Echo hunting: using reverberation mapping of active galactic nuclei to probe black hole growth and cosmic expansion

Abstract

Broad line active galactic nuclei (AGN) have been proposed as potential standardisable candles by Watson et al. (2011), using a technique called reverberation mapping. This thesis investigates whether AGN are useful high redshift standard candles and how to optimise the scientific output of the ongoing OzDES reverberation mapping survey. AGN are highly luminous, numerous, and are present over a huge range of cosmic history. Due to this, they can be used to probe unexplored regions in the Universe's expansion history and help constrain the properties of dark energy. To investigate the usefulness of AGN I consider whether a general high-redshift standard candle will be a powerful probe of dark energy, when compared to other types of planned dark energy measurements. I consider several models of how the dark energy equation of state, w(z), may evolve in time and assess the optimal redshift distribution that a high-redshift standard-candle survey could take to constrain these models. The more general the form of the dark energy equation of state w(z) being tested, the more useful high-redshift standard candles become. As part of the OzDES spectroscopic survey we are carrying out a large scale reverberation mapping study of >=500 AGN over five years in the 30 square degree area of the Dark Energy Survey (DES) supernova fields. These AGN have redshifts ranging up to 4 and have apparent AB magnitudes between 16.8 < r < 22.5 mag. The aim of the survey is to measure time lags between fluctuations in AGN continuum and broad emission line fluxes of individual objects, in order to measure black hole masses for a broad range of AGN and constrain the radius-luminosity (R − L) relationship, which makes them viable standard candles. I investigate the expected efficiency of the OzDES reverberation mapping campaign and its possible extensions. We expect to recover lags for ∼ 35−45% of the AGN. AGN with shorter lags and greater variability are more likely to yield a lag and objects with lags ≲ 6 month or ∼ 1 year are expected be recovered the most accurately. I show that the baseline OzDES reverberation mapping campaign should produce an unbiased measurement of the radius- luminosity relationship for the emission lines: Hβ, Mg ii, and C iv. However, enhancing the baseline survey by either increasing the spectroscopic cadence, extending the survey season, or improving the accuracy of the emission line flux measurement will significantly improve the R − L parameter constraints for all lines. For the first two years of OzDES we have observed 989 AGN but in the following three years the number of targets we can continue to monitor will be reduced, possibly down to a minimum of 500 AGN. We chose to prioritise quasars that are likely to recover a lag and/or enhance the scientific output of the survey. I quantify the quality of the emission line flux measurements using the signal-to-noise ratio of the flux measurement, and select the year 3 (2015) targets based on this value and their scientific merit. I also perform preliminary reverberation mapping analysis on a select sample of targets but the lag recovery was unsuccessful.