We have examined Pc 3-4 waves from Polar and mid-latitudes ground stations on May 11, 1999, when the solar wind density was well below 1 cm-3, and compared with those in the control day of May 14, when the solar wind density was normal and the IMF conditions were similar. The Pc 3-4 waves, clearly seen in the dayside magentosphere on May 14, were nearly absent on May 11 both in space as observed by Polar and on the ground at mid-latitudes, even though the foreshock geometry was favorable for the generation of Pc 3-4 waves in the magnetosphere for both days. The near absence of magnetospheric Pc 3-4 waves on May 11 is a result of weakened bow shock. Due to the low Mach number the bow shock was extremely weak on May 11. Combined with a very tenuous solar wind plasma, there was very few backstreaming particles in the foreshock, and thus, the upstream waves, although still present in the foreshock, were one order of magnitude weaker in power than usual. The upstream waves should be propagating sunward along the IMF in the plasma rest frame at Alfven velocity since there were very little compressional power on May 11. The Alfven velocity in the solar wind was unusually high due to the low solar wind density (VA=184 km/s for BIMF=5.9 nT and Nsw=0.5 cm-3). Thus the upstream waves were probably not being carried downstream to the magnetopause at the usual speed either. We believe that all these factors contributed to the Pc 3-4 waves not being observed in the magentosphere on May 11.
The simultaneous observations of Pc 3-4 waves in the upstream region, in the magnetosphere and on the ground in this study provide further support the solar wind source of magnetospheric Pc 3-4 waves. Previous observations have established the relationship between the IMF magnitude and the Pc 3-4 wave (as well as upstream wave) frequency and between the IMF cone angle and the Pc 3-4 wave occurrence [Bol'shakova and Troitskaya, 1968,Troitskaya et al., 1971,Gul'elmi, 1974]. Previous study also found correlations between the wave power and the solar wind velocity, which was attributed to the Kelvin-Helmholtz instability at the magnetopause [Singer et al., 1977]. This study demonstrates that the solar wind Mach number, which depends on both the solar wind density and velocity, may instead be a controlling factor of the wave amplitude. If true, the physical process that controls the wave amplitude take places at the bow shock rather than the magnetopause. The effect of solar wind Mach number on the amplitude of upstream waves and magnetic pulsations in the magnetosphere should be considered in future studies.
Acknowledgements. We wish to thank K. Ogilvie for providing data from WIND Solar Wind Experiment, R. Lepping for data from WIND Magnetic Fields Investigation. The IGPP/LANL Magnetometer Array was established under a grant from the Los Alamos branch of the IGPP. P. Chi, W. Goedecke, G. Le, and C. T. Russell were supported by the National Aeronautics and Space Administration (NASA) under research grant NAG5-7721. S. Petrinec was supported by NASA ISTP programs.