| The observational evidence acquired in the last decade suggests that the Universe is undergoing an accelerated expansion driven by an unknown energy component that has been labeled "Dark Energy". Weak gravitational lensing (WL) is a fundamental tool for probing dark energy. WL measures the distortions of the images of distant galaxies by the intervening matter along the line of sight. These distortions depend on both the matter distribution in the Universe and on its expansion rate, which is in turn sensitive to dark energy. One of the numerous applications of WL is cluster detection and counting. In Chapter 2 we present a Fisher matrix analysis of the constraints that WL-detected clusters can give on cosmology, and especially on dark energy. The study includes a halo model calculation of the projection effect, a source of systematic error for WL cluster counting which arises because the shear is an integral of the mass distribution along the line of sight. The most significant conclusion of this work is that the projection noise is usually subdominant to the shape noise from the intrinsic ellipticities of galaxies, and therefore it does not significantly affect the cluster cosmological constraints. In Chapter 3 we follow up these results with a detailed numerical investigation of the projection effect. In Chapter 4, we analyze the possibility of using high-redshift galaxy and 21-cm lensing surveys for CMB delensing and tensor-mode detection. Inflationary gravitational waves create tensor B modes in the CMB polarization maps. The tensor signal could reveal the energy scale of inflation, but it is contaminated by WL, which converts scalar E modes into B modes. Delensing means subtraction of the WL contamination from B-mode maps. Our delensing method optimally combines the lensing potential of lower redshift sources, such as galaxies and the 21-cm line, to obtain an estimator of the CMB lensing potential. The conclusion of this study is that future 21-cm surveys could reduce the WL contamination by an order of magnitude, without requiring the very high resolution in the CMB experiment needed by other delensing methods. |
