论文部分内容阅读
利用Hurst指数和小波分析工具对中国农作物干旱灾情的1950~2010年的变化动态作了分析,揭示了灾情序列的趋势持续性、特征时间尺度、周期特征和突变点,同时还定性地预测了农作物干旱灾情的未来走势。研究结果发现:1)根据Hurst指数计算结果,过去60年农作物干旱灾情变化有着很好的自相似性,受灾面积时间序列的Hurst指数为0.793;成灾面积和粮食损失时间序列的Hurst指数分别为0.8634和0.9449,过程均具有很强的正持续性。2)中国农作物干旱灾情都存在多时间尺度特征,不同的时间尺度表现为不同的循环交替,大尺度的周期变化嵌套着小尺度的周期变化。3)小波分析显示,受灾面积在6a和13a上分别存在着第1和第2主周期;成灾面积在6a和18a上分别存在着第1和第2主周期;粮食损失在8a和18a上分别存在着第1和第2主周期。
The Hurst index and wavelet analysis tools were used to analyze the dynamics of crop drought in China from 1950 to 2010, revealing the trend continuity, characteristic time scales, periodic characteristics and mutation points of the disaster series, and also qualitatively predicted the crop Drought disaster in the future trend. The results show that: 1) According to the calculation results of Hurst index, the change of crop drought over the past 60 years has a good self-similarity, Hurst exponent of time series of affected area is 0.793; Hurst exponents of disaster area and food loss time series are 0.8634 and 0.9449, the process has a strong positive continuity. 2) There are many time-scale features of crop disasters in China. Different time scales show different cycle alternations. Large-scale periodic changes are nested with small-scale periodic changes. 3) Wavelet analysis shows that the affected area has the first and second main periods on 6a and 13a, respectively; the first and second major periods exist on the areas 6a and 18a respectively; the loss of grain on the areas 8a and 18a There are first and second main periods, respectively.