Understanding and Managing Leaf Fall Patterns

‘Leaves on the line' are a key cause of low adhesion in autumn: what options do we have for managing leaf fall?

Rail leaf contamination can have potentially serious consequences where slipping trains pass designated stopping positions in stations or at signals. It can also affect the traction of trains on the track, resulting in problems with acceleration that end up disrupting service timetables and causing passenger delays. More accurate information about when leaves are expected to fall and the extent of leaf-fall at any given time could help industry to take actions, improving safety and reduce disruption and delays.

We have funded a PhD at the University of Birmingham for the development of a low-cost imaging solution that can be mounted on trains and could provide train operators and infrastructure managers with real-time route specific information about leaf fall. This innovative new device uses low-cost imaging technology – a fish-eye lens which is typically used in smartphones and a Raspberry Pi (a single board computer) – to measure and analyse differences in the spectral reflectance of trees as they move from full canopies to bare trees. The work has been developed in close cooperation with the Met Office to enable the inclusion of the data in their high-resolution adhesion forecast models.  

A complementary project is investigating the effect that weather has on the leaf-fall patterns of different trees and the entrainment of leaves between the wheels and the track. This work, due to be completed in 2022, will also look at the fundamental biochemistry of leaves and the leaf film they produce, with the aim of developing an improved adhesion prediction index. 


High resolution leaf fall monitoring and low adhesion forecasting (COF-E14-02)
Haven’t found what you’re looking for?
Get in touch with our expert for more information
Paul Gray
Tel: 020 3142 5602
Cookies help us improve your website experience.
By using our website, you agree to our use of cookies.