Applying Modern Statistical Methods in Geophysics: Example of Maximum Likelihood Estimator for Magnetotelluric Data
Speaker: Alan Chave, Complete MT Solutions
Summary: Geophysicists have kept up with technological and numerical innovations,but the one area where geophysics is behind the state-of-the-art is statistics. In this paper we show an example pertaining to the estimation of magnetotelluric (MT) response functions. The robust statistical model of a Gaussian core contaminated by outlying data that underlies robust estimation of the MT response function has been re-examined. The residuals from robust estimators are systematically long tailed compared to a distribution based on the Gaussian, and hence are inconsistent with the robust model. Instead, MT data are pervasively present as stable distribution family that has power law tails. We outline a maximum likelihood estimator (mle) that exploits this fact is outlined. Through comparison to a conventional robust estimator, we show that the mle outperforms it on an exemplar data set.
Adaptive Sensing of UAV-Borne Aeromagnetic Surveys for Improved Target Characterization in Geophysical Surveys
Speaker: Callum Walter, Queen's University
Summary: Advances in aeromagnetic sensor and unmanned aerial vehicle (UAV) technologies have allowed lightweight scalar magnetometers to be integrated with UAV platforms for several geophysical applications including mineral exploration, infrastructure detection, and unexploded ordinance detection. The reliable collection of high-resolution UAV-borne aeromagnetic data requires careful consideration and regard for the numerous physical relationships and dynamic interactions that occur between the UAV platform, the magnetometer payload, and the geophysical targets. The main benefit of UAV-borne aeromagnetic surveys is the ability to fly closer to the ground surface, allowing for the capture of smaller wavelength and weaker amplitude targets, while also being able to traverse flight lines at speeds up to 10 m/s or greater. This compromise in traverse speed and target resolvability can be extremely advantageous for specific project profiles. Presently, most UAV-borne aeromagnetic surveys are conducted by applying a constant flight speed by the operator. This is typically the fastest flight speed that the UAV is capable of and is done to minimize surveying time and cost, while maximizing coverage and efficiency. This approach is effective in reducing the operators overall time in the field, but it does not optimize UAV-borne aeromagnetic surveys based on target characteristics. As the target frequency (wavelength) is the result of both the flight speed and the target’s physical size, it is critical to optimize the resolvability relationship throughout flight by actively managing the flight speed. Real-time monitoring of the target character through adaptive sensing in-flight allows for the flight path and speed of the survey to be optimized, leading to the acquisition of higher quality magnetic observations. The adaptive sensing approach is also able to minimize total flight time and optimize the capture of target signals without the need to input detailed prior knowledge of target sources or locations. The real-time monitoring of the target allows for the operator to update the planned flight trajectory and flight speed to ensure that the expected geophysical target frequencies do not overlap with the inherent platform noise sources (magnetometer swing). The proposed procedure enables target focused surveys based on incorporating and optimizing specific surveying parameters (flight speed, flight time, trajectory, elevation, orientation) and will be unique for all combinations of UAV platforms, magnetometers, target sources, and site characteristics.
Expanding 3D DCIP Sensitivity: Applications for Archaean Precious Metal Exploration
Speaker: Jeff Warne, Quantec Geoscience
Summary: Employing additional transmits electrodes located in boreholes within the area of investigation for surface 3D DCIP surveys can provide substantial expansion of the region of sensitivity, and increased resolution in areas of interest. The presentation reviews developments required for implementation of the technique and examines results from application for precious metals targets at McEwen Mining’s Grey Fox zone located on the Porcupine-Destor deformation zone near Matheson, ON.
Signal Processing for a Three-Component Transmitting (3CTx) Electromagnetic Device
Speaker: Michael Finlayson, Laurentian University
Summary: An electromagnetic device with a three-component transmitter (3CTx) is an innovative mineral exploration tool. The three components are orthogonal and co-located, with each component emitting a field that excites the subsurface. When a 3CTx transmits simultaneously from all three transmitters, the signal measured in a receiver coil will be the sum of the three primary and secondary fields. In order to interpret the data, it is necessary to separate the signals from each transmitter. To that end, we began by constructing a synthetic signal comprised of the three transmitter signals, a powerline signal and a low frequency noise signal. Spectral analysis in the frequency domain showed that multiple combinations of base frequencies can allow the signals from each transmitter to be identified uniquely and separated using a stacking filter in the time domain.
Using a conducting loop-model, it is possible to predict the time-domain response at the three transmitter frequencies. When base frequencies of 30, 32.5 and 35 Hz were selected for the three transmitters and the pulse width and window times after the pulse end were identical, the differences between the responses for each frequency were typically less than 5% for delay times later than 1 millisecond and decay rates greater than 3 milliseconds. However, the discrepancy seems to vary linearly with frequency, so it might be possible to predict and hence reduce this discrepancy. This would allow combining the responses in such a way that extremely conductive features, such as nickel in the Sudbury area could be easily identified without knowing the transmitter-receiver offsets. Thus, building and field testing a 3CTx is warranted.
NEXT : SESSION 2