Surface wave dispersion analysis and inversion from 3D complex land seismic exploration data

Ivan Sanchez, Daniel O. Trad, William M. Agudelo

Near-surface characterization plays a crucial role in enhancing seismic imaging quality and developing accurate velocity models. This is especially important in complex geological settings, such as foothill regions, where near-surface conditions like irregular topography and small-scale heterogeneities can significantly impact seismic data quality. Surface waves are particularly relevant for near-surface characterization, but their use in complex land seismic exploration data is limited. This is primarily due to three factors: (1) Surface waves are often treated as coherent noise (ground roll), which is typically filtered out during processing rather than analyzed. (2) Scattering noise distorts surface wave signals, complicating their interpretation. (3) Conventional land acquisition geometries are not optimal for surface wave analysis due to coarse spatial sampling, which leads to aliasing that reduces the accuracy of surface wave dispersion analysis. To overcome these challenges, this study introduces a novel methodology for inverting surface wave dispersion curves extracted from 3D land seismic exploration data, addressing issues frequently encountered in complex geological settings. The proposed workflow includes 3D beamforming, topographic corrections, azimuth binning, and midpoint stacking to generate high-quality dispersion images. Dispersion curves automatically extracted from these images are then inverted to construct an accurate pseudo-3D S-wave velocity model of the near-surface by combining 1D inversion results.Application of this method to data from the Colombian Foothills demonstrates its effectiveness in addressing the challenges posed by complex near-surface conditions, resulting in a detailed S-wave velocity model that improves seismic energy coupling and can potentially enhance the resolution of deeper subsurface structures.