Innovative approach to diagnosing the development of net blotch in winter barley

Abstract

The article demonstrates the possibility of applying intelligent digital technologies to forecast the development of net blotch disease in winter barley caused by Pyrenophora teres. The obtained intelligent solution in the form of a binary decision tree has the ability to determine scenarios of net blotch development into three classes: depressive, moderate, and epiphytotic. To effectively train the model during the period from 2021 to 2023, field and laboratory experiments were conducted on the research sites of the Federal Research Center of Biological Plant Protection. The experiments collected data on the degree of leaf damage, the type of resistance of the variety, the growth phase at which primary infection occurred, and the average relative air humidity during the vegetation period in which infection occurred. The total sample size was 249 observations. The trained model demonstrated high classification accuracy on both the training and test sets, with an accuracy rate exceeding 0.96 and 0.92, respectively. The study shows that the innovative method discussed in the article for forecasting the development of net blotch in winter barley can be successfully integrated into the overall strategy of phytosanitary diagnostics of grain crops, including barley.