🔭 astrophysics

CORALIE radial-velocity search for companions around evolved stars (CASCADES) V. Three planetary companions and achievable precision

The CASCADES survey utilized long-term CORALIE radial velocity measurements to confirm three massive planetary companions orbiting two low-luminosity red giant stars while demonstrating an effective observing strategy to mitigate stellar pulsation noise.

Original authors: P. Figueira (Instituto de Astrofísica de Andalucía-CSIC, Observatoire Astronomique de l Université de Genève), E. Fontanet (Observatoire Astronomique de l Université de Genève), J. P. Faria (Observato

Published 2026-06-09✓ Author reviewed ⓘ

📖 4 min read☕ Coffee break read

CC BY 4.0

Original authors: P. Figueira (Instituto de Astrofísica de Andalucía-CSIC, Observatoire Astronomique de l Université de Genève), E. Fontanet (Observatoire Astronomique de l Université de Genève), J. P. Faria (Observatoire Astronomique de l Université de Genève), M. Esseldeurs (Instituut voor Sterrenkunde, KU Leuven), E. Friden (Observatoire Astronomique de l Université de Genève), A. Leleu (Observatoire Astronomique de l Université de Genève), R. Luque (Instituto de Astrofísica de Andalucía-CSIC), G. Ottoni (Observatoire Astronomique de l Université de Genève), D. Ségransan (Observatoire Astronomique de l Université de Genève), M. Stalport (Space sciences, Technologies and Astrophysics Research, Astrobiology Research Unit, Université de Liège), S. Tavella (Observatoire Astronomique de l Université de Genève, European Southern Observatory, Chile), S. Udry (Observatoire Astronomique de l Université de Genève)

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 by the authors. For technical accuracy, refer to the original paper. Read full disclaimer

Imagine the universe as a giant, bustling dance floor. For a long time, astronomers have been looking for planets (the dancers) around stars that are like our Sun (the steady, middle-aged hosts). But this paper is about a different kind of party: finding planets around stars that are getting older, bigger, and a bit more "wobbly" as they evolve into giants.

Here is the story of the CASCADES project, explained simply:

The Challenge: Dancing with a Wobbly Partner

Finding planets around these older stars is tricky. Think of a massive star like a giant, aging dancer. As they age, they expand and their surface starts to ripple and pulse, like a giant jelly wobbling on a plate. These ripples create "noise" that can look exactly like a planet tugging on the star.

Furthermore, these stars are often very hot and spin fast, which makes it hard to see the tiny "wobble" caused by a planet. It's like trying to hear a whisper in a hurricane.

The Mission: The "Stethoscope"

The team used a very precise instrument called CORALIE, which acts like a super-sensitive stethoscope listening to the stars. They listened to three specific stars (HD 125136, HD 127195, and HD 220218) for about 15 to 18 years. That's a long time to listen! They were looking for a rhythmic "tug" in the star's movement that would reveal a hidden planet.

The Discoveries: Finding the Dancers

After years of listening and filtering out the noise, they found three new planetary systems:

HD 125136: They found one massive planet, about 2.3 times the size of Jupiter. It takes about 850 days (a little over two years) to orbit its star.
HD 127195: This star has a family of two planets.
- One is about two-thirds the size of Jupiter (orbiting in 535 days).
- The other is about three-quarters the size of Jupiter (orbiting in 834 days).
HD 220218: They saw a signal that looked like a planet, but after closer inspection, it turned out to be a "fake." It was actually the star itself acting up (stellar activity), not a planet.

The Secret Weapon: The "Noise-Canceling" Strategy

One of the coolest parts of this paper is how they solved the "wobbly star" problem for HD 127195.

Imagine trying to take a photo of a hummingbird while the camera is shaking. If you take one quick picture, it's blurry. But if you take many quick pictures in rapid succession and blend them together, the shake averages out, and the image becomes clear.

The team did exactly this with the star's "wobbles" (pulsations). Instead of taking one long observation, they took three short observations in one night, spaced out by about 20 minutes. By averaging these three snapshots, they successfully smoothed out the star's natural ripples. This allowed them to get a very clear signal, proving that the planets they found were real and not just the star shaking.

Why Does This Matter?

Rare Territory: These planets are in a "neighborhood" of the universe that hasn't been explored much. They orbit stars that are heavier than our Sun but not yet fully giant, and they take a long time to orbit (over a year). It's like finding a new type of fish in a deep, dark part of the ocean that no one has fished in before.
The Future of the Solar System: The paper also looked at what will happen to these planets in the distant future. As the stars continue to age and swell up, they will eventually eat the inner planets. However, the outer planets might survive and drift further away as the star loses weight, only to be swallowed later when the star becomes a giant red ball.

The Bottom Line

This paper is a success story of patience and clever technique. By listening carefully for nearly two decades and using a smart "averaging" trick to ignore the star's natural wobbles, the team confirmed the existence of three massive planets around two aging stars. They proved that even with "noisy" older stars, we can still find the quiet rhythm of a planet's orbit.

