RGBfromGaiaEDR3

This Python script predicts RGB magnitudes from Gaia EDR3 photometric data. These magnitudes are given in the standard system defined by Cardiel et al. (2021a).

The code performs a cone search defined by coordinates right ascension and declination on the sky and a search radius. The predictions make use of the polynomial transformations given by Eqs. (2)-(5) in Cardiel et al. (2021b; hereafter C21)

The cone search is performed making use of the Astroquery coordinated package of astropy.

You need to have a live connection to the Internet for the script to work!

Installing the code

In order to keep your current Python installation clean, it is highly recommended to install a python 3 virtual environment and install all the dependencies there.

Creating and activating the python virtual environment

python3 -m venv venv_rgb
. venv_rgb/bin/activate

Downloading custom repos

We have experienced problems installing pyvo from pypi. However, downloading the latest version from GitHub worked fine

(venv_rgb) $ mkdir venb_rgb/repos/
(venv_rgb) $ cd venb_rgb/repos/
(venv_rgb) repos$ git clone [email protected]:astropy/pyvo.git
(venv_rgb) repos$ cd pyvo
(venv_rgb) pyvo$ python setup.py install

Installing additional packages

(venv_rgb) $ cd
(venv_rgb) $ pip install astroquery matplotlib

Downloading and executing the code

(venv_rgb) $ mkdir venb_rgb/repos/
(venv_rgb) $ cd venb_rgb/repos/
(venv_rgb) repos$ git clone https://github.com/nicocardiel/RGBfromGaiaEDR3.git
(venv_rgb) repos$ cd RGBfromGaiaEDR3

Besides the dependencies listed above, this script itself does not need further installation. Just execute it from the command line:

(venv_rgb) repos$ cd RGBfromGaiaEDR3
$ python RGBfromGaiaEDR3.py 56.66 24.10 1.0 12

The last instruction executes the RGBfromGaiaEDR3.pyscript providing the four positional arguments: right ascension, declination, search radius and limiting Gaia G magnitude. Note that the coordinates and search radius must be given in decimal degrees.

The first time you execute the code, the auxiliary file edr3_source_id_15M_allsky.fits (size 129 Mb), containing the source_idof the Gaia EDR3 stars belonging to the ~15 million star sample of C21, is automatically downloaded. When using the --starhorse, a larger file (size 289Mb), containing additional star parameters, is downloaded instead.

The script executes the following steps:

• Step 1: cone search in Gaia EDR3, gathering the following parameters: source_id, ra, dec, phot_g_mean_mag, phot_bp_mean_mag and phot_rp_mean_mag.

• Step 2: cross-matching of the previous EDR3 sample with the list of ~15 million stars from C21. This step determines the subsample of EDR3 stars for which the RGB photometric calibration is reliable.

• Step 3: cone search in Gaia DR2. This additional step is performed in order to retrieve the phot_variable_flag parameter indicating whether the star was flagged as variable in DR2. Note that this flag is not available in EDR3.

• Step 4: cross-matching between DR2 and EDR3 to identify the variable stars in EDR3. This step is required because it is not guaranteed that the same astronomical source will always have the same source identifier in the different Gaia Data Releases.

• Step 5: computation of the RGB magnitudes using the polynomial transformations given in Eqs. (2)-(5) of C21.

• Step 6: generation of the output files. Three files (in CSV format) are generated:

  • rgbsearch_15m.csv: stars belonging to the ~15 million star sample of C21 (with reliable RGB magnitude estimates).

  • rgbsearch_var.csv: objects flagged as variable in DR2.

  • rgbsearch_edr3.csv: remaining objects in EDR3. The RGB magnitude estimates of these objects can be potentially biased due to systematic effects introduced by interstellar extinction, or by exhibiting non-solar metallicity, or a colour outside the Gaia -0.5 < G_BP-G_RP < 2.0 interval. This file will typically contain more stars than the rgbsearch_15m.csv selection.

  The three CSV files provide the same 11 columns:

  • number: consecutive number of the star in each CSV file
  • source_id: identification in EDR3
  • ra: right ascension (from EDR3)
  • dec: declination (from EDR3)
  • b_rgb: blue RGB magnitude estimate
  • g_rgb: green RGB magnitude estimate
  • r_rgb: red RGB magnitude estimate
  • g_br_rgb: pseudo-green RGB magnitude estimate, defined in C21 as the arithmetic mean of the blue and red RGB magnitudes
  • phot_g_mean_mag: Gaia G magnitude (EDR3)
  • phot_bp_mean_mag: Gaia G_BP magnitude (EDR3)
  • phot_rp_mean_mag: Gaia G_RP magnitude (EDR3)

  The list of objects in those files is sorted by right ascension.

  When using --starhorse, the file rgbsearch_15m.csv contains 3 additional columns providing parameters derived by Anders et al. (2019):

  • av50: 50th percentile of the interstellar extinction
  • met50: 50th percentile of the metallicity [M/H]
  • dist50: 50th percentile of the distance (kpc)

• Step 7: generation of a finding chart plot (in PDF format): rgbsearch.pdf. The execution of the previous example generates a cone search around the Pleiades star cluster: The stars in this plot are color coded based on the Gaia G_BP - G_RP colour. A red circle has been overplotted on the stars belonging to the ~15 million star sample of C21, a blue square on the variable objects in DR2, and a grey diamond on EDR3 stars outside the Gaia -0.5 < G_BP - G_RP < 2.0 colour interval. Stars brighter than a pre-defined threshold are displayed with big star symbols. To facilitate the identification of each star, the consecutive star number in the three files (rgbsearch_15m.csv, rgbsearch_edr3.csv and rgbsearch_var.csv) is also displayed (in red, black and blue, respectively). These numbers are not displayed when using the parameter --nonumbers in the command line.

Note that the four output archives (1 PDF and 3 CSV files) share the same root name rgbsearch. This can be easily modified using the optional argument --basename <newbasename> in the command line.

Additional help

Some auxiliary optional arguments are also available. See description invoking the script help:

$ python RGBfromGaiaEDR3.py --help

...
...

positional arguments:
  ra_center             right Ascension (decimal degrees)
  dec_center            declination (decimal degrees)
  search_radius         search radius (decimal degrees)
  g_limit               limiting Gaia G magnitude

optional arguments:
  -h, --help            show this help message and exit
  --basename BASENAME   file basename for output files
  --brightlimit BRIGHTLIMIT
                        stars brighter than this Gaia G limit are displayed 
                        with star symbols (default=8.0)
  --symbsize SYMBSIZE   multiplying factor for symbol size (default=1.0)
  --nonumbers           do not display star numbers in PDF chart
  --noplot              skip PDF chart generation
  --nocolor             do not use colors in PDF chart
  --starhorse           include StarHorse av50, met50 and dist50 in rgbsearch_15m.csv
  --verbose             increase program verbosity
  --debug               debug flag

Citation

If you find this Python script useful, please cite Cardiel et al. (2021a) (to quote the use of the standard RGB system) and Cardiel et al. (2021b) (where the transformation between the Gaia photometry and the RGB magnitudes is derived).

Bibliography

Anders et al. (2019), https://ui.adsabs.harvard.edu/abs/2019A%26A...628A..94A/abstract

Cardiel et al. (2021a), MNRAS, in press, https://ui.adsabs.harvard.edu/abs/2021MNRAS.tmp..971C/abstract

Cardiel et al. (2021b), MNRAS, in preparation