Premium
An enhanced method for source parameter imaging of magnetic data collected for mineral exploration
Author(s) -
Smith Richard S.,
Salem Ahmed,
Lemieux Jean
Publication year - 2005
Publication title -
geophysical prospecting
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.735
H-Index - 79
eISSN - 1365-2478
pISSN - 0016-8025
DOI - 10.1111/j.1365-2478.2005.00494.x
Subject(s) - amplitude , economic geology , regional geology , a priori and a posteriori , geology , environmental geology , engineering geology , magnetic susceptibility , signal (programming language) , mineral exploration , igneous petrology , mineralogy , geophysics , hydrogeology , optics , computer science , seismology , physics , philosophy , geotechnical engineering , metamorphic petrology , epistemology , quantum mechanics , volcanism , telmatology , tectonics , programming language
We have developed a method for imaging magnetic data collected for mineral exploration to yield the following structural information: depth, model type (structural index) and susceptibility. The active nature of mineral exploration data requires we derive the structural information from a robust quantity: we propose that the first‐ or second‐order analytic‐signal amplitude is suitably stable. The procedure is to normalize the analytic‐signal amplitude by the peak value and then use non‐linear inversion to estimate the depth and the structural index for each anomaly. In our field example, different results are obtained depending on whether we inverted for the first‐ or second‐order analytic‐signal amplitude. This is probably because the two‐dimensional contact, thin sheet or horizontal cylinder models we have assumed are not appropriate. In cases such as these, when our model assumptions are not correct, the results should not be interpreted quantitatively, but they might be useful for giving a qualitative indication of how the structure might vary. With a priori information, it is possible to assume a model type (i.e. set the structural index) and generate estimates of the depth and susceptibility. These data can then be gridded and imaged. If a contact is assumed, the susceptibility contrast is estimated; for the dike model, the susceptibility‐thickness is estimated; for the horizontal cylinder, the susceptibility‐area is estimated. To emphasize that the results are dependent on our assumed model, we advocate prefixing any derived quantity by the term ‘apparent’.