The kinetic power of jets magnetically accelerated from advection dominated accretion flows in radio galaxies

TL;DR

This paper models the relation between bolometric luminosity and jet kinetic power in low luminosity AGNs using magnetically driven ADAFs, successfully reproducing observed correlations and providing insights into accretion and feedback processes.

Contribution

It introduces a numerical model of magnetically driven ADAFs that explains the observed kinetic power and luminosity relation in low luminosity AGNs, highlighting wind escape and feedback efficiency.

Findings

The model reproduces the observed L_kin/L_Edd ~ (L_bol/L_Edd)^0.49 relation.

Over 60% of accreted gas escapes as wind before reaching the black hole.

Mechanical efficiency is estimated between 10^-2 and 10^-3.

Abstract

There is a significant nonlinear correlation between the Eddington ratio (L_bol=L_Edd) and the Eddington-scaled kinetic power (L_kin=L_Edd) of jets in low luminosity active galactic nuclei (AGNs) (Merloni & Heinz). It is believed that these low luminosity AGNs contain advection dominated accretion flows (ADAFs). We adopt the ADAF model developed by Li & Cao, in which the global dynamics of ADAFs with magnetically driven outflows is derived numerically, to explore the relation between bolometric luminosity and kinetic power of jets. We find that the observed relation, L_kin/L_Edd ~ (L_bol=L_Edd)^0.49, can be well reproduced by the model calculations with reasonable parameters for ADAFs with magnetically driven outflows. Our model calculations is always consistent with the slope of the correlation independent of the values of the parameters adopted. Compared with the observations, our results show that over 60% of the accreted gas at the outer radius escapes from the accretion disc in a wind before the gas falls into the black holes. The observed correlation between Eddington-scaled kinetic power and Bondi power can also be qualitatively reproduced by our model calculations. Our results show that the mechanical efficiency varies from 10^-2 ~ 10^-3, which is roughly consistent with that required in AGN feedback simulations.