2021 Volume 90 Issue 3 Pages 261-268
The objective of this study was to clarify a vegetation index (VI) that could be the best measure for leaf nitrogen content of soybean canopy and identify the stages or periods at which the change in the VI positively correlates with seed yield. We investigated the relation between the nitrogen contents and VI in 2018 and 2019 and the relation between the VI and seed yield in 2019. In both years, plants were grown conventionally and in narrow rows, both at 2 or 3 planting densities. Soy fields were photographed with a multispectral camera and visible camera on an unmanned aerial vehicle. The images were reconstructed to generate ortho-images using the location information of the ground control points and VI maps of the fields were created. Nitrogen content of each organ, leaves, pods and others, were determined within 3 days after taking the photo. After seed maturation, we determined seed yield and VIs of the harvested areas. Among the VIs, GNDVI showed the highest coefficients of determination with the total nitrogen content and leaf nitrogen content but CIGreen was linearly related with those until the nitrogen content became high. Although CIGreen did not correlate with seed yield at any stage, we found highly significant positive correlations between seed yield and the change in CIGreen, delta CIGreen, in the periods of reproductive stages; the highest was the period between R1 and R5. Delta CIGreen (R1-R5) also showed highly significant positive correlations between the number of seeds and between 100-seed weight. These results suggest that seed yield of soybean could be speculated from delta CIGreen (R1-R5).
