Field-scale assessment of soil water dynamics using distributed modeling and electromagnetic conductivity imaging

Knowledge of the soil water balance is fundamental for improving crop water use in agricultural fields. Estimates are normally for representative and homogeneous areas where the variability of soil properties is neglected. However, this variability significantly impacts soil water dynamics at the field scale. In this study, the MOHID-Land distributed process-based model was used to compute the spatially explicit soil water dynamics in a 22.6-ha almond field located in southern Portugal. An electromagnetic induction survey was first performed to obtain electromagnetic images of the real soil conductivity in depth, which were related to soil texture. Then, pedo-transfer functions were used to convert soil texture into soil hydraulic data. MOHID-Land results included maps of the spatial distribution of soil water contents, actual crop transpiration, actual soil evaporation, percolation below the rootzone, and surface runoff. These allowed identifying preferential flow pathways as well as the main control factors influencing soil water dynamics at the field scale. Some development needs were identified, and overcoming them would enhance the significance of contributions such as this study to the field of precision agriculture.