Home » Zhelavskaya et al. 2017

Empirical Modeling of the Plasmasphere Dynamics Using Neural Networks

Zhelavskaya I. S., Y. Y. Shprits, M. Spasojević, (2017), Empirical Modeling of the Plasmasphere Dynamics Using Neural Networks, J. of Geophys. Res. [Space Physics], 122, 11,227-11,244, doi:10.1002/2017JA024406

Abstract

Abstract We present the PINE (Plasma density in the Inner magnetosphere Neural network-based Empirical) model a new empirical model for reconstructing the global dynamics of the cold plasma density distribution based only on solar wind data and geomagnetic indices. Utilizing the density database obtained using the NURD (Neural-network-based Upper hybrid Resonance Determination) algorithm for the period of 1 October 2012 to 1 July 2016, in conjunction with solar wind data and geomagnetic indices, we develop a neural network model that is capable of globally reconstructing the dynamics of the cold plasma density distribution for 2≤L≤6 and all local times. We validate and test the model by measuring its performance on independent data sets withheld from the training set and by comparing the model-predicted global evolution with global images of He+ distribution in the Earth's plasmasphere from the IMAGE Extreme UltraViolet (EUV) instrument. We identify the parameters that best quantify the plasmasphere dynamics by training and comparing multiple neural networks with different combinations of input parameters (geomagnetic indices, solar wind data, and different durations of their time history). The optimal model is based on the 96 h time history of Kp, AE, SYM-H, and F10.7 indices. The model successfully reproduces erosion of the plasmasphere on the nightside and plume formation and evolution. We demonstrate results of both local and global plasma density reconstruction. This study illustrates how global dynamics can be reconstructed from local in situ observations by using machine learning techniques.

Authors (sorted by name)

Shprits Spasojevic Zhelavskaya

Journal / Conference

Journal Of Geophysical Research (Space Physics)

Acknowledgments

This work was funded by the NASA Heliophysics Guest Investigator Program under NASA grant NNX07AG48G, P.I. Jerry Goldstein. The electron density data set was derived by the NURD algorithm and is available at ftp://rbm.epss.ucla.edu/ftpdisk1/NURD. This work was supported by Helmholtz‐Gemeinschaft (HGF) [10.13039/501100001656], NSF GEM AGS‐1203747, NASA grant NNX12AE34G, NASA grant NNX16AF91G, and project PROGRESS funded by EC | Horizon 2020 Framework Programme (H2020) [10.13039/100010661] (637302)). The research has been partially funded by Deutsche Forschungsgemeinschaft (DFG) through grant SFB 1294. The work at Stanford was supported by NASA award NNX15Al94G.

Grants

1294 637302 AGS-1203747 NNX07AG48G NNX12AE34G NNX15Al94G NNX16AF91G

Bibtex

@article{doi:10.1002/2017JA024406,
author = {Zhelavskaya, Irina S. and Shprits, Yuri Y. and Spasojević, Maria},
title = {Empirical Modeling of the Plasmasphere Dynamics Using Neural Networks},
journal = {Journal of Geophysical Research: Space Physics},
volume = {122},
number = {11},
pages = {11,227-11,244},
keywords = {plasmasphere, inner magnetosphere, models, neural networks, machine learning},
doi = {10.1002/2017JA024406},
url = {https://agupubs.onlinelibrary.wiley.com/doi/abs/10.1002/2017JA024406},
eprint = {https://agupubs.onlinelibrary.wiley.com/doi/pdf/10.1002/2017JA024406},
abstract = {Abstract We present the PINE (Plasma density in the Inner magnetosphere Neural network-based Empirical) model a new empirical model for reconstructing the global dynamics of the cold plasma density distribution based only on solar wind data and geomagnetic indices. Utilizing the density database obtained using the NURD (Neural-network-based Upper hybrid Resonance Determination) algorithm for the period of 1 October 2012 to 1 July 2016, in conjunction with solar wind data and geomagnetic indices, we develop a neural network model that is capable of globally reconstructing the dynamics of the cold plasma density distribution for 2≤L≤6 and all local times. We validate and test the model by measuring its performance on independent data sets withheld from the training set and by comparing the model-predicted global evolution with global images of He+ distribution in the Earth's plasmasphere from the IMAGE Extreme UltraViolet (EUV) instrument. We identify the parameters that best quantify the plasmasphere dynamics by training and comparing multiple neural networks with different combinations of input parameters (geomagnetic indices, solar wind data, and different durations of their time history). The optimal model is based on the 96 h time history of Kp, AE, SYM-H, and F10.7 indices. The model successfully reproduces erosion of the plasmasphere on the nightside and plume formation and evolution. We demonstrate results of both local and global plasma density reconstruction. This study illustrates how global dynamics can be reconstructed from local in situ observations by using machine learning techniques.},
year = {2017}
}