Velocity-inverted three-dimensional distribution of the gas clouds in the Type 2 AGN NGC1068

TL;DR

This study derives a 3D distribution of gas clouds in NGC1068's AGN, revealing a hollow-cone outflow structure that aligns with previous models and observations, enhancing understanding of AGN gas dynamics.

Contribution

It introduces a method to directly derive 3D gas cloud distributions from velocity maps, applied here to NGC1068, revealing a hollow-cone outflow structure.

Findings

Inner narrow-line region has a hollow-cone structure.

The 3D distribution aligns with previous models.

The geometry suggests a connection to the AGN torus and molecular outflows.

Abstract

Spatially-resolved velocity maps at high resolutions of 1-10 pc are becoming available for many nearby AGNs in both optical/infrared atomic emission lines and sub-mm molecular lines. For the former, it has been known that a linear relationship appears to exist between the velocity of the ionized gas clouds and the distance from the nucleus in the inner ~100 pc region, where these clouds are outflowing. Here we demonstrate that, in such a case, we can actually derive the three-dimensional (3D) geometrical distribution of the clouds directly from the velocity map. Revisiting such a velocity map taken by HST for the prototypical Type 2 AGN NGC1068, we implement the visualization of the 3D distribution derived from the map, and show that this inner narrow-line region has indeed a hollow-cone structure, consistent with previous modeling results. Quite possibly, this is the outer extended part of the polar elongated dusty material seen in the recent mid-IR interferometry at pc scale. Conversely, the latter small-scale geometry is inferred to have a hollow-cone outflowing structure as the inward extension of the derived 3D distribution above. The AGN obscuring "torus" is argued to be the inner optically thick part of this hollow-cone outflow, and its shadowed side would probably be associated with the molecular outflow seen in certain sub-mm lines. We discuss the nature of the linear velocity field, which could be from an episodic acceleration that had occurred ~10^5 years ago.