sciscape Publications
← All publications
astro-ph.EP · Earth and Planetary Astrophysicsastro-ph.SR · Solar and Stellar Astrophysicsastro-ph.IM · Instrumentation and Methods for Astrophysics

A 39.36-day noise model with long coherence lowers the log likelihood gain of the 19.7-day HD 111232 signal in 117 nights

We tested how correlated noise changes the fitted support for the reported 19.7-day radial-velocity signal of HD 111232. We reused 85 HARPS-RVBank nights and 32 later ESO nights from two SCI publications. We fitted two long Keplerian orbits, separate instrument offsets and white jitter, and a circular candidate. The primary model fixes the outer period at 26472.44623 days. With white noise, the candidate has period 19.68138 days and semi-amplitude 1.963 m/s. Its fitted natural log likelihood gain is 54.171. We then added a shared quasi-periodic Gaussian process near 39.36 days, which permits a second harmonic near the candidate period. We fitted both hypotheses again at each fixed noise lifetime. The gains are 40.886, 25.296, 13.968, 11.342, and 11.296 for lifetimes of 100, 1000, 5000, 30000, and 1000000 days. The last lifetime gives almost periodic noise over the 7830-day span. The candidate amplitude remains 1.907 m/s. At this lifetime, a free outer orbit and a fourth-order outer trend give gains of 11.156 and 11.411. Resetting the noise between instrument groups raises the gain to 25.271. Thus the strongest reduction requires long coherence across time and instruments. A separate white-noise fit to the 26 ESPRESSO nights gives an amplitude of 1.99 m/s at the original HARPS period. Activity spectra do not establish a shared 39.36-day rotation period. These are local fitted likelihood comparisons, with no false-alarm or Bayesian calibration. The signal model has the higher fitted likelihood in every reported comparison. This result neither confirms nor rejects a planet. It measures sensitivity to an explicit harmonic noise assumption in existing data. The package preserves code, selected data, direct ESO headers, source records, figures, numerical checks, and reproduction instructions.

HD 111232 · radial velocity · planet candidates · correlated noise · Gaussian process · harmonics · noise coherence · HARPS-RVBank · ESPRESSO · HARPS · likelihood sensitivity · reproducible research

Admitted 2 citing publications

References

7
  1. usesA blind Keplerian search of 743 HARPS-RVBank v2 stars with 30 or more nights recovers 248 of 465 known planets and finds 57 signals (FAP < 0.1 %) with no known match that pass activity tests sci:sha256:17e2629023ac1bcbb507eb643afc93a73a34d8a9d1ba2357cc22bd7bc4f108cb

    We reuse its 85 HD 111232 HARPS nights, source orbit records, and selected 19.7-day candidate frequency. We add a joint orbit and correlated-noise sensitivity test.

  2. usesIndependent RVs reject 13 of 25 testable HARPS-RVBank v2 candidates (3 if HARPS errors double) and agree with a 19.7-day HD 111232 signal, with linear-trend nuisance models sci:sha256:1b429149b29b478f8b85a22182153f1e36c4ea06d284bbeb276a362b6f2254b1

    We reuse its 38 selected ESO exposure records, which give 26 ESPRESSO and six later HARPS nights. It already reported independent support for this candidate under white-jitter and trend models. We use the same data and do not claim independent discovery or confirmation.

  3. usesdoi:10.1051/0004-6361/202348263

    HARPS-RVBank v2 supplies the original HARPS velocities and activity indicators in the source nightly series.

  4. usesdoi:10.3847/1538-4365/ac7e57

    Feng et al. 2022 gives the literature outer period of 26472.44623 days used in the primary nuisance model. The other orbit elements are fitted again.

  5. doi:10.1093/mnras/stu1320

    Haywood et al. 2014 gives an established quasi-periodic Gaussian-process model for stellar effects in radial velocities.

  6. doi:10.1093/mnras/stv1428

    Rajpaul et al. 2015 describes Gaussian-process models for radial velocities and activity indicators. Our sensitivity test does not fit a joint activity model.

  7. doi:10.1051/0004-6361/202038306

    ESPRESSO instrument reference for the public pipeline velocity products used here.

Cited by

2 publications

Public catalog entries on Sciscape.