Abstract
Strong gravitational lensing is a unique tool which can allow us to probe the structure of galaxies and infer the existence of dark matter subhaloes around lensing galaxies. Detections of substructure can allow us to constrain the halo mass function, and thereby the nature of dark matter. JWST provides high, multi-filter resolving power, allowing us to model the same mass across different light configurations, providing compelling results when models agree. However, higher resolving power means we require more complex models to account for increased detectability of internal mass complexity, otherwise subhalos may be incorrectly inferred to exist in models purely to compensate for any missing complexity.
I will discuss the use of PyAutoLens modelling tools, including methods to incorporate angular complexity through “multipole” perturbations in mass profiles, and how we apply this to the galaxy-galaxy strong lens SPT2147. In this lens we achieve a new possible detection of a dark subhalo with Bayes Factors up to 50 when adding an NFW substructure to an Elliptical Power-Law + External Shear model, however, when we include angular complexity in this model, the Bayes Factor decreases to around +10. This is a perfect example of the mass degeneracy question: is the multipole model absorbing true subhalo signal; is the subhalo model absorbing true multipole complexity; or is it a mixture of both, in which case how do we know where to draw the line?
Strong lensing has the potential to unlock the secrets of dark matter, but we must be vigilant for systematic errors that can arise from modelling that which we cannot see. I will include an explanation of how we are hoping to place priors on our angular complexity in lens galaxies using “ellipse fits” to their lens light. With these priors, we will hope to improve the reliability of dark subhalo detections with the next generation of strong lenses from e.g. CSST, Euclid, and JWST, giving us the confidence in our detections to allow us to advance our understanding of dark matter.
Anyone interested is welcome to attend.