Interception in a sustainable drainage design

Interception suds design means holding back the first few millimetres of rainfall so that small storms produce no discharge from the site at all. The SuDS Manual, CIRIA C753, sets the working target at the first 5mm of a rainfall event. It is a separate requirement from attenuation, it is assessed differently, and a drainage strategy can pass every peak flow calculation and still fail on it.

Plenty do.

What interception means hydraulically

Attenuation is about the peak. You store water and release it slowly, so the outflow hydrograph is flattened and the receiving system is not overwhelmed. Interception is about the small events at the other end of the scale, the ones that never trouble a peak calculation.

On an undeveloped site, light rain soaks in, is intercepted by vegetation or evaporates. Almost nothing reaches the watercourse. Pave the same site and every millimetre runs straight off. The total volume is modest, but the frequency is high and the water is dirty, because it has picked up whatever was sitting on the surface since the last shower.

Interception restores that behaviour. Rainfall up to the target depth is absorbed, stored in a substrate, evaporated or infiltrated, and no discharge leaves the site. Above the target depth the system behaves like any other, routing flow to attenuation and then to the outfall.

Why the first few millimetres matter

Two reasons, and the water quality one is stronger than the volume one.

The first flush from a paved surface carries the highest concentration of hydrocarbons, sediment, metals and whatever else has accumulated. Removing that flush from the discharge entirely does more for the receiving water than treating a larger volume less thoroughly. Lead local flood authorities and the Environment Agency both understand this, which is why interception appears in local design guides even though it has no place in the national non-statutory technical standards.

The second reason is frequency. Most rainfall events in the UK are small. A site that discharges on almost every day of rain has a very different relationship with its receiving watercourse from one that discharges a few dozen times a year. Baseflow, channel erosion and the ecology of a small stream all respond to that frequency rather than to the annual total.

Features that deliver it

Interception has to happen at or near the source, before the water enters a pipe. Below ground tanks with an outlet do nothing for it, however large they are.

• Green roofs, where the substrate holds the early rainfall and releases it to the atmosphere. A modest substrate depth handles the target depth comfortably.
• Permeable paving with a granular sub-base, where the early rainfall is held in the voids and lost to evaporation and infiltration between events.
• Rain gardens, bioretention areas and tree pits, where the growing medium absorbs the first rainfall before any underdrain is reached.
• Filter strips and grassed swales with check dams, where the depression storage and infiltration on the surface take the small events.
• Soakaways and infiltration trenches on sites where infiltration is viable at all.
• Water butts and rainwater harvesting, though only partially, because the tank has to be empty when the rain arrives.

Note: the last point is the one designers argue about. A harvesting tank contributes to interception only to the extent that demand empties it reliably between events, and a demand profile that assumes summer garden watering in February is not credible.

Showing that it has been achieved

Two routes are accepted in practice. The first is calculation, modelling the rainfall loss through the proposed features and showing that no discharge occurs for an event of the target depth. This is normal on larger schemes where a hydraulic model already exists.

The second is a design based demonstration. Where recognised source control features serve the whole of the contributing impermeable area, and where their storage above any underdrain exceeds the target depth applied to that area, interception can be taken as achieved without a separate model run. Set out the arithmetic anyway. A single line showing contributing area, target depth, resulting volume and available storage at source is enough for a reviewer to check in a minute.

The failure mode is partial coverage. A scheme with green roofs on two of five blocks and conventional drainage elsewhere has achieved interception for part of its area and nothing for the rest. Say which part, rather than claiming the principle has been applied.

Where it sits in the treatment train

Interception is the first stage of the treatment train and it is not a substitute for the rest of it. The Simple Index Approach in the SuDS Manual still requires the pollution hazard of the contributing surface to be matched by the mitigation indices of the components provided, and a green roof over a car park roof deck does not answer the hazard from the parking bays below.

Where interception is delivered by a component that also provides treatment, such as a bioretention area, the same feature can count in both assessments. Where it is delivered by evaporation from a green roof, the runoff from the paved areas still needs its own treatment stages. Keep the two assessments separate in the Drainage Strategy so a reviewer can follow each one.

Illustrative example: a 1.1 hectare residential scheme with 0.45 hectares of roof, 0.25 hectares of permeable paved parking and 0.12 hectares of adopted road. The roofs discharge to rain gardens with 150mm of growing medium above the underdrain. The parking is permeable with a 350mm sub-base. The road drains to a swale. Interception is achieved across the roof and parking areas by design, and the swale handles the road, so the whole contributing area is covered and the strategy says so in one short table.

Sites where it cannot be delivered

Some sites cannot deliver interception, and the honest answer is to say so and compensate rather than assert compliance.

Infiltration is restricted where the ground is contaminated and the water would mobilise pollutants, inside a groundwater source protection zone where the Environment Agency objects, where groundwater sits close to the surface, or where infiltration would destabilise a slope or affect a neighbouring basement. Structural constraints bite on podium decks and over basements, where substrate depth and loading limit what can be built. Heritage constraints bite on listed buildings where a green roof would alter the fabric or the roofscape.

Where interception is not achievable across part of the site, add treatment stages instead, and say which mechanism is being substituted for which benefit. A Green Infrastructure Plan can help carry the amenity and biodiversity side of the argument where the hydraulic side is constrained.

Recording it in the drainage strategy

A reviewer looking for interception wants five things in one place: the target depth applied and its source, the contributing impermeable area broken down by surface type, the features providing interception for each area, the storage available above any underdrain, and the maintenance that keeps it working.

That last item is the one that fails in service. Permeable paving clogs if it is never swept. A green roof substrate compacts and loses storage if the planting dies. Rain garden inlets silt up. A maintenance schedule with frequencies and a named responsible body is part of the interception case, not an appendix to it. Link it to the wider residual risk narrative in the Flood Risk Assessment where one is required, so the two documents describe the same system.

PlanScope drafts a Drainage Strategy structured to UK planning policy from your site data and guided answers, and flags site-specific figures such as infiltration rates and substrate depths as still to be confirmed rather than filling them in. Drainage design needs a suitably qualified author, and for specialist technical reports, qualified professional input may be required. This is general guidance rather than 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.