Abstract:

In the context of global climate change, the frequency, intensity, and duration of extreme climate events are increasing and strengthening, which greatly affects agricultural production. The three provinces in Northeast China are the main maize-producing areas in the country, and the region most significantly affected by climate change. It is crucial to explore the effects of extreme climate events on maize meteorological yield in the three provinces and safeguard China's food security and economic development. In the current study, a machine learning model was constructed based on the historical meteorological data and statistical maize yield data to clarify the impact extreme climate events on maize meteorological yield in northeast China during the historical (1981−2014) and future (2031−2060) periods. The results showed that high temperature and high- temperature-drought compound events had the greatest impact on maize meteorological yield during the historical period, with meteorological yield decreasing by 13.2% and 15.9%, respectively. Meanwhile, the extreme temperature events had a greater impact on maize meteorological yield compared to extreme precipitation events. In the future, the climate show a warming trend, Compared with the SSP1−2.6 (low−emission) scenario, the magnitude of maize meteorological yield reduction in Northeast China under the SSP5-8.5 (high-emission) scenario is more pronounced, and more attention needs to be paid to the impact of extreme precipitation events on maize meteorological yield in the future.

Key words:

Northeast China, Maize meteorological yield, Machine learning, Extreme temperature events, Extreme precipitation events