Hunting exomoons with a kilometric baseline interferometer

TL;DR

The paper argues that exomoons, as probes of planet formation, migration, and potential habitability, can be robustly detected via astrometric wobble with a kilometric-baseline optical interferometer achieving $1\,\mu$as precision. It presents a methodological framework and sensitivity analyses showing Earth-mass and sub-Earth-mass moons orbiting Jupiter-like planets at 50–200 pc could be detected within $0.3R_{\mathrm{Hill}}$ using 12–18 astrometric epochs. The work demonstrates that such an instrument would outperform current facilities and complement ELT-driven direct imaging, enabling new insights into moon survival, system architectures, and habitable environments in exoplanetary systems. Overall, it lays out a tangible path to expanding exomoon science from detection to characterization with transformational astrometric precision.

Abstract

Despite numerous search campaigns based on a diverse set of observational techniques, exomoons - prospective satellites of extrasolar planets - remain an elusive and hard-to-pin-down class of objects. Yet, the case for intensifying this search is compelling: as in the Solar System, moons can act as proxies for studying planet formation and evolution, provide direct clues as to the migration history of the planetary hosts and, in favourable cases, offer potentially habitable environments. Here, we present an investigation into how the search for exomoons would benefit from a new interferometric facility operating in the optical wavelength domain and leveraging baselines substantially longer than the ones the VLTI is currently equipped with. We find that an interferometer providing an astrometric precision of 1$\,μ$as would be able to robustly detect Earth-mass and sub-Earth-mass exomoons on dynamically stable orbits around Jupiter-like planets at distances between 50 and 200 pc.

Figures (2)

• Figure 1: Left: Schematic of the orbital wobble exhibited by a planet as a result of being orbited by a moon. The system parameters in the top right indicate the baseline 'plane parallel moon' case. The adjustments made for the alternative configurations are indicated in brackets in the bottom left. This is a reproduction of Fig. 1 in Winterhalder et al. (2025). Right: Schematic loosely sketching the parameter space accessible using the astrometric technique as compared to the transit methods of exomoon detection in the context of the moon survival rate as a function of planetary orbital period. Additionally, the expected peak of the gas giant occurrence around the water ice line as well as the predicted ELT planet detection range are indicated.
• Figure 2: 5$\sigma$ moon detection sensitivity curves around an exemplary exoplanet ($M_\mathrm{s}=1.5M_\odot$; $a_\mathrm{pl}=10AU$; $i_\mathrm{pl}=0^\circ$, i.e. face-on; $e=0$) using $N_\mathrm{e}=12$ and 18 astrometric epochs. The two line colours correspond to the different parallax settings of $\varpi=5$ and $20mas$ as indicated in the left panel, the solid and dashed lines relate to the host planet mass settings of $1$ and $10M_{Jup}$, respectively. The dynamical instability domain beyond $0.3$ times the respective Hill radius, $R_\mathrm{Hill}$, is shaded red for the different planetary masses. The mass ranges below one Neptune and one Earth-mass are shaded in light and dark blue.