Investigating the effect of maize roots under different nitrate applications using crosshole GPR

Non-invasive imaging of small-scale features within the soil–plant continuum can help to advance sustainable agriculture by optimizing agricultural treatments and protecting natural resources such as groundwater and soil. This study investigates the potential of different ground penetrating radar (GPR) frequencies with two primary objectives: monitoring soil water content (SWC) variations in maize root zones and detecting soil electrical conductivity variations caused by different nitrate concentrations. Therefore, weekly horizontal crosshole GPR measurements were conducted during a maize growing season using 200 MHz and 500 MHz GPR antennae at the upper field minirhizotron facility in Selhausen, Germany. Within the facility, horizontal rhizotubes are installed in three sets of three columns, with each column containing six rhizotubes at depths ranging from 0.1 m and 1.2 m and a horizontal rhizotube spacing of 0.75 m. These were used to acquire time-lapse measurements: horizontal zero-offset profiling (ZOP) collected between 0.2 m and 1.2 m and root images at all six depths. While variations in SWC and root presence are primarily linked to the permittivity, different nitrate concentrations are expected to cause variations in soil electrical conductivity, which affects the GPR signal attenuation resulting in a lower signal amplitude in areas of higher nitrate concentrations and vice versa. The permittivity of the soil is calculated for each position using the estimated travel time and the rhizotube spacing. Variations in GPR signal amplitudes are analyzed by calculating the envelopes and identifying their maximum at each position. For a time-lapse comparison, the static and dynamic influences are removed from the permittivity and maximum envelopes by using a statistical trend-correction approach. When both data are compared along the tube, the 500 MHz provides more details and structures than 200 MHz.  Pronounced root presence in the travel times are particularly evident at the 500 MHz frequency. Trend-corrected permittivity results show increased variability over time up to depths of 0.6 m and 0.8 m, correlating with greater root presence, while maximum envelopes shows greater variability only at 0.2 m. Preliminary results suggest that different nitrate concentrations affects the GPR data, with both frequencies indicating decreased maximum amplitudes in areas with higher nitrate concentration. At some locations in deeper layers, a decrease in maximum envelopes was observed while no increase in root presence was noticed, which could indicate zones of preferential flow of nitrate. The combined interpretation of permittivitiy and envelopes variations can help to disentangle the effect of SWC, roots and/or nitrate. These results highlight the potential of GPR as a non-invasive tool to accurately map root zones and to assess spatial variations in nitrate concentrations, thereby enhancing precision farming practices and promoting sustainable crop management.