Chapman and Ferraro (1931, 1932) were first to discuss the existence of a boundary to the Earth's magnetic field. During 1950's, as the concept of continuous solar wind emerged, it was obvious that such a feature should be a permanent feature of the magnetosphere. In the early 1960's, Explorer 10 and 12 provided the first measurements of this boundary (Cahill and Amazeen, 1963) that was to be called magnetopause. It plays an important role in space physics, since the coupling between the solar wind and magnetosphere occurs through it. Outside magnetopause we find the shocked solar wind region, magnetosheath, and just inside are the magnetospheric boundary layers.
Magnetopause is a direct consequence of solar wind interaction with magnetized planets. However, something similar is found also for weakly magnetized planets like Venus, although the physics differ. Then we talk about ionopauses (e.g., Luhmann, 1995).
In the first approximation, magnetopause is formed at a distance where the solar wind dynamic pressure equals the magnetic pressure of Earth's field. At this location, typically around 8 - 11 Re away on the Earth - Sun line, a large scale duskward (Chapman-Ferraro) current develops in the dayside magnetopause to cancel the Earth's field outside. At the same time, the dipole field inside is increased, being now about the two times the nominal dipole value. Similar current flows around the magnetotail, but there the direction has to be reversed in order to cancel the field outside. This current is closed via the cross-tail current. The thickness of the current layer is typically from several hundred to a thousand kilometers, which corresponds to several ion gyroradii (e.g., Berchem and Russell, 1982).
The magnetosphere presented above is of closed type. Even if simple, also it can describe some dynamic events relating to Sun-Earth connection. For example, solar wind pressure pulses push the magnetopause inside causing, e.g., SI/SSC's. However, when the effects of interplanetary magnetic field are taken into account, the magnetic reconnection complicates the physics of the magnetopause considerably by "opening" up the magnetosphere.
For empirical magnetopause models, see Sibeck et al. (1991) and Petrinec and Russell (1993). Elsen and Winglee (1997) have constructed a MHD model for the magnetopause, and compared the results with the above empirical models. See also Shue et al. (1997).