Marcs - Home

Welcome to MARCS

This is a grid of one-dimensional, hydrostatic, plane-parallel and spherical LTE model atmospheres. These may be used together with atomic and molecular spectral line data and software for radiative transfer to generate synthetic stellar spectra.

The MARCS site contains about 52,000 stellar atmospheric models of spectral types F, G and K in 3 different formats and also flux sample files indicating rough surface fluxes. The data files are downloadable in limited amounts in the form of tar archives after registration. We ask users to cite the basic reference containing a description of the models: Gustafsson B., Edvardsson B., Eriksson K., Jørgensen U.G., Nordlund Å., Plez B. 2008, Astronomy & Astrophysics 486, 951.

MARCS low-resolution sampled fluxes of F and G type models were compared to six observational data bases by Edvardsson (2008). Synthetic Strömgren uvby-Hβ colours of MARCS fluxes were compared to those observed for field- and globular cluster stars by Önehag et al. (2009).

File types

The ASCII data files are individually gzipped and the file types are identified by their file name extensions:

".mod" contain the most detailed lists of the model structure, thermodynamical quantities, and many partial gas pressures. Their sizes gzipped/unzipped are typically 10/25 kB. Here is a FORTRAN program to read these models.
".opa" contain the basic model structures and continuous opacities as a function of wavelength and optical depth useful for synthetic spectrum calculations. Their sizes gzipped/unzipped are typically 0.4/1.4 MB. Here is a FORTRAN program to read the opacity data files.
".krz" contain model files in a format explained HERE . Their sizes gzipped/unzipped are typically 2/6 kB.
".flx" contain statistical samples of the model surface fluxes derived in the model calculations. They give rough estimates of the surface fluxes and are NOT synthetic spectra why they should not be used for synthetic photometry. See below and Plez (2008) for a discussion of the effects on the model fluxes of the opacity sampling frequency. The flux units are erg/cm²/s/Å. The wavelengths range from 1300 to 200,000 Å in vacuum and come in a separate file "flx_wavelengths.vac.gz". The wavelength step is NOT equidistant but has rather a constant resolving power R=λ/Δλ=20,000. The file sizes are typically 0.4/1.4 MB.

Model Parameters

The stellar atmospheric model parameters range in effective temperature from 2500 to 8000 K in steps of 100 K from 2500 to 4000 K and 250 K between 4000 and 8000 K. The logarithmic surface gravities (cgs units) between −1.0 and 5.5 in steps of 0.5. Overall logarithmic metallicities relative to the Sun are between −5.0 and +1.0 in variable steps. The reference solar abundance mixture is that of Grevesse, Asplund & Sauval (2007). The atmospheres of more luminous stars with low surface gravities (−1.0 to +3.5) are calculated in spherical geometry and with different stellar masses: the standard mass is 1.0 solar mass with sparse grids at 0.5, 2.0, and 5.0 Msun. The spectral differences of using spherical rather than plane-parallell models and/or radiative transfer for luminous stars were explored by Heiter & Eriksson (2006). Plane parallel models are computed for gravities between 3.0 and 5.0 or 5.5 for the cooler models. Models with different choices of "microturbulence parameters" are also available.

Since the mixture of metals vary among stars there are presently seven different "metallicity classes": "standard", "alpha poor", "alpha enhanced", "alpha negative", "mildly CN cycled", "heavily CN cycled", and for solar [Fe/H] there is also a subgrid with abundances according to "Grevesse & Sauval (1998)". The "standard" mixture reflects the typical elemental abundance ratios in stars as a function of metallicity [Fe/H] in the solar neighbourhood. The model grids lack many models due to convergence problems, particularly for low-gravity models, and specifically the warm low-gravity corners of the sperical-geometry subgrids are all empty.

The tables below display the abundance mixtures for the different metallicity classes. The "pp" and "sph" columns display the available microturbulence parameter values (km/s) for plane parallell and spherical model geometries, respectively. The complete elemental composition of a model is given in the ".mod" files and in a different format in the ".krz" files.

