[Previous] | [Session 57] | [Next]
A.J. Dombard, W.B. McKinnon (Washington University)
We report continuing efforts to simulate crater relaxation on Ganymede and Callisto. The finite element method is used, employing an elastoviscoplastic (EVP) rheological model. Adding elastic and plastic (a continuum approximation for discrete, brittle faulting) components to the usually modeled viscous creep provides extra avenues through which topography can be relaxed. Hence, EVP relaxation can, in principle, proceed faster than viscous relaxation. Driving stresses are small enough, however, compared to the elastic and plastic strengths that these extra components do not augment relaxation much, and the process is strongly determined by the creep properties of the near-surface. Indeed, plastic strains associated with floor uplift appear limited to less than ~1%; such low strain faulting may not be visible. Radial faults seen in some dome craters are thus likely not due to relaxation, but to prompt collapse or dome formation. In our simulations, we use the most recent data for fresh crater dimensions and the most recently determined rheological parameters for water ice, including a grain size sensitive, grain boundary sliding creep mechanism. For reasonable choices of input parameters (crater diameter, thermal structure, and grain size), we find relaxation times (the time needed for initial apparent crater depth to decrease by a factor of 1/e) that span the lifetimes of the satellites' surfaces. For near-zero heat flow, relaxation times can exceed the age of the solar system, even for large craters (a finding that contrasts with previous analyses). For high heat flow (approaching 10 mW m-2), relaxation times for large craters can be on the order of years. This large variation in relaxation times will permit the simulation of the full range of observed crater relaxation states, with the goal of piecing together more refined thermal histories for Ganymede and Callisto. Funding provided by a NASA GSRP Fellowship to AJD.
The author(s) of this abstract have provided an email address for comments about the abstract: rew@wurtzite.wustl.edu