Misaligned Discs as Obscurers in Active Galaxies

We review critically the evidence concerning the fraction of Active Galactic Nuclei (AGN) which appear as Type 2 AGN, carefully distinguishing strict Type 2 AGN from both more lightly reddened .Type 1 AGN, and from low excitation narrow line AGN, which may represent a different mode of activity. Low excitation AGN occur predominantly at low luminosities; after removing these, true Type 2 AGN represent 58% of all AGN, and lightly reddened Type 1 AGN a further ~15%. Radio, IR, and volume-limited samples all agree in showing no change of Type 2 fraction with luminosity. X-ray samples do show a change with luminosity; we discuss possible reasons for this discrepancy. We test a very simple picture which produces this Type 2 fraction with minimal assumptions. In this picture, infall from large scales occurs in random directions, but must eventually align with the inner accretion flow, producing a severely warped disk on parsec scales. If the re-alignment is dominated by tilt, with minimal twist, a wide range of covering factors is predicted in individual objects, but with an expected mean fraction of Type 2 AGN of exactly 50%. This ’tilted disc’ picture predicts reasonable alignment of observed nuclear structures on average, but with distinct misalignments in individual cases. Initial case studies of the few well resolved objects show that such misalignments are indeed present.

💡 Research Summary

The paper revisits the long‑standing question of how many active galactic nuclei (AGN) appear as Type 2 objects and why. By carefully separating true Type 2 AGN from lightly reddened Type 1s and from low‑excitation narrow‑line AGN (which likely represent a different accretion mode), the authors find that, after removing the low‑excitation population, genuine Type 2 AGN constitute about 58 % of the total AGN population, with an additional ~15 % being modestly reddened Type 1s. This fraction is remarkably consistent across radio, infrared, and volume‑limited samples, showing no dependence on luminosity. In contrast, X‑ray selected samples display a decreasing Type 2 fraction at higher luminosities. The authors explore several reasons for this discrepancy, including selection biases, the difficulty of detecting Compton‑thick sources in X‑rays, and the different physical sensitivities of X‑ray versus IR/radio diagnostics.

To explain the observed 58 % Type 2 fraction with minimal assumptions, the authors propose a “tilted‑disc” model. In this picture, gas inflow from kiloparsec scales arrives at random orientations relative to the central supermassive black hole. As the material spirals inward, it must align its angular momentum with that of the inner accretion flow. The alignment process is assumed to be dominated by a tilt of the disc rather than a twist, producing a severely warped, but essentially untwisted, structure on parsec scales. If the tilt angle θ is uniformly distributed between 0 and 90°, the covering factor C = sin θ has a mean value of 0.5, predicting that, on average, half of all AGN should be observed as Type 2. The slight excess to 58 % can be accommodated by modest additional obscuration (e.g., dusty clouds) or by selection effects.

Unlike the classic torus model, which assumes a roughly fixed geometry and a single covering factor for all objects, the tilted‑disc scenario naturally yields a broad distribution of covering factors from object to object. Consequently, individual AGN can display a wide range of obscuration levels, while the ensemble average remains close to the observed value. The model also predicts that nuclear structures (e.g., the inner disc, maser emission, radio jets) should be roughly aligned on average, but that significant misalignments will be present in many cases.

The authors test this prediction using a handful of well‑resolved nearby AGN. In NGC 1068, the maser disc is inclined by ~30° relative to the radio jet. In Circinus, the ionisation cone, maser disc, and jet axis are similarly offset. In NGC 4151, high‑resolution IR interferometry shows a disc orientation that does not coincide with the optical narrow‑line region. These case studies provide concrete evidence for the expected misalignments, supporting the tilted‑disc hypothesis.

Finally, the paper discusses broader implications. A warped, tilted disc can affect the efficiency of AGN feedback, the re‑processing of radiation into the infrared, and the time‑dependent appearance of an AGN as Type 1 or Type 2. The model suggests that the observed luminosity dependence in X‑ray samples may be largely a selection artifact rather than an intrinsic change in covering factor. Future high‑resolution observations with facilities such as JWST, ALMA, and the upcoming Athena X‑ray observatory will be able to map the geometry of the obscuring material in many more AGN, providing a decisive test of the tilted‑disc scenario.