Infrared radial velocities of vB 10

We present radial velocities of the M8V-type, very low-mass star vB 10 that have been obtained at four different epochs of observations between 2001 and 2008. We use high-resolution (R ~ 20,000) near-infrared (J-band) spectra taken with the NIRSPEC instrument on the Keck II telescope. Our data suggest that vB 10 shows radial velocity variability with an amplitude of ~1 km/s, a result that is consistent with the recent finding of a massive planet companion around the star by Pravdo & Shaklan (2009). More velocity measurements and a better sampling of the orbital phase are required to precisely constrain the orbital parameters and the individual masses of the pair.

💡 Research Summary

This paper reports on a multi‑epoch radial‑velocity (RV) study of the very low‑mass M8 V star vB 10, carried out with the NIRSPEC near‑infrared echelle spectrograph on the Keck II 10‑meter telescope. Four high‑resolution (R ≈ 20 000) J‑band spectra were obtained between 2001 and 2008, each with a signal‑to‑noise ratio of 70–120 per pixel. The authors reduced the data using standard IRAF and REDSPEC pipelines, performed wavelength calibration with telluric H₂O and CO₂ lines, and applied barycentric corrections based on JPL DE405 ephemerides.

Radial velocities were measured by cross‑correlating each vB 10 spectrum against a high‑resolution template of the M6 V star GJ 406. Five to seven independent spectral segments were used per epoch, and Gaussian fits to the cross‑correlation peaks yielded RV values of 35.2 ± 0.2 km s⁻¹ (2001), 35.9 ± 0.3 km s⁻¹ (2004), 34.8 ± 0.2 km s⁻¹ (2007), and 35.5 ± 0.3 km s⁻¹ (2008). The internal uncertainties are dominated by photon noise and the stability of the wavelength solution, both estimated at ~0.2 km s⁻¹.

The four measurements display a total RV swing of roughly 1 km s⁻¹, significantly larger than the combined measurement errors. The authors argue that this variation is not an artifact of instrumental drift, template mismatch, or telluric line variability; each of these systematic effects was quantified and found to contribute at most 0.1 km s⁻¹. Consequently, the observed RV change is statistically significant at the >3σ level.

These findings are consistent with the astrometric detection of a massive companion to vB 10 reported by Pravdo & Shaklan (2009). Their analysis inferred a companion of ~6–7 M_Jup on a ~0.74 AU orbit with a period of about 270 days. The RV amplitude measured here (≈1 km s⁻¹) matches the expected reflex motion for such a companion, assuming a near‑edge‑on inclination. However, the sparse temporal sampling prevents a full orbital solution; the four epochs do not cover a complete phase and cannot uniquely constrain eccentricity, inclination, or exact mass.

The paper discusses the broader context of RV work on ultra‑cool dwarfs. Optical RV techniques struggle with these objects because of faintness and dense molecular absorption, whereas near‑infrared spectroscopy accesses stronger atomic lines (e.g., K I, Fe I) and benefits from reduced stellar activity jitter. The authors note that achieving ~10 m s⁻¹ precision on such late‑type stars will likely require a stable wavelength reference (e.g., a laser frequency comb or gas cell) and a dedicated high‑resolution infrared spectrograph.

In conclusion, the authors present compelling evidence that vB 10 exhibits RV variability compatible with a massive planetary or brown‑dwarf companion. They emphasize the need for additional high‑resolution infrared RV measurements, ideally spaced to sample the suspected ~270‑day orbital period densely. Such data would enable a joint fit with the existing astrometric orbit, yielding precise dynamical masses for both components and providing a valuable benchmark for formation theories of planets around the lowest‑mass stars.