What is the size of a galaxy?

My thesis has focused on measuring and understanding the sizes of galaxies. Galaxy size and how they correlate with other properties contain a wealth of information about the physical process which shaped how they formed. However, their are a few issues: How do we define the size of a galaxy? and What tracer do we use?. I have studied both of these questions, in papers discussed below.

Light is a biased tracer of galaxy stucture

Different regions of galaxies contain different stellar populations which have different colors. For example the Milky Way has a old red center and a young dusty outer disk. These distinct populations emit different amounts of light per unit stellar mass, biasing the light distribution away for the intrinsic mass distribution. When studying galaxy structure are interested in the intrinsic mass distribution, meaning these difference in colors can decieve us. In a recent study (arXiv link here), we used imcasacade (a flexible method we developed to measure galaxy profiles, see more info here) to measure the color profiles and half-mass radii for galaxies out to redshift 2. We confirm previous results which show that the ratio between mass and light weighted radii evolves significantly with redshift. This half-light to half-mass radii varies as a function of redshift, galaxy type and stellar mass, complicating our interpretation of the evolution of galaxy sizes.

Do we focus too much on the half-light radius?

Galaxies do not have an edge or cut-off so it is likely impossible to uniquely define the size of a galaxy. The most commonly used measure is the half-light, or effective radius, which is the radius that contains half of the total luminosity. In this study, we investigated using r₈₀ and r₂₀ the radii that contain 80% or 20% of a galaxy’s total luminosity and how it would impact out interpretation of galaxy sizes. In the figure below, taken from the paper, we show that the galaxy size - stellar mass relation looks markedly different depending on which definition of size is used. Specifically when we use r₈₀ star-forming and quiescent galaxies appear to have nearly the same size at all stellar masses. This is contrary to many previous works which suggest star-forming and quiescent galaxies follow different tracks in the size-mass plane.

Showing the galaxy size – stellar mass distribution using different definitions of radii from Miller et al. 2019