DPS 35th Meeting, 1-6 September 2003
Session 27. Planet and Satellite Origins II: Accretion, Terrestrial Planets
Oral, Chairs: F. Spahn and T. J. Ahrens, Thursday, September 4, 2003, 3:40-5:50pm, DeAnza I-II

[Previous] | [Session 27] | [Next]

[27.07] Planetesimal Collisions and Terrestrial Planet Formation

Z. M. Leinhardt, D. C. Richardson (University of Maryland)

During the past decade over 100 Jupiter-sized extrasolar planets have been discovered. Innovations in observational techniques will soon allow observers to detect Earth-sized planets and increase the extra-solar planet inventory by orders of magnitude. At the same time the growing capabilities of computers make large direct simulations of solar system formation possible. Due to computational limitations previous numerical simulations have simplified planetesimal collisions, the dominant growth mechanism in the protoplanetary disk. Past simulations of terrestrial planet formation have either assumed that colliding planetesimals merge completely, ignoring any fragmentation, or have extrapolated the collision outcome from a model based on laboratory impact experiments. In a real disk a range of collisions is expected, from slow collisions in which most of the mass ends up in one remnant, to fast collisions in which the mass ends up in several small fragments. For planetesimals large enough not to be affected by nebular gas, the most important force involved in collisions is gravity. At these sizes the material strength of the planetesimals is negligible compared to their gravitational binding energy. The first simplification, perfect merging, ignores the range of collision possibilities. The second simplification method,extrapolation of laboratory experiments, ignores the effect of gravity in the collision outcome. In both cases the numerical simulations produce terrestrial planet systems with eccentricities many times that of our own solar system. We have developed a more realistic planetesimal collision model in which gravity is the dominant mechanism in determining the collision outcome. This model will allow us to determine realistic timescales for terrestrial planet formation for various initial conditions and, by including a self-consistent model of fragmentation, it will test our current understanding of terrestrial planet formation. At this meeting we present preliminary results from incorporating our collision model into a direct numerical simulation of planet formation.

The author(s) of this abstract have provided an email address for comments about the abstract: zoe@astro.umd.edu

[Previous] | [Session 27] | [Next]

Bulletin of the American Astronomical Society, 35 #4
© 2003. The American Astronomical Soceity.