Daniel Sattel

Inverting airborne electromagnetic (AEM) data with Zohdy's method

Zohdy’s method for the inversion of dc-resistivity data has been adapted to the inversion of airborne electromagnetic (AEM) data. AEM responses are first transformed into apparent-conductivity depth profiles, followed by an iterative adjustment of layer thicknesses and interval conductivities. The start model, including the number of layers, is determined from the data. This approach optimizes model flexibility without the need for parameter regularization. Results from Zohdy’s inversion applied to TEMPEST, GEOTEM, and DIGHEM V data acquired in a range of conductivity scenarios including the Bull Creek prospect in Queensland, Australia; the Boteti area, Botswana; and the Reid-Mahaffy test site in Ontario, Canada, show well-delineated target zones. A comparison with Occam’s inversion shows good agreement between the conductivity-depth models recovered by the two methods, with Zohdy’s inversion being 25 to 80 times faster.

Modelling the superparamagnetic response of AEM data

Several lines of VTEM data flown at different system elevations across a known sulphide body and surface cover with elevated superparamagnetic (SPM) properties were analysed with MAXWELL, layered-earth inversions (LEI), LEROIAIR and LEROI. The SPM material was modelled with frequency-dependent magnetic susceptibilities at shallow depth. Due to their slow late-time decay, SPM responses can be confused with responses of deep conductors and vice versa. Depending on the parameter weighting used, 1D inversions model all late-time responses as deep conductive material or as surficial SPM material. However, the joint 1D inversion of data acquired at different system elevations manages to recover a deep conductor from the sulphide anomaly and elevated SPM values at the location of the SPM response. For the modelled parameters, the VTEM datasets from two elevations (at 70 and 80 m) require a vertical separation of ~10 m to allow for the discrimination between the SPM and sulphide responses. For lower system elevations, less sensor separation is necessary due to the strong gradient of the SPM response. Following the determination of SPM parameters from VTEM survey data, these values were used to hypothesise the SPM response for a range of system geometries, showing that larger transmitter loops and larger offsets between transmitter and receiver loops reduce SPM effects. We suggest that two vertically separated receivers could be used to measure the airborne electromagnetic (AEM) gradient and depending on the flying height of the transmitter, the vertical offset of the receivers should be between 2 and 40 m. If gradient data are not collected, then EM responses measured during the transmitter on-time and x-component data, if available, might offer some model discrimination. Whereas synthetic data of the examined helicopter TEM systems VTEM, AEROTEM and HELITEM indicate a fairly high sensitivity to SPM effects, fixed-wing MEGATEM data are much less affected, due to the higher transmitter elevation and large transmitter loop – receiver separation. SPM effects on data of frequency-domain systems such as the RESOLVE system are also small. VTEM data flown at different system elevations across a known sulphide body and surface cover with elevated superparamagnetic (SPM) properties were analysed. The results indicate that SPM responses can be distinguished from deep conductor responses if the vertical AEM gradient is measured, with EM sensors being offset vertically by 2–40 m.

An Analysis of ZTEM Data Over the Mt Milligan Porphyry Copper Deposit, British Columbia

ZTEM data acquired across the Mt Milligan Cu-Au porphyry system are compared with overlapping VTEM data. Conductivity-depth sections derived from both data sets show broad agreement, but indicate better spatial resolution for the VTEM data. Neither data set appears to show a significant response from the Cu-Au mineralization. Products derived from the ZTEM data, including apparent conductivity, phase and Karous-Hjelt filtered grids appear to map geologic structure.

