Ground movement and damage categories for basements

A basement scheme in a terrace has to show that predicted ground movement will not cause damage to neighbouring buildings beyond an acceptable damage category. The ground movement damage category is the output the auditor reads first, and everything upstream of it, the site investigation, the retaining wall stiffness, the construction sequence, exists to support that one classification. Get the category wrong and the design changes: more propping, a stiffer wall, smaller excavation stages, or a shallower basement.

The classification in general use follows the Burland scheme, which sorts aesthetic and serviceability damage into six categories from negligible to very severe, tied to approximate crack widths and to the ease of repair. It is a description of damage, not a safety limit, and that distinction matters when a neighbour reads the report.

Why a prediction is required at all

The planning system treats structural stability and the amenity of neighbouring occupiers as material considerations. Where a borough has adopted basement policy, it will normally require an assessment showing that the development will not cause harm to neighbouring properties, to the structural stability of the host building, or to local groundwater flow. The assessment is then audited, often by an engineer appointed by the council.

Separately, the Party Wall etc. Act 1996 gives adjoining owners rights over notifiable excavation and work to a party structure. The party wall surveyor's schedule of condition and the planning movement prediction are different documents serving different purposes, and neither discharges the other. Applicants who assume the party wall process will deal with movement lose time at validation.

Read the adopted policy and any supplementary planning document before modelling, because the maximum acceptable damage category is a policy choice. The difference between a limit of Category 1 and a limit of Category 2 changes the wall stiffness, the propping regime and the cost.

How the damage categories are defined

Crack width is an indicator rather than the definition. The categories are anchored in the ease of repair and the function affected, so a 3 mm crack through a lintel bearing is treated differently from a 3 mm crack in render. Report the category and the reasoning behind it, rather than the number alone.

The staged approach and what feeds it

Assessment normally runs in three stages, and stopping at the first is legitimate where the screening criteria are met.

1. Preliminary screening using predicted greenfield movements. Where maximum settlement and the slope of the settlement profile fall below the accepted screening thresholds, damage is classified as negligible and no further work is needed.
2. Second stage assessment, in which the greenfield settlement and horizontal strain profiles are applied to the neighbouring building treated as a simple beam, and the resulting tensile strain is read against the damage categories.
3. Detailed assessment, using soil-structure interaction analysis where the second stage indicates a category above the policy limit, or where the building is sensitive, listed, or already distressed.

The inputs that decide the answer are fewer than the length of the report suggests. Wall type and stiffness, the propping sequence and the timing of each prop, excavation depth and width, soil stiffness derived from the site investigation, groundwater conditions, and the position and form of the neighbouring foundation. A reviewer will test the soil stiffness profile and the construction sequence before anything else, because those two carry the most optimism.

Site investigation depth is the usual weak point. A basement to 3.5 metres with boreholes to 6 metres tells you little about the stratum the wall toe sits in, and nothing about groundwater response. Geotechnical design under Eurocode 7 expects ground parameters supported by investigation rather than by regional assumption, which is why the Geotechnical Assessment and the movement prediction should be written by the same team from the same data.

Where the prediction changes the design

Movement is dominated by wall deflection during excavation, and deflection is dominated by how long the wall stands unpropped. So the design levers are mostly about sequence.

Top-down construction, where the ground floor slab props the wall before the bulk dig, cuts predicted movement substantially compared with bottom-up with temporary props. Excavating in short hit-and-miss pin lengths reduces movement compared with a long continuous dig. Reducing the excavation width by pulling the basement footprint in from the party wall lowers strain at the neighbour. Increasing wall stiffness with a contiguous or secant pile wall in place of mass concrete pins does the same at greater cost.

But a stiffer wall with a poor sequence can predict more movement than a modest wall with a disciplined one. That is why the construction sequence has to appear as a drawn sequence in the submission rather than as a paragraph of intent, and why it should be carried into the construction phase controls alongside the matters covered when piling vibration is controlled by condition.

Dewatering is the other lever, and the one that most often bites later. Lowering groundwater to keep the excavation dry causes consolidation settlement over a wider area than the excavation itself, and may need an abstraction licence or an exemption. Deal with it in the assessment rather than leaving it to the contractor, because the licensing position affects whether the predicted movement is achievable at all. The practical thresholds are set out in more detail in the guidance on when dewatering a basement needs a permit.

Monitoring, trigger values and what gets reported

A prediction is only useful if it is checked. The monitoring regime should name the instruments, their positions, the reading frequency, the baseline period before work starts, and a set of trigger values with a defined action at each.

• Green: readings within prediction. Continue, report monthly.
• Amber: readings approaching a defined proportion of the predicted movement. Increase frequency, review sequence, notify the engineer.
• Red: readings at or above the predicted movement, or an unexpected pattern. Stop, prop, investigate.

Precise levelling on studs in the neighbouring facade, crack gauges across existing cracks, and inclinometers in the wall or in the ground behind it make up the usual set. Baseline readings taken after demolition has started are worthless, so the baseline period belongs in the pre-start condition.

Reporting to the council and to neighbours is where goodwill is won or lost. A regime that sends monthly readings to the adjoining owners, in a format a non-engineer can read, prevents most of the complaints that otherwise arrive as objections on the next application. The full set of documents the auditor works through is covered in the note on what the basement impact assessment auditor checks, and the ground movement chapter sits inside the Basement Impact Assessment rather than standing alone.

Illustrative example. A single-storey rear basement under a Victorian terrace, 3.2 metres deep, is screened at stage one and fails on settlement slope at the neighbour's flank wall. Stage two returns Category 2 with bottom-up construction and Category 1 with the ground floor slab cast first as a permanent prop. The applicant adopts the propped sequence and narrows the footprint by 600 mm from the party wall. Figures are illustrative.

A basement impact assessment is a report councils expect from a suitably qualified author, and professional review is strongly advised before it is submitted. 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.