Abstract:
Deep foundations are often constructed as end bearing members with a significant amount of their capacity derived from bedrock. Sites with highly variable bedrock conditions therefore present a unique challenge when attempting to estimate the design elevation of deep foundations. Typical subsurface characterization techniques such as drilling, standard penetration testing (SPT), and cone-penetration testing (CPT) may fail to adequately establish top-of-rock elevation in geologic settings prone to karst and/or significant amounts of residuum, saprolite, and deformed bedrock. Seismic surface wave geophysical methods have been increasingly used to map bedrock topography in such settings. However, the current standard of practice for these methods introduces resolution limitations in spatially variable subsurface conditions due to averaging effects from the analytical framework used to process the acquired data. This study explores the use of a tomographic approach with full waveform inversion (FWI) to map top-ofrock elevations at a simulated site representative of highly variable bedrock topography. Comparisons are made with a dispersion-based multichannel analysis of surface waves (MASW) approach as more commonly used in geophysical subsurface investigations. The results highlight that FWI provides a superior estimate of top-of-rock topography than MASW at a cost of a significant increase in the required computational resources and time. Other issues related to full waveform tomography, including non-uniqueness and regularization are also presented.