- 1GFZ Helmholtz Centre for Geosciences, Space Physics and Space Weather section, Germany
- 2Institute of Physics and Astronomy, University of Potsdam, Postdam, Germany
- 3Department of Earth, Planetary, and Space Sciences, University of California, Los Angeles, CA, USA
- 4Ludwig Maximilian University of Munich, Munich, Germany
- 5Graduate School of Natural Science and Technology, Kanazawa University, Kanazawa, Japan
- 6Planetary Plasma and Atmospheric Research Center, Tohoku University, Tohoku, Japan
- 7Institute for Space-Earth Environmental Research, Nagoya University, Nagoya, Japan
- 8Data Analysis Center for Geomagnetism and Space Magnetism, Kyoto University, Kyoto, Japan
- 9Graduate School of Engineering, Kyushu Institute of Technology, Kitakyushu, Japan
- 10Institute of Space and Astronautical Science, JapanAerospace Exploration Agency, Sagamihara, Japan
Abstract
PINE (Plasma density in the Inner magnetosphere Neural network-based Empirical model) [4] is
a previously developed neural network model that uses RBSP [3] data and geomagnetic indices to
capture the global dynamics of cold plasma density in the plasmasphere. In this study, we enhance
PINE by incorporating additional data from the ERG (Exploration of Energization and Radiation
in Geospace) ARASE mission [1, 2], alongside the existing RBSP dataset. The updated model
is rigorously validated using a withheld test set and further evaluated through comparison with
global hydrogen ion distribution images obtained by the IMAGE (Imager for Magnetopause-to-
Aurora Global Exploration) mission. Model performance is analyzed under varying geomagnetic
conditions, including quiet periods, disturbed intervals, and extreme space weather events. Inte-
grating Arase data improves modeling of the inner magnetosphere, extending PINE’s applicability
to lower altitudes in the ionosphere by covering low L-shell regions and enhancing predictions of
the plasmapause configuration.
References
[1] Atsushi, K., Fuminori, T., Hirotsugu, K., Shoya, M., Ayako, M., Mariko, T., Masafumi, S., Satoko, N., Masahiro, K., Yoshizumi, M., et al., 2021. Exploration of energization and radiation in geospace (erg) plasma wave experiment (pwe) high frequency analyzer (hfa) level-3 electron density data. DOI: 10.34515/DATA.ERG-10001.
[2] Kasahara, Y., Kumamoto, A., Tsuchiya, F., Kojima, H., Matsuda, S., Matsuoka, A., Teramoto, M., Shoji, M., Nakamura, S., Kitahara, M., et al., 2021. The pwe/hfa instrument level-3 electron density data of exploration of energization and radiation in geospace (erg) arase satellite. ERG Sci. Cent. DOI: 10.34515/DATA.ERG-10001 1.
[3] Mauk, B., Fox, N.J., Kanekal, S., Kessel, R., Sibeck, D., Ukhorskiy, a.A., 2014. Science objectives and rationale for the radiation belt storm probes mission. The van Allen probes mission, 3–27. DOI: 10.1007/978-1-4899-7433-4_2.
[4] Zhelavskaya, I.S., Shprits, Y.Y., Spasojevi´c, M., 2017. Empirical modeling of the plasmasphere dynamics using neural networks. Journal of Geophysical Research: Space Physics 122, 11–227. DOI: 10.1002/2017JA024406
How to cite: Shahsavani, S., Shprits, Y. Y., Bianco, S., Haas, B., Smirnov, A., Kasahara, Y., Tsuchiya, F., Kumamoto, A., Shinbori, A., Matsuoka, A., Teramoto, M., Yamamoto, K., Shinohara, I., and Miyoshi, Y.: Enhancing PINE Model Performance through Database Extension, EGU General Assembly 2025, Vienna, Austria, 27 Apr–2 May 2025, EGU25-17180, https://doi.org/10.5194/egusphere-egu25-17180, 2025.
