Engineering topics
Publications and source records attributed to Stevenson, D. J..
Galileo, the first artificial satellite of an outer planet, has been orbiting Jupiter since Dec 7, 1995. The spacecraft encounters one of the four Galilean satellites on each orbit.
Various arguments suggest that Venus probably has no asthenosphere, and it is likely that beneath the crust there is a highly depleted and highly viscous mantle layer which was probably formed in the early history of the planet when it was partially or completely molten. Models of crystallization of magma oceans suggest that just after crystallization of a hypothetical magma ocean, the internal structure of Venus consists of a crust up to about 70 km thickness, a depleted layer up to about 500 km, and an enriched lower layer which probably consists of an undepleted 'lower mantle' and heavy enriched accumulates near the core-mantle boundary. Partial or even complete melting of Venus due to large impacts during the formation period eventually results in differentiation. However, the final result of such a differentiation can vary from a completely differentiated mantle to an almost completely preserved homogeneous mantle depending on competition between convection and differentiation: between low viscosity ('liquid') convection and crystal settling at small crystal fractions, or between high viscosity ('solid') convection and percolation at large crystal fractions.
In the traditional view of planetary magnetism, a planet either has a core dynamo (Earth, Jupiter, Saturn, Uranus, Neptune, maybe Mercury) or does not (Mars, Venus, Moon...) I argue that this view is simplistic in two respects. First, mantle convection in terrestrial planets is invariably ata high enough Rayleigh number that it is time variable; this leads to the intermittent arrival of mantle 'cold fingers' at the core-mantle boundary promoting at least local core convection and dynamo action even when the planetary core is stably stratified on average. Thus, I predict an intermittent dynamo regime in addition to the simple dynamo-on (Earth) and dynamo-off regimes. Second, the mantle convection-driven horizontal temperature gradients just below the core-mantle boundary can lead to unstable flows that will convert thermoelectric or electrochemical toroidal fields into externally detectable poloidal fields, even when a dynamo is not possible. It is likely that Mars possesses an interesting core magnetic field of the latter kind, complex but with a dipole that might be approximately aligned with the rotation axis and a surface field of a few to tens of gammas.
A large self-gravitating body does not easily lose significant mass because the escape velocity is much larger than the sound speed of atmosphere-forming species under ambient thermal conditions. The most significant exceptions to this are giant impacts or impact jetting by fast-moving projectiles. A very small object (e.g. a comet) also does not easily lose significant volatile mass upon formation because the energy release associated with its accretion is so small. (It can however lose a great deal of mass if it is subsequently moved closer to the Sun.) I argue that there is an intermediate mass range (corresponding to bodies with radii of approximately 300-800 km) for which the ambient steady-state mass loss is a maximum. By ambient, I mean those conditions pertaining to the formation region of the body. By steady state, I mean to exclude infrequent traumas (giant impacts). The existence of a preferred intermediate mass arises through the competition of growing gravitational containment and growing energy release by accretion; it corresponds typically to GM/(Rc(sub s)(exp 2)) approximately equals 2 to 4, where M is the protoplanet mass of radius R, and c(sub s) is the sound speed. Several factors determine the amount of volatile loss is this vulnerable zone during accretion but in general the loss is a substantial fraction of the volatiles, sometimes approaching 100 percent. The principal implication is that bodies larger than a few hundred kilometers in radius will not have a 'primitive' (i.e. cometary) composition. This is relevant for understanding Triton, Pluto, Charon, and perhaps Chiron.
Laboratory simulations of Martian CO2 storage address whether carbonate formation could have reduced CO2 pressure from a hypothetical greater than 1 bar to the present 7 mbar in less than or equal to 3 to 4 billion years. This problem is addressed with experiments and analysis designed to verify and improve previous kinetic measurements, reaction mechanisms, and product characterizations, with the goal of improving existing models of Martian CO2 history. A sensitive manometer monitored the pressure drop of CO2 due to uptake by powdered silicate for periods of 3 to 100+ days. Pressure drops for diopside 1 and basalt show rapid short-term (approximately one day) CO2 uptake and considerably slower long-term pressure drops. Curves for diopside 2, olivine 1, and olivine 2 are qualitatively similar to those for diopside 1, whereas quartz and plagioclase show near-zero short-term pressure drops and very slow long-term signals, indistinguishable from a leak (less than 10(exp 11) mol/sq m/s).
