed by the SDS, who should arrange for the
equipment to be positioned taking account of the risks of incidents caused by low
adhesion. No TGA equipment had been installed on the Up or Down Salisbury
Main lines at the time of the accident.
Monitoring and reporting of low adhesion
107 Network Rail’s Wessex route’s autumn working arrangements document also
details how the WICC (figure 21) will respond to information about low adhesion
received through reports from train drivers, feedback from MOMs identifying
railhead contamination through their inspections, or notifications of wrong side
track circuit failures (figure 22). Wrong side track circuit failures can occur if
railhead contamination is such that it insulates trains’ wheels from the rails to a
degree that a train may be present but not detected. Such failures, which can be
indicative of low adhesion risk, are monitored within the WICC by the intelligent
infrastructure technician.
Incident control desk
Intelligent infrastructure Autumn control desk
technician (IIT)
Figure 21: Wessex integrated control centre showing the location of various autumn management roles.
Report 12/2023 36 October 2023
Salisbury Tunnel Junction
Analysis
WSTCF MPV
Autumn
IIT TGA
control desk
MOM
ROLA
WICC
Geographic
block working
Strategic (GBW)
Train driver / managment
signaller /
MOM GSM-R
Figure 22: Sources of information coming into the WICC from reports of low adhesion and wrong side
track circuit failures and mitigations that can be implemented to manage the low adhesion.
108 The need for drivers to report low adhesion conditions was set out in the railway
Rule Book GERT8000 module TW1, ‘Preparation and movement of trains’
(issue 16, March 2021) and in SWR’s train driver briefings. The Rule Book
classifies railhead adhesion as being ‘good’, ‘expected’ and ‘reportable’ with
drivers only required to report the ‘reportable’ conditions to the signaller. This is
defined as ‘Railhead adhesion is worse than would be expected for the location
and environmental conditions.’ Signallers then pass that information on to
operations control centres (such as the WICC) and also inform train drivers via a
GSM-R message if the reported location is on the approach to a signal that can
display a red aspect.
109 As well as section 28 of the Rule Book Module TW1, 'Preparation and movement
of trains', train drivers must also follow their train operating company’s processes
for reporting low adhesion, in this case the SWR Autumn brief for 2021. This
again will not require them to report low adhesion at locations where it is
expected. Consequently, a driver’s decision to report an incident is subjective
and dependent on their previous experiences, knowledge of the route, published
briefing notices (figure 23), and the environmental conditions on the day.
110 From 2020, the Wessex SDS implemented a scheme for an autumn control
desk within the WICC. Autumn controllers make sure that all autumn-related
information is collated and analysed in real time to identify any related emerging
risk and keep all relevant parties informed. As part of this, they monitor and
collate reports of low adhesion and wrong side track circuit failures from drivers
and signallers. They also have a role to play in managing these risks by deploying
available resources, such as MPVs, MOMs and maintenance resources, to
mitigate the emerging risks of low adhesion.
Report 12/2023 37 October 2023
Salisbury Tunnel Junction
Analysis
Figure 23: Weather and adhesion forecast notices placed alongside engineering notices in SWR's
Salisbury depot.
Network Rail audit regime
111 Network Rail’s management assurance processes are set out in its company
standard NR/L2/ASR/036, ‘Assurance Framework’ (issue 5, December 2017)
and are intended to provide assurance at every level of the organisation that risk
management systems are operating as intended. Network Rail has three levels
of assurance varying from level 1 (line manager assurance) to level 3 (the level
which provides the greatest independence).
112 A level 2 audit was carried out on Network Rail’s Wessex outer route MDU
between January and March 2021. A level 2 functional and management system
audit is conducted by an auditor independent from those with the responsibility
of implementing the risk controls. The level 2 audit highlighted observations and
non-compliances with the requirements of NR/L2/OTK/5201, ‘Lineside vegetation
management manual’ (see paragraphs 226 and 261). The audit team concluded
the overall rating was 'fair to good' with a 'low to moderate' probability of a risk
event occurring.
113 A separate level 2 audit of Network Rail’s Western route in September 2019
identified issues with the competency and training requirements for the role of an
SDS (see paragraph 257).
Report 12/2023 38 October 2023
Salisbury Tunnel Junction
Identification of the immediate cause
Analysis
114 Train 1L53 passed signal SY31 at danger and could not stop before colliding
with train 1F30.
115 Signal SY31 at Salisbury Tunnel Junction was at red (danger) as it was protecting
the paths of train 1F27 on the Up Dean line and train 1F30 on the Down Dean line
(figure 3).
116 Train 1F27 was running around five minutes late, having been due to traverse
Salisbury Tunnel Junction at 18:36 hrs. Train 1F30 was timetabled to pass
through the junction at 18:23 hrs but was running around 20 minutes late, while
train 1L53 was scheduled through the junction at 18:39 hrs and was running
around four minutes late.
117 The signaller believed that train 1F30 would pass through Salisbury Tunnel
Junction before train 1L53, and so they decided it would be more efficient to
provide train 1F30 with a clear route while holding train 1L53 at signal SY31 to
protect the movement.
118 Network Rail’s policy and instructions for train regulation at Salisbury require
signallers to give priority to passenger trains from the Dean lines unless those
trains are running late by five or more minutes. When trains are running late,
signallers use their experience to make decisions about the priority of services.
The signaller’s decisions and actions were consistent with these principles.
119 Train drivers are provided with the facility to make a railway emergency group call
which will cause other drivers who receive the message to bring their trains to a
stand immediately. Section 39.5 of the Rule Book17 states:
‘You must only use the emergency call facility when it is necessary to give
immediate advice for trains to be stopped or cautioned, or to call the emergency
services, in connection with an accident, obstruction or other exceptional
incident.’
120 The driver of train 1L53 stated that he was aware of this requirement but was
hoping that the train’s WSP and brakes would slow the train before the junction,
and he was concerned that a train could be bought to a stop across the junction.
By the time he saw signal SY31 was still at red he was becoming distressed.
Soon after, he saw train 1F30 approaching the junction and took action to leave
the cab (paragraph 68).
121 RAIB believes that, given the instruction, it is reasonable to assume that a driver
is unlikely to decide to make a railway emergency group call until they have
seen that the protecting signal has remained at red and their train is continuing
to travel uncontrollably. Analysis undertaken by RAIB indicates that, in such
circumstances, the driver of train 1F30 is unlikely to have received sufficient
warning to bring that train to a stop before a collision.
17
The Rule Book extract is taken from issue 16 of GERT8000-TW1 which came into force in June 2021.
Report 12/2023 39 October 2023
Salisbury Tunnel Junction
Identification of causal factors
Analysis
122 Train 1L53 passed signal SY31 and ran onto the junction due to a combination of
the following causal factors.
a. Wheel/rail adhesion was very low in the area where the driver of train 1L53
applied the train’s brakes (paragraph 123).
b. The driver did not apply train 1L53’s brakes sufficiently early on approach to
protecting signal SY31 to avoid running onto the junction, given the prevailing
levels of wheel/rail adhesion (paragraph 172).
c. The braking systems of train 1L53 were unable to mitigate the effects of the
prevailing wheel/ rail adhesion (paragraph 203).
Each of these factors is now considered in turn.
Very low wheel/rail adhesion
123 Wheel/rail adhesion was very low in the area where the driver of train 1L53
applied the train’s brakes.
124 Trains rely on friction between the wheel and rail to slow down and stop; this can
be affected by contaminants such as crushed leaves which, when combined with
moisture, reduce the level of wheel/rail adhesion (paragraph 81 and table 1).
While WSP systems (paragraph 40) can mitigate a reduction in wheel/rail
adhesion to a degree by maximising the use of available friction, ultimately, if
levels of friction are sufficiently low, a train will decelerate much more slowly than
normal, particularly on a descending gradient.
125 As the driver was braking, the very low level of wheel/rail adhesion in the area
where the driver braked meant that the train’s braking performance was lower
than normal (paragraph 62, figures 24, 36 and 40).
126 Evidence for low wheel/rail adhesion encountered by train 1L53 is provided by
data from the train and the driver’s account. OTDR data from train 1L53 shows
that the train’s WSP system was active throughout the brake application on
approach to signal SY31, as would occur when a train encounters low wheel/rail
adhesion conditions.
127 Following the accident, RAIB tested the train’s WSP system (see paragraph 203)
and used data from the train to derive the values of adhesion encountered by
train 1L53 from the point at which the brakes were applied on approach to signal
SY31 on the night of 31 October 2021. This analysis determined that the train was
likely to have encountered average levels of adhesion between 0.04 and 0.05.
This equates to very low wheel/rail adhesion conditions (as defined in table 1).18
128 The level of wheel/rail adhesion was very low due to contamination on the
railhead (see paragraph 129) and Network Rail’s Wessex route not effectively
mitigating the railhead contamination (see paragraph 156).
18
Recovery, tribology testing and analysis of railhead contamination samples was completed on site by RAIB and
University of Sheffield. Analysis of the train's systems and WSP was completed using the WSPER testing rig (see
paragraph 216). The combination of the results from the various testing and analysis indicated that the level of
adhesion on approach to signal SY31 is likely to have varied between low and very low.
Report 12/2023 40 October 2023
Salisbury Tunnel Junction
Railhead contamination
Analysis
129 The railhead was contaminated with fallen leaf debris, much of it as a result
of the weather conditions since the last railhead treatment run, coupled with
an increased density of vegetation, and wet conditions from the band of
drizzle that had recently passed over.
130 When RAIB deployed to the site of the accident on the night of 31 October 2021,
its inspectors undertook an initial visual examination of the Down Main line on the
approach to the accident site. The railhead generally showed signs of medium
contamination (level 2), but with some evidence of heavy contamination (level 3),
where the driver started braking and on the immediate approach to Fisherton
Tunnel. Comparison with the guidance given in Wessex route’s autumn working
arrangements document for assessing contamination indicated that the levels
of contamination found on the approach to the braking zone used by train 1L53
and on the approach to Fisherton Tunnel would be likely to have resulted in low
to very low adhesion (paragraph 84 and figure 24). All the contamination found
was comparable with that expected from successive train wheels crushing leaf
material and other contaminants onto the railhead, building up the thickness of
contamination in layers (paragraph 82). Testing undertaken in 2006 suggested
that a significant percentage of any railhead contamination would be removed
by water jetting (paragraph 83 and see paragraph 245). As such, much of the
contamination would have been a result of the weather conditions since the last
railhead treatment MPV run (see paragraph 133) coupled with a high density of
vegetation and topography on the various sections of the Down Main line (see
paragraph 136).
131 RAIB commissioned the University of Sheffield to conduct a detailed tribology
(friction) survey at 12 sample locations over 3.3 km on approach to the accident
site (figures 24, 25 and 26). This survey took place on 3 November 2021, while
the line remained closed and undisturbed. Measurements and samples were
taken under wet and dry conditions. In dry conditions, values of the coefficient of
friction were measured as being between 0.09 and 0.35, while for wet conditions
the values were measured as being between 0.02 and 0.20. The University of
Sheffield also noted that ‘most wet values for the contaminated rail head samples
that were analysed were below those required for effective braking’.
132 A sample taken at 81 miles 40 chains (where the train had full service braking
applied) showed an increase in the level of silica (a component of sand). This may
have been due to the activation of sanding systems on trains passing over the
line, including the one on train 1L53. Subsequent sample locations on approach
to the accident site had an appearance consistent with train wheels having slid
over them.
Weather conditions on the day of the accident
133 In addition to the leaf fall density provided by MetDesk (figure 18), Met Office data
going back to 2016 showed that, around the time the accident occurred, the trees
in the area were approximately two weeks behind the normal leaf fall schedule,
having around 65 to 70% of their leaves remaining. The wind and rain on the day
of the accident (paragraph 43) almost certainly resulted in increased leaf fall and
this is likely to have contributed to the high levels of contamination observed on
the section of line approaching Salisbury and the approach to the junction.
Report 12/2023 41 October 2023
Salisbury Tunnel Junction
Analysis
Level 0: Railhead has no
contamination
1
Level 1: Railhead has a lot of silver
showing - light contamination
2
SY29R
Level 2: Railhead has some silver
showing - medium contamination
Broken Cross bridge
Level 3: Railhead has no silver 3-5 Tree debris
showing - heavy contamination
Note: Photographic evidence
6 shows that contamination
BS1 extended thoughout the whole
7 BS2 length of the track, including the
FSB areas between the locations
where samples were taken from.
EB
8 BS1 Brake step 1 (81 m 27 ch)
SY29 BS2 Brake step 2 (81 m 28 ch)
Salisbury 9
10-11
Tunnel FSB Full service brake (81 m 38 ch)
12
Junction EB Emergency brake (81 m 49 ch)
SY31 © Crown Copyright. All rights reserved. Department for Transport 100039241. RAIB 2023
Figure 24: Post-accident sample locations and (below) Network Rail classification of contamination
levels (NR/L3/OPS/045/4.07).
Report 12/2023 42 October 2023
Salisbury Tunnel Junction
Analysis
Figure 25: Location (left) and railhead condition (right) at 80 miles 40 chains.
Figure 26: Location (left) and railhead condition (right) at 82 miles 27 chains beyond signal SY31 and
approaching Salisbury Tunnel Junction.
134 Network Rail records for 31 October 2021 showed that fifteen trees fell on railway
lines within the Wessex route (paragraph 48), causing major disruption. Although
trees do not naturally shed their leaves when they have been felled, the resulting
detritus will spread and may result in further localised contamination of the
railhead.
135 Wet wheel/rail adhesion levels on a contaminated railhead are significantly lower
than dry adhesion levels. The wheel/rail adhesion conditions encountered by
train 1L53 would have been adversely affected by a band of drizzle which passed
over the Salisbury area between 18:35 and 18:40 hrs (figure 6), around five
minutes before the train approached Salisbury Tunnel Junction. This also meant
that train 1L53 probably saw significantly reduced wheel/rail adhesion compared
to the two previous trains that had been required to stop at signal SY31 (see
paragraph 178).
Report 12/2023 43 October 2023
Salisbury Tunnel Junction
Vegetation management
Analysis
136 Vegetation management is undertaken by the off track section within Network
Rail’s maintenance organisation19 in accordance with the requirements of
Network Rail standard NR/L2/OTK/5201, ‘Lineside vegetation management
manual’ (issue 5, December 2020) (figure 27). This standard covers all elements
of vegetation inspection, assessment and management of the risks associated
with trees falling on or near the railway, ensuring necessary sightlines to signals,
autumn leaf fall risk and preserving biodiversity. When issues are identified
during vegetation inspections, off track staff raise a work order using a work
arising identification form (WAIF) which is entered into the Network Rail Ellipse
maintenance system.20 This form specifies the work to be undertaken and the
recommended timescales for completion, from immediate action up to 12 months,
depending on the associated risk. Inspections are to be carried out during periods
of vegetation growth in summer. Therefore, any actions that may have an impact
on low adhesion risk that year would need to be completed within three months of
the inspection so that the risk is mitigated before the autumn season.
137 Network Rail standard NR/L2/OTK/5201 Module 03, ‘Route Vegetation
Management Plans’ allows the section manager for off track to reprioritise
lineside vegetation work. NR/L3/MTC/MG0176, ‘Ellipse Management Handbook:
Prioritisations, Reprioritisations and Cancellations’ allows for deferment of
work on Ellipse up to five times before authorisation is required from the SAE.
Depending on the risk from the vegetation at the time of the reprioritisation, the
work can be delayed by up to 12 months on each occasion.
NR/L2/OTK/5201/Module 1
F3079 F3270 F3077 F3076
Vegetation inspection Lineside vegetation Remote survey Tree inspection Leaf fall inspection
On foot Cab ride / video On foot On foot
36 - 44 months 12 - 16 months 60 - 68 months 30 - 36 months 60 - 68 months
Figure 27: The various assessment regimes and periodicities for tree, vegetation and leaf fall within
Network Rail standard NR/L2OTK/5201/module 1.
19
Wessex Route Area Services team also managed contractors to undertake off track maintenance, including
vegetation clearance on the Wessex route.
20
Ellipse is an integrated asset and work management system which helps planning of work to be performed on
the railway infrastructure (see also figure 41).
Report 12/2023 44 October 2023
Salisbury Tunnel Junction
138 NR/L2/OTK/5201 specifies a suite of vegetation inspections, one of which is an
Analysis
on-foot leaf fall inspection to be undertaken every 60 months. These are to be
undertaken by off track inspection staff within the MDU or can be contracted
out. During these inspections, the inspector is required to complete form
NR/ L2/ OTK/5201/F3076, which scores each one-eighth mile (approximately
200 metres) using criteria including tree density, tree size, distance from the
track, the presence of other vegetation and topography. The output from a leaf
fall assessment is a numerical score from 0 to 32. This figure is also entered into
Ellipse.
139 Operation standard NR/L2/OPS/095, ‘High Risk Sites for Wrong Side Track Circuit
Failures in Leaf Areas and for Low Rail Adhesion’ (issue 6, June 2019) describes
Network Rail’s business process for identifying and managing the risks from
railhead contamination causing low adhesion and WSTCF. The standard requires
the SDS to convert the leaf fall inspection score from NR/L2/OTK/5021/ F3076 into
a risk level and to reassess each location on an annual basis. NR/ L2/ OPS/095
classifies a score above 26 as high risk (risk category 5). The standard requires
that high risk sites are added to Network Rail’s record of HRLA sites and that the
details are shared with the operational standards department of relevant train
operators. The standard also requires the SDS to lead and co-ordinate plans to
remove high risk sites (removal plans), although individual plans may be owned
by others, for instance the SAE or track maintenance engineer (TME). The SDS
is also required to consider appropriate mitigations in the interim to reduce the
probability of a low adhesion incident, such as a SPAD. RAIB observes that
NR/ L2/OTK/5201 and form NR/ L2/ OTK/5021/ F3076 both also classify a score
between 21 and 26 as being ‘high risk during the peak leaf fall period and wet
conditions’, however NR/ L3/OPS/095 does not require any action to be taken for
sites with these scores (paragraph 88).
140 Standard NR/L2/OPS/095 also requires that all HRLA and high risk of
WSTCF sites are subject to an adhesion risk assessment process using form
NR/ L2/ OPS/095/F01. This takes into account the associated probability and
impact of a low adhesion event and classifies the risk at locations on a scale of
between 2 (low risk) and 10 (high risk).
141 In 2018 and 2019 Network Rail revised standard NR/L2/OTK/5201 with the aim
of aligning its operational and vegetation management standards and to include
more detail on undertaking lineside inspections. It also rolled out an associated
training programme to facilitate more inspections by MDU staff, in particular the
leaf fall inspections. The training programme was provided to MDU managers
with an expectation that they would cascade the training to off track inspectors
undertaking lineside inspections. Witness evidence is that training was not
provided to arboriculture contractors used by Network Rail’s Wessex route,
although senior managers in Network Rail expressed an opinion that contractors
would be included in the cascading of such training. However, RAIB observes that
NR/L2/OTK/5201 states that ‘contractors are responsible for undertaking their
own technical and awareness briefings in accordance with their own processes
and procedures’.
