Active galactic nucleus outflows accelerate when they escape the bulge

TL;DR

The study investigates whether the rapid outer acceleration of AGN-driven outflows is a consequence of bulge clearing, by applying a 1D energy-driven wind model (Magnofit) to ten local AGN hosts with an isothermal bulge. The authors find that a simple wind-feedback scenario, characterized by two free bulge parameters ($f_g$ and $M_{ m bulge}$) and a constant $L_{ m AGN}$, can reproduce most observed radial velocity profiles, with acceleration beginning near the bulge boundary. The results generally place bulge masses within the SMBH–bulge relation and gas fractions within observationally plausible ranges, though a subset of cases shows faster-than-predicted outer acceleration likely due to cooling and star formation reducing outflow mass. These findings support an energy-driven wind propagation model and highlight testable predictions for future multi-phase, spatially resolved observations to validate the bulge-clearing mechanism and outflow evolution.

Abstract

Large-scale outflows driven by AGNs are an important element of galaxy evolution. Detailed analysis of their properties allows us to probe the activity history of the galactic nucleus and, potentially, other properties of the host galaxy. A recent paper presents detailed radial velocity profiles of outflows in ten AGN host galaxies and shows a common trend of approximately constant velocity in the centre followed by rapid acceleration outside $R_{\rm tr} \sim 1 - 3$ kpc. We show that this is a consequence of the AGN-driven outflows clearing the gaseous bulges of the host galaxies and beginning to expand into a region of negligible gas density. We used a 1D semi-analytical code to calculate outflow propagation in each of the ten galaxies, assuming a constant AGN luminosity and an isothermal bulge density profile, with a finite bulge radius, and leaving the gas fraction and total mass of the bulge as free parameters. We also considered the effect of different gas density profiles, variations in bulge velocity dispersion, AGN luminosity, and the effect of outflow fragmentation. Our simplest model can fit six outflow profiles essentially perfectly, while another can be fit if the bulge gas density profile is shallower than isothermal. A shallower density profile also improves the fit in the central regions of the remaining three outflows, but they accelerate faster than our models predict; this could be evidence of significant gas cooling and star formation that reduce the total mass of outflowing gas. We conclude that a simple AGN-driven wind feedback model can explain the detailed velocity profiles of real outflows in local AGN hosts. The free parameters of our model have values that fall well within reasonable ranges. This suggests that the simple scenario we envisioned is close to the true conditions governing the general trends of large-scale outflow expansion.

Figures (5)

• Figure 1: Observed (blue points) and modelled outflow velocity profiles for all ten galaxies. In each galaxy, all velocities are normalised to the mean observed outflow velocity in the central kiloparsec or within the modelled bulge radius, whichever is lower. Panels are arranged approximately from best to worst match. The black line and shaded region correspond to the $M_{\rm bulge}$ taken from the observed $M_{\rm BH}-M_{\rm bulge}$ relation and its $\pm 0.68$ dex scatter Schutte2019ApJ when the value of $f_{\rm g}$ is chosen to fit the velocity in the central region. The dashed red line is the result of choosing both $f_{\rm g}$ and the value of $M_{\rm bulge}$ to produce the best fit to the observed outflow profile. The vertical dotted red line shows the bulge radius given by the best-fit value of $M_{\rm bulge}$, which is also the point where the outflow begins to accelerate.
• Figure 2: Effect of varying the slope of the gas density distribution. Top panel: Observed outflow in NGC 1365 (blue points) compared with our fiducial model (dashed red line; same as in Fig. \ref{['fig:all_fit']}) and a model where the density profile is $\rho \propto R^{-1.55}$ (green dashed line). Other panels: Same as top panel but for NGC 2992, NGC 5643, and NGC 5728, from top to bottom. In these three cases, we chose the values of the free parameters to match the innermost observed point and the point with the lowest velocity (see the main text for our motivation). In each panel, we give the values of $M_{\rm bulge}$, $f_{\rm g}$, and the slope exponent, $\alpha,$ for each model.
• Figure 3: Black hole and bulge masses of our best-fit base models (red points), plotted against SMBH masses. Green points show the bulge masses in the fits with different bulge density profile slopes. The grey line shows the observed $M_{\rm BH} - M_{\rm bulge}$ relation from Schutte2019ApJ, with dashed lines representing the $\pm 0.68$ dex scatter. The blue line shows the equivalent relation from McConnell2013ApJ, with dashed lines representing the $\pm 0.34$ dex scatter.
• Figure 4: Effect of varying $L_{\rm AGN}$ on outflow velocity profiles, while keeping all other parameters fixed. Blue points are observational data from M25 and the dashed red line is our best-fit model (as in Fig. \ref{['fig:all_fit']}) using $L_{\rm AGN} = 0.5 L_{\rm Edd}$. The shaded region represents the effect of changing the AGN luminosity by a factor of 4, between $0.25 L_{\rm Edd}$ and $L_{\rm Edd}$.
• Figure 5: Same as Fig. \ref{['fig:varlagn']} but showing the effect of varying the bulge mass (grey shading) and velocity dispersion (purple shading). The bulge mass varies by a factor of 4 around the fiducial value, and the velocity dispersion by a factor of 2.