The Size-Luminosity Relationship of Quasar Narrow-Line Regions

TL;DR

This study models the size-luminosity relationship of quasar narrow-line regions, explaining observed sizes and saturation effects through cloud properties and ionization physics, and estimates the obscuring torus covering factor.

Contribution

It provides a parameter-free model of NLR sizes based on cloud properties and explains size saturation and the torus covering factor using ionization physics.

Findings

Model reproduces NLR sizes without free parameters for high-luminosity quasars.

Size saturation at ~10 kpc explained by optically thick clouds.

Estimated torus covering factor is approximately 0.5.

Abstract

The presence of an active galactic nucleus (AGN) can strongly affect its host. Due to the copious radiative power of the nucleus, the effects of radiative feedback can be detected over the entire host galaxy and sometimes well into the intergalactic space. In this paper we model the observed size-luminosity relationship of the narrow-line regions (NLRs) of AGN. We model the NLR as a collection of clouds in pressure equilibrium with the ionizing radiation, with each cloud producing line emission calculated by Cloudy. The sizes of the NLRs of powerful quasars are reproduced without any free parameters, as long as they contain massive ($10^5 - 10^7 M_\odot$) ionization-bounded clouds. At lower AGN luminosities the observed sizes are larger than the model sizes, likely due to additional unmodeled sources of ionization (e.g., star formation). We find that the observed saturation of sizes at $\sim 10$ kpc which is observed at high AGN luminosities ($L_\text{ion} \simeq 10^{46}$ erg/s) is naturally explained by optically thick clouds absorbing the ionizing radiation and preventing illumination beyond a critical distance. Using our models in combination with observations of the [O III]/IR ratio and the [O III] size -- IR luminosity relationship, we calculate the covering factor of the obscuring torus (and therefore the type 2 fraction within the quasar population) to be $f=0.5$, though this is likely an upper bound. Finally, because the gas behind the ionization front is invisible in ionized gas transitions, emission-based NLR mass calculations underestimate the mass of the NLR and therefore of the energetics of ionized-gas winds.