Urban environments require high-resolution, continuous subsurface imaging and monitoring to address challenges in infrastructure resilience, groundwater management, and seismic hazard assessment. However, traditional seismic networks are often limited by sparse spatial coverage, high deployment costs, and logistical constraints in urban areas. Distributed acoustic sensing (DAS) has emerged as a promising sensing technology, repurposing preexisting telecommunication fiber networks into dense, large-scale seismic arrays. This study demonstrates the potential of urban DAS for near-surface characterization and time-lapse monitoring using data from the Stanford DAS-2 experiment. A targeted interferometry workflow extracts high-quality virtual source gathers from vehicle-induced signals, enabling daily time-lapse elastic full-waveform inversion. Ambient noise interferometry extends both the offset range and low-frequency content beyond what targeted interferometry can achieve, while the nonrepeatability of ambient noise suggests the need for careful data selection to ensure robust subsurface monitoring. Additionally, earthquake recordings provide constraints on urban fault structures through the analysis of scattered wavefields. As DAS seismic arrays continue to expand in urban settings, they are poised to play a critical role in high-resolution, continuous near-surface characterization and monitoring.