Search

The Earth’s bow shock has been shown to have often a speed rather high (50–150 km s−1) on the basis of different methods of investigation (Formisano et al.,1971; Greenstadt et al., 1972; Formisano et al., 1973; Guha et al.,1972). The causes of the shock motion and speed are not well understood, the resonant oscillator proposed by Smit (1968) being not suitable for the Earth’s magnetosphere. Shen and Dryer (1972) studied the interactions of the Earth’s bow shock with interplanetary shocks as these are causes of bow shock motions; the frequency of interplanetary shocks is rather low (1 per month) so this kind of interaction cannot be invoked to explain the bow shock motion and speed.

The reconnection of magnetic field lines of force has been invoked as a key process for astrophysical and geophysical problems such as solar flares, geomagnetic storms, strong radio sources and star formations. The problem was first investigated by Dungey (1953), then by Sweet (1958) and Parker (1963). The main objective in most of these and other papers was to investigate a mechanism (magnetic field annihilation) capable of accelerating to high energies the large number of particles observed in large solar flares. The mathematical models developed therefore used mainly the conditions of the plasma in the solar atmosphere (see Petschek, 1963; Sonnerup, 1970; Yeh and Axford, 1970; Stevenson, 1972). Until recently, however, no experimental study of the reconnection problem was available.

A multisatellite study of the position of the magnetopause has allowed the first statistical study of the constant F 2/k, describing the kind of interaction that exists between the solar-wind plasma and the Earth’s magnetosphere. Using HEOS-1 plasma data and OGO-5 observations of the magnetopause, we have tried to study the dependence of the magnetopause position D, after normalisation for geometry and dynamical pressure, on the IMF latitude ϑ and on the solar wind β (plasma pressure/magnetic pressure ratio). Differences are found for the average value of D depending on whether ϑ> 0°, ϑ 0.9 or β < 0.9.

This paper summarizes our knowledge of the Earth’s bow shock fine structure. Review papers (Sagdeev, 1967; Paul, 1968, and Friedman et al., 1971) and books (Tidman and Krall, 1971) on collisionless shocks in plasmas, which have appeared so far, deal especially with laboratory produced collisionless shocks and theories to describe the corresponding observations, but only marginally with the Earth’s bow shock. This is probably due to the great complexity of the bow shock structure (see Figure 1), the intermixing of the many different phenomena occurring there and the difficulties of observation. Furthermore, although the available literature on bow shock observations is now very extensive, systematic investigations were not possible until recently since in general the different parameters observed were rather uncorrected. On many occasions for example magnetic field data on the bow shock crossings were reported without a corresponding knowledge of the plasma data and vice versa. In such cases unambiguous checking of the measurements against theories of collisionless shocks in plasmas was clearly practically impossible. Open image in new window Fig. 1. Sketch in the ecliptic plane of the interaction between the solar wind and the Earth’s magnetic field. The magnetopause and the bow shock are indicated together with the region where the particles reflected at the bow shock interact with the incoming plasma generating four kinds of wave phenomena (see Section 3).

Interplanetary-plasma data at 1 AU obtained with experiment S-73 on board of HEOS-A during the last week of February 1969, have been analyzed.