Leaves on the line – how many times have we heard or seen this offered up to our customers as a reason for delayed trains, cancelled trains, or missed station stops during the autumn season?

Do leaves actually fall on railway tracks and somehow get embedded there during autumn, causing untold and unfathomable problems with trains not being able to stop, not being able to start, and disappearing from the screens that railway signallers use to regulate and control the railway?

The recent published report by the Rail Accident Investigation Branch (RAIB) into the accident at Salisbury Tunnel Junction in October 2021 confirms that any or all of these problems can happen. Fortunately, they don’t need to happen every autumn day and the likelihood and consequences of them happening can be predicted, managed and controlled, even if, as at Salisbury Tunnel Junction two years ago, they are not yet totally eliminated.

What do we really mean by 'leaves on the line'?

Our climate, our geography, the vegetation that grows on and around our railway, the types and mixture of railway services that we run in this country – they all play a part in making our railway particularly prone to service disruption in the autumn, compared to many other railway-operating countries that do not share these specific features.

'Leaves on the line' is actually a shorthand for leaves and other vegetable matter being trapped between steel rails and steel railway carriage or wagon wheels and becoming compressed, spreading out as a thin dark layer of largely organic (carbon containing) material over, and adhering to, both the tops of the rails and the treads of the wheels. The composition, the extent of this thin layer and its propensity for adhering to rails and wheels can all be modified by the mixture of leaf species involved, highly local ground level air temperatures and humidities, whether oil, grease, or fuel spillage have occurred locally and recently etc. These thin layers are often both slippery and electrically insulating – both these features present their own particular railway safety and performance challenges.

Once a thin, dark, largely organic material containing layer has formed on the tops of rails, or on wheel treads, whether this is an issue or not depends on if a train is trying to brake or accelerate, especially away from standstill, over it, and what type of train detection and signalling technology is in place at this location. It is perfectly possible for an extended layer of this material to form and for it to play no, or only a very minor role in any autumn railway performance issues. It is also possible for a quite short layer of this material to form on one line and not form on the immediately adjacent line only a couple of metres away, and for this to impose a significant performance risk for the line on which it has formed, and none for the adjacent line.

So what are the risks of 'leaves on the line' and how can we mitigate them?

Safety and performance challenges posed by these thin layers of material include trains skidding over signals set at danger (red light showing), trains skidding past station platforms where they are scheduled to call, trains not being detected by some types of signalling systems, trains either not being able pull away, or taking an extended amount of time to pull away from signal or station stops. We refer to both trains skidding and trains not being able to pull away in these circumstances as suffering from 'low adhesion'. We refer to trains not being properly detected and confirmed at a particular location due to rail top or wheel tread contamination as 'loss of train detection'.

Mitigations that we use in the railways during autumn fall into two camps – those which try to prevent these thin layers forming, or which try to remove them, or negate their influence once formed – and those that try to reduce the extent and impact of consequences once they happen. Examples of the first type include Network Rail running Railhead Treatment Trains which use high pressure water jets to blast these thin layers off rail tops, running these railhead treatment trains at times, and over routes, according to adhesion predictions and adhesion reports received, cutting down and removing overhanging vegetation prior to autumn, using more sophisticated track or train mounted electronics to attempt to better allow the signalling system to detect and report the location of the train – examples of the second type include sanders and wheel slide protection systems fitted to trains, running longer trains, slowing trains down, reducing the need for trains to apply brakes as often through the use of minor timetable changes.

The future of adhesion

Our ability to both predict when, where and by how much low adhesion will occur – and to thus better and more efficiently deploy Railhead Treatment Trains has improved dramatically in recent years. We have improved and now better target pre-autumn vegetation clearance, whilst also taking increased account of maintaining and increasing lineside biodiversity, we have enhanced the design and the performance of wheel slide protection and braking systems fitted to new trains and have in some cases also begun to retrofit these to older trains – we are currently testing alternative methods of cleaning contaminated rails, including the use of dry ice CO₂ pellets.

A few years ago, the concept of a 'seasonal agnostic' railway where we can run the railway as reliably over the entire autumn season as we can in spring would have been unbelievable. Even though we do not run and do not plan to run a 'seasonal agnostic' railway this autumn, the concept of it is now firmly in sight for autumns of the future.