The evolution of an initially totally molten magma ocean is constrained on the basis of analysis of various physical problems in the magma ocean. First of all an equilibrium thermodynamics of the magma ocean is developed in the melting temperature range. The equilibrium thermodynamical parameters are found as functions only of temperature and pressure and are used in the subsequent models of kinetics and convection. Kinematic processes determine the crystal size and also determine a non-equilibrium thermodynamics of the system. Rheology controls all dynamical regimes of the magma ocean. The thermal convection models for different rheological laws are developed for both the laminar convection and for turbulent convection in the case of equilibrium thermodynamics of the multiphase system. The evolution is estimated on the basis of all the above analysis.
The prevailing view of very large impacts during earth accretion suggests an initial state for earth evolution that was totally molten or nearly so. The problem confronted is to understand the evolution from this state to an almost completely solidified mantle. Two crucial questions are asked by the author: (1) is the resulting endstate of magma ocean freezing compatible with geological record, inferred mantle structure and evidence from geochemistry; and (2) does the freezing event leave a signature that can be discerned in the present earth. The emphasis on this keynote introduction will be to set the stage for the more detailed analyses to follow and to clarify the crucial questions and uncertainties.
A large number of volcanic features exist on Venus, ranging from tens of thousands of small domes to large shields and coronae. It is difficult to reconcile all these with an explanation involving deep mantle plumes, since a number of separate arguments lead to the conclusion that deep mantle plumes reaching the base of the lithosphere must exceed a certain size. In addition, the fraction of basal heating in Venus' mantle may be significantly lower than in Earth's mantle reducing the number of strong plumes from the core-mantle boundary. In three-dimensional convection simulations with mainly internal heating, weak, distributed upwellings are usually observed. We present an alternative mechanism for such volcanism, originally proposed for the Earth and for Venus, involving Rayleigh-Taylor instabilities driven by melt buoyancy, occurring spontaneously in partially or incipiently molten regions.
The (NH3)x(H2O)1-x phase diagram for X from 0 to 0.50 has been reexamined at temperatures from 125 K to 400 K and at pressures from 6.0 GPa using diamond anvil cells, and the implications of the findings for icy satellites are addressed. Titan is likely to have a thicker NH3-H2O ocean than previously suspected, because the stability field of NH3-H2O is found to be smaller than previously supposed. The implications for methane and ammonia volcanism on Titan are briefly discussed. The experimentally observed reactivity between the liquid and iron may also have implications for planetary and satellite evolution.
Thermal history of Mercury's interior is examined using the model of Stevenson et al. (1983), extended to include the effects of tidal heating in Mercury's solid inner core. The implications of Mercury's thermal history for the source of the planet's magnetic field are discussed. It is shown that the major results of this model are similar to the results obtained with the Stevenson et al. model, except for the addition of inner-core tidal dissipation. It is concluded that the extended model properly characterizes Mercury's internal structure and thermal history, and that the criteria for dynamo generation are not properly satisfied. Alternative explanations, including the possibility of a weak thermoelectric dynamo, are examined.
A simple model of the equatorial zonal jet in the giant planets is studied in which the flow is assumed uniform on cylinders concentric with the spin axis, and viscous and magnetic torques on each cylinder are balanced. This 'Taylor constraint' is solved simultaneously with the dynamo equation to obtain the velocity and magnetic field in the equatorial plane. The model is used to reproduce the widely differing jet widths of Jupiter and Saturn eddy viscosity of 2500 sq cm/s. For Saturn, the model has a large magnetic Reynolds number where the Chandrasekhar number Q = 1, and hence exhibits substantial axisymmetrization of the field in the equatorial plane.