Report 12/2023 45 October 2023
Salisbury Tunnel Junction
142 Network Rail’s Wessex route was unable to provide any data from leaf fall
Analysis
inspections before 31 October 2018. However, witness evidence suggests
that from 2016 a section of track between Overton (55 miles 42 chains) and
Whitchurch (59 miles 8 chains), just over 20 miles from the accident site, had
been classified as a ‘supersite’, that is, one suffering from excessive leaf fall
resulting in railhead contamination causing frequent WSTCF. Figure 28 shows
the effects of vegetation work undertaken during 2019/20. There is no evidence
to suggest that the track in the vicinity of the accident had been classified as a
‘supersite’. In the absence of any historical documentation, RAIB was unable to
confirm the technical basis for categorising a site as a low adhesion ‘supersite’,
but it is possible that this was what later became known as an HRLA site.
Figure 28: The main lines between Overton and Whitchurch looking towards Salisbury showing the
condition of the vegetation before (taken in summer) and after (taken in winter) vegetation work was
completed in 2019/20.
143 Leaf fall risk inspections were completed between 31 October and 2 November
2018 from 50 miles to 83 miles on the Salisbury line. Although Network Rail could
not provide copies of the completed leaf fall inspection forms, RAIB was able
to view the leaf fall inspection scores that had been supplied to the then SDS
and SAE. These scores showed that sites on the main lines between 80 miles
30 chains and 80 miles 40 chains, and 81 miles 10 chains and 81 miles 20 chains
had been categorised as level 5 sites, with scores of 27 and 28, respectively.
Sites between 80 miles 50 chains and 80 miles 70 chains, and 81 miles 20 chains
and 81 miles 30 chains were categorised as level 4 sites. Rather than having
two individual one-eighth mile sections close together, the entire length of track
covering both level 5 sites and the track in between, almost a mile in total, was
designated as a single HRLA site (figure 30). This single site encompassed the
Down Main line starting at Broken Cross bridge and terminating around 940
metres on approach to signal SY29. RAIB found that these scores were not
entered into Ellipse. In addition, the site was not added to the table of HRLA sites,
and no vegetation plan or mitigation was actioned for it.
Report 12/2023 46 October 2023
Salisbury Tunnel Junction
144 In 2019 a further leaf fall inspection was undertaken, for which again only the
Analysis
scores were available. These indicated that the main lines between 80 miles
30 chains and 81 miles 20 chains had again been categorised as a level 5 site.
RAIB found again that these scores were not added to Ellipse, nor was the site
added to the table of HRLA sites, although it was subsequently added in 2020.
Despite the site being added to the table, no vegetation plan or mitigation was
actioned for it by Network Rail and the site was not included in SWR’s 2020
autumn briefing.
145 RAIB has been unable to determine why no removal plan or mitigations were
implemented for this HRLA site between 2018 and 2020. The SDS in post at the
time of the accident reported that they were unaware that no action had been
taken. However, RAIB found that the collaboration between the SDS, SAE, and
MDU TME and SM for off track staff had been ineffective and the SDS had not
checked on the presence or status of any removal plans because it had “always
been like that” (see paragraphs 229, 231 and 253).
146 In June 2021, arboriculture specialist contractors undertook a further leaf fall
inspection of the Down Main line. The results from this assessment showed
the presence of several level 4 sites between 80 miles 30 chains and 82 miles
30 chains but no level 5 sites (figure 29). The risk scores resulting from this
inspection for the sites that had previously been identified on the Down Main line
in 2018 and 2019 as level 4 or level 5 were scored at lower levels. Of particular
note was the section around 80 miles 30 chains. This was reassessed as a level
1 site having previously been scored as a level 5 site. This scoring meant that
none of the line was categorised as being an HRLA site and, as such, no actions
to reduce the trees or vegetation were required. However, RAIB found that this
section was not removed from the table of HRLA sites, likely because the SDS
risk assessment process (paragraph 139) is undertaken on an annual basis and
requires consideration of other operational experience as well as the leaf fall risk
assessment.
147 No action to manage the vegetation had been taken between 2018 and 2021
and video footage shows that vegetation continued to grow on the main lines
approaching Salisbury Tunnel Junction between 2019 and 2021. A comparison
between the 2021 scores and those from earlier inspections therefore brings into
question the robustness of the process. It is also unclear whether classification of
sites with an assessment score of between 21 and 26 as ‘high risk during peak
leaf fall period and wet conditions’ in NR/L2/OTK/5201/F3076 may have affected
the contractors' scoring. As explained in paragraph 139, operational standard
NR/L2/OPS/095 did not require any action to be taken for such sites, but the
contractors were unlikely to have been aware of this. Figure 32 shows the change
in the management of vegetation at Salisbury Tunnel Junction over the previous
decades.
Report 12/2023 47 October 2023
Salisbury Tunnel Junction
Analysis
To London
N BS1 Brake step 1 (81 m 27 ch)
BS2 Brake step 2 (81 m 28 ch)
FSB Full service brake (81 m 38 ch)
EB Emergency brake (81 m 49 ch)
80 m 30 ch: Embankment 80 m 30 ch: Flat
2 Ash-Shrub-Sycamore Open-Shrub-Sycamore-Brambles 1
80 m 40 ch: Embankment SY29R 80 m 40 ch: Step
1 Shrub-Sycamore Open-Shrub 1
80 m 50 ch: Cutting 80 m 50 ch: Cutting
3 Shrub-Sycamore Ash-Shrub-Sycamore 3
80 m 60 ch: Step 80 m 60 ch: Step
3 Shrub-Sycamore Shrub-Sycamore-Scrub 3
80 m 70 ch: Cutting 80 m 70 ch: Cutting
3 Shrub-Sycamore Ash-Shrub-Sycamore-Scrub 3
81 m 00 ch: Cutting 81 m 00 ch: Cutting
3 Open-Shrub-Scrub Oak-Shrub-Scrub 3
81 m 10 ch: Embankment 81 m 10 ch: Embankment
3 Ash-Shrub-Sycamore Shrub-Sycamore 4
81 m 20 ch: Flat 81 m 20 ch: Embankment
3 Ash-Shrub-Sycamore Shrub-Sycamore 3
81 m 30 ch: Flat 81 m 30 ch: Embankment
2 Ash-Shrub-Sycamore BS1 Ash-Shrub-Bramble 2
BS2
81 m 40 ch: Flat 81 m 40 ch: Embankment
2 Ash-Shrub-Sycamore Ash-Shrub-Sycamore 2
FSB
81 m 50 ch: Flat 81 m 50 ch: Embankment
2 Shrub-Sycamore-Brambles Ash-Shrub-Sycamore-Brambles 2
EB
81 m 60 ch: Flat 81 m 60 ch: Embankment
2 Ash-Shrub-Sycamore-Scrub Ash-Shrub-Sycamore-Scrub 2
81 m 70 ch: Flat 81 m 70 ch: Embankment
2 Ash-Shrub-Sycamore-Scrub Shrub-Sycamore-Scrub 2
SY29
82 m 00 ch: Flat 82 m 00 ch: Flat
2 Ash-Shrub-Sycamore-Scrub Ash-Shrub-Sycamore-Scrub 2
82 m 10 ch: Flat 82 m 10 ch: Flat
1 Shrub-Sycamore-Brambles Open-Shrub-Sycamore-Brambles 2
82 m 20 ch: Flat 82 m 20 ch: Step
3 Ash-Shrub-Sycamore-Brambles Ash-Shrub-Sycamore-Brambles 2
82 m 30 ch: Cutting SY31 82 m 30 ch: Cutting
4 Shrub-Sycamore-Scrub Ash-Shrub-Sycamore 4
To Southampton
To Salisbury © Crown Copyright. All rights reserved. Department for Transport 100039241. RAIB 2023
Figure 29: Network Rail (Wessex) data in diagrammatic format showing mileage, topography, tree
species and leaf fall risk assessment data scores (June 2021).
Report 12/2023 48 October 2023
Salisbury Tunnel Junction
Analysis
Figure 30: Table of high risk low adhesion sites showing details of the location, engineers line reference
(ELR), line, start and end details (miles and chains). Misspelling of Grateley and other names are from
the original document.
ACTION
IMMEDIATE
ACTION
ALERT ALERT
45° xx
xxx
6.0 m
Figure 31: RAIB diagram showing the action zones from Network Rail standard NR/L2/OTK/5201,
train profiles similar to class 159 trains, and tree growth as found on certain sections of the main lines
between signals SY29R and SY31 (red dotted line).
Report 12/2023 49 October 2023
Salisbury Tunnel Junction
Analysis
Figure 32: The Up and Down Dean and Main lines looking towards Fisherton Tunnel in 1963 (top left,
courtesy of Crecy publishing), in the 1970s (top right, courtesy of Crecy publishing), and on 1 November
2021 (main image).
148 Furthermore, an examination of the section of line from signal SY29R to beyond
signal SY31 after the accident (paragraph 130) showed there were areas of
medium to heavy railhead contamination which would have resulted in low
levels of wheel/rail adhesion, even on sections that had low scores in the 2021
risk assessment. The amount and condition of the vegetation and trees were
also seen to be non-compliant with the wider requirements of NR/L2/OTK/5201
(figure 31).
149 While the risks of trees falling or encroaching onto the railway are separate from
those associated with leaf fall, any vegetation management activity undertaken in
response to these risks will likely have associated benefits for preventing leaf fall
and reducing the likelihood of low adhesion.
Report 12/2023 50 October 2023
Salisbury Tunnel Junction
150 Witness and documentary evidence shows that an on-foot inspection of the
Analysis
vegetation was also undertaken by a member of Wessex outer MDU’s off track
staff on 14 June 2019. The inspection recommended trees and vegetation be cut
back as they had encroached into the ‘action zones’ as shown within Network Rail
standard NR/L2/OTK/5201 (figure 31). Records showed that the data was entered
into the Ellipse system in June 2019 with a timescale for action within 12 months,
although on the 12-monthly review this was deferred by a further 12 months to
June 2021.
151 The action timescales and their subsequent deferment may have been affected
by the lack of contextual information available to Wessex outer MDU off track
staff. Although MDU off track staff were provided with the Wessex autumn working
arrangements document showing the locations designated as HRLA sites and
where WSTCF had occurred as a result of low adhesion, this information was
not used by them, nor were they given any details of recommendations from
previous inspections. As a consequence of this, work could be reprogrammed
without taking account of the wider implications of leaf fall increasing the risk of
low adhesion.
152 The last vegetation inspection undertaken by the MDU before the accident was
carried out remotely between 13 and 14 June 2021 by the section manager
reviewing footage that had been taken the previous week from high-definition
trainborne cameras (figure 33).21 No WAIFs were raised from this inspection since
the section manager was aware that existing work orders were already in Ellipse
from the previous inspection in 2019. The tree and vegetation work to reduce the
risk was further deferred to April 2022 (paragraph 150).
Figure 33: Automated Intelligent Video Review software for cab/remote assessments of vegetation
(image courtesy of Network Rail and One Big Circle).
21
Remote monitoring was introduced as a result of new lineside inspection safety restrictions and cab ride
restrictions due to the COVID-19 pandemic.
Report 12/2023 51 October 2023
Salisbury Tunnel Junction
153 The off track section manager and the TME stated that much of the vegetation
Analysis
work in their area had been deferred due to other higher priority work, including
drainage assessments in response to the accident at Carmont on 12 August
2020 (see paragraph 326), as well as resource issues in the off track section (see
paragraph 225).
154 This lack of action was possibly also because the TME and off track section
manager were not aware of the risk from leaf fall, as the MDU off track staff had
not entered the leaf fall scores into Ellipse, and no vegetation management plan
for leaf fall risk had been developed.
155 However, witness evidence also suggests that, even if the leaf fall scores had
been entered into the Ellipse system, it is unlikely that this would have resulted in
actions to address the risk through vegetation management. This was because off
track staff in the MDU neither had the competence to interpret the leaf fall data,
nor did they accept that the leaf fall inspection process and managing the risk of
leaf fall reducing railhead adhesion was within their remit, despite it being defined
as such in Module 1 of NR/L2/OTK/5201. MDU off track staff believed instead that
this was a track maintenance issue (see paragraph 259).
Management of railhead contamination
156 Network Rail’s Wessex route had not effectively mitigated the railhead
contamination.
157 Besides the management of trees and vegetation, infrastructure managers
can take proactive and reactive steps to mitigate the risks arising from railhead
contamination. However, Network Rail’s Wessex route did not take effective
action to address these risks on the approach to Salisbury Tunnel Junction over
the weekend of 30/31 October 2021.
Proactive mitigation measures
158 As part of Network Rail’s planning for autumn, an SDS gathers information to
understand the status of lineside vegetation and whether the level of risk requires
proactive mitigation. At HRLA sites, this mitigation can include signage to warn
drivers of low adhesion, and TGA equipment, although neither of these were
actually installed on the Down Main line on the approach to Salisbury. Scheduling
runs of the railhead treatment MPV fleet using both water and adhesion modifier
can also be implemented. An SDS can also arrange for sections of track outside
of HRLA sites to be treated as needed.
159 The site where the driver braked train 1L53 on the Down Main line on the
approach to Salisbury Tunnel Junction did not include any HRLA sites. However,
the area was nevertheless scheduled for treatment with water using the MPV
throughout the leaf fall season.
160 In accordance with the 2021 autumn working arrangements, a railhead treatment
MPV was due to run over the Up and Down Main lines once on the morning of
Saturday 30 October 2021 and again on the afternoon of Sunday 31 October
2021. The MPV ran as scheduled over the Down Main line and Salisbury Tunnel
Junction at around 11:06 hrs on 30 October. However, although the MPV was due
to pass over this section of track again at 17:03 hrs on 31 October, around 1 hour
and 40 minutes before the accident, this run was postponed until 23:00 hrs on
31 October. This postponement was programmed on 25 October and was due to
planned engineering work, which would block the exit from the MPV depot.
Report 12/2023 52 October 2023
Salisbury Tunnel Junction
161 Witness and documentary evidence shows that Network Rail’s autumn mitigation
Analysis
measures, such as the railhead treatment programme, were given a lower priority
than timetabled passenger services and engineering work. This meant that MPV
treatment runs were frequently delayed or cancelled at weekends when much
scheduled engineering work takes place.
162 Because the delay to the MPV run on 31 October was planned in advance, and
the MPV would still run within the planned calendar day, Network Rail did not
consider the postponement to be a missed site. RAIB observes that the autumn
working arrangements document includes an aspiration for routes to be treated
once per 24 hours (paragraph 102) and therefore this decision was inconsistent
with that aspiration. However, the fact that it was not considered as a missed
site meant that no additional mitigations or control measures were considered or
adopted, even though Network Rail’s Wessex route control staff were aware that
poor weather with a high risk of leaf fall was predicted (see paragraph 250).
163 The MPV ran in accordance with the autumn working arrangements plan during
the previous week. RAIB found no faults with the water jetting equipment on the
MPV used on 30 October, and its OTDR and GPS records confirm that the Down
Main line, including approach to signal SY31, was treated, although water jetting
was suspended on the immediate approach to Salisbury Tunnel Junction. RAIB
cannot be certain of the effectiveness of the treatment on 30 October, however,
the performance of train 1L53 and inspections carried out after the accident
(paragraphs 130 and 131) suggest that any significant benefit of the treatment
had been nullified by the time of the accident.
164 Although the MPV normally treats the railhead with water jetting, treatment can
be enhanced with the addition of an adhesion modifier (paragraph 100). However,
the SDS did not consider the area in which the driver of train 1L53 braked
(81 miles 27 chains to 82 miles 35 chains) to present a risk that required the use
of adhesion modifier. The SDS stated that the previous SDS had had reservations
about the effectiveness and value for money of using an adhesion modifier
and the incumbent SDS did not believe that decision needed to be reviewed.
Furthermore, even if the SDS had seen the value of using an adhesion modifier,
they are unlikely to have requested its use outside of an HRLA site. This meant
that the railhead was only treated with water jetting. Although it is possible that the
use of an adhesion modifier on 30 October would have improved the wheel/rail
friction, any benefit would probably have been reduced by the passage of trains
and additional leaf fall by the time of the accident.
Reactive mitigation measures
165 On Friday 29 October 2021 the SDS chaired a tri-weekly teleconference to
discuss the autumn working arrangements for the next few days, in accordance
with the Wessex autumn working arrangements document. The meeting involved
several stakeholders, including MetDesk and the Met Office, and provided an
opportunity to consider any necessary operational mitigations that might be
required in response to forecast weather.
Report 12/2023 53 October 2023
Salisbury Tunnel Junction
166 At the meeting, the forecasters confirmed that adverse weather would occur
Analysis
over the weekend of 30/31 October, with inland wind gusts of 40 to 45 mph
(64 to 72 km/h). The forecasts predicted that it was likely that the rail network in
Hampshire and Wiltshire, including the area from Andover to Salisbury, would be
affected by tree and leaf fall. The forecast predicted ‘poor’ adhesion conditions,
the second- worst rating on the scale (figure 18).
167 Since no notes or actions from the meeting were required to be recorded, there
is no documentary evidence of the risks of the weather forecast being considered
during this meeting. Witness evidence indicates that some staff at the meeting
were aware that trees and vegetation on the Salisbury line had not been cut back
since 2019, and that the MPV run on 31 October 2021 had been postponed.
Despite this, no additional reactive mitigation measures were implemented in
response to the forecast.
168 The MetDesk weather forecast provided to Network Rail on 29 October 2021
predicted an ‘adverse’ weather hazard on the Wessex outer route. The MetDesk
adhesion index forecast for the period from 06:00 hrs 31 October to 06:00 hrs
1 November was emailed to the Wessex autumn controller at 12:31 hrs on
31 October.22 This predicted ‘red’ (poor) adhesion conditions, the second worst
category (figure 18).
169 Network Rail’s Wessex autumn working arrangements stated that the MetDesk
information is used in the decision to convene an EWAT and Met Office forecasts
are for information. Control room staff at the WICC are only required to take
action in response to extremely poor weather actually occurring; they are not
required to respond to forecasts of low adhesion. They are only required to
respond to low adhesion if they are alerted to it by WSTCF activations, which may
be caused by leaf fall railhead contamination, or reports from train drivers or from
MOMs inspecting HRLA sites (see paragraph 250).
170 On 31 October 2021, the weather itself did not trigger any extreme weather
response and there had been no WSTCF or reports of low adhesion from train
drivers. Furthermore, no MOM inspections were carried out on HRLA sites during
the weekend of 30/31 October (Wessex autumn working arrangements require
regular MOM inspections of HRLA sites), because there was no MOM on duty for
the Salisbury area, a known issue (see paragraph 241).
171 Although the impact of the weather on the infrastructure became apparent during
the day, not least by virtue of the number of weather-related incidents, the likely
implications of the weather on leaf fall and railhead contamination were not
identified by any managers within the WICC (see paragraph 250).
