Tillage practices affect weeds differently in monoculture vs crop rotation

Reduced tillage practices are widely considered to be more sustainable than conventional tillage practices, but many producers remain reluctant to reduce tillage due to difficulties controlling weeds. Crop rotation is often put forward as the best means to manage weeds in reduced tillage systems, but uncertainties remain around how different tillage practices and crop rotations interact. Here, we assess the effects of four different tillage practices on weed seedbank density and composition in wheat (Triticum aestivum) monoculture (WWWW), and two different rotations, wheat-medic-wheat-medic (annual medic, Medicago spp.; WMWM), and wheat-canola-wheat-lupin (Brassica napus, Lupinus spp.; WCWL). We use data across a whole four-year rotation period from a long-term experiment replicated at two sites in South Africa’s winter rainfall region. The four tillage practices assessed follow a gradient of soil disturbance: conventional tillage (CT, soil inversion through ploughing), minimum tillage (MT, shallow soil loosening), no tillage (NT, direct drilling with tine openers) and zero tillage (ZT, direct drilling with disc openers). Our results indicate that tillage type had no effect on weeds within the wheat monoculture. Both crop rotations generally had lower weed densities and reduced dominance of grass weeds than the monoculture, but under ZT weed seed bank density in both rotations was similar to that found in monoculture. Thus the use of ZT with crop rotation is antagonistic in this system, possibly due to more limited chemical weed control options than in CT, MT and NT, or due to crop residue cover promoting weed establishment. Subsequently, we recommend that producers in the region seeking to reduce tillage opt for NT rather than ZT, and avoid a wheat monoculture. Weed researchers and agronomists should be wary of other such antagonistic interactions between weed management practices in different systems.