Formation conductivity inverted from multi-frequency, multi-elevation electromagnetic induction data

Thursday, 13 June 2019: 09:20
Davie West Building, DW103 (Florida Atlantic University)
Lanbo Liu, University of Connecticut, Geosciences, Storrs, CT, United States, Rongyi Qian, China University of Geoscience, School of Geophysics and Information Technology, Beijing, China, Jianping Xiao, Central South University, Changsha, China and Hao Xie, China University of Petroleum, Beijing, China
Abstract:
Frequency-domain electromagnetic (FDEM) method is one of the most widely used geophysical electromagnetic induction (EMI) techniques for fast, near-surface geophysical exploration and investigations. Combination of the multi-frequency FDEM equipment (e. g, the Geophex GEM-2) with the multiple-elevation data acquisition capacity by using the unmanned aerial vehicle (UAV) platform enables us to conduct electromagnetic induction survey with an unprecedented efficiency. We call this FDEM data acquisition fashion the multi-frequency, multi-elevation (MFME) approach. The MFME combination generates more data, and provides more constraints for the inversion of the subsurface conductivity. Consequently, it reduces the non-uniqueness of the inverse problem. In this paper, we present an optimized inversion algorithm based on the Levenberg-Marquardt method and the simulated annealing using the one-dimensional layered model. We tested this algorithm by invert the synthetic MFME data generated by numerical model, as well as the real-world MFME data we acquired from a field survey. Through this study, we intend to provide some guidance for acquiring and inverting UAV-borne MFME data sets for environmental and engineering applications.