Palaeoenvironmental sampling in an archaeological evaluation

Palaeoenvironmental sampling is the part of an archaeological evaluation that gets specified in one line and then costs three months. A written scheme of investigation that says "environmental samples will be taken where appropriate" has specified nothing, and the reviewer who signs it off has accepted an open-ended commitment on behalf of the client. Palaeoenvironmental sampling in an evaluation needs a stated strategy, a named specialist, a processing route and a decision point between assessment and analysis, or the programme drifts.

This is the component that turns a list of contexts into a description of a past environment, which is usually where the archaeological significance of a site actually resides.

What the sampling programme is for

Three questions, distinct from the questions the excavation itself answers.

First, what was the environment in which the features formed: wooded, cleared, cultivated, wet, salt influenced. Second, what were the site's inhabitants doing with plants and animals, which is where charred cereal assemblages, chaff, weed seeds and fish bone carry the argument. Third, what is the date of the deposits, since carbonised single entity material recovered from a bulk sample is often the only radiocarbon dating opportunity a site offers.

The reason this matters at evaluation stage rather than only at excavation is that the results inform the mitigation strategy. A site where the sampling shows well preserved waterlogged organic sequences is a different planning proposition from one where the deposits are dry and truncated, and the difference changes the scale of any further work the authority will require by condition.

Deciding where and what to sample

Sampling is targeted, not uniform. The strategy should state, for each deposit type expected on the site, whether it will be sampled, by what method, and what question the sample is meant to answer.

Deposits worth sampling in most evaluations include hearths and ovens, pit and ditch fills with charcoal or ash, occupation surfaces, buried soils, waterlogged fills below the water table, and any organic sequence in a palaeochannel. Deposits rarely worth sampling include modern overburden, obviously mixed or bioturbated fills, and contexts with no stratigraphic relationship to anything dated.

Two rules keep the volume sensible. Sample by question rather than by context count, so ten pits of the same date and character do not need ten bulk samples. And record the contexts you decided not to sample, with the reason, because a reviewer or a county archaeologist will ask.

Where the desk study already identified the potential, the strategy should reference it directly. That is the practical link between an Archaeological Desk-Based Assessment and the fieldwork specification, and the sampling strategy is one of the places where a weak desk study shows.

Bulk samples, monoliths and column samples

Three different techniques, three different specialists, three different costs.

Bulk samples are the standard tool for plant macrofossils, small bones, molluscs and small artefacts. A working volume is 40 litres, or the whole context where it is smaller, taken from a single stratigraphic unit and never mixed across contexts. Processing is by flotation, producing a flot for the light fraction and a retained residue, both of which are sorted.

Monolith samples are taken in metal tins pushed into a cleaned section face, preserving the stratigraphic sequence intact. They support soil micromorphology and detailed description of the sequence, and they are the right choice where the relationship between layers is the question.

Column samples are contiguous sub-samples taken down a section face, usually for pollen, diatoms, testate amoebae or insect remains. They only work where the sequence is waterlogged or otherwise anoxic, because pollen does not survive in freely draining deposits.

Specify which techniques are in scope and which are contingent. A written scheme that includes monolith sampling as a contingency, with a unit rate and a trigger condition, is far easier to administer than one that requires a variation every time a section is worth recording properly.

Processing, assessment and analysis stages

These are three separate stages and they are frequently collapsed in project budgets, which is where overruns come from.

Processing is the physical work: flotation, wet sieving, drying, sorting of flots and residues. It is priced per sample and the throughput of a processing facility is finite.

Assessment is a specialist looking at the processed material and reporting on what is present, its preservation, its quantity and whether it is capable of answering the research questions. The output is a recommendation, not an interpretation. For an evaluation, assessment is normally the end point.

Analysis is full quantification and interpretation, with identification to species where possible, statistical treatment where the assemblage supports it, and comparison to regional datasets. It is commissioned only where assessment shows the material will repay it, and the decision is documented in an updated project design.

The reviewer's job at the assessment stage is to test whether the recommendation for analysis is proportionate. A specialist who recommends full analysis of every sample from an evaluation is not being careful, they are being unhelpful, and the condition discharge does not require it.

What the results add to the report

The sampling results should change the conclusions of the evaluation report, not sit in an appendix behind them. If they do not, the sampling was not targeted at the right questions.

Concretely, the results should feed the statement of significance, the assessment of preservation and survival, the dating framework, and the recommendations for mitigation. A report that says the site has moderate archaeological potential, and then includes an environmental appendix showing waterlogged organic preservation in a palaeochannel, has not been assembled properly.

Cross reference by sample number and context, so a reader can move between the context register, the sample register and the specialist report without guessing. Include the sample register as an appendix even where nothing was analysed, because it evidences the strategy.

Costs and how they escalate

The escalation pattern is predictable. Processing costs scale with the number and volume of samples, which is why an untargeted strategy is expensive before anyone has learned anything. Specialist assessment costs scale with the number of samples that produced material. Analysis costs scale with assemblage size and the number of specialist disciplines involved, and radiocarbon dating is charged per determination with a turnaround measured in weeks, sometimes longer at busy periods.

Four controls actually work:

1. Cap the number of bulk samples in the written scheme, with a mechanism for exceeding it by agreement.
2. Separate the processing fee, the assessment fee and any analysis fee, so the client can see the decision point.
3. Require a sample register with preliminary observations to be issued before processing is authorised.
4. State the number of radiocarbon determinations included, if any, and price further ones as a unit rate.

Where the evaluation is likely to lead on to excavation, the post-fieldwork programme is the part that delays discharge of the condition long after the building is finished. Our note on post-excavation assessment, archive and publication sets out how that sequence runs and what it does to a completion programme.

Specifying it in the written scheme

The sampling section of a written scheme of investigation should be about a page and should answer seven questions without ambiguity: which deposit types will be sampled, by which technique, at what volume, who the named environmental specialist is, where processing will be carried out, what the assessment deliverable is, and who decides whether analysis proceeds.

Add the discard policy. Processed residues and unanalysed flots cannot be retained indefinitely, and the receiving museum will have a collections policy that governs what it accepts. Agreeing the discard policy with the curator before fieldwork avoids an argument at deposition.

One honest uncertainty worth stating in the scheme itself. Until the deposits are seen, nobody knows whether the site will produce a usable assemblage, and a well specified programme can legitimately return very little. That is a result, not a failure, and a scheme that acknowledges it is easier to agree with a county archaeologist than one that promises outcomes.

The whole of this sits inside the Archaeological Field Evaluation and should be consistent with the significance argument in any Heritage Statement submitted with the application.

This is general guidance, not legal advice; check the position with your local planning authority or a qualified professional before relying on it.

PlanScope drafts the archaeological and heritage documents that accompany an application, structured to UK planning policy, with a clear marker wherever site specific specialist input is still needed. For specialist technical reports, qualified professional input may be required.

PlanScope provides AI-assisted draft reports for planning applications. Reports should be reviewed before submission. We do not guarantee planning approval.