22
RAIB understands that normally the autumn controllers would expect to receive the MetDesk adhesion forecast
by 06:00 hrs and that the late delivery on 31 October was due to staffing issues.
Report 12/2023 54 October 2023
Salisbury Tunnel Junction
The driver’s braking point
Analysis
172 The driver did not apply train 1L53’s brakes sufficiently early on approach
to protecting signal SY31 to avoid running onto the junction, given the
prevailing levels of wheel/rail adhesion.
173 The driver of train 1L53 stated that, when signal SY29R displayed a double yellow
aspect, he normally used Broken Cross bridge as a visual cue to start braking.
Broken Cross bridge is 560 metres beyond signal SY29R and 1,780 metres on
approach to signal SY29. It is also 2,560 metres on approach to signal SY31.
174 OTDR and witness evidence suggests that other SWR drivers normally start
braking at various locations between Broken Cross bridge and the permissible
speed warning indicator for the permanent speed restriction. This is positioned
1,890 metres on approach to signal SY31, meaning that the normal braking point
described by the driver of train 1L53 was consistent with the range of those used
by other drivers.
175 The driver stated that on the evening of the accident he intended to use the fallen
tree and associated debris that he had seen on his earlier journey to London as
his marker after which to start braking (paragraph 47 and see paragraph 181).
The tree was 250 metres beyond Broken Cross bridge and braking here was also
consistent with the range of braking points used by other drivers. However, OTDR
data shows that the driver actually applied the brakes approximately 750 metres
beyond the location of the fallen tree and associated debris and 1,000 metres
beyond Broken Cross bridge, with the train travelling at 86 mph (138 km/h)
(figure 35 and see paragraph 189).
176 Analysis of class 159 brake testing data showed that, from this speed and under
normal adhesion conditions (that is to say, with the railhead dry and free from
significant contamination) and adjusted for the gradient, the train would have
stopped in a distance of between approximately 600 and 700 metres using brake
step 3. This means that, in the absence of wheel slide, the train would have
stopped before it passed signal SY31 if a braking application had been made at
either the location of the fallen tree (2,300 metres on approach to signal SY31) or
at the actual braking point (1,560 metres on approach to signal SY31).
177 RAIB analysis based on results from the WSP Evaluation Rig (WSPER) testing
showed that, given the adhesion conditions on the evening of 31 October 2021,
the accident would probably have been avoided if the brakes of train 1L53 had
been applied at Broken Cross bridge and, in this case, it is possible that the
train would also not have passed signal SY31 at danger. If the brakes had been
applied at the fallen tree, around 250 metres beyond Broken Cross bridge, then it
is likely that the train would have passed signal SY31 at danger, but it is possible
that the collision would have been avoided or occurred at a much lower speed.
Report 12/2023 55 October 2023
Salisbury Tunnel Junction
178 Before the passage of train 1L53 and the change in railhead conditions as a
Analysis
result of the drizzle that passed over just before it, two earlier trains had stopped
at signal SY31, one at 16:29 hrs and the other at 16:45 hrs. Both started braking
earlier than train 1L53. The first of these trains, train 1L43 (formed of two
three- carriage class 159 units), started to brake before signal SY29R (before
the location of the fallen tree) while travelling at 78 mph (126 km/h) and took
approximately 3,250 metres to stop. This consisted of approximately 2,050 metres
in brake step 1, 80 metres in step 2 and 15 metres in step 3. For the remainder
of the distance, train 1L43 was coasting and stopped approximately 30 metres
before signal SY31. The weather was dry at this time and, although OTDR data
showed that WSP activity occurred, the train was able to stop before the red
aspect being shown at signal SY31 (figure 34). The driver of train 1L43 adopted a
‘lighter and longer’ braking technique that they described as normal for them, and
something they had been told to do by a former driver instructor. The driver of this
train also stated that they were aware of the weather conditions causing some low
adhesion during their journey.
179 The last train to encounter signal SY31 showing a red aspect before train
1L53 was train 1L45 (formed of a three-carriage class 159 unit). At 16:45 hrs,
when the weather was still dry, the driver of train 1L45 braked shortly before
the PSR warning board, beyond the location of the fallen tree, but around 420
metres before the actual braking point of train 1L53, while travelling at 84 mph
(135 km/h). Train 1L45 took approximately 1,980 metres to stop. This consisted
of approximately 790 metres in brake step 1 and 1,080 metres in brake step 2.
For the remainder of the distance, train 1L45 was coasting. Train 1L45 stopped
approximately 25 metres on approach to signal SY31. Although the driver of train
1L45 braked later that the driver of the earlier train 1L43, and encountered a
greater level of WSP activity, this train was also able to stop before passing signal
SY31 (figure 34).
180 After the accident, driver instructors with experience of driving the route on
the approach to Salisbury Tunnel Junction stated that, taking into account the
environmental conditions on the day, they would have expected a driver who had
been told about low adhesion and had received a cautionary aspect at signal
SY29R to be making a brake application no later than signal SY29R.
The driver’s braking decision
The driver’s understanding of the adhesion conditions on the day
181 The driver stated that he intended to use the fallen tree and associated debris
as his marker after which to start braking, because he believed that the leaves
and debris from the tree would create a risk of low adhesion and a risk of the
train’s wheels sliding. Wheel slide is undesirable in itself because it can cause
flat spots on train wheels. The driver stated that he was aware that the formation
of the train was shorter than normal, and that this would affect the train’s braking
capabilities,23 but that he believed there was still sufficient braking distance
beyond the tree debris, and up to the point where he actually braked, to stop
before passing signal SY31 (paragraph 175 and see paragraph 213).
23
In accordance with the SWR professional driving policy SWR train drivers are required to know how shorter train
formations can affect the performance of train braking and adapt their driving and braking technique accordingly.
Longer trains generally have better braking performance in low adhesion conditions due to the additional
conditioning of the railhead by the increased number of wheels. Also, trains operating in multiple have more braked
axles and additional sanders.
Report 12/2023 56 October 2023
Salisbury Tunnel Junction
OTDR speeds and brake steps for train 1L53 and two previous trains which did stop at signal SY31,
Analysis
including published low adhesion site and points of interest
120
Observed 50 mph PSR 1L53 braking point
Broken Cross bridge warning board 50 mph PSR commencement board
position of
SY29R tree debris SY29 SY31
100
80
OTDR speed (mph)
60
1L43 OTDR speed
1L45 OTDR speed
1L53 OTDR speed
40
AWS magnets
TPWS grids
Signals
20
Points of interest
Published low adhesion area
0
80.25 80.5 80.75 81 81.25 81.5 81.75 82 82.25 82.5
Mileage
Normal braking Location of Actual braking
position tree and debris position
1L53
1L45
1L43
80.25 80.5 80.75 81 81.25 81.5 81.75 82 82.25 82.5
Mileage
Release Brake step 1 Brake step 2 Full service braking Emergency braking
Figure 34: Speeds for trains 1L43 (blue), 1L45 (orange) and 1L53 (grey) as recorded by their respective
OTDRs, and important locations. The red dashed line indicates an approximation of train 1L53’s ‘ground
speed’. The OTDR has recorded the measured wheel as travelling more slowly because it is sliding.
A similar but less pronounced effect can be seen during part of train 1L45’s data. Braking techniques
employed by train drivers of trains 1L43,1L45 and 1L53 are shown in the graph below.
Report 12/2023 57 October 2023
Salisbury Tunnel Junction
Analysis
18:42:02 (86 mph) 18:42:03 (86 mph) 18:42:14
Brakes to step 1 Brakes to step 2, Emergency
axle starts to slide brakes applied
18:41:50
AWS warning for 50
mph PSR warning
acknowledged. Brakes not 18:42:08
applied for 13 seconds Brakes to full service
18:42:03. Step 2 braking
applied
18:42:56
Loss of brake continuity,
18:41:09 (91 mph) traction interlock, speed
Power moved signal and desk controls
from notch 2 to off
18:41:19 (90 mph) 18:42:47
AWS warning for AWS warning
SY29R acknowledged for SY31
acknowledged
18:42:49
SY31 TSS
18:41:50 (87 mph)
AWS warning 18:42:21
for 50 mph AWS warning for 18:42:18 18:42:42.
PSR warning SY29 acknowledged 50 PSR OSS (Below SY31 OSS (set speed 34.5
acknowledged several times set speed 81.5 mph) mph), TPWS brake demand
Figure 35: Annotated OTDR trace from train 1L53 showing braking and AWS/TPWS activations and
acknowledgements.
182 Other than the fallen tree, the driver’s understanding was that adhesion elsewhere
in the area was unremarkable and, as such, he did not need to drive more
defensively (paragraph 53). There was, however, information available to him
that could have raised a concern about low adhesion. This information included
weather reports, incorporating the poor adhesion forecast, which were posted at
Salisbury depot where the driver booked on for duty (paragraph 96). However, the
driver of train 1L53 and other drivers told RAIB that they did not routinely check
or look at these forecasts because they felt they did not have time to do so and
that they found the weather notices too detailed and difficult to understand.24 They
therefore relied on their own experience and visual observations when driving to
understand the prevailing adhesion conditions. This was the case with the driver
of train 1L53 on the day of the accident and was his normal approach.
24
The weather notices were simplified to their current format around 2014. SWR driver diagrams generally allow
16 minutes booking-on time, of which 10 minutes is provided for reading notices and briefing material.
Report 12/2023 58 October 2023
Salisbury Tunnel Junction
183 In addition to the weather notices, another driver warned the driver of train 1L53
Analysis
about the poor adhesion conditions on the Down Main line approaching Salisbury.
It is also possible that a different driver warned him of the general poor weather
conditions on the route (paragraph 51). As the driver had never previously
experienced problems in that area and had not knowingly encountered any low
adhesion on the journeys to and from London until the approach to Salisbury
Tunnel Junction, these reports of low adhesion were not a factor in his decision to
delay his braking until beyond the fallen tree and associated debris.
184 There were other potential opportunities to raise the driver’s awareness of
low adhesion in the area which were not taken. If staff at the WICC receive
intelligence about low adhesion, such as from driver’s reports of low adhesion
(ROLA), WSTCF or MOM inspections, then control staff can arrange for signallers
and the WICC to warn drivers by sending messages over the GSM-R train radio
system.
185 Witness and OTDR evidence shows that several other trains had encountered
low adhesion on the Down Main line between signals SY29R and SY31 on
the afternoon and evening of 31 October, and the drivers of these trains had
observed signs of wheel slide occurring. However, none of the drivers made a
report of low adhesion to the signaller. There were also no WSTCF reports and no
MOM inspections.
Signal spacing
186 The signalling on the approach to Salisbury Tunnel Junction provides an
excessive braking distance for passenger trains between signals SY29R and
SY31 (figure 36). Excessive braking distances mean that drivers must delay full
braking beyond the double yellow aspect signal, otherwise they could stop short
of the red signal and/or incur time delays on their journey.
187 The applicable Rail Industry Standard in force at the time of the accident,
RIS- 0703-CCS, ‘Signalling Layout and Signal Aspect Sequence Requirements’
(issue 1.1, March 2018) states that it is good practice for signal spacing to provide
between 110% and 150% of the required braking distance. The spacing between
signals SY29R and SY31 provides approximately 190% of the minimum required
braking distance.
188 The signalling at Salisbury Tunnel Junction substantially pre-dates standard
NR/ L2/SIG/30009/D220 which states: ‘If the distance is significantly greater than
the minimum then this may lead to drivers continuing at the permissible speed
and trying to judge when to brake with the risk of SPAD, or braking early and
coasting leading to an increase in journey time or loss of capacity'. As such, the
over‑long spacing between cautionary and danger aspects on the Down Main line
enables drivers to decide how far beyond the cautionary aspect to start braking
and increases the possibility of situational awareness loss.
The driver’s actual braking point
189 The driver decided to use the fallen tree as his marker after which he would
start braking. This was located 250 metres beyond his normal braking point of
Broken Cross bridge. However, the driver did not then start braking until around
1,000 metres beyond this bridge (paragraph 175).
Report 12/2023 59 October 2023
Salisbury Tunnel Junction
Analysis
Normal Actual 1L43 OTDR speed
1L45 OTDR speed
braking braking 1L53 OTDR speed
location location AWS magnets
TPWS grids
Signals
Broken Cross Observed position Points of interest
bridge of tree debris Published low adhesion area
50 mph PSR Point of
50 mph PSR commencement board collision
120
warning board
SY29R SY29 SY31
100
80
OTDR speed (mph)
60
40
20
0
80.25 80.5 80.75 81 81.25 81.5 81.75 82 82.25 82.5
BAE1 mileage
560 m 1000 m 780 m 270 m 510 m 200 m
1560 m 1050 m
Figure 36: Annotated OTDR traces from trains 1L43, 1L45 and 1L53 showing important locations and
distances.
190 Witness evidence indicates that the driver did not brake until this point because
he had not recognised that he had passed the fallen tree. Although RAIB cannot
be certain from the available evidence how accurate the driver’s appreciation was
of the fallen tree’s location, he had specifically been looking for the fallen tree on
his journey up to London earlier in the day, having been previously told about it.
In addition, he was a very experienced driver who had been driving the route for
many years, and the fallen tree was in close proximity to significant structures,
signal SY29R followed by Broken Cross bridge. RAIB has therefore concluded that
the driver would have had a reasonable appreciation of the fallen tree’s location.
191 Although the available evidence cannot conclusively explain why the driver did not
recognise the location of the fallen tree, RAIB considers that the two most likely
explanations are that he did not see the tree (even though he was looking for it),
or that he lost awareness of the driving task. It cannot be known how conspicuous
the fallen tree was to the driver of train 1L53 in the dark, with the train travelling
at 86 mph (138 km/h). However, the tree remnants were close to the line and on
the driver’s side and are visible in FFCCTV images (figure 37), although RAIB
recognises that a driver’s actual view of objects may differ from that shown in
CCTV images.
Report 12/2023 60 October 2023
Salisbury Tunnel Junction
Visual cues
Analysis
192 It is possible that the driver, despite consciously looking for the fallen tree and
associated debris, did not see it because it was not sufficiently conspicuous in the
environment on that evening. RAIB examined and tested the unit involved and
has ruled out any impairment of the forward view, such as a fault with the train’s
windscreen, wash/wipe or headlights which may have contributed to the driver not
seeing the tree.
Figure 37: (Main photograph) Image taken from train 1L53 FFCCTV system showing the tree debris
as train 1L53 passed by. (Inset) Daylight image showing the condition of the tree and debris when train
1L52 passed by on the Up line to London Waterloo.
193 In May 2022, RAIB undertook a reconstruction in daylight and in darkness to
understand the visual cues and other environmental influences present when
driving a train towards Salisbury Tunnel Junction. This showed that the section
of line approaching and beyond Broken Cross bridge is largely dark at night, with
little artificial light available from streetlights or buildings. Consequently, there are
few visual cues for a driver regarding their position after they have passed Broken
Cross bridge until they start to approach signal SY29; it is a dark ‘tunnel’ created
by the vegetation and tree canopies, only illuminated by the train’s headlights.
Report 12/2023 61 October 2023
Salisbury Tunnel Junction
194 Signal SY29 is approached around a left-hand curve after a straight section of
Analysis
railway. Along this straight section, there is an increase in artificial lighting from
streetlights as the train approaches the outskirts of Salisbury, as well as a local
garage forecourt and a supermarket car park (figure 38). The light cast from the
signal aspect at signal SY29 can also be visible, reflecting on the rails, although
on the evening of the accident the reflection from the railhead may have been
limited due to the railhead contamination.
195 Witness evidence is that the driver of train 1L53 decided to start braking when
he believed he would be at risk of passing signal SY31 at red if he delayed
braking any further and that this was before signal SY29 came into view. The
location where he started braking coincided with the appearance of the visual
cues described above. RAIB has concluded that it was these cues that probably
prompted the driver to his location and caused him to apply the train’s brakes.
Figure 38: RAIB reconstruction showing the environment on approach to signal SY29: (left) 798 metres
from signal SY29; (centre-enhanced) reflection from lighting from signal SY29 at danger on the
railhead; and (right) 773 metres from signal SY29 with street lighting and car forecourt lighting visible.
Driver awareness
196 Given the proximity of the tree to the railway, the time that elapsed between the
train passing the fallen tree and the driver braking, and the limited use of train
controls during this period, it is also possible that the driver did not see the fallen
tree because he experienced a temporary loss of awareness of the driving task.
197 Evidence from the train’s OTDR shows that the driver acknowledged the AWS
horn for the permissible speed warning indicator which was located 290 metres
(8 seconds) beyond the fallen tree and 540 metres beyond Broken Cross bridge.
Around 12 seconds later, when the train was around 1000 metres from Broken
Cross bridge and travelling at 86 mph (138 km/h), the driver started applying the
train’s brakes.
198 Although the driver’s acknowledgement of the AWS horn may imply some
awareness of the driving task, research25 has shown that such responses can be
automatic with little conscious awareness.
199 As described in paragraph 195, the location where the driver started braking
coincided with the appearance of the artificial lighting from streetlights, a local
garage forecourt and a supermarket car park. In this scenario, it may have been
these visual cues which caused him to become aware of the driving task again,
and start to brake.
25
McLeod, RW, Walker, GH and Mills, A (2005), 'Assessing the human factors risks in extending the use of
AWS'. In J. Wilson, B. Norris and A. Mills (Eds.), Rail Human Factors: Supporting the Integrated Railway. London:
Routledge.
Report 12/2023 62 October 2023
Salisbury Tunnel Junction
200 As the driver was over 63 years old, he was subject to an annual medical
Analysis
examination in accordance with RIS-3451-TOM, ‘Train Drivers – Suitability and
Medical Fitness Requirements’ (issue 1, dated December 2016). This standard
requires that medical examinations are carried out by, or under the supervision of,
a registered medical practitioner. Medical examinations include:
• a general health assessment to identify any health conditions or medications
that could cause sudden loss of consciousness, a reduction in attention or
concentration, sudden incapacity or a loss of balance or co-ordination
• vision tests, including colour vision and diseases of the eye, and hearing
• blood or urine tests to detect diabetes mellitus and other conditions, and the
presence of drugs of abuse
• an electrocardiograph examination to check the heart’s rhythm and electrical
activity.
201 RAIB has found no evidence of distraction from electronic devices or the cab
environment. Post-accident toxicology did not identify the presence of any drugs
or alcohol. RAIB has also found no evidence from the driver’s roster or sleep
pattern, or any medical evidence of a sleep disorder, to suggest that he may have
been suffering from fatigue.
202 RAIB also found no evidence of any pre-existing medical condition or medical
treatment that could have caused a temporary loss of awareness or that might
have affected the driver’s perception, memory, or attention, although the presence
of an unknown medical condition of this nature cannot be entirely discounted.
The train’s braking systems
203 The braking systems of train 1L53 were unable to mitigate the effects of the
prevailing wheel/rail adhesion conditions.
