Modeling Astrophysical and Large-Scale Structure Signatures in Axion Cosmologies

The quest to understand the fundamental nature of dark matter (DM) remains a challenge in contemporary physics. This thesis aims at elucidating astrophysical and cosmological imprints of fuzzy dark matter (FDM), a promising class of models which find ample motivation within theories of particle physics. Notably, the string axiverse hypothesis posits the existence of a multitude of pseudoscalar particles called axions with logarithmically distributed masses, potentially addressing both the DM and dark energy problems simultaneously. Combining theoretical analyses with numerical simulations, we unravel the distinct signatures of FDM across intergalactic and group/cluster scales. We begin by studying density, shape and weak lensing statistics in FDM cosmologies, finding strong departures from the universality of density profiles and the monotonicity of shape profiles observed in the vanilla cold dark matter (CDM) model. Our first-ever analysis of geometric and intrinsic alignments in FDM cosmologies reveals increased alignment strengths compared to CDM, holding particular relevance for weak lensing surveys like Euclid, and we embed our findings into a wider weak lensing analysis to predict convergence and ellipticity spectra. By publishing the COSMICPROFILES Python package we enhance reproducibility and accessibility of our results. To reveal the morphology of the cosmic web, we then apply an advanced segmentation algorithm (NEXUS+). We report high mean densities of filaments, sheets and voids in FDM compared to CDM, while cosmic skewness estimates also increase, which could be used as a testbed for constraining FDM. Moving to astrophysical implications, we deploy generative machine learning models (normalizing flows) to characterize neutral hydrogen distributions in post-reionization FDM model Universes. Our findings indicate that extreme FDM models can be ruled out based solely on their low neutral hydrogen (HI) abundance. We quantify the HI and damped Lyman-𝛼 (DLA) biases, and show that many small-halos in FDM models have high DLA cross-sections, which we trace to the high column density of cosmic filaments. We construct mock radio maps for the Square Kilometre Array and present prospects for imaging the brightest HI peaks. By interpolating the latent space of axion masses, normalizing flows predict HI distributions even for synthetic FDM cosmologies. Addressing the degeneracy between baryonic physics and FDM physics, we also aim to advance our understanding of baryonic feedback and the star-formation histories of high-redshift galaxies. We identify instances of (mini-)quenching in several state-of-the-art galaxy formation models and implement advanced spectral energy distribution models to match spectrophotometric data from the James Webb Space Telescope, suggesting higher levels of burstiness in some observed galaxies compared to simulated galaxies. Central to our work is the development of a robust simulation framework tailored to bona fide FDM. We find that not only do common approximations for FDM lead to unreliable estimates of key observables, but by pairing the gravity solver with the ILLUSTRISTNG galaxy formation module, we find that baryonic processes (adiabatic contraction and feedback) alter the structure of ground-state eigenmodes in the centers of halos called solitons. In particular, solitons dynamically relax into modified eigenmodes and experience baryon-induced growth. Finally, we extend our simulation framework to accommodate for mixed dark matter (MDM), where FDM constitutes only a fraction of the total DM content, in agreement with recent observational constraints. We present non-linear power spectra and halo mass functions in selected MDM cosmologies for the first time, concluding our advancements in FDM/MDM modeling.

Name dc:type.qualificationname

Doctor of Philosophy (PhD)

Level dc:type.qualificationlevel

Doctoral

Grantor dc:publisher.institution

University of Cambridge

Year dc:date.issued

2024

Author dc:creator

• Döme, Tibor

Advisors dc:contributor.advisor

• Fialkov, Anastasia
• Sijacki, Debora

• cosmology
• large-scale structure of Universe
• dark matter
• galaxy formation and evolution
• axion cosmology

Licence

• https://apollo8-f-pro.lib.cam.ac.uk/bitstreams/912c0385-b523-4aaf-90e0-fa99b68dde69/download
• https://creativecommons.org/licenses/by/4.0/

Language dc:language

eng

Author Identifier

0000-0003-2586-3702

OAI identifier oai:identifier

oai:www.repository.cam.ac.uk:1810/372527