UCLA

The Earth’s magnetic field is produced by a fluid dynamo in the molten iron outer core. This geodynamo is driven by fluid motions induced by thermal and chemical convection and strongly influenced by rotational and magnetic field effects. While frequent observations are made of the morphology and time-dependent field behavior, flow dynamics in the core are all but inaccessible to direct measurement. Thus, forward models are essential for exploring the relationship between the geomagnetic field and its underlying fluid physics. The goal of my PhD is to further our understanding of the fluid physics driving the geodynamo.

In order to do this, I have performed a suite of nonrotating and rotating convection laboratory experiments and developed a new experimental device that reaches more extreme values of the governing parameters than previously possible. In addition, I conduct a theoretical analysis of well-established results from a suite of dynamo simulations by Christensen and Aubert (2006). These studies are conducted at moderate values of the Ekman number (ratio between viscosity and Coriolis forces, ~ 10^−4), as opposed to the the extremely small Ekman numbers in planetary cores (~ 10^−15). At such moderate Ekman values, flows tend to take the form of large-scale, quasi-laminar axial columns. These columnar structures give the induced magnetic field a dipolar morphology, similar to what is seen on planets. However, I find that some results derived from these simulations are fully dependent on the fluid viscosity, and therefore are unlikely to reflect the fluid physics driving dynamo action in the core. My findings reinforce the need to understand the turbulent processes that arise as the governing parameters approach planetary values. Indeed, my rotating convection experiments show that, as the Ekman number is decreased beyond ranges currently accessible to dynamo simulations, the regime characterized by laminar columns is found to dwindle. We instead find a large variety of behavioral regimes ranging from axial columns to fully three-dimensional turbulence. By comparing these to direct numerical simulations and asymptotically-reduced models, we find broad agreement in both the heat transfer scaling properties and flow morphologies in these separate regimes. In particular, large, multi-scale axial vortices emerge consistently in numerical and asymptotic simulations. Such multi-scale structures in the core may be related to the Earth’s dipolar magnetic field structure. I have designed and fabricated a novel laboratory experimental device capable of characterizing these flow regimes in great detail using accurate heat transfer and velocity measurements and high-resolution flow imaging.