🔭 astrophysics

Astrometric exomoon detection by means of optical interferometry

This paper demonstrates that astrometric time series measurements, particularly from optical interferometry, can detect and characterize exomoons around giant planets and brown dwarfs, with current data already capable of inferring a massive satellite around AF Lep b and future facilities promising the detection of Earth-like moons in habitable zones.

Original authors: T. O. Winterhalder, A. Mérand, J. Kammerer, S. Lacour, M. Nowak, W. O. Balmer, G. Bourdarot, F. Eisenhauer, A. Glindemann, S. Grant, Th. Henning, P. Kervella, G. -D. Marleau, N. Pourré, E. Rickman

Published 2026-02-04

📖 5 min read🧠 Deep dive

CC BY 4.0

Original authors: T. O. Winterhalder, A. Mérand, J. Kammerer, S. Lacour, M. Nowak, W. O. Balmer, G. Bourdarot, F. Eisenhauer, A. Glindemann, S. Grant, Th. Henning, P. Kervella, G. -D. Marleau, N. Pourré, E. Rickman

Original paper licensed under CC BY 4.0 (http://creativecommons.org/licenses/by/4.0/). ✨ This is an AI-generated explanation of the paper below. It is not written or endorsed by the authors. For technical accuracy, refer to the original paper. Read full disclaimer

Imagine you are watching a dance partner on a stage. If they are dancing alone, they move in a smooth, predictable circle around the center. But if they are holding hands with a smaller partner (a moon) that they are spinning around, their own path isn't a perfect circle anymore. Instead, they wobble slightly, tracing a tiny, bumpy loop as they move.

This is exactly what astronomers are trying to do with exomoons—moons that orbit planets around other stars. So far, no one has successfully "seen" one directly. This paper proposes a new way to find them by watching for that specific "wobble."

Here is a breakdown of the paper's ideas using simple analogies:

1. The Problem: The Moon is Invisible

Usually, when we look for planets, we look for the planet itself or the shadow it casts. But moons are tiny and dim compared to their giant planet hosts. Trying to see a moon next to a planet is like trying to spot a firefly sitting on a car headlight from miles away. The light of the headlight (the planet) completely drowns out the firefly (the moon).

2. The Solution: Watching the "Wobble"

Instead of trying to see the moon, the authors suggest watching the planet's movement.

The Analogy: Imagine a heavy adult (the planet) spinning a small child (the moon) around them. Even though the adult is much heavier, they don't spin perfectly in place. They lean and shift their weight slightly to keep the child from flying off.
The Science: The planet is actually orbiting a shared center of gravity with its moon. This causes the planet to make a tiny, rhythmic "wobble" in its path around its star. If we can measure this wobble very precisely, we know a moon is there, even if we can't see it.

3. The Tool: A Cosmic Ruler

To see this wobble, you need a ruler that is incredibly precise.

Current Tool (GRAVITY): The paper uses an instrument called GRAVITY at the Very Large Telescope in Chile. It is like having a ruler that can measure the width of a human hair from 10 kilometers away. It is precise enough to detect the wobble of giant planets, but it's still a bit too "fuzzy" to catch the wobble caused by very small moons.
Future Tools (PLANETES & KBI): The authors look at future instruments that will be even sharper. One is like upgrading from a standard ruler to a laser measure. Another is a "kilometer-baseline" interferometer, which is like building a telescope the size of a city to get a view so sharp it could theoretically spot an Earth-sized moon.

4. The Strategy: Taking Many Snapshots

You can't catch a wobble with just one photo. If you take a picture of a spinning dancer once, you might catch them at a moment where they look perfectly still.

The Analogy: To prove someone is wobbling, you need to take a series of photos over time (an "astrometric time series").
The Paper's Plan: The authors simulated taking 12 or 18 "photos" (observations) over several months. They found that with current technology (GRAVITY), we can already detect large moons around giant planets. For example, they calculated that with 12 observations, we could prove the existence of a moon around the planet AF Lep b that is about 14% the mass of Jupiter.

5. What They Found

Right Now: We can find "Super-Moons" (moons that are huge, like mini-planets) around giant gas planets using current technology.
Soon: With the next generation of instruments (like the proposed PLANETES), we could find smaller moons, perhaps the size of Neptune or Saturn, around planets like Beta Pictoris b.
In the Future: If we build a massive, city-sized telescope (the Kilometer-Baseline Interferometer), we could finally detect Earth-sized moons in the "habitable zone" (where life could exist). This would be like spotting a tiny pebble wobbling on a spinning top from across the room.

6. The Catch

The paper notes that this method works best for planets that are far away from their stars (like Jupiter is from our Sun). Moons close to their stars are often unstable and get kicked out, so the best candidates for this "wobble hunt" are the distant, cold giants.

Summary

This paper is a "feasibility study." It says: "We have a new way to hunt for exomoons by watching planets wobble. Our current tools can find big moons, and our future tools will be sharp enough to find Earth-sized moons. We just need to take enough pictures over time to catch the dance."

It does not claim we have found a moon yet, but it proves that the math and the technology are finally lining up to make that discovery possible.

