GNGTS 2024 - Atti del 42° Convegno Nazionale

Session 3.3 GNGTS 2024 EEMverter, a new 1D/2D/3D inversion tool for Electric and Electromagnetc data with focus on Induced Polarizaton G. Fiandaca 1 , B. Zhang 2 , J. Chen 1 , A. Signora 1 , F. Daut 1 , S. Galli 1 , N.A.L. Sullivan 1 , A. Bollino 1 , A. Viezzoli 3 1 The EEM Team for Hydro & eXploraton, Department of Earth Sciences “Ardito Desio”, Università degli Studi di Milano, Milano (Italy) 2 Insttute of Earth exploraton, Science and Technology, Jilin University, Changchun (China) 3 EMergo S.r.l. , Cascina (Italy) Introducton The induced polarizaton (IP) phenomenon in airborne electromagnetc AEM data (AIP) presents a challenge to exploraton in many parts of the world. It is a well-known phenomenon since Smith and Klein (1996) frst demonstrated the presence of IP efects, which have been further discussed by several authors (e.g., Marchant et al. , 2014; Macnae, 2016; Viezzoli et al. , 2017). IP-afected AEM data are ofen interpreted in terms of the Cole-Cole model (e.g., Marchant et al., 2014; Viezzoli et al. , 2017; Lin et al. , 2019), but the inversion problem is partcularly ill-posed: for a 1D inversion of a single sounding four parameters have to be retrieved for each model layer. Furthermore, AIP and ground IP modelling are usually carried out in diferent inversion frameworks, making a direct comparison of the results difcult. In this study we present a novel inversion sofware, EEMverter, specifcally developed to model electric and electromagnetc data taking into account the IP phenomenon. Three distnctve features have been implemented in EEMverter: i) 1D, 2D and 3D forward modelling can be mixed sequentally or simultaneously in the iteratve process within multple inversion cycles, for diminishing the computatonal burden; ii) the joint inversion of AIP, ground EM-IP and ground galvanic IP data is fully supported with a common IP parameterizaton; iii) tme-lapse inversions of AIP, EM and galvanic IP data is possible with both sequental and simultaneous approaches. In the following, the implementaton of EEMverter is described, with examples of synthetc and feld inversion results. Method and results In EEMverter the inversion parameters are defned on model meshes which do not coincide with the forward meshes used for data modelling: the link between model and forward meshes is obtained interpolatng the model mesh parameters into the forward mesh discretzaton, as done for 1D AEM in Christensen et al. (2017), in 3D galvanic IP in Madsen et al. (2020) and in 3D EM in Zhang et al. (2021), Engebretsen et al. (2022) and Xiao et al. (2022a). This spatal decoupling allows