Three-dimensional analysis of impact processes on planets

[摘要] NOTE: Text or symbols not renderable in plain ASCII are indicated by [...]. Abstract is included in .pdf document.This thesis consists of four chapters. The first chapter describes the numerical method known as Smoothed Particle Hydrodynamics (SPH), that is used in present calculations relating to oblique impact on a planet with an atmosphere. Numerical test results are presented to demonstrate the validity of the present computer code. The SPH code is applied to Comet Shoemaker-Levy 9 impact on Jupiter, particularly the disintegration of the comet in the Jovian atmosphere and the evolution of the vapor plume. These calculations are described in chapter II. In order to interpret the phenomena observed in the impacts, numerical predictions are compared with the observational data. These comparisons are summarized in chapter III. Chapter IV considers analytical models of the formation of radar dark and/or bright halos surrounding impact craters discovered on Venus as a result of the Magellan mission. The surface features caused by the atmospheric shock waves associated with impacts are modeled and applied to the observed Venusian radar features.Paper I:The SPH method used for a series of impact calculations is tested. From the performance tests, we infer that the tree structure method for a vectorizing computer code can be applied when the number of particles, N, [...]. Most of the calculations are performed using a CRAY-YMP at Jet Propulsion Laboratory or a CRAY-C90 at Goddard Space Flight Center. Tests for one-dimensional shock propagation show that the developed code yields an error of less than several percent in density and pressure, upon comparison with the analytical impedance-match solution. For spherical, three-dimensional impact of a projectile upon a silicate half-space, the shock wave attenuation of pressure and density calculated by the SPH code is very similar to the results obtained with a finite-difference Lagrangian continuum code (previously obtained by Ahrens and O'Keefe, [1977]). For example, the shock pressures in the near field agree within 15 % differences, whereas the shock pressure in the far field, at 10 projectile radii, differs by a factor of 2 from that calculated previously using a two-dimensional axisymmetric Lagrangian code.Paper II:The impact of fragments of Comet Shoemaker-Levy 9 on Jupiter and the resulting vapor plume expansion are investigated by conducting three-dimensional numerical simulations using the Smoothed Particle Hydrodynamics (SPH) method. An icy body, representing the cometary fragments, with a velocity of 60 km/sec and a diameter of 2 km can penetrate to 350 km below the 1-bar pressure level in the atmosphere. Most of the initial kinetic energy of the fragment is transferred to the atmosphere between 50 km and 300 km below the 1-bar pressure level. The shock-heated atmospheric gas in the wake is totally dissociated and partially ionized. Scaling our SPH results to other sizes indicates that fragments larger than [...] 100 m in diameter can penetrate to below the visible cloud decks. The energy deposited in the atmosphere is explosively released in the upward expansion of the resulting plume. The plume preferentially expands upward rather than horizontally due to the density gradient of the ambient atmosphere. It rises ...] km in [...] seconds. Eventually the total atmospheric mass ejected to above 1 bar is [...] 40 times the initial mass of the impactor. The plume temperature at a radius ~ [...] km is > [...] K for [...] seconds, for a 2-km fragment. We predict that impact-induced plumes will be observable with the remote sensing instruments of the Galileo spacecraft. As the impact site rotates into the view of the earth some 20 minutes after the impact, the plume expansion will be observable using the Hubble Space Telescope and from visible and infrared instruments on ground-based telescopes. The rising plume reaches ~ 3000 km altitude in ~ ten minutes and will be visible from the earth.Paper III:Preliminary observational data from the impact of fragments of Comet Shoemaker-Levy 9 (SL9) are compared with smoothed particle hydrodynamic (SPH) and radiative calculations to determine the energies of individual SL9 fragments and the equivalent diameter of the SL9 progenitor. The Hubble Space Telescope (HST) images of the G impact-induced plume