Data assimilation techniques to probe the interiors of rotating MHD experiments and link with planets

Many experiments around the world, including in our lab in Zurich,
seek to investigate the fluid dynamics of rapidly rotating flows, in
either hydrodynamical (e.g. using water) or magnetohydrodynamical (e.g.
using sodium) settings. Such experiments are an exciting avenue with
which to understand convection and magnetic field generation in
planetary settings, and are often able to mimic the relevant physics in
the most realistic manner, often exceeding the reality of computer
simulations. Successful experiments include DTS in Grenoble, "Big
Sister" in Maryland and our own burgeoning experiment SpiNaCH in Zurich.
Of paramount importance is the ability to diagnose the flows that are
generated using non-invasive techniques. One successful technique is to
use magnetic sensors and detect the induced magnetic field that is
generated by the experiment. One must then use the observations together
with a dynamical model that represents the physics of the experiment.
One can then use the techniques of data assimilation (DA), which we have been
at the forefront of developing [1,2] in the context of planetary cores.
Our results thus far are very encouraging. We have never applied the technique
to experimental data sets. By applying the technique we hope to learn about
induction processes in the experimental interior. A rich and complex data set is
presented in [3], and it requires explanation. We will be part of an international
effort to develop DA in the context of this data set, and to develop
state-of-the art techniques that can be applied in our own experiment SpiNaCH.
The net aim is to improve the resolution of velocity fields in liquid
metal experiments, and as a result comprehend the fluid dynamics of
rapidly rotating fluid subject to strong magnetic forces.
The project will appeal to a candidate with a background in one of physics, geophysics,
engineering, mathematics or computer science. No prior knowledge of the subject area
is required.
[1] K Li, A Jackson, & P Livermore,
Variational data assimilation for the initial value dynamo problem,
Physical Review E, 84, 056321 (2011 ) doi: 10.1103/PhysRevE.84.056321

[2] Li, K., Jackson, A. & Livermore, P. W., 2014. Variational data assimilation for a forced,^Minertia-free magnetohydrodynamic dynamo model, Geophys. J. Int. 199, 1662-1676.

[3] Daniel S. Zimmerman, Santiago Andrés Triana, Henri-Claude Nataf and Daniel P. Lathrop
A turbulent, high magnetic Reynolds number experimental model of Earth’s core
ournal of Geophysical Research: Solid Earth
Volume 119, Issue 6, pages 4538–4557, June 2014
doi: 10.1002/2013JB010733