skip to main content

gaia data release 3 documentation

10.12 Short-timescale variables

10.12.5 Quality assessment and validation


A time series cleaning operator dedicated to per-CCD photometry has been applied (see Section 10.2.3), specific to the short-timescale analysis and based on the expected amplitude of the variation in the G band. Consistency of the G-band photometry with the BP and RP photometry was ensured by limiting the discrepancy between fluxes in the G versus GBP and GRP bands. Visual inspection of candidate light curves enabled us to further refine the selection criteria and clean suspicious short-timescale samples.


Given the sensitivity of short-timescale methods to changes in magnitude per unit time, variations typical of short-period objects (such as EW-type eclipsing binaries and δ Scuti stars) and/or sources with rapid changes in magnitude (such as the eclipses of EA-type binaries, the brightening of RR Lyrae stars of type RRab, and even low-amplitude outbursts) are naturally detected as short-timescale candidates.

About 40 % of the short-timescale candidates are found in the cross-match of Gaia DR3 sources with a compilation of variable stars from the literature (Gavras et al. 2023), among which about 80 % are classified as eclipsing binaries, 10 % as RR Lyrae stars, 6 % as solar-like variables (rotational modulation, BY Draconis, and RS Canum Venaticorum stars), and 4 % as δ Scuti stars.