UC Berkeley

Deep surveys have allowed us to precisely chart the evolution of galaxies from billions of years ago through to the present day. We’ve found that galaxies in the distant past are dramatically different than those in the local universe: the stellar masses, sizes, structures, star formation rates, gas contents, and kinematics of galaxies all evolve with redshift. One of the few constants is the bimodality between star-forming and quiescent galaxies, which persists across a wide redshift range. Despite these observational advances, the exact nature of the physical processes governing the growth and transformation of galaxies remain unknown.

The first section of this thesis concentrates on the growth of galaxies. Previous studies have used half-light radii to trace the evolution of galaxy sizes. However, radial color gradients cause the light and mass profiles of galaxies to differ, biasing these measurements. I develop and test multiple techniques for measuring the mass profiles of galaxies from high-resolution multi-band Hubble Space Telescope imaging, then use these techniques to create the largest- ever catalog of galaxy half-mass radii at 0 ≤ z ≤ 2.5. This unique catalog enables me to correct for biases in previous studies of size evolution and discover that galaxies grow much more slowly than previously thought. I then examine how the structures of galaxies change as they age along the quiescent sequence, providing support for a merger-driven “inside-out” growth mechanism. Finally, I investigate how the structures of galaxies change as they gradually move up the star-forming sequence towards quiescence. These results pave the way for a new interpretation of the star-forming size-mass relation, and provide evidence for multiple distinct pathways for galaxies to shut down their star formation.

The second section of this thesis addresses the transformation of galaxies from star-forming to quiescent. Because stars form out of cold molecular gas, most theoretical mechanisms for this “quenching” process rely on heating, depleting, or removing molecular gas reservoirs. Post-starburst galaxies, whose spectra indicate that they have recently and rapidly shut down a major burst of star formation, are the ideal laboratory to test these theoretical quenching mechanisms. I introduce the SQuIGGLE multi-wavelength survey of intermediate-redshift post-starburst galaxies. After selecting ∼1,300 of these rare recently-quenched galaxies by their Sloan Digital Sky Survey spectroscopy, I perform detailed Bayesian spectral modeling to constrain their star formation histories. I also present Atacama Large Millimeter/submillimeter Array measurements of the CO(2–1) line in two SQuIGGLE galaxies. I show that, in contrast to theoretical predictions, these recently-quenched galaxies can retain large molecular gas reservoirs. These puzzling observations challenge our understanding of how galaxies shut down their star formation.