Abstract
Accurate and timely detection of bow shock and magnetopause crossings is essential for understanding the dynamics of a planet’s magnetosphere. However, for Mercury, due to the variable nature of its magnetosphere, this remains a challenging task. Existing approaches based on geometric equations only provide average boundary shapes, and can be hard to generalise to environments with variable conditions. On the other hand, data-driven methods require large amounts of annotated data to account for variations, which can scale up the costs quickly. We propose to solve this problem with machine learning. To this end, we introduce a suitable dataset, prepared by processing raw measurements from NASA’s MESSENGER (MErcury Surface, Space Environment, GEochemistry, and Ranging) mission and design a five-class supervised learning problem. We perform an architectural search to find a suitable model, and report our best model, a Convolutional Recurrent Neural Network (CRNN), achieves a macro F1 score of 0.82 with accuracies of approximately 80% and 88% on the bow shock and magnetopause crossings, respectively. Further, we introduce an approach based on active learning that includes only the most informative orbits from the MESSENGER dataset measured by Shannon entropy. We observe that by employing this technique, the model is able to obtain near maximal information gain by training on just two Mercury years worth of data, which is about 10% of the entire dataset. This has the potential to significantly reduce the need for manual labeling. This work sets the ground for future machine learning endeavors in this direction and may be highly relevant to future missions such as BepiColombo, which is expected to enter orbit around Mercury in December 2025.
This is a preview of subscription content, log in via an institution to check access.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Notes
- 1.
The apoapsis of an elliptic orbit is the point farthest away from the planet.
- 2.
The position of the event in the window should not matter.
- 3.
\( \triangle ^{n-1} {:}{=}\{(p_1, p_2, \dots , p_n) \in \mathbb {R}^n \mid \forall i: p_i \ge 0, \sum _{i = 1}^{n}p_i = 1\} \subseteq [0, 1]^n\).
- 4.
All plots for the entire test set are made available in the code repository linked in Sect. 1.
References
Alexeev, I.I., Belenkaya, E.S., Bobrovnikov, S.Y., Kalegaev, V.V.: Modelling of the electromagnetic field in the interplanetary space and in the earth’s magnetosphere. Space Sci. Rev. 107(1), 7–26 (2003)
Alexeev, I.I., et al.: Mercury’s magnetospheric magnetic field after the first two messenger flybys. Icarus 209(1), 23–39 (2010)
Alexeev, I., et al.: A global magnetic model of saturn’s magnetosphere and a comparison with cassini soi data. Geophys. Res. Lett. 33(8), 1–4 (2006)
Amiriparian, S., et al.: Recognition of echolalic autistic child vocalisations utilising convolutional recurrent neural networks (2018)
Amiriparian, S., Cummins, N., Julka, S., Schuller, B.: Deep convolutional recurrent neural network for rare acoustic event detection. In: Proceedings of DAGA, pp. 1522–1525 (2018)
Anderson, B.J., et al.: The magnetic field of mercury. Space Sci. Rev. 152(1), 307–339 (2010)
Belenkaya, E., Bobrovnikov, S.Y., Alexeev, I., Kalegaev, V., Cowley, S.: A model of jupiter’s magnetospheric magnetic field with variable magnetopause flaring. Planet. Space Sci. 53(9), 863–872 (2005)
Benkhoff, J., et al.: Bepicolombo-comprehensive exploration of mercury: mission overview and science goals. Planet. Space Sci. 58(1–2), 2–20 (2010)
Fairfield, D.H.: Average and unusual locations of the earth’s magnetopause and bow shock. J. Geophys. Res. 76(28), 6700–6716 (1971)
