Dominant sources of uncertainty in simulating maize adaptation under future climate scenarios in China

The potential of climate adaptation has been widely investigated with a climate-crop modeling approach. Although different sources of uncertainty in projected crop yields have been quantified in climate change impact assessments, uncertainty in simulating the crop adaptation to future climate has not been fully assessed. The objective of this study was to determine the uncertainty in simulating maize adaptation to future climate change with two adaptation options (adjusting planting date and shifting cultivars) at four contrasting sites across China's Maize Belt. Maize yield with adaptation was simulated using three crop models (APSIM, DSSAT-CERES, and STICS) driven by 22 global climate models (GCMs) under four emission scenarios of future societal development pathway (SSP126, SSP245, SSP370, and SSP585) during two periods (2040–2069 and 2070–2099). We found that late planting had a greater potential to cope with climate change at most study sites. However, all sites required new cultivars with increased thermal time requirements. Under optimum management options at the four study sites, rainfed maize yields were likely to increase by 1.9%–68.3% compared with yields obtained without adaptation. For the adaptation simulation using adjusted planting date alone, GCM was the major source of uncertainty, accounting for 22.9%–36.7% of the total uncertainty at all sites except a high-altitude site where changing planting time was the major source of uncertainty (32.4%). For the adaptation simulation using shifting cultivar alone, crop model was the dominant source of uncertainty, accounting for 24.0%–38.0% of the total uncertainty at all sites except a high-latitude site where shifting cultivar was the major source of uncertainty (34.0%). These findings demonstrated that adaptation options have great potential for increasing maize yields, and the major source of uncertainty depends on study sites and adaptation type used. The results of this study advance the understanding of the dominant sources of uncertainty in crop yield under different climate adaptations, thereby improving our confidence in assessments of future climate impact on maize yields determined by different adaptation strategies.