Embodied carbon of foundations and ground works

On a site with poor ground, the substructure can account for more embodied carbon than the frame above it, and foundation embodied carbon is the line most assessments report last and understand least. Concrete and reinforcement are carbon-dense, piling adds plant emissions and spoil movements, and ground improvement trades one material for another. Any whole life carbon assessment that reports substructure as a single number has lost the argument before it starts, because every reduction available to you sits inside that number.

Why substructure dominates some assessments

The pattern is predictable once you have seen a few schemes. On a greenfield site with competent ground and strip footings, substructure is a modest share of upfront carbon. On a brownfield site with made ground, a high water table and a piled solution, it can rival the superstructure.

Three things drive that. Concrete volume rises with piling and pile caps. Reinforcement rates in piles and caps are high, and steel reinforcement carries far more embodied carbon per tonne than the concrete around it. And the site activities module picks up rig fuel, arisings transport and imported fill, which on a constrained urban site is a real quantity rather than a rounding.

So the ground investigation is a carbon document as well as a geotechnical one. A scheme that establishes bearing capacity early can often avoid piling altogether, which is why the parameters in a Geotechnical Assessment belong in the carbon conversation from the start.

Piling against raft against strip foundations

The comparison has to be made on the same basis, which means whole-system carbon rather than element carbon. A raft uses more concrete than pile caps and removes the piles. Piles reduce excavation and spoil but add rig emissions and reinforcement.

Driven piles generate no arisings, which matters more than people expect on a contaminated site where disposal is classified. Bored piles generate arisings that may need treating as waste under a classification you cannot change, and that disposal sits in the same module as the fuel.

And the loads come from above. A lighter superstructure, fewer storeys at the same floor area, or a structural grid that puts loads where the ground is better will cut substructure carbon more than any mix change.

Concrete mix design and cement replacement

Portland cement clinker is the carbon in concrete, so the single most effective specification change is replacing part of it. Ground granulated blast furnace slag and fly ash are the established options, and the designated concretes in BS 8500 accommodate both.

Replacement has consequences you have to design for. Higher replacement levels slow early strength gain, which affects striking times, winter working and programme. In large pours the lower heat of hydration is an advantage. In a thin slab poured in October it is not.

Specify by performance and let the supplier optimise. A specification that fixes a cement type removes the supplier's ability to offer the lowest-carbon compliant mix, and an assessment based on generic industry averages rather than the supplier's own environmental product declarations will understate what you have actually achieved.

Reinforcement deserves the same attention. Electric arc furnace steel from recycled feedstock has a materially lower footprint than primary steel, and the supplier's declaration will say which route produced it.

Ground improvement alternatives

Vibro stone columns, dynamic compaction and lime or cement stabilisation all substitute process energy and imported material for structural concrete. Sometimes that is a clear saving and sometimes it is not, and the honest answer depends on haulage distance for the aggregate.

Stone columns in particular are sensitive to this. The stone has to come from somewhere, and a long road haul can erase the saving against a shallow raft. Ask for the source and the distance before claiming the reduction.

Soil stabilisation using lime or cement to treat in-situ material avoids both import and disposal, which is the strongest argument for it on sites with marginal cohesive fill. Its carbon depends on the binder dose, and the dose comes from trial mixes rather than from an assumption.

Reusing existing foundations

On a redevelopment site, foundations already in the ground are the largest carbon saving available, and they are the hardest to realise.

The barrier is verification. You need as-built records, condition assessment, capacity testing in many cases, and a structural engineer prepared to take the design responsibility. Where records are poor the investigation cost can exceed the saving, and where they are good the saving is substantial.

Partial reuse is the realistic middle ground. Retaining a basement box, reusing a pile cap layout with new piles between, or keeping existing ground slabs as piling platforms all avoid work without requiring anyone to warrant a 1960s pile. The approach fits the same logic as other retained-fabric decisions, which retrofitting flats: fabric works and consents sets out for the building above.

Quantities and their sources

An assessment is only as good as its quantities, and substructure quantities are the ones most often estimated.

At planning stage you will usually have concept geotechnical advice, an indicative foundation strategy and no bar schedules. Say so. Record the basis of each quantity: measured from drawings, taken from an engineer's provisional schedule, or derived from a rate per square metre of footprint, and state which rate and where it came from.

Reinforcement rates are the weakest link. A rate in kilogrammes per cubic metre applied across all substructure elements will be wrong in both directions, and the assessment should flag it as the largest single source of uncertainty rather than present it as a measured figure.

Where arisings are involved, get the haulage assumption on the record too: volume, disposal classification and distance. Those three numbers move the site activities module more than anything else.

Reporting substructure separately

Report substructure as its own element, broken down by material, and show the modules separately so that product manufacture, transport and site activities are each visible. A single combined figure hides where the reduction came from and makes the assessment impossible to review.

The London Plan requires whole life-cycle carbon assessments for referable schemes, reported in the format the Greater London Authority guidance sets out, and that format expects element and module breakdown. Outside London the expectation varies by local plan, and a growing number of authorities ask for an assessment through a policy on carbon in new development. A Whole Life Carbon Assessment prepared to the RICS methodology will satisfy most of them, with the local reporting template checked first.

State your benchmark honestly. Published benchmarks exist for building types and they are not foundation benchmarks, so comparing a piled brownfield substructure against a whole-building figure proves nothing. Compare against the alternative you rejected, show the quantities for both, and let the saving stand on its own arithmetic. The same discipline applies where drainage and ground works overlap, a point covered in sustainable drainage and carbon.

Then tie the conclusion back to the planning case. A Sustainability Statement that claims a substructure saving without the rejected option quantified is an assertion, and reviewers treat it as one.

This is general guidance, not 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.