Specifications

University of Pretoria etd – Combrinck, M (2006)

• Speed

• No need for an initial model

• Although it is based on the same imaging principles as the CDI transform it is

easier to implement with fewer empirical factors that need to be included

• The late time approximation used in the algorithm is very applicable to the impulse

response, central-loop configuration this work is based on

• For the specific case of a conductive host rock environment the S-layer differential

transform shows good resolving capabilities

• As mentioned by Tartaras et. al. (2000) the main drawback of this method is the

fact that it requires a numerical differentiation scheme which tends to introduce

noise in the interpretation results. The implementation of a robust, yet accurate

numerical differentiation scheme based on the specific properties of TEM data

therefore receives high precedence. Further enhancements include a noise filter

based on data points satisfying the theoretical conditions for the S-layer transform.

The S-layer differential transformation results in two equations that have to be solved for

every time channel obtained at every station. Originally derived by Sidorov and Tikshaev

(1969) these are summarised by Tartaras et. al. (2000) as:

()

)2.4(

)1.4(

⎟

⎠

⎞

⎜

⎝

⎛

−

′

−=

⎟

⎠

⎞

⎜

⎝

⎛

′

⎟

⎠

⎞

⎜

⎝

⎛

∑