TY - JOUR
T1 - Development and Validation of a Mechanistic, Weather-Based Model for Predicting Puccinia graminis f. sp. tritici Infections and Stem Rust Progress in Wheat
AU - Salotti, Irene
AU - Bove, Federica
AU - Rossi, Vittorio
PY - 2022
Y1 - 2022
N2 - Stem rust (or black rust) of wheat, caused by Puccinia graminis f. sp. tritici (Pgt), is a re-emerging, major threat to wheat production worldwide. Here, we retrieved, analyzed, and synthetized the available information about Pgt to develop a mechanistic, weather-driven model for predicting stem rust epidemics caused by uredospores. The ability of the model to predict the first infections in a season was evaluated using field data collected in three wheat-growing areas of Italy (Emilia-Romagna, Apulia, and Sardinia) from 2016 to 2021. The model showed good accuracy, with a posterior probability to correctly predict infections of 0.78 and a probability that there was no infection when not predicted of 0.96. The model’s ability to predict disease progress during the growing season was also evaluated by using published data obtained from trials in Minnesota, United States, in 1968, 1978, and 1979, and in Pennsylvania, United States, in 1986. Comparison of observed versus predicted data generated a concordance correlation coefficient of 0.96 and an average distance between real data and the fitted line of 0.09. The model could therefore be considered accurate and reliable for predicting epidemics of wheat stem rust and could be tested for its ability to support risk-based control of the disease.
AB - Stem rust (or black rust) of wheat, caused by Puccinia graminis f. sp. tritici (Pgt), is a re-emerging, major threat to wheat production worldwide. Here, we retrieved, analyzed, and synthetized the available information about Pgt to develop a mechanistic, weather-driven model for predicting stem rust epidemics caused by uredospores. The ability of the model to predict the first infections in a season was evaluated using field data collected in three wheat-growing areas of Italy (Emilia-Romagna, Apulia, and Sardinia) from 2016 to 2021. The model showed good accuracy, with a posterior probability to correctly predict infections of 0.78 and a probability that there was no infection when not predicted of 0.96. The model’s ability to predict disease progress during the growing season was also evaluated by using published data obtained from trials in Minnesota, United States, in 1968, 1978, and 1979, and in Pennsylvania, United States, in 1986. Comparison of observed versus predicted data generated a concordance correlation coefficient of 0.96 and an average distance between real data and the fitted line of 0.09. The model could therefore be considered accurate and reliable for predicting epidemics of wheat stem rust and could be tested for its ability to support risk-based control of the disease.
KW - black rust
KW - disease onset
KW - disease progress
KW - epidemiological modelling
KW - model evaluation
KW - black rust
KW - disease onset
KW - disease progress
KW - epidemiological modelling
KW - model evaluation
UR - http://hdl.handle.net/10807/230956
U2 - 10.3389/fpls.2022.897680
DO - 10.3389/fpls.2022.897680
M3 - Article
SN - 1664-462X
VL - 13
SP - 897680-N/A
JO - Frontiers in Plant Science
JF - Frontiers in Plant Science
ER -