Project

geomagnetic field; data assimilation

Lead
Lay summary
The Earth’s magnetic field has been in existence for the last 3 billion years or more. Generated in the Earth’s fluid core, it provides a protective shield from solar radiation which would otherwise be harmful to life on Earth. Yet this protective shield is far from constant in time, having undergone many reversals or huge variations in intensity over much of its life. In this proposal we look at variations in the field over the last 4 centuries, and try to determine characteristics of the magnetic field within the core. We do this by using observed geomagnetic data (e.g. from mariners in the 17th century and onwards) together with a dynamical model that represents the physics of fluid flow in the core of the Earth. This technique is termed data assimilation. The rewards of this fusion are tangible: for the first time we will be able to image fluid flows and magnetic fields in the interior of the Earth’s core. A great unknown in geomagnetism is the strength and geometry of the magnetic field in the core. Our work will, for the first time, provide a 3-dimensional picture or model of this field. This has implications for Earth history, since it is intimately tied to the energy budget of the core.

Direct link to Lay Summary

Last update: 21.02.2013

Number	Title	Start	Funding scheme

Traditionally the analysis of observed geomagnetic data and the production of dynamical models of the secular variation have proceeded as separate activities. Here we describea fusion of these ideas, termed data assimilation, that brings the two fieldstogether. The rewards of this fusion are tangible: for the first time we will beable to image fluid flows and magnetic fields in the {\sl interior} ofthe Earth's core. A great unknown in geomagnetism is the strength and geometry of the magneticfield in the core. Our work will, for the first time, provide a 3-dimensionalpicture of this field. This has implications for Earth history, since it is intimately tied to the energy budget of the core. We build on work carried out during the previous SNF grant. In that grant we made considerable progress on developing solutions to a very non-linear inverse problem concerning the induction equation, governing the evolution of the magnetic field through time. Data are taken on the boundary of a spherical core and through the use of thesedata and the evolution equation for the magnetic field, the aim is to determine the initial conditions of the magnetic field in the entire corethat lead to the observed evolution. This study (published in Li et al, Phys. Rev. E, 2011) was a kinematic study in which the velocity field wasprescribed. The next logical step, that moves towards a proper description of the dynamics of the core, is to incorporate the variation of thefluid velocity in time as described by the Navier-Stokes equation. A certain version of the Navier-Stokes equation is used in which rotation and Lorentz (magnetic) forces play a key role.Our project is built on a clear division of work between one postdoctoral fellow and three students. Work on long-timescale dynamics will be carried out bythe student and postdoc together, and will use a dynamical model eminently suited to time scales of 100 years or more. The data to beassimilated are already extant, in the form of historical observations over the last 400 years, and archaeomagnetic data spanning somethousands of years. Our methods will be validated by a study of a controlled laboratory experiment, in which an assimilation schemeis used and the results compared to "ground-truth" values determined by non-invasive methods such as ultrasonic Doppler velocimetry andpressure sensors. Finally we aim to reap rewards from the opportunity afforded by the imminent launch of the European Space Agency's Swarm mission, due for launch in July 2012. We will use data from this mission, and from previous missions, in a special assimilation schemegeared towards short time scale dynamics. This scheme uses the quasi-geostrophic description of core dynamics, and will deliver certain averagedmeasures of interior flows and magnetic fields that are complementary and comparable to those estimated from longer time scale studies.