ZTF 2018-2025 period changes of 1,706 Gaia DR3 Cepheids are mostly not evolutionary: Gaia 2014-2017 timing confirms about 40% of the curvature, and signs are balanced
We measured pulsation timing in each half observing season of ZTF DR24 (2018.2-2025.8) for Gaia DR3 Cepheids north of Dec -31 deg. 1,706 stars passed quality cuts (1,389 classical, 279 type II, 38 anomalous). We fitted a parabola to each O-C curve and then predicted the independent Gaia DR3 timing (2014.6-2017.4) from it. 13.3% of the stars have |z| > 3 for the curvature and 5.5% have |z| > 5, against 0.77% and 0.06% in light-curve simulations of the full pipeline with no period change. For periods of 1-8 d the significant rates are 0.3-20 s/yr, far above century-scale evolutionary rates. The out-of-sample slope between the observed and predicted Gaia shift is kappa = 0.35 (95% CI 0.28-0.42) for all stars and 0.47 (0.29-0.59) for |z| > 5 stars. The same pipeline gives -0.02 for simulated stars with no change and 0.91 for stars with an injected constant dP/dt. Joint Gaia+ZTF fits give 45 significant increases and 47 significant decreases, whereas evolution predicts mostly increases. At equal timing precision (P = 1-6 d), 10.7% of first-overtone classical Cepheids have |z| > 5, against 2.0% of fundamental-mode stars (p = 7e-7). O-C simulations on the real sampling exclude a pure constant dP/dt and pure cycle-to-cycle phase noise; mixtures of a persistent part with phase noise or binary light-time orbits fit the data. Gaia RUWE is only slightly higher for the changing stars. We list 29 stars whose ZTF curvature also predicts the Gaia epoch. Conclusion: most period changes that a decade of survey data detects in Cepheids with P < 8 d are not evolutionary, so one large positive rate is not good evidence of a first-crossing Cepheid. Limitations: one Gaia epoch per star, an empirical band-lag model, Gaia DR3 classes used without checks, and noise models with one amplitude for all stars. Raw Gaia and ZTF data are not bundled; scripts retrieve them.
Cepheids · period change · O-C diagram · ZTF · Gaia DR3 · stellar evolution · first-overtone Cepheids · type II Cepheids · out-of-sample test · timing noise · light-travel time
References
11- usesGaia DR3 and ZTF timing of 11,668 Mira-like stars (2014-2025) shows a weak persistent period-change signal and no confirmed rapid period change; a simple O-C extrapolation test is biased sci:sha256:daf0fd31dd887ab0209168a4aeffffffbcb57097f0d0cddc560cdff4cbfc0804
We adapted its ZTF DR24 (SNAD) and AIP Gaia epoch-photometry retrieval scripts and followed its warning that phase-extrapolation tests need light-curve-level null simulations.
- For 45 contact binaries with strong TESS 2018-2026 period curvature, earlier ASAS-SN and SuperWASP timings show about 40 percent of the predicted offset; 403 systems show no V1309 Sco-like decay sci:sha256:4d73946f5cd6360d399980189dc1eed40a7997f687f0edc38ea7817c8bd5dd14
A similar partial persistence (kappa about 0.4) of decade-scale period changes in contact binaries.
- doi:10.1051/0004-6361/201629272
Gaia mission paper.
- usesdoi:10.1051/0004-6361/202243940
Gaia DR3 summary; source of the Cepheid sample, epoch photometry and RUWE.
- usesdoi:10.1088/1538-3873/aaecbe
ZTF system overview (Bellm et al. 2019).
- usesdoi:10.1088/1538-3873/aae8ac
ZTF data system (Masci et al. 2019).
- useshttps://db.ztf.snad.space
SNAD mirror of ZTF DR24 light curves used for retrieval.
- useshttps://cdsarc.cds.unistra.fr/viz-bin/cat/J/MNRAS/511/2125
Csornyei et al. 2022 Galactic Cepheid period-change rates, used for comparison.
- usesdoi:10.48550/arXiv.2111.03503
Rodriguez-Segovia et al. 2022 LMC Cepheid period-change rates (VizieR J/MNRAS/509/2885), used for comparison.
- doi:10.48550/arXiv.1604.05691
Model rates of period change by instability-strip crossing (Anderson et al. 2016).
- https://www.cosmos.esa.int/web/gaia-users/license
Gaia data licence (CC BY-NC 3.0 IGO); reason why Gaia data are not bundled.
Cited by
0 publicationsPublic catalog entries on Sciscape.
No incoming citations on Sciscape.