CFD modelling shows where bridge scour risk really develops

Computational fluid dynamics modelling is being used to improve bridge scour assessment across the rail network, helping engineers identify where erosion risk is actually developing around bridge foundations during flood events.

Bridge scour is the removal of sand, gravel and soil from around piers, abutments and foundations by fast-moving water and complex currents. It is one of the most serious risks for bridges over water and is widely recognised as a leading cause of bridge collapse. In Europe, around 500 scour-related incidents have been recorded over the past century, and the risk is expected to grow as more intense rainfall and flooding affect ageing infrastructure.

For many years, scour protection has often relied on conservative assessment methods. Where uncertainty exists, engineers may specify deep protection, heavy rock armour or concrete mattresses to cover both known and unknown risks. That cautious approach has value, but it can also lead to inflated risk ratings and expensive physical works at structures that may not need such extensive intervention.

Traditional scour assessment commonly uses 1D and 2D hydraulic models. A 1D model estimates average flow conditions along a river channel, while a 2D model allows depth-averaged flow to vary across a wider grid, such as a floodplain. These tools are useful, but they can miss the local three-dimensional flow behaviour that controls scour around real bridge geometry.

CFD modelling goes further by solving more complex fluid-flow equations and simulating how water moves at different depths, directions and velocities. It can represent turbulence, flow separation, vortices, superelevation and local acceleration around piers and abutments.

This matters because scour is often governed by local flow mechanisms, not only by average river velocity. During a flood, water may spread across flood relief arches, change direction, accelerate around bridge elements and attack a foundation from a location that is not obvious in a simplified model.

CFD can also include sediment behaviour. By using parameters such as the Shields parameter, engineers can assess when particles of different sizes are likely to be lifted and transported. This helps model how a scour hole develops, deepens and eventually becomes self-limiting as the exposed material changes with depth.

A case study at Shrewsbury railway station bridge over the River Severn shows why this matters. A conventional Stage 2 scour assessment using a 1D model predicted scour in one area near the middle of the bridge, with a depth of around 12 m.

The CFD model indicated a different result. It showed the critical scour in another location and at a much lower depth, slightly above 4 m for the assessed flood event. The important lesson was not only that the predicted depth was lower. It was that the conventional approach could have directed protection works to the wrong part of the structure.

This is where CFD becomes valuable for asset management. It can help engineers define the actual risk zone, foundation exposure, likely scour depth, armour stone extent and toe protection requirements with greater confidence.

Network Rail has incorporated CFD as an advanced three-dimensional modelling tier within its scour assessment standard NR/L2/CIV/295. Around 30 to 40 structures have reportedly been modelled so far, with more expected.

The potential benefit is significant. Scour protection is expensive, especially where bridges are remote, difficult to access or located within active river environments. Temporary access tracks, plant mobilisation, river working constraints and environmental controls can make physical intervention costly even before permanent protection is installed.

More accurate modelling does not remove the need for conservative engineering judgement. It helps place that judgement where it is most needed. If CFD confirms that a structure is genuinely at risk, the case for intervention becomes stronger. If it shows that the risk is lower or located elsewhere, resources can be redirected to assets that need protection more urgently.

CFD also helps test whether proposed works could increase flood risk, disturb flow paths, affect ecology or transfer risk downstream. That wider catchment view is becoming increasingly important for regulators and infrastructure owners.

Scour resilience depends on understanding the river, the foundation and the local flow field together. Better modelling can reduce uncertainty, avoid unnecessary works and support safer, more targeted protection of bridges exposed to future flood events.