Trench positioning in an archaeological field evaluation

Evaluation trench positioning is settled on paper before any machine arrives. The layout sits inside a written scheme of investigation that the local planning authority approves on the advice of its archaeological adviser, and it has to do two jobs at once: sample enough of the developable area to be defensible, and land on the anomalies that the desk-based work and geophysics have already flagged. Where those two aims pull against each other, the sampling argument tends to win, because that is what the adviser is measuring you against.

For a promoter carrying a 10 to 100 unit site, the cost of getting this wrong is rarely the fieldwork itself. It is the second phase, mobilised four months later, when the adviser reads the report and says the blank half of the site was never tested.

What the evaluation is trying to establish

An Archaeological Field Evaluation exists to answer three questions. Is there anything there. If so, how well preserved, how extensive and how significant is it. And can the scheme be designed around it, or will a full excavation be needed.

That third question is the commercial one. An evaluation that proves absence across the whole site is worth paying for, because it closes the risk. An evaluation that proves absence across two thirds of the site and never looked at the rest has bought you very little.

This is a different exercise from the Archaeological Desk-Based Assessment that precedes it. The desk study tells you where to look. The evaluation tests whether the desk study was right.

Sampling percentage and how it is agreed

Trench sampling percentage is expressed as a proportion of the area to be disturbed, and it is a matter for agreement rather than a national standard. In England, 2 to 5 per cent of the developable area is the range most advisers open with. Sites with a strong desk-based case for buried remains, or those inside an Archaeological Priority Area, attract higher figures.

Two points are worth settling in writing before the scheme is drafted.

1. What counts as the sampled area. Roads, gardens, attenuation basins and the areas of deepest cut all disturb ground. A percentage calculated against the built footprint alone will be challenged.
2. Whether the percentage is a floor or a target. Some advisers treat an agreed figure as a minimum to be exceeded if early trenches are productive, and say so in the scheme.

A useful discipline is to state the percentage, the total area, the number of trenches and the individual trench dimensions in one table, so nobody has to reverse-engineer the arithmetic later.

Trench width matters more than it looks. A 1.8 m trench opened with a toothless bucket lets a supervisor read a section face and define a feature edge. A narrower slot tells you something is there without telling you what.

Targeting trenches from geophysics and desk data

Geophysical survey targeting is where evaluation layouts earn their money. Magnetometer survey over a ploughed field will usually resolve ditches, pits and industrial activity; it is far weaker over made ground, near steel fences and across former hardstanding. Historic England's guidance on geophysical survey in field evaluation sets out what each technique can and cannot see, and it is the document your adviser will have read.

Position trenches so that each one crosses an anomaly at an angle rather than running along it. A trench laid down the length of a linear anomaly gives you one long fill and no stratigraphic relationship. A trench cutting across it at 45 degrees gives you both edges, the profile and, if you are lucky, a junction with something else.

Then keep some trenches deliberately blank. Advisers distrust a layout that only tests anomalies, because an all-targeted scheme cannot demonstrate absence anywhere. A blend of targeted trenches and a background grid over the apparently empty ground is what usually gets approved, and the scheme should say which trenches are which.

Constraints from services, trees and access

A layout drawn purely from the geophysics plot survives contact with the site for about an hour. Overlay the constraints before the trench plan is issued.

• Buried services. Statutory undertaker records, a CAT and genny sweep and hand-dug inspection pits where a trench crosses a recorded main. The Health and Safety Executive's HSG47 guidance is the reference point, and the CDM Regulations 2015 put the planning of this on the client as well as the contractor.
• Trees. Root protection areas calculated under BS 5837:2012 will rule out machine excavation close to retained trees. Offset the trench or reduce it to a hand-dug sondage, and say which in the scheme.
• Standing water and made ground. Trenches in a low-lying corner may need a pump or a different season.
• Access and spoil. A 360 excavator needs a route in and somewhere to put the spoil beside each trench. On a narrow site this alone can force a rotation of the whole layout.
• Livestock, crops and tenancy. Agricultural holdings often set the window, and a tenant's harvest date is a programme item.

Record each moved trench and the reason. An adviser who can see why trench 14 sits eight metres west of the anomaly will accept it. One who cannot will ask for another trench.

Recording and the written scheme of investigation

The scheme is the contract. It names the organisation, the project manager, the standards followed, the recording system, the sampling strategy for environmental and dating evidence, the treatment of human remains, the arrangements for reporting and archive deposition, and the health and safety framework. The Chartered Institute for Archaeologists publishes the standard and guidance for field evaluation that most schemes are written against, and registered organisation status is often what the adviser checks first.

A trench plan referenced to a coordinate system, with each trench numbered and dimensioned, belongs in the scheme rather than in an email. Our longer piece on the written scheme of investigation for field evaluation works through the section order and what each part has to contain.

Build in a variation mechanism. Something will change on site, and a scheme that allows the supervisor to agree minor repositioning with the adviser by telephone, recorded in the site diary, avoids a fortnight of correspondence.

Deciding what happens after evaluation

The evaluation report should end with a clear statement of what the results mean for the scheme. A description of the features alone leaves the decision to the adviser. Three outcomes are normal: no further work, a watching brief during groundworks, or targeted excavation of a defined area before construction.

Where excavation is needed, the trench results define its extent, so the positioning decision you took months earlier is what sets that cost. This is the point at which a layout that tested the whole site pays off, because the excavation area can be drawn tightly.

Scheduled monuments sit outside all of this. Works affecting a scheduled monument, including evaluation trenching, need scheduled monument consent under the Ancient Monuments and Archaeological Areas Act 1979, and that is a separate application to the Secretary of State with its own timescale.

Getting the scheme approved before mobilising

County archaeologist approval is normally secured through the local planning authority, which holds the decision even where the advice comes from a county or shared service. Send the scheme in as a formal submission against the relevant condition rather than as an informal draft, and allow time for one round of comments.

An illustrative sequence for a nine-hectare site with a four-hectare developable area, at an agreed 3 per cent sample, might run: desk-based assessment and geophysics in month one, draft scheme in month two, approval in month three, fieldwork in month four, report in month five. Treat those durations as indicative only. They compress if the adviser is well resourced and stretch considerably if the scheme arrives in August.

Two things reliably delay approval. A scheme with no named project manager, and a trench plan whose numbers do not reconcile with the stated percentage. Both are avoidable in an afternoon.

If you are still working out which reports your site will need before you commission any of them, browse the 66 PlanScope report types and start from the constraint list rather than the survey list.

This article 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.