Residual risk after a defence: breach and overtopping

A site behind a raised flood defence is still in the flood zone, because the published flood zones ignore the presence of defences. So the sequential and exception tests still apply, and the flood risk assessment has to address what happens if the defence fails or is overtopped. That residual risk case is what the Environment Agency reviews most closely on defended sites, and an assessment that says the defence provides a 1 in 100 year standard of protection and stops there will attract an objection.

Two failure mechanisms have to be considered, and they produce very different results. Overtopping, where the water level exceeds the crest and flow enters gradually along a length of defence. And breach, where the structure fails and a concentrated flow enters through an opening.

Why a defended site still needs assessment

Defences reduce the frequency of flooding rather than removing it. They have a design standard, an asset condition that changes, an owner who may or may not maintain them, and a finite life against rising design events.

Three consequences follow for a planning application. The zone remains as mapped, which fixes the vulnerability test that applies. The consequence of failure behind a defence is often more severe than on an undefended floodplain, because depths build in a confined area with little warning. And the lifetime of a residential development, taken as 100 years, is longer than the design horizon of most defences.

That last point is the one developers find hardest. A defence built to a 1 in 100 standard today will provide a lower standard in 60 years unless it is raised, and the Environment Agency has permissive powers rather than duties to maintain its assets.

Breach location and the modelling assumptions

Breach modelling is a two-dimensional exercise, and the assumptions matter more than the software. Every one of them should be stated and agreed with the Environment Agency before the model is run.

1. Breach location, usually chosen as the credible worst case for the site rather than the most likely point, often directly upstream or opposite the development.
2. Breach width, agreed with the Agency and typically measured in tens of metres for an embankment, with the invert level stated.
3. Timing of the breach, commonly at or shortly before the peak water level, which produces the greatest inflow volume.
4. The return period of the event during which the breach occurs, including the climate change allowance for the relevant epoch.
5. Ground model, including any raised ground, walls, buildings and the road network, which control where water goes.
6. Duration of the simulation, long enough to show the maximum depth at the site and the drain-down.

Report depth, velocity and the resulting hazard at the building thresholds, on the access route and across the site, as a map rather than as a table of numbers at a few points.

Overtopping against structural failure

Overtopping is usually modelled as a discharge over the crest for the duration that the water level exceeds it. On a tidal frontage that may be a short period around high water, and the volume can be limited. On a fluvial defence during a long duration event it can be substantial.

Assess both mechanisms and present them separately. A common shortcut is to model breach only, on the basis that it is the worst case, which leaves the reviewer unable to judge the more likely event.

Standard of protection and its expiry

Set out what is known about the defence: who owns it, its crest level to a stated datum, the standard of protection it currently provides, its recorded condition grade, whether it is in any published maintenance or capital programme, and whether it is designated under the relevant legislation so that alterations need consent.

Then state what happens to that standard across the development lifetime with the climate change allowances applied. Where the standard falls below the design event within the lifetime, the residual risk case becomes the primary case, and the application has to be built on the breach and overtopping results rather than on the present-day protection.

Do not assume a future defence improvement unless it is funded and programmed. An aspiration in a strategy document carries very little weight.

Rate of onset, warning time and evacuation

Hazard is a function of depth and velocity, and the recognised approach combines them with a debris factor to produce a hazard rating with categories running from low hazard through to danger for all. Apply it to the escape route as well as to the building.

Then address time. State whether the site is covered by the flood warning service, the likely warning lead time, how long evacuation of the proposed number of occupants would take, and where occupants go if the route floods first. Safe refuge above the modelled breach level inside the building is acceptable in some cases and unacceptable for vulnerable uses.

Flood warning and evacuation plans are the developer's responsibility, and the Environment Agency will not approve them. Where the local authority's emergency planning team is willing to comment, get that comment, because it carries weight that a consultant's assertion does not.

Mitigation that addresses residual risk

The measures that work are physical and hierarchical. Set finished floor levels above the modelled breach level where it can be done, with freeboard stated. Keep sleeping accommodation out of areas with a high hazard rating. Design safe refuge and an internal route to it. Specify flood resilient construction to the recognised standard for the ground floor where water entry is credible. Provide an emergency access route at a level that stays passable, or demonstrate a safe refuge strategy instead. And keep basements out of the scheme, as basement development and the flood risk objection explains.

Land raising can help and can harm. Raising within the defended area displaces water onto neighbours during a breach, so any raising needs compensation or a demonstration that the effect elsewhere is acceptable.

How the Environment Agency reviews the scenarios

Expect the review to check the assumptions first: breach location and width, timing, the event modelled, the allowance applied and the ground model. Then the outputs: depth, velocity and hazard at thresholds and on the access. Then the response: levels, layout, refuge, resilience and the evacuation position.

Where the site needs the exception test, the safety limb is built from exactly this evidence, and exception test evidence: the two limbs in practice sets out how it is presented. On coastal and estuarine frontages the mechanisms differ again, and tidal flood risk: breach modelling and coastal change covers those specifics.

Scope the modelling with the Agency before commissioning it, record the agreement in the Flood Risk Assessment, and expect the whole exercise to take months rather than weeks on a defended site.

For specialist technical reports, qualified professional input may be required. This is general guidance, not legal advice; check the position with your local planning authority or a qualified professional before relying on it. PlanScope provides AI-assisted draft reports for planning applications. Reports should be reviewed before submission. We do not guarantee planning approval.