New Advances for a Joint 3D Inversion of Multiple

SUMMARY Electromagnetic (EM) methods are routinely applied to image the subsurface from shallow to regional structures. EM methods differ in their sensitivities towards resistive and conductive structures as well as in their exploration depths. Joint 3D inversion of multiple EM data result in significantly better resolution of subsurface structures than the individual inversions. Proper weighting between different EM data is essential, however. We present a recently developed weighting algorithm to combine magnetotelluric (MT), controlled source EM (CSEM) and DC-geoelectric (DC) data. It is known that MT data are mostly sensible to regional conductive structures, whereas, CSEM and DC data are suitable to recover more shallow and resistive structures. Our new scheme is based on weighting individual components of the total data gradient after each model update. Norms of each data residual are used to assess how much weight individual components of the total gradient must have to achieve an equal contribution of all data in the inverse model. A numerically efficient way to search for appropriate weighting factors could be established by applying a bi-diagonalization to the sensitivity matrix. Thereby, the original inverse problem can be projected onto a smaller dimension in which the search of weighting factors is numerically cheap.