Specifications

University of Pretoria etd – Combrinck, M (2006)

The second type of target is considered to be a half space or layered earth. The theoretical

assumptions made in this case are that there are no finite conductors present and that the

subsurface layers are perfectly horizontal. Mathematically this means that all processing

reduces to one dimension. TEM surveys designed for this type of target emphasize vertical

variations in conductivity and are called sounding surveys. Fairly realistic models of the

subsurface are obtained by stitching together a number of sounding models to produce a

conductivity depth section and effectively resolving geological features in two dimensions

and not only one. Practical applications would include determining depth to bedrock or

mapping saltwater intrusions into aquifers.

In reality a combination of these two approaches are necessary to obtain a complete

subsurface conductivity distribution and it is towards this goal that most TEM research is

currently focused. The most important factor when dealing with the huge amounts of data

gathered in this instance is to automate as much of the processing and interpretation as

possible in order to keep survey time and costs competitive in the exploration industry. At

the same time it is important to resolve less conspicuous anomalies and map them more

accurately as the tradition of “bump-hunting” is not profitable in modern day exploration.

In the rest of this chapter the different techniques used in TEM interpretation will be

discussed with reference to the optimum targets for interpretation as well as the potential

for automation of these techniques.

3.2 Profiles versus soundings

Profiling and sounding can refer to different survey geometries or just to the way data is

viewed, processed and interpreted. The same data set (or parts of it) can therefore be

treated as either sounding- or profile data or both. The most relevant view or

interpretation strategy is dependent on the primary target or goal of the survey as described

in 3.1. Interpreting data in profile format involves looking at measured values as a function

of distance, removing a half space response if required, and interpreting anomaly shapes

associated with two- or three-dimensional conductors either through forward modelling,

inversion or curve matching. This is very similar to modelling potential field data, but with

some important differences.

There are as many profiles to consider as there were time channels measured (varying

anywhere from 7 to 100) and often it is not possible to obtain a good fit for a specific

conductor on all of these channels simultaneously. This can be ascribed to the fact that