204 RAIB’s investigation considered the capability of the train’s braking and sanding
systems to cope with the low adhesion conditions at the time of the accident.
205 Maintenance records for unit 159102 were up to date and no defects relevant
to the accident were present. The last scheduled test of the brakes and sanding
systems was conducted on 26 September 2021. On this occasion, a sand
discharge test was conducted. A container is placed beneath the sand discharge
nozzles, and a test button pressed for 30 seconds. The sand discharged into
the container is then weighed. The range of acceptable weights, as defined by
the vehicle maintenance instruction, is 0.6 kg to 1.0 kg. The values recorded
for the nozzles on 57803 (the carriage leading at the time of the accident, and
hence actively sanding) were 0.617 kg (left side nozzle in direction of travel) and
0.752 kg (right side nozzle). All other brake system air pressures were found to be
within acceptable limits.
206 Overnight from 29 October into 30 October, 159102 had a fuel point examination
at Salisbury depot. During that examination, the sandboxes were topped up.
Records show that 52803 had four bags of sand added, and 57803 had one bag
of sand added. This indicates that, before that examination, the sanding system
had been operative and discharging sand. SWR staff found no faults with the
sanding system during train preparation on the day of the accident.
Report 12/2023 63 October 2023
Salisbury Tunnel Junction
207 RAIB has therefore concluded that the maintenance and preparation of the train
Analysis
were unlikely to have had any influence on the accident.26
208 RAIB tested the sanding equipment fitted to carriage 57803, the leading carriage
of train 1L53 during the accident, as far as was possible following the accident,
although the extent of this testing was limited due to the damage caused by the
collision. Testing also took place following recovery activity, which may also have
affected the status of systems.
209 As part of this testing, it was established that the electrical controls of the sanding
system were fully operational, and that the system was not pneumatically isolated.
Clean, dry sand of the correct type was found to be present in the sandbox.
Accident damage meant that it was not possible to confirm the alignment of the
leading sanding nozzles with respect to the wheel/rail interface. It was also not
possible to eliminate the possibility of a blockage in the sand delivery pipework
that was dislodged during the collision and recovery. However, this pipework was
examined by RAIB and found to be unobstructed.
210 The operation of the pneumatic system which forms part of the sanding
equipment could also not be tested, as the unit’s engines could not be started.
However, given that the air system feeding the sanders is very simple, and was
apparently functioning correctly before the accident, it is considered unlikely by
RAIB that an undetected pneumatic system fault existed.
211 Railway Group Standard GMRT2045 ‘Compatibility Requirements for Braking
Systems of Rail Vehicles’ (issue 4, March 2016) states that the maximum
distance in which a three-carriage class 159 DMU should be able to stop in
normal adhesion conditions on level track from a speed of 85 mph (136 km/h) is
approximately 822 metres using full service (step 3) braking.
212 In February 2022, SWR carried out a series of brake tests using a three-carriage
class 159 unit running between Basingstoke and Salisbury on the Down Main line
approaching signal SY31. When adjusted for the gradient on the section of track,
the results of these tests showed that a train of the same type and configuration
as train 1L53 would be compliant to GM/RT2045.
213 On 22 December 2022, further tests were carried out by SWR using a
three- carriage class 159 unit on the Down Main line approaching signal SY31.
The weather conditions during these tests were overcast with light rain, and the
brakes were applied from similar speeds and locations as used by the driver of
train 1L53 on 31 October 2021. No railhead contamination was reported and
no WSP activity was recorded during the tests and the train was able to stop on
approach to signal SY31.
214 The driver of train 1L53 first applied the brakes at step 1, then almost immediately
step 2, when the train was 1,560 metres on approach to signal SY31. After
travelling another 210 metres, the driver increased the level of braking to step 3
(full service). The collision with train 1F30 occurred 1,760 metres from the point
where the driver had first applied the brakes (figure 36). This shows that the
braking distances achieved by the train during the accident exceeded both the
maximum range set out in standard GMRT2045 and those achieved during
SWR’s post-accident tests.
26
Post-accident wheelset checks for train 1L53 were completed and no contamination or damage was identified.
Report 12/2023 64 October 2023
Salisbury Tunnel Junction
215 Data from train 1L53 showed that the WSP system was active on all three
Analysis
carriages of the train from 18:42:04 hrs until the time of the collision at
18:42:56 hrs. Analysis of railhead samples taken after the accident found the
presence of silica (a component of sand) which showed that trains, likely including
train 1L53, had been discharging sand from on approach to signal SY29 through
to after signal SY31 (paragraph 132).
216 RAIB recovered the WSP equipment from train 1L53 (unit 159102). No defects or
abnormalities were found in the operation of the WSP components when tested
by RAIB. A detailed laboratory examination of this equipment was completed
using the WSPER (WSP Evaluation Rig) system. This showed that the WSP
system operated as expected for equipment of its age and type, and that its
performance was consistent with that recorded when it was originally approved
for use in 2006. Although the system was not fully compliant with the current
guidance27 (which is not applicable retrospectively), there is no evidence to
suggest that this non‑compliance affected the causes or the outcome of the
accident.
217 The tests undertaken by SWR during dry and light rain conditions demonstrated
that a train of the same type and configuration as train 1L53 would have been
capable of stopping within the braking distances set out in standard GMRT2045
under normal adhesion conditions. The braking distances actually achieved by
train 1L53 indicate that the WSP and sanding systems on the train were unable
to mitigate the very low wheel/rail adhesion encountered on 31 October 2021
(paragraphs 130 and 131).
218 The WSPER system was also used to approximate the braking performance of
six-carriage and nine-carriage class 159 formations. This analysis assumed that
the first three carriages would encounter similar adhesion levels to that derived
from train 1L53, but that subsequent carriages would encounter slightly higher
levels of adhesion. This would be due to the passage of previous wheelsets
modifying the available adhesion and the application of sand by the preceding
three-carriage vehicles in the train. A six-carriage class 159 unit would discharge
sand from the leading and fourth carriages; a nine-carriage train from the leading,
fourth and seventh carriages. Using WSPER, it was possible to store the data of
the modified adhesion conditions from each run and then apply this modified set
to the next test, which started from a slightly enhanced level of adhesion.
219 The results obtained from these tests suggest that neither a six-carriage nor a
nine-carriage formation braking from the same location as train 1L53 on the night
of the accident would have been able to stop before the collision point, assuming
they had encountered the same initial wheel/rail conditions.
27
GMGN2695 'Guidance on Testing of Wheel Slide Protection Systems Fitted on Rail Vehicles', issue 1, December
2010.
Report 12/2023 65 October 2023
Salisbury Tunnel Junction
220 The type of sanding system fitted to train 1L53 injects sand at a fixed rate
Analysis
under both wheels of the third wheelset from the front of the leading carriage.
When braking, this sanding system is controlled by the train’s WSP system
(paragraph 41). Research in the UK by RSSB has shown that enhanced sanding
systems are able to improve adhesion beyond that provided by a single fixed
rate sander, such as that fitted to train 1L53. Enhanced sanding systems include
distributed sanding, where more than one axle is equipped with sanders, and
variable rate sanding, where higher quantities of sand are delivered at higher
train speeds. Enhanced sanding can achieve a significant reduction of a train’s
stopping distance in reduced adhesion conditions, with the largest benefit being
achieved by variable rate sanders at two axles, known as double variable rate
sanding.
221 RSSB research illustrates the potential benefits of enhanced sanding systems, as
shown in figure 39. RAIB has undertaken a simple analysis, using the underlying
data from the RSSB research, to estimate the effect if train 1L53 had been fitted
with a DVRS system. The higher sand delivery rates of a DVRS system would
certainly have provided better low adhesion braking performance, which would
have led to a reduced collision speed. Although it is not possible to be certain of
the actual benefits of DVRS in the specific circumstances of this accident, the
analysis undertaken by RAIB suggests that it is probable that the collision would
have been avoided altogether. The analysis also suggests that it is possible that a
DVRS equipped train would not have passed signal SY31.
Figure 39: Diagram from RSSB research project T1107 showing the potential benefits of variable rate
sanding systems.
Report 12/2023 66 October 2023
Salisbury Tunnel Junction
Identification of underlying factors
Analysis
Management of low adhesion risk
222 Network Rail’s Wessex route did not effectively manage the risks of
low adhesion associated with the leaf fall season. This was a probable
underlying factor.
223 Documentary and witness evidence shows that before the accident Network
Rail’s Wessex route had not correctly identified sections of track as HRLA sites.
This was probably because the information (figure 40) that the SDS was provided
with and which they were using to identify such locations was inaccurate. These
inaccuracies arose because the process employed to inspect and risk assess
locations, and the subsequent dissemination of the information gathered, was
ineffective (paragraph 136).
224 Even if Network Rail’s Wessex route had identified the discrepancies with the
information, and the SDS had been able to appropriately identify the HRLA
sites, RAIB considers it unlikely that the route would have taken suitable action
to manage the associated risk. This was because there were several barriers to
effective low adhesion risk management present within the route. These included
issues relating to:
• resourcing (paragraph 225)
• liaison between departments (paragraph 231)
• track access constraints (paragraph 237)
• MOM inspections (paragraph 241)
• understanding of railhead treatment effectiveness (paragraph 243)
• weather response standards (paragraph 250)
• staff competences (paragraph 253).
Resourcing
Off track
225 Within Network Rail Southern region, off track asset management is a regional
role, supporting its Kent and Sussex routes, as well as the Wessex route.
However, at the time of the accident the role was vacant. Witnesses described
how the responsibilities for off track asset management within Wessex route
were being undertaken by the SAE, which led to a lack of strategic leadership in
relation to developing a vegetation management plan and co‑ordinating activities
between Wessex operations and maintenance off track staff.
Report 12/2023 67 October 2023
Salisbury Tunnel Junction
Analysis
LFR - 2 RCL - 2
Location 1 - 80 m 32.2 ch
LFR - 1 RCL - 1
Location 2 - 80 m 40 ch
LFR - 3 RCL - 2
Location 3 - 80 m 57.6 ch
LFR - 3 RCL - 2
Location 3.2 - 80 m 60.1 ch
SY29R
LFR - 3 RCL - 2
Location 4 - 80 m 70.2 ch
LFR - 4 RCL - 2
Location 5 - 81 m 0.7 ch
LFR - 3 RCL - 3
Location 6 - 81 m 21.4 ch
LFR - 2 RCL - 2
Location 7 - 81 m 39.8 ch
LFR - 2 RCL - 2 SY29
Location 8 - 81 m 64 ch
SY31 LFR - 2 RCL - 1
© Crown Copyright. All rights reserved.
Location 9 - 82 m 0 ch
Department for Transport 100039241. RAIB 2023
LFR - 2 RCL - 2
Location 10 - 82 m 0.5 ch
LFR - 2 RCL - 2
Location 11 - 82 m 15.8 ch
LFR - 4 RCL - 3
Location 12 - 82 m 28.7 ch 0 123 4 5 LFR - Leaf fall risk score
0 1 2 3 RCL - Railhead contamination level
Figure 40: Braking locations for train 1L53 with predicted leaf fall risk scores (0 to 5) from the leaf fall
assessment completed in June 2021 and corresponding physical railhead contamination level found on
site (0 to 3) on 31 October 2021.
Report 12/2023 68 October 2023
Salisbury Tunnel Junction
226 A level 2 audit was carried out on Network Rail’s Wessex outer route MDU
Analysis
between January and March 2021 which highlighted observations and
non- compliances with the requirements of NR/L2/OTK/5201 (paragraph 112).
As a result of the audit process, in February 2021, the SAE asked the off track
section manager within Wessex outer MDU to address the excessive vegetation
growth on the Up and Down Main lines at Salisbury. In June 2021, the off track
section manager reported that, although the COVID-19 pandemic had restricted
visual on-foot assessments of lineside vegetation, cab ride video assessment
of lineside vegetation would still be completed (paragraph 152). The section
manager reported that although vegetation growth was identified, this related to
work that had previously been identified and reprioritised.
227 Despite this report, MDU off track staff reported that its increasing workload
combined with a shortage of resources between May and August 2021 (figure 41)
meant that it could not address the existing work bank and that the MDU was
‘firefighting’ in response to arising incidents. Additionally, witness evidence is
that there was also a perceived problem with the availability of contractors.
Consequently, the MDU did not have the time or resources to prioritise and plan
the vegetation assessment and management work.
Work bank
2900
2700
2500
Number of tasks
2300
2100
1900
1700
1500
Mar-20
May-20
Jul-20
Mar-21
May-21
Jul-21
Nov-20
Nov-21
Sep-20
Sep-21
Jan-20
Jan-21
Jan-22
Figure 41: Ellipse work bank (number of tasks) levels (blue solid line) and mean average level (red
dotted line) between January 2020 and January 2022 (green triangle shows accident date).
Report 12/2023 69 October 2023
Salisbury Tunnel Junction
228 Before the accident, the off track section manager discussed the competence,
Analysis
workload and resource issues with the off track TME. The off track TME
subsequently wrote a paper to Network Rail’s Wessex route infrastructure
management team highlighting the significant increase in the MDU off
track section’s workload since 2020. This paper also noted that the lack of
competences held by permanent staff, such as working at heights, using rope
access and use of chainsaws, and the lack of resource was creating a risk of
non‑compliance with Network Rail’s standards, as highlighted by the 2021 audit.
Despite this, no action was taken by Wessex route to address the points raised
and the Ellipse data suggests that the off track section continued to see high
workloads until around August 2021.
Seasons delivery specialist (SDS)
229 During 2020 and 2021, the SDS realised they were struggling with workload and
that management of the autumn process could not be undertaken by one person
alone. This problem had also been identified by one of their predecessors in
2017. This predecessor prepared a proposal to increase the number of seasons
delivery staff to mitigate the risks of relying on a single, relatively junior member of
staff. Neither Network Rail’s Wessex route nor South Western Railway operations
departments formally responded to the proposal, although a lack of budget for the
new roles was stated in a subsequent conversation as a reason for not pursuing
the new roles.
230 In July 2021, the SDS was successful in obtaining the secondment of extra staff
into the WICC to operate an autumn control desk (paragraph 110). Four staff
were seconded to work early and late shifts from September to December 2021.
However, due to the number of incidents that occurred on the morning of 31
October, the members of staff on the autumn control desk were diverted from their
responsibilities to help the WICC manage the incidents.
Liaison between departments
231 The management of autumn preparedness requires planning and co‑ordination
between Network Rail’s operations and off track functions (figure 42). As outlined
below, witness and documentary evidence show there was no apparent strategy
to promote collaboration between these departments to manage leaf fall risk. This
led to what witnesses described as a ‘silo’ culture within Network Rail’s Wessex
route, with very little co‑ordination between, or information shared across, these
functions. This was an issue which was also identified by the Office of Rail and
Road (ORR) in 2020 (see paragraphs 299 to 305).
232 Network Rail has two principal standards relating to autumn preparedness.
These deal with vegetation management (NR/L2/OTK/5201) and operations
(NR/ L2/ OPS/095). The operations standard was amended in June 2019 to align it
with the maintenance activities in the vegetation management standard. However,
despite this, the standards do not promote liaison between the two functions
or mandate the sharing of information needed to support development and
implementation of the autumn working arrangements.
Report 12/2023 70 October 2023
Salisbury Tunnel Junction
Analysis
Outer Wessex Inner
Senior asset Seasonal delivery
Operations Off track Track
engineer (SAE) specialist (SDS)
Salisbury
WICC
(Eastleigh MDU)
Intelligent • Lubrication
Local operations Autumn Track
infrastructure Signaller • Ultrasonics
manager (LOM) controllers (x3) maintenance Track
tecnicians (IIT) • Drainage
engineer (TME) • Lineside vegetation
maintenance
(off track) • Boundary fencing
Mobile operations Poor track circuit Train driver Section 4x Track
manager (MOM) ‘shunt’ ROLA manager (SM) maintenance
(off track) engineers (TME)
Off track
Sub-contractor
supervisor
• MPV fleet scheduling
TGA maintenance • Missed sites 2x Off track
and low adhesion • Weather information inspectors
site inspection • Incident management
with incident controllers
Track team: Track team:
• Team leader • Team leader
• Operative • Operative
• Technician • Technician
Figure 42: Network Rail’s Wessex route departments involved with the co-ordination, planning,
maintenance and response to autumn.
233 Witness and documentary evidence show that very little co‑ordination took
place between the SDS and the relevant MDU off track section managers in
the preparation of the autumn working arrangements document for 2021. If the
collaboration had been effective, it might have become apparent to the SDS that
MDU off track staff had not implemented any actions to mitigate the leaf fall risk
and had repeatedly deferred other vegetation management actions since 2018.
234 Although the MDU off track had video footage of the vegetation growth since
2019, this had not been shared with the SDS. This meant that the SDS had little
awareness of the nature and status of the risk or risk mitigation measures relevant
to the vegetation on the Up and Down Main lines and other lines within Wessex
route. Likewise, the MDU off track staff had little appreciation of the importance of
the intelligence they had gathered about the condition of the vegetation and trees
to the leaf fall risk being managed by the SDS. Furthermore, the MDU off track
staff were not using information that was available from the SDS to help them
appropriately assess and prioritise the work that was required to manage the risk
from lineside trees and vegetation (paragraph 151).
235 Although the audit of the MDU undertaken in 2021 highlighted non‑compliances
with Network Rail standards, the report did not recognise any issues of concern
with the relationship between the SDS and the MDU off track staff, as this liaison
is not required by the standard (paragraph 112).
Report 12/2023 71 October 2023
Salisbury Tunnel Junction
236 Witness evidence also shows that the lack of technical training, competence
Analysis
management and non-technical skills training associated with the role of the SDS
was a factor in the SDS not appreciating the importance of engaging with staff in
other departments (see paragraph 253).
Track access
237 From June 2021, after the fatal track worker accidents at Margam (RAIB report
11/2020) and Surbiton (RAIB report 05/2022), Network Rail implemented safety
measures to restrict some forms of track access. The track access constraints
affected both the maintenance of vegetation and the assessment of HRLA sites.
Due to the topography of the Up and Down Main lines at Salisbury and locations
where limited clearance increased the risk to track workers, both MDU off track
staff and MOMs were required to obtain line blockages to inspect and maintain
the line.
238 When the SAE asked the off track section manager to address the excessive
vegetation growth on the Up and Down Main lines at Salisbury in February 2021
(paragraph 226), the off track section manager reported that the complexity of
the lines made track access difficult, so it was not possible to schedule such a
maintenance task to inspect or cut back the vegetation.
239 The Wessex route autumn working arrangements document required a MOM to
undertake regular inspections28 of HRLA sites for railhead contamination and to
submit inspection forms and images for review by the autumn controller or the
SDS. The objective of these regular inspections is to proactively monitor levels of
contamination at the sites and, if significant contamination is identified, to arrange
for additional treatment to be undertaken. MOMs are also required to respond to
reports of WSTCF activations and low adhesion.
