Changes in maize traits and yield under the cultivar, environment and management interactions across China’s Maize Belt in the past two decades

Abstract

It is important to investigate the response and adaptation mechanisms of crop traits and yield to ongoing climate change under genotype (G), environment (E), and management (M) interactions in order to develop climate-resilient cultivars and agricultural systems. Here, using long-term trial data at six comprehensive agricultural experiment stations across China’s Maize Belt, we investigated the changes in maize traits and yield during 1998–2019 under the interactions of G×E × M. The first-order difference, multi-collinearity detection, Pearson correlation, stepwise linear and a machine learning method were employed to construct the best statistical regression models between maize traits and climate variables and disentangle the relative contributions of climate, cultivar and management to grain yield changes. The results showed a warming trend but a decreasing trend in solar radiation at all the investigated stations. Climate variations affected maize traits and subsequently grain yield, and there were trade-offs among maize traits. With local agricultural management practices, about 20.5–73.8%, 1.2–42.7% and 2.3–37.0% of the yield changes were associated with climate change, cultivar shifts and management, respectively. The stations with higher latitudes exhibited greater variations in yield due to climate variations, compared to the stations with lower latitudes. Wind speed, radiation, and rainfall was found to affect maize yield in Northeast China by affecting evapotranspiration, soil moisture and drought occurrence. The ideal cultivars were characterized by a relatively long grain-filling stage, large hundred-grain weight and number of rows. This study gains new insights into the response and adaptation mechanisms of maize traits and yield to ongoing climate change, providing evidence for designing cultivar ideotypes and optimizing agricultural management in different environments.