Clustering of high-redshift quasars

John D. Timlin III

doi:10.17918/etd-7828

In this work, we investigate the clustering of faint quasars in the early Universe and use the clustering strength to gain a better understanding of quasar feedback mechanisms and the growth of central supermassive black holes at early times in the history of the Universe. It has long been understood (e.g., Hopkins et al. 2007a) that the clustering of distant quasars can be used as a probe of different feedback models; however, until now, there was no sample of faint, high-redshift quasars with sufficient density to accurately measure the clustering strength. Therefore we conducted a new survey to increase the number density of these objects. Here, we describe the Spitzer -IRAC Equatorial Survey (SpIES) which is a moderately deep, large-area Spitzer survey which was designed to discover faint, high-redshift (2.9 [less than or equal to] z [less than or equal to] 5.1) quasars. SpIES spans ~115 deg^2 in the equatorial "Stripe 82" region of the Sloan Digital Sky Survey (SDSS) and probes to 5-[sigma] depths of 6.13 [mu]-Jy (21.93 AB magnitude) and 5.75 [mu]-Jy (22.0 AB magnitude) at 3.6 and 4.5 microns. At these depths, SpIES is able to observe faint quasars, and we show that SpIES recovers ~ 94% of the high-redshift (z > 3.5), spectroscopically-confirmed quasars that lie within its footprint. SpIES is also ideally located on Stripe 82 for two reasons: It surrounds existing infrared data from the Spitzer-HETDEX Exploratory Large-area (SHELA) survey which increases the area of infrared coverage, and there is a wide range of multi-wavelength, multi-epoch ancillary data on Stripe 82 which we can use together to select high-redshift quasar candidates. To photometrically identify quasar candidates, we combined the optical data from the Sloan Digital Sky Survey and the infrared data from SpIES and SHELA and employed three machine learning algorithms. These algorithms were trained on the optical/infrared colors of known, high-redshift quasars. Using this method, we generate a sample of 1378 objects that are both faint (i > 20.2) and high-redshift (2.9 [less than or equal to] z [less than or equal to] 5.1) which we use to compute the angular two-point correlation function. We fit a single power-law model with an index of delta = 1.39 ± 0.618 and amplitude of theta_0 = 0.71 ± 0.546 arcmin to the correlation function, as well as a dark matter model with a bias of b = 6.78 ± 1.79. The bias in our investigation suggests a model of quasar feedback that considers quasar activity as an intermittent phase in galaxy evolution. If this model is correct, quasar feedback is strong enough to periodically halt the accretion of gas onto the central supermassive black hole of the quasar, which shuts down quasar activity and causes the black hole to stop growing, however it is not strong enough to completely shut down the quasar in the early Universe.

Clustering of high-redshift quasars

Files and links (1)

Abstract

Metrics

Details

Clustering of high-redshift quasars

Files and links (1)

Abstract

Metrics

Details

Drexel University Social media