Probabilistic modelling of substorm occurrences with an echo state network

Nakano, Shin'ya; Kataoka, Ryuho; Nosé, Masahito; Gjerloev, Jesper W.

doi:https://doi.org/10.5194/angeo-41-529-2023

Articles | Volume 41, issue 2

https://doi.org/10.5194/angeo-41-529-2023

© Author(s) 2023. This work is distributed under
the Creative Commons Attribution 4.0 License.

https://doi.org/10.5194/angeo-41-529-2023

© Author(s) 2023. This work is distributed under
the Creative Commons Attribution 4.0 License.

Articles | Volume 41, issue 2

Regular paper

|

21 Nov 2023

Regular paper |

| 21 Nov 2023

Probabilistic modelling of substorm occurrences with an echo state network

Shin'ya Nakano, Ryuho Kataoka, Masahito Nosé, and Jesper W. Gjerloev

Data sets

Geomagnetic Wp index, Data Analysis Center for Geomagnetism and Space Magnetism World Data Center for Geomagnetism, Kyoto and M. Nose https://doi.org/10.17593/13437-46800

Geomagnetic AE index, Data Analysis Center for Geomagnetism and Space Magnetism World Data Center for Geomagnetism, Kyoto, M. Nosé, T. Iyemori, M. Sugiura, and T. Kamei https://doi.org/10.17593/15031-54800

The SuperMAG data processing technique (https://supermag.jhuapl.edu/) J. W. Gjerloev https://doi.org/10.1029/2012JA017683

One min and 5-min solar wind data sets at the Earth’s bow shock nose J. King and N. Papitashvili https://omniweb.gsfc.nasa.gov/html/HROdocum.html

Short summary

Substorms are a phenomenon in the magnetosphere–ionosphere system, which are characterised by brightening of an auroral arc and enhancement of electric currents in the polar ionosphere. Since substorms are difficult to predict, this study treats a substorm occurrence as a stochastic phenomenon and represents the substorm occurrence rate with a machine learning model. We then analyse the response of substorm activity to solar wind conditions by feeding synthetic solar wind data into the model.