On Thursday 28 June 2012 a landslide occurred on the western slopes of Stob Coire Sgriodain by Loch Treig, in the Scottish Highlands approximately 6 km south–south-west of Tulloch Station.
At 20.00 hours that evening this landslide caused the derailment of a goods train on the Crianlarich to Fort William line, causing the engine to fall down the embankment towards the loch.
Earlier remedial work, immediately north of the landslide, took place in 2010 when rock armour was placed above the track to stabilise a retaining wall. This was not related to any landslide event.
The BGS has no record of earlier slope failure on the eastern shore of Loch Treig. Since the railway opened in 1894 there is no record of disruption and historic Ordnance Survey maps do not show any evidence for slope failure.
The landslide was reported to the BGS on Monday 2 July and a visit was arranged by Network Rail for BGS landslide specialists to examine the event on Monday 9 July. By this time, while the failure scarp and pathway could be examined, the landslide debris had been cleared from the trackway and the track sleepers and rails were being relaid. The line reopened on Wednesday 11 July.
The landslide was initiated by a shallow planar movement that developed into a debris flow. The landslide event is recorded as National Landslide Database ID NLD 18683/1.
BGS mapping shows superficial deposits comprising till (boulder clay) and morainic deposits on the loch side rising to a height of 350 m above OD on the hillside. These superficial deposits overlie the Loch Treig Schist and quartzite bedrock, metamorphic rocks of Neoproterozoic age. There is little evidence for the presence of till in the vicinity of the backscarp, however, and the failure appears to have occurred in saturated soil and head containing cobbles and large (>1 m diameter) boulders.
A landslide backscarp was examined on the western slope of Stob Coire Sgriodain. The backscarp is at an elevation of 325 m above OD and is roughly 15 m in width. Bedrock is not exposed within the scarp but is estimated to be within 1 m of the surface; detachment of material, therefore, is likely to have taken place close to the bedrock surface. The slide has excavated material to a length of roughly 20 m and forms a concave-upwards depression on the hillside.
A smaller scarp 18 m to the north of the failure was also examined (marked 'spring' on top photograph). This feature comprised a 3 m-wide scar below which a narrow gully containing cobbles and boulders extended downslope for 20 m. The scarp was associated with running water, and it appears that this feature is not the result of slope failure but rather a small spring, below which the fine-grained material had been washed away.
Below the backscarp, the debris flow track is roughly 20 m in width and extends down slope to the railway line. This pathway is marked by downslope-smoothed grass and scrub over which the debris has passed. The pathway contains fragments of bedrock (metasedimentary schist and quartzite) including boulders and is covered by a thin layer of reddened soil and head. The lower parts of the pathway forming the railway cutting have been stabilised, fenced and netted. Tilt meters have been installed around the slide to detect any further movement and new drains have been installed.
The majority of the debris from the landslide accumulated on the level ground forming the railway track some time before the evening of 28 June. It is understood from Network Rail that the derailment was caused by the train impacting a boulder, not from the quantity of debris on the track.
A small number of boulders were observed in the trees 10 m below the level ground but it is uncertain whether they were transported by the landslide event or removed during remedial work to restore the railway.