Testing gravity using the growth of large scale structure in the Universe

Future galaxy surveys hope to distinguish between the dark energy and modified gravity scenarios for the accelerating expansion of the Universe using the distortion of clustering in redshift space. The aim is to model the form and size of the distortion to infer the rate at which large scale structure grows. We test this hypothesis and assess the performance of current theoretical models for the redshift space distortion using large volume N-body simulations of the gravitational instability process. We simulate competing cosmological models which have identical expansion histories - one is a quintessence dark energy model with a scalar field and the other is a modified gravity model with a time varying gravitational constant - and demonstrate that they do indeed produce different redshift space distortions. This is the first time this approach has been verified using a technique that can follow the growth of structure at the required level of accuracy. Our comparisons show that theoretical models for the redshift space distortion based on linear perturbation theory give a surprisingly poor description of the simulation results. Furthermore, the application of such models can give rise to catastrophic systematic errors leading to incorrect interpretation of the observations. We show that an improved model is able to extract the correct growth rate. Further enhancements to theoretical models of redshift space distortions, calibrated against simulations, are needed to fully exploit the forthcoming high precision clustering measurements.