Estimating Height and Yield in Grain Sorghum Using Uav Systems and Their Application in Breeding.

Plant breeders must increase the rate of genetic improvement to meet future crop production goals. Utilizing new technologies in these programs is essential and a major limitation for crop scientists is limits on the number and frequency of phenotypic data points. Emerging technologies such as unmanned aerial systems (UAS) present an exciting opportunity to address this limitation. To assess their utility in grain sorghum, plant height was evaluated as an initial trait. In this study, we present an in-depth statistical analysis of the ability for UAS to estimate height in sorghum (Advanced and Early Generation material) and the application of these estimates in their respective breeding programs. The UAS height estimates account for genotypic variation similar to ground-truth methods with high repeatabilities (R = 0.61 – 0.99), indicating that both techniques are effective at differentiating genetic material. Additionally, correlations between ground-truth and UAS measurements were moderate to high for all tests (r = 0.4 – 0.9). In terms of accuracy to identify genotypes with acceptable heights, UAS methods were most effective in early generation tests and less so in advanced generations.