Supermassive Black Hole Masses with multi-object Reverberation Mapping

Abstract

Reverberation mapping uses light echoes to infer the masses of supermassive black holes (SMBH) in active galactic nuclei (AGN). By effectively substituting for spatial resolution with temporal resolution, this technique overcomes the limits faced by other methods reliant on high angular resolution instruments, allowing the study of SMBH evolution over cosmic history. Reverberation measurements have been used to constrain the relationship between the size of the broad-line region and luminosity of AGN. This Radius-Luminosity ($R-L$) relation is used to estimate single-epoch virial black hole masses, and has been proposed for use to standardise AGN to determine cosmological distances. However, the intensity of observational resources required has impeded greater application.

This has necessitated the commencement of multi-object spectroscopic surveys, monitoring an order of magnitude more AGN and extending to high redshifts. In this thesis, I explore the intricacies of performing reverberation mapping on this `industrial-scale', and deliver results from the six-year Australian Dark Energy Survey (OzDES) Reverberation Mapping Program; one of two multi-object surveys completed to-date.

There are complex interactions between the window function, emission-line lag and variability timescales, which are further complicated by time dilation with redshift. My comprehensive simulations demonstrate the impact of each component of the global observational window function on reverberation lag recovery for a diverse population of AGN. I find the presence of a significant seasonal gap dominates the lag recovery efficacy of any given campaign strategy; more so than the cadence or baseline of the data.

With an appreciation of the challenges imposed by the limited spectroscopic cadence and significant seasonal gaps in the OzDES data, I successfully recover reverberation lags for eight AGN at $0.12<z< 0.71$ from the OzDES H$\beta$ sample, probing higher redshifts than the bulk of H$\beta$ measurements made to date. The results from this multi-object spectroscopic survey are consistent with previous measurements made by dedicated source-by-source campaigns (which had significantly higher cadence), and with the observed dependence on accretion rate.

Given the overall lag recovery efficacy of only $\sim$10\% for the entire OzDES sample, I use the novel stacking technique to extract the maximum information from our data. The data sparsity is ameliorated through the stacking of cross correlations from physically similar sources to recover average reverberation lags. I present average lags recovered for the H$\beta$, Mg\textsc{ii} and C\textsc{iv} samples, and multi-line measurements for redshift bins where two lines are accessible. Our results demonstrate that stacking has the potential to improve upon constraints on the $R-L$ relation, which have been derived using only individual source measurements thus far.

This thesis reveals both the challenges and the potential of industrial-scale reverberation mapping. With the imminence of the next generation of multi-object surveys, including LSST, TiDES and SDSS-V, this work highlights the urgency to prioritise observing fields with longer seasonal visibility---not only to improve lag recovery for more sources, but also to mitigate signal aliasing. Although I demonstrate the ability of stacking to partially compensate for data sparsity, only by optimising the observational window function can we harness the full potential of reverberation mapping. Show more