Revealing brightness anomalies in the gravitationally-lensed ring of the cosmic horseshoe

Abstract

Strong gravitational lensing is a powerful tool for probing the observational signatures of small-scale structures (substructures) in cold dark matter (CDM), produced by sub-halos (invisible satellite galaxies), or small-scale structures (granulation) in wave-like dark matter ($\psi$DM), produced by self-interference of ultralight bosons. Both substructures and granulation will introduce slight perturbation to the magnification of lensed images, and hence brightnesses ratios between multiply-lensed image counterparts that differ from those predicted by a smooth-lens. In this thesis, I use gravitational lensing to study in detail the distribution of matter in the Cosmic Horseshoe (SDSS J1148+1930), which consists of a nearly complete Einstein ring formed by a background galaxy at $z_s = 2.38115$ that is highly magnified by a central lensing galaxy $z_l = 0.444$. From the \textit{Hubble Space Telescope (HST)} images, I robustly identify 47 bright multiply-lensed knots in the ring, corresponding to 18 distinct regions in the background galaxy, to serve as constraints for constructing parametric lens models for the central lensing galaxy. Adopting different mass profiles for the lensing galaxy, I find that the lens model that best reproduces the observed positions of the lensing constraints requires an elliptical power-law profile with density slope $n = 1.71$, along with up to fourth-order multipole perturbations for the lensing galaxy, as well as a minor external shear of $\gamma = 0.039$. A measure of the fidelity of the lens model is its ability to reproduce all of the multiply-lensed arcs, in which the bright knots are embedded, that make up the ring; the morphologies of these arcs are not used as lensing constraints, and therefore serve as independent tests of the lens model. Although able to reproduce the vast majority of the bright knots used as lensing constraints, three knots near the critical curve - whose flux ratios with their associated counterparts is $\sim30-50\%$ higher - stand out as being anomalously bright compared with that predicted by my lens model. By contrast, nearby knots in the same arcs have brightnesses approximately consistent with those predicted by my lens model. My work demonstrates the need for substructures or granulations in dark matter that will be difficult to be explained away as deficiencies in smooth lens models, as is possible in lensed systems in which much fewer lensing constraints are available. Implementing granulations onto the inferred global mass profile of the Cosmic Horseshoe, I demonstrate that $\psi$DM can plausibly explain the brightness anomalies that my smooth lens model finds for the aforementioned knots in the Einstein ring.