RAVEN – RAdial VElocity, Non-parametric

Non-parametric inference of the radial velocity distribution of stars and identification of binaries in clusters.

The statistical framework implemented in RAVEN is described in Rinaldi & Ramírez-Tannus (2024). The main result from this paper can be reproduced using the content of this repository: if you are interested in doing so, please find the instructions at the bottom of this page.

Basic usage

Install RAVEN with

git clone git@github.com:sterinaldi/raven.git
cd raven
pip install .

The folder structure expected by RAVEN is the following:

cluster
└── stars
    ├─ star_1.txt
    ├─ star_2.txt
    └─ star_3.txt

The naming convention does not matter as long as the folder structure is respected. Each .txt file must contain two columns with radial velocity measurement and error for each available epoch (0 if no error is available).

To analyse a cluster with RAVEN, simply run the command line instruction raven -i path/to/cluster/stars. If you want to specify some bounds for the output plot, add the option -b "[vmin, vmax]" (remember the quotation marks). A different number of draws can be set with -n and the non-parametric inference, once performed at least one time, can be skipped with -p. RAVEN accounts for the potential presence of multiple populations (--n_populations N, where N is the expected number of sub-populations) and/or outliers (--outliers). The complete list of available options can be displayed with raven -h. After the run, the folder will look like this:

cluster
├── stars
│   ├─ star_1.txt
│   ├─ star_2.txt
│   └─ star_3.txt
├── draws
│   ├─ star_1.json
│   ├─ star_2.json
│   ├─ star_3.json
│   ├─ draws_cluster.json
│   └─ posteriors_single_event.json
├── events
│   ├─ star_1.txt
│   ├─ star_2.txt
│   └─ star_3.txt
├── cluster.pdf
├── prob_cluster.txt
├── no_errors.txt
├── p_pop_cluster.txt
├── single_fraction_cluster.pdf
├── samples_fraction_cluster.txt
├── p_single_cluster.pdf
└── p_single_cluster.txt

Output files overview

Content of the output files produced by RAVEN:

• cluster.pdf: radial velocity probability density;
• single_fraction_cluster.pdf: posterior probability density for the fraction of single stars;
• samples_fraction_cluster.txt: posterior samples for the fraction of single stars;
• p_pop_cluster.txt: in case of multiple populations/outliers, a summary of the membership probability for each object included in the analysis;
• p_single_cluster.txt: objects included in the cluster ranked by their probability of being single stars;
• p_single_cluster.pdf: fancy plot displaying the probability of each object of being a single star;
• no_errors.txt: if some of the measurements do not have an associated uncertainty, this file contains the relative uncertainty that has been associated with the measurements.

M17

Rinaldi & Ramírez-Tannus (2024) presents the analysis of 20 O-type stars in the giant star-forming region M17 reported in Ramírez-Tannus et al. (2024). Our results can be reproduced using the code and data stored in this repository.

Once RAVEN is installed, move to the M17 directory and run the analysis:

cd M17
raven -i stars -b "[-20,40]" --sana_variability --n_populations 2 --outliers

The analysis should finish in around 10 minutes on a normal laptop. The plots included in the paper are M17.pdf and p_single_M17.pdf.

Acknowledgments

If you use RAVEN in your research, please cite Rinaldi & Ramírez-Tannus (2024):

@ARTICLE{2024arXiv241107676R,
       author = {{Rinaldi}, Stefano and {Ram{\'\i}rez-Tannus}, Mar{\'\i}a Claudia},
        title = "{Non-parametric identification of single-lined binary candidates in young clusters using single-epoch spectroscopy}",
      journal = {arXiv e-prints},
     keywords = {Astrophysics - Astrophysics of Galaxies, Astrophysics - Solar and Stellar Astrophysics},
         year = 2024,
        month = nov,
          eid = {arXiv:2411.07676},
        pages = {arXiv:2411.07676},
archivePrefix = {arXiv},
       eprint = {2411.07676},
 primaryClass = {astro-ph.GA},
       adsurl = {https://ui.adsabs.harvard.edu/abs/2024arXiv241107676R},
      adsnote = {Provided by the SAO/NASA Astrophysics Data System}
}