Comparison of invasive and non-invasive electromagnetic methods in soil water content estimation of a dike model

Abstract

Water infiltration through a dike model under controlled flooding and drainage conditions was investigated using the gravimetric soil water sampling technique and electromagnetic techniques, in particular ground penetrating radar (GPR) applied in different forms, time domain reflectometry with intelligent microelements (TRIME-TDR) and spatial-time domain reflectometry (S-TDR). The experiments were conducted on the model in two phases. In the first phase, the model was flooded with varying water levels between 0 and 1.25 m above the waterproof base of the model. In the second phase, the characteristics of the temporal water content changes were investigated over a period of 65 days as the flood water drained off from the 1.25 m level. The dike model was constructed with soil of the texture class loamy sand. The aim of the experiment was to investigate whether GPR-based invasive and non-invasive methods were able to quantitatively observe and correctly monitor temporal changes in the volumetric water content (VWC) within embankment dams. The VWC values from the various techniques corresponded very well, especially with low VWC values. A comparison with the VWC of gravimetric soil water sampling showed a satisfactory reproducibility. Characteristic discrepancies were recorded with higher values of the VWC. Under saturated conditions only the invasive methods were able to produce reasonable values of the VWC. After the release of the highest flood level, the drainage phase could be characterized by two invasive methods based on the TDR and GPR techniques.