Earthquake seismology is an increasingly data-rich science, with tens of thousands of sensors recording the motions of the Earth in near-real time. Despite these rich new data sets and the opportunities they present (Arrowsmith et al., 2022), we still lack several key forms of observations needed to advance our understanding of earthquake rupture processes and the hazards they produce. One of the most acute challenges of this type relates to the paucity of high-quality, on-scale recordings close to large ruptures, especially within normal faulting sequences. Large earthquakes are thankfully rare, and few seismometers manage to capture the fine-scale details of the rupture process from a close vantage point (10 km or less). This observational deficit is more than just an academic concern and indeed has important implications for hazard, as the empirical ground motions models used in typical hazard applications must extrapolate beyond the support of their training data, while physics-based rupture models cannot be calibrated to match robust observations (Baker et al., 2021;Douglas & Edwards, 2016;Gerstenberger et al., 2020). A better resolution of near-source ground motions would be transformative for earthquake scientists, engineers, and communities near active fault systems.