Scour is the removal of bed material from around a structure by moving water. It is the single most common cause of bridge failure in the United States, and it has a property that makes it unusually dangerous: it does most of its damage at the peak of a flood, underwater, where nobody can see it, and then it frequently fills back in as the flow drops.
A bridge that was nearly lost on Tuesday can look completely normal on Friday.
Scour is not the same as erosion
The two words get used interchangeably and the distinction matters, because the treatments differ.
Erosion is the general wearing away of a surface by flowing water, wind, or wave action. It is a broad, distributed process. A streambank losing material along its length is eroding.
Scour is localised, and it is caused by a structure. Put a pier in a river and you change the flow field around it. Water accelerates past the obstruction, turbulence increases, and the bed immediately around that pier is excavated in a way the rest of the channel is not.
Every scour problem is an erosion problem, but the reverse is not true, and a treatment designed for distributed bank erosion will not necessarily address a hole developing at a foundation.
The three components
Engineers evaluating a bridge do not compute a single number called “scour.” They separate it into three components that have different causes and different timescales. FHWA’s HEC-18, Evaluating Scour at Bridges, is the standard reference for this in the US.
1. Long-term aggradation and degradation
The channel bed as a whole rising or falling over years or decades. Causes include upstream dam construction, changes in sediment supply, gravel extraction, channel straightening, and land use change in the watershed.
This one is slow and easy to miss because no single event causes it. It is also the component that quietly reduces the margin on everything else: a bed that has degraded two feet over thirty years has given away two feet of foundation cover before any flood arrives.
2. Contraction scour
When a channel narrows, the same discharge has to pass through a smaller opening. Velocity increases, the flow’s capacity to carry sediment increases, and the bed lowers across the full width of the contracted section.
Bridges cause this routinely, because abutments and approach embankments narrow the floodplain. It affects the whole opening rather than one location.
3. Local scour
The hole immediately around a pier or abutment, caused by the structure’s own interference with the flow. This is what most people picture when they hear the word, and it is usually the deepest of the three at any given foundation.
The three are additive. Total scour at a foundation is degradation plus contraction plus local, which is why a bridge can be in trouble even when no single mechanism looks alarming on its own.
Why local scour happens at a pier
Flow arriving at a pier is stopped at the face. That stagnation drives a downward current along the upstream face of the pier, which strikes the bed and rolls into a vortex that wraps around the base like a horseshoe — this is the horseshoe vortex, and it is the primary excavating mechanism.
The vortex lifts bed material and the flow carries it away. The hole deepens, which enlarges the vortex, which deepens the hole. It is self-reinforcing until the geometry stabilises or the flow drops.
Downstream of the pier, separate wake vortices spin off either side and lift material as well, which is why the scour signature usually extends past the structure.
Bridge pier scour covers this mechanism and its consequences for layout in more detail.
Why it is so hard to see
This is the property that catches people out, and it is worth stating plainly.
Scour depth is greatest at or near the peak of the hydrograph, when velocity and turbulence are highest. As the flood recedes, the transport capacity of the flow drops, and the sediment being carried settles. Much of it settles into the hole that was just excavated.
The consequence is that an inspection carried out after the water clears may find a bed that looks essentially undisturbed, over a foundation that was briefly exposed to a scour hole several feet deep. The material now sitting in that hole is loose, uncompacted infill with none of the strength of the undisturbed bed it replaced.
This is why underwater inspection findings, fixed monitoring instruments, and scour evaluations based on hydraulic modelling matter more than a visual look from the bank after an event. What you can see is not the condition that governs.
What scour actually does to a structure
Scour does not usually fail a bridge by washing it away. It fails it by removing support:
- Loss of embedment. A pile or a spread footing depends on the material around and beneath it. Take that away and the foundation’s capacity drops.
- Undermining. A footing with a void beneath one edge is a footing that will rotate.
- Loss of lateral support. Even where vertical capacity remains, a pile with several feet of newly unsupported length is a different structural element from the one that was designed.
- Progressive settlement. Partial support loss produces movement, movement opens joints, and the structure redistributes load in ways it was not designed for.
The failure, when it comes, is often sudden, because the structure holds until it does not.
Where scour shows up besides bridges
The mechanism is the same wherever flow meets an obstruction or a discontinuity:
- Culvert outlets, where a concentrated jet exits into a channel and excavates a plunge pool that then works back toward the structure. See culvert outfall scour.
- Bulkhead and sheet pile toes, where a vertical face reflects wave energy downward into the bed at its own base.
- Berths and quays, where propeller wash from manoeuvring vessels excavates the bed in front of the structure.
- Marine piles, dolphins, and fenders, subject to the same vortex mechanisms as a bridge pier in a current.
- Subsea pipelines and cables, where flow around and under a line develops free spans.
- Grade control structures and low-head weirs, where the drop concentrates energy at the downstream toe.
What you do about it
Broadly, three approaches, and real projects often use more than one:
- Armor the bed so the flow cannot lift the material. Riprap, bagged armor, articulated block mats, grout bags, and gabions all belong here.
- Modify the flow so the excavating mechanism is weakened. Collars, sacrificial piles, vanes, and guide banks work this way.
- Design past it. Found the structure below the computed scour depth so that losing that material does not matter.
The third is the most robust and is usually only available at design stage. The first is what most existing structures get, because the foundation is already where it is.
The full menu of scour countermeasures covers the families and where each one fits, and when not to use rock bags is where we set out the cases our own product does not suit.
Where to go next
- Contraction, local, and degradation scour separates the three components properly
- Why scour protection fails covers winnowing, edge failure, and undermining of the armor itself
- Bridges and structures covers how the work gets staged over water
- The scour protection hub collects the rest
If you have a structure you are worried about, site photographs, whatever survey exists, and the flow conditions are enough to start a conversation. Send them through the quote form.