Spatially resolved physical properties of high-redshift galaxies observed with KMOS
Turner, Owen James
The formation and evolution of galaxies is governed by a set of complex non-linear physical processes. When we observe a galaxy at a particular epoch, we see a unique spatial distribution of stars, gas and dust, and can use the light emitted by these components to infer the chemical composition and dynamics of the galaxy. The spatial distribution of the different components, as well as the galaxy chemistry and dynamics, is shaped by the history of both the internal and external physical processes involving the galaxy until the time of observation. Although no two galaxies appear the same, the study of galaxy evolution is an attempt to understand generally the physical processes which lead to the observed distribution of galaxy properties at any stage across the history of the Universe. Many of these processes are still poorly understood, particularly in terms of the timescales over which they operate and their importance in the energetics of the galaxy. Much progress has been made in this area in both the local and distant Universe over recent years, thanks in particular to multi-object integral-field spectrographs, which have allowed for the spatial mapping of representative samples of the galaxy population. This thesis presents studies of the spatially resolved dynamical and chemical properties of high-redshift star-forming galaxy samples, using integral-field spectroscopic observations with the K-band Multi-Object Spectrograph (KMOS). In Chapter 2 I describe the data reduction pipeline developed to process these observations. Specifically, I have added processing steps which extend this pipeline beyond the standard recipes and describe these in the chapter. In Chapter 3, using a sample of 77 star-forming galaxies from the KMOS Deep Survey (KDS) at a median redshift of z ≃ 3:5, I demonstrate that the intrinsic, beam-smearing corrected velocity dispersions are ubiquitously high, with a sample median value of 70.8kms-1. At this redshift, the KDS sample is the largest from which these observations have been made and the measurements indicate that the trend of increasing intrinsic velocity dispersions continues out to at least z ≃ 3:5. In Chapter 4, by compiling 16 high-redshift comparison samples from the literature, I study the evolution of the stellar mass Tully-Fisher relation, demonstrating clearly that recent literature discrepancies can be explained by differing sample selection criteria. By also examining the rotation-velocity fields of the KDS galaxies, I discuss the possibility that a significant portion of gravitational support is provided by random motions, as evidenced by the redshift decline of the ratio of rotation velocity to velocity dispersion. This leads to the formulation of a ‘total velocity’ containing both velocity dispersion and rotation velocity terms. Combining the intrinsic velocity dispersion values with the rotation velocities of the galaxies is necessary to trace the galaxy potential wells at intermediate and high redshift. When this is done, the evolution of the normalisation of the total-velocity versus stellar-mass relation is consistent with a steady decline over cosmic time. Finally, using multi-band KMOS observations from the KMOS LEnsed galaxies Velocity and Emission line Review (KLEVER) survey, I explore the spatial distribution of emission-line ratios across individual galaxies at z ≃ 2:3, finding significant variation in the dust content and gas excitation conditions between galaxy cores and outskirts. The majority of KLEVER galaxies are observed to have negative Balmer decrement gradients (centrally concentrated dust distributions), with more-negative gradients belonging to the galaxies with higher stellar mass, lower Hα velocity dispersions and lower Hα luminosities. Similarly, the dust-corrected ratio of the [O iii]λλ4960; 5007/[O ii]λλ3727; 3729 emission lines (which may be used as a proxy for metallicity) for the majority of KLEVER galaxies has a radial profile with a positive gradient (i.e. higher metallicity cores), with more-positive gradients also correlating with higher stellar mass, lower Hα velocity dispersions and lower Hα luminosities. This work is still preliminary and requires further work to strengthen and build on the conclusions. The fundamental objective of the work presented throughout this thesis is to understand the spatially-resolved dynamical and chemical structure of high-redshift star-forming galaxies and to draw conclusions regarding how they evolve into the star-forming galaxies we observe in the local Universe.