Cosmological tests of gravity with MGCAMB

The discovery of the cosmic acceleration, along with the old cosmological constant problem and the still-unknown nature of dark matter, stimulated exploration of alternate theories of gravity. The interest in testing general relativity (GR) on cosmological scales is further fuelled by the opportunities offered by existing and forthcoming cosmological surveys. This thesis presents several original contributions to the actively developing area of cosmological tests of gravity. First, we present a new version of Modified Growth with CAMB (MGCAMB), a widely used numerical tool for cosmological tests of gravity. New features include a parameterization allowing for a simultaneous reconstruction of phenomenological functions characterizing departures from the background expansion and departures in the linear growth of cosmic structures from the ΛCDM prediction. Other new features include the option to test models with a scalar field coupled only to dark matter, and the option to include dark energy perturbations when working with alternative background expansion histories. This version of MGCAMB comes with a Python wrapper to run it directly from the Python interface, making it easy to use with commonly used Monte-Carlo Markov Chain samplers. Next, we show how cosmological observations could distinguish between a modification of gravity and additional dark matter interactions - a question that has not been quantitatively investigated before. We demonstrate that data from a next generation survey, such as the Square Kilometer Array, will make it possible to distinguish between the two possibilities through measurements of gravitational redshift. Finally, we extend the framework in MGCAMB with a phenomenological model that can capture the nonlinear evolution of cosmic structures in a broad range of modified gravity theories. The extension employs the halo model reaction code ReACT used for modeling the nonlinear corrections in extensions of the ΛCDM model. We demonstrate that using this extension allows one to derive stronger constraints on modified gravity from the existing Dark Energy Survey data. The nonlinear extension will be of particular importance for the next generation of high resolution surveys such as Euclid and the Large Synoptic Survey Telescope.