Numerical and experimental geodynamo models offer complementary insights into the dynamics of the Earth's liquid iron core. The goal of this research project is to leverage the strengths of both approaches to gain significant insights into the origin and evolution of the Earth's magnetic field. The experiments produce geophysically realistic turbulence in liquid metal flows that are well beyond the spatial resolution of direct numerical simulations. Experimental observations of highly nonlinear interactions between the flow and the magnetic field provide unique information about the complexity of important processes in the core. Parallel advances in computation have produced increasingly sophisticated models for unresolved (subgrid-scale) turbulence. These models have been implemented and tested in numerical geodynamo calculations with remarkable success. However, the effort to push these models to Earth-like conditions is presently hampered by the lack of "known" solutions to test the predictions of the subgrid-scale models. Use of carefully chosen diagnostics from the experiments enables more realistic tests of the numerical models. The models, in turn, aid the interpretation of the experiments, which establishes a foundation for refining the assumptions used in the construction of the subgrid-scale models. This synergy between computations and experiments is most effectively realized through a collaborative research program.
The experiments used in this study include spherical Couette flow and rotating thermal convection. In either case, an externally imposed magnetic field interacts with the liquid metal flow. The experimental setups are based on prior experience with the explicit goal of producing distinctive large-scale flows that are sensitive to the presence of turbulence. Experimental observations of these large-scale features are used to test the predictions of the numerical models. Spherical Couette flow permits strong (and realistic) interactions between the flow and the magnetic field. Three of the four key subgrid-scale models in the geodynamo problem are being tested under the conditions of rapid rotation and strong magnetic fields by measuring the pattern of induced magnetic field outside the liquid metal. Rotating convection experiments, with and without an imposed magnetic field, are being used to test the fourth subgrid-scale model. The primary observations in the convection experiments include the total heat flow, the large-scale velocity and the power spectra of temperature fluctuations. The expected outcome of this research project is a more realistic geodynamo model that incorporates sophisticated parameterizations for turbulence and a database of experimental results to stimulate other research groups to engage in similar comparisons using different modeling strategies.
