UAV Multispectral Imaging Potential to Monitor and Predict Agronomic Characteristics of Different Forage Associations

The capability of UAVs imagery to monitor and predict the evolution several forage associations was assessed during whole growing cycle 2019–20. For this purpose, eight different grown in triplicate were used: vetch-barley-triticale (VBT), vetch-triticale (VT), vetch-rye (VR), vetch-oats (VO), pea-barley-triticale (PBT), pea-triticale (PT), pea-rye (PR) pea-oats (PO). Six biophysical parameters monitored through six vegetation indices on seven measurements dates distributed along cycle. experiments carried out organic farm “Gallegos de Crespes” located municipality Larrodrigo (Salamanca, Spain). results obtained exploratory correlation analysis suggested that a predictive model (PLS regression) could be performed. Overall, vetch-based showed slightly higher values for both field than pea-based ones. Correlations very strong significant each association throughout their cycle, suggesting would from spectral indices. Integrating these multispectral observations PLS model, agronomic predicted with reliability more 50%. A single combination VNIR (or even only visible) bands able feed regression leading successful prediction parameters.