Performance of Three Different Vegetation Indices in Modelling the Grain Protein Content of Winter Wheat Cultivated on Small- and Medium-Sized Parcels

As a result of recent technological advances in remote sensing and geographic information systems (GIS) more and more farmers and agronomic consultants are using unmanned aerial vehicles (UAVs) equipped with multispectral cameras for monitoring, estimating and predicting crop-related traits with the help of different vegetation indices (VIs). Since winter wheat ( Triticum aestivum L.) is one of the most important crops worldwide, the relationships between VIs and different agronomic or physiological traits of winter wheat have already been investigated in several studies. However, the relationships between different VIs and the grain protein-content (GPC) of winter wheat is still not well-studied and several issues remain unclear. Therefore, we tested the accuracy of three VIs (= the NDVI, GNDVI and NDRE) in modelling the GPC. The accuracy of all three VIs was quite similar, but we found a saturation effect for the NDVI and GNDVI at higher VI-values, which did not occur in the case of the NDRE. Therefore, we recommend the use of the NDRE to farmers and agronomic consultants for modelling and predicting the GPC of winter wheat instead of the NDVI or GNDVI.