Depth (m) 0.88 1.88 2.88 3.88 4.88 Arrival times (ft) 2.90 6.17 9.45 12.75 16.0 Arrival Time (ms) 22.3 22.46 22.565 22.79 23.42 acquired at various SSROs and, while analyzing the data sets, 2D velocity models are derived for each subsequent offset by using previously established velocity values that are used whenever the ray path travels through an area that previously covered. For example, the ray path for offset X2 Depth (m) 0 to 0.88 0.88 to 1.88 1.88 to 2.88 2.88 to 3.88 3.88 to 4.88 (ft) 0 to 2.90 2.90 to 6.17 6.17 to 9.45 9.45 to 12.75 12.75 to 16.0 Estimated interval velocities years, the authors have performed extensive work to develop a technique that utilizes five different normalized parameters: 1. Linearity estimate (i.e., hodograms fitting straight lines) from the polari- zation analysis of the data trace 2. Cross-correlation coefficient of the full waveform of the data trace and at preceding depths 3. Deviation of the source wave frequency spectrum from a desirable bell-shaped curve 4. Signal-to-noise ratio of the data trace 5. Peak symmetry differential defining the skewing or time shifting of the peak source wave response Based on the value of these parameters, guidance can be obtained on how to best process the data sets by applying seismic signal processing tools, such as digital filtering, spectral isolation, polarization and time windowing. 2D Tomographic SCPT Imaging Using the Normal Moveout Seismic Cone Tomographic Testing (NMO-SCTT) algorithm, SCPT results can be used to generate 2D tomographic profiles for site characterization, which will illustrate site variability. Downhole seismic data sets are 106 • DEEP FOUNDATIONS • MAY/JUNE 2018 Internal Velocity SRA FMDSM (ms) (ms) 140 2588 5891 3313 1346 140.2 172.8 249.8 378.5 514.8 and depth Z2 might travel through areas V2D[1,2], V2D[1,1] and V2D[2,1]; the vel- ocity values for the last two areas obtained during the analysis of the data set for offset X1 are used and V2D[1,2] is estimated. The algorithm imple- ments an iterative technique based on the same mathematical tools used in the single source offset FMDSM technique but with additional slant plane interfaces for each source offset. As demon- strated in the figures, this approach can be used to illustrate site variability and that source waves do not have straight ray paths, emphasizing the impor- tance of using analytical techniques, which consider ray path refraction when estimating the in-situ velocity interval. Conclusion Geotechnical si te characterizat ion performed using applied seismic tools is increasing in use, even if only to compute the average shear wave velocity for the top 30 m (100 ft) of the soil needed for the site hazard classification. An overview of the theory behind various seismic testing methods (downhole seismic testing, crosshole seismic testing and spectral analysis of surface waves) was presented along with the authors’ explanation of why downhole seismic testing DST and, especially, seismic cone penetration testing are the preferred methods. Clearly, these methods cannot be used everywhere, but the most common reasons why seismic cone penetration testing is not used is based on perceived limitations with the method (e.g. difficulties in near-surface characterization). Larger offsets for the seismic sources and 2D tomography were introduced, which should shed a different light on perceived limitations and encourage wider use of seismic cone penetration testing and applied seismic tools, in general, for geotechnical site characterization. Schematic of NMO-SCTT testing and analysis configuration, after Baziw and Verbeek (2018) Estimated source raypaths for a 5 m SCPT sensor-source radial offset, after Baziw and Verbeek (2018) Estimated source raypaths for a 10 m SCPT sensor-source radial offset, after Baziw and Verbeek (2018) Erick Baziw received a B.A.Sc.in geophysics engineering, an M.A.Sc. in geotechnical engi- neering and a Ph.D. in seismic signal processing from the University of British Columbia. His primary interest is in the design of seismic signal processing algorithms for applied seismology applications. Gerald Verbeek received a B.Sc. in civil engi- neering and an M.Sc. in structural engineering, both from Delft University of Technology in The Netherlands. One of his interests is to increase the awareness and understanding of Cone Penetration Testing (CPT), including Seismic CPT, for site characterization.
