What archaeological geophysical survey finds and misses

An archaeological geophysical survey planning exercise is a screening tool, and it is a good one. It covers ground quickly, its cost per hectare is a fraction of trenching, and on the right geology it produces a plan of buried ditches and enclosures that looks almost like an excavation. What it does not do is prove absence, and the gap between those two statements is where developers lose money.

A blank magnetometer plot over heavy clay tells you very little. A county archaeologist knows that, and will still want trenches.

The techniques and what each detects

Three methods account for most commercial work, and they answer different questions.

Magnetometry, usually with a fluxgate gradiometer, measures small variations in the magnetic field caused by differences in magnetic susceptibility. It picks up cut features filled with topsoil, so ditches, pits and enclosure boundaries show well, and it is excellent at detecting burning, so kilns, hearths and fired clay stand out. It is the default for large rural sites because a team can cover several hectares a day.

Earth resistance measures the resistivity of the soil, which depends on moisture. It detects stone walls, metalled surfaces, rubble spreads and foundations better than magnetometry, and it is much slower, so it is normally used over targeted areas rather than whole sites.

Ground penetrating radar sends pulses into the ground and records reflections, giving depth information and the ability to work through hard surfaces. It is the method of choice in urban settings, inside buildings, over car parks and where stratified deposits need to be understood in three dimensions.

Specifying one method for a site that needs another is the most basic error, and it happens because magnetometry is quoted per hectare and everyone likes the price.

Geology and soils that defeat the method

Magnetometry depends on contrast, and some ground offers none.

Igneous and strongly magnetic bedrock generates a background signal that swamps archaeological responses. Heavy clay subsoils give weak contrast between feature fill and natural, so a ditch that would read clearly on chalk or gravel may be invisible. Alluvium and colluvium bury features below the effective depth of the technique, which is why floodplain sites and the foot of slopes routinely produce clean plots over real archaeology.

Made ground is worse. A site levelled with imported material in the 1960s will give you a plot of the 1960s.

Resistance survey has its own weather problem. In a drought or in waterlogged conditions the contrast between a wall and the surrounding soil collapses, so the timing of the survey affects the result. Radar attenuates rapidly in clay, which limits useful depth.

Check the geology before specifying the technique. The published geological mapping for the site is the first thing a competent contractor will look at, and it should be referenced in the survey report alongside soil data.

Modern interference and site clearance

Ferrous metal dominates magnetic data. Field boundary fences, pylons, buried services, pipelines, reinforced concrete, scrap and even a parked vehicle produce responses that mask whatever is underneath.

Practical consequence: a 10 metre strip along every fence line is usually unreadable, and in a small field that can be a third of the area. Say so in the report rather than presenting the whole plot as surveyed.

Vegetation decides whether the survey can happen at all. Short grass, stubble and bare ploughed ground are fine. Standing crops, long grass, scrub and brash are not. Livestock have to be moved. Booking a survey for July in a maize field is a wasted mobilisation, and the cost of a flail pass is far less than the cost of a second visit.

Agree with the landowner what will be cut and when, in writing, before the contractor is instructed.

Reading a greyscale plot without over-claiming

A geophysical survey produces anomalies. Features are an interpretation of those anomalies, and the two should never be conflated in a report.

Competent interpretation classifies responses: probable archaeology, possible archaeology, uncertain origin, agricultural in origin such as ridge and furrow or ploughing trends, natural such as geological banding or periglacial features, and modern including ferrous and service responses. Each class should be mapped separately and the reasoning given.

Three habits of good reports. They present the raw or minimally processed data as well as the interpretation, so a reviewer can see what the plot actually shows. They state the instrument, traverse interval and sample interval. And they say what confidence attaches to each interpretation rather than labelling everything as probable.

Where a report shows crisp rectilinear enclosures, the interpretation is usually safe. Where it shows a scatter of weak discrete anomalies, it is not, and committing a mitigation strategy to that reading is premature.

Using results to target trenches

Geophysics earns its fee by making the evaluation smarter, not by replacing it.

A targeted trenching scheme places trenches across anomalies to establish their date, character and condition, and across blank areas to test whether the survey missed anything. The blank-area trenches are the ones applicants try to delete, and they are the ones that protect you, because they are the evidence that the negative result is real.

The overall sample is agreed with the county archaeologist or the authority's archaeological adviser, expressed as a percentage of the application area, and it varies with the perceived potential. Agree it at scoping stage and record the agreement. A field evaluation designed this way, with a written scheme of investigation approved before fieldwork, is what allows the authority to reach a view on significance at application stage rather than deferring everything to condition.

Where burials are a possibility, note that magnetometry detects inhumation graves poorly, so the evaluation strategy rather than the geophysics has to address it. The procedural consequences if human remains are found are set out in human remains on site, licence and procedure.

What a negative survey does not prove

This is the section to quote at the client who wants to skip the trenches.

A survey that detects nothing has established that there are no detectable anomalies of the kind the chosen method can see, in the areas that could be surveyed, under the conditions on the day. It has not established that the site is archaeologically sterile.

Things that commonly do not show:

• postholes and small cut features, particularly in clay soils
• inhumation graves and cemeteries
• stone structures without magnetic contrast, on a magnetometer survey
• timber structures and organic deposits
• flint scatters and other artefact-based evidence with no sub-surface structure
• features truncated by ploughing to a shallow surviving depth
• anything beneath made ground, alluvium or deep colluvium

So the correct conclusion in a report is that no anomalies of probable archaeological origin were detected, with a statement of the limitations. A report that concludes the site has no archaeological potential is overreaching, and an archaeological adviser will treat the whole document with more suspicion as a result.

Cost and programme against evaluation

Geophysics is fast and low in cost relative to trenching, which is exactly why it comes first. A large rural site can be surveyed in days and reported in weeks, and used to shape the masterplan before the layout is fixed.

Trial trenching needs a written scheme of investigation, a machine, a team, spoil management, reinstatement and a post-excavation allowance. It takes longer to arrange and longer to report, and the archive and publication obligations continue after the fieldwork ends, as covered in post-excavation assessment, archive and publication.

The sequence that keeps a programme intact is a desk-based assessment first, then geophysics scoped to the geology, then targeted and blank-area trenching, then a mitigation strategy. Compressing it rarely saves time, because the authority will not accept a mitigation strategy built on a screening survey alone.

Start with an Archaeological Desk-Based Assessment so the geophysics specification is informed rather than generic, and carry the conclusions into the Heritage Statement where designated assets are also affected. Councils expect a suitably qualified author for archaeological work, and professional review is strongly advised before submission.

To see which heritage and archaeology documents your site is likely to need, enter your postcode on PlanScope and work through the guided questions.

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.