Detecting Planets Around Very Low Mass Stars with the Radial Velocity Method

The detection of planets around very low-mass stars with the radial velocity method is hampered by the fact that these stars are very faint at optical wavelengths where the most high-precision spectrometers operate. We investigate the precision that can be achieved in radial velocity measurements of low mass stars in the near infrared (nIR) Y-, J-, and H-bands, and we compare it to the precision achievable in the optical. For early-M stars, radial velocity measurements in the nIR offer no or only marginal advantage in comparison to optical measurements. Although they emit more flux in the nIR, the richness of spectral features in the optical outweighs the flux difference. We find that nIR measurement can be as precise than optical measurements in stars of spectral type ~M4, and from there the nIR gains in precision towards cooler objects. We studied potential calibration strategies in the nIR finding that a stable spectrograph with a ThAr calibration can offer enough wavelength stability for m/s precision. Furthermore, we simulate the wavelength-dependent influence of activity (cool spots) on radial velocity measurements from optical to nIR wavelengths. Our spot simulations reveal that the radial velocity jitter does not decrease as dramatically towards longer wavelengths as often thought. The jitter strongly depends on the details of the spots, i.e., on spot temperature and the spectral appearance of the spot. Forthcoming nIR spectrographs will allow the search for planets with a particular advantage in mid- and late-M stars. Activity will remain an issue, but simultaneous observations at optical and nIR wavelengths can provide strong constraints on spot properties in active stars.

💡 Research Summary

The paper addresses the challenge of detecting planets around very low‑mass stars—particularly late‑type M dwarfs—using the radial‑velocity (RV) technique. Optical spectrographs achieve the highest RV precision, but the faintness of these stars at visible wavelengths limits the signal‑to‑noise ratio (S/N) and thus the achievable precision. The authors evaluate whether moving to the near‑infrared (nIR) Y, J, and H bands can overcome this limitation, and they compare the attainable RV precision in the nIR with that in the optical for a range of M‑spectral types.

First, synthetic spectra for M0–M9 stars were generated and realistic observing conditions (telescope aperture, exposure time, detector efficiency) were applied to compute S/N in each band. While early‑M stars (M0–M3) still benefit more from optical observations because of the high density of deep atomic lines, the flux advantage of the nIR becomes decisive from about spectral type M4 onward. For M6–M9 stars the nIR flux exceeds the optical by a factor of five or more, allowing comparable or better RV precision with the same exposure time.

Second, the authors quantify the RV information content of each band. The optical region is rich in Fe I, Ca I, and other atomic lines, providing strong cross‑correlation peaks. In the nIR the spectrum is dominated by molecular bands (H₂O, CO, TiO). Although the line density is lower, the lines are broad and deep enough that, for late‑M dwarfs, the cumulative RV information surpasses that of the optical. The transition point where nIR becomes superior is found near M4.

Third, the paper investigates wavelength calibration strategies for nIR spectrographs. Although ThAr lamps have fewer lines in the nIR, the authors demonstrate that the existing lines are sufficiently bright and stable. Simulations of a ThAr‑calibrated, thermally and pressure‑stabilized spectrograph show that sub‑meter‑per‑second wavelength stability can be achieved without resorting to more expensive laser‑frequency combs. This makes ThAr a viable, cost‑effective solution for nIR RV work.

Fourth, the impact of stellar activity—specifically cool spots—on RV measurements is modeled across wavelengths. Traditional wisdom suggests that spot‑induced jitter diminishes dramatically toward longer wavelengths because the contrast between spot and photosphere drops. The authors’ spot simulations, however, reveal a more nuanced picture. The jitter amplitude depends strongly on spot temperature contrast, filling factor, and the spectral characteristics of the spot (e.g., molecular band strengths). For very cool spots (ΔT > 1000 K) the reduction in jitter in the nIR is modest, while for warmer, smaller spots the jitter can even be larger in the nIR than in the optical. Consequently, the assumption that nIR observations automatically mitigate activity noise is not universally valid.

The paper concludes with practical recommendations: (1) For early‑M dwarfs (M0–M3) optical RV remains optimal; nIR offers at best marginal gains. (2) From M4 onward, especially for mid‑ and late‑M dwarfs (M6–M9), nIR RV measurements can match or exceed optical precision, leveraging the higher stellar flux and the molecular line forest. (3) A stable spectrograph calibrated with a ThAr lamp can deliver the required m s⁻¹ stability in the nIR, providing a cost‑effective path forward. (4) Stellar activity will continue to limit RV precision; simultaneous optical and nIR observations are essential to disentangle planetary signals from spot‑induced jitter and to constrain spot properties.

Overall, the study provides a thorough quantitative framework for the design and operation of upcoming nIR high‑resolution spectrographs (e.g., CARMENES, SPIRou, NIRPS) and underscores their particular advantage in the search for Earth‑mass planets around the coolest M dwarfs, while also highlighting the persistent challenge of stellar activity across all wavelengths.