How did the first galaxies form in the universe? What drove the reionization of intergalactic hydrogen?

I study the formation and evolution of galaxies in the early universe, including those formed just a few hundred million years after the Big Bang, and their relationship to cosmic reionization.

My work combines empirical and semi-analytical theoretical modelling with statistical analyses of observations from Hubble Space Telescope imaging and grism spectroscopy, complemented by ground-based spectroscopy at Keck and the VLT. I am a member of the Brightest of Reionizing Galaxies (BoRG) Survey which searches for bright galaxy candidates at z > 8, and the Grism Lens-Amplified Survey from Space (GLASS) which is a spectroscopic survey of high-redshift galaxies lensed by clusters.


The universe underwent a phase transition from predominantly neutral at recombination (z ∼ 1100), to almost fully ionized by z ∼ 6. Reionization is thought to start by z < 20, but we have not identified the sources of ionizing photons – though the most likely candidates are young stars in galaxies. Lyman alpha emission from star-forming galaxies is absorbed by neutral gas, so observations of Lyα in high-z galaxies can constrain the timeline and topology of reionization. In Mason et al. (2017b) I presented a new Bayesian framework for directly inferring the neutral hydrogen fraction in the IGM from Lyα observations. I am leading the design and reduction of a ESO VLT Large Program with KMOS (PI: A Fontana) to confirm the largest sample Lyα emitting galaxies at z>6 to-date from GLASS.

Luminosity Function Evolution

Large groups of galaxies can be studied by their density distribution with luminosity, the galaxy luminosity function (LF). Observing changes in the LF with redshift allows us to see galaxy evolution in a statistical sense. In Mason, Trenti & Treu (2015) we show that if we assume halo growth drives the growth of galaxies we can explain the observed evolution of the LF and other global galaxy properties over 0 < z < 10 without requiring any evolution in feedback mechanisms.

Magnification Bias in Luminosity Functions

Light travelling from distant objects can be distorted by intervening mass distributions. For galaxies at high redshift the probability of being gravitationally lensed and magnified in this way is increased, which can distort the number of galaxies we detect at a given luminosity. In Mason et al. (2015) we show that this is not a significant effect in current surveys, but that it will dominate future wide field surveys such as Euclid and WFIRST.

KLASS Kinematics

Galaxies in the nearby universe tend to be either blue, star forming, disk galaxies, or red, dead, elliptical galaxies. How this bimodality arises and the transformation processes that can cause galaxies to change morphology is still an open question. KLASS (the KMOS Lens-Amplified Spectroscopic Survey) is providing 3D information about galaxies at 'Cosmic Noon' to help understand the evolution of star forming galaxies. By targeting galaxies lensed by massive clusters from GLASS we can resolve galaxies much fainter than ever previously studied. We find a huge diversity in the kinematics of galaxies in this epoch, though they are mostly dynamically hot disks.

Image Credit: Charlotte Mason (2015), Avi Loeb (2006)