Bayesian inversion for a global apparent magnetic susceptibility model based on quiet-time satellite magnetic field measurements

点击翻译按钮获取中文摘要

The study of the lithospheric magnetic field is one of the main goals of the Swarm mission. Its configuration has been designed to be especially beneficial for detecting the small wavelengths of the lithospheric magnetic signal. In particular, satellites Alpha and Charlie fly side by side at a longitudinal distance of 1.4 degrees, which allows for the approximation of the East-West gradient of the magnetic field. This gradient is mainly devoid of the large scale and rapidly varying magnetic field generated by the magnetospheric ring current. Effectively removing the ring current signature from the measurements is currently one of the biggest challenges in achieving high-resolution lithospheric field models and that is why Swarm magnetic field measurements are so valuable. However, gradients cannot effectively remove the unwanted signature of magnetic fields generated by currents flowing over the polar regions, including the Disturbance Polar 2 current system, the polar electrojet and Field-Aligned Currents. In the first part of this proposal, we suggest an innovative way to deal with the polar contamination of lithospheric field models. We suggest that the combined use of SuperMAG indices and data delivered by the AMPERE satellite mission can significantly improve the data selection scheme currently employed over the polar regions, therefore allowing the construction of a magnetically quiet polar data set. This new data selection scheme will be applied on the latest Swarm and CHAMP Level-3 magnetic field data. The additional step of a principal components analysis will help to model and remove any remaining external field noise over the poles. In the second part of the project, the generated data set will be inverted for a global visible magnetization model, along the lines of the project “Structure of the magnetic lithosphere from global analysis of satellite data” from the first phase of SPP1788. Methodological innovations for this part include employing a stochastic inversion approach that delivers posterior estimation of the model uncertainties. In the third part of the project, the magnetization model will be converted to an apparent magnetic susceptibility model. One of the main challenges towards this step is the inherent non-uniqueness of inverting magnetic field data for the apparent susceptibility. However, recently derived equations allow for the first time, under the assumption of induced magnetization, the estimation of the null space of the magnetization and therefore of the apparent magnetic susceptibility. Magnetic susceptibility is a physical quantity of the rocks that reflects properties like age, mineral composition, temperature and pressure. In this respect, it is of direct interest to a variety of disciplines, dealing, e.g., with geology, petrology, seismology, gravity, and tectonic reconstruction studies.