Map: Research / Habitability / Rocky planets	< Back		Next >

2.3 The structure and dynamics of rocky planets (Tackley, Aurnou)

	Figure 2.3. Plausible domain diagram of present-day planetary habitability (i.e., after billions of years of evolution) as a function of planet size and solar flux. Domain boundaries will be time-dependent.

Cycling of volatiles (particularly water and carbonate) between the interior and fluid envelope of a terrestrial planet could play a major role in the evolution of the fluid envelope, and hence the long-term habitability of the surface environment. This cycling is strongly influenced by tectonic mode (i.e., plate tectonics, rigid lid or episodic plate tectonics). For medium-sized planets, a magnetic field, when present, greatly reduces solar wind-induced escape of the atmosphere, aiding long-term habitability.

Previous habitability analyses have included only highly idealized, parameterized models of planet interiors (bringing into question their veracity) and have not included the generation and consequences of a planetary magnetic field. Tackley and Aurnou will study the interaction of the interiors of terrestrial planets (mantles and cores) with their fluid envelopes using advanced numerical models, with the goal of formulating predictions for the evolution of habitability of terrestrial planets as a function of size, stellar flux and initial conditions. The assumed criterion for planetary ‘habitability’ will be the usual: the existence of liquid water at the surface (e.g., Kasting et al. 1993b) which requires a surface temperature of between 273K and 373K (at 1 bar or 10⁵ Pa).

Both the mantle and the core of a terrestrial planet play important roles in determining the evolution and composition of its atmosphere. The fluid envelope of a planet in turn influences the dynamics of its interior, resulting in a system of feedbacks that requires careful analysis. Such an analysis requires consideration of coupled mantle-core-atmosphere systems to determine the likelihood of surface liquid water (hence habitability). Habitability is clearly time-dependent, with planets such as Mars possibly passing through a short-lived (e.g., 100s of Myr) habitable phase before reaching a long-term, non-habitable (too-cold) condition, while planets such as Earth experiencing a long-term, slowly-evolving habitable condition. It is notable that the known planets without plate-tectonics-related interior-atmosphere feedback (i.e., Mars and Venus) appear to have undergone a transient one-way evolution to a non-habitable mode.

The mantle of a rocky planet influences the planet’s fluid envelope by contributing volatile components (e.g., water, carbon dioxide gas) to the surface by way of volcanic activity, and by as a sink for volatiles from the surface where tectonic processes are capable of delivering surface material to the mantle. Indeed, a long term cycle that moves carbonate to and from the mantle (volcanic outgassing, tectonics) may provide a critical feedback mechanism that maintains surface temperature in a habitable range (Kasting et al. 1993b; Sleep and Zahnle 2001). Volatile recycling also affects the redox (reduction/oxidation) state of the mantle, which in turn, through outgassing, influences the redox state of the atmosphere (Delano 2001; Kasting et al. 1993a; Lecuyer and Ricard 1999), and may have been responsible for the rapid rise in atmospheric oxygen ~2 Gyr ago (Kump et al. 2001).

Plate tectonics appears to be a crucial component of such planet-scale cycles and feedbacks. It is by way of plate tectonics that carbonate, water and other volatiles are returned to the mantle by the process of subduction. Such an efficacious return mechanism is not possible in a planet with a rigid, unyielding outer lid (lithosphere). Apparently, the existence of plate tectonics may be important for long-term planetary habitability. The conditions necessary for plate tectonics to exist as a feature of other rocky planets of variable sizes and distances from their stars, and the scaling of key rates (e.g., outgassing, subduction) associated with plate tectonics on other planets, are not understood. In addition, the details of how recycled water and other volatiles circulate in the mantle to be returned to the shallow melting zones beneath volcanic centers are poorly understood even for Earth. The uncertainties are exacerbated by the fact that the mantle of a planet may be partly stratified in convection or in composition, and this stratification will have evolved with time (Tackely 2000a). Tackley and colleagues will study these aspects using numerical models of mantle convection and lithosphere dynamics, as discussed below.

Based on the known terrestrial planets of our Solar System, the core of a rocky planet is responsible for the generation of the magnetic field that shields the planet’s atmosphere from the stellar wind emanating from its star, thereby greatly reducing the rate of atmospheric escape (Shizgal and Arkos 1996; Yung and DeMore 1999). Existence of a magnetic field is particularly important for retaining atmospheres around smaller terrestrial planets such as Mars (larger ones such as Venus are more able to hold on to their atmospheres by gravity alone) and is also important for protecting primitive organisms from potentially lethal charged particles (Horneck et al. 1994). Existence of a magnetic field may be important for maintaining a habitable atmosphere over billions of years for smaller planets, but not all terrestrial planets have an internal dynamo of long duration.

