The WISSH Quasars Project IV. BLR versus kpc-scale winds

TL;DR

This study investigates winds in hyper-luminous quasars at redshift 2-4, revealing two distinct populations with different outflow properties and suggesting orientation and ionization effects influence observed wind features.

Contribution

It provides the first comprehensive analysis of BLR and NLR winds in WISSH quasars, highlighting a dichotomy linked to orientation and ionization, and supports radiatively driven wind models.

Findings

Two sub-populations with different outflow characteristics identified.

Correlation between bolometric luminosity and wind velocity established.

Evidence supports radiatively driven winds requiring strong UV and weak X-ray emission.

Abstract

We have undertaken a multi-band observing program aimed at obtaining a complete census of winds in a sample of WISE/SDSS selected hyper-luminous (WISSH) QSOs at z~2-4. We have analyzed the rest-frame optical (LBT/LUCI and VLT/SINFONI) and UV (SDSS) spectra of 18 randomly selected WISSH QSOs to measure the SMBH mass and study the properties of winds both in the NLR and BLR traced by blueshifted/skewed [OIII] and CIV emission lines, respectively. These WISSH QSOs are powered by SMBH with masses $\ge$10$^9$ Msun accreting at 0.4<$\lambda_{Edd}$<3.1. We have found the existence of two sub-populations characterized by the presence of outflows at different distances from the SMBH. One population ([OIII] sources) exhibits powerful [OIII] outflows, rest-frame EW (REW) of the CIV emission REW$_{CIV}\approx$20-40 A and modest CIV velocity shift (v$_{CIV}^{peak}$) with respect to the systemic redshift (<=2000 km/s). The second population (Weak [OIII] sources), representing ~70% of the analyzed WISSH QSOs, shows weak/absent [OIII] emission and an extremely large v$_{CIV}^{peak}$ (up to ~8000 km/s and REW$_{CIV}$<=20 A). We propose two explanations for the observed behavior of the strength of the [OIII] emission in terms of orientation effects of the line of sight and ionization cone. The dichotomy in the presence of BLR and NLR winds could be likely due to inclination effects considering a polar geometry scenario for the BLR winds. We find a strong correlation with L$_{Bol}$ and an anti-correlation with $\alpha_{ox}$, whereby the higher L$_{Bol}$, the steeper $\alpha_{ox}$ and the larger is the v$_{CIV}^{peak}$. Finally, the observed dependence v$_{CIV}^{peak}\propto L_{Bol}^{0.28\pm0.04}$ is consistent with radiatively driven winds scenario, where strong UV continuum is necessary to launch the wind and a weakness of the X-ray emission is fundamental to prevent overionization of the wind itself.