A Comparative Study of TESS Light Curve Extraction Methods Applied to Blazars

Blazars are characterized by largely aperiodic variability on timescales ranging from minutes to decades across the electromagnetic spectrum. The TESS (Transiting Exoplanet Survey Satellite) mission provides continuous sampling of blazar variability on timescales ranging from tens of minutes to 27 days for a single sector observation. Proper removal of the background, thermal ramping, and onboard systematic effects are crucial to the extraction of a reliable blazar light curve. Multiple publicly available procedures have been created to correct for these effects. Using ground based observations from the Zwicky Transit Facility (ZTF) and the Asteroid Terrestrial-impact Last Alert System (ATLAS) as “ground truth” observations, we compare 6 different methods (Regression, Cotrending Basis Vectors (CBV), Pixel Level Decorrelation (PLD), eleanor, quaver, and simple differential photometry (SDP)) to each other, and to our “ground truth” observations, to identify which methods properly correct light curves of a sample of 11 bright blazars, including the suspected neutrino source TXS 0506+056. In addition to comparing the resulting light curves, we compare the slopes of the power spectral densities, perform least-square fitting to simultaneous ZTF data, and quantify other statistical qualities generated from the light curves of each method. We find that only three of the six methods compared (Simple Differential Photometry, eleanor, and quaver) produce TESS light curves consistent with the ground-based ZTF and ATLAS observations.