11.4 Quality assessment and validation 11.4.5 Evolutionary Parameters 11.4.7 Groups of stars

11.4.6 Extinction, Dust & ISM

Author(s): Thavisha Dharmawardena, Douglas J. Marshall, Mathias Schultheis, Ronald Drimmel, René Andrae, Yves Frémat

We compare the extinction estimates from GSP-Phot and ESP-HS in Figure 11.87a. Evidently, if both modules agree on the temperature of a star they also agree on its extinction, which is the case for the vast majority of stars having results from both GSP-Phot and ESP-HS. In fact, for 1 073 423 stars both modules agree in temperature to within 10% and the RMS differences between both modules are 0.130 mag for $A_{0}$ , 0.121 mag for $A_{\rm G}$ , and 0.066 mag for $E(G_{\rm BP}-G_{\rm RP})$ . These comprise $\sim$ 68.5% of all stars that have results from both modules. The overall excellent agreement is also emphasised by Figure 11.87b, showing that 50% of stars with results from both GSP-Phot and ESP-HS differ in $A_{0}$ by 0.124 mag or less and 90% of stars by 0.470 mag or less.

Figure 11.87: Comparison of extinction estimates between GSP-Phot and ESP-HS. Panel a: RMS difference between azero_esphs and azero_gspphot as function of relative temperature difference between GSP-Phot and ESP-HS (black step function). The grey step function shows the (rescaled) density distribution. Panel b: Cumulative distribution of absolute differences between azero_esphs and azero_gspphot with dashed lines marking the 50% and 90% quantiles.

To further illustrate the quality of extinction estimates from GSP-Phot and ESP-HS, we compare extinction estimates to colours involving near-infrared photometry from AllWISE (Cutri et al. 2013). In particular, the $G_{\rm BP}-W_{2}$ colour has a very long wavelength leverage such that $A_{\rm BP}\gg A_{W2}$ and hence we can approximate its reddening as

E(G_{\rm BP}-W_{2})=A_{\rm BP}-A_{W2}\approx A_{\rm BP}

(11.50)

Indeed, Figure 11.88a shows that GSP-Phot’s $A_{\rm BP}$ estimate follows a trend with slope 1 with the $G_{\rm BP}-W_{2}$ colour for stars with similar effective temperatures. For ESP-HS, no estimate of $A_{\rm BP}$ is available. Yet, we can use ag_esphs and Figure 11.88b shows that it also follows a trend with slope 1 with the $G-W_{2}$ colour for stars with similar effective temperatures.

Figure 11.88: Comparison of near-infrared colours to extinctions from GSP-Phot and ESP-HS. Panel a: abp_gspphot vs $G_{\rm BP}-W_{2}$ colour using AllWISE (Cutri et al. 2013) colour-coded by teff_gspphot. The diagonal dashed line has slope 1 and is arbitrarily shifted. Panel b: ag_esphs vs $G-W_{2}$ colour using AllWISE colour-coded by teff_esphs. The diagonal dashed line has slope 1 and is arbitrarily shifted. The data shown in both panels represent randomly selected subsets of the full Gaia DR3 sample.

A comparison of the $A_{0}$ median measurements obtained for open clusters to the means obtained by Cantat-Gaudin et al. (2020) is provided in Fouesneau et al. (2023). Values derived by both modules tend to be overestimated in average by about $0.1$ mag. While for ESP-HS the offset increases with the extinction, in GSP-Phot its magnitude is expected to depend most on the stellar library.

Gaia Collaboration et al. (2023i) discuss the overall relation between the DIB equivalent width at 862 nm obtained from RVS spectra and the reddening computed by GSP-Phot and ESP-HS. It is shown that the two values are well correlated, and consistent in the cool and hot temperature domains.

Table 11.44: Extinction and DIB parameter estimates in Gaia DR3.

monochromatic at