AGN from Every Angle (and at Many Wavelengths)
Event date
Robert Nikutta
N232, room 103
Event Type

AGN unification and the clumpiness of the central obscurer ("torus") are mostly accepted, and both orientation and clumpiness designate an AGN as type-1 or type-2. Observations continue to add second-order complications, though. Small type-1 / type-2 samples have been compared through SED fitting, IR interferometry, and X-ray monitoring. It is, however, especially difficult to obtain large samples of type-2 AGN, since they are both dimmer and redder. To understand better the nuclear obscurer, we attack the problem at multiple wavelengths. In this talk I will focus on the IR, and on X-ray eclipse monitoring. WISE has produced a large catalog of IR-bright sources, among them ~10^6 QSOs. With only 4 bands and a small aperture, the analysis of single-source properties is challenging, but the large sample size allows statistical inference at a very significant level. We model the _ensemble_ properties of the entire AGN content in WISE using regularized linear regression, with orientation-dependent color-color-magnitude (CCM) tracks of CLUMPY torus models (Nenkova+2008) as basis functions. We are able to reproduce the observed number counts per CCM bin with very high accuracy, and simultaneously infer the probability distributions of torus parameters and redshifts. We reliably identify type-1 and type-2 populations through their IR properties alone, and for the first time are able to reveal the CCM location and number count distribution of a large obscured type-2s sample. Our analysis is extendable and would certainly benefit from including additional photometric points, _in particular_ in the FIR. Looking deeper into the nuclear region, we analyzed 16 years of X-ray AGN observations with RXTE, and found among 155 sources 12 secure eclipse events, plus several candidates. In Markowitz, Krumpe, Nikutta (2014) we show that these events are discrete gas+dust clouds crossing our line of sight. Due to RXTE's long mission duration, it was sensitive to eclipses caused by clouds much further away from their AGN than previous events, which have been observed with e.g. XMM/Newton and Chandra. Our clouds are located both in the BLR and also well into the dusty regions, the AGN torus. With the relatively large number of events we were able to derive "instantaneous" probabilities to observe an AGN in an obscured state. In a follow-up paper (in prep.) we constrain the properties of the clouds further, and test the predictions of most recent AGN torus models. Using Bayesian inference together with CLUMPY, we can derive some of the parameters that govern the distribution of clouds in the BLR and in the torus. This provides a natural connection between the X-ray signal (BLR/TOR) and the IR emission (TOR).