EXOPLINES: Molecular Absorption Cross-Section Database for Brown Dwarf and Giant Exoplanet Atmospheres

Citing the Data 

If you use this data be sure to cite the paper that describes the work, as well as this Zenodo DOI. 

Paper Abstract

Stellar, substellar, and planetary atmosphere models are all highly sensitive to the input opacities and subject to errors arising from incomplete line lists and the lack of appropriate pressure-broadening parameters. Generational differences between various state-of-the-art stellar/planetary models are primarily because of incomplete and outdated atomic and molecular line lists. Addressing this, here we present a database of pre-computed molecular absorption cross-sections for all isotopologues of key atmospheric absorbers relevant to late-type stellar, brown dwarf, and planetary atmospheres: MgH, AlH, CaH, TiH, CrH, FeH, SiO, TiO, VO, and H2O. The pressure and temperature ranges of the computed opacities are between 10-6--3000 bar and 75--4000 K, and their spectral ranges are 0.25--330 micron for many cases where possible. For cases with no published pressure-broadening data, we use collision theory to bridge the gap. We also probe the effect of absorption cross-sections calculated from different line lists in the context of Ultra-Hot Jupiter and M-dwarf atmospheres. Using 1-D self-consistent radiative-convective thermochemical equilibrium models, we report significant variations in the theoretical spectra and thermal profiles of substellar atmospheres. With a 2000 K representative Ultra-Hot Jupiter, we report variations of up to 320 and 80 ppm in transmission and thermal emission spectra, respectively, and differences of up to 130 K in the thermal profile between 0.1 mbar and 10 bars. For a 3000 K M-dwarf, we find differences of up to 125% in the spectra and 60 K in the thermal profiles between 1 mbar and 100 bars. We find that the most significant differences arise due to the choice of TiO line lists, primarily below 1 micron, with minor variations because of metal hydride differences in the near-infrared. In sum, we present (1) a database of pre-computed molecular absorption cross-sections to mitigate shortcomings for pre-existing incomplete line lists, and (2) quantify biases that arise when characterizing substellar/exoplanet atmospheres due to line list differences, therefore highlighting the importance of correct and complete opacities for eventual applications to high precision spectroscopy and photometry.

About the Data

All the molecules, except H2O, are computed on a 1460 pressure-temperature (P-T) grid. H2O is computed on a 1060 pressure-temperature grid. Each filename has an associated number that corresponds to a P-T point. You can find the grid information in grid1060.txt, and grid1460.txt. 

Reading Data on a 1460 Grid Python

import pandas as pd
import numpy as np

g1460 = pd.read_csv('grid1460.csv')

#example reading in P-T point T=1400 K , P=1.e-03 bar
file_number = g1460.loc[((g1460['pressure_bar']==1e-3) &
(g1460['temperature_K']==1400)) ,'file_number'].values[0]
numw = g1460.loc[(g1460['file_number']==file_number) ,'number_wave_pts'].values[0]
delwn = g1460.loc[(g1460['file_number']==file_number) ,'delta_wavenumber'].values[0]
start = g1460.loc[(g1460['file_number']==file_number) ,'start_wavenumber'].values[0]

file_to_read = '/data/weighted_cxs/weighted_AlH_1460/p_{0}'.format(file_number)

cx_data = np.fromfile(file_to_read)
wavenumber_grid = np.arange(numw)*delwn + start

Reading Data on a 1060 Grid with Python

import pandas as pd
import numpy as np

g1060 = pd.read_csv('grid1060.csv')

#example reading in P-T point T=1400 K , P=1.e-03 bar
file_number = g1060.loc[((g1060['pressure_bar']==1e-3) &
(g1060['temperature_K']==1400)) ,'file_number'].values[0]

file_to_read = '/data/weighted_cxs/weighted_H2O_1060/H2O_H2HE_POKAZATEL_1060.{0}'.format(file_number)

data = pd.read_csv(file_to_read, delim_whitespace=True, header=None,
names = ['wavenumber','cx'])
cx_data = data['cx']
wavenumber_grid = data['wavenumber']