The integration of different geophysical data has the potential to provide more accurate estimate of subsurface rock properties. Several methodologies and attempts have been developed over the years with the objective of reducing exploration risk. We have developed a cooperative joint-inversion approach intended to facilitate recovery of acoustic impedance (AI) using seismic and magnetotelluric (MT) data. In this approach, the MT data provided a pathway for iteratively building large-scale low-frequency information content not directly recoverable from the seismic data themselves. The MT data provided complementary information to seismic, especially in seismically complex terrains such as overthrust belts, subbasalt and subsalt, carbonate reefs or for targets below deep cover containing limestone, concretionary layers, or basalt. On the other hand, the seismic data provided structural information necessary to derive accurate resistivity models from MT inversion and small-scale features during seismic impedance inversion. The connections between resistivity and the elastic property of rocks are obtained from petrophysical relationships derived from available borehole data, or if not available, from empirical relationships. We tested our technique on synthetic and field data. The application of cooperative joint inversion to 3D seismic and MT data sets acquired in a mineral province made it possible to recover AI distribution across a wide range of geologic environments. The resulting rock property images provided a direct link to geology that is exceedingly difficult, if not impossible, to extract from the individual data sets.

Five audiofrequency magnetotelluric AMT soundings were collected northwest-southeast along the Manyu river in the Mamfe sedimentary basin of southwestern Cameroon. The soundings were performed with frequencies in the range 3 to 2500 Hz and covered a distance of approximately 28 km. Sounding curves and geoelectric and geological sections were processed, and the results were compared with rocks' resistivity to characterize the lithostratigraphy of the eastern part of the basin. The results show above 1000 m depth, sedimentary layers with resistivities in the range of 1 to 100 Ohm-m, which decrease with depth. We identified three types of sedimentary rocks: laterite-clay mixture, shale, and sandstones. Various faults were also identified, illustrating the structural complexity of the Mamfe basin, along the Manyu River.

Natural source electromagnetic methods have the potential to recover rock property distributions from the surface to great depths. Unfortunately, results in complex 3D geo-electrical settings can be disappointing, especially where significant near-surface conductivity variations exist. In such settings, unconstrained inversion of magnetotelluric data is inexorably non-unique. We believe that: (1) correctly introduced information from seismic reflection can substantially improve MT inversion, (2) a cooperative inversion approach can be automated, and (3) massively parallel computing can make such a process viable. Nine inversion strategies including baseline unconstrained inversion and new automated/semiautomated cooperative inversion approaches are applied to industry-scale co-located 3D seismic and magnetotelluric data sets. These data sets were acquired in one of the Carlin gold deposit districts in north-central Nevada, USA. In our approach, seismic information feeds directly into the creation of sets of prior conductivity model and covariance coefficient distributions. We demonstrate how statistical analysis of the 123Surv Geophys (2016) 37:845-896 DOI 10.1007/s10712-016-9377-z distribution of selected seismic attributes can be used to automatically extract subvolumes that form the framework for prior model 3D conductivity distribution. Our cooperative inversion strategies result in detailed subsurface conductivity distributions that are consistent with seismic, electrical logs and geochemical analysis of cores. Such 3D conductivity distributions would be expected to provide clues to 3D velocity structures that could feed back into full seismic inversion for an iterative practical and truly cooperative inversion process. We anticipate that, with the aid of parallel computing, cooperative inversion of seismic and magnetotelluric data can be fully automated, and we hold confidence that significant and practical advances in this direction have been accomplished.

To obtain a higher resolution quantitative P-wave velocity model, 2D waveform tomography was applied to seismic reflection data from the Queen Charlotte sedimentary basin off the west coast of Canada. The forward modeling and inversion were implemented in the frequency domain using the visco-acoustic wave equation. Field data preconditioning consisted of f-k filtering, 2D amplitude scaling, shot-to-shot amplitude balancing, and time windowing. The field data were inverted between 7 and 13.66 Hz, with attenuation introduced for frequencies ≥ 10.5 Hz to improve the final velocity model; two different approaches to sampling the frequencies were evaluated. The limited maximum offset of the marine data (3770 m) and the relatively high starting frequency (7 Hz) were the main challenges encountered during the inversion. An inversion strategy that successively recovered shallow-to-deep structures was designed to mitigate these issues. The inclusion of later arrivals in the waveform tomography resulted in a velocity model that extends to a depth of approximately 1200 m, twice the maximum depth of ray coverage in the ray-based tomography. Overall, there is a good agreement between the velocity model and a sonic log from a well on the seismic line, as well as between modeled shot gathers and field data. Anomalous zones of low velocity in the model correspond to previously identified faults or their upward continuation into the shallow Pliocene section where they are not readily identifiable in the conventional migration.

No abstract