TESS photometry shows that under 1% of Gaia DR3 SB1 orbits at 0.25-0.3 d and 2.6-6% at 0.3-0.5 d are real, for dwarf hosts, minimum companion mass below 0.15 Msun and ellipsoidal model scale 0.22-1

Gaia DR3 gives single-lined spectroscopic (SB1) orbits for many stars. Prior work shows that some short-period SB1 orbits are not correct. We used TESS light curves to test 7784 Gaia DR3 SB1 orbits with periods of 0.25 to 10 d, dwarf host stars and minimum companion mass below 0.15 Msun. If an orbit is real, the tide of the companion causes an ellipsoidal light variation at half the orbital period. We calculated the predicted amplitude from the Gaia orbit and the stellar parameters. We measured the amplitude, or an upper limit, in TESS Quick-Look Pipeline light curves. An orbit is ruled out if no signal has SNR 5 or more and the upper limit is less than 0.5 times the prediction for 99.7% of isotropic inclinations. TESS detected 696 signals and ruled out 1312 orbits. We could not test 5776 orbits. At 0.25-0.3 d, TESS rules out 338 of 376 orbits, and a likelihood model gives a real fraction of 0.000 (95% upper limit 0.008). At 0.3-0.5 d, TESS rules out 501 of 642 orbits, and the real fraction is 0.040 (0.026-0.060). These two fractions do not change when the scale of the amplitude model changes from 1 to 0.22. At 0.5-3 d the fraction is 0.11-0.26 and depends on the model. The SB1 periods pile up at the 0.25 d search limit of Gaia: 80 orbits are at 0.250-0.252 d, 2.4 times the mean of the near bins. The ruled-out orbits contain no known TESS eclipsing binary and no TOI. Signals that we injected in the real light curves never gave an upper limit below 0.5 times the injected amplitude (0 of 1775). A mixture fit of the signal phases shows that 79% (75-82%) of the detections have the phase of the Gaia conjunction. The quality score of Bashi et al. (2022) separates detections from ruled-out orbits with AUC 0.81, but only 0.67 below 1 d. Limitations: detections are on average 0.35 times the model amplitude, for an unknown cause, and a ruled-out orbit can still have a companion at a different period. The package gives the verdict table with Gaia source_id and TIC, the code, the figures and the validation tests. It does not contain the raw Gaia values because of their license. A script downloads them again and rebuilds the full tables.

Gaia DR3 · spectroscopic binaries · SB1 orbits · TESS · ellipsoidal variation · period aliasing · brown dwarfs · low-mass companions · catalogue validation · false positives · light curves · Quick-Look Pipeline · population fraction · injection-recovery

