The magnetopause location as observed by the INTERBALL-1 satellite

We evaluate the uncertainty of the empirical magnetopause models Petrinec and Russell, (1996); Shue et al. (1997, 1998) used in forecasting of the magnetospheric boundary positions. The data set includes single magnetopause crossings and multiple ones with various duration, which were determined from INTERBALL-1 measurements during 1995–1997 years. The solar wind conditions were obtained from plasma and magnetic field measurements from WIND spacecraft. We show that the best prediction gives the model Shue et al. (1998), but the differences between considered models are rather small. Approximately 50% of the observed magnetopause crossings have deviations from the model predictions more than 1Re. Sometimes these deviations may be as large as 5–6Re. For high latitude magnetopause crossings due to the cusp influence more often the real measured boundary is located more closer to the Earth than predicted one.