Haaland, S., et al.: Characteristics of the flank magnetopause: mms results. J. Geophys. Res.: Space Phys. 125(3), e2019JA027623 (2020)
Jelínek, K., Němeček, Z., Šafránková, J.: A new approach to magnetopause and bow shock modeling based on automated region identification. J. Geophys. Res.: Space Phys. 117(A5) (2012)
Johnson, C.L., et al.: Messenger observations of mercury’s magnetic field structure. J. Geophys. Res.: Planets 117(E12) (2012)
Kraeft, S.K.: Detection and analysis of cancer cells in blood and bone marrow using a rare event imaging system. Clin. Cancer Res. 6(2), 434–442 (2000)
Lin, R., Zhang, X., Liu, S., Wang, Y., Gong, J.: A three-dimensional asymmetric magnetopause model. J. Geophys. Res.: Space Phys. 115(A4) (2010)
Nguyen, G., Aunai, N., Michotte de Welle, B., Jeandet, A., Fontaine, D.: Automatic detection of the earth bow shock and magnetopause from in-situ data with machine learning. In: Annales Geophysicae Discussions, pp. 1–22 (2019)
Nikolaou, N., et al.: Lessons learned from the 1st ariel machine learning challenge: correcting transiting exoplanet light curves for stellar spots. arXiv preprint arXiv:2010.15996 (2020)
Philpott, L.C., Johnson, C.L., Anderson, B.J., Winslow, R.M.: The shape of mercury’s magnetopause: the picture from messenger magnetometer observations and future prospects for bepicolombo. J. Geophys. Res.: Space Phys. 125(5), e2019JA027544 (2020)
Ren, P., et al.: A survey of deep active learning. ACM Comput. Surv. (CSUR) 54(9), 1–40 (2021)
Settles, B.: Active learning. Synth. Lect. Artif. Intell. Mach. Learn. 6(1), 1–114 (2012)
Shue, J.H., et al.: A new functional form to study the solar wind control of the magnetopause size and shape. J. Geophys. Res.: Space Phys. 102(A5), 9497–9511 (1997)
Sibeck, D.G., Lopez, R., Roelof, E.C.: Solar wind control of the magnetopause shape, location, and motion. J. Geophys. Res.: Space Phys. 96(A4), 5489–5495 (1991)
Slavin, J.A.: Mercury’s magnetosphere. Adv. Space Res. 33(11), 1859–1874 (2004)
Wang, Y., et al.: A new three-dimensional magnetopause model with a support vector regression machine and a large database of multiple spacecraft observations. J. Geophys. Res.: Space Phys. 118(5), 2173–2184 (2013)
Winslow, R.M., et al.: Mercury’s magnetopause and bow shock from messenger magnetometer observations. J. Geophys. Res.: Space Phys. 118(5), 2213–2227 (2013)
Zhong, J.: Mercury’s three-dimensional asymmetric magnetopause. J. Geophys. Res.: Space Phys. 120(9), 7658–7671 (2015)
Zurbuchen, T.H., et al.: Messenger observations of the spatial distribution of planetary ions near mercury. Science 333(6051), 1862–1865 (2011)
Acknowledgements
The authors acknowledge support from Europlanet 2024 RI that has received funding from the European Union’s Horizon 2020 research and innovation programme under grant agreement No. 871149.
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2023 The Author(s), under exclusive license to Springer Nature Switzerland AG
About this paper
Cite this paper
Julka, S., Kirschstein, N., Granitzer, M., Lavrukhin, A., Amerstorfer, U. (2023). Deep Active Learning for Detection of Mercury’s Bow Shock and Magnetopause Crossings. In: Amini, MR., Canu, S., Fischer, A., Guns, T., Kralj Novak, P., Tsoumakas, G. (eds) Machine Learning and Knowledge Discovery in Databases. ECML PKDD 2022. Lecture Notes in Computer Science(), vol 13716. Springer, Cham. https://doi.org/10.1007/978-3-031-26412-2_28
Download citation
DOI: https://doi.org/10.1007/978-3-031-26412-2_28
Published:
Publisher Name: Springer, Cham
Print ISBN: 978-3-031-26411-5
Online ISBN: 978-3-031-26412-2
eBook Packages: Computer ScienceComputer Science (R0)
