Subsurface conductivity structure as approximated by conductivity-depth transforms

Abstract

Conductivity-depth transforms (CDTs) are widely used as a means of quickly approximating the subsurface conductivity structure as sampled by electromagnetic systems. Using synthetic data from forward modelled layered-earths, the behaviour of CDT profiles with changing conductivity structure is explored to assess how approximate these methods are. It is shown that CDT profiles exhibit a characteristic behaviour that prevents literal interpretation. Specifically, profiles show a propensity to `bulge'. The depth, size and shape of this bulge is directly related to conductivity structure encountered with depth but does not mimic it. Consequently, CDT products (sections, interval conductivities or volumes) have variable levels of accuracy with depth, dependent on what part of the bulge is being examined. Accurate interpretation of CDT products then requires that the interpreter be mindful of CDT behaviour. Alternatively, new products based on profile characteristics, can be generated and interpreted in terms of CDT behaviour. To this end, an algorithm has been developed to automatically provide a quasi 3-layer simplification of CDT profiles. The algorithm shows promise as a means of finding depth to base of the conductive sequence and depth to the top of the first significant conductive layer.