Comparison of Different Coupling Methods for Joint Inversion of Geophysical Data: A Case Study for the Namibian Continental Margin

Abstract

 imaging of rift related volcanic processes at the Namibian margin through joint analysis of magnetotelluric, gravity and seismic data  3D inversion of marine magnetotelluric data is improved by cross-gradient coupling with fixed structural density modelESSOAr |

Formation and geophysical character of transitional crust at the passive continental margin around Walvis Ridge, Namibia

“…In order to identify specific parameter ranges and -relationships representing the different stages of continental breakup at the Namibian margin, we analyse electrical resistivity and density models which have been derived from a joint data inversion with cross-gradient coupling, based on marine magnetotelluric-(MT) and satellite gravity data, respectively (Franz et al, 2021). Marine MT stations were deployed along two perpendicular profiles along-(profile P100) and across (profile P3) Walvis Ridge, with a stations spacing of ~10 km (Fig.…”

“…Through a comparison to literature parameter values and geophysical studies, we show that the clusters can indeed distinguish lithologies or rock types linked to the continental breakup stages. The applied data, inversion scheme, sensitivity-, and model analysis of the congruent resistivity and density models are described in much detail in Franz et al (2021); Jegen et al (2016); and Moorkamp et al (2011). GMM and other fuzzy clustering methods have also been used to guide joint inversion schemes by adding petrophysical data and improving subsequent geological interpretations (e.g.…”

“…Fig. 4 in Franz et al, 2021). Moho depth was derived from a regional density model (Maystrenko et al, 2013) and a global crustal thickness reference (Laske et al, 2013).…”

“…In each of these three domains, we perform clustering using logarithmic electrical resistivity (log 10(ρa) in Ωm) with absolute density (d in kg m -³) of the same model cell. Although the inversion is conducted on a 3D model cube and both data sets have 3D sensitivity, previous tests indicate, that the resistivity model's resolution capabilities are limited away from the MT stations (Franz et al, 2021). We therefore use only model cells with a maximum horizontal distance of 10 km to the MT station location (Fig.…”

“…In this study, we complement previous geophysical studies by analysing combined parameter models of electrical resistivity and density. The models, which form the basis for this study, are published in Franz et al (2021), where data, model resolution, and inversion scheme are presented in detail. Here, we focus on a clustering analysis of the relationship of density and electrical resistivity from two models derived in a cross-gradient coupled joint inversion.…”