Infiltration testing to BRE Digest 365 in practice

An infiltration test to BRE Digest 365 is three fills of a trial pit, each timed as the water level falls from 75 per cent to 25 per cent of the effective storage depth, with the design value taken from the slowest of the three. That is the whole method, and it is the number a lead local flood authority will look for before it accepts a soakaway on a minor residential scheme. What goes wrong is rarely the arithmetic. It is a pit dug in the wrong place, at the wrong depth, in August, with one fill instead of three.

Get the fieldwork right and the calculation takes ten minutes.

The test method and the three fills

Dig the trial pit to the depth of the proposed soakaway base. Fill it rapidly with water to the full effective depth, then record the time taken for the level to fall from 75 per cent to 25 per cent of that depth. Refill and repeat, twice more, without a long gap between fills.

The repetition is the point. The first fill wets up the soil and often drains faster than the ground will manage in a wet winter, so the Digest asks for three and directs you to design on the slowest result. A single test, reported as the answer, is the most common reason a drainage strategy comes back with a request for further information.

Measure from a fixed datum across a lid or board, not from the ground surface, because the sides slump. Record the fill times, the water volume used, the start and finish levels and the times for all three fills, even the ones you would rather not show. Photographs of the open pit with a staff in it are worth including.

Two practical warnings. A tanker discharging straight into a pit in gravel can empty faster than you can read the level, so plan the measurement interval before the water goes in. And in clay the 75 to 25 fall can take longer than a working day, which means a logger or an overnight return visit rather than an estimate.

Trial pit depth and finding the water table

The pit goes to the proposed base of the soakaway, because the soil that matters is the soil the structure will sit in. Testing a shallow pit in permeable topsoil and designing a soakaway 2.5 metres down into clay gives a number that is worse than useless.

The Digest expects the base of the soakaway to sit clear of the water table, with a working margin of at least a metre. That means the trial pit has to establish where the water table is, not merely confirm that the pit was dry on the day. Leave the pit open, covered for safety, and return to record any standing water after 24 hours. Where the site has a history of wet ground, a monitored standpipe over several weeks is the only honest answer, and the same evidence supports the flood risk work described in evidencing groundwater flooding in a flood risk assessment.

British Geological Survey mapping will tell you what to expect before you hire a machine. River terrace gravels over clay, made ground of unknown depth, or a chalk site with a seasonally high water table each need a different testing plan.

Testing season and why winter matters

Infiltration in the same pit varies through the year, and it varies in the direction that matters. A test in late summer on a dried clay with open shrinkage cracks can return a rate an order of magnitude better than the same pit in February.

Reviewers know this. A report dated August, with no groundwater monitoring and no seasonal caveat, invites the question of whether the result would survive a winter. Where the programme allows it, test between November and March. Where it does not, say so plainly, apply a larger factor of safety and offer a condition requiring confirmatory testing before construction.

Working out the infiltration coefficient

The soil infiltration rate is the volume of water lost between the 75 per cent and 25 per cent levels, divided by the product of the internal surface area of the pit up to half the effective storage depth and the time taken. The Digest defines that area term precisely, including the base, so take the definition from the Digest itself rather than from a spreadsheet you inherited.

Illustrative example, and it is illustrative only. A pit 1.5 metres long by 0.5 metres wide with an effective depth of 1.0 metre loses 0.375 cubic metres of water as the level falls from 0.75 metres to 0.25 metres. The area term comes to 2.75 square metres. If the slowest of the three fills takes one hour, the rate is roughly 3.8 x 10-5 metres per second, which is a sandy gravel sort of answer.

Then apply a factor of safety. The SuDS Manual, CIRIA C753, sets out factors that rise with the consequence of failure and the confidence in the data, and a single pit tested once in summer sits at the cautious end of that range. Quoting a raw measured rate as the design rate is the second most common review comment after the single-fill problem.

How many tests, and where

One pit tells you about one place. Infiltration varies over a few metres in made ground and over tens of metres in natural deposits, so the number of tests follows the number of proposed structures and the variability of the geology.

A workable rule for a small site:

1. One trial pit at the location of each proposed soakaway, tested three times.
2. An additional pit where the geology mapping shows a boundary crossing the site.
3. Repeat testing where results between pits differ by more than a factor of about two, because at that point you do not know which one governs the design.
4. No extrapolation across a made ground boundary, ever.

Larger sites and adoptable systems attract more. Where the design relies on infiltration for the whole of the surface water discharge, the evidence base needs to be proportionate to that reliance.

When infiltration fails

Plenty of sites will not infiltrate, and discovering that early is a good outcome rather than a bad one. A rate below roughly 10-6 metres per second, a water table within a metre of the proposed base, or contaminated ground where infiltration would mobilise pollutants all point away from soakaways.

The design then moves to attenuation with a controlled discharge: a tank, permeable paving with a lined sub-base, a basin or a blue roof, releasing at the greenfield rate or at a rate agreed with the sewerage undertaker. Discharge hierarchy still applies, so the strategy has to show that infiltration was tested and rejected on evidence before a connection to a sewer is proposed. Where there is no mains connection at all the foul side needs its own treatment, and foul drainage with no mains sewer covers what to show.

Note that percolation testing for a foul drainage field under Approved Document H is a different test with a different pit, a different fill and a different result. The two are not interchangeable, and submitting one in place of the other is a validation problem rather than a technical one.

Reporting the result so it is accepted

The report needs to let a reviewer reproduce the number without ringing you. That means a site plan with the pit locations marked and levelled, a log of each pit showing the strata and the depth, a table of all three fills with times and levels, the calculation with the area term shown, the factor of safety applied and why, and the resulting design rate.

Add the groundwater observation and the date of testing. Add a one-line statement of what you would do if confirmatory testing at construction returned a worse rate, because that is the question a drainage engineer will ask at the discharge of conditions stage.

Where infiltration is accepted, expect a condition requiring as-built verification. Offer the wording yourself rather than waiting for a version you cannot meet.

You can draft the narrative of a Drainage Strategy or a Flood Risk Assessment in PlanScope, with the test results and calculations dropped into the right sections, then send it for professional review. PlanScope provides AI-assisted draft reports for planning applications. Reports should be reviewed before submission. We do not guarantee planning approval. This is general guidance, not legal advice; check the position with your lead local flood authority or a qualified professional before relying on it.