Galaxy formation and evolution

Abstract:

Galactic-scale tests have proven to be powerful tools in constraining fundamental physics in previously under-explored regions of parameter space. In this thesis we use astrophysical systems to test some of the fundamental principles governing our current theories of the Universe, through the development of source-by-source, Monte Carlo-based forward models.

We consider modifications to the propagation of light by one of three effects: quantum gravity (QG), a non-zero photon mass and a violation of the Weak Equivalence Principle (WEP). We use spectral lag data of Gamma Ray Bursts from the BATSE satellite to constrain the photon mass to be $m_\gamma < 4.0 \times 10^{-5} \, h \, {\rm eV}/c^2$ and the QG length scale to be $\ell_{\rm QG} < 5.3 \times 10^{-18} \, h \, {\rm \, GeV^{-1}}$ at 95\% confidence, WEP to $\Delta \gamma < 2.1 \times 10^{-15}$ at $1 \sigma$ confidence between photon energies of $25 {\rm \, keV}$ and $325 {\rm \, keV}$, and we demonstrate that these constraints are robust to how one models other contributions to the signal.

We investigate Galileon modified gravity theories by studying the offsets between the centre of a galaxy and its host supermassive black hole (BH). We constrain the Galileon coupling to be $\Delta G / G_{\rm N} < 0.16$ at $1\sigma$ confidence for Galileons with crossover scale $r_{\rm C} \gtrsim H_0^{-1}$. Inspired by the aforementioned test of modified gravity, we study spatially offset BHs in the Horizon-AGN simulation and compare these to observations, finding i) the fraction of spatially offset BHs increases with cosmic time, ii) BHs live on prograde orbits in the plane of the galaxy with an orbital radius that decays with time but stalls near $z=0$, and iii) the magnitudes of offsets from the galaxy centres are substantially larger in the simulation than in observations.

By cross-correlating dark matter density fields inferred from the spatial distribution of galaxies with gamma ray data from the \textit{Fermi} Large Area Telescope, marginalising over uncertainties in this reconstruction, small-scale structure and parameters describing astrophysical contributions to the observed gamma ray sky, we place constraints on the dark matter annihilation cross-sections and decay rates. We rule out the thermal relic cross-section or $s$-wave annihilation for all $m_\chi \lesssim 7 {\rm \, GeV}/c^2$ at 95\% confidence if the annihilation produces $Z$ bosons, gluons or quarks less massive than the bottom quark. We infer a contribution to the gamma ray sky with the same spatial distribution as dark matter decay at $3.3\sigma$. Although this could be due to dark matter decay via these channels with a decay rate $\Gamma \approx 3 \times 10^{-28} {\rm \, s^{-1}}$, we find that a power-law spectrum of index $p=-2.75^{+0.71}_{-0.46}$ is preferred by the data.

Finally, we outline a framework for assessing the reliability of the methods used in this thesis by constructing and testing more advanced models using cosmological hydrodynamical simulations. As a case study, we use the Horizon-AGN simulation to investigate warping of stellar disks and offsets between gas and stars within galaxies, which are powerful probes of screened fifth-forces.

Abstract:

Supermassive black hole binaries (SMBHBs) produced in galaxy mergers are thought to complete their coalescence, below separations of r_GW=10^{-3} (M_BH/10^8 M_sun)^{3/4} pc, as their orbit decays due to the emission of gravitational waves (GWs). It may be possible to identify such GW-driven inspirals statistically in an electromagnetic (EM) survey for variable sources. A GW-driven binary spends a characteristic time T_GW at each orbital separation r_orb < r_GW that scales with the corresponding orbital time t_orb as T_GW = (const) t_orb^{8/3}. If the coalescing binary produces variations in the EM emission on this timescale, then it could be identified as a variable source with a characteristic period t_var = t_orb. The incidence rate of sources with similar inferred BH masses, showing near-periodic variability on the time-scale t_var, would then be proportional to t_var^{8/3}. Luminosity variations corresponding to a fraction f_Edd<0.01 of the Eddington luminosity would have been missed in current surveys. However, if the binary inspirals are associated with quasars, we show that a dedicated survey could detect the population of SMBHBs with a range of periods around tens of weeks. The discovery of a population of periodic sources whose abundance obeys N_var = (const) t_var^{8/3} would confirm (i) that the orbital decay is indeed driven by GWs, and (ii) that circumbinary gas is present at small orbital radii and is being perturbed by the BHs. Deviations from the t_var^{8/3} power-law could constrain the structure of the circumbinary gas disk and viscosity-driven orbital decay. We discuss constraints from existing data, and quantify the sensitivity and sky coverage that could yield a detection in future surveys.

Abstract:

Cosmic filaments are the channel through which galaxy groups assemble their mass. Cosmic connectivity, namely the number of filaments connected to a given group, is therefore expected to be an important ingredient in shaping group properties. The local connectivity is measured in COSMOS around X-Ray detected groups between redshift 0.5 and 1.2. To this end, large-scale filaments are extracted using the accurate photometric redshifts of the COSMOS2015 catalogue in two-dimensional slices of thickness 120 comoving Mpc centred on the group's redshift. The link between connectivity, group mass and the properties of the brightest group galaxy (BGG) is investigated. The same measurement is carried out on mocks extracted from the lightcone of the hydrodynamical simulation Horizon-AGN in order to control systematics. More massive groups are on average more connected. At fixed group mass in low-mass groups, BGG mass is slightly enhanced at high connectivity, while in high mass groups BGG mass is lower at higher connectivity. Groups with a star-forming BGG have on average a lower connectivity at given mass. From the analysis of the Horizon-AGN simulation, we postulate that different connectivities trace different paths of group mass assembly: at high group mass, groups with higher connectivity are more likely to have grown through a recent major merger, which might be in turn the reason for the quenching of the BGG. Future large-field photometric surveys, such as Euclid and LSST, will be able to confirm and extend these results by probing a wider mass range and a larger variety of environment.

Abstract:

We present HST WFPC2 I band imaging for a sample of 9 Hyperluminous Infrared Galaxies spanning a redshift range 0.45 < z < 1.34. Three of the sample have morphologies showing evidence for interactions, six are QSOs. Host galaxies in the QSOs are reliably detected out to z ~ 0.8. The detected QSO host galaxies have an elliptical morphology with scalelengths spanning 6.5 < r_{e}(Kpc) < 88 and absolute k corrected magnitudes spanning -24.5 < M_{I} < -25.2. There is no clear correlation between the IR power source and the optical morphology. None of the sources in the sample, including F15307+3252, show any evidence for gravitational lensing. We infer that the IR luminosities are thus real. Based on these results, and previous studies of HLIRGs, we conclude that this class of object is broadly consistent with being a simple extrapolation of the ULIRG population to higher luminosities; ULIRGs being mainly violently interacting systems powered by starbursts and/or AGN. Only a small number of sources whose infrared luminosities exceed 10^{13}Lsun are intrinsically less luminous objects which have been boosted by gravitational lensing.