240 However, direct access to the track to allow these MOM inspections of HRLA sites
was restricted because of the limited availability of line blockages. This meant
that MOMs were required to make railhead observations to identify contamination
from vantage points such as fences and bridges. In July 2021, MOMs within
Network Rail’s Wessex route challenged the decision with their line managers
about the adequacy of the process of carrying out inspections from a vantage
point, because this was seen to be ineffective due to the length of track to be
inspected and limited sighting that was available (figure 43). However, MOMs
were instructed to continue with the current process of inspection. RAIB is
unaware of any action taken in response to these challenges.
MOM inspections
241 As well as restrictions on track access, the process of using MOMs to inspect
for railhead contamination within Network Rail’s Wessex route appears to have
been ad hoc. This was because there was no guarantee that a MOM would be
available at any given time, due to the requirement for them to respond to other
operating incidents. This was exacerbated by the lack of MOM resources within
the Wessex route.
28
Witness evidence indicates that the expectation was that MOM inspections of HRLA sites would be undertaken
on a daily basis by the early or late rostered MOMs.
Report 12/2023 72 October 2023
Salisbury Tunnel Junction
Analysis
Figure 43: Image taken from a vantage point by a Network Rail MOM inspecting an HRLA site
(post- vegetation clearance).
242 Witness and electronic evidence shows that, before the accident, HRLA
inspections by MOMs were only being sporadically undertaken for the reasons
given above. Furthermore, witness and documentary evidence shows that the
inspection forms and images that were being submitted were not being reviewed
in accordance with the Wessex route autumn working arrangements document.
The SDS was unable to explain why the review process was not being followed.
Understanding of railhead treatment effectiveness
243 Network Rail’s Wessex route adopted 60 mph (96 km/h) as the standard speed
for its railhead treatment vehicles. Witness evidence shows that this decision on
treatment speed was due to the number of route miles needing to be covered, the
resources required to staff the MPV and the water refilling opportunities.
244 In 2006, RSSB and the rail industry AWG sponsored research, including testing,
into low railhead adhesion, its prevention, and the mitigation of associated risks
(paragraph 83). This was after the incidents at Lewes and Esher in 2005 (RAIB
report 25/2006). The research included consideration of the factors affecting
build-up of contamination, and testing the effectiveness of railhead treatment.
245 The focus of the 2006 railhead treatment testing was twofold. It investigated
whether the speed of the railhead treatment could be increased to integrate
the service more effectively into the passenger service timetables and reduce
delays. It also investigated the effects of increasing the pressure of water
jetting in removing contamination. The research concluded that it was justifiable
to increase the speed of railhead treatment services from 40 mph (64 km/h)
to 60 mph (96 km/h) with a recommendation for the water pressure to be
increased from 1000 to 1500 bar. At the time of the testing, it was noted that
treatment undertaken at 60 mph (96 km/h) would remove around of 85% of the
contamination.
Report 12/2023 73 October 2023
Salisbury Tunnel Junction
246 Despite the testing and research, the relationship between environmental
Analysis
factors, track usage, and the rate of build-up of contamination affecting railhead
adhesion is not well understood. Furthermore, RAIB found no evidence that the
understanding of the longevity of the benefits from a given railhead treatment
regime was clearly understood. Limited further research has been undertaken
on this subject since the 2006 testing. Witness evidence suggests that the rail
industry’s understanding of the limitations and benefits of such regimes has not
progressed significantly since that date.
247 This means that there is uncertainty associated with predicting where very low
adhesion will occur. This uncertainty makes the sharing of information between
train operators and Network Rail critical to managing the risks associated with low
adhesion, especially during the autumn leaf fall season.
248 The lack of clear guidance on how long a railhead treatment remains effective
has led to standard treatment patterns being used. Network Rail’s Wessex route
applies a regime of two railhead treatment runs each weekday, and one per
day at weekends. However, there appears to be little scientific justification for
this pattern. The SDS did not consult others, such as Network Rail’s Technical
Authority, when determining it and witness evidence shows there was a lack of
understanding within Wessex route as to whether the aspiration at weekends was
for one run each calendar day or a maximum interval between treatment runs of
24 hours (paragraph 102).
249 Witness and documentary evidence also shows the planned treatment at
weekends was frequently delayed or cancelled in Network Rail’s Wessex route
without any assessment of the risk, or alternative mitigations being implemented.
This was the case for the MPV treatment on 31 October, despite the poor weather
forecast for the weekend being received on 29 October. This unmitigated delay
had become the accepted practice and provides further evidence of a lack of
understanding of how effective railhead treatment was at controlling low adhesion
risk.
Weather response standards
250 Network Rail standards NR/L3/OPS/045/3.1T7, ‘Weather Arrangements’ (issue 3,
June 2020) and NR/L3/OPS/021/01 outline the actions to be taken following
receipt of a weather warning. The forecast for 31 October correctly predicted a
high risk of leaf fall and a ‘red’ (poor adhesion) day (paragraph 168). However,
neither of these two standards requires, or provides guidance for route control
or the SDS to assess the possible impact of the weather forecasts on the risk of
low railhead adhesion. In essence, there is nothing to drive consideration of any
necessary additional control measures for low adhesion in Network Rail weather
response standards.
251 This meant that, even though the weather forecast on 29 October highlighted
an ‘adverse’ weather warning and a red day for low adhesion, no additional risk
assessment was undertaken. If such an assessment had been undertaken, it
might have highlighted the significance of the delayed MPV run and additional
mitigation might have been considered (paragraph 160).
Report 12/2023 74 October 2023
Salisbury Tunnel Junction
252 The weather forecast for 31 October 2021 did not trigger an EWAT
Analysis
(paragraph 94). The process for instigating an EWAT does not consider how
the weather conditions during the day could affect the risk of low adhesion.
This meant that there was no recognition by the WICC of the need to consider
the consequences of the delayed MPV railhead treatment run or to consider
additional mitigation, such as a GSM-R message to warn train drivers (see
paragraph 326).
Competence management
Seasons delivery specialist (SDS)
253 In order to effectively implement the requirements of Network Rail standards
such as NR/L2/OPS/095, staff undertaking the role of SDS require an appropriate
level of training and competence. However, RAIB found that there was no
formal competence development framework for the SDS role, that training was
effectively on-the-job in nature and that the Wessex route SDS in post at the time
of the accident was primarily relying on the working arrangements set up by their
predecessor.
254 The Wessex route SDS in post at the time of the accident entered the role in
early 2020, having joined Network Rail on the graduate entry scheme in 2019
(paragraph 25). On taking up the position, the SDS completed a one-day
e‑learning course and was provided with a short period of mentoring support from
a seasonal specialist in Network Rail’s Technical Authority.
255 Witness evidence shows that this lack of a formal framework apparently led to
gaps in the breadth and depth in the professional knowledge of this and other
SDSs. For example, the Wessex route SDS at the time of the accident was
confused as to which risk assessment model should be used to drive the action
plan for autumn preparedness. RAIB also found that none of the Wessex route
SDSs in post from 2014 to 2021 were aware of Network Rail’s signal overrun
risk assessment process, possibly affecting their appreciation of how low
adhesion affects the risk of SPADs. The lack of formal training in technical and
non-technical skills also meant that the SDS at the time of the accident neither
appreciated the importance of, nor had the confidence to engage in, co‑ordination
activities with other departments (paragraph 231).
256 Witness evidence indicates that the role of the SDS has primarily been seen
as a short-term secondment filled by a graduate, as a ‘stepping-stone’ to gain
experience over a wide range of disciplines. This may explain the lack of a formal
training framework and suggests that the importance of the SDS role was not fully
recognised within Network Rail.
257 An internal audit of Network Rail’s Western route in September 2019
(paragraph 113) identified the need to improve the competency and training
requirements for the role of SDS. In particular, the audit report noted that the
e-learning did not capture the wide range of knowledge in seasonal risks,
hazards and mitigations required for the role. Although witnesses stated that
this report would have been shared with Network Rail’s Wessex route, neither
the SDS nor their manager was aware of it. This meant that none of the report’s
recommendations were considered or applied in Network Rail’s Wessex route.
Report 12/2023 75 October 2023
Salisbury Tunnel Junction
258 Witness evidence is that the title of ‘specialist’ also affected how other members
Analysis
of staff interacted with the SDS. In some circumstances, staff were not willing to
challenge the SDS as they believed them to be the expert in that field. However,
the SDS did not regard themself as an expert in these terms.
Off track staff
259 Network Rail’s vegetation management standard NR/L2/OTK/5201
(paragraph 136) requires that relevant assessment staff have the correct skills
and knowledge to understand what they are inspecting. During 2018 and 2019
Network Rail updated this standard. These revisions and the associated training
(paragraph 141) were supposed to help facilitate MDU off track staff to undertake
the leaf fall risk assessments.
260 However, witness evidence suggests that off track staff did not feel that they had
been given sufficient training in leaf fall risk assessment and that they did not
fully understand when or how to apply the revised standard. Training consisted
of a one-day e-learning course, covering leaf fall assessment and how to identify
levels of railhead contamination. Furthermore, staff did not have the underlying
arboriculture knowledge to undertake the assessments or to understand the
consequences of leaf fall on low adhesion risk. Consequently, the MDU off track
section manager refused to allow their staff to undertake leaf fall assessments
until adequate training had been provided.
261 The internal level 2 audit of the Wessex outer MDU in 2021 (paragraph 112)
noted a good practice item associated with the off track section manager’s
understanding of their team’s competence, as well as the need for new staff to
shadow others and to be mentored. Although the auditor identified the lack of tree
and leaf fall management plans, they classed these only as observations. As such
these were not tracked and no action was taken before the accident to resolve
them. Furthermore, the auditor did not identify that the lack of tree and leaf fall
strategies stemmed from the gap in the competence of staff to undertake leaf fall
risk assessments.
262 In 2020, ORR identified deficiencies in the training and competence of off track
staff undertaking this type of work (see paragraph 305). ORR’s recommended
actions had not been carried out by the time of the accident on 31 October 2021.
Despite this, witness evidence indicates that Network Rail had nevertheless
identified the issue and was developing a training and competency framework for
off track staff at that time.
Report 12/2023 76 October 2023
Salisbury Tunnel Junction
South Western Railway’s arrangements for dealing with low adhesion
Analysis
263 South Western Railway had not effectively prepared its drivers for assessing
and reporting low adhesion conditions. This is a possible underlying factor.
264 SWR Operations Manual (AP21, issue 1) includes an outline of the
implementation of arrangements to minimise the risk to safe operation of trains
during the autumn leaf fall period. Specifically SWR’s professional driving policy
(PDP)29 Section 15 ‘Autumn’ requires drivers to ‘drive defensively according
to the prevailing conditions, braking capabilities of the traction driven, utilising
WSP sanding and being proactive to the available advice regarding adhesion
information/forecasts’. The policy also requires drivers to be aware how stock
formation can impact on stopping distances and advises that ‘shorter formations
have a higher risk of sliding so often requires a greater stopping distance’.
265 In addition, the PDP also requires that drivers ‘must also be aware of identified
low rail adhesion locations on the routes you sign, adjusting the driving technique
accordingly. These locations are listed in the SWR Route Maps and identified by
trackside signage’ (although the signage was missing, paragraph 158).
266 The SWR Autumn train driver brief (2021) provided guidance, as per the Rule
Book (see paragraph 272), on the levels of rail adhesion and their classification
as ‘good', 'expected' and 'reportable’. The brief required a driver encountering
low adhesion to inform the signaller immediately, provided the conditions were
worse than expected for the location and environment. Witness and OTDR
evidence shows that several train drivers during the day had encountered varying
degrees of low wheel/rail adhesion. However, none of these drivers had formally
reported the presence of low adhesion to the signaller and RAIB has no evidence
to confirm whether the level of adhesion should have been considered as
‘reportable’.
267 It is also possible that an opportunity was missed earlier in the day to identify
railhead contamination as the cause of low adhesion in the area, when train 1L13
collided with the tree just beyond Broken Cross bridge (paragraph 47). During
their conversation reporting the incident to the signaller, the driver stated that
the train’s braking was affected by “slipping”, but they did not identify railhead
contamination as a factor. The signaller did not enquire further as to whether the
driver considered the low adhesion conditions to be a cause of the ‘slipping’ or
if the low adhesion should be classified as reportable, and no further action was
taken.
268 The knowledge required by drivers to identify and deal with low adhesion should
be obtained through training and briefings based on SWR’s professional driving
policy and autumn briefing, which is given to all drivers. New, post qualified
and drivers who had been involved in a low adhesion incident in the previous
12 months are required to have training in addition to the annual autumn briefings.
This requirement did not apply to experienced drivers. In line with the policy, all
SWR train drivers are expected to make themselves aware of weather forecasts
and adhesion conditions through reading notices placed at the booking‑on points
(paragraph 96). The autumn briefing is produced each year by SWR and is based
on its own autumn strategy and Network Rail’s Wessex route autumn working
arrangements document.
29
South Western Railway Professional driving Policy version 1.0 issued August 2021 and Autumn brief for drivers
and guards (reference SB003-2021) issued 08/10/21.
Report 12/2023 77 October 2023
Salisbury Tunnel Junction
269 The 2021 autumn briefing included guidance on reporting low adhesion, levels of
Analysis
contamination and details of known HRLA sites. The briefing generally focused
on reactive measures for drivers experiencing low adhesion and only provided
limited guidance on proactively detecting it. The briefing also contained less
information than in previous years, and some topics such as carrying out frequent
running brake tests to test railhead adhesion levels were omitted altogether.
270 Data provided by Network Rail shows that, from 2018 to November 2021, there
had been an ongoing reduction in the number of reports of low adhesion from
drivers to signallers in Network Rail’s Wessex route.
271 From 2018 to 2020, SWR autumn briefings took place face-to-face in groups.
However, in early 2021 SWR Driver Operations decided to make use of the
available technology and introduce an electronic briefing process. As such,
in September 2021, the autumn briefing was published on the company’s
intranet, with notices posted at drivers’ booking-on points that the briefing was
available; drivers could request a hard copy. This meant there was no immediate
opportunity for drivers to ask questions30 or for SWR managers to emphasise
critical learning. Train drivers could still seek clarification from their managers by
email or when the opportunity arose.
272 Section 28 of Rule Book Module TW1 requires that drivers report certain levels
of low railhead adhesion to the signaller. Before 2018, low adhesion had been
classified as either ‘low’ or ‘exceptionally poor’, depending on where a driver
experienced difficulties in stopping or starting. Both ‘low’ and ‘exceptionally poor’
adhesion were required to be reported to the signaller. This classification was
changed in 2018 to three categories, with drivers only required to report the
‘reportable’ conditions (paragraph 108). This was defined as ‘Railhead adhesion
is worse than would be expected for the location and environmental conditions’.
While both sets of requirements need a subjective judgement to be made, the
newer requirements are more limiting in terms of what would be considered
reportable.
273 Evidence suggests that this change created confusion among many drivers over
the correct terminology and reporting criteria. SWR driver managers identified
this confusion in 2020 and introduced a reminder to drivers that ‘Track Is Good,
Expected or Reportable’ (TIGER) within the autumn training programme. It is
nevertheless possible that this confusion may have persisted and contributed to
drivers not reporting conditions on the day to the signaller, particularly with the
need for subjective judgement.
274 During autumn 2021, SWR’s driver managers installed flip charts at booking- on
points as an informal means for drivers to communicate to each other any
adhesion issues during their journeys (figure 44). Witness evidence shows that
some drivers did not consult the flip charts, either because they did not book on
for duty at the booking-on point or because they believed they were out of date by
the time they read them. Some drivers were under the impression that reporting
low adhesion on the flip chart meant that they were not required to formally report
it to the signaller.
30
If the briefing had been face-to-face, drivers may also have reported that there was no on-track signage to
identify the HRLA sites shown within the autumn briefing.
Report 12/2023 78 October 2023
Salisbury Tunnel Junction
275 RAIB has concluded that the general reduction in reporting of low adhesion
Analysis
was likely to have been a result of a combination of SWR’s briefings to drivers,
changes in and confusion over reporting requirements and the introduction of
the flip charts. There may also have been some effect due to the changes in
rail services seen in Autumn 2020 due to the pandemic. Similarly, if drivers had
experienced ‘reportable’ levels of adhesion on the day of the accident, some
of these factors may have contributed to them not making any formal ROLAs
to the signaller. If there had been a ROLA that day, the signaller would have
been required to broadcast GSM-R messages to all drivers warning them of the
conditions, and this might have altered the risk perception of the driver of train
1L53 and caused him to start braking earlier on approach to signal SY31.
Figure 44: Flip chart (dated 28 November 2021) used by drivers at Salisbury depot to report low
adhesion conditions and other safety issues.
Observations
Network Rail TPWS design criteria
276 The TPWS installation at signal SY31 was not compliant with Network Rail
requirements for new signalling design.
277 Design of a TPWS installation was governed by Network Rail standard
RT/ E/S/10138, ‘Train Protection and Warning System (TPWS) Transmitter Loop
Requirements and Positioning’ (issue 3, April 2004). To be effective in controlling
the risk of an overrunning train, the designer of a TPWS installation must consider
the likely speed of an approaching train. The above standard required that, where
there was a reduction in the permissible speed on the approach to the signal
being assessed it had to have occurred more than 450 metres on the approach
to the TPWS installation for it to be considered in the design. However, technical
instruction, TI022 ‘Provision of TPWS at signals’ (issue 4, April 2019) extended
this distance to 800 metres. Incorporating a higher approach speed in the TPWS
design calculations improves the effectiveness of TPWS in preventing collisions
involving trains travelling at the higher speed. However, Network Rail did not
make it necessary to apply TI022 retrospectively to existing TPWS installations
such as that at Salisbury Tunnel Junction, where the 90 to 50 mph (145 to
80 km/h) permissible speed reduction occurred approximately 500 metres on
approach to signal SY31 (figure 5).
Report 12/2023 79 October 2023
Salisbury Tunnel Junction
278 To retrospectively apply TI022 to the TPWS installation at Salisbury Tunnel
Analysis
Junction would likely require changes to signals SY31, SY29 and SY29R. This
might include changes to the signal location, the controls applied to the signals, or
a combination of both, so it is not certain what effect adopting the higher approach
speed would have on the TPWS design at signal SY31, or on the circumstances
of this accident. However, it is probable that the required changes to the signalling
would have also reduced the unevenness of signal spacing between signals
SY29R, SY29 and SY31 (paragraph 186).
Network Rail’s overrun risk assessments
279 Network Rail’s processes for standard signal overrun risk assessment do
not require HRLA sites to be accounted for, nor does the identification of a
new HRLA site trigger a reassessment of signal overrun risk.
280 Network Rail quantifies the risk of a train passing a stop signal in accordance
with standard NR/L2/SIG/14201/Mod04, ‘Signalling Risk Assessment Handbook:
Prevention and Mitigation of Overruns – Signal Overrun Risk Assessment Tool
Specification’ (issue 3, December 2020). This requires a ‘standard’ assessment
to be undertaken using the Signal Overrun Risk Assessment Tool (SORAT).
