Can machine learning reveal precursors of reversals of the geomagnetic axial dipole field?

• Published in: Geophysical Journal International
• Main Authors: Gwirtz, K., Davis, T., Morzfeld, M., Constable, C., Fournier, A., Hulot, G.
• Other Authors: Institut de Physique du Globe de Paris (IPGP (UMR_7154)), Institut national des sciences de l'Univers (INSU - CNRS)-Université de La Réunion (UR)-Institut de Physique du Globe de Paris (IPG Paris)-Centre National de la Recherche Scientifique (CNRS)-Université Paris Cité (UPCité), ANR-19-CE31-0019,revEarth,Modélisation réaliste des inversions du champ magnétique terrestre(2019)
• Format: Article in Journal/Newspaper
• Language: English
• Published: HAL CCSD 2022
• Subjects: Reversals: process; time scale; magnetostratigraphy; Magnetic field variations through time; Paleointensity; Time-series analysis; Dynamo: theories and simulations; Machine learning; [SDU]Sciences of the Universe [physics]; North Pole; South pole
• Online Access: https://insu.hal.science/insu-03748536; https://insu.hal.science/insu-03748536/document; https://insu.hal.science/insu-03748536/file/ggac195.pdf; https://doi.org/10.1093/gji/ggac195

International audience It is well known that the axial dipole part of Earth's magnetic field reverses polarity, so that the magnetic north pole becomes the south pole and vice versa. The timing of reversals is well documented for the past 160 Myr, but the conditions that lead to a reversal are still not well understood. It is not known if there are reliable "precursors" of reversals (events that indicate that a reversal is upcoming) or what they might be. We investigate if machine learning (ML) techniques can reliably identify precursors of reversals based on time series of the axial magnetic dipole field. The basic idea is to train a classifier using segments of time series of the axial magnetic dipole. This training step requires modification of standard ML techniques to account for the fact that we are interested in rare events - a reversal is unusual, while a non-reversing field is the norm. Without our tweak, the ML classifiers lead to useless predictions. Perhaps even more importantly, the usable observational record is limited to 0-2 Ma and contains only five reversals, necessitating that we determine if the data are even sufficient to reliably train and validate an ML algorithm. To answer these questions we use several ML classifiers (linear/nonlinear support vector machines and long short-term memory networks), invoke a hierarchy of numerical models (from simplified models to 3-D geodynamo simulations), and two paleomagnetic reconstructions (PADM2M and Sint-2000). The performance of the ML classifiers varies across the models and the observational record and we provide evidence that this is not an artifact of the numerics, but rather reflects how "predictable" a model or observational record is. Studying models of Earth's magnetic field via ML classifiers thus can help with identifying shortcomings or advantages of the various models. For Earth's magnetic field, we conclude that the ability of ML to identify precursors of reversals is limited, largely due to the small amount and low frequency ...