The James Webb Space Telescope (JWST) heralded a new era in extragalactic astronomy with its spatial resolution and sensitivity in the near infrared. This thesis leverages data collected during the first three years of the JWST mission to address some of the open questions concerning high redshift galaxies.
We first develop a software pipeline that produces science-ready photometric catalogs from reduced JWST imaging mosaics. Combining public imaging data across legacy fields, we compile a rest-frame optically selected sample of >30,000 galaxies at z~4-9 and calculate stellar mass functions (SMFs). We find that UV-red galaxies dominate the high mass end of the SMF to z~6, but that their abundance drops rapidly from z~4 to z~6. At z~7-9, the SMF evolves slowly, suggesting an increasing star formation efficiency (SFE).
Through the cycle 2 program RUBIES, we obtain a NIRSpec/PRISM spectrum of a source initially selected as a candidate massive galaxy at z~7. Modeling of the spectrum reveals that the galaxy is massive, lies at a redshift of z=7.29, and ceased forming stars 50-100 Myr before the time of observation, making it the most distant massive quiescent galaxy (MQG) known to date. Its high stellar mass density suggests some of the equally dense cores of local elliptical galaxies are in place less than 700 Myr after the Big Bang. Intriguingly, at z~7, the number density of MQGs in models and simulations is >100 times smaller than suggested by the discovery of RUBIES-UDS-QG-z7.
To get an unbiased view of the galaxy population at z>9, we combine data from the cycle 1 NIRCam pure parallel imaging program PANORAMIC with legacy imaging to a total of 35 independent sightlines. A careful search for Lyman break galaxies at z~10, z~13, and z~17 yields 86, 11, and 0 candidates respectively. Our inferred UVLF is consistent with literature results at z~10, but lies below most literature results atz~13. Together with our upper limits at z~17, this suggest a rapid drop in the cosmic star formation rate density by a factor of >50 from z~17 to z~10, and a rapid build-up of UV-bright galaxies in the first few hundred Myr.
These measurements are consistent with various theoretical models proposed to explain the high UVLF at z~10. To get a complementary probe of galaxy physics, we use our z~10 galaxy sample to measure cosmic variance, which is directly related to galaxy clustering. We implement models in the UniverseMachine that mimic the effect of proposed theoretical explanations for the high UVLF and explore their effect on cosmic variance. All models decrease cosmic variance by placing galaxies at fixed UV luminosity in lower mass halos but do so to varying degrees. Our measurements tentatively disfavor a global increase in the SFE or bursty star formation as the main driver of the high UVLF at z~10. While more data is needed to confirm these results, they emphasize the potential of future pure parallel imaging to constrain galaxy physics at cosmic dawn through clustering constraints.
This thesis highlights some of the progress that JWST has enabled in its first years of operation by shedding light on previously inaccessible or poorly characterized red and distant galaxies. It provides a more complete view of the galaxy census from z~4-17, presents the unexpected discovery of a MQG at z=7.3, and sketches avenues for future research building on the rich and growing data set provided by JWST.
