Sizing attenuation storage and the critical duration

A single design storm does not size an attenuation tank. Attenuation storage sizing depends on finding the storm duration that fills the storage hardest for a given return period, and that duration changes with the discharge rate, the contributing area and the shape of the system. A short intense burst overwhelms a small orifice. A long moderate event delivers more total volume. Which one governs is a question you answer by testing, and the answer is often not the duration anyone guessed.

Why one storm is not a design

Storage volume is the difference between what arrives and what leaves. Inflow depends on rainfall intensity and duration. Outflow is usually throttled to a fixed rate by a flow control, so it is near constant once the system is operating.

That means the critical duration is the point at which the gap between cumulative inflow and cumulative outflow is widest. On a site discharging at a high rate, short durations govern because the control empties the tank quickly between bursts. On a site throttled to a few litres per second, which is normal where a greenfield rate applies, long durations govern because the outflow barely keeps up.

So two sites with the same impermeable area and the same rainfall can have critical durations an order of magnitude apart. This is why borrowing a duration from a previous job produces a tank that is either undersized or absurd.

Finding the critical duration

The method is mechanical. Run the model for a range of durations at the design return period, record the storage required for each, and identify the duration producing the greatest volume.

Test a wide range. Fifteen and thirty minutes at the short end, through one, two, four, six and twelve hours, to twenty four hours and beyond where the discharge rate is very low. Long duration events measured in days matter on sites with volumetric control, and omitting them is a frequent reason a drainage design is rejected.

Three habits make the output defensible.

1. Present every duration tested rather than the critical one alone, so a reviewer can see the curve turning over.
2. State the rainfall dataset and the storm profile used, since both change the answer, as which rainfall dataset to cite in a drainage calculation sets out.
3. Repeat the exercise for each return period the design has to pass, because the critical duration is not the same for all of them.

If the storage requirement is still rising at the longest duration you tested, you have not found the critical duration. Extend the range.

The return periods the design has to pass

The Defra non-statutory technical standards for sustainable drainage systems and the local lead flood authority's own requirements set the events a design must accommodate. Three thresholds do most of the work.

The climate change uplift applies through the peak rainfall intensity allowances, which are published by catchment and are a separate set of figures from the peak river flow allowances used in flood risk assessment. Mixing the two is a recurring error.

Lead local flood authorities add their own requirements, and they vary. Some ask for interception of the first 5 mm of rainfall, some set a minimum discharge rate below which they will not require further restriction, and many have a standing position on infiltration testing. Ask before designing rather than after.

Half drain time, freeboard and recovery

Storage has to be empty again before the next event. The widely used benchmark is that attenuation storage should be half emptied within 24 hours of the end of the design event, and the calculation output should show the drain down curve rather than asserting it.

Freeboard matters on open features. A basin designed to the design water level with no allowance will overtop in reality, and 300 mm above the top design water level is the usual provision. On tanks the equivalent question is the level at which the system surcharges to ground and where that water goes.

Recovery is the point people forget on back-to-back events. Where a system takes 36 hours to empty and the design assumes it starts empty, the assessment is optimistic. Say so in the report where it applies.

Long term storage and volumetric control

Peak rate control alone does not deal with volume. Where a site discharges more total water than it did in its greenfield state, that extra volume has to go somewhere, and the standards address it through volumetric control or long term storage.

The usual approach compares the run-off volume from the developed site for a specified long duration event against the greenfield volume for the same event, and provides storage for the difference, released at a very low rate. That release rate is what drives the long durations into the critical duration test, and it is why a design can need a modest attenuation tank and a much larger long term storage component.

Infiltration removes the problem where ground conditions allow it, since water leaving the site vertically is not a volume to control. That makes infiltration testing worth doing early, to the recognised method, rather than assuming a rate. Where infiltration is not viable, the discharge arrangement itself needs designing properly, including the outfall, as outfall design, bank stability and scour protection describes.

Exceedance above the design event

Everything above the design event has to go somewhere, and the honest assumption is that it will. Exceedance routing shows where water flows when the system is full or blocked, and a scheme without it will be queried.

Route it deliberately. Roads and car parks as shallow temporary conveyance, levels set so overland flow passes between buildings rather than into them, thresholds raised where a flow path is unavoidable, and no low point without an outlet. Gardens and footpaths can hold water. Doorways cannot.

Blockage is a related scenario rather than the same one. A flow control partially obstructed behaves differently from an exceedance event, and the consequences are often worse because the system fails at a lower rainfall, which is explored in modelling a partially blocked flow control.

Output a reviewer can check

The approval decision is usually made by a drainage engineer reading a summary, so the summary is the deliverable. A Drainage Strategy that gets through first time generally contains the contributing areas with a plan showing them, the rainfall dataset and location, the climate change allowance with its source and date, the agreed discharge rate and its derivation, a full table of durations tested with the critical duration identified, the storage provided against the storage required for each design event, a drain down curve, the exceedance plan, and a maintenance schedule.

That is ten items. Most rejected submissions are missing three or four of them, and the common omission is the full duration table, which is the one piece of evidence showing the critical duration was found rather than assumed.

Drainage design is engineering work. A draft can structure the strategy, set out the standards and list what has to be demonstrated, and it cannot run the model or choose the discharge rate, so councils normally expect a suitably qualified engineer as the author.

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. For specialist technical reports, qualified professional input may be required. Enter your postcode on PlanScope to see whether your site needs a drainage strategy.