Fertilisation and weed control effects on yield and weeds in durum wheat grown under rain-fed conditions in a Mediterranean climate

Summary A field study was undertaken to examine the effect of fertiliser type and weed control system on grain yield, yield components and weed population in durum wheat ( Triticum turgidum. var. durum) grown in a 4-year rotation bare fallow–barley ( Hordeum vulgare )–vetch ( Vicia sativa )–wheat]. Fertilisation treatments were: no fertilisation, organic fertilisation with 2500 kg ha^?1 of compost (sheep manure and cereal straw) and chemical fertilisation with 100–60–60 kg ha^?1 of NPK. Weed-control treatments were: no control, herbicide, harrowing with long-flex spring tines and strip sowing with inter-row hoeing. Chemical fertilisation increased grain yield with respect to the other treatments, amongst which no significant differences were noted. Weed-control systems afforded no improvement in yield compared with controls; indeed, yields obtained using inter-row hoeing were lower. The weed population consisted of a large variety of species, of which the most important were Convolvulus arvensis and Polygonum aviculare . Application of weed-control systems reduced weed density. Herbicide was the most effective, reducing weed biomass by 80%, whilst harrowing and hoeing reduced weed biomass by 40% and 52% respectively. Inter-row hoeing does not appear to be a viable alternative to herbicides, when used as the sole weed control method in a non-diverse cropping system in Mediterranean climates. The lack of response to compost suggests a need for further long-term research.