Application of GPR early-time signal analysis for mapping and monitoring of soil moisture in the shallow vadose zone

Understanding the distribution of soil water content (SWC) in the vadose zone is important for agricultural purposes, allowing farmers to maximize irrigation efficiency. SWC also influences biological and chemical processes in the vadose zone, and understanding SWC distribution is instrumental climate modeling. Geophysical methods, including ground penetrating radar (GPR), are commonly employed for rapid, non-invasive estimation of SWC distribution from the field to global scale. This thesis seeks to further develop recent methodologies that allow for the use of GPR in traditionally non-viable field sites, such as those with a clay-rich subsurface or saline intrusion, to estimate SWC in the very shallow subsurface.Through two field irrigation studies and one tank study, we compare three statistics derived from the GPR early-time signal (ETS), which are the overlapping air- and ground-waves: the average envelope amplitude (AEA) and carrier frequency amplitude (CFA) statistics, and the crosstalk frequency (CTF). We collected measurements using antennae ranging in frequency from 200 MHz to 1200 MHz, and supported them with electrical resistivity (ER), soil sample gravimetric analysis, and time-domain reflectometry (TDR) and capacitance probe measurements. In our experiments, we find that the AEA and CFA correlate strongly with SWC, including spatial and temporal changes. While the AEA and CFA both provide excellent qualitative estimates of moisture content variations, the CTF provides the potential ability to make quantitative estimates of relative dielectric permittivity (ε_r) across multiple materials using a system calibration. We demonstrate that the CTF statistic successfully estimates changes in ε_r in 0%, 5%, and 10% clay mixed with sand, which suggests it could be used in heterogeneous environments and multiple field sites to estimate ε_r via a single calibration performed in a lab environment. Site-specific calibrations would be required to convert ε_r to SWC. This thesis demonstrates that the ETS extends the usefulness of GPR to conductive materials by allowing for rapid estimation of changes in the SWC of the very shallow subsurface. Additionally, the CTF statistic provides a means by which to directly and quantitatively estimate SWC via measurement of ε_r, using a system calibration that is unaffected by changes in soil type.Ph.D.Includes bibliographical reference