The Earth's magnetic field is generated by fluid motions in the outer core (the geodynamo). The field varies dramatically on a wide range of timescales (e.g. years, decades, centuries, millennia). By studying geomagnetic field variations at the Earth's surface we can learn more about the deep Earth where the field is generated. Thus surface measurements, and models derived from them, provide a window in to the deep Earth.

Geomagnetic field reconstructions for the last four centuries, based on historical observations, have provided insights into the field variations on decadal to centennial timescales. When projected down to the core-mantle boundary, these models reveal the presence of four high intensity patches of flux at high latitudes symmetric about the equator, which have remained relatively stationary over the past 400 years. These, so-called flux lobes, are thought to represent the signatures of whole core convection patterns. Westward drift of the magnetic field is a natural effect of Earth's rotation and the fact that these flux lobes have remained more or less in the same places suggests some form of exterior forcing mechanism on the geodynamo, e.g. temperature variations in the lower mantle. Another characteristic feature of the historical geomagnetic field is the concentration of strong secular variation in the Atlantic (as opposed to the Pacific) Hemisphere, which in turn has been linked to asymmetric growth of the inner core. To confirm these hypotheses we need longer geomagnetic field time-series data.

To look further back in time we rely on indirect observations of the magnetic field recorded in geologic and archaeologic material, the study of palaeomagnetism. The problem is that palaeomagnetic data are not as numerous or as well distributed over the globe as direct observations, in fact, almost 90% of the data are from the Northern Hemisphere. In addition, the majority of palaeomagnetic data come from sediment records that are often naturally smoothed in time, which make modelling and geophysical inferences from the models more challenging.

Alternative estimates of the past geomagnetic field intensity can be obtained from cosmogenic radionuclides (e.g. Beryllium-10 and Carbon-14) deposited in high-resolution geological archives. Cosmogenic radionuclides are produced in the atmosphere by interactions with cosmic rays at a rate that is inversely related to the strength of the Earth's and the Sun's magnetic fields. Extraction of the geomagnetic field component in radionuclide records is difficult and therefore, these data have not previously been incorporated into geomagnetic field models.

In my research project I aim to (i) generate new palaeomagnetic data from key locations in the Southern Hemisphere, (ii) assess timescale uncertainties and adjust for natural smoothing of sedimentary data and (iii) incorporate, for the first time, geomagnetic field intensity estimates from cosmogenic radionuclide records to construct new high-resolution geomagnetic field models for the past 10,000 years. I will use these models to test whether flux lobes are generally symmetric about the equator, which would imply that they are surface expressions of large-scale convection columns in the core. By establishing if the flux lobes are preferentially situated at specific longitudes, I will test if they are tied to lower mantle heterogeneities. Finally I will investigate whether or not the Atlantic-Pacific asymmetry is a persistent feature of the geomagnetic field, which could be linked to asymmetric growth of the inner core. The results from the proposed research will be of direct relevance for a wide range of research topics, including core dynamics, geomagnetic field forecasting, solar activity reconstructions and archaeological dating.