ExoCross keywords


The position of the keywords is not important. The input is not key-sensitive.

  • GNS: Nuclear statistical weight gns 2.0
  • Temperature: Temperature (K):
Temperature  3000.0

Aliases: Temp, Temperature

  • Temperature can be split into rotational temperature and vibrational temperature, which in input is given as
Temperature  (Rotational) 700 Vib (vibrational) 2000 (K)

In this case it is also important to provide the addresses of the vibrational columns using the QN section.

  • A reference temperature (used e.g. with SPECTRA) can be specifed using the REF keyword as part of the Temperature line, e.g:
Temperature 1000 Ref 296
  • QN: The designation of the quantum numbers (QN) to columns can be done using the QN section. In this section vib is used for the range of the vibrational QNs; K is for the column with the rotational “K”; Sym is for symmetry; Nsym is number of the symmetries; Nmodes stands for the number of vibrational modes. The column numbering starts after the J column (4th). The vibrational addressing is required for when vibrational temperature is used.

density can be used to give the non-LTE vibrational populations. Lifetime is to define the column with lifetimes, needed for the pre-dissociation line broadening calculations.

  Vib 2 7
  Nmodes 9
  K  7
  Vib 2 7
  density 9
  • PF: input value of the partition function.
PF  1000.0
  • A reference partition funcion (used e.g. with SPECTRA) can be specifed using the REF keyword as part of the PFe line, e.g:
PF 4000 Ref 100
  • Range: Wavenumber (cm-1) or wavelength (um or micron) range, i.e. from nu1 to nu2:
Range 0.0 2000.000

Here is example for microns (wavelength), which however does nor work will all profiles:

Range 100.0 200.000 um

Aliases: Window, Frequencies, Wavenumbers, Range

  • um or micron: units of wavelength used in combination with Range
  • cm-1: units of wavenumber used in combination with Range, can be omitted as default.
  • abundance: is the molecular abundance, which is the factor used in intensities or cross-sections.
abundance 0.97
  • Cut-off: to use the HITRAN-cut-off scheme (see HITRAN 2012 paper). nu_crit is to give the corresponding parameter in the HITRAN cut-off scheme.
cutoff HITRAN (S_crit) 1e-29  nu_crit 2000
  • Cutoff exp: is used together with trans for partitioning into strong and weak parts for a reference temperature using the following dynamic threshhold: \(I_{\rm cut} = I_0 \exp(-\nu/\alpha)\).


cutoff Exp 1e-25  alpha  2000
  • Npoints: Number of grid points (usually an odd number)
Npoints 100001

> Aliases: Npoints, Number-of-points

  • Absorption: type of spectra
  • Emission: type of spectra`
  • lifetime: type of spectra, to compute life time of Ej as 1/sum_i A_ji; saved into a new .life file.

Only one of these three types is needed, it can appear outside any sections anywhere in the program. ‘Absorption’ is default.

  • Partfunc or Partition-function: To compute the partition function and its contributions for a set of temperature; it also starts a section defining the corresponding parameters
  • Ntemps: Number of temperature steps
Ntemps 500
  • TempMax, maxtemps: Maximal temperature
TempMax 5000.0
  • Moment: Moment to compute (0=none,1=Q1,2=Q2,3=CP)
moment 1
  • CP: the same as Moment 3
  • output: File name for crosssections, stick-spectra etc.
output CH4_stick_T300K
  • Cooling is a section name to compute the cooling function on a grid of temperatures. The parameters used in the Cooliing section are TempMax (maxtemps) and Ntemps, see Partfunc.
  • States, StatesFile, StateFile, States_file, States-File: The name of the .states file States CH4_linelist.states
  • Transitions, TransitionFiles: The name of the .trans file or the section name for the list of the .trans files.
Transitions CH4_linelist.trans



In combination with the keyword histogram-J the transition filename can be followed by the J-value this file is associated with.

