Prediction of field capacity and permanent wilting point using rapid soil sensing approaches

Field Capacity (FC) and Permanent Wilting Point (PWP) are required in various biophysical models. Reliable prediction of PWP and FC from rapid sensing techniques such as Mid-Infra Red (MIR) spectroscopy and proximal sensing have been investigated for 2-paddocks representative of major dryland cereal producing soil/landscapes of North West Victoria. MIR derived calibrations were undertaken where the performance of data pre-treatment steps and uncertainties affecting the reliability and accuracy of the MIR predictions were considered. Geophysical (EM38, EM31 and gamma-ray spectra) and terrain derivatives were examined using spatial modelling approaches to predict FC and PWP. Individual paddock calibrations for MIR prediction of PWP and FC using Partial Least Squares (PLS) performed best in comparison to a state derived calibration. Terrain derivatives including elevation and relative elevation have proven significant variables in the spatial modelling of PWP and FC while gamma ray total count was positively correlated and EM38v negatively correlated.