Full-wave inversion of ground-penetrating radar data for soil characterization
(2013)
Files
PhDthesisArdekani.pdf
Restricted Access - Adobe PDF
- 110.8 MB
Details
- Authors
- Supervisors
- Lambot, Sébastien
- Abstract
- Measuring shallow soil water content is important for estimating water and energy fluxes for providing the knowledge of spatial distribution and dynamics in agricultural, hydrological, meteorological, and climatological research and applications. In that respect, the local point-scale and large-scale sensors are widely used, but the scientific demand for revealing the soil moisture dynamics at a field-scale with high spatial resolution is gaining interests. In this thesis, full-wave inversion of ground-penetrating radar (GPR) data for soil characterization was studied. The advanced GPR full-wave models of Lambot et al. (2004) and Lambot and André (2013) were presented and studied in the laboratory and field conditions. The stability and repeatability of GPR systems and effects of GPR instability on inversion results were studied. Different GPR systems and antennas operating in near-field and far-field conditions were employed in the field conditions for high-resolution mapping of surface soil moisture. In addition, the effects of vegetation cover on retrieving the surface soil moisture were investigated in the laboratory and field conditions and an effective model of scattering was developed. The GPR systems based on vector network analyzer showed a high degree of stability while the impulse commercial GPR systems presented significant instability (drift), which was corrected by signal post-processing. Comparison of the field-scale soil moisture maps between the advanced far-field GPR model and traditional on-ground approaches (direct-ground wave) showed different characterization depths. Furthermore, employing different GPR systems and antennas as well as different GPR models (far-field/near-field) provides different characterization scales but the same soil moisture patterns at field scale. The numerical experiments for spatial soil moisture variability within the GPR antenna footprint showed the effective antenna footprint of about -9 dB, which may measure the arithmetic average of the local soil moisture values. Also, numerical experiments resulted in deeper soil moisture sensing from the near-field GPR data than the far-field one. Finally, the vegetation scattering model developed in the laboratory condition was able to significantly correct the GPR soil moisture retrieval over a vegetated field.
- Affiliations
Citations
Mahmoudzadeh Ardekani, M. R. (2013). Full-wave inversion of ground-penetrating radar data for soil characterization. https://hdl.handle.net/2078.5/25302