  • Threshold: Intensity threshold. A line is skipped from line profile evaluation, or simply from the output if the corresponding absorption coefficient/emissivity is smaller than Threshold.
Threshold 1e-28
  • Enermax: Energy Threshold used to select states below some energy value
enermax  20000.0
  • Gaussian, Gauss, Gauss0, Doppler, Doppl, Doppl0, Box, Bin, Rect, Sticks, Stick, Voigt, pseudo, pse-Rocco, pse-Liu, Voig-Quad, trans: The type of the line profile.




  • Sampling: used together with the line profile to indicate that a sampling (not averaging) version will be used. For example
gaussian sampling

Currently this makes sense only in combination with Gaussian and Doppler.

  • HWHM, HalfWidth: Half width at the half maximum, used for Gauss, Lorenz.
`HWHM  0.1`
  • Mass, Masses: Effective molecular mass (u0), used for Doppl.
mass 16.0
  • Verbose: verbose level, defines the amount of data to be printed out in the output Verbose 3
  • histogram computing crosssections using intensities saved on (usually) an equidistant grid.
  • histogram-J computing cross-sections using intensities saved on a grid with J-dependent histogram files.
  • NRAM or Ncash: number of transitions to put into RAM; Alias is LINES-TO-CASH
  • Nprocs: Number of OMP processes. Ideally should be the same as the number of omp-processors allocated, but would work with any number. Aliases are OMP_NUM_PROCS and OMP_PROCS.
  • mem: maximal memory allocated for the job. It is used to estimate how many transitions can be put into RAM. Should be less than or equal too the memory of the system.
  • offset (alias line-cutoff) is to define line-width cut-off in wavenumbers (cm-1). The default value is 25 cm-1.
offset 25 (cm-1)
  • line-cutoff (alias offset) is to define line-width cut-off in wavenumbers (cm-1). The default value is 25 cm-1.
offset 25 (cm-1)

offset or line-cutoff (line-wings cut-off) can be defined in terms of HWHM:

offset 50 HWHM

offset`s can allow using different values for different regions, see the multi-grid section `grid.

  • pressure value in bar must be specified (otherwise P=1 bar is assumed) used for Voigt.
  • Species or Broadener starts a section to define the Voigt-type broadening parameters

\(\gamma = \sum_i \gamma_i (T^0_i/T)^n P/P^0_i r_i\)

gamma or gamma0 is the reference HWHM (cm-1) for Voigt used in Species

n is the exponent n_i for Voigt used in Species

delta is the pressure shift.

T0 is the reference T (K), usually 298 for Voigt used in Species

P0 is the reference pressure in bar, usually 1 for Voigt used in Species

ratio is the mixing ratio of the species (unitless) for example the solar mixing ratio of H2 and He is 0.9 and 0.1 (species).

file is the name of the file with broadening parameters

model is the broadening model, e.g. J (alias a0) or JJ (alias a1).

nquad is the number of quadrature points used for Voigt-Quad

The name of the species should be the first thing on the line.


mass 16.0
pressure 0.5
Temperature 1300.0
  H2  gamma 0.05 n 0.4 t0 298.0 ratio 0.9
  He  gamma 0.04 n 1.0 t0 298.0 ratio 0.1

Example 2:

  H2  gamma 0.0207 n 0.44 t0 298.0 file 1H2-16O__H2__a1.broad model JJ ratio 0.84
  He  gamma 0.043 n 0.02 t0 298.0 file  1H2-16O__He__a1.broad model JJ ratio 0.16
  • unc is the keyword for specifying the uncertainty threshold. it should appear in the filer section.

  • filter is the section name for specifying the filters

  • Phoenix is the keyword for converting ExoMol line list to the Phoenix format. The species are expected to specify the Voigt parameters of the broadeners. Phoenix should appear anywhere in the main body of the input in the way as a line profile keyword.

  • HITRAN is to use the HITRAN-format of the transition file or output. Reading from hitran (.par) requires also the definition of the partition function pf and the isotopologue number iso. No .states is needed. To read from HITRAN use HITRAN READ

  • OXFORD is to convert to the Oxford-format. The input structure is the same as used for writing in the HITRAN format.

