LPP - Laboratoire de Physique des Plasmas - UMR 7648 - Machine learning paves the way for a massive multi-mission statistical study and analytical modeling of the Earth’s magnetopause

Aller au menu
Aller à la recherche
Aller au contenu
Aller à la navigation

Accueil > A propos du LPP > Communication > Actualités archivées > 2022 > Machine learning paves the way for a massive multi-mission statistical study and analytical modeling of the Earth’s magnetopause

Machine learning paves the way for a massive multi-mission statistical study and analytical modeling of the Earth’s magnetopause

Toutes les versions de cet article : [English] [français]

The Earth magnetopause results from the interaction between the solar wind and the Earth magnetic field. It acts like an obstacle for the upcoming supersonic solar wind and is located downstream a collisionless bow shock across which the solar wind becomes subsonic.
Despite decades of observations of this boundary by spacecraft orbiting the Earth, the researchers still poorly understands how the solar wind and interplanetary magnetic field (IMF) properties control the location and shape of the magnetopause. . Gautier Nguyen (who defended his PhD in February 2021 at LPP in the space plasma team) and his collaborators at LPP, IRAP in Toulouse and the Laboratory of Astrophysics of Bordeaux recently performed a massive multi-mission statistical study and brought new answers to this question in a series of papers published in Journal of Geophysical Research.

Detecting in the data all possible times at which satellites have crossed the magnetopause is the first step towards a statistical representation of its location and shape. Such a task has so far mostly been performed through visual inspection of the data, which, considering its variability and complexity, results in quite subjective, hardly reproducible and small lists of crossing, thus hampering the building of large scale statistics.

Gautier Nguyen and his colleagues have solved this issue with an innovative approach, based on teaching an algorithm to recognize in which region of the near-Earth environment a spacecraft is at a given time. The principle is detailed in a newly published technical report in JGR Space Physics (Nguyen et al. 2021a). Using their new method on the 83 years of cumulated data from the missions THEMIS, ARTEMIS, Cluster, Magnetospheric MultiScale and Double Star they identified 15062 1 hour time intervals enclosing a magnetopause crossings, constituting the most exhaustive catalog existing so far and which spatial representation is shown in the Figure 1.

With this unique and now publicly accessible catalog, the authors show evidences that the magnetopause twists as the IMF orientation rotate around the Sun-Earth axis, a phenomenon already suggested by global magnetohydrodynamics numerical models but never seen in data. They also brought a final answer to the long standing debate regarding the presumed - and now confirmed - indentation of the magnetopause in the cusp region, where the geomagnetic field lines converge towards the magnetic dipole axis. Results also confirmed for the first time that the shape of the magnetopause, as predicted by analytical models, is consistent with its observations in the far night side of the magnetosphere, in the so-called geomagnetic tail, as far as the lunar orbit. Following their new findings, the authors finally proposed a new way to predict analytically the shape and location of the magnetopause in a more accurate fashion than previous models.

A representation of this model, accounting for the time variation of the solar wind and IMF conditions, is shown in the Figure 2.

The four mentioned articles form a series of companion papers recently published in the same issue of JGR : Space Physics under the name Massive Multi-Mission Statistical Study and Analytical Modeling of the Earth’s Magnetopause. These articles can either be considered separately or as a whole.

References :

Nguyen, G., Aunai, N., Michotte de Welle, B., Jeandet, A., Lavraud, B., & Fontaine, D. (2022). Massive multi-mission statistical study and analytical modeling of the Earth’s magnetopause : 1. A gradient boosting based automatic detection of near-Earth regions. Journal of Geophysical Research : Space Physics, 127, e2021JA029773. https://doi.org/10.1029/2021JA029773

Nguyen, G., Aunai, N., Michotte de Welle, B., Jeandet, A., Lavraud, B., & Fontaine, D. (2022). Massive multi-mission statistical study and analytical modeling of the Earth’s magnetopause : 2. Shape and location. Journal of Geophysical Research : Space Physics, 127, e2021JA029774. https://doi.org/10.1029/2021JA029774

Nguyen, G., Aunai, N., Michotte de Welle, B., Jeandet, A., Lavraud, B., & Fontaine, D. (2022). Massive multi-mission statistical study and analytical modeling of the Earth’s magnetopause : 3. An asymmetric non indented magnetopause analytical model. Journal of Geophysical Research : Space Physics, 127, e2021JA030112. https://doi.org/10.1029/2021JA030112

Nguyen, G., Aunai, N., Michotte de Welle, B., Jeandet, A., Lavraud, B., & Fontaine, D. (2022). Massive multi-mission statistical study and analytical modeling of the Earth’s magnetopause : 4. On the near-cusp magnetopause indentation. Journal of Geophysical Research : Space Physics, 127, e2021JA029776. https://doi.org/10.1029/2021JA029776

Dans la même rubrique :

Rechercher

Menu

Machine learning paves the way for a massive multi-mission statistical study and analytical modeling of the Earth’s magnetopause