Surface water flood forecasting and early warning system for National Rail

Location: UK                                                                                                              Client: Network Rail


In July 2020 we were awarded funding to develop a first of a kind surface water flood forecasting and early warning system for Network Rail to support optimised railway operations through a real-time decision-support system.

The 2020 Rail innovation programme, funded by the Department for Transport and delivered by Innovate UK in partnership with the Knowledge Transfer Network, awarded £9.4 million to 25 outstanding projects that would make railways cleaner, greener and more passenger friendly.

Our challenge was to provide evidence that by combining our Flood Foresight technology with high resolution numerical weather prediction data, we could create a first of a kind technology that could be used by railway operators to accurately forecast the likelihood of surface water flood events.


Railway weather risk on the GB mainline is managed using passive systems, with data presented as hazards. A new system, making use of the best available high resolution rainfall forecasts, can improve the provision of risk information at asset level.

Network Rail has been a long-term client of ours and previously we have provided services including detailed mapping of locations such as earthworks, cuttings and embankments that were vulnerable to surface water hazards. By combining this asset level data with Flood Foresight probabilistic estimates of heavy rainfall and surface water accumulations ahead of time, we were able to create a system that produces consistent flood information on surface water events, alerts and warnings, nationwide.

Using this first of a kind technology to provide risk information across the network we’re also able to highlight any assets that may be partially or fully compromised. This means the Infrastructure Control Centre (ICC) (operational management) and Safety, Technical & Engineering (ST&E) teams (Asset Management) are able to identify if failures may be expected if inspection or maintenance is required.


Users of this technology will benefit from asset-specific surface water flood information, alerts, and warnings, in addition to being able to view surface water flood information in map format. Generated by integrating the forecast flood extent with existing asset metadata (location, elevation, asset type, etc.), this information can be used to prioritise early action to mitigate flood and heavy rainfall impact and minimise disruption. Ultimately, this will enable proactive flood management to reduce the overall impacts experienced by operators, maintainers (including planning by third party suppliers) and users of the national network. The resulting mapping outputs will also be available to inform where rail infrastructure might need priority attention ahead of flood events.

This first of a kind approach has several benefits:

  • It uses the best available rainfall forecast data that can resolve the type of rain that creates very intense, often short duration, rainfall events (convective rainfall cells)
  • It has a method of managing the uncertainty in the rainfall forecasts (multiple forecasts, referred to as a forecast ensemble, are used to create a probability of flooding map)
  • It uses existing, mature technology and data to map the surface water flood likelihood ahead of the event and link the mapped surface water flood information to Network Rail’s assets to assess flood risk in real-time.

By utilising high-resolution rainfall forecasts for the UK and Ireland and drawing on products and knowledge from the suite of other projects that Network Rail has been developing, there is also a unique opportunity to combine previous and current investments. Ultimately allowing Network Rail to drive the use of these projects and processes through Control Period (CP) 6 (2019-2024) and develop a more sustainable and reliable rail network.

During the term of this project, in response to the tragic August 2020 Carmont derailment, Network Rail established a Weather Advisory Taskforce (WATF) to identify improvements in how they manage weather-related risks. The WATF strongly recommend the development, trialling and integration of state-of-the-art ensemble weather prediction models into Network Rail for management of the UK’s railways. The findings of our project are consistent with these recommendations and this project provides a demonstration of how these models can be merged with flood hazard data to provide users with asset-specific information on damaging events ahead of their occurrence. We will continue to engage with Network Rail and the WATF to further validate our method and trial its adoption in an operational context.

Beyond this project, this new heavy rainfall and surface water capability within Flood Foresight can also be applied to other sectors where asset-specific alerting can help inform proactive interventions to reduce impacts of surface water flooding, including local authorities, highways, (re)insurance and utilities.

Want to know more?

For more information about this project please contact Andrew Gubbin or John Bevington.

Visit the Flood Foresight webpage for more on details on effective flood management.

Leave a Reply