Publication Date:
2023-02-09
Description:
When interpreting geophysical models, we need to establish a link between the models’ physical parameters and geological units. To define these connections, it is crucial to consider and compare geophysical models with multiple, independent parameters. Particularly in complex geological scenarios, such as the rifted passive margin offshore Namibia, multi-parameter analysis and joint inversion are key techniques for comprehensive geological inferences. The models resulting from joint inversion enable the definition of specific parameter combinations, which can then be ascribed to geological units. Here we perform a user-unbiased clustering analysis of the parameters electrical resistivity and density from two models derived in a joint inversion along the Namibian passive margin. We link the resulting parameter combinations to break-up related lithology, and infer the history of margin formation. This analysis enables us to clearly differentiate two types of sediment cover. Namely, one of near-shore, thick, clastic sediments, and a second one of further offshore located, more biogenic, marine sediments. Furthermore, we clearly identify areas of interlayered massive, and weathered volcanic flows, which are usually only identified in reflection seismic studies as seaward dipping reflectors. Lastly, we find a distinct difference in the signature of the transitional crust south of- and along the supposed hot-spot track Walvis Ridge. We ascribe this contrast to an increase in magmatic activity above the volcanic centre along Walvis Ridge, and potentially a change in melt sources or depth of melting. This characterizes a rift-related southern complex, and a plume-driven Walvis Ridge regime. All of these observations demonstrate the importance of multi-parameter geophysical analysis for large-scale geological interpretations. Furthermore, our results may improve future joint inversions using direct parameter coupling, by providing a guideline for the complex passive margins parameter correlations.
Type:
Article
,
NonPeerReviewed
Format:
text
Permalink
