In the second flyby of the Earth, Galileo approached from the
southern dusk hemisphere of the tail lobe. The trajectory is
plotted in Figure 5 in terms of 
= ( y
+
z
)
versus x in an aberrated GSE coordinate system.
The spacecraft was well outside the nominal position of the distant
bow shock as inferred from the fit of Greenstadt et al.
[1990], but it repeatedly crossed the shock surface, moving back
and forth between the solar wind and the magnetosheath. The
multiple magnetosheath encounters are evident as the shaded
portions of the magnetometer data acquired near 300 R
downtail
that is plotted in Figure 6. The total field shows abrupt changes
between states that differ in amplitude by factors less than 1.5,
indicating an extremely weak shock. Prior to each entry into the
magnetosheath and following each exit, the field in the solar wind
rotated. As described above for the distant Venus bow shock [
Khurana and Kivelson, 1994], for steady solar wind conditions,
these rotations of the IMF are causally linked to the motion of the
bow shock past Galileo, with the bow shock cross section standing
farthest from the tail axis when the IMF is perpendicular to the
local shock normal. Indeed, over the period of several hours shown
in Figure 6, during which the solar wind plasma conditions appear
to have remained fairly steady, the repeated crossings into the
magnetosheath coincided with field rotations into a
quasi-perpendicular orientation. The inbound crossings reveal quite
monotonic and abrupt variations, consistent with the properties of
weak quasi-perpendicular shocks. Additional bow shock crossings
have been identified between x 
-360 R
, 
190 R
and x 
-120 R
, 
70
R
. The positions of these crossings are shown in Figure 5 as
crosses (for inbound) and circles (for outbound). The solar wind
plasma conditions seem to have remained roughly constant between
the first (quasiperpendicular) shock crossings on December 5, 1992
and the crossing near -170 R
early on December 7, 1992, with
the crossings between 
-360 and -300 R
corresponding to
encounters with the quasiperpendicular bow shock and crossings
between 
-230 and -170 R
corresponding to encounters
with the quasiparallel bow shock. The remaining crossings occurred
later on December 7, 1992 during anomalous solar wind plasma
conditions which appear to have compressed the entire magnetosphere
and driven the bow shock to positions unusually close to the tail
axis.
These encounters add extensively to the data on the distant
terrestrial bow shock. The previous data on the distant bow shock
surface are those of Greenstadt et al. [1990]. They
concluded on the basis of 7 segments of ISEE3 orbits between -20
and 
-100 R
that Fairfield's [1971] hyperbolic
model of the shock surface (with the terms introducing dawn-dusk
asymmetry dropped) provides a satisfactory representation of the
shock locus at antisolar distances out to 100 R
. Fairfield's
work was based on analysis of a much larger number of shock
crossings but they were mostly sunward of 20 R
. The additional
crossings in the Galileo data set constrain the asymptotic opening
angle of the shock surface to distances of order -350 R
and
reveal its dependence on the IMF orientation. Figure 5 shows the
hyperbolic fits of Greenstadt et al. [1990] and the proposed
fits to the quasiperpendicular shocks from the Galileo flyby. The
parameters [A. Prevost et al., personal communication, 1994] are
given in the figure caption. The new shock models, based on the
insights provided by the Venus flyby, are consistent with the
properties of MHD (magnetohydrodynamic) perturbations. They
account for the frequent observation of multiple shock encounters
along a spacecraft trajectory even in a steady solar wind.