UC Berkeley

Mapping the distribution of matter and light in the Universe is key to unlocking some of its most fundamental secrets. What drives the acceleration of cosmic expansion? Does Einstein's Theory of General Relativity fail on very large scales, requiring a new model of gravity to account for an accelerating expansion? Or is there some exotic and invisible form of "dark energy" that dominates the composition of our Universe? Beginning with the Dark Energy Spectroscopic Instrument (DESI), a series of next-generation galaxy surveys will revolutionize our understanding of dark energy, providing an unprecedented wealth of data to solve these and other cosmic mysteries. Several major challenges directly determine the ultimate success of ambitious missions such as DESI: how tightly systematic sources of error and contamination can be controlled, how well the biases and properties of the galaxy samples can be understood, and how accurately their positions can be mapped. This thesis presents contributions I have made towards addressing these challenges and doing early science with DESI, for which I received "Builder Status" by the DESI Collaboration. Approximately two-thirds of this thesis are devoted to performing the first major analysis of the systematics and clustering of DESI samples selected from deep imaging. In the final part of this thesis, I present a cross-correlation between DESI galaxies and the lensing of the cosmic microwave background, one of the most significant detections of this type of signal to date, from which I further characterize the DESI samples and also study how the accuracy of their positions impacts science goals. In addition to enabling future cosmology with DESI, the methodologies and frameworks developed in this thesis have broader applications in future dark energy experiments and, more generally, cosmological studies using deep imaging data.