6.2.3 Auxiliary data

The ground-based radial velocities of standard stars are used in wavelength calibration to fix the zero point (Section 6.3).

The calibration dataset consists of 2576 standard stars described in Soubiran et al. (2018b) (CU6GB-cal), for which the radial velocity has been obtained with high accuracy and precision ($\sigma (V_{\mathrm{rad}})\sim 0.1$$\mathrm{km}{\mathrm{s}}^{-1}$). In addition, another 5929 standard stars were selected from XHIP measurements, having uncertainties  $\mathrm{km}{\mathrm{s}}^{-1}$ and quality A and B, and having a match in the master list in Crifo et al. (2010). The XHIP stars zero point was corrected to the one of the CU6GB-cal stars. The analysis of the RVS observations of the calibration stars acquired during the dry run, permitted to identify and exclude from the standard-star list those stars having variable radial velocity, those having a velocity different from the one in the catalogue, and those showing double-lined spectra.

In addition to the standard stars used in wavelength calibration, there are also auxiliary ground-based radial velocities of validation stars. The validation dataset is used in the Automated Verification steps (Section 6.5.1).

Table 6.3 lists the catalogues of standard stars used for calibration, together with the catalogues used for validation.

The knowledge of the atmospheric parameters, $T_{\mathrm{eff}}$, $\log g$, and $[\mathrm{Fe}/\mathrm{H}]$, of the stars is essential for the RVS pipeline. The atmospheric parameters are used to select the appropriate synthetic spectrum to generate a template as much as possible similar to the RVS spectrum, which is then used to estimate, via correlation with the RVS spectrum (Section 6.4.8), the spectroscopic radial velocity and the broadening velocity of the star. The atmospheric parameter information is also used to select the stars having a spectral type required by the calibration module. For example, wavelength calibration needs spectra with deep lines, and the input spectra selection criterion uses bright (), non-saturated spectra with $T_{\mathrm{eff}}$: 4500 $T_{\mathrm{eff}}$  6500 K.

The auxiliary atmospheric parameter file contains ground-based parameters from the literature for about 2.3 million stars with $G_{\mathrm{RVS}}$ $\le$ 14 (see Table 6.4). The brightest star parameters are from Soubiran et al. (2014) (cleaned from some inaccurate data), and the large majority of the star parameters, about 1.6 million, are from LAMOST-DR4 Xiang et al. (2017) (the data with valid error, SNR$>$20 and $T_{\mathrm{eff}}$ in range 4000–7500 K were selected). The auxiliary atmospheric parameters are associated with the $\sim$2.3 million stars for which they are available.

In Gaia DR3 the spectroscopic processing took place before the atmospheric parameters processing (Chapter 11) and could not have access to the atmospheric parameters produced for DR3; therefore a preliminary processing was done to provide atmospheric parameters to the RVS pipeline.

The RVS pipeline associates atmospheric parameters to each star with the following priority order: 1) the auxiliary atmospheric parameters, 2) the parameters obtained by the preliminary processing of GSP-Spec (Section 11.3.4); 3) the parameters obtained by the preliminary processing of GSP-Phot (Section 11.3.3); 4) the parameters obtained by the RVS pipeline method called DetermineAp and described in Section 6.4.5; 5) the default solar parameters ($T_{\mathrm{eff}}$ =5500 K; $\log g$ = 4.5 dex and $[\mathrm{Fe}/\mathrm{H}]$ = 0 dex).

The auxiliary atmospheric parameter catalogue is input to the RVS pipeline, while the preliminary GSP-Spec and GSP-Phot parameters are input via the DPAC data model CompleteSource, containing the data available from the other processing pipelines (astrometric and photometric processing).

The information on how the atmospheric parameters associated with the star were obtained is stored in the column rv_atm_param_origin. The vast majority of the stars, $\sim$ 75 %, have their parameters from GSP-Phot (i.e. rv_atm_param_origin=111), while approximately 6 % (2.2 million) have the auxiliary ground-based atmospheric parameters associated (i.e. rv_atm_param_origin=222).