SORAT is a software tool which takes account of the railway layout and timetable
to quantitatively assess hazards and potential conflicts. It produces a risk score
for each assessed signal on the rising scale from M4 (lowest risk) to A1 (highest
risk). Signal SY31 was assessed on 12 May 2018 and returned a risk score of ‘I2’.
281 The standard SORAT assessment assumes a good level of adhesion when
calculating overrun risk, using a generic multiplier to quantify the additional
risk presented by autumn adhesion conditions. This multiplier is applied to all
assessed signals and no account is taken of site-specific contamination or low
adhesion issues such as the presence of HRLA sites.
282 The signalling risk assessment handbook requires all signals to be reassessed
using SORAT every five years to capture incremental changes. A SORAT
assessment can also be triggered by changes including altered permissible
speeds, multiple signals passed at danger incidents, accidents, timetable
changes or a new or modified track layout. When the accident occurred, signal
SY31 had not yet reached the five-year review following its last assessment, nor
had any infrastructure change occurred since then which would have required
reassessment.
283 Where a standard SORAT assessment determines the signal to have a risk
score that is above a set threshold, the process requires that signal to be further
assessed in a workshop environment. This process, known as variSPAD, allows
the workshop to consider a wider range of hazards and potential mitigations.
Those present are prompted by a list of questions/prompts posed by the
assessment proforma.
Report 12/2023 80 October 2023
Salisbury Tunnel Junction
284 Hazards assessed in a variSPAD include whether the signal approach includes
Analysis
irregular signal spacing, significantly greater than necessary braking distance,
falling gradients and known areas of low adhesion. All these issues were present
on the approach to Salisbury Tunnel Junction. However, because the risk score
given to signal SY31 did not reach the required threshold, it was not subject to
a variSPAD workshop. This means that, even if the potential presence of an
HRLA site had been identified it would not have formed part of the overrun risk
assessment for signal SY31.
Factors affecting the severity of consequences
The driver’s cab of train 1L53
285 There was a loss of survival space in the driver’s cab of train 1L53.
286 The initial impact between the two trains was between the front left corner of
the driver’s cab of train 1L53 and a point around the leading right-hand side
passenger doors of the fourth carriage of train 1F30 (paragraph 69). The driver’s
cab of train 1L53 suffered a loss of survival space31 that was likely to have led
to serious or fatal injuries to the driver, had he not vacated his seat just before
impact (paragraph 68). Examination of the surviving parts of the cab structure
from the leading carriage (57802) of train 1L53 showed that the front left cab
corner pillar became overloaded and failed (figure 45). This pillar would have
carried much of the force resulting from the collision.
Figure 45: Cab of train 1L53 (additional damage was caused by the fire and rescue service during the
extraction of the injured train driver).
31
The volume of the carriage body containing the occupants that is to be maintained during a collision to protect
the occupants and limit the likelihood of injuries.
Report 12/2023 81 October 2023
Salisbury Tunnel Junction
287 The design of the class 158 DMU structure (from which the class 159 involved
Analysis
in the accident was derived) was based on UIC 566 ‘Loading of coach bodies
and their components’ which was applicable at the time of construction. UIC 566
defined a range of proof loading requirements to apply on the structure when
considering its structural integrity. This included longitudinal loads applied to
the cab structure at various heights from floor to cantrail, where the bodyside
meets the roof, which had to be sustained without permanent deformation. A
review of the corresponding requirements contained in the current standard,
BS EN 12663- 1:2010 ‘Structural requirements of railway vehicle bodies’, shows
that the magnitude of the longitudinal loads applied to the cab structure are the
same. This suggests that the proof strength of the cab structure of a modern cab
would be similar to the proof strength of a class 158 cab structure.
288 Unlike modern standards, such as BS EN 15227 ‘Crashworthiness requirements
for rail vehicles’, UIC 566 did not define specific collision scenarios to assess the
crashworthiness performance of the carriage in the event of an accident, that is,
what happens to the body structure when subjected to loads greater than its proof
strength. As such, the class 158 body structure was not designed to meet any
specific crashworthiness requirements.
289 Nevertheless, as the class 158 body structure was an early application of
welded aluminium to UK train construction, a considerable amount of work was
undertaken at that time to understand its crashworthiness performance. Part
of this work included assessing its crashworthiness performance against an
emerging internal British Railways Board standard, CP-DDE-116 ‘Structural
requirements for the bodies of multiple unit vehicles’, which contained collision
scenarios. A British Rail report published in 1989 (TR-VST-004) concluded that,
when assessed against CP-DDE-116 requirements, ‘the class 158 is not a good
crashworthy structure: high force levels and low energy absorption at low impact
velocities are evident with deceleration levels approximately twice the suggested
values’. In 1991, the class 158 cab structure was crush tested in a controlled
environment. British Rail report RR-VST-002 concluded that ‘the energy absorbed
by the cab ends in the tests was sufficiently high to meet British Rail’s latest
standard but the peak forces generated during the test were much higher than in
comparable steel vehicles and the mode of deformation rather aggressive’.
290 As the proof strength of a class 158 cab structure is likely to be similar to a
modern equivalent, RAIB has not made a specific recommendation relating to the
driver’s cab of the class 158. The more general question on the crashworthiness
performance of traction and rolling stock which predates modern crashworthiness
requirements was identified in recommendation 19 of the RAIB investigation into
the accident at Carmont, Scotland (see paragraph 326).
Internal sliding vestibule doors
291 Damage to the internal sliding doors on the train obstructed evacuation
routes and prevented train crew from accessing carriages to assist
passengers.
292 Class 158 and class 159 trains have their exterior doors located at each end
of the body. The area adjacent to these doors, commonly referred to as the
vestibule, is separated from the main saloon area by ‘bi-parting’ sliding doors. The
purpose of these doors is to improve passenger comfort, by isolating the saloon
area from draughts when the exterior doors are opened.
Report 12/2023 82 October 2023
Salisbury Tunnel Junction
293 Each saloon to vestibule door has two door leaves (figures 46 and 47). These
Analysis
meet in the middle of the doorway and are power operated such that each leaf
opens to its respective side. The vestibule doors are electrically operated, and
each leaf moves on runners located above the door aperture.
Figure 46: Image showing jammed vestibule door of the
leading carriage of train 1L53.
Figure 47: Internal CCTV image showing passengers trying to prise vestibule door open in the rear
carriage of train 1L53.
Report 12/2023 83 October 2023
Salisbury Tunnel Junction
294 Analysis of CCTV footage recovered from the carriages and interviews with
Analysis
witnesses identified that, in the immediate aftermath of the accident, at least three
of these doors were unavailable to passengers. These were:
• The leading vestibule door of carriage 57803, the leading carriage of train
1L53.32 This area of the train, adjacent to the cab, suffered severe damage
during the collision. The damage to this door did not obstruct a viable egress
route. The significant damage to the leading end of carriage 57803 meant that
the exterior doors at that end of the carriage were unusable. In the immediate
aftermath of the accident, the trailing vestibule door of 57803 was operable,
although this door was subsequently found in a jammed condition (see
paragraph 296).
• The trailing door of carriage 52803, the rearmost carriage of train 1L53. CCTV
images show passengers forcing this door open and that this clearly required
considerable physical effort. If these passengers had been unable to open
this door, then the guard of the train would have been unable to move into the
train from the vestibule/rear cab area, where they were immediately before the
accident, to offer assistance to passengers. This would have also obstructed a
viable egress route.
• The leading vestibule door of carriage 52763, the fourth and last carriage of
train 1F30. This prevented the guard of this train (who was in the third carriage
of train 1F30 at the time of the accident) gaining access to this carriage in the
aftermath of the collision. It also meant that passengers who were in the third
carriage of this train had only one viable egress route available to them, which
was through the cab-end door of the carriage and thence via Fisherton Tunnel
to Salisbury station.
295 Although no injuries occurred because of these doors becoming jammed, the
carriages were in a tunnel and some diesel fuel had been spilled. Blockage of
otherwise viable egress routes is therefore evidently undesirable and may result
in passengers panicking and/or injuring themselves when seeking alternative
routes as part of an uncontrolled evacuation.
296 Examination of the carriages after recovery found two further jammed vestibule
sliding doors. They were the trailing door of carriage 57803, leading carriage of
train 1F30, and the trailing door of carriage 57763, the third carriage of train 1F30.
There is no evidence to indicate when these doors became jammed, either during
the accident or afterwards in the recovery of the carriages.
297 Railway Industry Standard RIS-2730-RST33 sets out requirements for railway
vehicle fire safety and evacuation. This standard states that internal sliding doors
shall:
a) Slide open in opposing directions at each end of the passenger saloon;
or
b) Be bi-parting; or
32
The leading vestibules on Class 159 trains are normally inaccessible to passengers. However, a release is
available to allow access in an emergency. The release for the leading vestibule door of 57803 had not been
operated.
33
RIS-2730-RST issue 1.1 dated September 2021 is the current issue, although this standard and predecessors
have existed for many years, with requirements being substantially unchanged.
Report 12/2023 84 October 2023
Salisbury Tunnel Junction
c) Be fitted with a means of escape within the door opening which allows
Analysis
through egress in the event of the door becoming jammed, so as to
comply with clause 4.3.2.1 (e) of BS EN 45545-4:2013.
298 The bi-parting doors fitted to class 158 and class 159 trains comply with clause ‘b’
of this standard. It is evident that the doors became jammed due to them derailing
from their runners. However, RAIB has not investigated the specific mechanism/s
that caused them to derail. Nor has it considered whether similar doors are fitted
to carriages beyond the class 158/159 fleets.
The role of the safety authority
299 The Office of Rail and Road (ORR) regulates the railway industry’s health
and safety performance. Its role includes the monitoring of health and safety
performance, carrying out assessments and taking action to enforce compliance
with health and safety law. ORR is also tasked with ensuring that appropriate
action is taken in response to RAIB recommendations.
300 ORR plans its routine inspection work on the basis of strategic risk priorities
and its analysis of where it can secure the most significant improvements in
safety management. Its inspections and assessments aim to draw systemic
conclusions which will promote improved safety arrangements across a wide
range of activities, rather than identifying specific shortcomings. However, if any
shortcomings were to be identified during inspections and assessments, then
these would be raised by ORR with the duty holder concerned.
301 ORR reported to RAIB that it has been challenging Network Rail regarding
its management of vegetation since 2014. This included the monitoring of the
company’s performance across its regions and routes.
302 In 2017, ORR escalated this challenge and required all routes within Network
Rail to develop and implement long‑term plans and interim mitigation measures
to manage vegetation risks. Each route demonstrated to ORR that plans were
in place and were being implemented. Network Rail’s Wessex route submitted a
15-year plan to bring the route back to compliance with Network Rail’s standard
by 2034. ORR therefore moved Network Rail onto a routine monitoring plan
regarding its management of vegetation risks in early 2020.
303 During this routine monitoring, ORR identified declining performance in the
management of vegetation within Network Rail’s Southern region, which includes
Wessex route. This triggered an ORR targeted review of the route’s performance
at the start of financial year 2021 to 2022.
304 ORR’s assessment of Network Rail’s Southern region was that there were varying
levels of maturity of vegetation management across the routes forming the region,
namely, Wessex, Kent and Sussex. While all three of these routes recognised
the ongoing risks from a lack of vegetation management, no evidence was found
of any immediate risks to safety. However, ORR was particularly concerned that
Network Rail’s Wessex route was not able to provide sufficient evidence on how
it effectively resourced, prioritised and delivered works against plans to control
vegetation risks.
Report 12/2023 85 October 2023
Salisbury Tunnel Junction
305 ORR’s review indicated significant vulnerabilities with risk controls in the following
Analysis
areas:
(a) Communication, co-ordination, and liaison between disciplines
(such as operations, off track and track) responsible for vegetation
management, which could result in risks being unmanaged.
(b) The training and competence framework for lineside staff, and the
lack of senior management accountability for lineside, in particular,
issues related to tree assessment and vegetation management, had
the potential for risks to go unrecognised.
306 This led to ORR’s decision in July 2021 to consider vegetation management
within Network Rail’s Southern region, including Wessex route, as an emerging
issue, with the potential for further escalation. ORR provided a corresponding
assessment of the route’s vegetation management performance as part of its
2021 to 2022 Network Rail Annual Assessment.34
307 By the end of October 2021, ORR considered that Network Rail had made
insufficient progress in addressing senior leadership for lineside management and
committing to the delivery of additional vegetation management works to manage
high risk leaf fall sites. This resulted in ORR escalating this topic with Network
Rail with further engagement and monitoring.35 After the accident, in December
2021 ORR took further action in response to evidence of the emerging themes
from the Salisbury accident (see paragraph 356).
34
Annual Assessment of Network Rail - April 2021 to March 2022, 20 July 2022 (www.orr.gov.uk).
35
In June 2022, ORR considered Network Rail’s Southern region had made sufficient progress to return to routine
monitoring.
Report 12/2023 86 October 2023
Salisbury Tunnel Junction
Previous occurrences of a similar character
Previous occurrences of a similar character
Autumn Adhesion Investigation
Part 1: Esher, 25 November 2005
308 On 25 November 2005, a South West Trains (the predecessor train operating
company to SWR) service from Alton to Waterloo passed signals at danger on
the Up Fast line between Esher and Hampton Court Junction by approximately
200 metres. After passing the signal, the train came within 200 metres of another
service from Woking to Waterloo, which had crossed from the Up Slow to the Up
Fast line at Hampton Court Junction.
309 The investigation into the incident (RAIB report 25/2006 (Part 1)) was
undertaken in parallel with the investigation into the SPAD that occurred at
Lewes on 30 November 2005 and a general investigation into the causes of
adhesion- related station overrun and SPAD incidents during Autumn 2005. RAIB
made three recommendations in this report which were not directly linked to the
causes of the Salisbury accident.
Part 2: Lewes, 30 November 2005
310 On 30 November 2005, a train passed a signal located at the end of platform 3 at
Lewes station at danger. After passing the signal at danger, the train ran through
a set of points which were not set for the passage of the train. Another train that
had departed from platform 5 at Lewes station on time was approaching the
points when the driver heard the other train approaching and, realising that the
two trains were on a conflicting route, stopped some 30 metres from the point of
conflict. RAIB’s investigation report (RAIB report 25/2006 (Part 2)) made three
recommendations which were not directly relevant to the causes of the Salisbury
accident.
Part 3: Review of adhesion-related incidents in Autumn 2005
311 Part 3 of RAIB’s review (RAIB report 25/2006 (Part 3)) found that the immediate
cause of the SPAD incidents at Esher and Lewes was poor adhesion between
wheel and rail. The two incidents occurred against a backdrop of an increase
in the number of adhesion-related SPAD incidents and a significant increase
in the number of adhesion-related station overrun incidents on the national rail
network during autumn 2005 compared with autumn 2004. The purpose of the
class investigation was to establish the causes of this increase and any common
themes. A number of recommendations were made as a result of the class
investigation.
312 Short-term recommendations related to the following:
• sanding systems
• monitoring of the level of sand
• defensive driving techniques
• vegetation management on approaches to junctions
• undertaking low adhesion railhead treatment
• proposing research into technology to estimate levels of contamination and/or
adhesion.
Report 12/2023 87 October 2023
Salisbury Tunnel Junction
313 Longer‑term recommendations related to the following:
Previous occurrences of a similar character
• research to determine how long low adhesion endures
• possible methods for identifying the presence of low adhesion
• methods for preventing railhead contamination formation and dispersal
• fitment of automatic sanding equipment
• braking performance of trains
• WSP equipment and modelling techniques
• research to determine whether magnetic track brakes were a cost-effective
solution for new-build rolling stock and/or retrofitting to existing rolling stock
• how information and intelligence and use of wireless remote data transmission
from rolling stock can help in providing details of low adhesion conditions in real
time.
314 By 2014, ORR considered that sufficient action had been taken against all 19
recommendations for them to be considered as implemented. RAIB has stated
in its annual reports that it had concerns that the actions taken, or proposed,
to address six of these recommendations were inappropriate or insufficient to
address the risks identified.
Passenger train collision at Darlington, 3 October 2009
315 On 3 October 2009, a passenger train arriving at Darlington station collided
with the rear of another passenger train which had just started to depart from
the same platform. The collision caused a small number of minor injuries, and
two passengers were taken to hospital for treatment. Both trains suffered minor
damage. The primary cause of the accident was contamination of the railhead
at the south end of the station, with high levels of leaf/vegetation material being
found.
316 RAIB’s safety bulletin on this accident (RAIB bulletin 01/2010) found that it
was likely that the poor weather and high winds resulted in leaves and other
vegetation being blown onto the line, from where they could be picked up by the
wheels of passing trains and transported short distances. This material, having
been deposited from trees during the early part of autumn, is likely to have had
a high moisture content, which would have significantly affected the existing low
level of adhesion. Bulletins had a similar role to RAIB’s current safety digests and
did not include recommendations.
Norwich Road level crossing, New Rackheath, Norfolk, 24 November 2019
317 On 24 November 2019, the barriers at Norwich Road level crossing, near New
Rackheath, Norfolk, lifted as a passenger train from Norwich to Sheringham was
approaching (RAIB report 15/2020). Two road vehicles crossed the railway in front
of the train, which reached the crossing less than half a second after the second
road vehicle was clear. The investigation found that leaf fall and atmospheric
conditions had resulted in contamination of the railhead and low adhesion
conditions. The contamination had not been removed because there were no
railhead treatment trains on the Norwich to Sheringham line at weekends. The
incident involved new rolling stock running on a different band on the railhead to
the previous trains and as such the contamination had not been cleared resulting
in poor electrical contact.
Report 12/2023 88 October 2023
Salisbury Tunnel Junction
318 As a result of the recommendations made by RAIB, Network Rail reported it had
Previous occurrences of a similar character
made an emergency change and modified its approach to autumn management,
specifically regarding the items to consider when planning the use of railhead
treatment trains as part of its national operating standard NR/L3/OPS/045.
This instruction was sent to the seasonal delivery managers and SDSs and
required them to liaise with train operators through the joint season management
group to identify any changes in rolling stock. ORR has determined that these
recommendations have been implemented.
Train collision with fallen tree and derailment near Balderton, Cheshire, 26 November
2021
319 On 26 November 2021, an empty passenger train collided with part of a fallen
tree while travelling at 46 mph (74 km/h) and derailed (RAIB safety digest
03/2022). The train was a class 150 DMU travelling between Wrexham General
and Chester. There were no injuries, but minor damage was caused to the train
and to local signalling equipment. The tree was felled by poor weather. The
Meteorological Office report about Storm Arwen stated that trees, including large
mature trees, were felled across the north of the UK due to the unusual direction
of the wind.
320 The safety digest highlighted the recommendations made to Network Rail in
RAIB’s Carmont report (RAIB report 02/2022) (see paragraph 326) to address
some of the factors present in the Balderton accident. These related to improving
processes for mitigating the effects of extreme weather conditions and enhancing
route control staff skills and resources to improve incident management.