Time-constant analysis of frequency-domain EM data

Summary Deriving time-constants from EM data allows a fast analysis of survey data, but historically time-constants were only derived from transient EM data. The relationship between time-constant and conductance is examined for plate and sphere models for frequency-domain EM data of the RESOLVE system. The results confirm published relationships between the two parameters and indicate that time-constants are underestimated for targets with elevated magnetic permeabilities. Further, the plate model results suggest that RESOLVE data detect highly conductive structures, but don’t resolve their conductances and timeconstants if those values exceed 500 S and 8 ms, respectively. A RESOLVE data set flown initially for kimberlite exploration has been re-analyzed for nickel sulphides. Since the latter are highly conductive, the standard processing products such as apparent resistivity grids, deemed useful for mapping kimberlites, did not provide the optimum resolution for data analysis. The survey area is characterized by strong magnetic and EM responses due to the presence of banded-iron formations and pyrrhotite-rich massive sulphides. In order to determine the strongest conductors, time-constants were derived from the RESOLVE survey data. Due to the strong magnetic response, correction of the RESOLVE data for magnetic permeability resulted in better-resolved time-constants. This was achieved by inverting the data for the conductivity σ and magnetic permeability µr of a layeredearth, followed by the forward modeling of the layer conductivities with µr=1. A comparison of independently derived time-constants and apparent conductances suggests that reliable values derived from the survey data do not exceed 0.4 ms and 200 S, respectively. Conductors with values above 200 S / 0.4 ms are detectable, but not resolvable.

Novel ways to process and model GEOTEM and MEGATEM data

Summary The high-frequency (i.e. near-surface) information of GEOTEM and MEGATEM data is contained in the data recorded during the transmitter pulse, which can be difficult to model with conductivity-depth algorithms. Data processing methods originally developed for the TEMPEST system allow GEOTEM and MEGATEM halfsine data to be transformed to GEOTEM and MEGATEM square-wave data, respectively. The advantages of the transformed square-wave data are that they refer to a standardised waveform that does not vary through a survey, the high-frequency information in the received signal can be readily utilised and the data can be easily corrected for variations in the transmitter height, pitch, roll and receiver offset. Modelling results from MEGATEM data acquired at the Reid Mahaffy test site indicate that the transformation works well for survey data. An alternate way to make use of the GEOTEM and MEGATEM on-time data for conductivity-depth modelling is their inclusion in the modelled transient response. Model resolution is further improved by the inversion of multicomponent data sets. In highly conductive terrain such as above seawater where system parameters such as the bird position are hard to derive reliably from the primary field estimate, the joint inversion of multicomponent data helps to correctly resolve layeredearth parameters. Jointly inverting the 3-component onand offtime data of a GEOTEM bathymetry survey in the Torres Strait showed that the data fit can be greatly improved by allowing the inversion to determine the receiver offset and attitude. This results in greater confidence in the derived conductivity-depth values.

The analysis of ZTEM data across the Humble magnetic anomaly, Alaska

ZTEM data acquired across the Humble magnetic anomaly of almost 30 000 nT were analysed for the presence of a magnetic gradient response and the effects from elevated magnetic susceptibilities. Mag3D inversion of the magnetic data indicates magnetic susceptibility values as high as 2.0 (SI). The response of moving the receiver coil through the magnetic-field gradient peaks at 0.01 Hz and drops off strongly with frequency. Lacking information about the field strength at the base station precludes the comparison of amplitudes between computed gradient responses and the survey data, but the comparison of response shapes suggests that the gradient responses are too small to have a noticeable effect on the survey data. ZTEM responses were forward modelled with a 3D algorithm developed at the University of British Columbia Geophysical Inversion Facility (UBC-GIF) that takes into account electric conductivities σ and magnetic susceptibilities κ, in order to assess the impact of the elevated κ−values derived from the Mag3D inversion. Computing the ZTEM response for these κ-values combined with resistive half-spaces indicates that the response amplitudes and shapes strongly depend on the background resistivities. Ignoring the elevated κ-values during an inversion can result in patterns that resemble crop circles. The approximate conductivity structure of the survey area was derived with a UBC-GIF 3D ZTEM inversion, which models κ = 0. Forward-model results of these conductivities combined with the elevated κ-values derived from the Mag3D inversion indicate that the conductivities are underestimated with the κ = 0 assumption. For an environment such as Humble, with deep-seated zones of elevated κ-values, the shallow inverted conductivity structure appears to be reliable, but the deeper structure should be interpreted with caution. ZTEM data acquired across a magnetic anomaly of almost 30 000 nT were analysed for the presence of a magnetic gradient response and the effects from elevated magnetic susceptibilities. Modelling results indicate distortions in the conductivity structure recovered by 3D inversion when elevated magnetic susceptibility values are ignored during the inversion.

