气候学
中国
环境科学
气候变化
生态系统
全球变暖
地理
气候模式
极端气候
降水
气象学
全球变化
中国南方
强度(物理)
陆地生态系统
地球系统科学
自然地理学
大气环流模式
基线(sea)
气候系统
作者
Rui Zhao,Xiao Xie,Yanling Chen,Hong Wang,Jing Li
摘要
ABSTRACT Climate change has increased heatwave frequency, duration, and intensity globally, which is expected to worsen under warmer conditions. Understanding heatwave variabilities is crucial for the ecosystem and human health they impact, yet comprehensive assessments over China at different spatiotemporal scales remain limited. This study evaluates 12 NASA Earth Exchange Global Daily Downscaled Projections' CMIP6 (NEX‐GDDP‐CMIP6) downscaled data in simulating historical heatwaves (1981–2010) and their projections under SSP2‐4.5 and SSP5‐8.5 scenarios for near‐term (2031–2060) and late‐term (2071–2100) periods, using CN05.1 observations and key heatwave metrics. The NEX‐GDDP‐CMIP6 ensemble mean (MME) outperforms individual models in simulating heatwave frequency (HWF) and duration (HWD), with 65%–87% of China showing biases within ±10%. While the MME effectively captures HWF trend patterns across China, it fails to simulate decreasing HWD trends over southern China and intensity metrics trends over the Tibetan Plateau. Among the individual models, CMCC‐ESM2, GFDL‐ESM4, and GFDL‐CM4 emerge as the three best‐performing models. Under SSP5‐8.5, about 87% (74%) of regions will experience increases in HWF (HWD). Notably, heatwave frequency over China will exceed 15 events annually during 2071–2100. The entire country will experience intensification of heatwave intensity with larger increments in high‐latitude regions. Furthermore, the likelihood of extreme heatwave events under SSP5‐8.5 is projected to be double under SSP2‐4.5. Northern China faces the highest risk of extreme heatwaves with over 50% probability under late‐term SSP5‐8.5, followed by Southeastern China. Our results offer comprehensive insights into NEX‐GDDP‐CMIP6 simulated heatwave variations, necessary for refining global climate models and understanding future risks.
科研通智能强力驱动
Strongly Powered by AbleSci AI