Technical Summary: CORALIE Radial-Velocity Search for Companions around Evolved Stars (CASCADES). V. Three Planetary Companions and Achievable Precision

Problem Statement
The search for exoplanets has historically focused on low-mass main-sequence stars, leaving the population of planets around more massive stars ( $>1.1 M_\odot$ ) poorly characterized. While approximately 20% of known exoplanets orbit such stars, the study of planets around evolved stars (subgiants and giants) presents specific challenges. Massive stars often exhibit higher effective temperatures, reducing the number of photospheric lines available for precise radial velocity (RV) measurements, and higher rotational velocities that broaden spectral lines. Furthermore, evolved stars exhibit intrinsic RV variability ("jitter") caused by pulsations and granulation, which can mask planetary signals or mimic them. The CASCADES survey aims to expand the planetary census around these evolved hosts, specifically targeting low-luminosity red giants which exhibit comparatively lower RV jitter than their higher-luminosity counterparts.

Methodology
The study analyzes high-precision RV data acquired over 15–18 years using the CORALIE spectrograph on the 1.2 m Swiss Euler telescope at La Silla. The targets are three evolved stars: HD 125136, HD 127195, and HD 220218.

Stellar Characterization: Stellar parameters were derived using a combination of Gaia DR3 data and the ARIADNE package (Bayesian inference on spectral energy distributions). The targets are classified as low-luminosity giant stars based on their position in the Hertzsprung–Russell and Kiel diagrams.
Data Reduction: CORALIE data were reduced using an automatic data reduction software (DRS) with different versions corresponding to instrument upgrades (COR98, COR07, COR14, COR24). RVs were calculated via the cross-correlation function method using a K5 template mask. Activity indicators (FWHM, contrast, bisector inverse slope, and H $\alpha$ ) were monitored to distinguish planetary signals from stellar activity.
Signal Analysis: The authors employed the Bayesian inference tool kima to model the RV time series. This approach allows for the simultaneous estimation of planetary orbital parameters, systemic velocities, instrument-specific offsets, and jitter terms. The number of planets ( $N_p$ ) was treated as a free parameter, with model selection based on Bayes factors (threshold of 150). Dynamical stability was enforced using the angular momentum deficit criterion.
Pulsation Mitigation: A dedicated observing campaign was conducted on HD 127195 to test a strategy for averaging out stellar pulsations. This involved multiple intra-night integrations (three 300 s exposures separated by 1200–1500 s) to reduce the impact of pulsation-induced noise.

Key Results

Planetary Detections:
- HD 125136: A single planetary companion is detected with a minimum mass ( $M \sin i$ ) of $2.26 \pm 0.08 M_{Jup}$ and an orbital period of $850 \pm 5$ days. The signal is well-modeled by a Keplerian curve with low eccentricity ( $e \approx 0.05$ ).
- HD 127195: A two-planet system is identified. Planet b has a minimum mass of $0.67 \pm 0.02 M_{Jup}$ and a period of $535 \pm 3$ days. Planet c has a minimum mass of $0.65 \pm 0.02 M_{Jup}$ and a period of $837 \pm 10$ days. Both orbits are nearly circular.
- HD 220218: A signal with a period of $\sim 1035$ days was detected but is interpreted as spurious. The signal shows correlations between semi-amplitude and other orbital parameters, and the required jitter term ( $\sim 26$ m s $^{-1}$ ) is significantly higher than typical for CORALIE observations of giants. The authors attribute this to stellar activity rather than a planetary companion.
Activity vs. Planets:
- For HD 125136 and HD 127195, no significant correlations were found between the RVs and activity indicators (FWHM, BIS, contrast, H $\alpha$ ). The activity indicators themselves showed no significant periodicity at the detected planetary periods.
- The authors note that while long-term magnetic cycles (periods $\sim 500$ –$1000$ days) cannot be fully ruled out as false positives, the lack of correlation with activity indicators and the stability of the Keplerian fits support a planetary origin.
Achievable Precision:
- The dedicated observing campaign on HD 127195 successfully demonstrated that stellar pulsations in low-luminosity giants can be averaged out. By using multiple short exposures per night, the nightly scatter was reduced to a median of 4 m s $^{-1}$ , with a total residual RMS of 5 m s $^{-1}$ after modeling. This aligns with theoretical predictions that pulsation contributions can be reduced to $\sim 3$ m s $^{-1}$ under these conditions.
Orbital Evolution:
- Simulations of the planetary systems' evolution through the star's future asymptotic giant branch (AGB) phase suggest that while the planets will survive the red giant branch (RGB) phase, the inner planets of both systems (HD 125136 b and HD 127195 b) will likely be engulfed by the star during the AGB phase due to tidal interactions. HD 127195 c, being less massive, may survive to the white dwarf phase.

Significance and Claims
The paper claims to detect three massive planets around two low-luminosity red giant stars, filling a gap in the parameter space regarding stellar mass and planetary orbital periods. These systems are located in a region of the period distribution ( $\sim 500$ –$900$ days) where occurrence rates for giant planets are reported to be high, yet the sample remains sparsely populated due to observational difficulties.

The study highlights that:

Low-luminosity giants are viable targets for RV surveys, provided that pulsation noise is managed.
The specific observing strategy of intra-night averaging can effectively mitigate pulsation noise, achieving precisions below 5 m s $^{-1}$ , a threshold rarely reached for giant stars.
The detected planets provide crucial constraints on the formation and migration efficiency of planets around intermediate-mass stars, bridging the demographic gap between solar-type stars and higher-mass hosts.

The authors remain modest regarding the definitive nature of the detections, acknowledging that long-term magnetic cycles could theoretically mimic the signals, though current evidence favors a planetary interpretation. The work underscores the necessity of long-term, stable RV monitoring to characterize the planetary populations around evolved stars.