• Cross-cutting
  • Convection
  • Large Scale Dynamics

Under Construction

We are as interested in uninhabitable planets as in habitable ones, maybe more so, since uninhabitably hot planets are easier to observe than cooler Earthlike ones. Very hot planets also give us a window into the composition of rocky planets, as various minerals volatilize into the atmosphere where they can be detected by spectroscopic methods.

  • Subsaturation of condensible species
  • Cloud formation and their optical properties
  • Non-idea equations of state, and atmospheric dynamics near the critical point of phase transitions of condensible substances
  • Generation of spectrally resolved synthetic phase curves from simulations of planetary atmospheres
  • Outgassing of volatiles from magma oceans (e.g. on lava planets such as 55 Cancri e) and their effect on atmospheric dynamics and thermal structure
  • Atmospheric chemistry required to model cloud formation. In constrast to formation of water clouds by simple condensation as on Earth, exoplanet clouds often form from a condensate that is produced by chemical reactions, making incorporation of atmospheric chemistry essential. This is the case for most mineral-vapor clouds, such as appear on very hot planets.

The term convection refers to motions in a fluid driven by positive buoyancy. In planetary atmospheres, convection can be driven by heating of a planetary surface, outflow of heat from planetary interiors, stellar heating of internal absorbing layers, or infrared radiative cooling of upper layers of an atmosphere. When condensible substances are present, they profoundly alter convection by introducing an asymmetry between upward plumes -- which condense and release latent heat -- and downward plumes which do not. Convection is a central driver of both precipitation of condensible species and formation of clouds. It has been extensively studied for the case of water vapor condensation on Earth, but the subject needs to be reinvented to deal with the much greater range of condensible configurations expected in exoplanet atmospheres. That is one of the goals of EXOCONDENSE. Convection is practically ubiquitous in planetary atmospheres, so it is essential that we learn enough about it to represent it in global circulation models. A better understanding of convection can even help in formulating better 1D radiative-convective models.

We seek to understand the fundamental nature of generalized "moist" convection employing theoretical concepts, idealized models, and full three dimensional nonhydrostatic eddy-resolving models of convection. Convection involves scales down to tens of meters or less, and so cannot be resolved in global circulation models. In order to make progress on global simulation of convecting exoplanet atmospheres, it is necessary to learn how to statistically represent ("parameterize") convection within global models.

The following topics and projects are of particular interest:

  • Compositional Buoyancy:On Earth water vapor contributes slightly to buoyancy, because the molecular weight of water is less than that of dry air. In contrast, water vapor is strongly negatively buoyant in a hydrogen atmosphere, leading to great accumulation of potential energy before convection is triggered, leading to violent release of energy. This situation is believed to happen on Saturn, but stable compositional buoyancy is expected to be a common situation in general planetary atmospheres. It would apply, for example, to an N2 atmosphere over a CO2 ocean. Understanding the behavior of atmospheres with strong compositional buoyancy is critical to the prospects for a number of future modeling endeavours.
  • Nondilute Convection: Water vapor in Earth's atmosphere is dilute in the sense that it makes up at most a small portion of the mass of the atmosphere. However, it is a common situation in planetary atmospheres that the condensible species makes up a substantial portion of the atmosphere. This happens, for example, at the inner edge of the habitable zone as planets are approaching a runaway greenhouse, but there are many more applications of nondilute behavior. Nondilute moist convection behaves very differently from the dilute case, and poses a number of challenges. For example, the mass loss due to precipitation can cause a substantial reduction in surface pressure and cause strong dynamical feedbacks which would be small in the dilute limit. Together with recent postgraduate student Feng Ding, we have made a start on exploring nondilute convection, but we have only scratched the surface of this fascinating and important regime.
  • Convective Clustering: Convective clustering refers to a phenomenon in which upward motions involving condensation are limited to a small portion of the nominally convecting domain, leaving most of the domain subsiding and dry. It has been extensively studied for water vapor in Earthlike climates, but is likely to be a general phenomenon in atmospheres with condensible substances. When convective clustering occurs, it has dramatic consequences for the planetary radiation budget, both via condensible subsaturation and the cloud distribution. One of our goals is to develop an understanding of convective clustering in general planetary atmospheres.
  • 3D resolved convection modeling: To help test hypotheses about convection, and generate ideas about general behavior, we plan to generalize the convection model CM1 to a broad range of planetary atmospheres containing condensible substances. The model will serve as an important tool for all of our planetary convection studies.
  • Generalized moist convection parameterizations: We will attempt to generalize the Emanuel convection parameterization scheme to a broad range of planetary atmospheres, including atmospheres with nondilute condensibles. It is not clear that the scheme can properly handle systems that are far out of equilibrium (i.e. have built up a great deal of convective available potential energy), so it may be necessary to explore other schemes, or formulate entirely new ones.

Loosely speaking "large scale dynamics" refers to motions whose characteristic length scale is an appreciable fraction of the scale of the planet itself -- say, from 1% of the radius of the planet up to the scale of the whole globe. These motions need not be quasigeostrophic (i.e. dominated by Coriolis effects), and indeed many exoplanets of interest are slow-rotators. Typically, though large scale dynamical phenomena can be adequately represented by hydrostatic models (the "primitive equations"). Typical phenomena that come under the heading of "large scale dynamics" include Rossby and Kelvin waves (both forced and free), baroclinic and barotropic instability, the Hadley circulation and the Walker circulation. All of these phenomena have counterparts in exoplanet atmospheres, and indeed the study of tide-locked exoplanets draws much inspiration from theory of Earth's tropical circulation.

Interaction with effects of condensible substances adds a great deal of richness -- and a great deal of challenge -- to the study of large scale dynamics. The dynamical effects are critical to interpretation of the new generation of observations coming online for terrestrial-sized planets, notably the use of phase curves to infer dayside-nightside temperature contrast and phase shift of thermal features relative to the substellar point. The dynamical effects on clouds and subsaturation profoundly influence the interpretation of phase curves. They also feed back on planetary climate through their effects on the planetary radiation budget, for example altering the inner edge of the habitable zone through altering the conditions under which a runaway greenhouse occurs.

Aside from theoretical analyses, the principal tool we are developing for studying generalized large scale moist phenomenon is the Exo-FMS general circulation model. This model has already been used in a wide range of planetary climate studies. It has been extended to handle nondilute dynamics (see definition in the Convection theme) and general planetary instellation configurations. Next steps include incorporation of more general thermodynamics, general real-gas radiation and cloud processes.

Here are a few of the large-scale phenomena which EXOCONDENSE will investigate:

  • Collective behavior arising from parameterizations of generalized moist convection embedded in global circulation models (e.g. convectively coupled waves, or Rossby waves coupled to generalized moist convection)
  • Behavior of Hadley and Walker circulations, and equatorial waves in the nondilute limit
  • Dynamics of atmospheres on planets undergoing a runaway greenhouse
  • Generalized moist baroclinic instability; effect of compositional buoyancy (see Convection theme for definition).
  • Cloud and subsaturation patterns on tide-locked vs. Earthlike rotators, and influence on thermal and reflected-light phase curves
  • Dynamics of lava planet atmospheres incorporating mineral vapor condensation.