Mapping shallow soil moisture profiles at the field scale using full-waveform inversion of ground penetrating radar data

Abstract

Full-waveform inversions were applied to retrieve surface, two-layered and continuous soil moisture profiles from ground penetrating radar (GPR) data acquired in an 11-ha agricultural field situated in the loess belt area in central Belgium. The radar system consisted of a vector network analyzer combined with an off-ground horn antenna operating in the frequency range 200-2000 MHz. The GPR system was computer controlled and synchronized with a differential GPS for real-time data acquisition. Several inversion strategies were also tested using numerical experiments, which in particular demonstrated the potentiality to reconstruct simplified two-layered configurations from more complex, continuous dielectric profiles as prevalent in the environment. The surface soil moisture map obtained assuming a onelayered model showed a global moisture pattern mainly explained by the topography while local moisture patterns indicate a line effect. Two-layered and profile inversions provided consistent estimates with respect to each other and field observations, showing significant moisture increases with depth. However, some discrepancies were observed between the measured and modeled GPR data in the higher frequency ranges, mainly due to surface roughness effects which are not accounted for. The proposed GPR method and inversion strategies show great promise for high-resolution, realtime mapping of soil moisture at the field scale. * julien.minet@uclouvain.be

Validation of ground penetrating radar full-waveform inversion for field scale soil moisture mapping

“…2 (b). This time-windowing permitted to identify the surface soil dielectric permittivity and correlated soil moisture from the shallow soil layer (Lambot et al, 2006b;Minet et al, 2011).…”

“…Using the same GPR setup, the effect of surface roughness on the retrieval of the dielectric permittivity was analyzed in laboratory in Lambot et al (2006a) and widely discussed in case of field acquisition in Minet et al (2011). Owing to the relatively low-frequency range used in the GPR inversion, surface soil roughness may significantly impact only for roughness larger than 5 cm.…”

“…The effect of vertically-varying soil moisture was investigated through numerical and laboratory experiments in Minet et al (2010a) and in field conditions in Minet et al (2011). In the presence of a shallow soil layer of a different moisture content than the sublayer, a RMSE of 0.05 m 3 m −3 in the soil moisture estimation was obtained (Minet et al, 2010a, laboratory experiment).…”

“…Few studies have applied such an off-ground GPR approach for soil moisture sensing in field conditions (Chanzy et al, 1996;Redman et al, 2002;Or, 2003, 2005). Based on a full-waveform inversion of the GPR data and an accurate GPR system modeling, the off-ground GPR system developed by Lambot et al (2004Lambot et al ( , 2006b) has shown excellent potentialities for surface soil moisture sensing and mapping in field conditions (Weihermüller et al, 2007;Lambot et al, 2008;Jadoon et al, 2010a;Jonard et al, 2010;Minet et al, 2011). The method relies on an accurate radar model that, in particular, accounts for the antenna and antenna-soils interactions.…”

Integrating coarse-scale uncertain soil moisture data into a fine-scale hydrological modelling scenario

“…First, a scale gap exists when the distributed model is employed at a finer resolution than the field scale, for instance as to meet precise agricultural needs or to model rainfall/runoff in a small hydrological catchment. The within-field variability of soil moisture may be significant (Minet et al, 2011b;Hupet and Vanclooster, 2002;Western and Blöschl, 1999) and may have a significant impact on field-scale hydrological behaviour (Minet et al, 2011a;Mallants et al, 1996), which justifies the spatial distribution of hydrological parameters within a field plot for accurate hydrological modelling. Second, the possibility distribution of soil moisture content represents the possible field-averaged soil moisture contents for a given field, whereas an empirical probability distribution, reflecting the within-field soil moisture variability can be computed on the basis of the fine-scale soil moisture content values predicted by the hydrological model.…”