岩溶地区生态环境脆弱性与水资源承载力评价方法研究——以云南省西畴县为例

构建岩溶地区生态环境脆弱性与水资源承载力评价的指标体系和分级标准,建立基于GRNN与RBF神经网络算法原理的生态环境脆弱性以及水资源承载力评价模型,采用内插法构造网络训练样本,对典型岩溶地区西畴县不同规划水平年生态环境脆弱性与水资源承载力进行评价分析。结果表明:①西畴县不同规划水平年生态环境脆弱性处于中度脆弱与强度脆弱之间,水资源承载力处于基本可承载与可承载之间,评价结果符合西畴县现状,可为岩溶地区生态环境治理和水资源开发利用提供参考。②GRNN与RBF网络模型对西畴县各水平年生态环境脆弱性与水资源承载力评价结果相同,表明研究建立的生态环境脆弱性与水资源承载力评价模型和评价方法均是合理可行的,同BP等网络算法相比,GRNN与RBF神经网络模型具有收敛速度快、预测精度高、调整参数少,不易陷入局部极小值等优点。

Research on the evaluation method for ecological environment vulnerability and the carrying capacity of water resources in karst areas:taking Yunnan Province Xichou for example

The paper built the evaluation index system and grading standards for the vulnerability ecological environment and the carrying capacity of water resources in karst areas, established the evaluation model for the ecological environment vulnerability and carrying capacity of water resources based on GRNN and RBF neural network theory,constructed using interpolation network training sample of a typical karst area Xichou County planning level in different eco-environmental vulnerability evaluation and analysis of water resources carrying capacity.The results showed that: ①the eco-environmental vulnerability of Xichou County in different level years is between moderate and strong,the water carrying capacity is fundamental to carry and can carry,the evaluation results accord with the status quo Xichou County and can provide reference for the karst regional ecological environment and water resources development and utilization.② the evaluation results of GRNN and RBF network model for the fragile ecological environment and water resources carrying capacity in Xichou county level is the same,indicating that the study to establish a fragile ecological environment and water resources carrying capacity of assessment models and evaluation methods is reasonable and feasible,compared with BP and other network algorithm,GRNN and RBF neural network model are possessed of fast convergence,high prediction accuracy,less adjustable parameters and differcult to fall into local minimum and so on.