Mass density inferred from toroidal wave frequencies and its comparison to electron density

[1] We have estimated the magnetospheric local mass density using the frequency of the toroidal standing Alfven waves detected with the magnetic ( B) and electric ( E) field experiments on the Combined Release and Radiation Effects Satellite (CRRES), which had a low-latitude elliptical orbit (6.3 R-E apogee and 18 degrees inclination). A 73-day period in 1991, which included several geomagnetic storms, was chosen for analysis because the spacecraft was located on the afternoon sector where toroidal waves are routinely excited. Dynamic spectra of the B and E fields were generated for each CRRES orbit for visual identification of toroidal mode ULF waves excited at a number of harmonics. The fundamental mode is the most persistent feature in the spectra, and this mode appears more clearly in the electric field than in the magnetic field. For statistical analysis we determined the fundamental frequency f(1) from a separate set of electric field spectra that were computed in a moving 20-min time window with 10-min overlap. The frequency is in the range of 1.7 - 14.2 mHz. The local mass density consistent with f(1) was obtained by numerically solving a magnetohydrodynamic wave equation that incorporated the Tsyganenko magnetic field model and an analytical form of field-line mass density variation that was derived from a recent study of toroidal harmonics. We combined the estimated mass density with the local electron number density n(e) derived from the plasma wave spectra at CRRES to obtain the local average ion mass M (= rho/n(e)). A total of 1094 M samples were obtained, with the majority ( 95%) coming from the plasmatrough region (n(e) < 100 cm(-3)). Our technique for the mass density is confirmed to be reliable because only 3.5% of the M samples lie outside the physically plausible range of 1 amu (all-H+ plasma) to 16 amu (all-O+ plasma). The average ion mass has a median value of 3.0 amu and depends on geomagnetic activity. For example, M is 2.5 amu for Dst = 0 and 4.1 amu for Dst = - 60 nT. There is a similar dependence of M on Kp, but only when the Kp values are averaged over a few days. There is no strong dependence of M on the magnetic field L shell in the range L = 4 to L = 8 covered by the f1 samples, although M is significantly low (< 2 amu) for the plasmaspheric regime (n(e) > 100 cm(-3)).