Estimation of spikelet number per area by UAV-acquired vegetation index in rice (Oryza sativa L.)

ABSTRACT The objectives of this study were to find the best vegetation index (VI) associated with plant nitrogen content at the reproductive stage in rice, to associate the number of spikelets with this VI and solar radiation, and to estimate the number of spikelets. Rice cultivars Ishikawa 65 and Koshihikari were grown in the field in 2019 and 2020 at various nitrogen application rates and transplanting densities. From 30 days before heading to just after heading, the field was imaged with a multispectral camera. The images were processed with predefined ground control point data to create VI maps. From the maps, VI data were retrieved from the canopy area where plants were harvested for the determination of plant nitrogen content at the reproductive stage and of the number of spikelets at maturity. Among 6 VIs tested, the chlorophyll index green (CIgreen) had the highest coefficient of determination (R 2) with plant nitrogen content at the reproductive stage and was the only VI with a linear relation with plant nitrogen content. The number of spikelets per unit area was well explained by multiple regression with CIgreen at 15 days before heading (CIG15) and cumulative solar radiation in the 15 days before heading (CSR15) as independent variables. A higher CIG15 would increase the number of spikelets differentiated and a higher CSR15 would reduce the rate of degeneration by increasing dry matter production. Graphical abstract