Proposal for a new standard area diagram to quantify the severity of orange rust in sugarcane.

Abstract

Orange rust (Puccinia kuehnii), a fungal disease of major concern in the sugarcane industry, can reduce the productivity of susceptible sugarcane varieties by up to 50% in many sugarcane-producing countries. The use of a standard area diagram (SAD) enhances the assessment of rust severity, improving the accuracy and consistency of disease quantification by evaluators. The present study aimed to develop and validate a new SAD for assessing the severity of orange rust in sugarcane. The SAD was constructed from symptomatic leaves collected in the field, with severity levels of 0, 1.5, 3.8, 7.3, 12.9, 20.7, 31.5, 44.4, and 63.2%. Validation was conducted with 30 evaluators with no prior experience, divided into two groups of 15. Disease severity was estimated from 100 leaf images, initially without the use of a SAD and subsequently with the aid of a SAD. One group used the proposed SAD, while the other used an old SAD for orange rust assessment. The proposed SAD reduced inter-evaluator variability, showed lower systematic bias, and had narrower limits of agreement, although Lin's concordance correlation coefficient (CCC, ρc) remained low, suggesting that factors beyond the SAD itself - such as the intrinsic complexity of orange rust symptoms and evaluator-related factors - may influence visual assessment accuracy. Analysis of bias parameters showed significant reductions in location (u) and scale (v) biases, indicating higher reproducibility and accuracy of the proposed SAD.