Spectral analysis in airborne electromagnetics

Abstract

Interpretation of airborne electromagnetic (AEM) data is mainly concerned with determining the position, dip, and quality (or conductance) of an anomalous geologic feature. A new method of interpretation using spectral analysis has certain advantages over conventional methods. It is specifically applied to the interpretation of AEM anomalies observed with the INPUT1 time‐domain electromagnetic system that are related to thin rectangular conductive plates. Profiles of field data are converted to the wavenumber domain by applying a Fourier transform. Because the amplitude spectrum varies exponentially with the distance from the target to the receiver, the target depth is easily interpreted from the slope of a semilogarithmic plot of the spectrum. Analytical expressions for the amplitude spectrum of an AEM anomaly can only be obtained for a rudimentary model where the target is represented by a single filament of induced current. The spectra of more realistic models such as the dipping infinitely conductive half‐plane and the thin rectangular plate of finite conductance must be obtained numerically. The computed spectra are relatively insensitive to the target dip, except for the case of a horizontal target of finite width. For a flat‐lying target, the spectrum exhibits a distinctive modulation pattern from which the target width can be determined. A low wavenumbers the time‐decay characteristics of a given target are preserved so the its conductance can be determined. In most cases, however, this parameter as well as the target dip are just as easy to evaluate directly from the observed field data. The results obtained with the spectral analysis method of interpretation applied to field examples agree well with the known geology. Interpretation of field data requires correction for the velocity of the aircraft and the time constant of the receiver equipment. In the wavenumber domain, however, these effects are easily eliminated by a simple division. The spectral analysis method of electromagnetic interpretation is useful for determining the depth of the target because it is unaffected by dip, size, conductance, or profile position. One application of this method is in automated interpretation of portions of AEM data profiles, with large data sets over many conductors. Anomalies of interest can be picked out and interpreted automatically.