Simultaneous inversion for mantle conductivity and source field: towards a 3-d global reference model

Oregon State University, Corvallis OR

Investigators

Abstract

Although mantle electrical conductivity primarily increases radially with depth, a growing body of evidence indicates there is significant lateral heterogeneity in the upper and mid-mantle, reflecting variations in composition, physical state, temperature, and volatile content. Conductivity complements seismically-derived density and elastic parameters; through co-registration and joint interpretation of these parameters, one can obtain better constraints on the composition and physical state of Earth's interior. The goal of this project is to develop significantly improved images of these 3D variations, especially in the transition zone and upper mantle. The project combines new approaches to data analysis with numerical modeling and geophysical inverse methods. An empirical orthogonal function approach is being applied to geomagnetic observatory data to extract the dominant modes of external excitation of the Earth's conducting mantle at daily variation and longer periods. These modes, which correspond to the most symmetric large scale sources, are then fit by simultaneously adjusting external source structure and 3-d Earth conductivity. A key constraint in this joint inversion is a priori knowledge of physically reasonable external source spatial structure. Such structure is enforced through use of parametric models, and through multivariate spatial covariances developed from knowledge of source geometries and outputs of ionospheric general circulation models. Co-registered electrical and elastic models will be useful to geodynamicists, to separate better temperature and compositional effects in 3-d tomographic images, and by mineral physicists developing equations of state relevant to deep Earth materials. The research will also result in an improved characterization of external source fields, which, together with the 3D conductivity model, will be very useful to those working with satellite magnetic data, e.g., to provide improved external/induced field corrections for modeling of the main or crustal fields. Current and planned efforts to use magnetic satellite data to probe mantle conductivity in particular depend critically on improved spatio-temporal modeling of external sources, and will ultimately benefit significantly from the results of this project. The project contributes to diversity and training of the next generation of Earth science researchers by providing post-doc support for an early-career female scientist. A global 3-d conductivity reference model is being produced that will be available as a resource to a broad community of Earth Science researchers, e.g., to provide boundary conditions, as well as large scale context, for modeling and inversion of data from regional magnetotelluric and geomagnetic induction studies, including the EarthScope backbone and transportable MT arrays.