Focused time-lapse inversion of radio and audio magnetotelluric data

Abstract

Geoelectrical techniques are widely used to monitor groundwater processes, while surprisingly few studies have considered audio (AMT) and radio (RMT) magnetotellurics for such purposes. In this numerical investigation, we analyze to what extent inversion results based on AMT and RMT monitoring data can be improved by (1) time-lapse difference inversion; (2) incorporation of statistical information about the expected model update (i.e., the model regularization is based on a geostatistical model); (3) using alternative model norms to quantify temporal changes (i.e., approximations of l 1 and Cauchy norms using iteratively reweighted least-squares), (4) constraining model updates to predefined ranges (i.e., using Lagrange Multipliers to only allow either increases or decreases of electrical resistivity with respect to background conditions). To do so, we consider a simple illustrative model and a more realistic test case related to seawater intrusion. The results are encouraging and show significant improvements when using time-lapse difference inversion with non l 2 model norms. Artifacts that may arise when imposing compactness of regions with temporal changes can be suppressed through inequality constraints to yield models without oscillations outside the true region of temporal changes. Based on these results, we recommend approximate l 1 -norm solutions as they can resolve both sharp and smooth interfaces within the same model.

Covariance-constrained difference inversion of time-lapse electrical resistivity tomography data

“…In consequence, this parameter only acts as a multiplication constant of C Δm in the inversion process and its effect is overcome by the optimization made on the regularization parameter λ (equation 3). This is also observed by Hermans et al (2012) andRosas Carbajal et al (2012) and will be illustrated for the second synthetic case. We invert for model parameter changes that can explain the variations observed in the data.…”

“…In the present application, we are instead interested in the ability of the covariance matrix to regulate the correlation length of the sought model changes (Hermans et al, 2012). We thus accept using the concept in situations, in which theoretical statistical assumptions are not strictly verified (Day-Lewis et al, 2002;Doetsch et al, 2010;Rosas Carbajal et al, 2012).…”

Time-lapse inversion of one-dimensional magnetotelluric data

“…Standard MT modelling tools such as inversion can also be adapted, for example Rees et al (2016) use cascading two-dimensional (2D) inversions to model a coal-seam gas depressurisation, where the results from the preinjection inversion are used as a prior-model for the postinjection. An improvement on cascaded inversion has been implemented by Rosas-Carbajal et al (2012), where differenced 2D MT inversions have achieved high accuracy by subtracting the prior-model response from all data and reducing the data error. High-accuracy simultaneous time-lapse 1D MT inversions have also been approximated using 2D MT codes (e.g.…”

Probabilistic 3-D time-lapse inversion of magnetotelluric data: application to an enhanced geothermal system

“…Various geophysical techniques have been applied in time-lapse studies that aim at inferring temporal changes in the near subsurface (e.g. LaBrecque & Yang 2001;Day-Lewis et al 2002;Ajo-Franklin et al 2007;Miller et al 2008;Doetsch et al 2010;Rosas Carbajal et al 2012). Tailored inverse formulations that reduce noise and model parametrizations that focus on temporal changes make time-lapse inversions more suitable than simple differencing of models obtained from separate inversions.…”

“…A similar strategy was applied by Ajo-Franklin et al (2007) to better resolve subsurface variations related to CO 2 injection with crosshole seismics, and by Doetsch et al (2010) combined with joint inversion of crosshole ERT and ground-penetrating radar (GPR). More recently, Rosas Carbajal et al (2012) applied this type of inversion approach to time-lapse electromagnetic (EM) data, specifically radio magnetotelluric (RMT) and audio magnetotelluric (AMT). Inspired by the work of Falgàs et al (2009), who monitored saltwater intrusion in a coastal aquifer, Rosas Carbajal et al (2012) demonstrated significant improvements in the resulting models by incorporating information about the expected temporal changes in the subsurface and by removing systematic errors.…”