Abstract
Interaction of solar events propagating throughout interplanetary space with the magnetic eld of the Earth may result in disruption of the magnetosphere. Disruption of the magnetic eld is followed by the formation of the time-varying electric eld and thus electric current is induced in Earth-bound structures such as transmission networks, pipelines or railways.
In that case it is necessary to be able to predict future state of the magnetosphere and magnetic eld of the Earth. The most straightforward way would using geomagnetic indices.
Several studies are investigating the relationship of the response of the magnetosphere to changes in the solar wind with motivation to give a more accurate prediction of geomagnetic indices during geomagnetic storms. To forecast these indices, dierent approaches have been attempted from simple correlation studies to neural networks.
We studied the eects of interplanetary shocks observed at L1 on the Earth's magnetosphere with a database of 29 shocks between 2009 and 2019. Driving of the magnetosphere is described as integral of reconnection electric eld, Ey = vx x Bz for each shock.
The response of the geomagnetic eld is described with the SYM-H index. We created an algorithm in Python for prediction of the magnetosphere state based on the correlation of solar wind driving and magnetospheric response and found that typical time-lags range between tens of minutes to maximum 2 hours.