Global reallocation of rainfed crops can boost production and reduce climate risk

The global food system faces escalating risks to the production of major rainfed agricultural crops. Here we used modern portfolio theory to explore Pareto-optimal spatial arrangements of global rainfed cropping that explicitly account for trade-offs between total crop production and its interannual variability (risk). We analysed production/risk trade-offs under recent historical climate and yields for 2010, as well as under business-as-usual (RCP 6.0) and ambitious mitigation (RCP 2.6) climate futures for 2050. We found that optimizing the spatial distribution of rainfed crops could increase global production by 10.1% while maintaining the same level of risk as in 2010 or, alternatively, reduce production variability by 33.1% while maintaining the same level of production level as in 2010, all without additional land or water. Optimal global rainfed cropping patterns could meet future food demand with enhanced resilience under increased climate variability but only with yield-gap closure and more open and globalized trade, highlighting the need for coordinated production strategies, strengthened trade cooperation and sustainable intensification within climate adaptation policy.