BRE365 Soakaway Testing; Discussion on Safety and Alternatives

- by
Tags: Featured

Since 2007 when the first SuDS manual was published, AGS members have increasingly been asked to provide services to support design of sustainable urban drainage systems. The author of this article has been working with the BRE365 test since the 1990s and is still supervising, designing and reporting infiltration rate results to designers from trial pit and borehole methods.

AGS has recently been investigating how our industry manages undertaking and reporting infiltration rates for design of devices for discharge of surface water into land including; soakaways, basins, swales and permeable paving. Amongst members, the almost exclusive method requested by clients is to follow the BRE DG365 procedure as discussed in the AGS article (ref AGS magazine Oct 2021).

BRE DG365 is an empirical methodology for the design of soakaways, which includes the test procedure to provide observational data on infiltration rates. It was not intended for either basins or permeable paving. Because of the sparse nature of the test instructions, how it is carried out and reported can vary between practitioners. Concerns were first raised in the AGS safety working group at the same time as general trial pitting methods came under renewed scrutiny. The geotechnical working group was also requested to look at data quality in context, which resulted in the previous article. It is recognised that AGS members’ risk assessments and method statements on trial pitting may have been extended to cover BRE365 by written procedures to keep inexperienced staff safe in conducting these in-situ tests. Innovations and modifications may also have been made to the BRE365 method, to standardise in-house practice and to avoid taking up excessive resources during what can be a logistical challenge of labour, plant and materials.

This article is intended to build on the previous article, briefly cover the test origins and discuss what alternatives might be possible.

The development of technology of infiltration testing can be traced through publications. The following list shows an incomplete chronology.

1973 BRE151 Soakaways – (also NHBC 5.3.11) Soakaways BRE Digest 151, Watford. Building Research Establishment (Digest 151 is presented currently as a method in NHBC Standards )
1991 SR 271 The Hydraulic Design of Soakaways Report The Hydraulic Design of Soakaways Report SR271 1991 by DC Watkins published by HR Wallingford is available as a free resource here and describes the modelled water flow processes in a soakaway.
1991 (latest revision 2016) BRE365 Soakaway Design Empirical method of soakaway design including test for infiltration rate.
1996 R156 Infiltration drainage – Manual of good practice CIRIA report Report 156 1996 Infiltration drainage – Manual of good practice Roger Bettess BSc PhD MCIWEM. Provides an infiltration test procedure similar to BRE365 but with some variations and extra information.
2007 (Version 6 including 2016, 2018, 2019) C753 The SuDS Manual The current reference for Suds Approval Boards/Lead Local Flood Authorities (SAB/LLFA) in the UK is the The SuDS Manual (C753)

BRE 151 & NHBC 5.3.11 involves recording the time for a fixed height of water to drain in a small trial hole. The time t is graphically translated to a soakaway size using correlations which have uncertain/unknown justification.

SR271 indicates that flow in unsaturated ground is different to saturated ground in four main respects (head in the fluid is due to suction, storage coefficient and conductivity are variables, and gravity induces vertical flow), and compares in-situ test results with those obtained from numerical modelling using Richards modified equation. By relying on testing at the proposed location of the device and use of bulk factors of safety the design philosophy is confirmed to be experimental. Uncertainties do not include the possibility of significant variations in soil type at the location of the test (and by implication the soakaway).  This may be a point of contention with some practitioners used to UK shallow geology.

The work in R156 was carried out by HR Wallingford under contract to CIRIA in the period October 1991 to March 1995. R156 cites many other sources of information (BRE were part of the steering group represented by Dr John Powell). The extra information in R156 does help to confirm the experimental nature of the test specific to the location of the proposed infiltration device and the intrinsic adoption of bulk safety factors. This requires that the location of the device (test site) must be chosen by a designer.

C753 probably deserves an AGS article on its own. In Chapter 01 – The philosophy of SuDs, the 3-D conceptual model in Figure 1.2 models the underlying geology as a grey monoblock rather than layers of strata for example which is a shame. In nearly 1000 pages of C753 the word geology occurs 16 times, frequently associated with land stability; ecology is listed 42 times and archaeology once. Possibly geology is not that important to the delivery of SuDs, but is regarded as a constraint. AGS members are probably most interested in SuDs Section 25 Infiltration design methods, and specifically 25.3 Infiltration Testing Methods and Appendix B.4 Infiltration Assessment. Appendix B.4 includes a checklist (Table B6) intended for use by the approving bodies (and the designers) which assumes competence in ground assessments, which may not be available in all design teams. There are plenty of apparently excellent SuDS case studies, but they don’t generally include details of the geology or infiltration tests and how infiltration challenges were overcome. Competent ground assessors would surely benefit designers.


