Most real soils are not uniform. They have a surface layer, an intermediate one, and perhaps rock or a water table. A single apparent resistivity value does not describe that structure; a depth sounding is required.
How it is measured
A four-electrode array in line is used, such as the Wenner or Schlumberger array, and the spacing is progressively increased. Each spacing explores a larger volume of soil, so the apparent resistivity obtained reflects progressively deeper layers. Apparent resistivity is plotted against spacing, normally on logarithmic scales.
How the curve is interpreted
- Nearly flat curve: approximately uniform soil.
- Rising curve: deep layers more resistive than the surface layer.
- Falling curve: deep layers more conductive.
- Curves with a maximum or minimum: more than two layers.
To go from the curve to the model, a layer structure (each layer with its resistivity and thickness) is fitted so that the calculated curve matches the measured one. Historically this was done by comparison with standard curves; today inversion programs are used. The usual result is a two-layer model, sometimes with more layers.
Limitations
The inverse problem has no unique solution: different models can produce nearly identical curves, especially if data are missing at certain spacings. It is therefore advisable to measure in several directions, with well-distributed spacings, and to cross-check against available geological or borehole information. Buried metals, recent moisture or lateral heterogeneities distort the curve.
What the model is for
The layer model is the input for realistically calculating the resistance of a system, its potential rise, and step and touch voltages. Using the value of a single layer where the soil is stratified can underestimate or overestimate those results. Measurements at sites with energized installations must be performed by qualified personnel.