Durum wheat quality under Mediterranean conditions as affected by N rate, timing and splitting, N form and S fertilization

Three different experiments were designed to study the effects of N fertilizer rate, timing and splitting, and the response to combined application of N and S fertilizer on the quality of durum wheat (Triticum turgidum L.) over a 3-year period in Vertisols under rainfed Mediterranean conditions. The following parameters were analyzed: grain yield, test weight, grain protein concentration, gluten index, yellow index, ash content and vitreousness. The N rate experiment included rates of 0, 100, 150, and 200 kg N ha−1 carried out on 4 different sites. The experiment was designed as a randomized complete block with four blocks. For the experiment on N timing and splitting, a single rate of 150 kg N ha−1 was used, different fractions being applied at sowing, tillering and stem elongation, at a single site; again, experimental design was a randomized complete block with four blocks. Finally, for the experiment on the response to combined application of N and S fertilizer, a single fertilizer dose of 150 kg N ha−1 was applied in two forms (urea + ammonium nitrate and urea + ammonium nitrosulfate) with one leaf application at ear emergence (zero, 25 kg S ha−1, 25 kg N ha−1, 25 kg S ha−1 + 25 kg N ha−1, and 50 kg N ha−1), also at a single site, using a split-plot design with four replications. Year-on-year variation in rainfall led to marked variations in wheat yield, grain protein concentration and quality indices. Rainfall and temperatures during grain formation and ripening (April and May) were critical for durum wheat quality. N fertilizer rate modified the behavior pattern of some quality indices as a function of climate, and had a more consistent effect on durum wheat quality than on grain yield. The highest values for grain yield were recorded at an N rate of 100 kg ha−1, while maximum grain protein concentration values were recorded at 200 kg ha−1. Increasing N rates also prompted an increase in vitreous kernel count, which was highest at 150 kg N ha−1. The timing and splitting of N fertilizer had no clear effect on either grain yield or quality indices. Leaf application of N at ear emergence increased only grain protein concentration, vitreous kernel count and grain ash content. Soil or leaf application of S had no effect on quality indices, with the exception of ash content, which rose with leaf application at ear emergence. The response of grain yield and grain protein concentration to fertilizer N differed from that reported for temperate climates.