Abstract: Passive surface wave imaging has been a powerful tool for near-surface characterization in urban areas, which extracts surface wave signals from ambient seismic noise and then estimates subsurface shear wave velocity by inversion of the measured phase velocity. The high-frequency (approximately >1 Hz) seismic noise fields in urban environments are dominantly induced by human activities such as the vehicle traffic. Traffic seismic sources are nonrandomly distributed in time and space. Applying standard interferometric techniques to recordings from these nonrandom noise sources makes the Green’s function liable to estimation errors. We analyze the influence of using nonrandom traffic seismic sources for surface wave imaging. With nonrandom traffic seismic sources in time, spurious signals are generated in the cross-correlation function. With nonrandom traffic seismic sources in space, surface-wave phase velocities could be overestimated in the dispersion measurement. We provide an overview of solutions for surface-wave imaging with nonrandom traffic seismic sources in time and space, aiming to improve the retrieval of high-frequency surface waves and achieve reliable results from ultrashort (tens of seconds) observations for near-surface characterization.