Using UAV Borne, Multi-Spectral Imaging for the Field Phenotyping of Shoot Biomass, Leaf Area Index and Height of West African Sorghum Varieties under Two Contrasted Water Conditions

Meeting food demand for the growing population will require an increase to crop production despite climate changes and, more particularly, severe drought episodes. Sorghum is one of cereals most adapted that feed millions people around world. Valorizing its genetic diversity improvement can benefit from extensive phenotyping. The current methods evaluate plant biomass, leaves area and plants height involve destructive sampling are not practical in breeding. Phenotyping relying on drone based imagery a powerful approach this context. objective study was develop validate high throughput field phenotyping method sorghum growth traits under contrasted water conditions imagery. Experiments were conducted Bambey (Senegal) 2018 2019, test ability multi-spectral sensing technologies on-board UAV platform calculate various vegetation indices estimate characteristics. In total, ten (10) varieties West African collection selected arranged randomized complete block design with three (3) replicates two (2) treatments (well-watered stress). This focused leaf index (LAI) measured weekly emergence maturity. Drone flights performed just before each images taken by visible cameras. UAV-derived exhibited their capacity estimating LAI biomass calibration data set, particular: normalized difference vegetative (NDVI), corrected transformed (CTVI), seconded modified soil-adjusted (MSAVI2), green normalize (GNDVI), simple ratio (SR) (r2 0.8 0.6 respectively). Developed models validated 2019 data, showing good performance 0.92 0.91 accordingly). Results also promising regarding estimation (RMSE = 9.88 cm). Regression plots between image-based showed r2 0.83. validation results similar treatments. first successful application development context characterized occurrence. developed could be used as decision support tool breeding programs characterization adaptive traits.