基于贝叶斯网络的矿业城市生态环境质量影响因子研究——以大冶市为例

Study on factors affecting the ecological environment quality of mining cities based on Bayesian network-a case study of Daye city

大冶市是典型的矿业城市,生态环境压力尤为突出,识别生态环境质量影响因子,能够及时发现生态环境中存在的问题,对于生态环境规划具有重要的参考意义。以研究区格网数据为样本,根据生态胁迫、生态助力、自然条件3类影响因子计算生态环境质量,并构建贝叶斯生态环境质量网络分析模型,探讨了3类影响因子对生态环境质量的作用大小和敏感程度,结果表明:大冶市生态环境质量由西向东逐渐降低,南北较高而中心较低,生态环境质量受人为活动影响较大;生态胁迫是影响大冶市生态环境质量的主要驱动因素,生态助力次之,二者熵减百分比分别为24.6%和2.31%,而自然条件仅为0.473%;贝叶斯网络较传统评价模型能够更加直观地量化各个变量,使评价结果更具说服力。

Daye city is a typical mining city with extremely high ecological environment stress.Identifying the impact factors of the ecological environment quality helps us timely find the problems in the ecology and is of great significance to make a plan for the ecological management.With the grid data from the study area,the ecological environment quality was calculated as per the three types of impact factors as follows:ecological stressors,ecological assistance,and natural conditions thus the Bayesian ecological environment quality network analysis model was constructed.The role and sensitivity of the three types of impact factors to the ecological environment quality were discussed.The results show that the ecological environment quality of Daye City gradually dropped from west to east,but high at north and south,and low in the center.The quality of ecological environment was greatly subjected to human activities.Ecological stressors were the leading factor affecting the ecological environment quality of Daye City,while ecological assistance took the secondary place,which presented with the entropy reduction percentage of 24.6%and 2.31%respectively,but the natural conditions accounted for only 0.473%.Bayesian network is proposed to be used for ecological environment evaluation because of the more intuitive quantification of each variable and availability of more convincing evaluation results than traditional models.