Half drain time and consecutive storm testing

Half drain time is the time an attenuation feature takes to fall from full to half its stored volume once inflow has stopped. It is a recovery measure, not a capacity measure, and it answers a question that peak storage volume cannot: if a second storm arrives tomorrow, how much of the tank is available to receive it.

Reviewers who check volume and outflow rate and stop there are checking half the design. A system sized correctly for a single event can still flood on the second of two, and the model output will show it if anyone looks.

What half drain time measures

The convention across most guidance is that attenuation storage should recover to half its volume within 24 hours of the end of the design event. For infiltration systems, the long-standing soakaway design method in BRE Digest 365 uses the same 24 hour half-emptying criterion, and that figure has carried across into wider practice.

The number matters less than what it stands for. A tank that takes four days to empty is unavailable for most of the week in a wet autumn, and the design event it was sized for is no longer the worst case it will see.

Half drain time is reported by the model, not calculated by hand, and it varies with the outflow control. A vortex control that throttles hard at low head will give a long tail, which is exactly the behaviour that produces a poor half drain time while meeting the peak discharge requirement comfortably.

Why recovery between storms matters

Design storms are single events. Real rainfall arrives in sequences, and the sequences that cause flooding are usually two or three moderate events in close succession rather than one extreme one.

The mechanism is simple. The first event fills the storage. The outflow is restricted to the agreed rate, often a small fraction of the greenfield equivalent on a constrained site. The second event arrives with the tank still part full, the available volume is less than designed, and the system surcharges at a return period well below the one it was tested at.

Autumn and winter are when this happens, and winter design storm profiles with their longer durations and lower intensities are the ones that expose it. A system that performs in a summer 30 minute profile and fails in a winter 24 hour profile has been tested in the wrong season.

Consecutive and nested storm testing

Three tests are worth running, and most models will do all of them.

1. The full range of durations for each return period, from the short intense events through to the long duration events, to identify the critical duration rather than assuming it.
2. Summer and winter profiles for the critical duration, since the critical case often differs between them.
3. A consecutive event test: the design event followed by a second event after a defined dry period, with storage carried over.

The third is the one that gets skipped. There is no single prescribed interval, and the practical approach is to run the design event followed by a repeat after 6, 12 and 24 hours, and report the available storage at the start of each second event.

Nested storms, where a short intense burst sits inside a longer event, are a related test and matter more where the system has two stages with different response times, such as permeable paving discharging to a downstream basin.

Where the receiving system is itself surcharged during the second event, the outfall may not discharge at the assumed rate at all. On sites draining to a combined sewer, ask the undertaker whether tailwater conditions should be modelled rather than assuming free discharge, and say in the Drainage Strategy what was assumed.

Where a long drain down is defensible

Not every long recovery time is a fault. Three cases come up regularly.

Where the permitted discharge rate is set at the minimum practicable rate for the control device, the drain down time follows arithmetically from the storage volume and there is nothing the designer can do about it short of reducing impermeable area. Say that explicitly rather than letting the reviewer find it.

Where the feature has a permanent water body, such as a wetland or a retention pond, the half drain time applies to the attenuation volume above the permanent level, not to the whole feature. Reports that conflate the two produce alarming numbers for a system behaving perfectly well.

Where the system relies on infiltration into low permeability ground, recovery is slow by nature and the design should carry a risk assessment of the consequences of a partly full system rather than a claim that it recovers quickly.

In each case the test is whether the consequences of the reduced availability have been assessed, not whether the number meets a threshold.

Infiltration and pumped systems behave differently

For infiltration, the governing variable is the infiltration coefficient and the safety factor applied to it. A design using an unfactored field test result will show an attractive drain down time and will not perform. Check that the test method and the number of tests match the guidance, and that a factor of safety appropriate to the consequences has been applied.

Groundwater level is the other input. A soakaway with its base within a metre or so of the seasonal high groundwater level will not empty in winter regardless of the summer test result, which is the same issue that arises with groundwater flooding evidence.

For pumped systems, recovery depends on the pump running. The reviewer should look for the storage provided for pump failure, whether duty and standby pumps are specified, the power supply arrangements, and who maintains the station. A pumped outfall with a single pump and no emergency storage is a system with a half drain time of infinity on the day it fails.

The checks a reviewer should run

Working through a submitted model, these are the points that produce findings:

• The reported half drain time for each return period, including the climate change scenarios, not just the 1 in 100 event.
• The range of durations tested and whether the critical duration sits at the edge of the range, which means the range was too narrow.
• Winter as well as summer profiles.
• Available storage at the start of any consecutive event test.
• The outflow control curve at low head, since many controls discharge far less than their nominal rate in the last part of the drain down.
• Whether the reported volume includes freeboard, filter media voids or permanent water, all of which inflate it.
• Whether the maintenance regime is compatible with the recovery assumed, because a silted filter drain does not behave like a clean one.

That last point is where design and adoption meet. A system whose performance depends on annual jetting needs someone obliged to do the jetting.

Reporting the result so it can be checked

State the half drain time in a table alongside peak storage and peak outflow for each event modelled, with the consecutive event results in the same table. Include the time series for the critical case as a figure. Give the model version, the rainfall dataset and the urban creep and climate change allowances applied.

Then write one paragraph saying what the numbers mean for the site. A reviewer can read a model output. What they cannot do is guess whether the designer understood it.

For smaller schemes the same logic applies at a smaller scale, as the note on drainage evidence for a single new dwelling sets out, and the drainage position feeds directly into the Flood Risk Assessment where one is required.

Browse the 66 PlanScope report types to see how the drainage and flood documents fit together.

PlanScope provides AI-assisted draft reports for planning applications. Reports should be reviewed before submission. We do not guarantee planning approval.