Modelling using receiver waveform and the importance of system geometry

Conductivity-depth sections (CDI) produced from a recent airborne TEM exploration survey showed a poor fit to the expected geology of the area (a known conductive layer was appearing deeper than expected). The source of the problem was found to be the use of an incomplete description of the system geometry which had the effect of dramatically scaling the secondary field. In many modelling programs, including in this case EMFlow, the system geometry may be used to determine transmitter-receiver coupling which is used to compute the apparent primary field. This paper explains why system geometry can be critical for precise modelling of TEM data. By specifying the correct transmitter orientation and de-rotating receiver pitch for both primary and secondary fields, the match between known geology and CDI depth was greatly improved.

The effect of highly magnetic material on ZTEM data

ZTEM data acquired across the Humble magnetic anomaly of almost 30,000 nT were analyzed for the presence of a magnetic gradient response and the effects from elevated magnetic susceptibilities. The response of moving the receiver coil through the magnetic-field gradient peaks at 0.01 Hz and drops off strongly with frequency. Lacking information about the field strength at the base station precludes the comparison of amplitudes between computed gradient responses and the survey data, but the comparison of response shapes suggests that the gradient responses are too small to have a noticeable effect on the survey data. The 3D inversion of the magnetic survey data indicates magnetic susceptibility values as high as 2.0 (SI). Forward-modeling the ZTEM response for these K-values combined with resistive half-spaces indicates that the response amplitudes and shapes strongly depend on the background resistivities. Ignoring the elevated K-values during an inversion can result in the underestimation of conductivities and other artifacts, such as the mapping of patterns that resemble crop circles. For an environment such as Humble, with deep-seated zones of elevated K-values, the shallow inverted conductivity structure appears to be reliable, but the deeper structure should be interpreted with caution.

The superparamagnetic response of transient AEM data

Several lines of VTEM data flown at different system elevations across a known sulphide body and surface cover with elevated superparamagnetic (SPM) properties were analysed with MAXWELL, layered-earth inversions, LEROIAIR and LEROI. The SPM material was modelled with frequency-dependent magnetic susceptibilities at shallow depth. Due to their slow late-time decay, SPM responses can be confused with responses of deep conductors and vice versa. Depending on the parameter weighting used, 1D inversions model all late-time responses as deep conductive material or as surficial SPM material. However, the joint 1D inversion of data acquired at different system elevations manages to recover a deep conductor from the sulphide anomaly and elevated SPM values at the location of the SPM response. For the modelled parameters, the VTEM data sets from two elevations (at 70 and 80 m) require a vertical separation of about 10 m to allow for the discrimination between the SPM and sulphide responses. For lower system elevations, less sensor separation is necessary due to the strong gradient of the SPM response. We suggest that two vertically separated receivers could be used to measure the AEM gradient and depending on the flying height of the transmitter, the vertical offset of the receivers should be between 2 and 40 m.

The 3D joint inversion of MT and ZTEM data

MT and ZTEM data were inverted with a number of 2D and 3D algorithms to recover the subsurface conductivity structure of an area of interest. A 2D inversion algorithm was used to model the magnetotelluric TM and TE mode impedances and the ZTEM tipper data, separately. The derived conductivity-depth sections don’t show much agreement, possibly indicating the conductivity structure of the area to be highly three-dimensional. A 3D inversion algorithm was used to invert the MT and ZTEM data, separately and jointly. Overall, there is good agreement between the derived conductivity structures. This suggests that a joint inversion can extract successfully the combined subsurface conductivity information from the two data sets.