Exoplanet search with astrometry

Searching for extrasolar planets by direct detection is extremely challenging for current instrumentation. Indirect methods, that measure the effect of a planet on its host star, are much more promising and have indeed led to the discovery of nearly all extrasolar systems known today. While the most successful method thus far is the radial velocity technique, new interferometric instruments like PRIMA at the VLTI will enable us to carry out astrometric measurements accurate enough to detect extrasolar planets and to determine all orbital parameters, including their orbit inclination and true mass. In this article I describe the narrow-angle astrometry technique, how it will be realized with PRIMA, what kind of planets we can find, and what kind of preparatory observations are required.

💡 Research Summary

The paper begins by reviewing the state of exoplanet detection, emphasizing that direct imaging remains extremely difficult with current instrumentation because of limited contrast, angular resolution, and sensitivity. Consequently, almost all known exoplanets have been discovered through indirect techniques that measure the influence of a planet on its host star. The most successful of these techniques to date is the radial‑velocity (RV) method, which detects the Doppler shift of stellar spectral lines caused by the reflex motion of the star. However, RV only yields the product M sin i (minimum mass) because the orbital inclination i is unknown, and it becomes less sensitive for planets on wide orbits or with low masses. Transit photometry provides inclination but requires a fortuitous alignment and is limited to short‑period planets. The authors argue that a complementary method—high‑precision astrometry—can overcome these limitations by directly measuring the tiny sky‑plane displacement of a star caused by an orbiting companion.

The core of the paper describes how the PRIMA (Phase‑Referenced Imaging and Micro‑arcsecond Astrometry) facility at the VLTI (Very Large Telescope Interferometer) will implement narrow‑angle astrometry. PRIMA uses two 1.8‑m auxiliary telescopes that can be placed on baselines up to 200 m. By observing a target star and a nearby reference (or “frame”) star within a field of view of ≤ 1 arcminute simultaneously, PRIMA measures the differential optical path length between the two beams with a laser metrology system. A fringe‑tracking subsystem stabilizes the interferometric phase, while a set of prisms provides real‑time phase referencing. The combination of these subsystems allows the relative position of the target star to be determined with an accuracy better than 10 µas (micro‑arcseconds). The paper details the optical layout, the metrology scheme, atmospheric turbulence mitigation (including dual‑star fringe tracking and real‑time atmospheric modeling), and the data reduction pipeline that extracts astrometric signals from the interferometric observables.

With this level of precision, the authors calculate the detection space for exoplanets. PRIMA will be sensitive to planets with masses down to a few Earth masses (≈ 5 M⊕) on orbital radii between roughly 0.5 AU and 5 AU around nearby (≤ 30 pc) solar‑type stars. This regime includes “super‑Earth” and “mini‑Neptune” planets that are largely inaccessible to RV because their induced velocity amplitudes fall below current detection thresholds, especially at larger orbital distances. Moreover, because astrometry measures the full three‑dimensional reflex motion, it yields the true planetary mass and orbital inclination, removing the sin i ambiguity inherent to RV. In multi‑planet systems, PRIMA can disentangle the individual astrometric signatures, enabling precise determination of mutual inclinations and dynamical interactions, which are critical for testing planet‑formation theories such as core accretion, migration, and dynamical scattering.

The paper also discusses the preparatory observations required to maximize PRIMA’s scientific return. First, the target star’s activity level must be characterized through high‑resolution spectroscopy and photometric monitoring to assess the impact of starspots, flares, and granulation on the astrometric signal. Second, suitable reference stars must be identified; they should be close enough on the sky to share the same atmospheric column, have comparable brightness, and exhibit minimal intrinsic motion. The authors present a selection algorithm that combines Gaia astrometry, spectral type, and photometric stability to generate a catalog of optimal frame stars. Simulations indicate that careful pre‑selection can improve the final astrometric precision by more than 30 % and reduce systematic errors caused by differential chromatic refraction.

In the concluding section, the authors argue that PRIMA’s narrow‑angle astrometry will complement existing RV and transit surveys, providing the missing piece of the exoplanetary puzzle: the true mass and three‑dimensional orbital architecture. This capability will allow a statistically robust census of planetary systems, especially for long‑period, low‑mass planets that dominate the planet‑formation models but remain under‑sampled. The paper also outlines technical challenges that still need to be addressed, such as long‑term phase stability, calibration of the metrology laser, and handling of systematic errors arising from instrumental flexure. Future upgrades—such as incorporating multiple reference stars, extending the baseline, and integrating PRIMA data with Gaia astrometry—are proposed to push the detection threshold toward Earth‑mass planets in the habitable zones of nearby stars. Ultimately, the authors envision PRIMA as a pathfinder for the next generation of astrometric missions, bridging the gap between indirect detection and the eventual direct imaging of Earth‑like worlds.