Effects of rate and timing of nitrogen fertilizer on disease control by fungicides in winter wheat. 1. Grain yield and foliar disease control

The effects of nitrogen (N) rate and timing on need for fungicide application in winter wheat (Triticum aestivum) were investigated in 3 years of field experiments on loamy sand soils in Denmark. A two-factor completely randomized experimental design was used, comprising seven combinations of different N fertilizer rates and application times, and five doses of fungicide (co-formulation propiconazole and fenpropimorph). Two different varieties of winter wheat with high susceptibility, to powdery mildew (Blumeria graminis) were used, Florida in the first season and Pepital in the last two seasons. The severity of powdery mildew and septoria leaf spot (mainly Septoria tritici) varied between seasons from slight to moderate with powdery mildew dominating in the first season and septoria leaf spot in the last season. The severity of both powdery mildew and septoria leaf spot assessed as the Area Under the Disease Progress Curve (AUDPC) was increased by application of N in all years, and more so by early applied N. Grain yields increased with increasing N rate and fungicide dose. However, the observed grain yields did not reveal any N x fungicide interactions. Regression models were therefore fitted, relating grain yield to rate and timing of N fertilizer and to AUDPC of powdery mildew and septoria leaf spot, and relating AUDPC to rate and timing of N fertilizer and to fungicide dose. They demonstrated that septoria leaf spot had a considerably higher impact on grain yield than mildew. The optimal fungicide dose and N rate were defined as those giving the highest economic return. The regression models were used to estimate the effect of N rate and timing on optimal fungicide dose, and the effect of fungicide application on optimal N rate. The optimal fungicide dose increased almost linearly with N rate above a minimum N rate, but with a large dependency on price relations. Early applied N caused a higher demand for disease control. The fungicide applications in the model were mainly driven by the need to control septoria leaf spot, whereas powdery mildew gave a poor net return for control. The estimated optimal N fertilizer rate for untreated diseased crops was 60 kg N/ha lower than for crops without disease. The use of fungicides with an efficacy twice that of the EBI-fungicides used in this experiment would increase the optimal N rate by c. 20 kg N/ha.