Ground parameters inside a Geotechnical Assessment

The useful part of a geotechnical report is a page or two: the ground model and the table of geotechnical design parameters that comes out of it. Everything else, the borehole logs, the laboratory schedules, the photographs, exists to support those two things. If the parameters cannot be traced back to the data, the report is a filing exercise.

Eurocode 7 is explicit about this. A characteristic value is a cautious estimate of the value affecting the occurrence of the limit state. Not the average of the test results, and not the lowest number in the set.

From ground investigation data to a ground model

A ground model is a three dimensional description of what is under the site, expressed so that an engineer can design against it. It sets out the strata sequence with thicknesses and levels, the extent and nature of made ground, groundwater conditions, and the parts of the site where the model is uncertain.

Built properly, it draws on more than the boreholes. Published geological mapping and the national borehole record give the regional picture, historical mapping and site history explain the made ground, and the investigation confirms or corrects both. Where the model departs from the published geology, say so and explain why.

The output a reviewer wants is sections, not just logs. Two or three long sections through the site with the proposed finished levels and foundation levels drawn on them show at a glance where the problems are. A report with 24 borehole logs and no section forces every reader to build the model in their head, differently each time.

Eurocode 7 splits the reporting into a factual investigation report and an interpretative design report. Keep them distinct. Mixing measured data with interpretation is how an assumption becomes a fact between one project stage and the next.

Selecting characteristic values

This is the judgement the whole report turns on, and a good report shows the reasoning rather than only the answer.

Four things govern the selection:

• the volume of ground that controls the behaviour being checked, since a pile in stiff clay averages the properties over its length while a shallow footing responds to a thin weak layer
• the scatter in the data, and whether it reflects real variation in the ground or variation in sampling and testing quality
• the number and distribution of tests, both across the site and with depth
• the limit state being considered, because the cautious value for bearing resistance and the cautious value for settlement are not the same number

Present it as a table with, for each stratum, the parameter, the range measured, the number of results, the value adopted, and a short note on why. Undrained shear strength from triaxial and hand penetrometer results, bulk density, effective stress parameters, compressibility from oedometer testing, and permeability from in situ testing each need that treatment.

Correlations deserve a health warning in the text. Deriving a friction angle from standard penetration test blow counts is normal practice and it carries real uncertainty, so name the correlation used and the assumption behind it rather than presenting the result as measured.

And report the chemical suite alongside the physical parameters: sulfate content and pH for the concrete specification class, and organic content where it affects ground improvement.

Groundwater and its seasonal variation

Groundwater kills more schemes than bearing capacity. A single set of readings taken on the day the rig left the site tells you almost nothing about the level in February.

What a defensible report contains: standpipes or piezometers installed in the relevant strata, readings across several visits spanning different seasons and rainfall conditions, a stated design groundwater level with the reasoning for it, and identification of perched water in made ground separately from water in the underlying strata.

The design level matters in three places. Below ground structures need it for uplift and waterproofing, which is where a Basement Impact Assessment picks the question up. Infiltration drainage needs an unsaturated zone beneath the base of any feature, normally around a metre, which the monitoring has to demonstrate at the worst time of year. And excavation needs it for dewatering, temporary works and any permit that follows.

Where monitoring covers only a dry autumn, say so and allow for it, rather than quoting the lowest level observed as the design level.

Foundation options and their triggers

Planning is not the stage at which foundations are designed, and the report still needs to show that a workable solution exists. Each option has a trigger in the ground model.

Shrinkable clay with trees nearby is the commonest driver of deeper foundations on housing sites, and the depths come from the industry guidance house builders design to rather than from Eurocode 7 directly. Where trees are being removed, long term heave rather than shrinkage becomes the issue, and the foundation and slab design has to allow for it.

Slopes, retaining structures and heave

Where the site has a slope, a retained cut, or an existing wall of unknown construction, the parameters feed a stability analysis and the report should say which analysis was done and what factor of safety resulted.

Three situations that need flagging at planning stage because they change the developable area. A slope that requires a setback or a stabilising structure. A retaining wall on a boundary whose construction and condition are unknown, particularly where a neighbour's land is retained. And ground with a history of instability, which national policy expects to be considered before permission is granted.

Heave protection, compressible fill, slip membranes and void formers are detailed design matters, and their cost is not trivial, so the appraisal benefits from knowing at concept stage that they are required.

Interaction with drainage and infiltration

The drainage engineer needs three things from the geotechnical work, and often gets none of them: measured infiltration rates from tests at the depth and location of the proposed feature, a design groundwater level, and a statement on whether infiltration is acceptable given contamination, stability and groundwater protection.

Infiltration rates taken from soil association mapping are not evidence, and a lead local flood authority will say so. Tests need to follow a recognised method, be carried out at the right depth, and be repeated. Where contamination is present, infiltration may mobilise it, and where the site sits in a source protection zone the Environment Agency's position governs.

Say plainly in the report whether infiltration is feasible. A Drainage Strategy built on a geotechnical report that hedged the point is the one that comes back at condition discharge stage.

Contamination sits alongside all of this rather than inside it. A Contaminated Land Assessment is expected to be prepared in accordance with BS 10175 by suitably qualified and experienced persons, and that is a legal expectation rather than a preference. An AI-assisted draft is a starting point for that professional, never a substitute.

What the planning submission needs against the design

Planning and design want different things from the same investigation, and conflating them produces a report that serves neither.

The planning authority needs to know that the site is suitable for the proposed use, that land stability has been considered, that contamination risks have been identified and can be dealt with, that drainage is achievable, and that the development can be built without unacceptable effects on neighbours. It does not need the bearing resistance calculation.

Illustrative example of the split on a 1.8 hectare former works site. For planning: a factual report, an interpretative report with the ground model and parameter table, a stability statement, a contamination risk assessment, an infiltration statement and a note on foundation feasibility. For the conditions that follow: a remediation strategy, a verification plan, a gas protection scheme and a piling risk assessment. For detailed design: everything else, after permission.

A Geotechnical Assessment is the report type carrying the ground model and parameters. Councils usually expect a suitably qualified author, and professional review is strongly advised.

One honest uncertainty, which good reports state: how much of the site the investigation actually characterised. Fourteen boreholes on two hectares is a sample. Say what the residual risk is, where it sits, and what further work would resolve it, in the same way that a well written baseline chapter of an environmental statement declares the limitations of its data.

To see which ground and drainage reports a site is likely to need, enter the postcode on PlanScope.

This is general guidance rather than 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.