The Gold Standard of Archaeology | Induced Polarisation, Chargeability and Clay–Metal Ambiguity

1. Define the archaeological target before the electrical property

Back to contents

Describe the feature, expected size and depth, matrix, moisture, salinity, clay content, redox conditions and modern disturbance. A kiln, ditch fill, slag scatter or corroded object may create different resistivity and polarisation patterns. Decide whether the question is detection, mapping or material discrimination. A survey cannot uniquely identify a target that several plausible ground models reproduce.

2. Understand time-domain and frequency-domain IP

Back to contents

In time-domain induced polarisation, current is applied and the voltage decay after switch-off is integrated over chosen windows. Frequency-domain or spectral IP measures amplitude and phase across frequencies. Both describe delayed charge storage and release, but the reported parameters depend on waveform, windows, frequencies and processing. Record these settings; chargeability values from different systems are not automatically interchangeable.

3. Relate chargeability to interfaces and pore processes

Back to contents

Polarisation can arise at metallic grain–electrolyte interfaces and at mineral or clay surfaces where ions accumulate or move slowly. Grain size, surface area, saturation, pore-water chemistry and conductivity affect the response. Normalised chargeability or complex conductivity may help separate conduction from polarisation, yet interpretation remains model-dependent. Avoid translating one high value directly into a material name.

4. Design electrode arrays for resolution and signal quality

Back to contents

Array geometry, spacing, contact resistance, stacking and current strength control sensitivity and depth coverage. IP measurements are especially vulnerable to electromagnetic coupling, electrode polarisation and late-time noise. Plan reciprocal measurements, repeat stacks and reference lines. Use non-polarising electrodes where appropriate, keep cables and connectors documented, and avoid claiming depth from a smooth inversion image alone.

5. Distinguish clay, metal and mixed deposits

Back to contents

Metallic particles can produce strong chargeability, but clay-rich material can also polarise through surface conduction and cation exchange. Corroded metal changes both the object and surrounding soil. Compare resistivity, chargeability, normalised chargeability, decay shape or spectral parameters rather than one map. Geological and sediment evidence may rule out interpretations that the electrical data alone cannot.

6. Worked example: a workshop anomaly beside a clay ditch

Back to contents

In an invented survey, a 20-by-15-metre workshop area contains a compact high-chargeability anomaly beside a broad conductive band. The compact response has moderate resistivity and a slow decay; the band is very conductive with lower normalised chargeability. Trial cores find corroded iron fragments in the compact zone and clay-rich ditch fill in the band. The electrical pattern guided sampling, but neither chargeability map alone could have assigned those materials.

7. Process decay curves before trusting the inversion

Back to contents

Inspect raw voltages, primary response, decay windows, sign, stacking variance, contact resistance and reciprocal error. Remove bad quadrupoles under declared thresholds and retain flags. Fit a decay or spectral model only where data support it. Joint or structurally coupled inversion can sharpen interpretation but can also force artificial agreement between resistivity and IP. Show data fit, sensitivity and alternative regularisation.

8. Changed-context transfer check: saline coastal sediment

Back to contents

Move from a well-drained inland site to saline coastal clay. High conductivity shortens useful signal windows and can increase coupling or suppress interpretable IP contrast. The same workflow transfers—target model, reciprocal checks, raw-decay inspection and validation—but electrode spacing, current, windows and companion sediment analyses must change. Failure to detect metal does not exclude it when the conductive host masks the response.

9. Know the limits and common misconceptions

Back to contents

IP is not a unique detector of metal, archaeology or contamination. Chargeability depends on acquisition and normalisation, not only the ground. A colour maximum can be an inversion artefact or a noisy late-time measurement. Greater depth of investigation does not mean equal resolution at depth. Clay and metallic responses can overlap, while dry ground and poor contacts may erase a real target.

10. Audit field acquisition, inversion and validation

Back to contents

Archive electrode coordinates, array sequence, current, waveform, windows or frequencies, contact resistance, stacks, reciprocal pairs, instrument timing, raw voltages and exclusion rules. Publish inversion mesh, error model, regularisation, convergence and sensitivity. Link cores, excavations, magnetic data and laboratory mineral or grain analyses to predictions made before validation, including anomalies that proved non-archaeological.

11. Frequently asked questions

Back to contents

Is IP the same as resistivity? No; it measures delayed or phase-shifted response in addition to conduction. Does high chargeability mean metal? Not uniquely. Can clay polarise? Yes. Why use normalised chargeability? It can help compare polarisation with conduction, but it is not a universal separator. Can IP give exact depth? No; depth and resolution depend on geometry, noise and inversion.

12. The gold standard keeps material claims conditional

Back to contents

Credible archaeological IP begins with a target model, measures decay or phase with adequate controls, and interprets chargeability alongside resistivity and sediment context. It makes clay–metal ambiguity explicit and uses cores, excavation or companion geophysics to test material predictions. The best result is not a dramatic chargeability image; it is an evidence chain from injected current to a bounded archaeological claim.