Zhang Yongxiang, Wang Hongli, Shao Xuemei, Liu Hongbin, Zhu Haifeng, Wang Lily, Liang Eryuan, Hao Zhixin, Fang Xiuqi, Zhang Qibin, Li Jinjian, Shi Xinghe, Yuan Yujiang, Chen Feng, Zhang Tongwen, Zhang Ruibo, Shang Huaming, Peng Jianfeng
National Climate Centre, China Meteorological Administration, Beijing 100081, China.
Key Laboratory of Land Surface Pattern and Simulation, Institute of Geographic Sciences and Natural Resources Research, Chinese Academy of Sciences, Beijing 100101, China; Beijing New Epoch School, the Administrative Bureau of Chinese Academy of Sciences, Beijing 100086, China.
Sci Bull (Beijing). 2024 Sep 15;69(17):2756-2764. doi: 10.1016/j.scib.2024.06.034. Epub 2024 Jun 28.
Extreme drought events have increased, causing serious losses and damage to the social economy under current warming conditions. However, short-term meteorological data limit our understanding and projection of these extremes. With the accumulation of proxy data, especially tree-ring data, large-scale precipitation field reconstruction has provided opportunities to explore underlying mechanisms further. Using point-by-point regression, we reconstructed the April-September precipitation field in China for the past ∼530 years on the basis of 590 proxy records, including 470 tree-ring width chronologies and 120 drought/flood indices. Our regression models explained average 50% of the variance in precipitation. In the statistical test on calibration and verification, our models passed the significance level that assured reconstruction quality. The reconstruction data performed well, showing consistency and better quality than previously reported reconstructions. The first three leading modes of variability in the reconstruction revealed the main distribution modes of precipitation over China. Wet/drought and extremely wet/drought years accounted for 12.81%/10.92% (68 years/58 years) and 1.69%/3.20% (9 years/17 years) of the past ∼530 years in China, respectively. Major extreme drought events can be identified explicitly in our reconstruction. The detailed features of the Chongzhen Great Drought (1637-1643), the Wanli Great Drought (1585-1590), and the Ding-Wu Great Famine (1874-1879), indicated the existence of potentially different underlying mechanisms that need further exploration. Although further improvements can be made for remote uninhabited areas and large deserts, our gridded reconstruction of April-September precipitation in China over the past ∼530 years can provide a solid database for studies on the attribution of climate change and the mechanism of extreme drought events.
极端干旱事件有所增加，在当前气候变暖的条件下给社会经济造成了严重损失和破坏。然而，短期气象数据限制了我们对这些极端事件的理解和预测。随着代用资料的积累，尤其是树木年轮数据，大规模降水场重建为进一步探索潜在机制提供了机会。利用逐点回归方法，我们基于590条代用记录重建了中国过去约530年4月至9月的降水场，其中包括470个树木年轮宽度年表和120个旱涝指数。我们的回归模型解释了降水变化的平均50%。在校准和验证的统计检验中，我们的模型通过了确保重建质量的显著性水平。重建数据表现良好，显示出一致性且质量优于先前报道的重建结果。重建中前三个主要变化模态揭示了中国降水的主要分布模式。在中国过去约530年中，湿润/干旱年和极端湿润/干旱年分别占12.81%/10.92%（68年/58年）和1.69%/3.20%（9年/17年）。在我们的重建中可以明确识别出主要的极端干旱事件。崇祯大旱（1637 - 1643年）、万历大旱（1585 - 1590年）和丁戊奇荒（1874 - 1879年）的详细特征表明存在可能不同的潜在机制，需要进一步探索。尽管对于偏远无人居住地区和大沙漠还可以进一步改进，但我们对中国过去约530年4月至9月降水的网格化重建可为气候变化归因研究和极端干旱事件机制研究提供坚实的数据库。
