Maize yields in varying rainfall regimes and cropping systems across Southern Africa: A modelling assessment

Abstract

Rainfall variability, which ultimately leads to climate change, is a major threat to smallholder agriculture. It affects time of sowing time and productivity, amongst other challenges. There is therefore need to evaluate the different strategies for their effectiveness in managing climate variability. This study assessed the effects of different strategies on sowing date, season length and maize yields under variable rainfall conditions. Maize (Zea mays L.) yield simulations for Southern Africa were conducted using the DSSAT model. Simulated conservation agriculture (CA)-based cropping systems included basins prepared early (CA-Basins early) and late (CABasins late), draught powered planter (CA-Direct seeder), ripper (CA-Ripper) and Dibble stick (CA-Dibble). Conventional systems were mouldboard ploughing early (CMP-early) and late (CMP-late). Rainfall seasons were classified into low, medium and high based on the total rainfall amount. Results showed that high-rainfall seasons were seeded earlier and had a greater season length compared to low rainfall seasons in drier agro-ecologies, translating to higher yields and vice versa. Reduced labour requirements and use of draught power, enabled early seeding of CA-ripper, direct seeder, basins early and CMP-early systems compared to CA-Basins late, Dibble stick and CMP-late systems. However, performance of cropping systems did not vary across season types suggesting that there was thus no evidence of higher yield advantages from CA technologies even during low rainfall seasons. This puts the merits of drought mitigation by CA technologies into doubt despite enabling early planting.