Stellar-activity radial-velocity noise at 10 cm/s: granulation and convective blueshift for Earth twins

The Extreme Precision Radial Velocity Working Group's final report (arXiv:2107.14291) identifies stellar variability, not instrument stability, as the main barrier at 10 cm/s, and recommends Sun-as-a-star observations as a testbed. Gaussian-process frameworks tied to activity indicators (Rajpaul et al., arXiv:1506.07304) are standard tools. Solar-telescope observations with HARPS-N, NEID and similar instruments have shown that convective-blueshift suppression can be partly modelled but a residual of 0.5–1 m/s on rotation and magnetic-cycle timescales remains for the Sun. No Earth-mass planet in the habitable zone of a Sun-like star has been detected by radial velocity. Combining Sun-as-a-star data from HARPS, HARPS-N, EXPRES and NEID, the Extreme Stellar-Signals Project found instrument-to-instrument residuals of about 15–30 cm/s on intraday timescales (Zhao et al., arXiv:2309.03762), so even the instrumental floor is not yet at 10 cm/s. The ten-year Terra Hunting Experiment with HARPS3 on the Isaac Newton Telescope (simulations: arXiv:1806.00518) suffered a cryostat failure late in commissioning in June 2026 and is now expected to start in 2027, according to the project site.

Detecting Earth twins is required to determine the occurrence rate η⊕ that sets the target lists and designs of future direct-imaging missions, and the activity-filtering methods apply to all precision RV science. No other application.

Resolved when a Sun-like star shows a habitable-zone Earth-mass planet at ≥5σ with a signal confirmed in at least two independent instruments and an analysis demonstrated, by injection–recovery tests on solar data, to achieve a residual RV scatter ≤30 cm/s on orbital timescales.