基于多目标遗传算法的层级生态节点识别与优化——以常州市金坛区为例

有效生态网络构建可保障区域生态安全,是实现区域可持续发展的重要路径。基于生态节点的内涵解析,建立了“资源型战略点-结构型战略点-结构型薄弱点”的多层级生态节点识别体系,利用金坛区2015年的遥感影像、土地利用数据、POI数据等,通过多目标遗传算法、最小阻力模型等方法,构建了金坛区层级生态网络,并采用表征网络拓扑结构和节点效用性等指标,定量评价生态节点优化前后的生态网络性能,取得以下主要研究结论：（1）多层级生态网络在节点效用、网络整体性能上显著优于一般网络,且在复杂生态水网区域具有较强适用性;（2）金坛区现状生态网络分布不均匀,生态节点布局亟待优化,经优化节点覆盖率提升了17.70%,节点分布均匀度降低45.45%,平均聚类系数提升了87.36%;（3）多层级生态节点体系具有实践应用性,应针对不同类型生态节点采取差别化管理策略。

Identification and optimization of hierarchical ecological nodes based on multi-target genetic algorithm: Take Jintan district of Changzhou as an example

Effective ecological network construction can guarantee regional ecological security, which is an important path to achieve regional sustainable development. Based on the analysis of the connotation of ecological nodes, this paper establishes a multi-level ecological node identification system based on "resource-based strategic point - structural strategic point - structural weak point", and uses remote sensing images, land use data, POI data, etc. of Jintan district in 2015. This paper also employs multi-target genetic algorithm, minimum resistance model and so on. The hierarchical ecological network in the ideal state of Jintan district was constructed, and the ecological network was quantitatively evaluated by using indexes such as the network topology and nodes utility. Based on the above, the following main conclusions were obtained: (1) Multi-level ecological network is significantly better than the general network in terms of nodes utility and overall network performance. And the water network area has a strong applicability. (2) As the current situation of the ecological network distribution is uneven, the ecological nodes' layout needs to be optimized. The optimized node coverage increased by 17.70%, the uniformity of node distribution decreased by 45.45%, and the average clustering coefficient increased by 87.36%. (3) Multi-level ecological node system has practical application, and different management strategies should be adopted for different types of ecological nodes.