Additionally, RSSB started a project to assess the effectiveness of blanket speed
restrictions in mitigating risks from trains colliding with trees or landslips.36
36
'Effectiveness of blanket speed restrictions in managing and mitigating risks from trains running into trees or
landslips', available at https://www.rssb.co.uk/research-catalogue/CatalogueItem/T1252.
Report 12/2023 89 October 2023
Salisbury Tunnel Junction
Summary of conclusions
Summary of conclusions
Immediate cause
321 Train 1L53 passed signal SY31 at danger and could not stop before colliding with
train 1F30 (paragraph 114).
Causal factors
322 The causal factors were:
a. Wheel/rail adhesion was very low in the area where the driver of train 1L53
applied the train’s brakes. This causal factor arose due to a combination of the
following:
i. The railhead was contaminated with fallen leaf debris, much of it as a
result of the weather conditions since the last railhead treatment run,
coupled with an increased density of vegetation; and wet conditions
from the band of drizzle that had recently passed over (paragraph 123,
Recommendations 1, 2 and 3).
ii. Network Rail’s Wessex route had not effectively mitigated the railhead
contamination (paragraph 156, Recommendations 1, 2, 4, and 5).
b. The driver did not apply train 1L53's brakes sufficiently early on approach to
protecting signal SY31 to avoid running onto the junction, given the prevailing
levels of wheel/rail adhesion (paragraph 172, Recommendation 8).
c. The braking systems of train 1L53 were unable to mitigate the effects
of the prevailing wheel/rail adhesion conditions (paragraph 203,
Recommendation 9).
Underlying factors
323 The underlying factors were:
a. Network Rail’s Wessex route did not effectively manage the risks of low
adhesion associated with the leaf fall season. This is a probable underlying
factor and included issues relating to:
i. resourcing (paragraph 225, Recommendations 2 and 3)
ii. liaison between departments (paragraph 231, Recommendations 1, 2
and 3)
iii. track access constraints (paragraph 237, Recommendation 1)
iv. MOM inspection process (paragraph 241, Recommendation 1)
v. understanding of railhead treatment effectiveness (paragraph 243,
Recommendations 1 and 2)
vi. weather response standards (paragraph 250, Recommendations
1, 4 and 5 and Recommendation 7 of RAIB report 02/2022, see
paragraph 326)
Report 12/2023 90 October 2023
Salisbury Tunnel Junction
vii. staff competences (paragraph 253, Recommendations 2 and 3 see also
Summary of conclusions
paragraphs 337 to 342).
b. South Western Railway had not effectively prepared its drivers for assessing
and reporting low adhesion conditions. This is a possible underlying factor
(paragraph 263, Recommendation 8, see also paragraphs 345 to 347).
Additional observations
324 Although not linked to the accident on 31 October 2021, RAIB made the following
two observations:
a. The TPWS installation at signal SY31 was not compliant with Network Rail
requirements for new signalling design (paragraph 276, Recommendation 7).
b. Network Rail’s processes for standard signal overrun risk assessment do
not require HRLA sites to be accounted for, nor does the identification of a
new HRLA site trigger a reassessment of signal overrun risk (paragraph 279,
Recommendation 6).
325 Factors that affected the consequences were:
a. There was a loss of survival space in the driver’s cab of train 1L53 (paragraph
285, no recommendation made as this is covered by Recommendation 19 of
RAIB report 02/2022, see paragraph 326).
b. Damage to the internal sliding doors on the train obstructed evacuation routes
and prevented train crew from accessing carriages to assist passengers
(paragraph 291, Recommendation 10).
Report 12/2023 91 October 2023
Salisbury Tunnel Junction
Previous recommendations that had the potential to
Previous recommendations that had the potential to address one or more factors identified in this report
address one or more factors identified in this report
Carmont, Aberdeenshire (2020) (published March 2022)
326 On 12 August 2020, a passenger train collided with debris washed from a drain
onto the track near Carmont, Aberdeenshire, following very heavy rainfall (RAIB
report 02/2022). The passenger train service from Aberdeen to Glasgow was
returning towards Aberdeen due to a blockage that had been reported on the
line ahead. There were nine people on board, six passengers and three railway
employees (one of whom was travelling as a passenger). The train was travelling
at 73 mph (117 km/h), just below the normal speed for the line, when the collision
occurred. This caused the train to derail and deviate to the left, before striking a
bridge parapet which caused the carriages to scatter. Tragically, three people died
as a result of the accident.
327 Following the accident, Scotland’s Railway established a permanently staffed
weather desk position. Network Rail has informed RAIB that suitably qualified
people have been recruited to cover this position, which is responsible for
monitoring weather conditions and advising controllers on the necessary
precautionary actions. In the light of the likelihood that climate change will
exacerbate this risk still further, Network Rail also decided to commission two task
forces to advise on earthworks management and weather.
328 The RAIB report into the accident at Carmont was published in March 2022, five
months after the accident at Salisbury. RAIB made 20 safety recommendations in
its report, two of which, recommendations 7 and 19, are relevant to the accident
at Salisbury Tunnel Junction. Recommendation 7 reads as follows:
Recommendation 7
This recommendation is intended to enhance the ability of route control
staff to contribute to the safe operation of a modern railway by making
good safety decisions in difficult circumstances based on a holistic
assessment of the most relevant information. It is intended to build on
the work already undertaken as part of Network Rail’s 21st Century
Operations programme.
Network Rail, in conjunction with train operating companies, should
review the capability of route control rooms to effectively manage
complex, widespread and unusual situations such as abnormal weather
conditions and multiple infrastructure failures. This review should
consider the steps needed to ensure that route controls have sufficient
staff with appropriate skills (technical and non-technical), experience and
knowledge, all with clearly defined responsibilities and accountabilities.
The review should therefore examine how Network Rail ensures that
route control staff are provided with appropriate training, learning and
professional development for their roles, supported by means of a
comprehensive competence management system, that enables them to
feel confident and empowered to make difficult decisions.
Report 12/2023 92 October 2023
Salisbury Tunnel Junction
As part of this review, Network Rail should also compare its railway
Previous recommendations that had the potential to address one or more factors identified in this report
control safety-related decision-making frameworks with those in other
organisations (such as off-shore exploration and air traffic management)
to determine if good practices can be imported into the railway
environment.
The review should be used to inform the development of a timebound
programme for the implementation of the measures that are needed to
develop the incident management capability of route controls.
329 Network Rail reported to ORR that a plan had been implemented setting out
various workstreams to address recommendation 7, relating to a training and
competence framework for more effective decision-making training for control
room staff and to improve understanding of the impact of abnormal weather
conditions.
330 Network Rail has also established a working group to review the critical
activities in Route and National Operations Control, with the aim of producing
a better process for managing assurance. The new process will take account
of the ORR guidance document Railway Safety Publication 1 ‘Developing and
Maintaining Staff Competence’.37 The work is closely linked to the development
of a competence framework to identify critical decision-making activities for staff
referred to in recommendation 7. The recommendation remains open at the time
of this report.
331 The second relevant recommendation, recommendation 19, 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.
37
https://www.orr.gov.uk/media/10885.
Report 12/2023 93 October 2023
Salisbury Tunnel Junction
c) In consultation with operators of other pre-1994 passenger rolling stock,
Previous recommendations that had the potential to address one or more factors identified in this report
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.
332 ORR reported that it had hosted a meeting on 6 April 2022 with owners and
operators of class 43 high speed trains (HST), together with government bodies
and RSSB, to consider how recommendation 19 should be addressed. The
initial consideration of the recommendation by relevant parties was done by the
Carmont Recommendation Steering Group (CRSG), co-ordinated by RSSB.
333 After reviewing the information provided by the bodies and organisations
responsible for implementing recommendation 19, ORR concluded that, in
accordance with the Railways (Accident Investigation and Reporting) Regulations
2005, RSSB has taken the recommendation into consideration and is taking
action to implement it. The recommendation remains open at the time of this
report.
Report 12/2023 94 October 2023
Salisbury Tunnel Junction
Actions reported as already taken or in progress relevant to
Actions reported as already taken or in progress relevant to this report
this report
334 Network Rail and SWR jointly undertook an internal rail industry investigation into
the accident. The report from this investigation recommended actions for both
Network Rail and SWR. Those relevant to RAIB’s findings are reported below.
Network Rail
335 The joint industry review placed the following actions on Network Rail:
a. Network Rail’s Head of Operations, Principles and Standards to undertake the
following:
• Liaise with RSSB to review rail industry standard RIS/8040/TOM, ‘Managing
Low Adhesion’ (then at issue 1) to include a requirement for the introduction
of restrictive working arrangements on the control of trains when poor or
exceptional railhead conditions are forecast, and the planned mitigation has
not been provided (see also paragraph 350).
• Review Network Rail standard NR/L2/OPS/095 and the process of risk
assessment that is used to identify HRLA sites. The review will consider
the consequences of incidents such as SPADs, overruns and derailments
arising from low adhesion conditions; this review is still in progress.
• Review compliance with Section 5 of NR/L3/OPS/021/01 for the installation
of trackside signage to indicate an HRLA site across all routes; this review is
still in progress.
b. In conjunction with (a), Network Rail’s Head of Seasons, Weather and
Resilience to review and align the standards relating to how Network Rail
manages the risk from low rail adhesion. The work is continuing and is
expected to be completed by the end of 2023.
c. Network Rail’s Head of Operations, Principles, and Standards to enhance
the training of controllers to provide them with the knowledge, skills and
competency to manage complex or extreme situations38 (paragraph 328).
d. Route Operations Directors to review the existing command structure,
available resources and support within Network Rail and integrated control
centres to assess the effectiveness of staff to deal with significant disruptive
events, including low adhesion. Any learning or recommendations is to
be included within a plan to address any shortfalls in resources. The work
is continuing and is expected to be completed by the end of 2023 (see
paragraph 342d).
e. Network Rail’s Head of Seasons, Weather and Resilience to arrange for
research to be undertaken to better understand the effectiveness of water
jetting by the MPV at differing speeds, and from this to review the water jetting
criteria of the MPV on Wessex and other routes. This work is ongoing at the
time of writing this report (see paragraphs 344d and 345g).
38
This is a duplication of a recommendation into the train derailment at Carmont.
Report 12/2023 95 October 2023
Salisbury Tunnel Junction
f. Network Rail Southern region appointed a new regional asset manager for
Actions reported as already taken or in progress relevant to this report
off track and a principal engineer for off track. However, maintenance delivery
units on Southern region remain the responsibility of the track maintenance
engineer (off track), with two TMEs currently in post in Wessex route and an
additional three planned for Kent and Sussex routes.
336 Other actions that Network Rail has reported that it has taken are discussed in the
following paragraphs.
Training and competence for staff
337 Network Rail has reviewed its training and competence framework for off track
staff and has introduced a new standard, NR/L2/CIV/1000/01 Module 1,
‘Competence Management for Drainage and Lineside’. This standard defines a
competence management system for off track staff, including a definition of the
knowledge, skills and behaviours required and the process for assessment.
338 Network Rail has stated that the assessment framework will:
a. define the knowledge, skills and behaviours of the roles across the
drainage and lineside (off track) activities such as inspection, assessment,
maintenance, evaluation, design and construction, assurance and
management
b. define competences and capabilities to support the development of a career
pathway
c. support decisions to minimise the impacts of safety, performance, reputational
and environmental risk.
339 Although Network Rail has introduced this new standard, the associated training
programme for off track staff remains under development. This training is planned
to be introduced in 2024, and so this action is still in progress.
340 Network Rail has also produced a training framework to develop the role of the
SDS in technical and non-technical skills. This training is planned to be introduced
in late 2023, and so this action is still in progress.
Autumn management
341 Network Rail has reported that it will revise NR/L3/OPS/021/01, ‘Autumn
Management’ (which will be retitled as ‘Adhesion Management’) to require a
risk assessment to be undertaken when planned railhead treatment is not to be
carried out. The revision will also require MOM inspection checks of HRLA sites to
be undertaken at least weekly and on a forecast ‘red’ or ‘black’ adhesion risk leaf
fall day.
342 In 2021 Network Rail established a working group to review the factors that had
been identified by RAIB and its internal investigation. In August 2023 Network Rail
provided an update on these actions:
a. Updates to standards and processes within NR/L2/OPS/021,
NR/ L3/ OPS/021/01 and NR/L3/OPS/045/3.17 are in progress and these
changes in conjunction with any learning from the RAIB report will be
implemented following the five-yearly review of RIS-3708-TOM in 2023.
Report 12/2023 96 October 2023
Salisbury Tunnel Junction
b. The title of the autumn working arrangements document, NR/L3/OPS/021/01,
Actions reported as already taken or in progress relevant to this report
has been changed to ‘Adhesion Management’ to capture the risks outside
of autumn (as part of the recommendation 1 from the RAIB Llanharan
investigation (RAIB report 03/2023).
c. The SDS and off track e-learning training for adhesion has been enhanced.
d. It is providing training and support to control room and seasons delivery staff
with weather academy workshops, building incident scenarios to facilitate the
rehearsal of weather plans and responses.
e. It has introduced meteorology training with MetDesk for seasons delivery staff
and their managers.
343 Network Rail held its first national low adhesion conference in February 2023,
bringing together its SDS community, MetDesk and research/railhead treatment
providers to share good practice and research.39
Network Rail’s Wessex route
344 Network Rail’s Wessex route reports that it has taken the following actions as a
result of the joint industry investigation:
a. The SAE and the newly appointed region asset manager (off track) are
reviewing and clarifying the required responsibilities for delivering leaf fall risk
control processes.
b. The route operations risk advisor and route asset manager (signalling) are
reviewing the SORAT assessment process for signal SY31 and other signals
in close proximity to junctions where sections of HRLA sites are also in the
vicinity to identify any learning potential and any recommendations (currently
in progress).
c. Ensured vegetation and leaf fall risk assessment data is included within the
Ellipse maintenance database.
d. The speed of MPV railhead trains through designated HRLA sections has
been reduced to 40 mph (64 km/h) to improve the effectiveness of the railhead
treatment. The railhead treatment programme has also been changed so that
all of its lines are treated twice a day on both weekdays and weekends, where
possible, during the autumn period.
e. Implemented assurance checks on the route to ensure compliance with
the '24-hour missed sites' criterion and ensure that when the railhead is not
treated for over 24 hours, these sites are identified as a missed site and
specified mitigation is provided by the development of site-specific plans.
f. Implemented assurance processes to ensure checks at high risk sites
are being undertaken at least weekly and following a red or black leaf fall
adhesion risk forecast, and that there is a process to verify that these
checks are being completed.
39
Similar conferences were held in previous years under the title of ‘National Autumn Review’.
Report 12/2023 97 October 2023
Salisbury Tunnel Junction
g. Undertaken a review of the resources for the autumn control desk within
Actions reported as already taken or in progress relevant to this report
the WICC. This will enable them to be more proactive in monitoring and
managing emerging risks from incoming data, including real-time weather
information and information from the intelligent infrastructure monitoring of
high risk WSTCF sites. This can then inform additional mitigations identified in
site- specific plans.
h. Installed new lineside low adhesion signage for all HRLA sites, including those
previously without any signage, to assist train driver’s identification of the
sites.
i. Included consideration of timing and repeat periodicity for GSM-R
transmissions during the EWAT calls.
j. Network Rail’s operations director Wessex, in conjunction with the train
operating companies, is reviewing the adequacy of the timetabling of
passenger trains at Salisbury Tunnel Junction with the objective of reducing
the frequency of trains that approach signal SY31 at red. This review will
also consider the appropriateness of the regulating policy at Salisbury Tunnel
Junction.
Network Rail and SWR
345 Network Rail’s Wessex route and SWR report that they have taken the following
actions as a result of the joint industry investigation:
a. Jointly agreed appropriate mitigation measures for the different levels of low
adhesion, including consideration of short formation trains. An additional
risk matrix was developed for various service disruptions and included
consideration of alternative mitigations where railhead treatment MPVs were
unable to cover their circuits. This has already been successfully used by
SWR to arrange operational mitigations during Autumn 2022.
b. Before Autumn 2022, they jointly updated the annual autumn working
arrangements to ensure that all HRLA sites were identified, reassessed,
managed and monitored.
c. Jointly reviewing and redeveloping the ROLA forms.
d. In conjunction with a contractor, commenced a trial on SWR’s class 158/159 of
on-train rail adhesion monitoring cameras to identify levels of contamination.
Network Rail has also fitted similar cameras with GPS technology to railhead
treatment MPVs for the same purpose. This work is currently ongoing.
e. Before Autumn 2022, introduced bi-weekly seasonal weather meetings with
Network Rail which become weekly during the autumn.
f. Developing software to correlate low adhesion forecasts that work with
real- time on-train data. This work is currently ongoing.
g. In collaboration with academia, undertaken testing and research into the
factors that affect the rate of build-up of all contaminants that affect low
adhesion. The trial40 looked at the effectiveness of current railhead treatment
methodology (water treatment and adhesion modifier) and alternative
technologies (laser, plasma and cryogenics). Network Rail is also working with
industry to develop other potential long-term solutions.
40
NR/RPT/6365 – The Future Technologies Trial Close Out Report.
Report 12/2023 98 October 2023
Salisbury Tunnel Junction
SWR
Actions reported as already taken or in progress relevant to this report
346 The actions that SWR reports that it has taken as a result of the joint industry
investigation are discussed in the following paragraphs.
Driver training
347 SWR has implemented the following actions relating to training and briefing of its
drivers:
a. Enhanced its guidance for driving in conditions of low adhesion.
b. Reintroduced face-to-face briefing on autumn arrangements to train drivers to
ensure that information has been effectively briefed and understood.
c. Enhanced its train driver briefing process to ensure its drivers are familiar
with the format of low adhesion signage. It has also revised the information
describing the location of HRLA sites to reference it to visual cues, such as
signals and bridge numbers, rather than mileages, thereby making it more
effective and memorable.
d. The 2022 autumn brief included the requirement to formally report a low
adhesion incident to the signaller, if the driver puts any low adhesion
information on the flip chart at the depot booking-on point. SWR has also
placed stickers on the flip chart to remind drivers of this requirement. The
autumn brief also included HRLA locations and risk scores.
e. Introduced the requirement to publish on the SWR intranet, on a weekly basis,
a list of sites where low adhesion incidents have occurred during the previous
week. This list will also highlight ‘hot spots’ where there have been two or
more incidents of low adhesion within 24 hours.
f. In November 2022, issued a traction notice to instruct all train drivers to
complete a running brake test on approach to known areas of HRLA as
documented in the Autumn brief.
g. Stopped the use of the pre-recorded GSM-R message that referred to
reporting ‘exceptional’ railhead conditions on the SWR driver training
simulator. The simulator operator will instead make a scripted general
broadcast alerting the train driver to ‘reportable’ adhesion conditions.
