Urban structure imaging using noise-based techniques has rapidly developed in recent years. Given the complexity of the cross-correlation function in high-frequency signals, here, the beamforming (BF) method was used to analyze one data set taken from a dense array in the Jizhong Depression and obtain multi-mode dispersion curves. Multi-mode surface waves improved inversion stability, reduced non-uniqueness, and yielded a one-dimensional shear wave (S-wave) velocity model. Interpolation yielded a high-resolution three-dimensional (3D) S-wave velocity model for the study area. The model shows that velocity gradually changed in the horizontal direction and greatly increased in the vertical direction, which is largely consistent with changes in the sedimentary environment related to the continuous subsidence of the Jizhong Depression since the Quaternary. A low-velocity anomaly at a depth of ~300–400 m was revealed and determined to be caused by either a deep-buried ancient river course or low-lying area. This study demonstrates the potential of the BF method for processing dense array data sets of urban exploration. The high-resolution 3D S-wave velocity model provides a new reference for studying the Quaternary structure of the Jizhong Depression, as well as groundwater resources, urban infrastructure, and underground spaces.

Utilizing short-period dense seismic arrays, ambient noise tomography has proven effective in delineating continuous geological structures, a task critical for characterizing shale gas reservoir configurations. This study deployed 153 short-period seismic stations across the Xiahuayuan District in Zhangjiakou, a region with prospective shale gas deposits, to perform an ambient noise tomography survey. Through a meticulous process involving cross-correlation analysis, dispersion curve extraction, and subsequent inversion, a three-dimensional velocity structure model of the area was constructed. The model discerns subtle velocity changes within the 0–3 km depth interval, achieving a horizontal resolution of approximately 1.5 km in the 0–3 km stratum, thereby effectively delineating the shale reservoir structure. Integration of the velocity model with regional geological data facilitated a comprehensive interpretation and structural analysis of the prospective shale gas zone. Low-velocity anomalies observed within the velocity structure correspond to the spatial distribution of the Xiahuayuan Formation, likely attributable to the prevalent stratum of mudstone shale deposits within this formation. Employing a binary stratigraphic model, the study predicted shale content based on the velocity structure, with predictions exhibiting a moderate correlation (correlation coefficient of 0.58) with empirical data. This suggests the presented method as a viable rapid estimation technique for assessing the shale content of target strata.