Linking horizontal crosshole GPR variability with root image information for maize crops

Abstract

Non-invasive imaging of processes within the soil–plant continuum, particularly root and soil water distributions, can help optimize agricultural practices such as irrigation and fertilization. In this study, in-situ time-lapse horizontal crosshole ground penetrating radar (GPR) measurements and root images were collected over three maize crop growing seasons at two minirhizotron facilities (Selhausen, Germany). Root development and GPR permittivity were monitored at six depths (0.1–1.2 m) for different treatments within two soil types. We processed these data in a new way that gave us the information of the “trend-corrected spatial permittivity deviation of vegetated field,” allowing us to investigate whether the presence of roots increases the variability of GPR permittivity in the soil. This removed the main non-root-related influencing factors: static influences, such as soil heterogeneities and rhizotube deviations, and dynamic effects, such as seasonal moisture changes. This trend-corrected spatial permittivity deviation showed a clear increase during the growing season, which could be linked with a similar increase in root volume fraction. Additionally, the corresponding probability density functions of the permittivity variability were derived and cross-correlated with the root volume fraction, resulting in a coefficient of determination (R²) above 0.5 for 23 out of 46 correlation pairs. Although both facilities had different soil types and compaction levels, they had similar numbers of good correlations. A possible explanation for the observed correlation is that the presence of roots causes a redistribution of soil water, and therefore an increase in soil water variability.