Cosmology with large-scale structure: a sub-percent-accurate model for galaxy clustering – COLSS

What is the origin of the accelerated expansion of the Universe? What is the nature of dark matter? What is responsible for the accelerated expansion that took place before the Big Bang? These questions address the most puzzling mysteries of cosmology and more largely some of the most important issues in fundamental physics. Crucial hints may be hidden in the large-scale structure of the Universe and its evolution. Upcoming observations by cosmic surveys, such as the ESA Euclid, Vera Rubin and DESI, will bristle with important information on these questions. Our task in the next few years is to prepare ourselves to extract this information from their data.

The goal of this project is to use the most advanced theoretical developments in cosmology and particle physics to build a numerically fast and most accurate model of the redshift-space galaxy clustering, one of the main probes of future surveys. The speed is important to make it practical when sampling the likelihood in the multi-dimensional parameter-space. We aim at a sub per-cent accuracy to avoid systematic biases on the inferred parameters and fully exploit the data. Most importantly, accuracy can be controlled: Errors can be estimated by higher-order terms.

Advances in this direction have been undertaken and have led to publicly available codes. We will build on one of these codes in partnership with one of its authors. The constructed modeling and its associated code will allow us to measure the standard cosmological parameters and to test or detect physics beyond the standard model with unprecedented accuracy. For that, we will include five essential features crucial to address the above fundamental questions. Ultimately, we will provide a fast and accurate perturbation theory code able to:
1) Constrain or measure modifications of gravity with scale dependence;
2) Measure the mass of neutrinos;
3) Constrain or measure the presence of non-standard dark matter components;
4) Achieve 2-loop precision in perturbation theory in the power spectrum;
5) Constrain or measure non-standard initial conditions, such as non-Gaussianities and isocurvature perturbations.
The inclusion of these new features will require novel theoretical tools, beyond the state of the art. We will take advantage of different areas of theoretical physics, for instance the effective field theory of dark energy and scattering amplitudes techniques.

The project will coincide with important phases of Euclid: The launch of the satellite and the preparation of the analysis of the first data release, and it will closely follow related science activities. Importantly, there will be a transfer of knowledge: the partner institute will share its know-how on the most advanced existing models and numerical analysis of galaxy clustering. We will share our knowledge on modeling new physics. Beyond the immediate need of the final outcome of this project, this proposal aims at strengthening the participation of the French theory community within the Euclid collaboration, valorize its expertise in this very competitive field for Euclid and, more largely, in the international cosmological/astrophysical community.