Other actions
348 SWR has also implemented the following actions:
a. Enhanced its risk assessment process to consider all relevant information,
including the use of shorter train formations during periods of disruption
caused by industrial action or weather-related issues.
b. Started a review of the WSP sanding system on diesel multiple unit trains to
consider the following:
• sand discharge rates and deployment of a sander air flow control
modification
• implementing the use of nozzles that better direct sand, thereby
improving adhesion.
c. Started a programme to retrofit single variable rate sanders to its
class 158 and class 159 fleet during 2023 (see paragraph 355).
Report 12/2023 99 October 2023
Salisbury Tunnel Junction
d. Consulted with Porterbrook, the owners of the class 158 and 159 units
Actions reported as already taken or in progress relevant to this report
operated by SWR, to review the risk assessment undertaken before fitment
of the current WSP system controller. This review will consider:
• practicable adhesion benefits of the sanders maintained to comply
with the sand discharge limits set in the Porterbrook overhaul
specification
• realistic sander air consumption rates
• the presumed maximum length of low adhesion track that a train
might encounter.
e. Updated its Wessex route risk assessment and posted it on the SWR
intranet, which will also include all information associated with route
knowledge to make it easier for crews and managers to provide regular
updates if information is changed.
f. Appointed a route training manager to ensure that all route risk assessments
are up to date. The review process requires the documents to be updated
following any major change to the infrastructure, serious operational incidents
or every five years, whichever comes sooner.
g. Introduced the ‘Notus’ system to assist in advising SWR drivers of WSP
activity on Network Rail’s Wessex route. This information is made available
on drivers’ tablets and displays a map which highlights HRLA sites and areas
where there have been low adhesion reports (defined as wheelslip events
over 7 seconds which have been captured on OTDRs in the previous 24
hours or locations where multiple wheelslip events have been recorded).
h. Training has been delivered to both existing and trainee guards in the use of
GSM-R. SWR is also undertaking assessments of the competence of guards
in making an emergency call via the GSM-R radio system, and is considering
providing training to assist guards in making these calls in emergency
situations.
i. Purchased the GSM-R training app that staff can access on their mobile
devices at any time to refresh their knowledge.
j. Approached RSSB and the Adhesion Research Group regarding the
following (this work is still in progress):
• proposing research into the development and potential use of on-train
technology to detect and automate the process of reporting locations where
the adhesion is lower than expected
• proposing a review of the current Rule Book section on the reporting
terminology for low or poor railhead adhesion
• proposing a review of the suitability of the braking instructions
recommended in the independent report produced by the Adhesion
Working Group in April 2006 on braking in adverse rail conditions, and the
instructions to a driver when a forecast has identified a ‘Red’ or ‘Black’
weather or adhesion day
Report 12/2023 100 October 2023
Salisbury Tunnel Junction
• proposing a review of RIS-8040- TOM, ‘Low Adhesion between the Wheel
Actions reported as already taken or in progress relevant to this report
and the Rail - Managing the Risk’ (issue 1, December 2016) to include
additional information or a description on how to identify an HRLA location,
the factors that must be considered in the site-specific risk assessment
process and providing guidance on the control measures to apply in the
event of degraded or exceptional conditions.
Great Western Railway
349 GWR reported it has taken the following actions:
a. Reviewed the content of its train driver briefing material to ensure material
meets the standards set out in the SCSG’s ‘GB Rail Approach to Managing
Low Adhesion’ document (see paragraph 353).
b. Ensured information on HRLA sites (as identified using the new
RIS- 0840- TOM methodology) is available for its train drivers and confirmed
with Network Rail Wessex route that low adhesion signage for each location is
present.
c. Provided clarity of wording on braking instructions for its train drivers, and
rebriefed the requirements for running brake tests and other driver actions in
low adhesion conditions.
d. Ensured the process for late notice notification of ‘red’ and ‘black’ adhesion
risk days is effective, including notification if the railhead treatment MPV has
been delayed or cancelled.
e. Driver managers have been instructed to brief new drivers on assessing
autumn risks. Driver managers have received ‘Brief the briefer’ training to
support this strategy. New drivers will also have to complete a revised training
session. All other drivers will be required to read and acknowledge that they
have received and understood the briefing document issued on a bulletin.
f. Undertaken to audit items (a) to (e) regularly to provide assurance that
information has been understood.
Other industry bodies
RSSB
350 As a result of the joint industry investigation, Network Rail and SWR had further
discussions with RSSB. In June 2022, RSSB revised Rail Industry Standard
RIS-8040-TOM, ‘Managing Low Adhesion’ to issue 2. This revision incorporates
relevant content from issue 1 of RIS-8040-TOM and GEGN8540, ‘Guidance on
Low Adhesion between the Wheel and the Rail - Managing the Risk’ (issue 2,
June 2015). The revised standard includes a simple structured framework
(figure 48) and clarifies the responsibilities of the infrastructure owner and train
operator in preparing for autumn. It also provides guidance for infrastructure
managers to develop, implement, monitor and review the effectiveness of
site- specific plans to manage low adhesion at high risk sites and to assist train
operators in controlling low adhesion risks and improving safety and performance
in low adhesion conditions.
Report 12/2023 101 October 2023
Salisbury Tunnel Junction
351 RSSB discussions with SWR also led to further changes to RIS-8040-TOM in
Actions reported as already taken or in progress relevant to this report
respect of low adhesion signage. Additionally, it now requires that when control
measures are identified as ineffective or unavailable, such as missed runs of an
MPV or RHTT, or where planned vegetation work has not been undertaken, the
infrastructure manager shall implement alternative mitigation measures without
delay.
352 RSSB has delayed the 12-month post-publication review of RIS-8040-TOM to
allow for any recommendations identified from the RAIB investigation into the
accident at Salisbury to be included in the review.
Figure 48: Diagram within Rail Industry Standard
RIS- 8040, ‘Managing Low Adhesion’.
Seasonal Challenge Steering Group
353 The Seasonal Challenge Steering Group (SCSG), a working group of industry
professionals and academics, issued a document ‘GB Rail Industry Approach to
Railhead Adhesion Management’ (version 1.1, May 2022). The document sets
out the recommended approach to the management of railhead adhesion on the
rail network using proven control measures and potential future control measures
currently being developed by the rail industry.
Rail Delivery Group
354 Following RSSB research project T1107 ‘Trials of sanders and sand laying rates’
(paragraph 221) undertaken by the Adhesion Research Group, the Rail Delivery
Group prepared a national business case for retrofitting variable rate sanders
(single and double) to mainline trains, funded by the performance innovation fund
(PIF).
355 Following this research, several successful PIF applications were made by:
a. Northern Trains to retrofit double variable rate sanders to their class 323 fleet
b. ScotRail to retrofit double variable rate sanders to their Class 170 fleet
c. SWR to retrofit single variable rate sanders to their class 158 and class 159
fleet during 2023.
Report 12/2023 102 October 2023
Salisbury Tunnel Junction
Office of Rail and Road (ORR)
Actions reported as already taken or in progress relevant to this report
356 In December 2021, ORR issued an improvement notice to Network Rail. The
notice required Network Rail to:
a. Review all available data sources and identify those which can usefully help to
shape or refine the deployment of railhead contamination risk controls.
b. Have in place and act upon an up-to-date autumn risk assessment.
c. Review the extent, type, frequency and speed of railhead treatment
operations with a view to optimising the allocation of treatment resource
according to identified risk. The review will include consideration of the risk of
contamination being transferred into untreated sites and what action will be
taken in the event of a missed or deferred planned railhead treatment.
d. Review the vegetation control programme with the aim of ensuring that it is
aligned with the need for railhead contamination risk control, as identified
above.
e. Demonstrate there is effective communication between the relevant disciplines
involved in vegetation management and control of seasons risk. Recognising
legislative constraints during the bird nesting season, implement a plan to
undertake the highest priority vegetation clearance.
f. Reassess the purpose and value of sample site inspections by MOMs and
confirm arrangements to enable any required safe access to the track. If
sample site inspections are a useful part of risk management, provide the
appropriate level of resource and review how inspections are prioritised
to those sites where lost adhesion presents a greater risk of collision or
derailment.
g. Review the effectiveness of the use of autumn adhesion forecasting in
shaping local responses and providing information in a useable format to train
operating companies and other operations stakeholders.
h. Have in place arrangements to undertake assurance on the effectiveness of
autumn controls.41
357 ORR has reported that Network Rail has since complied with the improvement
notice and has taken action to manage the risk from adverse railhead conditions
caused by leaf fall. The actions taken include:
a. Engaging with stakeholders and reviewing data to update its risk assessments
and identify new and existing sites at high risk of low adhesion.
b. Targeted vegetation removal work at the highest risk sites.
c. Implementing changes to the railhead treatment programme including revised
treatment circuits and speed, and the trialling of new treatment products.
d. Strengthening of the mitigation process when railhead treatment is missed.
e. Reviewing, publishing and briefing the autumn working arrangements to
relevant people. The arrangements include the inspection of high risk sites
and audits of railhead treatment effectiveness.
41
A copy of the improvement notice can be found at: https://orrprdpubreg1.blob.core.windows.net/docs/IN-
KL-20211224-01%20Network%20Rail%20improvement%20notice.pdf.
Report 12/2023 103 October 2023
Salisbury Tunnel Junction
f. Strengthening assurance arrangements for autumn management, including
Actions reported as already taken or in progress relevant to this report
engaging in the planning of future engineering works with the aim of planning
engineering work around the railhead treatment circuits and trialling new
technology to identify railhead contamination.
Report 12/2023 104 October 2023
Salisbury Tunnel Junction
Recommendations
Recommendations
358 The following recommendations are made:42
1 The intent of this recommendation is for Network Rail to have autumn
working arrangements that more effectively manage the low adhesion
risk, as a result of leaf fall.
Network Rail should consider the findings from this report to inform
a review of the processes, standards and guidance documents and
supporting management arrangements relating to the management of
leaf fall low adhesion risk. The review should result, where appropriate,
in the creation or revision of documents suitable to support Network Rail
staff in having an appropriate understanding of the risks when creating
autumn working arrangements. It should also identify the necessary
resource and competence required for their effective implementation.
The review should examine both the roles of operations and
maintenance (track and off track) and specifically include consideration
of:
a. leaf fall risk assessments, including consistency in their
implementation
b. capture, sharing and tracking of data and planned mitigations,
especially those related to vegetation management
c. definition of responsibilities and necessary competences, including
knowledge of the factors affecting leaf fall risk and low adhesion from
contamination build‑up and the effectiveness of mitigation measures
d. required resource to effectively undertake the main roles
e. alignment of the requirements and processes across all related
departments to promote a co-ordinated approach and a common
understanding of the risks and mitigations.
Network Rail should ensure that any revised processes, standards
and guidance are produced to a timebound plan, and supported by
appropriate training and briefing and that this includes any contracting
staff involved in the process (paragraphs 322a (i) and (ii), and 323a (ii) to
(vi)).
42
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, these recommendations are addressed to the Office of Rail and Road to enable it to carry out its 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.raib.gov.uk.
Report 12/2023 105 October 2023
Salisbury Tunnel Junction
Recommendations
2 The intent of this recommendation is for Network Rail to have seasons
delivery specialists that are more effective in managing Network Rail’s
seasonal risk.
Network Rail, building on the work that has already started in this area,
should develop an appropriate competency framework for the role of the
seasons delivery specialist.
This framework should include:
a. a job description that accurately reflects the responsibilities of the
role
b. the necessary technical skills required to undertake the role
effectively
c. the necessary non-technical and management skills needed to
undertake the communication and co‑ordination required of this role
d. appropriate training material
e. arrangements to confirm that staff have achieved, and continue to
have, the required level of competence.
Network Rail is to arrange for provision of the necessary staff to fulfil
the roles and develop a time-bound programme for implementation of
the associated training, supported by suitably qualified assessment staff
(paragraphs 322a (i) and (ii) and 323a (i), (ii), (v) and (vii)).
3 The intent of this recommendation is that Network Rail off track staff are
sufficiently competent and confident to undertake the tasks assigned to
them by Network Rail standards.
Network Rail should produce a time-bound programme to train
and assess the competence of off track maintenance staff in the
requirements of standard NR/L2/CIV/1000/01 Module 01, ‘Competence
Management for Drainage and Lineside’ (paragraphs 322a (i) and 323a
(i), (ii) and (vii)).
Report 12/2023 106 October 2023
Salisbury Tunnel Junction
Recommendations
4 The intent of this recommendation is for Network Rail to be able to make
more effective decisions regarding the management of emerging and
potential low wheel/rail adhesion conditions.
Network Rail, working in co‑operation with train operators, Rail Safety
and Standards Board and other relevant stakeholders, should undertake
research into real-time data that could be used to give an indication of
the wheel/rail adhesion conditions on its network and how this could be
used to support operational decisions to implement mitigation measures.
This review should include consideration of the following:
a. monitoring data, including that drawn from on-train data recorders,
wheel slide protection activity, and records of wrong side track circuit
failures
b. reports of low adhesion from train drivers and staff
c. weather and low adhesion forecasts.
This review should take account of good practice in other parts of the rail
sector both in the UK and abroad (paragraphs 322a (ii) and 323a (vi).
5 The intent of this recommendation is for Network Rail to improve
wheel/rail adhesion conditions through the application of improved
understanding of the effectiveness of railhead treatment regimes.
Network Rail should undertake research to better understand:
a. the factors that affect the rate of build-up of leaf fall contamination,
for instance, the environment, meteorological conditions, topography,
tree species and railway operations
b. the relationship between different types of contamination and low
railhead adhesion
c. the effectiveness and longevity of currently available alternative
railhead treatment regimes.
The findings from this research are to be used to support the seasons
delivery specialist in decision‑making relating to the necessary frequency
of railhead treatment and understanding the impact of missed or delayed
treatment (paragraphs 322a (ii) and 323a (vi).
Report 12/2023 107 October 2023
Salisbury Tunnel Junction
Recommendations
6 The intent of this recommendation is to enable the effective assessment
by Network Rail of the risk of overrun at signals which have HRLA sites
on their approach.
Network Rail should review its signalling standard
NR/ L2/ SIG/14201/ Mod04, ‘Signalling Risk Assessment Handbook’ to
ensure that signal overrun risk assessments appropriately consider the
impact of any high risk of low adhesion sites on approach to the signal.
Network Rail should also consider if the reassessment of signal overrun
risk is required when a new high risk of low adhesion site is identified on
approach to any signal capable of displaying a red aspect.
Any revised standard or process should be suitably briefed to all relevant
parties and consideration should be given to whether a revised overrun
risk assessment against the new standard should be required where
existing signals capable of displaying a red aspect have a high risk of
low adhesion site on their approach (paragraph 324b).
7 The intent of this recommendation is to reduce the risk of overrunning
signals at danger where there is a line speed change on the approach
after the preliminary caution signal.
Network Rail should review the decision not to retrospectively apply
technical instruction TI022 ‘Provision of TPWS at signals’ issue 4 to
existing signals. Should retrospective application of TI022 be found
appropriate, Network Rail should implement the required changes to
existing Train Protection and Warning System equipment (paragraph
324a).
8 The intent of this recommendation is that South Western Railway drivers
are able to identify areas of low adhesion and report them, if appropriate.
South Western Railway should review its arrangements for training and
briefing drivers to ensure that they are able to effectively identify areas of
low adhesion and that they report them if appropriate. This review should
specifically understand the effectiveness of the relevant provisions
of the railway Rule Book in informing drivers as to the requirements
for reporting low adhesion, as well as other methods. South Western
Railway should evaluate its processes for monitoring and reviewing the
reporting of low adhesion by drivers to ensure that these arrangements
remain effective (paragraphs 322b and 323b).
This recommendation may apply to other transport undertakings.
Report 12/2023 108 October 2023
Salisbury Tunnel Junction
Recommendations
9 The intent of this recommendation is for industry to realise the potential
benefits of future technologies to enable trains to better cope with low
wheel/rail adhesion when braking.
The Rail Delivery Group working with the train operating companies
and Rail Safety and Standards Board should create a framework and
mechanism for the assessment of future technologies to enable trains to
better cope with low adhesion when braking.
The framework should set out criteria and establish the process for cost
benefit analysis to apply to the assessment of future technologies as
they arise (paragraph 322c).
10 The intent of this recommendation is to minimise the risk that
passengers are unable to evacuate from class 158 and 159 carriages.
Porterbrook, Angel Trains and Eversholt Rail, working in conjunction with
the operators of class 158 and class 159 trains, should review the design
of the internal sliding doors on these carriages and determine if there
is a practicable means to prevent these doors becoming jammed in the
event of a collision.
They should develop a time‑bound plan to implement measures
identified by this review (paragraph 325b).
Report 12/2023 109 October 2023
Salisbury Tunnel Junction
Appendices
Appendices
Appendix A - Glossary of abbreviations and acronyms
AWG Adhesion Working Group
AWS Automatic Warning System
CCTV Closed-circuit television
DMU Diesel multiple unit
DVRS Double variable rate sanding
EWAT Extreme weather action team or Extreme weather action
teleconference
FFCCTV Forward-facing closed-circuit television
GBW Geographic block working
GWR First Greater Western Ltd, trading as Great Western Railway
HRLA High risk of low adhesion
HST High speed train
IIT Infrastructure Intelligent Technicians
MDU Maintenance delivery unit
MOM Mobile operations manager
MPV Multi-purpose vehicle
ORR Office of Rail and Road
OSS Overspeed system
OTDR On-train data recorder
PDP Professional driving policy
PIF Performance innovation fund
RHTT Railhead treatment train
ROLA Report of low adhesion
RSSB Rail Safety and Standards Board
SAE Senior asset engineer
SDS Seasons delivery specialist
SORAT Signal Overrun Risk Assessment Tool
SPAD Signal passed at danger
Report 12/2023 110 October 2023
Salisbury Tunnel Junction
SWR First MTR South Western Trains Ltd, trading as South Western
Appendices
Railway
TGA Traction gel applicators
TME Track maintenance engineer
TPWS Train Protection and Warning System
TSS Train stop system
WAIF Work arising information form
WSTCF Wrong side track circuit failure
WICC Wessex Integrated Control Centre
WSP Wheel slide protection
WSPER WSP Evaluation Rig
Report 12/2023 111 October 2023
Salisbury Tunnel Junction
Appendix B - Sources of evidence
Appendices
RAIB used the following sources of evidence in this investigation:
• information provided by witnesses
• information taken from the train’s OTDR and remote condition monitoring equipment
• examination of the carriages quarantined
• CCTV recordings taken from train 1L53
• site photographs and measurements
• weather reports and observations at the site
• research and analysis
• a review of previous RAIB investigations that had relevance to this accident.
Report 12/2023 112 October 2023
Salisbury Tunnel Junction
This report is published by the Rail Accident Investigation Branch,
Department for Transport.
© Crown copyright 2023
Any enquiries about this publication should be sent to:
RAIB Email: enquiries@raib.gov.uk
The Wharf Telephone: 01332 253300
Stores Road Website: www.raib.gov.uk
Derby UK
DE21 4BA