Estimating black hole spin from AGN SED fitting: the impact of general-relativistic ray tracing

Accretion disc model fitting to optical/UV quasar spectra requires that the highest mass black holes have the highest spin, with implications on the hierarchical growth of supermassive black holes and their host galaxies over cosmic time. However, these accretion disc models did not include the effects of relativistic ray tracing. Here, we show that gravitational redshift cancels out most of the increase in temperature and luminosity from the smaller radii characteristic of high spin. Disc models which include the self-consistent general relativistic ray tracing do not fit the UV spectra of the most massive quasars (log M/M_⊙ ≥ 9.5), most likely showing that the disc structure is very different to that assumed. We extend the relativistic ray tracing on more complex disc models, where the emission is not limited to (colour-temperature-corrected) blackbody radiation but can instead be emitted as warm and hot Comptonization. We demonstrate this on the broad-band (UV/X-ray) spectrum of Fairall 9, a local intensively monitored 'bare' active galactic nucleus (no significant intrinsic cold or warm absorption). We show that including relativistic corrections does make a difference even to these more complex models, but caution that the inferred black hole spin depends on the assumed nature and geometry of the accretion flow. Additionally, we make our model code publicly available, and name it RELAGN.