Tracing the formation and merger-driven growth of massive black holes with the Zwicky Transient Facility

Thumbnail Image


Publication or External Link





The dawning of low--frequency gravitational wave (GW) astronomy via pulsar timing arrays and space--based GW interferometry will provide new opportunities for the study of the supermassive black hole (SMBH) binaries which form as galaxies merge through cosmic time. The onset of observational GW studies has coincided with the expansion of wide--field optical time--domain surveys such as the Zwicky Transient Facility (ZTF), which provide a complementary way to detect and analyze SMBHs when they accrete gas and emit at optical wavelengths. In this thesis I describe how high cadence surveys like ZTF can be used to discover rare populations of massive black holes which inform our understanding of early massive black hole seeding channels and their subsequent growth through mergers to produce the SMBH populations we see today.

In the first part of this thesis I present a search for variable active galactic nuclei (AGN) which are spatially offset from their host galaxies using time-resolved imaging data from ZTF and deeper, higher resolution imaging data from the Legacy Surveys. I present a population of 52 variable AGN in merging galaxies in addition to 9 candidates for gravitational wave recoil of remnant SMBHs which may be used to constrain SMBH binary merger rates and spin alignment efficiencies. I also examine the dramatic rebrightening of a previous recoiling SMBH candidate SDSS1133, and conclude from spectroscopic follow--up that it is more likely an outbursting luminous blue variable star.

In the second part of the thesis, I present a population of 190 low--mass AGN in dwarf galaxies discovered by their optical or mid--infrared variability in deep ZTF difference imaging and forward--modeled photometry of {\it WISE} image stacks. These intermediate mass black hole (IMBH) candidates can be used to constrain the low--mass end of the $M_{BH}-\sigma_*$ relation and dwarf galaxy occupation fractions in order to better understand the origins of the first massive black holes. Only $9$ candidates from my search had been detected previously in radio, X-ray, and variability searches for dwarf galaxy AGN. I find that spectroscopic approaches to AGN identification would have missed 81% of my ZTF IMBH candidates and 69% of my {\it WISE} IMBH candidates, showing the promise of variability searches for discovery of otherwise hidden low--mass AGN.

In the third part of this work, I present 299 variable AGN in ZTF which have double--peaked Balmer broad lines from the motion of gas in their accretion disk, increasing the number of known double--peaked emitters (DPEs) by a factor of $\sim$2. DPEs can arise as false positive candidates in both spectroscopic and variability--based searches for SMBH binaries, so it is important to characterize the properties of their spectra and light curves. I find that 16% of variable broad line AGN in ZTF are DPEs and that $\sim$50% of the DPEs display dramatic changes in the relative fluxes of their red and blue peaks over long $10-20$ year timescales. I show that a number of DPEs exhibit apparently periodic and chirping signals in the optical and mid--infrared and discuss how this arises naturally from their power spectra. I show that DPE light curves have slightly steeper power spectra than their standard broad line counterparts and are $\sim$1.5 times more likely to have a low frequency turnover. I compare the variability and spectroscopic properties of the ZTF DPE population with the recently discovered inspiraling SMBH binary candidate SDSSJ1430+2303 (ZTF18aarippg) and conclude that the variable velocity--offset broad lines and periodic behavior of ZTF18aarippg are not unusual compared to other DPEs, and it is therefore more likely to be a single AGN rather than an SMBH binary.

I conclude this thesis by outlining how the transient detection and image forward--modeling techniques presented in this thesis can be used to find populations of low accretion rate, off--nuclear AGN with the upcoming Legacy Survey of Space and Time at the Vera Rubin observatory in order to produce much better constraints on massive black hole seeding channels and GW recoil rates. I also discuss how these techniques can be applied to new science cases, such as the analysis of strongly gravitationally lensed supernovae and quasars, for cosmological studies with LSST.