Seminar by Michael Barth, Univ of Montreal
Dynamical mass measurements are the foundation for our understanding of the co-evolution of supermassive black holes (SMBHs) and their host galaxies. However, precise dynamical measurements have typically been limited to nearby galaxies due to the high spatial resolution necessary to resolve the black hole’s gravitational sphere of influence. Strong gravitational lensing offers a route to extending spatially resolved gas-dynamical measurements into the early Universe, but in practice, the most promising target galaxies are highly dust obscured and often host active galactic nuclei (AGN). This poses a challenge for existing modelling methodologies, since obscuration and AGN luminosity can compromise stellar-light imaging, making it difficult to disentangle the gravitational contributions of stars and the SMBH. We introduce an imaging-free method that fits flexible axisymmetric stellar mass profiles directly to molecular-gas kinematics, without requiring observations of the stellar light. Implemented in SuperMAGE, our fully differentiable gas-dynamical simulator, and paired with VisCube, a visibility-space gridder and uncertainty estimator, this approach enables Bayesian inference of SMBH masses and stellar mass profiles directly from interferometric visibilities. We benchmark our method against imaging-based measurements in the nearby galaxy NGC 4697 and quantify systematic biases introduced by assumptions about the gas emission profile. Our results demonstrate that imaging-free modelling can yield SMBH mass constraints competitive with those obtained using traditional stellar-light-based methods. Finally, we combine our dynamical modelling framework with gravitational lens modelling and present preliminary dynamical constraints on the central compact mass in the strongly lensed galaxy ID141 at z = 4.2.
Host: Mary Putman via Greg Bryan