Comparing galaxy-galaxy(-galaxy) lensing in semi-analytic models and observations to study galaxy evolution

Abstract

One of the main challenges in cosmology is to understand the properties of dark matter, its distribution in the Universe, and its connection with baryonic matter. An ideal method to study the relation between baryonic matter and dark matter is the so-called "gravitational lensing". It relies on the fact that the light emitted from a background source in the distant Universe is deflected by the foreground matter distribution or "lens'', leading to distortions in the observed image of the source. By studying these image distortions, one can obtain information about the mass distribution associated with the lens. In the weak gravitational lensing regime, the lensing effect is too small to create a detectable lensing signal from a single image. One thus needs to examine the distortions in a large number of sources in order to derive statistical properties about the lenses mass. <br /> In the case where both the source and the lens are galaxies, this technique is known as ``galaxy-galaxy lensing'' (GGL). <br /> Distortion patterns around lens galaxy pairs instead of individual galaxies can also be analysed, <br /> a method known as "galaxy-galaxy-galaxy lensing'' (G3L) which gives information on the matter environment of galaxy pairs. <br /> In order to be able to interpret GGL and G3L measurements, a theoretical understanding of these statistics is required. A common approach is to use semi-analytic models (SAMs) which combine the results from dark matter N-body simulations with analytical prescriptions for the physical processes governing galaxy formation and evolution. Comparing the outcomes of SAMs with observations therefore offers an opportunity to connect observed properties of galaxies with the underlying physical processes leading to those features.<br /> In this thesis, we first use synthetic galaxy catalogs from two SAMs, the Garching and Durham models, and their predictions of GGL and G3L for various galaxy populations. These SAMs are all implemented on one of the largest dark matter simulations, the Millennium Simulation. However, they differ in several details which lead to different predictions of GGL and G3L. Therefore, comparing the SAMs predictions against each other allows us to gain information on the physical processes involved and how the different treatments used in the models impact the signal. Moreover, comparisons between the SAMs predictions of GGL and G3L suggest that G3L provides new information which cannot be obtained from the second-order GGL statistics alone.<br /> In order to identify shortcomings of the SAMs and obtain valuable information on how to improve the models, one needs to compare the SAMs results with observational measurements. Therefore, in the second part of this thesis, we investigate the ability of three SAMs, the Garching and Durham models as well as an updated version of the Garching model, to reproduce observations of GGL and G3L. For this purpose, we use measurements from the Canada-France-Hawaii Telescope Lensing Survey (CFHTLenS) which is a multi-color optical survey optimised for weak lensing analysis. We study the GGL and G3L signals for galaxy samples selected according to their stellar mass and redshift, and analyze the clustering properties of galaxies and galaxy pairs of these samples. Our results indicate that not all models can quantitatively reproduce the GGL and G3L observations although there is an overall qualitative agreement between the models and CFHTLenS data.