EMGS visiting researcher at Scripps Marine EM laboratory

The current measurement accuracy of marine electromagnetic data facilitates using the technology to study fainter targets and also smaller geological details. Resolving these data responses by imaging software can be challenging when the magnitude of data variations are dominated by variations in other geological features. This can be the case e.g. for a rough seafloor topography, presence of islands, or a coastline near the survey area. Although e.g. the seafloor topography can be measured to very high a ccuracy, the processing and imaging tools must accurately determine the effect on measured responses. Unlike the more commonly used rectangular grids, the use of unstructured meshes can accurately represent the effect of tilts and roughness in the geomo del. This is achieved by the arbitrary placement of grid nodes that facilitates tilted structure, and mesh spacing can be made denser in proximity of local geometrical distortions. In this project, we will develop new 3D inversion software based on 3D un structured tetrahedral mesh representations and the finite-element approach to modeling. The computational complexity of such an approach is a significant R&D challenge, and we will develop an efficient inversion algorithm that can utilize the new modelin g approach to process large-scale datasets. The new software will be applied to process a 3D real field dataset where current modeling approaches are known to be a limiting factor. The project will generate new knowledge both in the domain of numerical m ethods for processing of marine EM data and marine geology. The resulting software will enable processing of data acquired in the most complex environments, and the results from real field data processing will be submitted for publication in an internatio nal journal.