Modeling the early phases of epidemics by Phakospora pachyrhizi in Brazilian soybean

Abstract

Asian soybean rust, caused by the biotrophic basidiomycete Phakospora pachyrhizi, is a foliar disease that often causes considerable damage to soybean crops. The purpose of our work was to create a mechanistic model that can reliably represent epidemics of ASR in commercial soybean fields in Brazil. The most important inputs for the model are weather data (observations and forecast) and the initial observation of disease (or uredospore arrival). Our focus is on the first two or three cycles of infection after immigration into a soybean field. The model includes state variables for latent, infectious and senesced lesions, disease severity, uredospores, and soybean leaf area. Processes modeled include maturation through the latent and infectious periods, germination, sporulation, and processes affecting uredospores in the canopy. The model results were tested against field observations from trials at four locations in Brazil for the 2019/2020 growing season. The predictions generally matched the daily dynamics of disease progress in the field trials. The predictions reproduced the observed severity well with R2 value of 0.84. This high correlation indicates that our model is accurate enough to be used as a tool to predict the dynamics of ASR epidemics during the first few cycles after uredospore invasion into a soybean field. A sensitivity analysis was performed that showed that the model is sensitive to time and duration of the initial spore arrival. This indicates that spore traps or other observations should measure not only the first day of arrival but also subsequent days.