Whispers of the North

Paleomagnetism is the study of Earth’s magnetic field through geological time, as recorded by rocks. Most rocks contain magnetic minerals that record the magnetic field at the time of their formation. Retrieving this stored field is a robust method for reconstructing the past positions and motions of tectonic plates, providing essential insights into geodynamic processes and the long-term evolution of system Earth. However, traditional paleomagnetic methods cannot always recover the Earth's magnetic past because rocks have a geological history. Processes such as alteration, metamorphism, or tectonic processes can affect the mineralogy and the magnetic anisotropy. These processes can alter the preserved magnetic direction or intensity of some minerals, obscuring or erasing the primary magnetic record needed to reconstruct Earth’s past magnetic field. To overcome this problem, I advance and apply the novel approach Micromagnetic Tomography (MMT) in this thesis. Unlike traditional paleomagnetic methods, which measure the bulk magnetic signal of many minerals together, MMT targets the magnetic moment of individual minerals. The conditions in which magnetic minerals grow, determine their shape and size and thus their ability to record a magnetic signal. When measuring the average magnetic signal of millions of these minerals, this bias can hamper the measurement. MMT enables us to select magnetic moments of the reliable magnetic recorders in a rock, thereby overcoming the averaging inherent to bulk measurements. This offers more precise reconstructions of Earth’s past magnetic field direction and intensity than previously possible. This thesis advances the methodology of MMT through two main developments. I tested the use of a microlithography technique to prepare MMT samples. This method simplifies the co-registration of the different datasets required for MMT. Previously, co-registration was dependent on the presence of recognizable features on the sample surface. Microlithography deposits easily recognizable structures on the surface of samples lacking such features, enabling precise co-registration and therefore improving the precision with which magnetic moments are assigned to individual minerals. Furthermore, I present the design and development of an automated sample placement system. This innovation allows high-precision, reproducible positioning during magnetic measurements, greatly improving the robustness of MMT data acquisition. Building on these methodological improvements, MMT is applied here for the first time to a geological case study, focusing on Middle Devonian pillow basalts. Both traditional paleomagnetic analyses and MMT are used to compare the insights gained from bulk and individual approaches. Middle Devonian rocks are typically challenging to analyze, which is also evident in this dataset. The results demonstrate the added value of MMT, particularly when traditional techniques yield ambiguous signals. Finally, I discuss potential future directions for further improvement of MMT. Priorities are refining the statistical framework in which MMT operates and establishing MMT as a standard procedure and a versatile tool for the wider paleomagnetic community. With continued development, MMT has the potential to transform paleomagnetism by unlocking high-resolution magnetic records at the level of individual minerals, offering a new window into the history of Earth’s magnetic field by listening to the whispers of the North.