Solar origin of geomagnetic storms and prediction of storms with the use of neural networks

This review deals with how the changes of the large-scale solar magnetic fields are related to the occurrence of solar phenomena, which are associated with geomagnetic storms. The review also describes how artificial neural networks have been used to forecast geomagnetic storms either from daily solar input data or from hourly solar wind data. With solar data as input predictions 1–3 days or a month in advance are possible, while using solar wind data as input predictions about an hour in advance are possible. The predictions one hour ahead of the geomagnetic storm indexD_st from only solar wind input data have reached such high accuracy, that they are of practical use in combination with real-time solar wind observations at L1. However, the predictions days and a month ahead need to be much improved in order to be of real practical use.