
Tripotential data processing for HES interpretation
Author(s) -
Pietro Cosentino,
Dario Luzio
Publication year - 1994
Publication title -
annals of geophysics
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.394
H-Index - 60
eISSN - 2037-416X
pISSN - 1593-5213
DOI - 10.4401/ag-4174
Subject(s) - merge (version control) , electrical resistivity and conductivity , inversion (geology) , experimental data , interpretation (philosophy) , geology , mathematics , statistics , computer science , physics , paleontology , quantum mechanics , structural basin , information retrieval , programming language
In this paper some methods are proposed and compared to correct the experimental measurements for preliminary processing of tripotential data which are acquired for HES prospecting. However, the use of those methods should be based upon an accurate analysis of all experimental data. Such an analysis ought to involve: 1) an estimate of the averaged measurement errors with their variance and distribution in both the space and the three apparent-resistivities domains; 2) the choice of a resistivity model capable of describing the actual volume under study. The differences among the three values of apparent resistivity measured on a point are generally influenced both by the resistivity distribution below ground as well as by the eventual measurement errors. The proper choice of the method of correction which may be useful to merge the resistivity values and minimize the measurement errors is also linked to the separation of modelling effects. Consequently, the model chosen should be selected in relation to the above mentioned analyses. It thus becomes useful to know the general relations among the three apparent resistivity values for some simple structures, e.g. the two-layer model with a slowly changing first layer thickness. This theoretical model is presented and discussed using two new composed apparent resistivities, namely p? and p?, which seem to be useful tools in HES interpretation. The behaviour of the calculated responses can be useful also for a fast data inversion