2021A&A...650A.154P

Aims. Modelling the low-ionisation lines (LILs) in active galactic nuclei (AGNs) still faces problems in explaining the observed equivalent widths (EWs) when realistic covering factors are used and the distance of the broad-line region (BLR) from the centre is assumed to be consistent with the reverberation mapping measurements. We re-emphasise this problem and suggest that the BLR 'sees' a different continuum from that seen by a distant observer. This change in the continuum reflected in the change in the net bolometric luminosity from the AGN is then able to resolve the above problem.
Methods. We carefully examine the optical FeII and near-infrared (NIR) CaII triplet (CaT) emission strengths with respect to Hβ emission using the photoionisation code CLOUDY and a range of physical parameters. Prominent among these parameters are (a) the ionisation parameter (U), (b) the local BLR cloud density (n_H), (c) the metal content in the BLR cloud, and (d) the cloud column density. Using an incident continuum for I Zw 1 -a prototypical Type-1 narrow-line Seyfert galaxy- our basic setup is able to recover the line ratios for the optical FeII (i.e. R_FeII) and for the NIR CaT (i.e. R_CaT) in agreement with the observed estimates. Nevertheless, the pairs of (U,n_H) that reproduce the conforming line ratios do not relate to agreeable line EWs. We therefore propose a way to mitigate this issue. The LIL region of the BLR cloud does not see the same continuum emitted by the accretion disc as that seen by a distant observer; rather it sees a filtered version of the original continuum which brings the radial sizes into agreement with the reverberation mapped estimates for the extension of the BLR. This is achieved by scaling the radial distance of the emitting regions from the central continuum source using the photoionisation method in correspondence with the reverberation mapping estimates for I Zw 1. Taking inspiration from past studies, we suggest that this collimation of the incident continuum can be explained by the anisotropic emission from the accretion disc, which modifies the spectral energy distribution such that the BLR receives a much cooler continuum with a reduced number of line-ionising photons, allowing reconciliation in the modelling with the line EWs.
Results. (1) The assumption of the filtered continuum as the source of BLR irradiation recovers realistic EWs for the LIL species, such as the Hβ, FeII, and CaT. However, our study finds that to account for the adequate R_FeII (FeII/Hβ flux ratio) emission, the BLR needs to be selectively overabundant in iron. On the other hand, the R_CaT (CaT/Hβ flux ratio) emission spans a broader range from solar to super-solar metallicities. In all these models, the BLR cloud density is found to be consistent with our conclusions from prior studies, that is, n_H∼10¹²cm^–3 is required for the sufficient emission of FeII and CaT. (2) We extend our modelling to test and confirm the co-dependence between metallicity and cloud column density for these two ionic species (FeII and CaT), further allowing us to constrain the physical parameter space for the emission of these LILs. Adopting the estimates from line ratios that diagnose the metallicity in these gas-rich media -which suggest super-solar values (≥5-10Z_☉)-, we arrive at cloud columns that are of the order of 10²⁴ cm^–2. (3) Finally, we test the effect of inclusion of a micro-turbulent velocity within the BLR cloud and find that the FeII emission is positively affected. An interesting result obtained here is the reduction in the value of the metallicity by up to a factor of ten for the R_FeII cases when the microturbulence is invoked, suggesting that microturbulence can act as an apparent metallicity controller for the FeII. On the contrary, the R_CaT cases are relatively unaffected by the inclusion of microturbulence.