Finding existing borehole records before you commission a survey

Before you sign off a ground investigation specification, spend an hour looking for logs that already exist. The British Geological Survey holds a national index of onshore boreholes, wells and shafts, with scanned records that are free to view where the confidentiality period has expired. On an urban site with any history of construction, the chances of finding something within a few hundred metres are good, and what you find changes the investigation you commission.

This is not a substitute for intrusive work. It is how you stop paying for boreholes that tell you what was already known, and how you make sure the ones you do sink go deep enough and in the right places.

What the national index holds

The index covers boreholes sunk for water supply, mineral exploration, foundations, site investigation and research, going back to the nineteenth century. Each entry has a location, a depth, a date where known, and in many cases a scanned log showing the strata the driller recorded.

Coverage is uneven in a predictable way. Cities, coalfields, chalk aquifers and major infrastructure corridors are dense with records. Agricultural land and small settlements are sparse. Records deposited under statutory notification duties, which cover boreholes over a certain depth, are better represented than shallow foundation holes from the 1970s that nobody reported.

Alongside the borehole index sit the digital geological maps, which give superficial deposits, bedrock and artificial ground at national scale. The two together let you build a working ground model before anyone mobilises a rig.

Searching for logs near your site

Open the onshore map viewer, find your site, and switch on the borehole records layer. Then work outwards.

1. Take everything within about 250 metres first, then widen to 500 metres if the geology is uniform.
2. Note the depth of each record before opening it, because a 3 metre water well tells you nothing about founding conditions at 12 metres.
3. Download the scans you want and name them by index reference, not by "borehole 1", so you can cite them.
4. Record the horizontal accuracy stated for each record. Many pre satellite entries are plotted to the nearest 10 metres or worse, and some are simply wrong.
5. Check the digital geological map for the site and for each borehole, so you know whether a log 300 metres away sits on the same mapped deposit.

Then go somewhere most people forget: the council's planning register. Applications on adjoining sites often have full ground investigation reports attached as submitted documents, with modern logs, in situ testing and laboratory results. That is better data than anything in the national index, it is free, and it is almost never searched. Look at the last fifteen years of applications within a couple of streets.

Reading an old borehole log critically

An old log is evidence of what one driller recorded on one day. Treat it that way.

• Check the purpose. A water well log records the aquifer and skims everything above it. A mineral borehole records the seam. Neither describes made ground carefully.
• Check the descriptive terms. Modern soil descriptions follow the current British Standard for site investigation, with a defined vocabulary. Older logs use words like "loam", "marl", "ballast" and "brick earth" that do not map cleanly onto current terms.
• Look for what is missing. No standard penetration test results, no undrained shear strengths, no particle size distributions and no chemical testing is normal on older logs, and it means the log gives you stratigraphy without engineering parameters.
• Watch the made ground. Thickness of fill is the single most valuable thing an old log can tell you on a brownfield site, and also the thing most loosely recorded. A log saying "2.4 m rubble fill" is worth more than a page of interpretation.
• Check the datum. Ground level is often given as an approximate height or not at all, so a depth below ground level cannot safely be converted to a level.
• Note the termination reason. A hole that stopped on "hard ground" at 6 metres may have hit an obstruction rather than rockhead.

Superficial deposits and bedrock mapping

The mapped geology is where you turn scattered logs into a prediction. Identify the superficial deposit at your site, its likely thickness, the bedrock beneath, and any artificial ground already mapped.

Certain combinations carry known consequences worth flagging early. Alluvium and peat in a valley floor mean compressible soils and shallow groundwater. London Clay means shrink and swell and tree influence. Made ground over a former dock, canal arm or brick pit means variable fill and possible contamination. Chalk means dissolution features and, in some areas, historic chalk mines. Sand and gravel means good infiltration for drainage and good bearing, and also means groundwater moves quickly, which matters for contamination.

Write the ground model down as a short table of expected strata with depths, then let the investigation test it. A specification written against a stated model is much easier to defend and much easier to price.

Using the records to scope a new investigation

The point of all this is a better specification. Existing data should let you do five things.

Set the depth of exploration to reach the stratum that governs your foundation solution, rather than to a round number. Choose the technique, because cable percussion, rotary open hole, window sampling and trial pitting each suit different conditions and the existing logs tell you which. Place the holes to test uncertainty, concentrating on the areas where the mapped geology or the old logs disagree. Decide whether groundwater monitoring is needed and how long for. And identify where contamination testing should be concentrated, which feeds directly into a Land Contamination Phase 1 report and its Phase 2 scope.

Give the existing records to the contractor with the tender documents. A contractor pricing blind will either over specify or hit an obstruction and issue a variation.

Groundwater levels recorded historically

Historic water strikes are the most misread part of an old log. A level noted during drilling is a strike, not an equilibrium level, and in low permeability soils the two can differ by metres. Levels also vary seasonally and over longer cycles, and in several former industrial cities groundwater has risen since abstraction and dewatering ceased, so a 1960s level may understate today's.

If groundwater governs your design, whether for a basement, a soakaway or a buried structure, monitor it. Standpipes read over at least several months, ideally spanning a winter, are the only defensible basis for a design level. The Environment Agency's published groundwater level data can supplement that where a monitoring well exists nearby.

Where a basement is proposed, this feeds straight into a Basement Impact Assessment, which will be rejected if the groundwater regime rests on a single historic reading.

Limits of relying on third party data

Three limits to state plainly in any desk study.

The records carry no duty of care to you. They are archive material supplied as found, and errors in location, depth and description are common. You cannot rely on them in the sense a funder means by reliance.

They do not discharge the requirement for investigation. The design standards for geotechnical work expect an investigation proportionate to the structure and the ground risk, and construction health and safety duties require the ground risk to be identified before work starts. A desk study narrows uncertainty, it does not remove it.

And they age. A log from 1958 records ground that has since been dug up, filled, dewatered or contaminated. On a site with any post war industrial history, historic mapping matters as much as the borehole record.

Used properly, the existing data still pays for itself many times over. A Geotechnical Assessment built on a proper desk study, with a stated ground model and a targeted investigation, costs less overall than a generic grid of boreholes and tells you more. If you have not used the national viewer before, the walkthrough on using the geoscience index for borehole and ground data covers the mechanics, and the constraint layers belong on the same site constraints plan as everything else.

PlanScope provides AI-assisted draft reports for planning applications. Reports should be reviewed before submission. We do not guarantee planning approval. Ground investigation and geotechnical design need a suitably qualified and experienced practitioner, and this is general guidance rather than legal advice.