Fungicide treatments in winter wheat: The probability of profitability

The marginal return of fungicide applications against leaf blotch diseases in winter wheat fields were analysed based on data from field trials performed in 1996–2011 in Sweden. Yield increases from fungicide treatments were compared with data on growing conditions (precipitation, previous crop, nitrogen fertilization, soil type, and disease severity) using logistic regression. After identification and quantification of single factors with good predictive ability, multiple factor models were analysed. A model with five factors; rain days in April/May and three weeks before ear emergence, disease severity at ear emergence, soil type and previous crop, identified situations when a fungicide treatment gave a positive marginal return. The sensitivity and specificity of the model was evaluated by using receiver operating characteristic (ROC) curves. •The marginal return of fungicide application varies with weather and crop management.•Logistic regression is a useful tool in the analysis of results from field trials.•Evaluation with ROC curves shows the sensitivity and specificity of models.