摘要
干旱会引发植被绿度下降、农作物减产、生物栖息地破坏等诸多问题,对人类和自然环境均具有显著影响.以植被状态指数作为干旱指标,利用极点对称模态分解 EPSMD方法对 1999～2019 年全国植被状态指数序列进行周期与趋势识别,采用网格化趋势检验法揭示网格化的干旱趋势特征,运用多变量小波相干方法探讨干旱与多种遥相关因素的内在关系.结果表明,1999～2019 年基于植被状态指数的全国干旱大体呈减轻趋势,植被状态指数的线性倾向率为 0.044/10a;不同分区的干旱变化特征各不相同,其中黄土高原区的干旱减轻趋势最为明显,植被状态指数线性倾向率为 0.215/10a;网格尺度的植被状态指数趋势特征Z 均值在春季、夏季、秋季、冬季分别为 0.41、0.79、0.50、0.11,各季度旱情均呈减轻态势;遥相关因素组合PNA-ENSO对全国干旱的影响最为明显.未来可将PNA、ENSO 作为干旱预警输入因子提高干旱预报精度.
Abstract
Drought will cause vegetation greenness decline,crop yield reduction,biological habitat destruction and other problems,which have a significant impact on human and natural environment.In this paper,the vegetation condi-tion index(VCI)was adopted as a drought indicator,and the extreme-point symmetric mode decomposition(EPSMD)method was used to identify the periodicity and trend of the VCI series from 1999 to 2019 across China.Subsequently,the gridded trend test method was applied to reveal the drought trend characteristics at the grid scale.The multivariate wavelet coherence technique was utilized to explore the internal relationship between drought and various teleconnection factors.The results indicate that the VCI-based drought showed a decreasing trend during 1999-2019 across China,with a linear tendency rate of VCI of 0.044/10a;The characteristics of drought were different in various sub-regions,among which the drought mitigation trend was the most obvious in the Loess Plateau with a linear tendency rate of VCI of 0.215/10a;The gridded average Z-values of VCI trend characteristics at the grid scale were 0.41,0.79,0.50,and 0.11 in spring,summer,autumn,and winter,respectively,indicating that drought showed a decreasing trend in each season;The combination of remote correlation factors PNA-ENSO has the most obvious impact on drought in China.PNA and ENSO can be used as drought warning input factors to improve the accuracy of drought forecasting.
基金项目
中国水科院开放基金(IWHR-SKL_KF202207)
国家自然科学基金(52179015)
NETL
NSTL
维普
万方数据