On 14 January 2013, a routine run by Network Rail’s New Measurement Train on the Up Midland main line between Sheffield and Derby threw up an anomaly. There were signs of a very minor track fault on the cess rail resulting in a slight twist in the track. The fault was flagged up so that an eye could be kept on it. Writes Clare Brint
On 10 June, another run showed that the defect was slightly worse but the amount of twist was about the same. Further runs on 11 November and 9 December showed that the twist was slowly worsening and the track maintenance engineer (TME) made a comment on the fault during his periodic review of the traces, noting that it appeared ‘static’ and should be monitored.
There was no other indication at that time that anything out of the ordinary was happening until, following a prolonged spell of heavy rain over the Christmas period, a ‘rough ride’ was reported on 1 January 2014. The fault was manually corrected by the permanent way team but, unfortunately, re-appeared several times over the next three weeks. Every time that the track was lifted and packed to correct the fault, there was rapid deterioration afterwards.
By 21 January, a 50 mph temporary speed restriction (TSR) was imposed. Track patrollers noted that the cutting slope appeared to have lowered, the catchpits were leaning and they suspected that the earthwork might be the underlying cause.
Meganne Paul, Network Rail’s asset engineer, visited the site on the following day and the scale of the problem became clear. Network Rail’s Infrastructure Projects (IP) team were engaged to start work the same day and Principal Contractor Construction Marine Limited (CML) was on site the next morning.
Understanding the problem
It was clear from the initial visits that an area of the hillside about 100 metres wide and 100 metres long, the bulk of which was outside of Network Rail’s land, was moving down the slope in a series of slips. The lowest slip had moved into the cess area and was pushing the cess rail up.
There were some spoil heaps of recently tipped material in the upper slope immediately below the backscar. Aerial photographs taken using Network Rail’s helicopter were very useful in helping the team to visualise and understand the slip.
It had happened before
Records retrieved from Network Rail’s National Record Group (NRG) in York revealed some very interesting history. In 1969, the old Broomhouse Tunnel, located a few hundred yards south of this site, was opened out into a cutting and the contractor who carried out that work was paid to remove the spoil. The notes stated that “after a winter’s experience it was found necessary to carry out further excavation to stabilise the cutting”, and this work was carried out by the same contractor as an extension of the original contract.
The site for the tipping of the spoil from the original contract had been restored before the additional work was authorised, so the contractor therefore found it necessary to negotiate for another site for this additional spoil. A private arrangement was made with a local landowner to place the material in an area of ground adjacent to the railway known as Thorpe Spring Wood or Spring Bank Wood.
The slope was stepped and the spoil material placed in September 1970. Three months later, a landslip occurred which resulted in a caution being put on the trains due to slight movement in the rails.
Urgent remedial works discovered that the spoil had been placed on top of two natural springs, one in the upper slope and one in the lower slope. Ground investigation and monitoring instruments installed at the time indicated that movement in the upper slope was occurring at the interface between the fill and the natural ground and that movement at track level was caused by the weight of the additional spoil on to the natural weathered rock which had previously been softened and saturated by the lowest spring. Movement of the lower slip was therefore occurring through the natural soils. The remedial works included removal of some of the tipped material and installation of drainage designed to intercept the natural springs and dry out the lower slope.
Ongoing monitoring of the slope continued from 1970 and included measurement of groundwater levels, installation of slip indicator tubes to pick up depths of deep-seated movement, and the monitoring of pegs to pick up surface movement.
Over the next few years, it was found that, although further slight movements were recorded in the hillside, the slip was no longer affecting the railway and that the previous remedial works and drainage measures appeared to be effective.
The British Rail civil engineering department continued to monitor the slope and, by 1977, movements in the hillside had become sufficiently large that further work was proposed to safeguard the railway. These included additional drainage improvements and the installation of an automatic trip wire landslip detector, both of which were put in place during 1978/79.
It appears from the available archive information that, following these works, British Rail considered the landslip was no longer affecting the railway. There are no further records of monitoring of the slip indicators or groundwater levels after 1986 and the landslip detector had became redundant by the 1990s.
Back to today
Although it was hoped that the line could be kept open with the movements being managed by the track maintenance team, the Train and Freight Operating Companies (TOCs / FOCs) were advised within the first week that they should start preparing plans for line closure. Track maintenance engineer Mark Owens and his section manager Phil Milner developed a detailed and clear plan to react to deteriorating twist levels and ensure the safety of trains. In this particular situation, and given the number of unknowns, the plan included limits for reduction in speed and line closure which were set at more conservative values than those mandated by Network Rail Standards.
Phil Milner and his team visited site daily and, when movements increased, maintained a 24-hour presence on site, monitoring for track movement and lifting and packing as required. In addition, CML were on site surveying the hillside daily. As an extra precaution, a system of remote monitoring was installed on the sleepers to give 24-hour twist values and text alerts if these deteriorated. All of this monitoring information was reviewed, as it was received, by the geotechnical, track and project teams so that correct decisions could be taken regarding safe line speed.
To make sure that the operators could fully understand the situation and the potential impact it might have on their customers, Clare Brint, Network Rail’s senior geotechnical asset engineer, provided a daily email update and attended frequent teleconferences to keep everyone informed. These updates included information on the amount of movement in the hillside and at track level, and predicted the deterioration rate so that an estimate could be made on how much longer the line could be kept open.
As understanding of the landslip, its history, behaviour and effect on the track increased, CML and their design engineers developed a plan to address the lowest part of the hillside first, in order to halt further movements at track level. This would be followed by remediation work to the whole hillside.
Closure and reopening
Following several more days of heavy rainfall, there was significant further movement at track level. The decision was taken to give the TOCs and FOCs 36 hours notice that the Up line would be closed
on the morning of Tuesday 18 February 2014. At that stage, the programme for the works to allow the line to be re-opened was six weeks from the date of closure, which would mean it would not reopen until the end of March.
This was due to the anticipated length of time to excavate the shear trench in short lengths to protect the stability of the hillside and constraints both with on-site materials storage and vehicle movements into and out of the site.
The lowest slip was remediated by the installation of a ‘shear key’ – a 100 metre long stone-filled excavation, five metres wide and five to six metres deep, which went below the level of the slip.
This work took just three weeks and successfully prevented further movements at track level.
It had been anticipated that a full track renewal would be required due to the clay material having been pushed up, contaminating the ballast. This would have required plant, equipment and staff to be diverted from planned track renewals as well as up to two days of full line blockade for both lines. However, in response to a request to speed up the process, a number of various other options were explored. It was decided to use the High Output Ballast Cleaner (HOBC) to excavate and replace the contaminated ballast as an interim measure.
Following confirmation that the shear key was effective and the slip had stopped affecting the track, the HOBC was used on the night of Friday 7 March, allowing the line to be re-opened on Saturday 8 March 2014 at 50 mph and returned to line speed the following week.
The hillside continues to move, and work is still underway to remove up to 100,000 tonnes of soil and ensure that these slips do not affect the railway again. It is estimated that this will take a further three months, after which a full track renewal is being planned for 2015/16.
That should put an end to the story of Unstone’s slippery slope, a tale that started forty-five years ago.