Abstract
Estimates of the geomagnetic indices made with real-time solar wind measurements form the basis of many space weather forecast techniques. We analyze 20 years of hourly AL and OMNI solar wind data to determine geomagnetic importance of various solar wind and IMF parameters. Besides the solar wind driver of primary importance ( VB s), the first-order contributions, significantly increasing the quality of the model are: solar wind velocity, 2 h of solar wind prehistory and 1 h of AL history. The factors of secondary importance, marginally contributing to overall statistical quality, are IMF B y , solar wind density, and IMF fluctuations. The dynamic pressure is geomagnetically effective only if the pressure is lower than the average. Modelling of the same data set with an artificial neural network (ANN) confirmed our selection of important factors. Statistically the ANN model was just marginally better than our analytic expression E = V B y 2 / 2 + B z 2 sin 4 ( θ / 2 ) + α V 2 sin 0.5 ( θ / 2 ) . The AU index dependence is principally different from AL in several respects; therefore modelling of the AE composite index is physically misleading.
Published Version
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call