Impact of Groundwater Level Fluctuations on In-Situ Soil Properties using Automated Wave Detection Methods

Abstract

Understanding the impact of groundwater level fluctuations on soil properties is crucial in geotechnical engineering, as it influences soil strength, stiffness, and stability. Dynamic soil properties such as Poisson's ratio, Young's modulus, Shear modulus, and Bulk modulus determine soil behaviour under dynamic loads from earthquakes, machine vibrations, and other disturbances. Thus, this study aims to assess the effects of water table fluctuations on dynamic soil properties by measuring P–wave velocity (VP) and S–wave velocity (VS) in a series of seismic cross-hole surveys, each characterised by varying water table depths, followed by meticulous analysis of the acquired data; however, data analysis continues to rely on manual processes, which are time-consuming, prone to human error, and biased by expertise. To address this concern, two automated algorithms, one for P-waves and the other for S-waves, were developed and applied to the recorded data. Further, the developed algorithms were applied to the series of recorded cross-hole experimental data to precisely capture the changes in seismic wave velocities arising from fluctuating groundwater levels. The entire vertical profile was dichotomised into two distinct regions based on the depth of the water table: above and below. To quantify the changes in seismic wave velocities above and below the water table, a test depth-to-water table depth ratio was introduced across all experiments and depths. A ratio between 0 and 1 indicates the zone above the water table, while a value greater than 1 denotes the zone below the water table. Continuous piecewise linear regression was performed on the recorded P-wave and S-wave velocities against these ratios. Both primary wave velocities (VP) and shear wave velocities (VS) exhibited a marked increase above the water table level as the test depth-to-water table depth ratio transitioned from 0 to 1; however, the increase was more subtle for VS. Beyond the water table, with nearly full saturation (ratios between 1 and 2), the P-wave velocity (VP) stabilised, indicating more uniform conditions below the water table. In contrast, the S-wave velocity (VS) exhibited a modest increase in this range. Beyond a ratio of 2, the S-wave velocity rose more significantly due to increased effective confining pressure, while the P-wave velocity remained constant. These regression equations were then integrated with established relationships for subsurface dynamic soil properties to comprehensively examine the variations caused by fluctuating groundwater levels.