HD208487c: Saturn-Mass Exoplanet Candidate

• HD208487c is a candidate Saturn-mass exoplanet identified via decades-long radial velocity monitoring and robust Bayesian methods.
• The detection leverages 219 RV measurements over 23 years from multiple high-stability spectrographs with precise cross-calibration.
• Its orbital period of approximately 923 days and moderate eccentricity of 0.19 support a dynamically complex multi-planet system architecture.

HD 208487c is a candidate Saturn-mass exoplanet in orbit around the nearby bright star HD 208487, discovered through high-precision radial velocity (RV) monitoring and subjected to rigorous Bayesian analysis over more than two decades of observations. Its reported orbital period is approximately 923 days, situating it as the central component in a dynamically complex system that also includes at least two additional planet candidates. Detection and characterization of HD 208487c leverages extended RV time series acquired from multiple instruments, advanced periodogram and Bayesian methodologies, and a broad suite of stellar activity and photometric diagnostics to ensure robustness against false positive signals arising from stellar variability (Rubenstein et al., 17 Aug 2025).

1. Observational Data and Instrumental Baseline

RV measurements of HD 208487 span a temporal baseline of 8474 days (JD 2451034.18–2459508.64, about 23.2 years), yielding 219 independent data points. These were acquired with a suite of high-stability spectrographs:

Instrument	# RVs	Typical $\sigma_{\mathrm{RV}}$	Key Features
UCLES (AAT)	48	$2$–$3 \ \mathrm{m\,s}^{-1}$	Iodine cell
HARPS-TERRA (pre-2015)	20	$1 \ \mathrm{m\,s}^{-1}$	Fiber-fed, $R\approx115,000$
HARPS-TERRA2 (post-2015)	106	$1\ \mathrm{m\,s}^{-1}$	Hardware upgraded
PFS (Magellan/PFS)	45	$1\ \mathrm{m\,s}^{-1}$	Iodine cell

Extensive instrument cross-calibration and homogeneity correction are central for time-series combination fidelity. The data’s long temporal coverage and densely populated epochs are essential for resolving multi-year periodicities and for distinguishing planetary signals from activity cycles (Rubenstein et al., 17 Aug 2025).

2. Signal Identification: Periodogram and Bayesian Detection

Periodic signature extraction commenced with high-power Generalized Lomb-Scargle (GLS) periodograms applied to mean-subtracted, detrended RVs. Known planets with periods near 129.4 d and 493 d were first removed. Residuals revealed a strong peak at $P_3 \approx 939.7$ d with normalized power ≈ 0.30 and false-alarm probability (FAP) < 0.1%.

Splitting the RV dataset by S-index (a proxy for stellar activity) demonstrated that this $\sim$937 d signal is prominent in the low-activity subset (ΔBIC ≫ 5), but absent in the high-activity subset. This behavior is inconsistent with activity-driven origin, further supporting a planetary interpretation.

Bayesian evidence was quantified with the EMPEROR Markov Chain Monte Carlo (MCMC) analysis, modeling the RVs as:

$$y_{i,ins} = \gamma_{ins} + \dot\gamma\,t_i + f_k(t_i) + \sum_n c_{n,ins}\,\xi_{n,i,ins} + MA_{ins}(t_i) + \epsilon_{i,ins}$$

with $2$0 representing $2$1 Keplerian components, $2$2 stellar activity indices, $2$3 a moving-average red noise term, and $2$4 encompassing white and additional instrumental noise. Thermodynamic integration with 12 temperatures computed marginal likelihoods $2$5, and model comparisons adopted a strong detection threshold $2$6 (Bayes factor $2$7), consistent with the Kass & Raftery (1995) criterion.

3. Orbital Solution and Physical Parameters

HD 208487c was recovered as the system’s “planet 2” within the preferred “CorrO” model enforcing low orbital eccentricities. The planet’s orbital and physical parameters are:

Parameter	Value
Period $2$8	$2$9 d
RV semi-amplitude $3 \ \mathrm{m\,s}^{-1}$0	$3 \ \mathrm{m\,s}^{-1}$1
Eccentricity $3 \ \mathrm{m\,s}^{-1}$2	$3 \ \mathrm{m\,s}^{-1}$3
Argument of periastron $3 \ \mathrm{m\,s}^{-1}$4	$3 \ \mathrm{m\,s}^{-1}$5 rad
Mean anomaly at JD 2450000 $3 \ \mathrm{m\,s}^{-1}$6	$3 \ \mathrm{m\,s}^{-1}$7 rad
Minimum mass $3 \ \mathrm{m\,s}^{-1}$8	$3 \ \mathrm{m\,s}^{-1}$9

The periastron passage epoch $1 \ \mathrm{m\,s}^{-1}$0 relates to $1 \ \mathrm{m\,s}^{-1}$1 via:

$1 \ \mathrm{m\,s}^{-1}$2

No significant RV signal aliases at $1 \ \mathrm{m\,s}^{-1}$314 d or $1 \ \mathrm{m\,s}^{-1}$428 d—periods associated with stellar rotation and previously challenged candidates—remain, indicating precise separation of planetary and spurious periodicities (Rubenstein et al., 17 Aug 2025).

