Electrical engineers require geoelectric models to design the earthing grids of substations. The traditional 1D earth model is built from horizontal layers, each defined by its thickness and electrical resistivity. Such models are constructed from geophysical surveys carried out within the planned substation area, representing the average electrical structure of the shallow subsurface. This paper discusses the challenges of resistivity surveys and geoelectric modeling for wide-area earthing grids. Modern power generation facilities, such as thermal, hydroelectric, solar, and windfarm plants, include high-voltage substations and earth metallic plant that extend across large areas. The same applies to power-consuming infrastructure, including industrial and commercial complexes as well as transport systems. To characterize the subsurface for geoelectric modeling, three main geophysical sounding methods can be applied at different scales:
โข Shallow surveys (down to tens of meters) – electrical resistivity techniques, such as 2D ERT profiles and vertical electrical soundings (VES) using Schlumberger or Wenner arrays
โข Near-surface surveys (down to hundreds of meters) – audio-magnetotelluric (AMT) and transient electromagnetic (TEM) soundings, preferably collocated, and
โข Deep surveys (kilometres) – magnetotelluric (MT) soundings. For conventional earthing system simulations under short-circuit conditions, a combination of shallow and near-surface methods usually provides sufficient resolution, since low-frequency current penetration reaches depths of only a few kilometres. However, for HVDC earthing systems, particularly under monopolar operation with earth return, models must extend to depths of at least 20 km, requiring complementary MT soundings. High-resistivity terrain poses an additional challenge, even in not-so-large areas. In such cases, different geophysical methods must be combined to mitigate data distortions caused by resistive subsurface contacts. Field techniques, such as dampening the soil surface with water, can temporarily lower resistivity and improve current injection. The quality of ERT 2D profiles improves significantly after about an hour of water absorption. Once the survey data for individual traverses has been analysed apparent resistivity curves obtained from different methods must be reconciled. The technical literature notes that 1D AMT curves tend to be downward-biased, leading to models that appear more conductive than reality. In contrast, VES/ERT curves are generally upward-shifted. Therefore, correction factors are needed to align the tail of the VES curve with the head of the AMT/TEM curve. TEM data are often the most reliable reference, as they are purely magnetic measurements and are unaffected by static shifts, which are linked to electric-field observations.