Comparing the Effects of Rust on Plot Yield, Plant Yield, Yield Components, and Vegetative Parts of Soybean

Authors:

Abstract
To determine the plant variable for coupling a disease model to a soybean growth model, effects of soybean rust (caused by Phakopsora pachyrhizi) on soybean were quantified at four levels: vegetative parts, yield components, plant yield, and plot yield. Path analysis was conducted to determine the most correlated components to disease in each level. The responses of vegetative parts, such as shoot weight and branches/plant, to disease effects were inconsistent among cultivars. Yield components responded consistently to disease effects among the cultivars, indicated by the high values of correlation coefficients between disease and the components. Path coefficients indicated that reduction of 100‐seed weight was the most important cause of yield loss in this level. No significant within‐plant compensation was detected using path coefficient analysis except for that between pod length to pod width. The most consistent response to disease effect was at population level. Correlation of disease to plant yield was as significant as these to plot yield. Correlations between disease and these plant levels were in the order of plot yield, plant yield, yield components, and vegetative parts. Regression between plot yield and green leaf area duration (GLAD) from flowering to seed full size was highly significant (P < 0,001). This study suggests that green leaf area at population level was the best variable for coupling disease model to crop model.