Transforming electrical resistivity tomographies into volumetric water content ones: a strategy for optimizing irrigation in horticulture.

Agricultural water management is becoming an increasingly actual issue in a period of severe climate changings. Sustainable water management requires adequate knowledge of soil water availability and its storage capacity.

The correct management of irrigation water requires a good knowledge of the volumetric water content (VWC) in the soil. VWC is a parameter that can be measured using soil humidity sensors, and it can help understanding how the irrigation water distributes in the soil. However, these sensors are point humidity monitoring systems, that means they provide information limited to the vertical where they are installed and do not allow a 2D or 3D reconstruction of the water content in the subsoil. As well known, Electrical Resistivity Tomographies (ERT), a non-invasive geophysical method, estimates the spatial and temporal variations of soil resistivity (and thus of its inverse, i.e., conductivity), which is linked to parameters such as water content. Unlike point-based soil moisture sensors, ERT provides a broader view of water distribution in the soil. Thus, the goal of this study was to use electrical conductivity (EC) by full 3D-ERTs and moisture sensors to estimate the volumetric water content in the soil.

The study was conducted in a field dedicated to melon cultivation, where a detailed study of the irrigation system has been carried out over the years. It was determined that the three-drip-line system with a capacity of 4.1 lh/m² is the best irrigation system for this field located in Braccagni (GR, Italy). Therefore, one plot of the field was equipped with a three-drip-line irrigation system, and 72 electrodes were installed to perform full 3D-ERT measurements. Additionally, two PVC tubes, sealed at the base and with an opening at the surface, were installed to allow the insertion of the Diviner2000 probe and to measure soil moisture every 10 cm down to a maximum depth of 70 cm. Two ECH2O 10 HS sensors were also installed, connected to a data logger capable of recording temperature and moisture measurements every 30 min. The sensors, 15 cm in length, were installed vertically in the soil, allowing the measurement of VWC in a soil volume of 0.001 m³.

Between June and August 2023, six measurement campaigns of electrical conductivity were conducted. It is known that there is a direct relationship between EC and VWC. Therefore, the VWC data recorded by the Diviner2000 for all six acquisition times were correlated with the EC data acquired by ERTs. The two datasets (EC from ERT and VWC from Diviner2000) are in perfect agreement, showing a linear relationship with a R² of 0.96. Using the obtained regression law, it is possible to convert EC tomographies into VWC tomographies, thereby visualizing the variation of water content in the subsoil. This made it possible to understand the water distribution within the plot and to determine the percentage of water present throughout the entire root zone, not just at the points where the moisture sensors are installed.