Seismic image-guided 3D inversion of marine controlled-source electromagnetic and magnetotelluric data

Abstract

Geological interpretation of resistivity models from marine controlled-source electromagnetic (CSEM) and magnetotelluric (MT) data for hydrocarbon exploration and reservoir monitoring can be problematic due to structural complexity and low resistivity contrasts in sedimentary units typically found in new frontier areas. It is desirable to reconstruct three-dimensional (3D) resistivity structures that are consistent with seismic images and geological expectations of the subsurface in order to reduce uncertainty in the evaluation of petroleum ventures. Structural similarity is achieved by promoting a cross-gradient constraint between external seismically-derived gradient fields and the inversion resistivity model. The gradient fields come from coherency weighted structure tensors computed directly from the seismic volume. Consequently, structural similarity is obtained without the requirement for any horizon interpretation or picking, thus reducing significantly the complexity and effort. We demonstrate the effectiveness of this approach using CSEM, MT, and seismic data from a structurally complex fold-thrust belt in offshore northwest Borneo.

Deep Structure of the Santos Basin, Offshore Brazil From 3D Inversion of magnetotelluric Data

“…For geological verification, we extracted resistivity‐versus‐depth profiles from each of the 3D models (derived from different starting half‐space) at the borehole locations and compared them with resistivity logs from 11 wells that are located nearby or at shortest offsets from the MT lines using a threshold distance of 5 km (see Figure 7 and Texts S7 and S8 in Supporting Information S1). We consider this as a geologically valid way of ground‐truthing any inversion model (e.g., Karpiah et al., 2022; Mackie et al., 2020; Meju et al., 2019; Saleh et al., 2022). Ideally, this will enable us to select which of the MT models best describes the subsurface geology.…”

“…Marine magnetotelluric (MT) method is an established tool for offshore resistivity mapping (e.g., Constable & Weiss, 2006;Gallardo et al, 2012;Hoversten et al, 1998Hoversten et al, , 2015Karpiah et al, 2020;Key et al, 2004;Mackie et al, 2020;Saleh et al, 2022;Zhdanov et al, 2011). It is envisaged that resistivity models from threedimensional inversion of MT data may provide important constraints to address the challenges about the mapping of the basement topography, crustal nature, and Moho variation.…”

Using Large-Size Three-Dimensional Marine Electromagnetic Data for the Efficient Combined Investigation of Natural Hydrogen and Hydrocarbon Gas Reservoirs: A Geologically Consistent and Process-Oriented Approach with Implications for Carbon Footprint Reduction

“…Important and, in some respects, unique information about a hydrocarbon gas-charged or hydrogen-filled reservoir can be obtained by measuring its electrical resistivity. Resistiv-ity is a function of porosity, fluid saturation, temperature, and chemical composition, and hence the accurate 3D mapping and monitoring of this property in a reservoir is of practical importance and is remotely possible if the subsurface structural configuration is correctly specified using seismic, borehole, and other types of a priori information [2,6,7]. Figure 2 shows the relationship between bulk resistivity and resistive gas saturation ( [8] Archie 1942) for different porosities (0.15%-0.3%), representing typical unconventional reservoirs (shales and tight sands) and porous clastic reservoirs (permeable sands and carbonates).…”

Magnetotelluric Investigations of the Kīlauea Volcano, Hawaii

“…The MT modeling code used in this work is a proprietary commercial code that has a long history of use and has been benchmarked against semi‐analytical results and other numerical codes (e.g., Mackie & Watts, 2012; Miensopust et al., 2013). It has been used extensively for commercial exploration projects including oil and gas, geothermal, mining, and marine EM (e.g., Hoversten et al., 2021; Mackie et al., 2020; Soyer et al., 2018).…”

“…The proprietary inversion software used here is a nonlinear conjugate gradient algorithm. Details can be found in Hoversten et al (2021), Mackie et al (2020), Rodi and Mackie (2001) and in Supplement 2. Kauahikaua et al (1986).…”