Mapping the spatial variation of soil water content at the field scale with different ground penetrating radar techniques

Two ground penetrating radar (GPR) techniques were used to estimate the shallow soil water content at the field scale. The first technique is based on the ground wave velocity measured with a bistatic impulse radar connected to 450 MHz ground-coupled antennas. The second technique is based on inverse modeling of an off-ground monostatic TEM horn antenna in the 0.8-1.6 GHz frequency range. Data were collected on a 8 by 9 m partially irrigated intensive research plot and along four 148.5 m transects. Time domain reflectometry, capacitance sensors, and volumetric soil samples were used as reference measurements. The aim of the study was to test the applicability of the ground wave method and the off-ground inverse modeling approach at the field scale for a soil with a silt loam texture. The results for the ground wave technique were difficult to interpret due to the strong attenuation of the GPR signal, which is related to the silt loam texture at the test site. The root mean square error of the ground wave technique was 0.076 m3 m-3 when compared to the TDR measurements and 0.102 m3 m-3 when compared with the volumetric soil samples. The off-ground monostatic GPR measured less within-field soil water content variability than the reference measurements, resulting in a root mean square error of 0.053 m3 m-3 when compared with the TDR measurements and an error of 0.051 m3 m-3 when compared with the volumetric soil samples. The variability between the two GPR measurements was even larger with a RSME of 0.115 m3 m-3. In summary, both GPR methods did not provide adequate spatial information on soil water content variation at the field scale. The main reason for the deviating results of the ground wave method was the poor data quality due to high silt and clay content at the test site. Additional reasons were shallow reflections and the dry upper soil layer that cannot be detected by the ground wave method. In the case of off-ground GPR, the high sensitivity to the dry surface layer is the most likely reason for the observed deviations. The off-ground GPR results might be improved by using a different antenna that allows data acquisition in a lower frequency range.