High-resolution shallow-seismic experiments in sand, Part I: Water table, fluid flow, and saturation

A high-resolution, very shallow seismic reflection and refraction experiment was conducted to investigate the seismic response of groundwater level changes in beach sand in situ. A fixed 10-m-long receiver array was used for repeated seismic profiling. Direct measurements of water level in a monitoring well and moisture content in the sand were taken as well. The water table in the well changed by about 1 m in slightly delayed response to the nearby ocean tides. In contrast, inversion of the seismic data yielded a totally different picture. The reflection from the water table at high tide appeared at a later time than the reflection at low tide. This unexpected discrepancy can be reconciled using Gassmann's equation: a low-velocity layer must exist between the nearsurface dry sand and the deeper and much faster fully saturated sand. This low-velocity layer coincides with the newly saturated zone and is caused by a combination of the sand's high density (close to that of fully saturated sand), and its high compressibility (close to that of dry sand). This low-velocity zone causes a velocity pull-down for the high-frequency reflections, and causes a high-tide reflection to appear later in time than low-tide reflection. The calculated velocities in the dry layer show changes with time that correlate with sand dryness, as predicted by the temporal changes of the sand's density due to changing water/air ratio. The results show that nearsurface velocities in sand are sensitive to partial saturation in the transition zone between dry and saturated sand. We were able to extract the saturation of the first layer and the depth to the water table from the seismic velocities. The high-resolution reflections monitored the flow process that occurred in the sand during the tides, and provided a real-time image of the hydrological process.