COVER CROPS in South African Vineyards

CHAPTER 3 COVER-CROP SELECTION

Olifants River valley Trials in the Olifants River valley were conducted at the ARC Infruitec-Niet­ voorbij research farm near Lutzville. In this semi-arid region, a cover crop can only be cultivated in the grapevine work row under full-surface sprinkler or micro-sprinkler irrigation. Open-land trials Twenty-six cover-crop species and cultivars were grown in open fields under full-surface micro-sprinkler irrigation.The seeding density is summarised in Table 3.1. The cover crops were sown annually. They received 18 millimetres of water per week for the first ten weeks after sowing, and 18 millimetres every two weeks thereafter. Rainfall was supplemented by full-surface sprinkler irrigation until chemical control was applied at the end of August. Dry-matter production and weed suppression were assessed for two years. Results are summarised inTable 3.5.All the grasses except barley and perennial phalaris (bulbous canary grass) produced 5.0 tonnes per hectare or more dry matter, and restricted winter-growing weeds to less than 0.5 tonnes per hectare. Grazing vetch, Parabinga bur medic, Paraggio bur medic, and pink serradella all produced 5.9 tonnes per hectare or more dry matter, which restricted winter-growing weeds to 0.72 tonnes per hectare or less, and is enough to supply a weed-suppressing mulch for the grapevine-growing season. In contrast with grasses, the dry-matter production of legumes did not al- ways correlate well with weed suppression, suggesting that the growth habits of legumes may play a role in weed suppression. Vineyard trials Selected annual grasses (rye, black oats, Overberg oats, and triticale) and legumes (grazing vetch, Parabinga bur medic, Paraggio bur medic, and pink serradella) were evaluated in a vineyard. Each cover crop was sown annually in the work row. Full-surface chemical weed control was applied from just before bud-break until harvest. All cover crops except triticale were assessed from 1993–2002.Triticale was assessed from 1993–1998, after which it was replaced with perennial dwarf fescue ( Festuca arundinacea ).

68