Probing gravity with weak lensing and clustering from mock galaxy surveys

Abstract

The quest for a theoretical explanation of the Universe's observed accelerated expansion has necessitated the inclusion of the cosmological constant (Λ) as a dark energy component. This constant behaves as a negative pressure, or effectively as "repulsive" gravity, on large cosmological scales. Alternatively, several modified gravity models have been proposed to account for this acceleration without relying on a cosmological constant. The proliferation of these gravity theories has highlighted the need to develop robust methods for testing their validity. As part of the main project of this thesis, we present the first computation of the gravity model testing parameter EG using realistic simulated galaxy mocks. The study focuses on measuring the EG estimator within the framework of General Relativity (GR) and f(R) gravity models, leveraging high-fidelity simulated galaxy catalogs. Our primary aim is to assess the potential of future galaxy surveys to detect deviations from standard gravity using this widely adopted estimator that combines galaxy clustering and weak gravitational lensing. Our findings indicate that, for an all-sky galaxy survey and without accounting for observational systematics, EG can be estimated accurately and with minimal bias for both gravity models across all redshifts. However, the error bars are too large to definitively distinguish between the theories. Alternatively, we propose a straightforward null test of gravity based on redshift-space distortion (RSD) clustering. This test suggests that, with precise modeling of small-scale behavior in future galaxy surveys, significant departures from standard gravity could potentially be detected. We developed tools to probe gravity from 2-point correlation functions for galaxy clustering and galaxy weak lensing. For galaxy clustering, we primarily conducted linear galaxy bias calculation using angular power spectra and growth rate estimates derived from multipoles of the correlation function. For gravitational lensing, we computed the magnification bias factor and performed 3x2pt analyses on GR and f(R) galaxy mocks. This expertise enabled us to deliver accurate calculations for various forecasts for the Euclid mission and some minor contributions for the DESI mission. In this thesis, we detail the methodologies and tools developed throughout the PhD research, which facilitated the achievement of these results. These tools include advanced techniques for calculating galaxy clustering and lensing statistics, which also served to validate the simulated mocks used in the study. During this process, we identified and resolved errors in the generation of the dark matter and galaxy catalogs through persistent testing and troubleshooting. While these setbacks delayed progress and complicated the production of final results, they provided valuable insights into theoretical implementations and cosmological simulations, ultimately enriching our understanding of these topics.

Date of Award	11 Feb 2025
Original language	English
Supervisor	Pablo Fosalba Vela (Director)