Standard, 30% of all model
[Fe/H]	[α/Fe]	[O/Fe]			sph
+1.00	0.00	0.00	1	2		2	5
+0.75	0.00	0.00	1	2		2	5
+0.50	0.00	0.00	1	2		2	5
+0.25	0.00	0.00	1	2		2	5
0.00	0.00	0.00	1	2	1	2	5
−0.25	+0.10	+0.10	1	2		2	5
−0.50	+0.20	+0.20	1	2		2	5
−0.75	+0.30	+0.30	1	2		2	5
−1.00	+0.40	+0.40	1	2		2	5
−1.50	+0.40	+0.40	1	2		2	5
−2.00	+0.40	+0.40	1	2		2	5
−2.50	+0.40	+0.40	1	2		2	5
−3.00	+0.40	+0.40	1	2		2	5
−4.00	+0.40	+0.40	1	2		2	5
−5.00	+0.40	+0.40	1	2		2	5

Alpha enhanced, 15% of all models
[Fe/H]	[α/Fe]	[O/Fe]			sph
+1.00	+0.40	+0.40	1	2	2	5
+0.75	+0.40	+0.40	1	2	2	5
+0.50	+0.40	+0.40	1	2	2	5
+0.25	+0.40	+0.40	1	2	2	5
0.00	+0.40	+0.40	1	2	2	5
−0.25	+0.40	+0.40	1	2	2	5
−0.50	+0.40	+0.40	1	2	2	5
−0.75	+0.40	+0.40	1	2	2	5

Alpha poor, 14% of all models
[Fe/H]			sph
−0.25	1	2	2	5
−0.50	1	2	2	5
−0.75	1	2	2	5
−1.00	1	2	2	5
−1.50	1	2	2	5
−2.00	1	2	2	5
−2.50	1	2	2	5

Alpha negative, 23% of all models
[Fe/H]	[α/Fe]	[O/Fe]			sph
+1.00	−0.40	−0.40	1	2	2	5
+0.75	−0.40	−0.40	1	2	2	5
+0.50	−0.40	−0.40	1	2	2	5
+0.25	−0.40	−0.40	1	2	2	5
0.00	−0.40	−0.40	1	2	2	5
-0.25	−0.40	−0.40	1	2	2	5
-0.50	−0.40	−0.40	1	2	2	5
-0.75	−0.40	−0.40	1	2	2	5
-1.00	−0.40	−0.40	1	2	2	5
-1.50	−0.40	−0.40	1	2	2	5
-2.00	−0.40	−0.40	1	2	2	5

Mildly CN cycled, ¹²C/¹³C=20, 8% of all models
[Fe/H]	[α/Fe]	[C/Fe]	[N/Fe]	[O/Fe]	sph
+0.25	0.00	−0.13	+0.31	0.00	2	5
0.00	0.00	−0.13	+0.31	0.00	2	5
−0.50	+0.20	−0.13	+0.31	+0.20	2	5
−1.00	+0.40	−0.13	+0.31	+0.40	2	5
−1.50	+0.40	−0.13	+0.31	+0.40	2	5
−2.00	+0.40	−0.13	+0.31	+0.40	2	5

Heavily CN cycled, ¹²C/¹³C=4, 8% of all models
[Fe/H]	[α/Fe]	[C/Fe]	[N/Fe]	[O/Fe]	sph
+0.25	0.00	−0.38	+0.53	0.00	2	5
0.00	0.00	−0.38	+0.53	0.00	2	5
−0.50	+0.20	−0.38	+0.53	+0.20	2	5
−1.00	+0.40	−0.38	+0.53	+0.40	2	5
−1.50	+0.40	−0.38	+0.53	+0.40	2	5
−2.00	+0.40	−0.38	+0.53	+0.40	2	5

Grevesse & Sauval (1998), 2% of all models
[Fe/H]	[α/Fe]	[C/Fe]	[N/Fe]	[O/Fe]	pp			sph
+0.05	+0.11	+0.08	+0.09	+0.12	0	1	2	2	5

About the MARCS model flux files

The flux data files are not proper spectra, merely one-dimensional files of sampled photospheric surface fluxes (in ergs/cm²/s/Å) from 1300 Å to 20 μm sampled with a constant λ/Δλ = 20,000. This means that spectral lines are not resolved and that much information is missing. The convolution of a flux file will not help and still gives seriously erroneous line strengths. The average flux in a limited wavelength region may also give a poor representation of the integrated flux in that region.

The wavelength scale of the flux sample data is in vacuum, and the wavelengths (about 100,000 points) are supplied in a separate one-dimensional file, which is the same one for all models.

References

Edvardsson B. 2008, Physica Scripta, Vol T133, 014011
Grevesse N., Asplund M., Sauval A.J. 2007, Space Science Review 130, 205
Grevesse N., Sauval A.J. 1998, Space Science Review 85, 161
Gustafsson B., Edvardsson B., Eriksson K., Jørgensen U.G., Nordlund Å., Plez B. 2008, Astronomy & Astrophysics 486, 951
Heiter U., Eriksson K. 2006, Astronomy & Astrophysics 452, 1039.
Önehag A., Gustafsson, B., Eriksson K., Edvardsson B. 2009, Astronomy & Astrophysics 498, 527
Plez B. 2008, Physica Scripta, Vol T133, 014003