Home » Zhelavskaya et al. 2021

A Combined Neural Network- and Physics-Based Approach for Modeling Plasmasphere Dynamics

Zhelavskaya I. S., N. A. Aseev, Y. Y. Shprits, (2021), A Combined Neural Network- and Physics-Based Approach for Modeling Plasmasphere Dynamics, J. of Geophys. Res. [Space Physics], 126, e2020JA028077, doi:10.1029/2020JA028077, e2020JA028077 2020JA028077

Abstract

Abstract In recent years, feedforward neural networks (NNs) have been successfully applied to reconstruct global plasmasphere dynamics in the equatorial plane. These neural network-based models capture the large-scale dynamics of the plasmasphere, such as plume formation and erosion of the plasmasphere on the nightside. However, their performance depends strongly on the availability of training data. When the data coverage is limited or non-existent, as occurs during geomagnetic storms, the performance of NNs significantly decreases, as networks inherently cannot learn from the limited number of examples. This limitation can be overcome by employing physics-based modeling during strong geomagnetic storms. Physics-based models show a stable performance during periods of disturbed geomagnetic activity if they are correctly initialized and configured. In this study, we illustrate how to combine the neural network- and physics-based models of the plasmasphere in an optimal way by using data assimilation. The proposed approach utilizes advantages of both neural network- and physics-based modeling and produces global plasma density reconstructions for both quiet and disturbed geomagnetic activity, including extreme geomagnetic storms. We validate the models quantitatively by comparing their output to the in-situ density measurements from RBSP-A for an 18-month out-of-sample period from June 30, 2016 to January 01, 2018 and computing performance metrics. To validate the global density reconstructions qualitatively, we compare them to the IMAGE EUV images of the He+ particle distribution in the Earth's plasmasphere for a number of events in the past, including the Halloween storm in 2003.

Authors (sorted by name)

Aseev Shprits Zhelavskaya

Journal / Conference

Journal Of Geophysical Research (Space Physics)

Bibtex

@article{https://doi.org/10.1029/2020JA028077,
author = {Zhelavskaya, I. S. and Aseev, N. A. and Shprits, Y. Y.},
title = {A Combined Neural Network- and Physics-Based Approach for Modeling Plasmasphere Dynamics},
journal = {Journal of Geophysical Research: Space Physics},
volume = {126},
number = {3},
pages = {e2020JA028077},
keywords = {data assimilation, Kalman filter, machine learning, neural networks, plasmasphere, plasma density},
doi = {https://doi.org/10.1029/2020JA028077},
url = {https://agupubs.onlinelibrary.wiley.com/doi/abs/10.1029/2020JA028077},
eprint = {https://agupubs.onlinelibrary.wiley.com/doi/pdf/10.1029/2020JA028077},
note = {e2020JA028077 2020JA028077},
abstract = {Abstract In recent years, feedforward neural networks (NNs) have been successfully applied to reconstruct global plasmasphere dynamics in the equatorial plane. These neural network-based models capture the large-scale dynamics of the plasmasphere, such as plume formation and erosion of the plasmasphere on the nightside. However, their performance depends strongly on the availability of training data. When the data coverage is limited or non-existent, as occurs during geomagnetic storms, the performance of NNs significantly decreases, as networks inherently cannot learn from the limited number of examples. This limitation can be overcome by employing physics-based modeling during strong geomagnetic storms. Physics-based models show a stable performance during periods of disturbed geomagnetic activity if they are correctly initialized and configured. In this study, we illustrate how to combine the neural network- and physics-based models of the plasmasphere in an optimal way by using data assimilation. The proposed approach utilizes advantages of both neural network- and physics-based modeling and produces global plasma density reconstructions for both quiet and disturbed geomagnetic activity, including extreme geomagnetic storms. We validate the models quantitatively by comparing their output to the in-situ density measurements from RBSP-A for an 18-month out-of-sample period from June 30, 2016 to January 01, 2018 and computing performance metrics. To validate the global density reconstructions qualitatively, we compare them to the IMAGE EUV images of the He+ particle distribution in the Earth's plasmasphere for a number of events in the past, including the Halloween storm in 2003.},
year = {2021}
}