Jupiter-like planets might be common in a low-density environment

A central question in exoplanetary science is the prevalence of Solar System-like planetary systems. While models predict that giant planets should form readily around solar-type stars via core accretion, radial velocity (RV) surveys have yielded surprisingly low occurrence rates (6-20%) of Jupiter-like planets (defined here as planets with masses ≥1 Jupiter mass and located 3-12 AU from their stars). These RV studies, however, involve diverse populations of stars formed in various, largely unknown environments, and evolved over billions of years, making comparison with theoretical predictions challenging. Observing Jupiter-like planets in young associations provides an alternative approach, offering insights into their early formation and evolution. The β Pic Moving Group (BPMG), being the nearest and one of the youngest (∼20 Myr) associations of A-F stars, is ideal for direct imaging studies. The detection of multiple Jupiter-like planets in BPMG through high-contrast imaging and astrometry prompts a re-evaluation of the frequency of these planets. This paper further investigates the evidence from BPMG to resolve the discrepancy between RV survey results and direct imaging observations.

Several studies have explored the frequency of Jupiter-like planets using different methods and definitions. Cumming et al. (2008) estimated a 5.5% frequency, while Wittenmyer et al. (2016) reported a higher incidence (∼25%). Zhu (2022) found a frequency of 2.2 ± 0.7%, significantly lower. Fernandes et al. (2019) obtained an even lower value (1.7%). Studies considering stars with inner planets reported much higher occurrence rates (Bryan et al., 2016), but this is not necessarily representative of the overall frequency. Microlensing studies (Gould et al., 2010) also indicated a relatively low frequency (∼16%). These discrepancies highlight the challenges in estimating planet frequencies across different stellar populations, ages and detection methods. The inherent biases and limitations of each technique need to be carefully considered.

This study focuses on 30 stars in the BPMG with masses >0.8M☉. The authors compiled a comprehensive list of stellar and substellar companions to these stars from various sources: Gaia DR3, visual binary catalogs, proper motion anomalies (PMa) from Kervella et al. (2018), high-contrast imaging (HCI) data from SPHERE and GPI, and radial velocity variations. The stability of Jupiter-like planet orbits was assessed using Holman et al.'s (1997) criteria, considering the presence of stellar or brown dwarf companions. To account for the orbital sampling effect in HCI observations, the authors estimated the probability of detecting a planet at a given epoch, considering the coronagraphic mask and the number of observations. Two approaches were employed: one based on the completeness of existing HCI data, and another utilizing the astrometric signal from Hipparcos and Gaia to detect the presence of unseen planets. The methodology also accounts for the fact that many Jupiter-like planets are expected to be below the detection threshold of current instruments. By combining these techniques and using various statistical methods and models, they were able to calculate the actual frequency of Jupiter-like planets in the BPMG. They also considered the mass distribution of giant planets to correct for observational biases.

The analysis revealed that 20 out of the 30 stars might host a stable Jupiter-like planet. Four Jupiter-like planets have been directly imaged in the BPMG. Applying an orbital sampling correction to account for the incompleteness of HCI surveys, the frequency of stars hosting planets with masses >4MJupiter and semi-major axes between 3-12 AU is estimated to be 36 ± 19%. Combining direct and indirect detections (via PMa), this frequency increases to 41 ± 12%. Averaging these estimates, the frequency is 39 ± 12%, considerably higher than the 2.0 ± 1.0% obtained from RV surveys. Furthermore, a mass correction, considering the expected distribution of giant planet masses (1-4MJupiter), suggests that the actual frequency of Jupiter-like planets in the BPMG is likely a factor of 2-3 times higher. Statistical analysis confirms a high probability (99%) that the BPMG stars have at least one Jupiter-like planet with a 95% confidence level of >50%. The high frequency of debris disks around BPMG stars further supports the idea of efficient planet formation in this environment.

The significantly higher frequency of Jupiter-like planets in the BPMG compared to the solar neighborhood is discussed in terms of several factors. First, the BPMG stars are more massive than those typically studied in RV surveys; second, BPMG's young age (20 Myr) compared to the ages of the RV survey stars (several Gyr) implies that older systems might have undergone planet loss; third, the treatment of outer companions differ between RV surveys and this study; and finally, the BPMG's low density (total mass ∼94M☉) might have enabled more efficient planet formation. The lower density minimizes disruptive encounters with other systems and avoids premature disk dispersal caused by radiation from massive stars.

This study strongly suggests that the formation of Jupiter-like planets around stars with masses >0.8M☉ is common in low-density environments like the BPMG. This aligns with predictions from core-accretion models. The discrepancy between BPMG findings and RV survey results highlights the importance of considering stellar environment and age when evaluating planet occurrence rates. Future Gaia data releases are expected to provide a more definitive answer on the frequency of Jupiter-like planets in diverse environments.

The study acknowledges several limitations. The sample size of 30 stars is relatively small. There are uncertainties associated with estimating the orbital eccentricities and applying the mass correction. The model assumes a uniform eccentricity distribution for Jupiter-like planets; however, this distribution may not represent reality perfectly. Despite the limitations, the results provide compelling evidence for a high frequency of Jupiter-like planets in low-density stellar environments.