基于改进主成分分析的区域水资源承载力评价研究

水资源承载力评价,对推行最严格的水资源管理制度具有重要意义.本文在构建评价指标体系的基础上,尝试将Copula函数应用于水资源承载力评价.针对多元变量导致Copula函数参数求解困难的问题,采用主成分分析法(principal component analysis,PCA)提取主成分因子,形成新的指标体系,从而建立PCA-Copula评价方法,并以新疆和田地区为例,对区域水资源承载力(water resources carrying capacity, WRCC)进行综合评价.结果表明:PCA-Copula评价方法的评价结果与熵值法以及突变理论评价方法的结果排序完全一致,评价结果具有一致性,能够客观反映研究区水资源承载力水平.同时,PCA-Copula评价方法的评价值在0~1之间分布均匀,最大差值为0.818,明显高于另外两种方法,且相邻排序的综合评价值梯度明显,有利于更直观地区别水资源承载力的高低.

Evaluation and research on regional water resources carrying capacity based on improved principal component analysis

Water resources carrying capacity evaluation has important significance to carry out the most strict water resources management system.In the paper, Copulas function is applied to the water resources carrying capacity evaluation on the basis of constructing evaluation index system.Principal component analysis (PCA) is adopted for absorbing principal component factor and forming a new index system, thereby establishing PCA-Copulas evaluation method.Xinjiang Hotan is adopted as an example for comprehensive evaluation of regional water resources carrying capacity (WRCC).Results show that PCA-Copula evaluation method is completely consistent with the result order of entropy value method and catastrophe theory evaluation method.The evaluation result is consistent, which can objectively reflect the level of water resources carrying capacity in the study area.Meanwhile, PCA-copulas evaluation method is evenly distributed between 0 and 1.The maximum difference value is 0.818, which is significantly higher than the other two methods.In addition, comprehensive evaluation values have prominent gradient in adjacent sorting, which is beneficial for distinguishing water resources carrying capacity level more intuitively.