Geotechnical assessment and Eurocode 7 in planning

A geotechnical assessment submitted with a planning application is doing a narrower job than the one a structural engineer needs. Planning wants to know that the ground will not prevent the development being built as drawn, that neighbours and infrastructure are not put at risk, and that any ground-related constraint has a solution capable of being conditioned. Eurocode 7, BS EN 1997, governs the design that follows. BS 5930 governs how the investigation is carried out. Neither is a planning document, and quoting them does not by itself satisfy a case officer.

Get the sequence right and one investigation serves both purposes. Get it wrong and you pay for a second round of boreholes after permission.

What a geotechnical assessment covers at planning stage

For most applications the Geotechnical Assessment answers five questions.

What the ground is made of, from a desk study of published geology, historic mapping and any previous investigation. Whether the site has a stability issue, which includes slope stability, mining legacy, solution features, made ground of unknown origin and shrink-swell clay with nearby trees. What foundation approach is realistic, stated in enough detail to show the development is feasible without committing to a design. How the works affect anything adjoining, which on a tight urban site means the neighbour's foundations, a retaining structure, a sewer or a railway. And what further work is needed, with the stage at which it should happen.

That last point is what the case officer is really buying. A planning assessment that identifies residual uncertainty and proposes a condition is more useful than one that claims to have settled everything from three trial pits.

Where a basement is proposed, the ground work feeds a Basement Impact Assessment, which asks additional questions about groundwater flow, movement predictions and damage categories. Several London boroughs require one by local policy. Check the authority's published requirements rather than assuming.

Where Eurocode 7 sets the framework

Eurocode 7 has two parts: Part 1 covers geotechnical design, Part 2 covers ground investigation and testing, both read with the UK national annex. Two features of it matter at planning stage.

First, the geotechnical categories. Category 1 covers small, simple structures with negligible risk. Category 2 covers conventional structures with no exceptional risk or difficult ground. Category 3 covers everything unusual, including very large structures, structures in highly seismic or unstable areas and sites with difficult ground conditions. The category drives the extent of investigation and who has to do the design, and stating the category you have assumed tells a reviewer how much work you think the site needs.

Second, the deliverables. Eurocode 7 expects a ground investigation report presenting the factual results and the derived parameters, and a geotechnical design report setting out the design assumptions, the verification and a plan of supervision and monitoring. The ground investigation report is what belongs with a planning application. The geotechnical design report belongs to the building control and construction stage, and submitting it at planning is not wrong but is rarely necessary.

Characteristic values are the part applicants get wrong. Eurocode 7 asks for a cautious estimate of the value affecting the limit state, not an average of the test results, and a report that presents means without discussing the cautious estimate is incomplete on its own terms.

Investigation scope and borehole provision

Eurocode 7 Part 2 gives indicative spacings and depths by structure type rather than a single rule, and the sensible approach is to scope from the development rather than from a table.

Three principles hold on most sites.

1. Investigate to the depth of influence of the proposed foundations, which for piles means well below the anticipated toe level and for a raft means several times the loaded width.
2. Place exploratory holes where the loads are and where the risk is, not where the rig can reach most easily. A hole in the car park tells you about the car park.
3. Cover the variability. Two holes on a site with a buried channel, a backfilled quarry or a former watercourse will not find it, and the desk study should have told you to look.

Trial pits are quick, low cost and limited to shallow depths. Window samples suit soft ground and shallow foundations. Cable percussion boreholes remain the workhorse for most residential sites. Rotary coring is needed in rock and where you want continuous recovery. In situ testing, standard penetration tests, cone penetration tests and plate load tests all have their place, and the report should justify the methods chosen rather than list them.

Interaction with contamination work

The economics argue strongly for a combined geo-environmental investigation, and the technical requirements pull slightly against it.

Geotechnical sampling wants undisturbed samples for strength and consolidation testing. Contamination sampling under BS 10175 wants representative samples preserved for chemical analysis, with specified containers, headspace and holding times. Gas monitoring under the recognised ground gas guidance needs installed wells and repeat visits across a range of atmospheric pressures, which turns a one-week investigation into a monitoring programme of months.

Plan the two together and the same holes serve both. Decide which holes carry installations, agree the analytical suite before mobilisation, and set the monitoring rounds in the programme rather than adding them later. The contaminated land assessment and the geotechnical report can then share a single factual appendix, which also removes the risk of the two documents describing the same stratum differently.

Asbestos in made ground deserves a specific mention in the scoping, because it changes sampling, handling and disposal, and because it is found far more often than it is looked for. The same discipline applies to demolition and asbestos surveys on the standing building.

What planning needs against what building control needs

The division is worth setting out in the report itself, because it manages expectations on both sides.

Conditions follow this split. A typical permission requires a scheme of further investigation and a remediation strategy to be approved before works start, then verification before occupation. Where ground movement could affect a neighbour or a utility, a monitoring regime and trigger levels get conditioned too, and on sites near a railway or over a sewer the asset owner's approval runs in parallel. Temporary works and the protection of adjoining interests are a related exercise, and the constraints are similar to those described for temporary works and tree protection on tight sites.

Who is competent to prepare it

Councils usually expect geotechnical work from someone with demonstrable ground engineering competence, and professional review is strongly advised before submission. In practice that means a chartered engineer or geologist with geotechnical experience, and for anything in geotechnical category 3 a specialist. The registered ground engineering professional scheme run through the engineering institutions is the recognised way of demonstrating that competence at adviser, specialist and professional levels, and naming the author's registration is a short route to credibility with a reviewer.

Contamination work sits under its own competence expectations, with reports prepared in accordance with BS 10175 by suitably qualified and experienced persons, and a licensed or competent professional is legally required for the land contamination reports that support planning decisions. An AI-assisted draft can assemble the structure, the desk study framework and the list of evidence, and it will flag where site data is missing. It is a starting point for that professional rather than a substitute for them.

One honest caveat about scope. A planning-stage geotechnical assessment is based on a limited number of holes, and the thing that most often changes the design is a stratum found later between them. Say which parameters the scheme is sensitive to, and the later discovery becomes a variation rather than a redesign.

This is general guidance, not legal advice. Ground conditions are site specific and the consequences of getting them wrong are structural, so take advice from a qualified geotechnical professional before relying on anything described here.

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