Tomographic imaging of permafrost using three-component seismic cone-penetration test

Abstract

We conducted seismic cone-penetration tests (SCPT) and tomographic imaging in a permafrost mound in northern Quebec, Canada, to study the cryostratigraphy and assess the seismic properties of permafrost at temperatures near [Formula: see text]. A swept impact source generating both P- and S-waves and penetrometer-mounted three-component accelerometers were used to acquire surface-to-depth first-arrival times as input to produce 2D images of P- and S-wave velocities. Based on the three-component accelerometer records and the propagation modes of body waves, the P- and S-wave first arrivals were detected and discriminated. The inversion of the first-arrival times was based on the simultaneous iterative reconstruction technique. The multioffset surface-to-depth geometry used in this study limits the lateral resolution of tomographic imaging. However, the vertical variation in seismic velocities in the permafrost mound shows good reproducibility and can be compared to the cone data. The gathering of cone data such as cone resistance, friction ratio, electrical resistivity, and temperature, along with the seismic velocities, provides new insights into the cryostratigraphy of permafrost. While the cone data are affected by the vertical heterogeneity because of the complex sequence of ice lenses and frozen soil layers of a few centimeters thickness, the smooth velocity variations of P- and S-waves characterized by a wavelength of a few meters depend on the bulk physical properties of permafrost. The P- and S-wave velocities varied from [Formula: see text] and from [Formula: see text], respectively, for a temperature range between [Formula: see text] and [Formula: see text]. At this temperature range, the variations in unfrozen water content are important and affect directly the seismic properties of permafrost. The decrease in P- and S-waves velocities in depth with the permafrost mound depends nonlinearly on the increase of unfrozen water content from 9% to 30% for a temperature increase from [Formula: see text].