Mid-infrared interferometry of Seyfert galaxies: Challenging the Standard Model

TL;DR

This study uses high-resolution mid-infrared interferometry to test if the standard torus model can explain the diversity of Seyfert galaxy observations, finding that multiple configurations are necessary.

Contribution

It demonstrates that a single torus model cannot account for all Seyfert galaxy MIR data, suggesting at least two distinct dust configurations are needed.

Findings

Different Seyfert types require different torus models.

High and low filling factor models better explain different subsets.

A continuous transition between types cannot be ruled out.

Abstract

Aims: We aim to find torus models that explain the observed high-resolution mid-infrared (MIR) measurements of active galactic nuclei (AGN). Our goal is to determine the general properties of the circumnuclear dusty environments. Methods: We used the MIR interferometric data of a sample of AGNs provided by the instrument MIDI/VLTI and followed a statistical approach to compare the observed distribution of the interferometric data with the distributions computed from clumpy torus models. We mainly tested whether the diversity of Seyfert galaxies can be described using the Standard Model idea. In addition to line-of-sight (LOS) effects, we performed different realizations of the same model to include possible variations that are caused by the stochastic nature of the dusty models. Results: We find that our entire sample of AGNs, which contains both Seyfert types, cannot be explained merely by an inclination effect and by including random variations of the clouds. Instead, we find that each subset of Seyfert type can be explained by different models, where the filling factor at the inner radius seems to be the largest difference. For type I objects we find that about two thirds of our objects could also be described using a dusty torus similar to the type II objects. For the remaining third, it was not possible to find a good description using models with high filling factors, while we found good fits with models with low filling factors. Conclusions: Within our model assumptions, we did not find one single set of model parameters that could simultaneously explain the MIR data of all 21 AGN with LOS effects and random variations alone. We conclude that at least two distinct cloud configurations are required to model the differences in Seyfert galaxies, with volume-filling factors differing by a factor of about 5-10. A continuous transition between the two types cannot be excluded.