Comparative evaluation of nonlinear disease progress models for quantitative assessment of peanut leafspot disease

Abstract

Leafspot disease causes up to 70% yield losses in peanut production areas across the globe. One way to curb this menace is to use leafspot resistant peanut ( Arachis hypogaea L.) varieties. In breeding for disease resistance, quantitative characterization of disease dynamics, including initial inoculum ( y ₀ ) and apparent infection rate ( r ), is essential for improving disease assessment, resistance evaluation, and the development of future decision‐support systems in agriculture. This study evaluated the performance of commonly used nonlinear models for characterizing leafspot disease incidence (DI) and severity in peanut genotypes. Field‐based time‐series data on leafspot DI, early leafspot (ELS) severity, and late leafspot (LLS) severity were collected on 10 peanut genotypes over two growing seasons. Disease progress was analyzed using both linearized and nonlinear forms of the Gompertz, logistic, monomolecular, and exponential models. Model‐derived parameters revealed clear genotypic differences: resistant genotypes (Nkatiesari, Sarinut‐1) exhibited low y ₀ and r , moderately resistant genotypes (L010A1, L027B, L076J) showed intermediate dynamics, and susceptible genotypes (L030, L046, L104B, Chinese, Sarinut‐2) displayed rapid epidemic amplification. The Gompertz and exponential models generally provided the best fit for DI and ELS severity, whereas LLS severity exhibited greater interannual variability, with logistic and monomolecular models performing best in different years. These findings provide a comparative modeling framework to support the quantitative evaluation of crop disease dynamics and highlight the value of nonlinear disease progress models for agricultural disease assessment and resistance evaluation.