Submitted to: Meeting Abstract
Publication Type: Abstract Only
Publication Acceptance Date: 11/13/2019
Publication Date: N/A
Citation: N/A

Interpretive Summary:

Technical Abstract: Oat panicles of landraces, old cultivars, and newly developed germplasm were sampled at physiological maturity from a total of about 750 experimental plots during 2014-2018 growing seasons. The duration of vegetative and reproductive phenological stages were digitally documented and contrasted with the classical Zadoks classification system. The architecture and structure of five random panicles per plot were digitally documented under fully-illuminated laboratory conditions, then converted into measurements using the fractal dimension and hull and circle modules in ImageJ software. A database was constructed using phenotypic measurements per year, germplasm entry and sample. The extensive measurements (about 20 per panicle) describing the architecture and structure of the panicles were used to estimate their phenotypic plasticity and relationships to the duration of vegetative and reproductive growth stages expressed in growing degree days. Then they were used to predict groat yield and yield components over years. A Bayesian clustering method and structural equation modeling, respectively, identified phenotypically discrete and statistically unique subpopulations in relation to their panicle architecture, structure; and their direct and indirect relationships with phenological growth stages and their effects on groat yield. Of special interest are genotypes with unique three-way functional relationships between phenotypic panicle structure, duration of growth phases and groat yield.