Unveiling the growth of structure in the Dark Universe

Lead Research Organisation: University College London
Department Name: Physics and Astronomy

Abstract

The accelerated expansion of the Universe in recent cosmic times is one of the most important unexplained observations in modern physics. It has profound implications in that the cosmos could be permeated by a new, exotic matter component known as dark energy, or that one of the most successful theories of physics, Einstein's general relativity, breaks down on cosmological scales.
Emerging tensions between measurements in the late and early Universe are tantalising signs of a potential paradigm change away from the current cosmological standard model. The most promising route to shedding light on this key open problem is to explore the cosmic large-scale structure with multiple probes extracted from large galaxy surveys. I will map the growth of fluctuations in the matter distribution over an unprecedentedly wide range of cosmic history at per-cent level precision, enabling decisive conclusions on cosmic tensions and broad classes of alternative cosmological models. To this end, I will exploit two of the leading galaxy surveys of the decade, which are about to start delivering data: the ESA Euclid space mission and the DESI project, the largest spectroscopic galaxy survey ever undertaken. I will employ combined measurements of the clustering of galaxies, the gravitational lensing effect on
galaxies, as well as the spatial distribution of quasars and of imprints of large-scale structure in their spectra. The joint analysis in overlapping cosmic volumes ensures a maximum of robust constraints on cosmology and on the astrophysical processes that link galaxies to the underlying dark matter distribution. The fidelity of results is advanced through dedicated calibration measurements pushed far beyond the state of the art, and by abandoning the conventional data analysis in favour of a novel forward-modelling approach via fast, highly multiplexed simulations and machine learning-assisted inference techniques.

Publications

10 25 50