Technical Summary: Astrometric Exomoon Detection by Means of Optical Interferometry

Problem Statement
Despite the steady growth in confirmed exoplanet detections, the search for exomoons (satellites orbiting exoplanets) has yielded no conclusive detections to date. Existing methods, such as transit timing variations or direct imaging, face significant limitations. Transit studies are biased toward short-period planets where moons are prone to tidal disruption or ejection, while direct imaging struggles with contrast and resolution for smaller companions. Furthermore, radial velocity (RV) methods applied to planets often lack the precision required to detect the subtle reflex motion induced by a moon, particularly for wide-orbit systems. The authors posit that monitoring the astrometric position of a planet relative to its host star over time offers a viable "middle ground" to detect the periodic "wobble" induced by an orbiting moon.

Methodology
The authors developed an analytical framework to model the astrometric signal of a star–planet–moon system and simulate detection scenarios based on current and future instrumental capabilities.

Analytical Modeling:
- Signal Amplitude: The study defines the astrometric signal amplitude ( $A$ ) as the maximum radial deviation of the planet from a Keplerian orbit, derived from the moon's mass ( $M_m$ ), the planet's mass ( $M_{pl}$ ), the moon's semi-major axis ( $a_m$ ), and the system distance ( $d$ ).
- Time-Series Simulation: Two nested models were implemented: a two-body star–planet model and a three-body star–planet–moon model. The latter superimposes the planet–moon barycentric motion onto the planet's orbit around the star. This approach avoids full three-body numerical integrations by assuming stable systems where the moon's semi-major axis is within 0.531 times the planet's Hill radius.
- Wobble Extraction: By comparing the trajectories of the two models, the authors extract the time-dependent astrometric perturbation in Right Ascension (RA) and Declination (Dec).
Detection Simulation:
- Mock Data Generation: The authors generated mock astrometric epochs by adding Gaussian noise (based on instrument precision) to the fiducial planetary positions.
- Observing Strategies: Two sampling strategies were tested: 12 epochs (spaced by ~29.5 days with a 6-month gap) and 18 epochs (two gaps). The time of periplanet passage was randomized to avoid aliasing effects where regular sampling coincides with the moon's orbital phase.
- Statistical Significance: To distinguish the moon model from the null hypothesis (star–planet only), the authors compared the reduced $\chi^2$ values of the two models. An F-test based on the Fisher-Snedecor distribution was used to convert $\Delta\chi^2_{red}$ into a detection significance (p-value), penalizing the increased complexity of the moon model.
Instrumentation Context:
- The study evaluates sensitivity for VLTI/GRAVITY (50 $\mu$ as precision), the future VLTI/PLANETES (10 $\mu$ as), and a hypothetical 3 km baseline interferometer (1 $\mu$ as).

Key Contributions and Results

Sensitivity Curves: The paper presents the first astrometric exomoon sensitivity curves, mapping detection limits (at 5 $\sigma$ confidence) as a function of moon mass and semi-major axis for various instruments and epoch counts.
Current Capabilities (GRAVITY): The authors demonstrate that with 12 epochs, VLTI/GRAVITY can currently detect massive moons around giant planets. Specifically, they show it is possible to infer the presence of a $0.14\,M_{Jup}$ satellite at a separation of 0.39 AU around AF Lep b, or a $0.34\,M_{Jup}$ moon at 0.59 AU around HR 8799 d.
Future Capabilities:
- VLTI/PLANETES: Improved precision (10 $\mu$ as) would allow the detection of moons with masses around $0.06\,M_{Jup}$ ( $\sim19\,M_{Earth}$ ) at closer separations.
- Kilometre-Baseline Facilities: A future facility achieving 1 $\mu$ as precision could detect Earth-mass moons ( $\sim6\,M_{Earth}$ ) at separations of $\sim0.06$ AU, effectively probing the regime of potentially habitable exomoons.
Case Study ( $\beta$ Pic b): The authors analyze the "P24-moon" hypothesis (a moon suggested to explain $\beta$ Pic b's obliquity). They find that while GRAVITY's current precision (50 $\mu$ as) is likely insufficient to detect the predicted 10–20 $\mu$ as signal, the planned PLANETES instrument (10 $\mu$ as) would be capable of detecting such a signature.
Distinction from Other Perturbations: The study notes that while both inner planets and moons cause astrometric wobbles, their periods differ significantly (years vs. days/months). RV monitoring of the host star can further distinguish between a moon and an inner planet, as a moon does not induce a measurable stellar RV modulation.

Significance and Claims
The paper claims that the astrometric method, particularly when applied to interferometric observations, provides a promising avenue for the first definitive detection of exomoons. The authors emphasize that:

Feasibility: The technique is already viable for characterizing massive moons around giant exoplanets and brown dwarfs using existing facilities like GRAVITY.
Population Building: This method allows for the efficient building of a sample of confirmed objects, which is crucial for calibrating moon formation models (e.g., in-situ formation vs. capture).
Future Potential: With next-generation facilities achieving 1 $\mu$ as precision, the detection of Earth-like moons within the habitable zones of their host stars becomes a realistic proposition.
Limitations: The authors modestly note that current and near-future capabilities do not yet overlap with the theoretical mass ratio limit ( $\sim10^{-4}$ ) for regular, in-situ formed moons, suggesting that initial detections will likely target irregular or massive moons. They also acknowledge that optimal sampling strategies and robust fitting frameworks (e.g., nested sampling) require further development to handle the specific signal-processing challenges of these time-series data.