Activity clusters list of lines isolated

However, the PCA is highly sensitive to contamination from telluric residuals in the spectra. Therefore, we employ a filtering technique based on machine learning tools such as unsupervised Dimensional Reduction (DR) algorithm and Support Vector Machine (SVM) to remove faulty lines. We assess the performances of this filtering method using a simulation of observations of the per-line $dLW$ variations before applying it on near-infrared (nIR) high resolution spectroscopic observations using SPIRou at the Canada-France-Hawaii Telescope (CFHT) on five targets, with various levels of stellar magnetic activity, spectral types and rotation periods, from the SPIRou Legacy Survey, namely AU Mic, EV Lac, GJ1286, GJ1289 and Gl 410.

The filtered W_1_ signal is modulated with a period consistent with the rotation period retrieved from activity indicators, corresponding to the magnetic activity for all the stars studied. It also correlates with the small- scale magnetic field for all the five stars, with a direct Pearson correlation coefficient greater than 0.80. Additionally, we identified 201 stellar lines that are particularly sensitive to magnetic activity that could be valuable for studying of magnetic fields.