Unraveling the contribution of G x E x M interactions to maize productivity in Central Mozambique amid climate change

Maize ( Zea mays L.) is the most important cereal staple in sub-Saharan Africa. Yet, smallholders only produce about 20% of what is biophysically possible with agronomic best practices. On-farm researcher- and farmer-managed experiments were conducted in two consecutive cropping seasons in Central Mozambique to disentangle the relative contribution of genotype-by-environment-by-management (G x E x M) interactions to smallholder maize yields in relation to improved varieties, sowing time, and fertilization regime. In the 2022-2023 cropping season, maize yield variability on fertile fields was explained by a three-way interaction between fertilizer regime, sowing date, and variety, whereas on infertile fields, it was explained by fertilizer regime only. The highest yields were obtained with early sowing of a medium- (3.6 t ha ^-1 ) or early- (3.1 t ha ^-1 ) duration variety, whereas late sowing yielded the least (1.6 t ha ^-1 ). In the 2023-2024 cropping season, characterized by El Nino-induced drought, yield variability on fertile fields was explained by variety, with the highest yield associated with the medium-duration variety, and on infertile fields by fertilizer regime. On farmer-managed experiments, maize yield variability was attributed to variety and fertilizer. Profitability with the improved varieties and mineral fertilizers tested depended upon soil fertility and the magnitude of water limitation on maize growth, with trade-offs arising between food security and return on investment in improved genetics and agronomic management. Targeting medium-duration maize varieties in combination with early sowing and nutrient inputs was found to be critical for food security amid climate change.