(1) Characterisation of the pc-scale environment of the AGN: The primary goal was to establish the presence of dusty winds as a genuine feature in AGN. This has been achieved via a combination of VLT interferometry, single telescope and ALMA observations. While I originally relied on a specific new instrument (MATISSE), due to a delay plans needed to be changed. I have joined forces with the team building another interferometry instrument (GRAVITY). This included the first spatially resolved observations of the sub-parsec scale region of a quasar and the first reconstructed interferometric images of AGN in the infrared. In addition, I made use of the extended ESO science archive and we analysed archival data, both unpublished and previously published. Finally, a co-founded a team (GATOS) that undertook a survey of nearby AGN with ALMA in the sub-mm. As a result of these efforts, were were able to show that the obscuring structure around the AGN is dynamically and geometrically more complex than what may be expected from a single torus. We showed that dense, extended, geometrically thin disks are present in most of the Seyfert galaxies we observed and that they show some level of polar extended features in the dust continuum and molecular emission. With the archival data, we were able to show that the strength of polar features relative to the disks depends on Eddington ratio, as one would expect for a wind driven by radiation pressure.
(2) The physics of dusty wind features: To achieve this objective, advancements on the physical and radiative transfer modelling needed to be made. As part of the project, I revised my existing radiative transfer model of AGN to represent the configuration of a disk and polar wind. The model has been independently tested and been found to be overall the best to represent the infrared emission of AGN, especially in the mid- to high-luminosity regime where dusty winds are probably most common. The second, main part of this objective was building a new 3D radiation hydrodynamics code to simulate the parsec scale AGN environment and wind launching mechanisms. This required a huge effort, but resulted in a model that covers the widest dynamical range in all relevant parameters compared to other models and provides the resolution to resolve the wind launching mechanisms. We were able to show that dusty winds are a natural consequence of AGN and even if the AGN radiation pushes radially, will eventually result in polar extended infrared emission. We were also able to identify a mechanism that makes the winds and part of the accretion flow clumpy. Finally, I combined the bits of individual physics we learned from observations and simulations and proposed a full physical framework to explain observed phenomena, including predictions for the mass loss in dusty (molecular) winds due to radiation pressure on the dust.
(3) Dust parallax distances to AGN: This was the most challenging objective on the technical side. The original idea was to use an existing interferometric instrument at the VLTI (AMBER) and an established optical+infrared camera (SMARTS-ANDICAM). As AMBER suffered a calibration issue in the low-flux regime that could not be resolved, I used the new GRAVITY instrument instead with the team established for (1). Second, the SMARTS-ANDICAM camera broke about half way through the multi-year observing campaign without being replaced. As this inevitable meant that not the entire sample of AGN originally planned for would be covered, I reverted to extend the project from AGN as standard rulers to AGN as standard candles. For that, I established the VEILS public survey at the ESO 4m VISTA telescope. This survey uses the lag-luminosity relationship in AGN to use them in a similar way for cosmology as supernovae. The survey produced massive amounts of data that we are still in the process of analysing, with delays coming from temporary closure of the observatory due to COVID and staff changing jobs. In the end, we were able to use GRAVITY data and existing light curves to calculate direct distances to 5 AGN. However, the science prospects sparked interested beyond our group and is part of the science case for the new GRAVITY+ instrument that is currently being built.
