Constraining nearby sub-stellar companion architectures using high-contrast imaging, radial velocity, and astrometry

¹ ETH Zurich, Institute for Particle Physics and Astrophysics, Wolfgang-Pauli-Strasse 27, 8093 Zurich, Switzerland
² Department of Astronomy, University of Michigan, Ann Arbor, MI 48109, USA
³ ETH Zurich, Department of Earth and Planetary Sciences, Sonneggstrasse 5, 8092 Zurich, Switzerland
⁴ Département d’Astronomie, Université de Genève, 1290 Versoix, Switzerland

^★ Corresponding author: lia.sartori@phys.ethz.ch

Received: 20 November 2024
Accepted: 3 July 2025

Abstract

Context. Nearby stars offer prime opportunities for exoplanet discovery and characterisation through various detection methods. By combining high-contrast imaging (HCI), radial velocity (RV), and astrometry, it is possible to better constrain the presence of sub-stellar companions, as each method probes different regions of their parameter space. A detailed census of planets around nearby stars is essential to guide the selection of targets for future space missions seeking to identify Earth-like planets and potentially habitable worlds. In addition, the detection and characterisation of giant planets and brown dwarfs is crucial for understanding the formation and evolution of planetary systems.

Aims. We aim to constrain the possible presence of sub-stellar companions for seven nearby M-dwarf stars (GJ 3325, GJ 1125, GJ 367, GJ 382, GJ 402, GJ 465, and GJ 357) using a combination of new SPHERE/H2 HCI and archival RV and astrometric data. We investigate how combining these techniques improves the detection constraints for giant planets and brown dwarfs compared to using each method individually.

Methods. For each star and each dataset, we computed the mass limits as a function of the semi-major axis or projected separation using standard techniques. We then used a Monte Carlo approach to assess the completeness of the companion mass vs. semi-major axis parameter space probed by the combination of the three methods, as well as by the three methods independently.

Results. Our combined approach significantly increases the fraction of detectable companions. We quantify improvements of up to ∼ 60% over RV alone (improvement achieved for GJ 3325), ∼ 50% over HCI alone (for GJ 3325), and ∼ 12% over astrometry alone (for GJ 367). Although no new companion was detected, we could place stronger constraints on potential sub-stellar companions.

Conclusions. The combination of HCI, RV, and astrometry provides significant improvements in the detection of sub-stellar companions over a wider parameter space. Applying this approach to larger samples and lower-mass companions will help to constrain the search space for future space missions aimed at finding potentially habitable or even inhabited planets.