Estimating yield in commercial wheat cultivars using the best predictors of powdery mildew and rust diseases

IntroductionThis four-year research determined the best predictors of black, brown and yellow rusts and powdery mildew development in different wheat cultivars and planting dates across 282 experimental field plots. MethodsParameters estimated by exponential (for black rust and powdery mildew) and Gaussian (for brown and yellow rusts) models, area under disease progress curve (AUDPC), and maximum disease severity were considered as disease progress curve elements. Factor analysis determined the most predictive variables among 19 indicators in order to describe wheat yield. ResultsAccording to principal component analysis (PCA), 11 selected wheat diseases and yield predicators accounted for 60% of total variance in datasets. This PCA test described four principal components involving these selected predictors. Next, multivariate regression model, which developed according to four independent principal components, justified a noticeable part of yield variability over and within growing seasons. DiscussionPresent findings may improve accuracy of future studies to examine seasonal patterns of powdery mildew and rusts, predict wheat yield and develop integrative disease management programs.

Automated summary

This four-year research identified the best predictors of wheat diseases and yield across different cultivars and planting dates. By using exponential and Gaussian models, as well as factor analysis and multivariate regression, the study provided valuable insights for disease management and yield prediction in wheat cultivation.