Vertical-component seismic waveforms and RTK ground-truth trajectories for kinematic tracking of urban anthropogenic sources, Bailu Lake, Huizhou, 2025

This dataset contains the field data underlying the manuscript "Resolving Kinematic Characteristics of Anthropogenic Seismic Sources using Dense Nodal Arrays and Spatiotemporal Deep Learning" (Chai, Yu et al.), submitted to the Journal of Applied Geophysics. Field experiments were conducted at Bailu Lake Kite Square, Ruhu Town, Huicheng District, Huizhou, Guangdong Province, China (approximately 23.215 N, 114.451 E), using four 4x4 dense nodal arrays (3 m station spacing, 9 m aperture). Contents:- data/single/ : continuous vertical-component seismic waveforms (.mat) of single moving pedestrian sources along EW, NS, and oblique traverses.- data/station_delay5/: inter-station timing-calibration files (.mat) for the four arrays.- RTK/single/ : RTK differential-GPS ground-truth trajectories (.csv) for the single-source recordings.- RTK/double/ : RTK ground-truth trajectories (.csv) for the dual-source motion scenarios used in evaluation.- RTK/2025040Xarray1-4.csv : RTK coordinates of the four array layouts.- RTK/20250401hammer.csv : RTK positions of the hammer-shot grid used for site-velocity (azimuthal) correction. The multi-source training samples used by the ConvLSTM network are not included here: they are generated by linear superposition of the single-source beam-energy sequences, and can be fully reproduced from these waveforms using the accompanying code archive. Processing and modeling code: https://doi.org/10.5281/zenodo.20618958