Lightning source data mining for audio-frequency magnetotellurics

Hennessy, L 2018, Lightning source data mining for audio-frequency magnetotellurics, Doctor of Philosophy (PhD), Science, RMIT University.


Document type: Thesis
Collection: Theses

Attached Files
Name Description MIMEType Size
Hennessy.pdf Thesis application/pdf 28.37MB
Title Lightning source data mining for audio-frequency magnetotellurics
Author(s) Hennessy, L
Year 2018
Abstract Recently in Australia, there has been an unprecedented increase in the use of low frequency magnetotellurics to explore for mineral systems on scales of tens to hundreds of kilometres, as exemplified by the Australian Lithospheric Magnetotelluric Project (AUSLAMP). To take full advantage of these efforts and to connect the crustal scale interpretations to the near surface, many explorers will carry out higher frequency lightning sourced Audio-Frequency Magnetotelluric (AMT) soundings. Lightning strikes generate powerful electromagnetic waves, known as sferics, which can penetrate a 1000 Ω.m half space to depths between 100 m and 5.7 km at frequencies of 25 kHz and 7.8 Hz, respectively. However, AMT data are adversely affected by the highly variable signal to noise ratios (S/N) and distortion of the regional (primary) electromagnetic fields. These two problem areas are strongly linked to the lightning strike parameters which determine the primary AMT source fields, such as strike location, peak-current, and current polarity. It is commonly assumed that prolonged acquisition can resolve these issues, but due to the non-stationary statistics of lightning signals, this assumption is incorrect and may reduce survey coverage and data quality. Global lightning networks detect sferics and catalogue the time and location of up to four million lightning strikes per day. In this thesis, I show that through data mining of GLD360 lightning network data, it is possible to extract AMT source information to improve data processing and interpretation. Lightning network data and a model of the Earth-ionosphere waveguide predict the time of arrival, azimuth, and amplitude for each GLD360 detected sferic in our time series EM data. A window of extracted time series data around each predicted sferic is extracted and stored into a structured database with associated lightning network data, such as lightning peak current, polarity, geographical coordinates, and arrival azimuth. For large lightning-peak current, I observed a strong correlation between source proximity and increased S/N, particularly at dead-band frequencies (1.5 to 5 kHz), which often correlate to depths of interest to mineral explorers (200 to 400 m a 1000 Ω.m half space). When sferic signals are small or infrequent, measurement noise in electric and magnetic fields causes errors in estimated apparent resistivity and phase curves, leading to great model uncertainty. Thus, I chose to use our lightning network data mining approach to investigate relationships between lightning strike location, peak current, and the quality of the estimated apparent resistivity and phase curves using the bounded influence remote reference processing code. I investigated two empirical approaches to pre-processing of time-series AMT data before estimation of apparent resistivity and phase: stitching and stacking (averaging). For single-site AMT data, bias can be reduced by processing sferics from the closest and most powerful lightning strikes and omitting the lower amplitude signal-deficient segments in between. The results indicate that the best approach to reduce bias is to stitch the highest amplitude data from the closest sources. Without source information, identification and removal of galvanic distortion is a fundamentally ill-posed problem, unless data are statistically decomposed into determinable and indeterminable parts. Realistic assumptions of the Earth-ionosphere waveguide propagation velocity accurately predict the arrival azimuth for every significant sferic in the relevant time-series data. For each sferic with large amplitude, I calculate the rotation of the electric field from the measured to the predicted arrival azimuth. This rotation of the electric field is a primary parameter of distortion. These results demonstrate that a rudimentary model for near-surface galvanic distortion consistently fits observed electric field rotations. When local features rotate regional electric fields, then counter-rotating data to predicted arrival azimuths is proposed as a means to correct the directional dependence of static shift.
Degree Doctor of Philosophy (PhD)
Institution RMIT University
School, Department or Centre Science
Subjects Electrical and Electromagnetic Methods in Geophysics
Keyword(s) Electromagnetics
Geophysics
Lightning
Magnetotellurics
Conductivity
Versions
Version Filter Type
Access Statistics: 26 Abstract Views, 52 File Downloads  -  Detailed Statistics
Created: Wed, 06 Feb 2019, 09:49:58 EST by Keely Chapman
© 2014 RMIT Research Repository • Powered by Fez SoftwareContact us