Selection of Quasar candidates based on multi-band photometric data

Abstract

The J-PAS survey of quasars and AGNs has the potential to become the largest, most complete survey of its kind, due to its narrow-band photometric system that can resolve the broad emission lines of quasars. With this BEPE project we will test our techniques (an optimized photo-z code and a machine-learning based spectrum classifier) and determine sets of criteria that would provide us complete and accurate quasar catalogs with J-PAS data, in order to unravel the potential of quasars as tracers of the large-scale structure. This task is even more critical now, as J-PAS is considering a joint effort with the multi-object spectrograph WEAVE that is based on the exploitation of quasar data. In fact, the leader of the WEAVE quasar efforts is Dr. Mat Pieri, the external advisor.Another critical issue, both for J-PAS and for the WEAVE-QSO surveys, is to determine which minimal set of filters could be employed in order to obtain an accuracy (both in the identification of targets, as well as in photometric redshifts) that matches a given requirement. We also plan to enhance our methods in such a way that they are able not only to find photometric redshifts for our quasar candidates, but also to fit for the line widths of the main broad emission lines, so that we can estimate physical properties for these objects. Finally, in addition to fine-tuning her own code, the student will check the results against other methods, in particular (1) a new code which has been developed by Dr. Pieri's group, and (2) the well-known photo-z code LePhare. The student is already proficient in LePhare, and the main developers of that code are also at LAM, which will provide a unique opportunity for the student to acquire greater depth in this field. In particular, with respect to LePhare, one of our goals is to adapt that code to run with the full J-PAS filter system - something that is not possible with the present public version of the code.