研究水文系统复杂度的计算方法。根据排列熵、近似熵、样本熵和Lopez-Ruiz复杂度,建立了三种基于不同熵算法的水文系统复杂度计算模型。定量描述水文系统的复杂性。以渭河流域林家村、张家山和状头水文站的月平均流量序列为例,计算结果表明:1970年和1990年前后,各站的月平均流量序列复杂度均出现不同程度的变异现象,20世纪70年代初和90年代初,渭河流域人类活动显著增强,使得径流过程复杂性增加。滑动t检验法和斯米尔诺夫定理检验与文中计算结果一致,表明文中模型是合理的。通过本文的研究,以期为定量识别水文系统复杂性提供一种新的方法。
This paper presents a method for calculation of the complexity of hydrological system. The authors derived three complexity models by integrating the algorithms of permutation entropy, approximate entropy and sample entropy with the Lopez-Ruiz complexity theory. These models were applied to calculation of the observed data series of monthly mean runoff at three gauge stations of Linjiacun, Zhangjiashan and Zhangtou on Wei River. The results show that changes in the runoff complexity of all the stations appeared in around 1970 and around 1990. It is also shown that in the two periods of early 1970s and early 1990s, the human activities in this watershed were rapidly enhancing, resulting in greater runoff complexity. A t-test and the K- S theory have been used for model validation. These models provide new methods for analysis of hydrological system complexity.