Characterization of the bridge pillar foundations using 3d focusing inversion of DC resistivity data N. Yıldırım Gündoğdu ⁎, İsmail Demirci, Cem Demirel, M. Emin Candansayar Ankara University, Faculty of Engineering, Department of Geophysical Engineering, Geophysical Modelling Group (GMG), 06830 Gölbaşı, Ankara, Turkey abstract article info Article history: Received 4 December 2018 Received in revised form 25 October 2019 Accepted 25 October 2019 Available online 02 November 2019 We investigated the effectiveness of a focusing regularization technique for the inversion of direct current (DC) resistivity data for a typical engineering problem. A smoothing stabilizer (Laplacian of model parameters) is gen- erally preferred in the inversion (OCCAM's inversion) of DC resistivity data. Smooth reconstructions may be pro- duced with this stabilizer, but some specific problems might require more focused images for adequate interpretations. For this reason, we investigated the capabilities of the minimum gradient support (MGS) stabi- lizer for providing shaper results. This stabilizer allows the a sharper reconstruction because its main effect is to minimize the area where strong differences occur between adjacent model parameters. We also analyze the ef- fects of the focusing parameter, which is the parameter in the MGS expression controlling the level of sharpness of the final result. Our strategy for the selection of the optimal focusing parameter allows the resolution of distinct resistivity contrasts. Moreover, some artifacts that may arise in the use of the a very small focusing parameter dis- appear while using the normalized focusing parameter. We demonstrate these results by using both synthetic and field data examples. In the field data test, the subsurface image reconstructed using the proposed MGS ap- proach matches well with the lithology inferred from borehole drillings. © 2019 Elsevier B.V. All rights reserved. Keywords: DC resistivity Focusing inversion Minimum gradient support 1. Introduction The direct current (DC) resistivity method is a well-established tech- nique, which is used to tackle geotechnical and engineering problems where a complex subsurface is present (Cardarelli et al., 2007; Crawford et al., 2018; Danielsen and Dahlin, 2009; Kul Yahşi and Ersoy, 2018; Santarato et al., 2011). As with most geophysical methods, one of the main reason why DC resistivity is preferred in such studies is that it is a method that is (almost) non-invasive and non-destructive (Park et al., 2003; Rucker et al., 2013; Sentenac et al., 2018). Although two-dimensional (2D) resistivity imaging routines are probably the most used, the best way to investigate complex 3D geology is via three-dimensional surveys. However, 3D resistivity surveys are time- consuming and costly when compared to 2D imaging (Rucker et al., 2009). Nevertheless, the popularity of 3D resistivity surveys has been rapidly increasing due to recent developments in field equipment and interpretation software over the last 20 years (Chambers et al., 2006; Jones et al., 2012). Nowadays, thousands of data units can be collected in a few hours with multi-channel and multi-electrode resistivity mea- surement systems (Loke et al., 2013). Moreover, large data sets can be processed by using academic or commercial inversion algorithms. In these algorithms, the widely used regularized optimization is applied since the DC resistivity inversion problem is clearly ill-posed. The usual approach is that second-order derivative regularization is used to maximize the smoothness of in the final model (Binley and Kemna, 2005; Ellis and Oldenburg, 1994; Günther et al., 2006; Loke and Barker, 1996; Marescot et al., 2008; Pain et al., 2002; Papadopoulos et al., 2011; Sasaki, 1994; Yi et al., 2001). Journal of Applied Geophysics 172 (2020) 103875 ⁎ Corresponding author. E-mail addresses: gundogdu@eng.ankara.edu.tr (N.Y. Gündoğdu), idemirci@eng.ankara.edu.tr (İ. Demirci), cdemirel@ankara.edu.tr (C. Demirel), candansayar@ankara.edu.tr (M.E. Candansayar). Fig. 1. Synthetic model representing the bridge pillar (1000 Ωm) and the clayey zone (20 Ωm) below it. The homogeneous resistivity is 100 Ωm (the homogeneous medium is transparently presented for emphasizing the anomaly structures). https://doi.org/10.1016/j.jappgeo.2019.103875 0926-9851/© 2019 Elsevier B.V. All rights reserved. Contents lists available at ScienceDirect Journal of Applied Geophysics journal homepage: www.elsevier.com/locate/jappgeo