A new method for the correction of micro-tellurics effects in radial velocity data

We present a new method for the optimization of data reduction of near-infrared (nIR) radial velocity (RV) measurements.

Our work used data from the Spectro-Polarimètre Infra Rouge (SPIRou) on the Canada-France-Hawaii Telescope (CFHT). Nmerous telluric absorption lines are present in the nIR which creates an important challenge in high-precision RV measurements, as the barycentric velocity of the star varies through the year so do the telluric lines which leads to systematic RV offsets.

By taking advantage of the line-by-line (LBL) algorithm (Artigau et al., submitted), a recent tool for precision velocimetry, we were able to estimate the impact of micro-tellurics on the RV precision through Monte-Carlo simulations: the LBL enabled us to compute a RV timeseries for each lines and quantify how sensitive its RMS is to the addition of tellurics.

The lines that we consider too sensitive are consequently removed, we typically keep ~75-80% of the total spectral domain and even most of the lines deep in the tellurics bands are conserved. The main results are the removal of spurious yearly signals that we had in our data and a better final RV accuracy for our targets, decreasing their RMS from 0.5 to 1 m/s which is a non-negligible gain in order to detect small amplitude exoplanets.