Permanent magnetism and conventional dynamo theory are possible but problematic explanations for the magnitude of the Mercurian magnetic field. A new model is proposed in which thermoelectric currents driven by temperature differences at a bumpy core-mantle boundary are responsible for the (unobserved) toroidal field, and the helicity of convective motions in a thin outer core (thickness of about 100 km) induces the observed poloidal field from the toroidal field. The observed field of about 3 x 10 to the -7th T can be reproduced provided the electrical conductivity of Mercury's semiconducting mantle approaches 1000/ohm per m. This model may be testable by future missions to Mercury because it predicts a more complicated field geometry than conventional dynamo theories. However, it is argued that polar wander may cause the core-mantle topography to migrate so that some aspects of the rotational symmetry may be reflected in the observed field.
Thermodynamic models are developed for the processes which controlled the evolution of the surface Ganymede, an icy Jovian satellite assumed to have a rock-rich core surrounded by a water-ice mantle. Account is taken of a heat pulse which would have arisen from a Rayleigh-Taylor instability at a deep-seated liquid-solid water interface, rapid fracturing from global stresses imposed by warm ice diapiric upwelling, impacts by large meteorites, and resurfacing by ice flows (rather than core formation). Comparisons are made with existing models for the evolution of Callisto, and the difficulties in defining a mechanism which produced the groove terrain of Ganymede are discussed.
Theoretical models of lunar origin involving one or more collisions between the earth and other large sun-orbiting bodies are examined in a critical review. Ten basic propositions of the collision hypothesis (CH) are listed; observational data on mass and angular momentum, bulk chemistry, volatile depletion, trace elements, primordial high temperatures, and orbital evolution are summarized; and the basic tenets of alternative models (fission, capture, and coformation) are reviewed. Consideration is given to the thermodynamics of large impacts, rheological and dynamical problems, numerical simulations based on the CH, disk evolution models, and the chemical implications of the CH. It is concluded that the sound arguments and evidence supporting the CH are not (yet) sufficient to rule out other hypotheses.
Mechanisms to explain the mobilization of ice on the Uranian satellites Miranda and Ariel at the very low temperatures prevailing on those bodies are considered. A form of pressure solution creep is proposed in which very fine-grained water ice or clathrate hydrate is mobilized by a small amount of intergranular cryogenic fluid (CH4, CO, or N2). Viscosities as low as 10 to the 12th P are possible for a limited time, sufficient to allow flooding of rift valleys and perhaps even substantial lateral flows (glaciers).
Io is presently considered in light of a simple model for the coupled thermal and orbital evolution of a tidally heated satellite in an orbital resonance, demonstrating quantitatively how a feedback mechanism between the orbital and thermal energy of such a satellite may yield periodic surface heatflow and orbital eccentricity variations. The model predicts that the mean motion of Io may currently be increasing, as suggested by recent estimates of Io's mean motion on the basis of eclipse data. It is further inferred that the tidal stresses in the ice shell of Europa, whose eccentricity mimics that of Io, may recently have been sufficiently great to generate the fracturing observed.
Voyager IRIS brightness temperature measurements of Titan at a wavelength of 530/cm are crudely indicative of ground or lower tropospheric temperatures and indicate 93 K for the equator and 91 K for both northern and southern high latitudes. The symmetry between north and south is unexpected for the time of Voyager encounter (Northern Titan spring). It is shown that this near-symmetry can arise naturally in a model where the poles are 'pinned' year-round at the dew point of CH4-N2 lakes or, more probably, a CH4-N2 rich surface layer on a deep ethane-rich ocean. For a polar temperature of 91 K, the model implies that the atmosphere contains somewhat less than 8 percent mole fraction of CH4.
Solar system origin and planetary formation are discussed with emphasis placed on accretion disk dynamics, disk instabilities, giant gaseous protoplanets, condensation, sedimentation, coagulation, planetesimal swarm evolution, giant planet formation, and implications for satellites. Disk formation and the dynamics of a protosatellite disk are considered as well as satellite accretion, impact disruption and ablation, and satellite capture. Possible explanations for each of the satellite systems are offered. It is concluded that satellite formation involves a variety of processes.