Magnetic field lines inside Earth’s core for 2001, as estimated by data assimilation

North polar view of Earth's core surface flow and density anomaly

North Polar view of Earth’s core, with the core surface flow estimated in 2015 (arrows) and the corresponding density anomalies (green-brown, brown is lighter fluid) driving this flow (from Aubert, 2020)

Inverse problems and data assimilation

By combining the available geomagnetic data with statistics extracted from numerical models of the geodynamo, it is possible to determine the most likely internal core state given the data and the numerical model behaviour. This forms the core of the geomagnetic inverse problem: determine the internal magnetic field, fluid flow and density anomaly field of the geodynamo from the available data. Data assimilation introduces a temporal dimension to the geomagnetic inverse problem. This can be used for instance to improve our knowledge of the geodynamo in the past and present (reanalysis), or in the future (forecast).

Forecast for the evolution of the geomagnetic field intensity (contours in microteslas) at the surface of the Earth in the next century. The South Atlantic anomaly (in blue) will enlarge, deepen and drift westward in the future (from Aubert, 2015)

Operational modelling of the geodynamo

The geomagnetic field interacts with a number of human technological activities, such as low-earth orbiting satellite operations, navigation systems, exploration geophysics and the protection of large-scale electrical infrastructure. Questions of societal interest thus include: is the geomagnetic dipole decay that is observed since 1840 set to continue? Will the low-intensity anomaly in the South-Atlantic region continue to deepen? Obviously, the geodynamo is a nonlinear system which suffers from the well-known butterfly effect also present in the atmosphere: infinitesimal errors in the determination of the initial state lead to macroscopic errors after some time. That time, which is a few days in the atmosphere, is on the order of a few decades for the geodynamo.

Geomagnetic data assimilation combining Earth-like, state of the art numerical models together with high-quality data such as provided by ESA’s Swarm mission provides answers to these questions. Our data assimilation frameworks are able to provide insight into the kinematic and dynamic mechanisms responsible for the dipole decay, and to predict for how long the presently observed mechanisms will continue to operate. For the next century we predict a continuation of the geomagnetic dipole decrease at a rate close to that observed during the last 175 years, together with an enlargement, deepening and westward drift of the South Atlantic anomaly (see above), which will correspondingly enlarge the problematic flight zone for satellite, where the failure hazard is increased due to insufficient protection from solar wind by the magnetosphere. 

Numerical geodynamo simulations and data assimilations can also help diagnosing the present magnetic field, and since 2015 enter the production of the International Reference Geomagnetic Field, which is the main tool used for industrial purposes requiring knowledge of this field.

Selected references

Aubert, J.: Recent geomagnetic variations and the force balance in Earth’s core, Geophys. J. Int. 221, 378-393, 2020, doi: 10.1093/gji/ggaa007.

Finlay, C., Aubert, J. and Gillet, N.: Gyre-driven decay of Earth’s magnetic dipole, Nature Communications 7, 10422, 2016, doi: 10.1038/ncomms10422.

Aubert, J.: Geomagnetic forecasts driven by thermal wind dynamics in Earth’s core, Geophys. J. Int. 203, 1738-1751, 2015, doi: 10.1093/gji/ggv394 

Thébault, E., and al.: International Geomagnetic Reference Field: the 12th generation. Earth, Planets, Space 67, 79, 2015, doi: 10.1186/s40623-015-0228-9 

Hulot, Lhuillier, Aubert: Earth's dynamo limit of predictability, Geophys. Res. Lett 37, doi:10.1029/2009GL041869, 2010.

Fournier et al. : An Introduction to Data Assimilation and Predictability in Geomagnetism , Space Science Reviews, 155, 247-291, 2010, doi: 10.1007/s11214-010-9669-4