The distribution of the atmospheric parameters associated with the stars for which the RVS pipeline computed the radial velocity is shown in Figure 6.3.

The synthetic spectra are used to produce the templates (i.e. the synthetic spectrum is convolved with the RVS LSF-AL, and resampled, see Section 6.4.8).

A large set of synthetic spectra has been calculated (Sordo et al. 2011), for both the low-resolution spectrophotometry and the medium-resolution RVS data. The RVS spectra have been recalculated in 2016; in particular the MARCS library grid has been extended to the cool stars, several line lists have changed and the continuum opacity treatment has been improved. For more information on the MARCS and A-type synthetic spectra used in the spectroscopic pipeline, see Blomme et al. (2017). The OB-type library has been recalculated in 2020, it is included in the library, but has not been used to estimate the radial velocities published DR3, which are limited to rv_template_teff $\le \mathrm{14\hspace{0.17em}500}$K (Section 6.5.2).

The auxiliary synthetic spectra library contains 6772 spectra selected among the following grids:

• •

  MARCS: 4306 spectra $T_{\mathrm{eff}}$: 2500–8000 K, step 100 K in the range 2500–3900 K and 250 K in the range 4000–8000 K; $\log g$: -0.5–+5.0 dex, step 0.5 dex; $[\mathrm{Fe}/\mathrm{H}]$: -5.0, -4.0, -3.0, -2.5, -2.0, -1.5, -1.0, -0.75, -0.50, -0.25, 0.0, +0.25, +0.5, +0.75, +1.0 (dex);

• •

  A-type: 304 spectra: $T_{\mathrm{eff}}$: 8500–14 500 K, step 500 K; but missing 12 500 K. $\log g$: 2.0–5.0 dex, step 0.5 dex; $[\mathrm{Fe}/\mathrm{H}]$: -0.5–+0.25 dex, step 0.25 dex;

• •

  OB-type: 2162 spectra (not used for the data published in DR3) $T_{\mathrm{eff}}$: 15 000–55 000 K, step 1000 K in the range 15 000–30 000 K, step 2500 K in the range 30 000–50 000 K; $\log g$: highest value: 4.75 dex, lowest: approximately linearly from 1.75 dex at 15 000 K to 4.0 dex at 55 000 K; $[\mathrm{Fe}/\mathrm{H}]$: -0.3, 0.0, +0.3 (dex).

To reduce the computation time in some of the software modules, a subset of the synthetic spectra dataset (Section 6.2.3) is used, consisting of the 28 models listed in Table 6.5. This selection is based on the expected number of stars of a given spectral type/luminosity class, determined from the Gaia Universe Model Snapshot (Robin et al. 2012, specifically their Tables 16 and 17).

The subset is used notably in the modules DetermineAP (Section 6.4.5) and TodCor (Section 6.4.8).

The $G_{\mathrm{RVS}}^{\mathrm{ref}}$ magnitudes of the stars contained in this input catalogue are used as the reference to estimate the $G_{\mathrm{RVS}}$ zero-point (Sartoretti et al. 2023). $G_{\mathrm{RVS}}^{\mathrm{ref}}$ has been computed for 110 843 Hipparcos stars using the formula $G_{\mathrm{RVS}}^{\mathrm{ref}}=V-0.0501-1.1667*(V-I)+0.0052*{(V-I)}^2+0.0011*{(V-I)}^3$, where $V$ and $I$ are the magnitudes in the Hipparcos catalogue (Jordi et al. 2010). The stars flagged as photometric variable in DR2 (phot_variable_flag) were removed from the reference stars.

This auxiliary catalogue contains 1335 high-resolution spectra that are used for the calibration of the LSF-AL (Section 6.3.4). 706 spectra have been obtained with NARVAL (Bernard Lyot telescope at Pic du Midi), 470 with ESPaDonS (Canada France Hawaii Telescope), 135 with UVES (VLT), and 24 with the Potsdam Echelle Polarimetric and Spectroscopic Instrument (PEPSI).