Summarising AEM data for mapping applications

Abstract

AEM Geophysical surveys are excellent tools for mapping conductivity variation over large areas. Common workflows involve inverting data using 1D models on a per-station basis, then gridding those results over lines to produce maps. In the absence of other filtering, gridding operations must combine finely-sampled along-line data with sparsely-sampled between-line data. Independent of the choice of gridding technique and map cell dimension, such maps will obscure the two scales. Obtaining a spatially coherent map will often involve reducing the resolution along the survey lines. Indeed, it could be argued that if the goal of an AEM survey is mapping, then inverting the data on a station by station basis is not necessary. We show that large-scale structures are preserved when data are summarised using the arithmetic mean over a number of stations. This allows practitioners to objectively determine map cell dimensions since it provides an indication of the distance over which 1D models which are typically used to process large data sets are valid. Practically, it makes little difference whether this summary takes place before data are inverted or after. Summarising data before inversion may provide a practical estimate of spatial and temporal variation of the data at a particular scale. In contrast, summarising inverted models may provide an estimate of model variability at a particular scale.