Defining those characteristics of a rocky planet that lead to its ability to generate a magnetic field (by virtue of a dynamo in the core) is of the utmost importance for determining the likelihood that the planet could sustain life over the long term. A first-order criterion for the existence of a dynamo is that the heat flux extracted from the core by the mantle must be greater than that conducted down the core adiabat. However, it may be more complicated than this criterion implies. For instance, studies by (Kutzner and Christensen 2002) and (Olson and Christensen 2002) suggest that the existence and behavior of the magnetic field depends strongly on the strength and pattern of core heat loss as well as the planetary rotation rate. Assuming that the heat flux out of the core is sufficient to drive a dynamo (Gubbins 2001; Nimmo and Stevenson 2992), these studies find that generating an Earth-like dipole dominated magnetic field is sensitive to the ratio of buoyancy forces and rotational forces. It is not understood how this ratio evolves over the thermal history of a planet. In addition, the solid inner core tends to germinate and grow outward at the expense of the liquid outer core during the lifetime of a terrestrial planet (Labrosse et al. 1997). It is not well understood how the magnetic field generation process changes as the core solidifies (Al-Shamali et al. 2002). Tacklye and Aurnou intend to investigate these aspects using three-dimensional numerical geodynamo models and parameterized models of coupled core-mantle evolution, as discussed below.

Tackley, Aurnou and lead-team colleagues intend to investigate the coupled system comprising the fluid envelopes, mantles and cores of terrestrial planets by first performing fundamental research on key aspects of the problem (e.g., magnetic field generation, generation and scaling of plate tectonics convection) using multidimensional numerical models and planetary evolution calculations using either numerical modeling of the mantle with parameterized core and atmosphere models, or fully parameterized models. This improves on previous research which has used only simple parameterizations, about which there is considerable uncertainty, for planetary interior behavior. The new models will be applied to known terrestrial planets prior to extrapolating to all possible terrestrial planets. Terrestrial planets from Mars size to ~10 Earth masses will be considered, with the core making up different proportions of the total mass.

A key output of this will be to establish a predicted domain diagram of exoplanet habitability as a function of size, incident stellar flux (related to steller brightness and distance from the star) and time. The sensitivity of this to initial conditions (e.g., initial concentration of carbon dioxide in the atmosphere) will also be investigated. A preliminary example of a hypothesized domain diagram at a time corresponding to the present-day solar system is shown in Figure 2.3.

Numerical modeling (in two- and three-dimensions) of mantle convection in a plate tectonic regime will be performed. The models will include chemical transport and differentiation associated with crust production. This numerical modeling is necessary because there is considerable doubt as to how convective quantities scale with convective vigor in a plate tectonic regime due to the dissipation associated with subduction (Conrad and Hagar 1999), to the changing compositional buoyancy of oceanic plates (Davies 1992), and to changes in internal dynamics associated with phase transitions (Christensen and Yuen 1985; Davies 1995; Tackley 1995). The basic modeling technology already exists (Tackley 2000b; Tackley and Xie 2002) but will be expanded to include tracking of the subducted volatiles carbonate and water. Tracing the dispositions of these key volatiles will permit Tackley to monitor the evolution of the redox state and water-dependent viscosity of the mantle. A parameterized atmosphere-ocean 1-D radiative model will be overlain on these calculations in order to account for feedbacks between the two systems. A parameterized model of core heat loss (e.g., Labrosse et al. 1997) will also be included to evolve the lower boundary condition and determine core heat loss. Key issues that will be addressed in this way are:

• What are the parameter ranges in which plate tectonics is expected, when fluid envelope, crust, mantle are taken into account? What type of tectonic regime is likely to have existed early in Earth’s history? What changes in tectonic regime may take place over a planets’ history?

• How do convective vigor, outgassing rate, and recycling rate change with time as a planet evolves?

• How does core heat flux change as with time, and thus, when would a dynamo be possible?

• How does the geochemical structure of the mantle, including redox state and potential layering, evolve with time?

• How do the feedbacks between fluid envelope and interior, affect interior dynamics?

Fundamental research into core dynamics of prospective terrestial planes will be performed via 3-D numerical dynamo simulations (Wicht 2002) to understand the effect of the main parameters on the type and strength of dynamo generated. The main parameters are the size of the core, the size of the solid inner core, cooling rate, and rotation rate. These experiments will give us a systematic understanding of how a planetary dynamo is likely to vary in strength and form (e.g, the relative importance of quadropole and higher components) for different sized planets at different stages of their evolution. Using these results, the rate of escape to space of different atmospheric species can be calculated, and this will form an important component of the atmosphere evolution model.

The abundances of the main atmospheric and oceanic constituents will be tracked over a planet’s evolution, with inputs coming from mantle outgassing, and outputs to rock formation and subduction, and escape to space. A parameterized, 1-D atmospheric model will be used (as in e.g., Kasting et al. 1993b) to determine surface temperature and any other necessary quantities. Tackley and Aurnou do not plan to develop more sophisticated atmospheric models because this is being done by other NAI groups (e.g., Meadows et al. at JPL-2) and thus Tackley and Aurnou will collaborate with them if more sophistication is required. The initial condition (after major impacts, etc.) may have an important influence on subsequent planetary evolution, and this will be a major focus. Clearly, if life starts it will also exert a major influence, so the parameterized models will also include this possibility and compare the signature of planets with and without life.

< Previous Section	Next Section >