Yield response of maize (Zea mays L.) to conservation agriculture cropping system in Southern Africa

The effect of different conservation agriculture (CA) systems on maize grain yield was studied across four countries in southern Africa. Maize yield data was obtained from plots under no-tillage as well as from conventionally tilled plots. Crop residues were retained in no-till plots, whereas they were removed from conventional tillage plots in line with current farmer practices. Rotations or intercropping systems with grain legumes were introduced at all sites. Fertiliser treatments were uniform across tillage treatments at each trial location but varied across countries, based on local fertilizer recommendations. Focus group discussions were conducted with farmers in the study sites to understand the constraints related to the successful integration of CA into the farming systems and to document farmers' perceptions about CA. In the majority of cases (80%), yield responses from a range of CA systems were greater than those of the conventional control plot at the respective site. In 20% of the cases there was a negative response to CA, due to lack of experience by farmers in the initial year, slow increase in soil fertility at the respective site and waterlogging in some years with too much rainfall. Yield advantages on two manual CA systems, planted with a dibble stick with sole maize and maize-legume intercropping in Malawi were 1152 kg ha(-1) and 1172 kg ha(-1), respectively. Animal traction CA systems had slightly smaller yield benefits (458 kg ha(-1) on a ripline seeded system and 761 kg ha(-1) on an animal traction direct seeding systems) as compared to a ploughed control treatment. Yield benefits increased with increasing years of practicing CA, highlighting the need to gain experience to master critical management steps such as timely planting, weeding, fertiliser application and crop harvest residue management. Yields from CA system responded better to increasing clay and silt content in the top soil and were more resilient to seasonal rainfall variability than conventional control treatments. Results suggest that the niche for CA in southern Africa is larger than expected although rainfalls regimes below 600 mm are challenging to sustain large maize biomass production to provide effective soil cover in CA systems. The success of CA implementation will largely depend on addressing critical challenges observed in the field, which will need adaptation of CA system to the site and farmer circumstances. (C) 2014 Elsevier B.V. All rights reserved.