16. Automated removal of outlier observations
Raw light curves often contain imperfections resulting from atmospheric conditions (clouds entering the imaging field), equipment issues (telescope mount tracking problem, image focusing failure) and so on. The result of this are so called outlier observations, which should be removed from the light curve before starting any further analysis.
In this tutorial, we will learn how to let Peranso search for outlier observations in a light curve and how to automatically deactivate them. As an illustration, we will use a simple light curve of T Aur, obtained on 2021, Jan 18th by T. Vanmunster, observing from CBA Extremadura Observatory in Spain. The automated outlier removal is particularly efficient when working with large and/or complex light curves, but for this tutorial we prefer to work with a simple example.
T Aur or Nova Aurigae 1891 was discovered in 1891 by the Scottish amateur astronomer Thomas Anderson. It reached mag 4.1 and was the first nova to be observed spectroscopically. We now know that it an eclipsing binary, with a period of 4.9 hours, and an eclipse depth of 0.18 magnitudes.