A JWST investigation into barred galaxies at 1 < z < 4

Galaxies are complex systems that preserve the history of the Universe, revealing a violent, chaotic phase in the early Universe. During this phase, galaxies frequently interacted and merged, laying the foundations for the Universe's hierarchical structure. As the Universe expanded, galaxy interactions became less frequent and internal evolutionary processes dominated. It is unknown when this second evolutionary phase commences; however, stellar structures in nearby galaxies have revealed the driving mechanism behind it. In dynamically cold and rotationally supported disc galaxies, a non-axisymmetric potential will propagate and trap stars onto elongated stellar orbits. A disc galaxy with this central stellar structure is known as a barred galaxy, and they are one of the most common features found in nearby discs. However, their frequency decreases as we look back to earlier epochs in the Universe, demonstrating that internal galaxy evolution dominates at z < 1. This thesis presents one of the first studies of barred galaxies at z > 1, using the improved sensitivity and longer wavelength range of the James Webb Space Telescope (JWST) to identify the onset of bar formation and, if present, to determine whether these barred structures differ from those in nearby galaxies.

I investigate the evolution of the bar fraction by selecting a nearly mass-complete disc galaxy sample in the redshift range 1 < z < 4 from the JWST Cosmic Evolution Early Release Science (CEERS) survey. The barred galaxies were identified via visual classification. To understand instrumental biases and wavelength dependence, the methodology was repeated for the same sample of galaxies imaged with the Hubble Space Telescope (HST), and in different JWST filters. For the first time, a population of high-redshift barred galaxies was identified with JWST, yielding a bar fraction twice that from HST images. Additionally, the bar fraction was consistent across the observed JWST wavelengths. Thus, the bar fraction can be limited by instrument sensitivity, and the bar fraction reported here could still be a lower limit.

In addition, I employ image analysis techniques to measure the evolution of bar structural properties. This includes ellipse fitting and photometric decompositions to obtain the bar length, ellipticity, Sersic index and bar-to-total luminosity ratio. I verify that ellipse-fitting techniques yield shorter, rounder bars than photometric decomposition, and that bluer-wavelength images yield shorter bars than redder-wavelength images. Overall, the bars at z > 1 have a similar distribution in bar lengths to those in nearby galaxies, but lack a longer population of > 8 kpc bars, which exist at z < 1. Overall, these bars reside in smaller disc galaxies; thus, the bar length normalised by disc size is constant from the local Universe out to Cosmic Noon. This remarkable result demonstrates that bars grow in tandem with their discs. Furthermore, the distributions of other structural properties, such as ellipticity, Sersic index, and bar-to-total luminosity, are similar to those of nearby and intermediate redshift barred galaxies.

One of the most exceptional results of this thesis was the discovery of a nuclear disc in a strongly barred galaxy at z ~ 1.5. Nuclear discs form shortly after the bar and are key indicators that bar-driven evolution has commenced. Using image analysis techniques, I measure the radial extent of this nuclear disc to be ~ 1 kpc, comparable to the largest nuclear discs in nearby galaxies. I identify possible features of a nuclear bar and spiral arms. This is the first evidence of a nuclear disc beyond the local Universe and reinforces the idea that bar-driven evolution must be rapid to form the structures we observe in nearby galaxies.

This thesis presents evidence that bar formation occurs much earlier than previously thought, as early as a few billion years after the Big Bang. Additionally, for bars in young galaxies to develop properties that are consistent with those we observe in nearby galaxies, bar-driven galaxy evolution must be rapid. In general, we must re-evaluate our understanding of disc settling, bar formation, and bar-driven evolution in the early Universe.