demonstrate that it achieved a height of ~3300 km [Hammel et al., 1994]. This is in close agreement with the SPH calculations for a 7 x [...] erg or 2 km ± 0.3 km diameter solid ice impactor at 60 km/sec. Comparison of the R fragment impact-induced plume brightness in the 8 to 12 [...] band, as recorded by the NASA Infrared Telescope Facility (Mauna Kea, HI) [Orton et al. 1994], with our radiative predictive calculations yields a kinetic energy of 4 x [...] ergs or a diameter of 1.6 ± 0.3 for the R fragment. Using the G and R impacts to calibrate Weaver et al.'s [1994] detailed photometric determination of the relative diameters of pre-Jovian impact projectiles in the SL9 chain from HST images yields revised values for the 11 largest fragment diameters. Adding the inferred masses from the plume and radiative calibration of Weaver et al.'s catalog yields an SL9 progenitor equivalent (ice) diameter of 4.1 ± 0.6 and 3.5 ± 0.5 km, respectively. This compares to a 7.7 km diameter progenitor inferred by Weaver et al. and a 2 km diameter progenitor obtained using tidal break-up modeling by Scotti and Melosh [1993] and Asphaug and Benz [1994]. We also examined the inferred position of Jupiter derived H2O impact ejecta from the hypothetical (3 to 5 bar) cloud deck and find that although Jupiter derived impact induced plumes having radius of > [...] km and achieved altitudes of 3 x [...] km, the contained Jovian water ejecta is restricted to the interior of the plume and would be masked from Earth observations by the surprisingly opaque plume. On the basis of Bjorker et al.'s [1994] observations of thirty minutes' water emissions from the plumes of the impact sites of fragments G and K Field and Ferraro's [1994] analysis which indicates that even if SL9 fragments are 50% porous, upon being subjected to entry, the ram pressures, upon reaching the level of 1 bar, squeeze-out virtually all porosity and the present, self-consistent kinemetric and radiative coupling of the SL9 impactor, we conclude: 1) there is no evidence that SL9 was anything but a comet as first suggested by the coma, 2) penetration of 2 km diameter fragments occurs to depths of ~ 300 km, even for a porous comet, as calculated by the present SPH method and by other groups using finite difference methods [Zahnle and MacLow, 1993, Boslough et al., 1994].Paper IV:A paraboloidal bow shock model was developed in order to estimate the surface distribution of gas shock-induced modifications surrounding venusian impact craters. We applied two-dimensional oblique shock dynamics to describe a three-dimensional paraboloidal - shaped bow shock impinging upon an assumed incompressible venusian surface. The effects of the hypersonic atmospheric shock acting on the venusian surface are considered in terms of maximum gas pressure, the density, the particle velocity, and the temperature, for varying angles and velocities of impact. The maximum boulder size that can be saltated by the shockwave induced gas flow and the degree of mutual collision of the surface materials are also considered. The present calculations quantitatively predict the areal extent of the gas shock perturbed surface for normal and oblique impact as a function of impact angle and velocity, and radii of impactors. For a 1 km-radius stony meteorite impacting at 20 km/s, the radius of the disturbed area extends ~ 10-17 times the 3-5 km crater radius. The perturbed surface affects the surface radar properties and the present results can provide an explanation of the wide "dark/bright halos" surrounding some of venusian impact craters observed via Magellan imagery. For example, a ~ 50 km radius bright halo within a ~ 20 km dark halo is observed around the 3.1 km-radius crater located at 16.5° north latitude and 334.4° longitude. The average value of the radar backscatter cross section of the ~ 20 km radius dark halo indicates that ~ 50 cm-thick layer of porous lithologic material is superimposed upon the undisturbed surface. The occurrence of the bright halo indicates that the surface roughness in this region is ~ 30% greater than that of the surrounding original surface. The present model can relate the observed crater halo radii to the impact parameters, such as impact velocity and angle, and the impactor radius.

[发布机构] University:California Institute of Technology;Department:Geological and Planetary Sciences

[关键词] Planetary Science and Geaphysics