Krzysztof Hryniewicz, National Centre For Nuclear Research (NCBJ), Warsaw, Poland — April 20, 2026

Broad emission lines superimposed on a blue continuum are the most characteristic spectral signatures of unobscured Active Galactic Nuclei (AGN). The blue continuum was associated with accretion onto a supermassive black hole, borrowing the concept from X-ray binaries. By contrast, the debate about the geometry and origin of the line-emitting Broad Line Region (BLR) has continued.
Extended reverberation mapping campaigns have since yielded several key insights, most notably measurements of the mean BLR radius as a function of AGN luminosity. These measurements allow us to probe the underlying accretion disc properties and to identify how matter emerges from the disc into the BLR.
We interpret the BLR origin within the Failed Radiatively Accelerated Dusty Outflow (FRADO) framework, which ties the accretion disc directly to the line-emitting region through a radiatively driven, gravitationally bound failed wind. This model naturally explains why BLR clouds form at the dust sublimation radius and why their radial extent follows the observed size–luminosity relation.
Using a cloud distribution motivated by standard disc model, we further account for the radial extension of the line emitting regions and show how the BLR and Intermediate Line Region (ILR) arise naturally on top of the disc. Photoionization simulations robustly support the accretion-disc origin of the BLR, reproduce reverberation-mapped distances for different ionic species, and explain the distinct profile shapes of low- and high-ionization lines.
In the final part, I describe how X-ray absorption spectroscopy provides an independent test of this picture. Using an unprecedented
900 ks XMM-Newton RGS spectrum of Mrk 509, we apply Constant Total Pressure (CTP) photoionization models, which reproduce major part of the observed multi-phase warm absorber as a single thermally stratified slab. The high spectral resolution preserved by the full co-added exposure allows us to address individual line profiles, placing the absorber at the inner BLR scale and consistent with a FRADO-launched outflow. Both FRADO and CTP suggest that the radial density distribution of the matter producing spectral lines is strongly non-linear.