HR: 0830h
AN: P51B-0450    [PDF]
TI: Callisto and Titan: What Lies Beneath?
AU: * McKinnon, W B
EM: mckinnon@wustl.edu
AF: Dept. Earth and Planetary Sci. \& McDonnell Center for Space Sci., Washington Univ., One Brookings Dr., St. Louis, MO 63130 United States
AB: Assuming that Callisto is hydrostatic, the Galileo-derived value of C$_{22}$ yields a normalized moment-of-inertia of 0.3549 $\pm$ 0.0042. While clearly larger than that of Ganymede, this moment-of-inertia value is significantly lower than that of a completely undifferentiated Callisto (0.38). The rock+metal fraction in Callisto must increase with depth, but gravity data alone are unable to constrain the exact nature of this increase (i.e., whether it is continuous or step-wise). A continuous increase is ruled out, because such would suppress internal convection by solid state creep of the ice fraction, and the resulting conductive temperature gradient would intersect the melting curve and promote further differentiation. Callisto, then, must be layered in terms of its ice/rock ratio (except perhaps for small, restricted regions). The simplest layered model for Callisto consists of a denser (more rock and dense-ice-phase rich) interior surrounded by a less rock-rich and more low-density-ice-polymorph-rich shell. A broad range of shell thicknesses are possible. The rationale for two-layer models is that rock(+metal) can separate from ice if the ice melts or if the rock is in massive enough fragments (or concentrations) that they sink slowly through the ice. The downward Stokes velocity of the rock fragments must exceed interior convective velocities for the latter separation to be effective, but not be so great that the rock escapes remixing with deeper ice-rock, if the two-layered structure is to be maintained. Rock released by melting need not sink with respect to the ice, as long as water can escape to higher levels, but the rock must also remix with deeper ice-rock if the two-layered structure is to be maintained. Whether descending rock fragments (or concentrations) remix with deeper ice-rock depends on the fragment or concentration size and ice viscosity, which are unknown. If, however, the rock descends to the center of Callisto, then a rock core should form, surrounded by a mixed ice-rock layer and an exterior ice mantle. Such 3-layer models, in which the mixed layer has the same ice/rock ratio as the bulk satellite, indicate that a Callisto with the moment-of-inertia above corresponds to a body with $\sim$20 percent of its total rock in a central core $\sim$900 km in radius. New structural models indicate that under such conditions the boundary between the clean ice and mixed ice-rock layers is very close to the depth of ice I-ice III transition pressure (209 MPa). This depth is also the natural level for an ocean to perch, so it is tempting to imagine that thermal conditions that led to melting within Callisto operated at the ice minimum melting temperature, and that the rock+metal separated was able to descend to form a core. Here I examine the issue of whether Callisto can differentiate in this manner (via continuous partial melting in ascending convection currents) or whether a thermal runaway and complete differentiation ensues. The latter possibility has been used as an argument that Callisto must have differentiated in the solid state. These arguments will also be applied to Titan.
DE: 5430 Interiors (8147)
DE: 6220 Jupiter
SC: Planetary Sciences [P]
MN: 2003 Fall Meeting