This project is a simple Python wrapper for the ATLAS LTE stellar modelling code by Robert L Kurucz as well as satellite utilities including the spectral synthesis library SYNTHE and opacity distribution (ODF) calculator DFSYNTHE. Refer to the literature references below:

• Kurucz (1970): detailed description of an older version of the code (version 5)
• Kurucz et al. (1974): more on opacity distribution functions
• Kurucz & Avrett (1981): standard reference for the SYNTHE code
• Castelli & Kurucz (2003): description of the "new" ODF format, currently used by ATLAS
• Castelli (2005): user guide to DFSYNTHE
• Kurucz (2005a): more information on modern versions of ATLAS as well as SYNTHE
• Kurucz 2005b: bound-bound (line) opacity treatment in the code and line lists
• Kurucz (2014): brief overview of the modern versions of the code (9 and 12)

Also see the websites of Robert L Kurucz and Fiorella Castelli.

ATLAS 9 and ATLAS 12 differ in the opacity sampling method (opacity distribution functions vs direct sampling) which results in the former being considerably faster and, in principle, somewhat less accurate although (to my understanding) the significance of the uncertainty introduced by the chosen opacity treatment is not well established. BasicATLAS works with ATLAS 9 (support for ATLAS 12 may be added in the future).

If you are using BasicATLAS in your research, please cite Larkin et al. (2022)

• Roman Gerasimov (University of Notre Dame, University of California San Diego)
• Mikaela Larkin (University of California San Diego)
• Tianxing Zhou (University of California San Diego)
• Philipp Edelmann (Los Alamos National Laboratory) @ AAS 242 Hack Together
• Paul Barrett (George Washington University) @ AAS 242 Hack Together
• Efrain Alvarado (University of California San Diego)
• Evan Kirby (University of Notre Dame)

BasicATLAS uses a grid of pre-computed model atmospheres as initial guesses for temperature-pressure profiles in structure calculations. This repository contains a reduced grid of models; however, we recommend downloading the full grid from Google Drive and placing it in BasicATLAS/restarts/light.h5.

Run the wget-based bash script download.com to retrieve the original ATLAS/SYNTHE/DFSYNTHE Fortran codes and required data files (line lists, chemical constants etc):

source download.com

To compile the legacy code, both Intel and GNU Fortran compilers are required. Intel compilers are typically not available by default, but can be installed as part of the oneAPI Toolkit. Follow the instructions on the oneAPI website to install the Toolkit for your operating system (both GUI and command line installers are available). Once installed, the Toolkit must be initialized before every use by running the setvars.sh script:

sh ~/intel/oneapi/setvars.sh

Once installed and initialized, you must be able to use the ifx command in your terminal. Then run the compile script in BasicATLAS:

source compile.com

Finally, verify that the installation completed successfully by running

python test.py

The test is clean if no output is produced.

BasicATLAS is shipped with the PyTLAS sub-module, which implements a modified version of the SYNTHE suite that runs entirely in memory. PyTLAS is necessary e.g. for abundance fitting with chemfit. To enable PyTLAS:

cd PyTLAS
source compile.com

This should generate spectrv.so, synthe.so and xnfpelsyn.so in BasicATLAS/PyTLAS/bin/.