Results of the PCA for NRS1 (left three panels) and NRS2 (right three panels). We created 20 independent eigenvectors using the PCA functionality in scikit-learn. We show the results for a single spectroscopic channel (λ_cen = 4.143 ± 0.017). Top: black is the reduced spectroscopic light curve. Red is the light curve after it has been correlated against the four “best” eigenvectors, shown in the middle panel. Bottom: results of the Lomb–Scargle periodogram on the four eigenvectors in the middle plot.