Near-surface Characterization Using Seismic Refraction and Surface-wave Methods

Khaled Al Dulaijan

Varying thicknesses and velocities of near-surface layers cause serious problems for seismic reflection imaging of the deeper subsurface. Static corrections, calculated from near-surface velocity models, are used to remove the effects of the variable topographic near surface. In this thesis, near-surface layers are studied and characterized using two methods: Generalized Linear Inversion (GLI) of first arrival times and Multichannel Analysis of Surface Waves (MASW). The first is a widely used and proven method for obtaining near-surface velocity models for both P and S waves. The second is mainly used for geotechnical engineering purposes and is based on the dispersion properties of Rayleigh waves.

Two sites (at the Rothney Observatory and Spring Coulee, Alberta) are investigated. In the Rothney case, the S-wave velocities of the near surface (obtained by the MASW method) range from 200 to 1200 m/s and roughly correlate to the lithology log of a well drilled on site. In Spring Coulee, the S-wave near-surface models (obtained by GLI of first arrival times method) had different velocity layering and base of weathering from those of P waves. The MASW S-wave velocity model of Spring Coulee, Alberta correlates well to the GLI model. P-wave velocities range from 900 to 3600 m/s, while S-wave velocities range from 420 to 1966 m/s. Static corrections are calculated using near-surface models and are applied prior to stacking of PP and PS data. P-wave static corrections range from -25 ms to 27 ms, and those for S-wave range from -65 ms to 12 ms.