Running ExoPlaSim’s aerosol module
ExoPlaSim’s aerosol module allows the user to prescribe a source of either dust or photochemical haze in the planet’s atmosphere. The aerosol particles are transported by the horizontal and vertical winds and settle under gravity. If optical data is available, there is an option to include the aerosols in the radiative transfer calculation (absorption and scattering), incorporating their feedback on the climate into the simulation.
The aerosol module can be controlled through the ExoPlaSim
Python API in the configuration step. The arguments aerosol
and aerorad turn on the aerosol transport and radiative
transfer codes, respectively. For the aerosol radiative transfer,
the model requires optical data to be read in from a .dat file.
A Python script is available which inputs an ExoPlaSim-appropriate
stellar spectrum and a complex refractive index of the aerosol
and outputs a formatted .dat file containing the optical constants.
Configuration variables controlling the aerosol module
The free parameters controlling the aerosol run are:
- aerosol
Trueturns on the aerosol transport scheme only. Default isFalse.- rhop
A float specifying the density of the particles in kg/m3.
- apart
A float specifying the radius of the particles in meters.
- asource
The aerosol source. An integer that can take a value of
1or2.
1places the source at the top model level and sets it proportional to either the cosine of the solar zenith angle (if radiative transfer is off) or the downward shortwave flux (if radiative transfer is on). This set-up represents a photochemical haze source.
2places the source at the bottom level and switches it on wherever a gridbox contains land. This set-up represents a dust source.- fcoeff
A float that represents the prescribed mass mixing ratio (kg/kg) of the aerosol source.
If
asourceis set to 1, this coefficient is multiplied by the cosine of the solar zenith angle or the downward shortwave flux.If
asourceis set to 2, this coefficient is multiplied by the binary land-sea mask.- aerobulk
An integer controlling the background gas. Options are:
1: nitrogen gas
2: hydrogen gas
3: carbon dioxide (untested)- aerorad
Integer. If set to
1, the aerosols are included in the radiative transfer calculation. Default is0(aerosol transport without radiative transfer).- aerofile
String. Path to a 8x1 (row x col) .dat file containing the aerosol optical constants extinction efficiency, scattering efficiency, backscattering efficiency, and asymmetry factor for the two shortwave bands.
Example configuration set-up
Below is an example of how to use the configuration variables to set up a simulation with aerosols:
>>> myModel.configure(aerosol=True, rhop=1000, apart=1000e-09, asource=1,
fcoeff=10e-13, aerobulk=1, aerorad=1,
aerofile=‘/home/user/exoplasimproject/hazeconstants/titanconstants.dat’)
This creates particles of density 1000 kg/m3 (equivalent to soot) and radius 1000 nm (1 um), with a photochemical source at the top of the atmosphere and an initial source similar to the haze production rate per second in Titan’s upper atmosphere. The background gas is nitrogen and the aerosol radiative effects are switched on. The optical constants are from laboratory measurements of Titan’s photochemical organic haze.
Incorporating these aerosol settings into the simulation of TOI-700d set up in the tutorial would look like this:
>>> toi700d.configure(startemp=3480.0, flux=1167.0, # Stellar parameters
>>> eccentricity=0.,obliquity=0.,fixedorbit=True, # Orbital parameters
>>> synchronous=True,rotationperiod=37.426, # Rotation
>>> radius=1.19,gravity=11.9,aquaplanet=True, # Bulk properties
>>> pN2=1.47*(1-360e-6),pCO2=1.47*360e-6,ozone=False, # Atmosphere
>>> timestep=15.0,snapshots=720,physicsfilter="gp|exp|sp", # Model dynamics
>>> aerosol=True,rhop=1000,apart=1000e-09,asource=1, # Aerosol properties
>>> fcoeff=10e-13,aerobulk=1,aerorad=1,
>>> aerofile=‘/home/user/exoplasimproject/hazeconstants/titanconstants.dat’)
>>> toi700d.exportcfg()
Running the aerosol module typically requires reducing the timestep for numerical stability (here to 15 minutes).
Module output
The output of the aerosol module is a 4-D data cube called ‘mmr’ (for mass mixing ratio) and is in units of kg/kg. Here’s a visualisation of an example output, from a simulation of TRAPPIST-1 e:
The plot shows higher concentrations of haze particles at the high latitudes near the eastern terminator (lon 270), where the particles collect in standing Rossby gyres. The equator is comparatively free of haze.
Generating new optical constants
The hazeconstants folder in the exoplasim package directory already contains haze constants from (He et al. 2023) and `(Corrales et al. 2023) https://iopscience.iop.org/article/10.3847/2041-8213/acaf86>`_ for a 300K water-rich atmosphere calculated for TRAPPIST-1e and Wolf 1061c for 10 different particle sizes. The existing .dat files assume an N2 atmosphere (refractive index 1.00029, aka 1). The haze constants were calculated using the Python package MiePython assuming spherical haze particles.
The user can create a new
aerofile based on their own optical data for dust, other kinds of haze, etc.
The aerofile must be a .dat format file with 1 column and 8 rows. The file
structure is as follows:
SW band 1 extinction efficiency |
SW band 1 scattering efficiency |
SW band 1 backscatter efficiency/4PI |
SW band 1 asymmetry factor |
SW band 2 extinction efficiency |
SW band 2 scattering efficiency/4PI |
SW band 2 backscatter efficiency |
SW band 2 asymmetry factor |
An example script for generating new constants is available (contact Maureen Cohen to request). The script requires as input:
A stellar spectrum generated by ExoPlaSim’s makestellarspec.py script
The complex refractive index of the aerosols binned to 10 nm
The aerosol particle radius
The refractive index of the background gas
Alternately, users can generate optical constants by their own means, as long as the input file follows the structure shown in the table above.
Caveats
If MiePython is used to generate efficiencies, the backscattering efficiency must be divided by 4PI because of a difference in the way backscattering cross-sections are defined in optics versus atmospheric science. Please refer to Cohen et al. 2024 for details.
The code only simulates one particle size, not a particle size distribution. This could be conceived as the mean or peak size of a distribution which is not significantly changing in time, and the radiative effects as the mean effect of an aerosol distribution on the atmosphere.
The amount of aerosol (source) is arbitrarily specified in the configuration step and should be taken from observational/laboratory/1-D aerosol code literature to get a realistic source.
Aerosol scattering and absorption is only taken into account for the cloud-free portion of the atmosphere. In cloudy areas, only cloud scattering and absorption is considered. This means the radiative effect of the aerosols on the climate will be understimated if the aerosols cause significantly higher extinction than clouds.