The SUDS manual is probably the most important document for members and designers to understand. Unfortunately, members may only be requested to perform the BRE365 test and nothing else.

In my experience, over-winter (wosrt-case) water level monitoring has been more frequently requested by designers indicating a more favourable approach to ground modelling.

It seems the current SuDS manual was written with Eurocodes in mind, but apparently not integrated possibly due to the embedded modelling from SR271 between test, device design and bulk safety factors. Mention of 14688 & 14689 in the B6 checklist are assumed to enable initial permeability assessments to be made by designers from descriptions correlated to published permeability values. In such cases, ground practitioners might note subtle fines content descriptors in coarse soils which could lead to significantly different engineering properties in tests. Designers and regulators might just see either SAND or CLAY and get a wrong outcome.

It is tempting to assume that permeability and infiltration approximate to the same value but that may not be accurate. Scale & geology variations mean that measurements of permeability in small laboratory samples cannot be representative unless part of a fully developed ground characterisation. In any event permeability cannot be used in exactly the same way as infiltration rate due to the modelling/empirical link between test and the device design. It is the device (soakaway/basin) that has a characteristic design value of infiltration (usually the tested soil worst-case), and potentially too conservative if water will find preferential pathways and bulk safety factors are deployed. Whilst the SuDS manual includes falling head tests in ISO 22282-2:2012 as acceptable in principle, there are conditions and limitations. There may be parallels here with the recent AGS presentation on sample disturbance where taking a few high quality and well-selected data sets might be considered “better” than many more data of variable/unknown control. We might assume that falling head test data should be presented as a part of a full Eurocode characterisation of the ground to be compliant. Simply replacing isolated infiltration data from a trial pit with more falling head tests in a borehole is likely inadequate, although potentially safer for operatives and might use less resources.

Use of boreholes as a device for obtaining infiltration data is a natural ambition for AGS members seeking compliance with standards and health and safety. However, if the depth of infiltration is limited, for example, by the Environment Agency through planning to less than 2m depth (as been the experience of the author) then the preference for using trial pits is understandable. A daisy-chain or in series hierarchy approach from initial water harvesting to bulk basin attenuation with limited shallow infiltration finally to deep soakaways for overflow situations might be good design but is not necessarily supported by case study.

There is also Geotechnical investigation and testing – Geohydraulic testing – Part 5: Infiltrometer tests (ISO 22282-5:2012), which exclusively describes the various types of ring infiltrometer test; single or double ring, open and closed. In these tests, flow through the side is not included, therefore not compatible with the SR271 modelling. However, the ring infiltrometer would logically be more appropriate for plane devices such as permeable pavements if the modelling is different. Again in my experience, It seems shallow depth BRE365s is preferred by designers for permeable paving.


From this discussion, there are no obvious “off the shelf” replacement alternatives to the BRE365 test to recommend to members.  Noting the fundamental link in the digest between the design output and the in-situ test. However, practitioners might note that although BRE365 was most recently updated in 2016, Bettes 1996 (R156) is the primary reference in the SUDs manual.

The alternatives to BRE365 should be ISO 22282-2 or ISO 22282-5 neither is wholly recommendable (or likely to be accepted nationwide) as a direct replacement.

To support a sustainable agenda wider use of understanding ground models at the initial stage is recommended and not to rely on limited study & a small data set of BRE365 tests. A site where there is a 3m thick layer of clay over 10m of unsaturated permeable sands; three 2m deep tests would indicate infiltration is not possible resulting in the design of an attenuation basin occupying potentially unnecessary space and wasting other resources. A conceptual model, initial investigation and targeted deeper BRE365 tests may result in a much more sustainable scheme.

BRE365 and the SuDS manual could be updated to include the modern context such as health and safety including the gravel filling of test pits, responsible use of resources/logistics and adoption of dataloggers. Guidance could be provided on the benefits of a good geological characterisation and what benefits could be gained from having representative/characteristic high quality data rather than the adoption of worst-case values because of limited data. More case studies might be expanded to include problem solving where poor draining geology has been overcome by engineering for example using linear features which can intercept preferential pathways in variable soil/rock. Or combined systems featuring infiltration devices of limited/known capacity linked to bulk attenuation and final overspill into deep borehole soakaways.

The AGS is keen to hear examples of good practice for safely undertaking the BRE365 methodology and examples where SuDs design has been informed by alternatives to the conventional BRE365 method.  Please contact the AGS at with examples or comments.

Article provided by James Harrison of 4D Geo Limited