Population growth and economic development in China has increased the demand for food and animal feed, raising questions regarding China’s future maize production self-sufficiency. Here, we address this challenge by combining data-driven projections with a machine learning method on data from 402 stations, with data from 87 field experiments across China. Current maize yield would be roughly doubled with the implementation of optimal planting density and management. In the 2030 s, we estimate a 52% yield improvement through dense planting and soil improvement under a high-end climate forcing Shared Socio-Economic Pathway (SSP585), compared with a historical climate trend. Based on our results, yield gains from soil improvement outweigh the adverse effects of climate change. This implies that China can be self-sufficient in maize by using current cropping areas. Our results challenge the view of yield stagnation in most global areas and provide an example of how food security can be achieved with optimal crop-soil management under future climate change scenarios.

Drought is viewed as a naturally recurring phenomenon in many Australian agricultural systems. Identifying regional changes in frequency and severity of drought induced by climate change is required to develop regionally specific adaptation strategies. In this study, we provided a first look at the impacts of climate change on 21st century drought characteristics over the New South Wales wheat belt of southeastern Australia. These impacts were assessed from an ensemble of 28 statistical downscaled global climate models under representative concentration pathway (RCP8.5). A modified relative standardized precipitation and evapotranspiration index (rSPEI) at the seasonal scale (3 months) was used to analyse temporal and spatial changes in drought. Results indicated that there was a tendency towards more frequent and severe winter–spring droughts over the study area. Moreover, winter–spring drought prone areas were expected to expand from west to east. Until the end of the 21st century, more than half the wheat belt would be vulnerable to winter–spring drought. The combined effects of reduced precipitation and increased temperature during future winter and spring seasons were the main reasons causing these changes of drought. In addition, summer and autumn droughts would have both slight temporal and spatial changes across the study region. This study also revealed that traditionally dry areas would likely experience an increased frequency of drought compared to wetter areas when subjected to a same increase in temperature or decrease in precipitation. Furthermore, the western part of the wheat belt might be unsuitable for winter crops in the future, or at least exposed to an increased risk of variable yield and would require a gradual transformation which might include summer crops or pastures. Investments in cropping land should be focused on the east part of the wheat belt to achieve more consistent financial returns.