气候变化领域集成服务门户

North China Plain has undergone severe warming trends since the 1950s, but whether this trend is the same during different growth phases for crops remains unknown. Thus, we analyzed the non-linear changes in the minimum temperature (T-min), mean temperature (T-mean) and maximum temperature (T-max) using the Ensemble Empirical Mode Decomposition method during each growth stage of summer maize based on daily temperature data from 1960 to 2014. Our results strongly suggest that the trends and fluctuations in temperature change are non-linear. These changes can be categorized into four types of trend change according to the combinations of decreasing and increasing trends, and 8 fluctuation modes dominated by the fluctuations of expansion and shrinkage. The amplitude of the fluctuation is primarily expansion in the sowing-jointing stage and shrinkage in the jointing-maturity stage. Moreover, the temperature changes are inconsistent within each growth stage and are not consistent with the overall warming trend observed over the last 55 years. A transition period occurred in both the 1980s and the 1990s for temperatures during the sowing-tasseling stage. Furthermore, the cooling trend of the T-max was significant in the sowing emergence stage, while this cooling trend was not obvious for both T-mean and T-min in the jointing-tasseling stage. These results showed that temperature change was significantly different in different stages of the maize growth season. The results can serve as a scientific basis for a better understanding of the actual changes in the regional surface air temperature and agronomic heat resources.