High-resolution near-surface electromagnetic mapping for the hydrological modeling of an orange orchard

Abstract. While above-ground precision agriculture technologies provide spatial and temporal datasets ever-increasing in density and precision, below-ground information lags behind and has been typically limited to time series. As recognized in agrogeophysics, geophysical methods can address the lack of subsurface spatial information. This study focuses on high-resolution Frequency-Domain ElectroMagnetic induction (FDEM) mapping as an ideal complement to above- and below-ground time series that are commonly available in precision agriculture. Focused on a Sicilian orange orchard, this study first investigates some methodological challenges behind seemingly simple FDEM survey choices and processing steps, as well as their interplay with the spatial heterogeneity of agricultural sites. Second, this study shows how the detailed FDEM-based spatial information can underpin a surface/subsurface hydrological model that integrates time series from soil moisture sensors and micro-meteorological sensors. While FDEM has long been recognized as a promising solution in agrogeophysics, this study demonstrates how the approach can be successfully applied in a orchard, whose 3D subsurface variability is a complex combination of root water uptake, irrigation, evapotranspiration, and row-interrow dynamics. The resulting hydrological model reproduces the observed spatiotemporal water dynamics with parameters that agree with the results from soil laboratory analysis, supporting gamma-ray and electrical resistivity tomography datasets. The implementation of a hydrological model positively aligns with the increasing number and variety of methods in precision agriculture, as well as with the need for better predictive capability.