The residual risk section of a flood risk assessment

The residual flood risk section is the part of a flood risk assessment that explains what can still go wrong after everything you have proposed has been built and is working. It is not a restatement of the design event conclusions. If the section says "with finished floor levels set at 300 mm above the modelled 1 in 100 year plus climate change level, residual risk is low", it has answered nothing, and the lead local flood authority or the Environment Agency will say so.

Done properly, the section names each mechanism by which the site could flood despite the mitigation, states the consequence for people and property, and shows what manages it. That is a short list on most sites and a long one on a defended urban site next to a main river.

What residual risk actually means

Mitigation changes the probability or the consequence of flooding. It never removes either. Residual risk is what remains once the design has done its work, and it falls into three groups.

Design exceedance comes first: the event larger than the one the scheme was designed for, whether that is rainfall above the 1 in 100 year plus allowance or a river flow beyond the modelled extent. Second is failure, where the mitigation exists but does not perform, covering blocked culverts, jammed trash screens, pumps that stop and defences that breach. Third is the gap between assumption and reality, including a climate change allowance that proves too low, a third party asset nobody maintains, or an exceedance route across land the applicant does not control.

A Flood Risk Assessment that treats these three groups as headings and works through them is more persuasive than one that dedicates a paragraph to the phrase "residual risk" and moves on.

The sources to address explicitly

Which mechanisms apply depends on the site, and the assessment should say why each one is in or out of scope. On a typical defended site the list runs: breach of the defence, overtopping of the defence during an event above its standard of protection, blockage of a culvert or screen upstream, exceedance of the surface water system, failure of any pumped element, backing up of the public sewer, groundwater emergence, and flooding from a reservoir or canal where one is within range.

Rule out what does not apply, and give the reason. "There is no pumped drainage element, so mechanical failure is not a source of residual risk on this site" takes one line and shows you considered it. Silence reads as an omission.

Defence breach and overtopping

Where a site relies on a flood defence, the breach scenario is the one consultees look for first. A breach assessment considers where the defence could fail, what depth and velocity would result on the site, and how quickly the water would arrive.

The output that matters is hazard to people rather than extent on a plan. The Defra and Environment Agency flood hazard rating, used widely in breach work, combines depth and velocity with a debris factor, so that a shallow fast flow can present a greater hazard than deeper still water. Illustrative example: a breach producing 0.3 m depth at 1.0 m per second with a debris factor of 0.5 gives a hazard rating of 0.45, which sits in the lower hazard bands, while the same depth at 2.5 m per second takes it to 0.9. The point of the calculation is that the design response differs between those two cases, not that any single number is a pass mark.

Overtopping is the quieter sibling of breach and gets less attention than it deserves. A defence built to a 1 in 100 year standard is overtopped by the 1 in 200 year event even if it never fails structurally, and the assessment should state the standard of protection, where that figure came from, and what happens above it.

Blockage, exceedance and system failure

For surface water, residual risk is mostly about where the water goes when the system is full. The assessment should show an exceedance route on a drawing, not describe it in prose, and that route should lead away from doors, basements and neighbouring land.

Three checks carry most of the weight here.

1. Run the drainage model with the control blocked, partially and fully, and show where the resulting volume ponds. Work on modelling a partially blocked flow control sets out how the partial case is usually the one that governs.
2. Test the event above the design standard, commonly the 1 in 100 year plus climate change allowance with an additional urban creep factor, and record the depths in the ponding areas.
3. Confirm that overland flow at exceedance stays within the site or discharges to a defined route, with finished levels and kerb lines doing the work rather than a maintenance promise.

A Drainage Strategy and the flood risk assessment should agree on these numbers. Where they do not, and they often do not when different consultants produce them, the consultee reads the lower figure as the honest one.

Safe access and egress as risk management

Safe access and egress is where residual risk stops being a modelling exercise. The question the authority is asking is whether people can leave the site during a flood, or stay in it safely, without needing to be rescued.

Set out the route, its length, the depth and hazard rating along it during the relevant event, and who would use it. Dry pedestrian access at the design event is the comfortable answer. Where it cannot be achieved, the assessment has to make the case for a safe refuge instead, with floor levels above the flood level, an internal route to that refuge, and no sleeping accommodation below it.

Two things weaken this section reliably. Relying on emergency services to rescue occupants is not a flood risk management measure and will be rejected as one. And an egress route across third party land, with no legal right to use it, is worth nothing to an officer who has to defend the decision.

Linking the section to the evacuation strategy

The flood warning and evacuation plan is where the residual risk analysis turns into instructions for an occupier. The two documents should be written together, because the plan is only credible if it responds to the mechanisms the assessment identified.

A breach with 20 minutes of warning time needs a different plan from a fluvial event with 12 hours of Environment Agency flood alerts. Say which applies. Name the warning service the site can register for, the trigger for action, who holds responsibility, where people go, and how the arrangements pass to future occupiers through a management company or a condition. Where the scheme required an Exception Test, the safety limb of that test rests on this material, so vagueness here undermines the whole application.

Wording that answers the question being asked

Consultee responses on residual risk are usually a variation of one sentence: the assessment has not demonstrated that the development will be safe for its lifetime. Four habits head that off.

Quantify where you can and say so where you cannot. A depth and a hazard rating beat an adjective. Where no model output exists, state the limitation and the conservative assumption adopted instead of implying precision you do not have.

Tie every residual mechanism to a management measure and to the document or condition that secures it. Give the lifetime of the development explicitly, because the climate change allowances you apply depend on it, and the allowances are revised, so cite the version you used and the date you took it from.

Then write it for a reader who will be challenged at appeal. The residual risk section is the part of the assessment an inspector quotes when deciding whether the safety case was made, and a section that reads as a formality invites exactly that scrutiny.

You can draft an FRA structured to these sections and have the flood modelling and the safety case reviewed by a qualified professional before it goes in.

This is general guidance, not legal advice, and flood risk evidence requirements vary between lead local flood authorities. Check the position with your 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.