2. Numerical Model |
|
- Framework of the Whole System and Assumptions
- The model domain consists of an atmosphere and a ground soil
layer. The effect of planetary rotation is not included. The
atmosphere is set to obey the ideal gas law.
CO2 is only assumed to be the atmospheric constituent
and its condensation and sublimation are not considered. The
values of soil density and soil thermal properties are
horizontally uniform. There is no surface topography.
- Atmospheric Model
- The wind and temperature fields of the atmospheric model are
described by a 2D version of an anelastic system developed by
Ogura and Phillips (1962).
According to results from vertical 1D models (e.g.,
Flasar and Goody, 1976;
Pollack et al., 1979),
the thickness of the convection layer in the Martian atmosphere
with dust-free conditions should be equivalent to the scale height
of the Martian atmosphere calculated with a radiative equilibrium
temperature (Zurek et al., 1992).
The anelastic system enables convection to be properly described,
whose depth is nearly equal to the scale height of the atmosphere,
since the anelastic system includes the effects of density
stratification of basic fields.
- Turbulence Model
- Subgrid turbulent mixing (Turbulence
parameterization) is evaluated via the formula by
Klemp and Wilhelmson (1978).
Surface momentum and heat fluxes
(Surface flux
parameterization) are given by the bulk formula of
Louis (1979).
In the present model, the turbulent
mixing coefficient and the bulk coefficient for heat transport
have the same values of those for momentum. The roughness length
for the bulk coefficients is set to 1 cm
(Sutton et al, 1978).
Previous turbulent models have been developed to simulate
turbulence in the terrestrial atmosphere. In the present study, we
have assumed that these turbulent models are also applicable to
the turbulence of the Martian atmosphere.
- Dust Transport
- The spatial distribution of dust is calculated by an advection
diffusion equation that includes the gravitational settling of
dust. The representation of dust terminal velocity follows
Conrath (1975).
We have assumed that the radius of a dust particle is
constant (0.4 μm) in calculating dust terminal velocity.
The dust flux from the surface is that of a
wind tunnel experiment by
White et al. (1997).
- Radiation
-
Radiation of CO2
is calculated by the Goody narrow band model. A
15 μm band in the infrared wavelength region and 4.3, 2.7, 2.0
μm bands in the near infrared wavelength region are
included. Values of absorption line intensity and the width of
each band are adopted from Houghton (1986).
Radiation of dust is
calculated by the δ-Eddington approximation model. We have
included two bands (5-11.6, 20-200 μm) in the infrared wavelength
region and one band (0.1-5 μm) in the solar wavelength
region. The locations of these bands and the values of extinction
efficiency, single scattering albedo, and the asymmetry factor of
each band are adapted from Forget et
al. (1999).
- Ground and surface layer
- Ground temperature (i.e., subsurface temperature) is calculated by
a 1D thermal conduction equation. The values of soil density, thermal
conductivity and specific heat are adopted from the standard model of
Kieffer et al. (1977).
|
A Numerical Simulation of Thermal Convection in the Martian Lower Atmosphere with a Two-Dimensional Anelastic Model
Odaka, Nakajima, Ishiwatari, Hayashi,
Nagare Multimedia 2001
|
|
|