Admitted 2026-11-09 18:06 UTC3 citing publications

References

usesCatalog values for 2,668 Gaia DR3 SB1 entries with periods below 10 days fail a detached-star test under a periastron Roche model sci:sha256:8df05a84d11a9aa0d6da7b79e3a0663f5158596f8fd43b84e26fe40637edb9c0
We use its Roche-test flag table. TESS rules out 330 of its 351 flagged orbits in our sample, which supports the flags.
supportsGaia eclipse periods disagree with 32 of 39 SB1 entries flagged by a detached-star test in a 1,643-source sample sci:sha256:6ce1fde432bc3216b9e2a416e8b8e9b19a6fe0263e063e4bb571d6a9792b2347
It finds that Gaia eclipse periods disagree with most flagged short-period SB1 entries. Our TESS test finds that most SB1 orbits below 0.5 d are not real.
usesGaia DR3 SB1 orbits predict eclipse times with errors 1.45 times too small and 2.8% gross outliers, in a test with 1582 Gaia and 416 TESS eclipsing binaries of period mostly below 10 d sci:sha256:dd865fb0fe9055c98a3460d9be2bbf502e8b89f6a76ac3e6eb991affe46753d7
We multiply the Gaia SB1 ephemeris errors by its factor 1.45 in the phase test of the detections.
usesdoi:10.1051/0004-6361/202243940
Gaia DR3 summary (Vallenari et al. 2023).
usesdoi:10.1051/0004-6361/202243782
Gaia DR3 non-single-star catalogue (Arenou et al. 2023), source of the SB1 orbits.
usesdoi:10.1051/0004-6361/202450600
Gaia SB1 processing and the 4 per day frequency limit (Gosset et al. 2025).
usesdoi:10.1093/mnras/stac2928
We recompute and test the SB1 quality score of Bashi et al. (2022).
usesdoi:10.3847/1538-3881/ab3467
TIC 8 (Stassun et al. 2019) for TIC identifiers, radii and contamination.
usesdoi:10.3847/2515-5172/abca2e
TESS Quick-Look Pipeline light curves (Huang et al. 2020a).
usesdoi:10.3847/2515-5172/abca2d
TESS Quick-Look Pipeline light curves (Huang et al. 2020b).
usesdoi:10.3847/1538-4365/ac324a
TESS eclipsing binary catalogue (Prsa et al. 2022) for the cross-match.
usesdoi:10.1086/173488
Ellipsoidal amplitude formula (Morris and Naftilan 1993).
supportsdoi:10.3847/1538-3881/ad0135
Schmidt et al. (2023) verify some SB1 solutions with transiting companions; we extend the check to all short-period low-mass SB1 orbits.
supportsdoi:10.48550/arXiv.2310.12380
It uses the NSS catalogue against TESS candidates; we test the NSS orbits themselves with TESS.
doi:10.1051/0004-6361/202244207
Prior work on Gaia and TESS together for binary orbits (Panahi et al. 2022).
supportsdoi:10.48550/arXiv.2412.02082
It states that aliasing is the main cause of bad SB1 solutions; we find a period pile-up at the 0.25 d limit and few real orbits below 0.5 d.
doi:10.48550/arXiv.2207.05086
Prior work that uses photometry to examine a small number of Gaia binary orbits.
doi:10.48550/arXiv.2401.09531
Prior work that uses photometry to examine a small number of Gaia binary orbits.
doi:10.48550/arXiv.2510.05982
Prior work that uses photometry to examine a small number of Gaia binary orbits.

Cited by

3 publications

Public catalog entries on Sciscape.

supportsTESS shows too few eclipses at the Gaia DR3 SB1 periods of 13-80 MJ companions of FGK dwarfs: at least 27% of these orbits at 0.5-20 d are not real (95%, stellar-control calibration)
An ellipsoidal test with TESS found that few Gaia DR3 SB1 orbits at P < 0.5 d are real. Our Roche test and eclipse test agree.
supportsFewer than 10 % of Gaia DR3 SB1 orbits with m2min < 0.08 Msun are real: TESS shows 14 transits where 404 are expected (P ≤ 16 d) and 65 ellipsoidal signals where 922 are expected (r = 1)
Their QLP ellipsoidal test finds that few Gaia DR3 SB1 orbits with m2min < 0.15 Msun are real. Our independent PDCSAP ellipsoidal test and TESS transit test give the same result for m2min < 0.08 Msun (F < 0.1) and F = 0.37 for 0.08-0.15 Msun. We use their verdict table for a star-by-star comparison (section 4.6). Our calibration with known transiting companions gives c = 0.63, and our results give possible causes of their scale of 0.35.
supportsTESS eclipse-like dips of 116 of 240 Gaia DR3 SB1 systems with M2min < 0.15 Msun occur at the Gaia-predicted conjunction, against 47 by chance, with the Gaia time uncertainties multiplied by 1.85
That work tested the same type of Gaia DR3 SB1 orbits for TESS ellipsoidal variation. It rules out none of the 22 vetted candidates that are in both samples. In the orbits that it rules out, our timing matches occur at the chance rate (2 against 1.97). In the orbits where it detects the signal, 136 match against 53.0 by chance.