454 Encyclopedia of the Solar System
conditions can begin to convect, with warmer, lower-density
ice rising toward the surface, exchanging with cooler higher-
density ice sinking into the interior. This glacially slow solid
ice circulation is similar to what occurs in the Earth’s rock
mantle. The important point is that it is much more efficient
at transporting heat than conduction alone. In simple terms,
as the ice heats up from the radioactive energy from below,
it will begin to convect, taking heat to the surface, but never
allowing the temperature within the ice to rise above the
melting point. Under these conditions, even if an ocean
formed early in the satellite’s history, the convection process
in the ice crust could rapidly freeze it solid.
A final complication is the issue of ammonia, NH 3 .In
many formation models, ammonia is a possible minor con-
stituent of the icy satellites. If present, it has a major effect
on the melting point of a water–ammonia mixture, depress-
ing the temperature at which a liquid can exist to about
173 K, a hundred degrees below the point at which pure wa-
ter melts. Although ammonia has not been detected on the
surfaces of the satellites, even small amounts can affect the
results of theoretical thermal and convection calculations,
and most discussions of the satellites’ interiors include both
ammonia and nonammonia cases.
Whether convective cooling “wins” over heating deter-
mines whether a liquid ocean at the present time can exist.
The calculations for interior models including convection
are quite complex and depend on some properties of ice
that are poorly known. Current models for Ganymede and
Callisto show that liquid layers are possible under some con-
ditions, but these models cannot definitively demonstrate
their existence. Measurements pointing strongly to the
presence of liquid oceans in both satellites came from an un-
expected source—magnetic field measurements made by
the Galileo mission, which is the subject of the next section.
2.3 Magnetic Fields
TheGalileomagnetometer experiment had two major ob-
jectives for studying Ganymede and Callisto: (1) to deter-
mine whether they possess intrinsic magnetic fields of their
own and (2) to study the interactions of the satellites with
Jupiter’s huge and powerful magnetosphere. These two ob-
jectives are closely coupled because the satellites orbit deep
within the region of space controlled by Jupiter’s magnetic
field and its associated trapped radiation and plasma (a ten-
uous ionized gas made up of electrons, protons and posi-
tively charged ions). Measurements of magnetic fields in the
vicinity of the large satellites thus must take into account
the large background field from Jupiter, which is contin-
ually changing due to Jupiter’s fast rotation sweeping the
field past the satellites, and magnetic perturbations from
large-scale electrical currents flowing within the magneto-
sphere. Once these effects are measured and understood,
the experimenters can search for the smaller perturba-
tions in the local magnetic field produced by any intrin-
sic field and from local currents set up by the interactions
of the satellites and their tenuous atmospheres with the
magnetosphere.2.3.1 INTRINSIC FIELDS
On the very firstGalileoclose encounter with Ganymede,
the space physics instruments detected strong evidence
for both an intrinsic field and complex interactions with
Jupiter’s environment. As the spacecraft flew by the satellite,
the magnetometer recorded a marked change in both the
magnitude and direction of the magnetic field. At the same
time, the plasma wave spectrometer (which receives natu-
ral radio “noise” produced by the interactions of charged
particles and magnetic fields) showed sharp changes in the
nature of the radio signals it received, coinciding closely
in time with the observed magnetic deflections. To the in-
vestigators on these experiments, these observations were
familiar, a “fingerprint” indicating the spacecraft had passed
though a planetary magnetosphere. Due to the complexi-
ties discussed earlier, it took observations on subsequent
flybys to confirm the discovery, but it soon became clear
that Ganymede possesses a relatively strong intrinsic field,
oriented in the opposite sense to Jupiter’s field, which pro-
duces a “mini-magnetosphere” embedded within Jupiter’s
magnetosphere.
An intrinsic field at Ganymede was not totally unex-
pected. UCLA space physicist Margaret Kivelson, the head
of the Galileo magnetometer team, suggested prior to the
Galileomission that the big satellites might be able to gener-
ate their own internal fields. Nevertheless, the discovery of
an intrinsic field at Ganymede raises a number of issues for
our understanding of planetary magnetic field generation
[seePlanetaryMagnetospheres].
How planetary fields, including the Earth’s, are gener-
ated and maintained is an active area of research. It is be-
lieved that some form of what is called a “geodynamo” is
responsible for producing a magnetic field within a plan-
etary core. The exact requirements for generating a field
by this dynamo process in a given planet are the subject of
debate. Ganymede’s internal field is consistent with its high
degree of differentiation and favors a three-layer model with
a metallic iron/iron-sulfide core. However, merely having a
metallic iron core is not sufficient to produce a planetary
magnetic field. Although the Earth’s field and other plan-
etary fields are frequently described in textbooks as “bar
magnet” fields, this only describes the field’s mathematical
description (having a dipolar—N and S—configuration with
field lines connecting the poles). The bar magnet analogy
is misleading in terms of the source of the field, since it
has long been known that iron will lose its magnetization at
the temperatures typical of planetary cores (temperatures
above the Curie point, at which a magnetic material loses
its magnetism).
Current theories of planetary dynamos suggest that the
basic requirement for generating a field is continual convec-
tive motion of an electrically conducting fluid. Theoretical