GNGTS 2024 - Atti del 42° Convegno Nazionale

Session 3.3 GNGTS 2024 Figure 2. EEMverter mult-mesh inversion scheme for Time-Lapse inversion. Top lef) Model mesh corresponding to the frst Time-lapse acquisiton (red polygons). Top right) Model mesh corresponding to the second Time-lapse acquisiton (blue polygons), identcal to the frst model mesh despite of the diferent sounding positons. Botom lef) Forward meshes (grey bars) of the frst acquisiton (red frames). Botom right) Forward meshes (grey bars) of the second acquisiton (blue frames). Fig. 3 presents the resistvity secton of a synthetc model that mimics the electrical propertes (both conducton and polarizaton) of sand, clay and consolidated formatons, based on the petrophysical relatons described in Weller et al. (2015), together with the inversion model of inductve and galvanic data. In partcular, four diferent inversion results are presented: direct current and full-decay induced polarizaton (DCIP) galvanic data, with 10 m electrode spacing and 2D gradient sequence; AEM + ground EM data, with sounding distance of 40 m; AEM+ground EM + tTEM data (Auken et al., 2019), with tTEM soundings every 10 m; all data together in a joint inversion scheme. The joint inversion presents much beter resoluton capability, with the inductve and galvanic data complementng each other in resolving both conductve and resistve layers. The same kind of improvement is found in Signora et al. (2024) with feld data. Another example of joint inversion of AEM and galvanic VES data in EEMverter, without IP modelling but with integraton with resistvity logs is presented in Galli et al. (2024), where the asymmetric minimum support norm (Fiandaca et al., 2015) is used for an automated rejecton of