High‐Throughput Phenotyping of Cotton in Multiple Irrigation Environments

• Published in: Crop science 2015-03, Vol.55 (2), p.958-969
• Main Authors: Sharma, Bablu, Ritchie, Glen L.
• Format: Article
• Language: English
• ISSN: 0011-183X; 1435-0653
• DOI: 10.2135/cropsci2014.04.0310

ABSTRACT Rapid screening of plant growth provides additional phenological information for cotton (Gossypium hirsutum L.), which may then be linked to productivity analysis in breeding and agronomy. We tested automated measurements of plant height, ground cover fraction (GCF), normalized difference vegetation index (NDVI), and canopy temperature (Tc) using a ground‐based platform mounted on a research sprayer on four cotton cultivars (DP1212, DP1219, FM2484, and PHY499) grown under 10 different irrigation levels at Lubbock, TX. Measurements consistently identified all irrigation treatments throughout the season in 2012, a dry year, and the lower irrigation treatments in 2013, a wet year, as well as differences in growth habits among cultivars in both years. The coefficient of determination (r2) of each in‐season parameter had a positive correlation with lint yield during both years. Lint yield values were compared with in‐season measurements at each sampling date. All cultivars had higher correlations between in‐season measurements and lint yield in 2012 than in 2013 because of the increased rainfall in 2013. In 2012, canopy temperature had highest correlation with lint yield, followed by plant height, GCF, and NDVI. In 2013, GCF, NDVI, and plant height all had higher correlations than Tc with lint yield. All the methods showed value in determining growth and yield, but selection of methods for screening may require adaptations based on the growth environment of the study.