Magnetometry and Soil Magnetic Susceptibility

A fluxgate is a sensor that measures the component of a magnetic field along its axis by driving a ferromagnetic core through saturation and measuring the asymmetry of the resulting voltage. A gradiometer pairs two fluxgate sensors vertically, typically 0.5 m or 1 m apart. Each sensor measures the total field at its height. The instrument subtracts the upper reading from the lower, producing the vertical gradient. This gradient is sensitive to near-surface local sources but cancels the slowly varying geomagnetic background and its diurnal variation, both of which affect both sensors equally.

The sensitivity needed for forensic grave detection is approximately 1-2 nanotesla (nT). Modern instruments such as the Bartington Grad601 achieve sensitivity of around 0.1 nT at survey speeds compatible with systematic field scanning. The operator typically walks parallel traverses 0.5 m apart, recording a measurement every 0.25 m along each traverse. The resulting grid of gradient values is then plotted as a greyscale or colour-scale map, on which anomalies show as patches of locally different gradient relative to the background.

In most natural soil profiles, magnetic susceptibility decreases with depth. The surface horizon is enriched in fine iron-bearing minerals through weathering, biological activity, and the accumulation of magnetically enhanced particles from atmospheric deposition. The subsoil below is coarser, less weathered, and magnetically quieter. This gradient is the fundamental reason magnetometry detects grave cuts.

When a grave is excavated, the spoil pile contains material from multiple depths mixed together. When the grave is backfilled, this mixed material is inverted: topsoil ends up in the base of the fill, subsoil fragments end up near the surface. The result is a column of material with an average susceptibility higher than the undisturbed subsoil alongside it. This susceptibility anomaly produces a positive gradient anomaly detectable from the surface, and it persists for years after burial because the mixed fill does not re-sort itself.

The amplitude of the anomaly decays with time since burial as weathering and biological processes begin to homogenise the fill. Published studies using controlled burials suggest that the susceptibility contrast is strongest in the first two to five years and then slowly diminishes, though it may remain detectable for decades in stable, undisturbed conditions. This time-dependence means that the absence of a magnetic anomaly cannot rule out a burial, particularly an older one.

When soil is heated above the Curie temperature of its iron-bearing minerals (typically 580 degrees Celsius for magnetite, 680 degrees for haematite), those minerals lose their existing magnetic order. As they cool, they reacquire magnetisation aligned with the ambient field at the time of cooling. This is thermoremanent magnetisation. For forensic purposes, the more significant effect is that heating also converts weakly magnetic goethite and ferrihydrite into strongly magnetic magnetite and maghemite, permanently increasing the susceptibility of the burned material.

This means that a bonfire, a pyre, or a deliberate attempt to burn evidence at a crime scene leaves a distinct magnetic footprint in the ground that is detectable long after the physical ash has been removed or dispersed. Gradiometer surveys at conflict-era mass burial sites have repeatedly identified burning horizons as high-susceptibility anomalies that aid in locating associated burial pits. In more recent criminal investigations, fire-affected ground around a scene has been used to map activity zones that witness accounts or other physical evidence did not fully delineate.

Resistivity-based methods measure how easily electrical current flows through the soil. Clean dry sand resists current flow (high resistivity). Moist clay with dissolved ions conducts it easily (low resistivity). Decomposing organic matter contributes both moisture and ions to the pore water, lowering resistivity in the zone immediately around a burial. This makes electrical methods a complement to magnetometry in environments where the magnetic contrast is weak but moisture and organic contrasts are strong.

ERT produces a cross-section of resistivity with depth, which can be compared to the GPR section for the same traverse. A low-resistivity zone at a depth consistent with the GPR anomaly is strong evidence of a genuine target. The combination of two independent imaging methods with consistent results at the same location is the foundation for high-confidence anomaly ranking.

The usefulness of a resistivity measurement depends not just on the absolute value but on the contrast between the target zone and the surrounding host material. A grave in a uniformly low-resistivity clay terrain may produce a resistivity anomaly of only a few ohm-metres against a background of tens of ohm-metres, which is still detectable with good electrode coupling. A grave in a high-resistivity sandy gravel terrain may produce a contrast of hundreds of ohm-metres, which is easier to see but may be confused with natural moisture pockets or organic lenses in the substrate.

Soil moisture also varies seasonally, and with it the background resistivity. A survey run in a dry summer gives a different background value from the same site surveyed in winter. This seasonal variation is well-documented and means that a single survey represents a snapshot. Where results are equivocal, a repeat survey in a contrasting season can clarify whether an anomaly is stable (consistent with a genuine target) or merely tracking seasonal moisture variation.

The 0.5 m traverse spacing used for forensic magnetometry is a compromise between coverage speed and target resolution. A single operator with a Grad601 walking 0.5 m traverses at a normal pace can cover roughly 1,000 m2 per hour in open ground with good footing. A 10 m x 10 m plot takes about 6 minutes. A 50 m x 50 m plot takes roughly 2.5 hours, including time to set up the grid, mark traverses, and record data.

ERT is much slower. Laying a 40-electrode cable, acquiring a full tomographic dataset, running the inversion software, and interpreting the section takes roughly 1-2 hours for a single 39 m transect. For a 30 m x 30 m area with transects at 2 m spacing, that is 15 transects and approximately 15-30 hours of field and processing time. ERT is therefore used selectively, after gradiometry or GPR has identified candidate zones, rather than as a first-pass reconnaissance tool.

• Grid establishment: stakes and string at consistent spacing, tied to a site datum by total station. The grid must be documented photographically so that anomaly positions can be reported as reproducible coordinates.
• Metal-free zone: operators must remove metal objects from their person (belt buckles, keys, phones) to avoid instrument noise. Fences, gates, and metal-framed greenhouses impose a standoff distance of 2-5 m depending on gradiometer sensitivity.
• Data download and export: most instruments store data internally and export via USB to software such as Terrasurveyor or Geoplot. The exported file must be retained as part of the site record for evidential continuity.