4. Stellar Activity and Photometric Diagnostics

Robust stellar activity vetting was undertaken, including GLS analysis of:

• S-index (PFS, HARPS; chromospheric activity)
• BIS (HARPS; cross-correlation line bisector)
• FWHM (HARPS; cross-correlation function width)

No significant peak consistent with the planet candidate periods was detected (all FAP < 0.1%), and Pearson RV-activity correlations $1 \ \mathrm{m\,s}^{-1}$5 with $1 \ \mathrm{m\,s}^{-1}$6 indicate no statistical links. Bayesian modeling of FWHM returned a principal periodicity of $1 \ \mathrm{m\,s}^{-1}$7 d, more than $1 \ \mathrm{m\,s}^{-1}$8 offset from the 923 d candidate.

Analysis of ASAS and Hipparcos photometry found no power at planetary periods, with both datasets exhibiting a strong $1 \ \mathrm{m\,s}^{-1}$929 d signal attributable to stellar rotational modulation. The lack of correlated activity signals and periodic photometric variations at 923 d supports a Doppler (planetary) rather than spurious or stellar-origin scenario.

5. Dynamical Architecture and Evolutionary Scenario

System architecture and stability were investigated via direct numerical integrations using REBOUND. A hypothetical initial configuration of six approximately equal-mass ($R\approx115,000$0) Neptunes in a long 3:2 resonant chain, small inclinations, and eccentricities $R\approx115,000$1 was evolved for $R\approx115,000$2 yr. The majority rapidly destabilized; in $R\approx115,000$3 of runs, dynamical scattering produced mergers yielding:

• An inner eccentric giant ($R\approx115,000$4) at $R\approx115,000$5 AU,
• A central planet at $R\approx115,000$62 AU with $R\approx115,000$7,
• An outer survivor at $R\approx115,000$82.5 AU and $R\approx115,000$9.

These outcomes closely mirror the inferred orbital architecture:

Planet	Semi-major axis (AU)	Eccentricity
Inner gas giant	$1\ \mathrm{m\,s}^{-1}$00.53	$1\ \mathrm{m\,s}^{-1}$10.37
HD 208487c (central)	$1\ \mathrm{m\,s}^{-1}$21.94	$1\ \mathrm{m\,s}^{-1}$3
HD 208487d (outer)	$1\ \mathrm{m\,s}^{-1}$42.54	$1\ \mathrm{m\,s}^{-1}$5

This dynamical pathway plausibly accounts for the observed planet spacing, eccentricities, and mass hierarchy (Rubenstein et al., 17 Aug 2025).

6. Context, Significance, and Prospects

The refined detection of HD 208487c replaces earlier Bayesian claims of a $1\ \mathrm{m\,s}^{-1}$6 d planet (Gregory 2007), eliminating aliases at $1\ \mathrm{m\,s}^{-1}$714 d and $1\ \mathrm{m\,s}^{-1}$828 d that confounded prior analyses. The Bayes factor for the 923 d signal addition exceeds 150, indicating strong model preference. Data-split, activity-index, and photometry tests reinforce robustness. A third planet candidate, HD 208487d ($1\ \mathrm{m\,s}^{-1}$9 d, $1\ \mathrm{m\,s}^{-1}$0), is identified and positioned near a 3:2 period resonance with HD 208487c. Periodicities at $1\ \mathrm{m\,s}^{-1}$1460 d, close to the 2:1 harmonic of HD208487c, remain under investigation for possible additional companions.

Continued high-precision, long-baseline RV monitoring is projected to be decisive for confirming candidate planets, refining orbital architectures, and elucidating system dynamical evolution.

HD 208487c is thus established as a Saturn-mass planet ($1\ \mathrm{m\,s}^{-1}$2) on a nearly circular 923 d orbit, with detection substantiated via advanced statistical, activity, and dynamical analyses and consistent with a scenario of gravitationally processed multiple-planet system formation (Rubenstein et al., 17 Aug 2025).