  • TRANS is used to partitioning the line list (.trans) files into weak and strong part defined for a reference temperature. Currently used in combination with

    cutoff exp.

  • WATT or WATT/STR/MOLECULE can be used after the emission keyword to switch to the watt/str/molecule untis. E.g.

    emission watt

  • PHOTONS or PHOTONS/S can be used after the emission keyword to switch to the photons/s untis. E.g.

    emission photons/s

  • SPECTRA is to use the SPECTRA-format (IAO.ru, Tomsk) of the transition file. This will also require the definition of the (i) reference temperature, (ii) partition function for the target temperature, (iii)

partition function for the reference temperature, and (iv) the molecule/isotope pair (iso).


Temperature  500.0  ref 296.0
Range 0.0  10000.0

Npoints 10001


iso 26 1
pf 1000.0 ref 500

output CH4_voigt_T500K

Transitions  SpectraMol_CH4_296K.txt
  • iso is to define the isotopologue for HITRAN or SPECTRA molecule, e.g. 26 1 for 12C2H2.


   iso 26 1
   pf 1000.0
   output C2H2_ab_g0.5
   Transitions  26_hit12.par

`HITRAN` can also form a section where there HITRAN-error codes are specified, e.g.
 error-E  qn 4 ierr 4  vmax  10 ierr 3  vmax  20 ierr 2  vmax  30  ierr 1  vmax  40 ierr 0  vmax  100
 error-S  qn 4 ierr 5  vmax  10 ierr 4  vmax  20 ierr 3  vmax  100
 error-Air   ierr 4
  • grid is to define a multi-grid with different resolutions in different sub-grids as follows


  Range   0    100   Npoints 10000 offset 25.
  Range 100   1000  Npoints 1000
  Range 1000 10000  Npoints 100

Here offset can be substituted by LINE-CUTOFF.

  • error-E and error-S are used to specify the ranges for the quantum numbers for different Energy and intensity error-codes, respectively.
  • qn is the quantum number (the number of the QN-column after J) used for the error-specification.
  • ierr is followed by the error code, followed by
  • vmax (keyword) followed by the maximum value of qn this error code applies, which is followed by another error code.
  • error-Air ierr to give the error code for the HITRAN air-broadening (one constant value).
  • error-self ierr to give the error code for the HITRAN self-broadening (one constant value).
  • error-n ierr to give the error code for the HITRAN n-broadening (one constant value).
  • error-delta ierr to give the error code for the HITRAN line-shift (one constant value).
  • non-LTE: A name of the section for non-LTE calculations containing the infomation on the columns with the vibrational quantum numbers, vibrational densites as well as the reference vaue of \(J\).
  • Jref: a keyword specifying the reference \(J\) value used to define reference vibrational energies taken as rovibrational (rovibronic) term values from the States file at this value of \(J\) as part of the non-LTE calculations.
  • density: a keyword specifying the column that contains the custom vibrational densities used in the non-LTE calculations.
  • Vald: a single keyword trigering the conversion to the VALD format with log10(gf) factors using the astronomical convention for the statistical weight.
    To convert to the Astrophysical convention of the degeneracy factors, use gf_factor. For example, for water the convertion factor is 0.25. This will modify both gf and the g-factor.
gf_factor 0.25
  • GF: computer gf-factors. See the note about the gf_factor in Vald to convert to the astrophyical convention.
  • gf_factor: a keyword to convert gf and g-factors to the Astrophysical convention of the degeneracy. For example for water it should 0.25 (see note for Vald).
  • predissociation: a free floating single-valued keyword to invoke the pre-dissociative line broadening, where the lifetimes are used to estimate the pre-dissociative HWHM. For this option, the States file must contain the lifetimes and its column number must be specified in the QN (nono-LTE) section using the keyword lifetime, unless it is the default column 6:

 lifetime 5


The lifetime column specification can be combined with the non-LTE section, since QN and non-LTE are essentially aliases of each other.