MSc defence by Sofie Helene Bruun
Supervisor: Jens Hjorth
External examiner: Hans Kjeldsen
Title: Variability classification of PTF lightcurves
Abstract: Determining the nature of astrophysical sources can be done in many ways, and one of them is by the study of variations in luminosity. This thesis analyses how variability of lightcurves can be used for classification of sources. This could be useful in e.g. studying quasars or cataloguing standard stars for improved photometric calibrations. Four samples of lightcurves from the Palomar Transient Facility (PTF) are used: All sources in 3 patches covering 10 × 10 deg2 each, quasars spectroscopically confirmed by SDSS, quasar candidates selected from WISE colours and known lensed quasars. Their magnitude changes over different time scales are fitted to a power law with an affine-invariant Markov chain Monte Carlo ensemble sampler. The fit parameters then describe their variabilities and can be used for classification. A suggested cut in variability for quasar selection in SDSS lightcurves is tested on the PTF lightcurves, and while it does select more sources in the quasar samples than in a random sample, it still only classifies 14 % of the spectroscopically confirmed quasars correctly – but 51 % out of those consistent to two σ with variabilities following a power law with non-zero exponent and amplitude. So while variability classification in PTF does have potential, optimisation of methods for data selection and utilisation of the variability information is needed.