Maximizing realized yield by breeding for disease tolerance: A case study for Septoria tritici blotch

Disease-tolerant cultivars maintain yield in the presence of disease. When disease intensity is high, they can improve a grower's net return compared to less tolerant cultivars. Many authors report a trade-off, whereby higher fully protected yields are correlated with a lower disease tolerance. We analyse the question for breeders: to what extent should they breed for tolerance when it compromises maximizing fully protected yield? Field trials with 147 progeny from five parental crosses of wheat were used to measure yield and tolerance under a range of disease intensities from Septoria tritici blotch (STB; causal organism Zymoseptoria tritici) at a range of sites and seasons. The data define the variation for these traits from which breeders can select. A simple data-driven descriptive model was used to calculate the combination of tolerance and fully protected yield that maximizes actual yield for any given level of disease-quantified by loss of healthy canopy area duration (HAD-loss). This model was combined with data on the year-to-year variability of HAD-loss in the UK to calculate the tolerance and fully protected yield that maximizes the mean actual yield. We found that even when an effective fungicide treatment programme is applied, breeding for tolerance increases the mean actual yield. Some commercially available cultivars were found to have a level of tolerance that leads to yields close to the maximum yield in the presence of disease, others had a lower tolerance leading to suboptimal yields.