Archaeologists use GPR to understand a site before intrusive work begins, so digging can be focused on the right areas and archaeology can be left in place where excavation is not needed. Small test surveys beside excavations in progress gave us a rare chance to check GPR against ground that had already been opened.
The test surveys
We offered short GPR surveys beside live excavations, to add to the understanding of the sites without widening the digs.
- A grid beside the excavation. An area of roughly 15 m by 35 m was surveyed on an orthogonal grid, with lines run in two directions at right angles. That improves coverage and picks out features whatever their orientation.
- A survey next to an open trench. The trench is part of an ongoing investigation into a site used for waterfront industry, possibly from as early as the Iron Age, on silt laid down by a river. It had produced burnt masonry, suggesting a kiln or oven nearby.
Testing GPR against a trench
Most GPR surveys are interpreted without being able to see the ground. Here the trench lay directly beside the survey grid. Its walls showed the real sequence of layers and the archaeology within them, so we could compare what the radar recorded with what the archaeologists could see.
In the radargrams, the individual profiles of data along each survey line, GPR picked out the natural layers formed as the land built up. That matched the stratigraphy recorded in the trench and gave confidence in the wider interpretation.
Tomography and iso projection
We processed the data in GPR-Slice software.
- Tomography takes the areas of high reflection at a series of depth slices and overlays them into one plan, giving a single picture of the strongest features at all depths.
- Iso projection builds the same high-reflection material into a 3D model, giving each feature a depth and form that a flat plan cannot. It is clearest on an interactive 3D viewer, where it can be rotated and sliced.
These are interpretation tools. They do not replace the radargrams, which an experienced interpreter still checks, but they make patterns much easier to see and to explain to the excavation team.
What the surveys found
- several circular features that appear to be buildings, possibly Iron Age
- a semi-circular structure close to the trench, with a smaller circular feature inside it and a projection off one side, the form expected for a kiln or oven
- at least one other feature that may be a kiln, alongside others of similar size already excavated on site
These are interpretations from the data. Only excavation can confirm them.
Why clay in silt works well
Clay within sand or silt gives a strong contrast for GPR. The materials hold water differently and have different electrical properties, so the radar signal reflects clearly at the boundary between them. The contrast is strongest where a built-up deposit lies over an older ground surface.
The limits
These conditions suited GPR well. Results vary from site to site. Heavy, wet clay soils absorb the signal and reduce how deep GPR can see. Features made of material similar to the surrounding soil may give little contrast. Calibrating against a trench, a borehole or a known feature, as we could here, improves depth estimates and interpretation.
Planning a survey
If you are running an excavation or evaluation and want to target trenches or understand the ground around them, see our archaeological GPR survey page or contact us.