Multi-Line 1D Inversion of Frequency-Domain Helicopter-Borne Electromagnetic Data with Weighted 3D Smoothness Regulariza
- PDF / 5,392,716 Bytes
- 25 Pages / 547.087 x 737.008 pts Page_size
- 115 Downloads / 192 Views
Pure and Applied Geophysics
Multi-Line 1D Inversion of Frequency-Domain Helicopter-Borne Electromagnetic Data with Weighted 3D Smoothness Regularization: A Case Study from Northern Iran HOSSEINALI GHARI,1 BEHROOZ OSKOOI,2 Abstract—An efficient pseudo-3D Occam’s inversion scheme is proposed here to stabilize the traditional 1D inversion for the frequency-domain helicopter-borne electromagnetic (FHEM) data. In this scheme, multiple flight lines are inverted simultaneously for layered 1D models minimizing a common objective function with lateral, vertical, and cross-line in the model regularization function. Applying the lateral, vertical, and cross-line weighting factors into the regularization matrix yields a more stable solution and produces geologically more realistic results. In addition, we investigate how the errors of height measurements obscure the FHEM response and affect recovered resistivity models. In this inversion, attempt is made to recover a correct altitude that deals with distortions caused by the presence of measurement height errors in the reconstructed resistivity models. The comparison among 1D, pseudo-2D, and pseudo-3D Occam’s inversions is made through the analysis of data from two different 3D synthetic models and one field dataset acquired from the north of Iran. The results indicate that pseudo-3D Occam’s inversion provides fewer inversion artifacts, better model recognition, and smoother and more continuous models, while, reduces the effects of data noise in an effective manner. Keywords: Electrical resistivity, FHEM data, height errors, pseudo-3D Occam’s inversion.
1. Introduction During the last two decades the FHEM systems have been and are being improved dramatically. Holladay and Lo (1997), Siemon (2009), Pfaffhuber et al. (2012), and Legault (2015) introduced a
1 Department of Mining and Metallurgical Engineering, Yazd University, Yazd, Iran. 2 Department of Earth Physics, Institute of Geophysics, University of Tehran, Kargar shomali, 1439955961 Tehran, Iran. E-mail: [email protected] 3 Geological Survey of Sweden, Box 670, 75128 Uppsala, Sweden. 4 Department of Earth Sciences, Uppsala University, Villa¨ vagen 16, 75236 Uppsala, Sweden.
and MEHRDAD BASTANI3,4
considerable volume of information on the evolution and the state of the art of the commercial FHEM instruments available worldwide. The FHEM method is applied as a rapid reconnaissance technique in identifying resistivity structures of a shallow subsurface on a regional scale with measurements at more than one hundred thousand stations in a short period of time. The FHEM applications are commercially viable in the presence of a quite fast and accurate algorithm to simultaneously invert a large amount of data to model near-surface resistivity variations. Despite the great progress made in the computational power of today’s processors, an inversion with reasonable accuracy based on the full 2D and 3D solutions requires high computational cost (Yu and Haber 2012; Boesen et al. 2018; Ghari et al. 2020). Therefore, 1D inversion
Data Loading...
