Building a scientific and reasonable ecological network is the key for optimizing the pattern of territorial development and protection, and is of great significance for ensuring regional ecological security and promoting the virtuous cycle of ecosystems. In previous studies, nodal attack method (destruction of ecological source area) was often used in the "robustness" evaluation of ecological networks. Actually, the ecological corridor is more fragile than the source area, and thus the nodal attack method is not reasonable. In this study, taking Jiuquan City as the research area, based on the circuit model to construct the ecological network, we carried out the topology optimization of ecological network by using three strategies (random edge increase, node degree and priority edge increase with low node intermedium number) in complex network theory. We compared and analyzed the "robustness" of ecological network before and after optimization by constructing edge attack strategy, and selected the best network optimization strategy. The results showed that 65 ecological source areas were identified in Jiuquan City, with a total area of 20275.15 km2, and that grassland accounted for 89.5% of the source area. We identified 179 ecological corridors with a total length of 6387.16 km, 158 ecological barrier points with a total area of 1385.5 km2. The unused land accounted for 92.2% of the total barrier points area. We identified 63 ecological pinch points, mainly concentrated in the source edge and corridor intersection. Among them, the spatial distribution of 11 barrier points and pinch points was consistent, which was the key area to be repaired in ecological network optimization. The three optimization strategies had significantly improved the stability of ecological network in Jiuquan City. The relative size of the maximum connected subgraph and the edge connected rate of the ecological network of the optimization strategy of adding edges according to degree were all the most stable under random attack mode and deliberate attack mode, which was the best optimization scheme for ecological network in Jiuquan City.
构建科学合理的生态网络是优化国土空间开发保护格局的关键,对保障区域生态安全和促进生态系统良性循环具有重要意义。既有研究在生态网络“鲁棒性”评价中多采用节点攻击法(破坏生态源地),而现实中生态廊道相比源地更脆弱,节点攻击法尚缺乏合理性。本研究以酒泉市为研究区,基于电路模型构建生态网络,采用复杂网络理论中的随机增边、节点度和节点介数低者优先增边3种策略对生态网络进行拓扑结构优化,并构建了边攻击策略对优化前后的生态网络“鲁棒性”进行对比,遴选最佳网络优化策略。结果表明: 酒泉市共识别出生态源地65个,总面积20275.15 km2,草地占源地面积的89.5%。识别生态廊道179条,全长6387.16 km。识别生态障碍点共158个,总面积1385.5 km2,未利用地占障碍点总面积的92.2%。识别生态夹点63个,主要集中在源地边缘及廊道交汇处。其中,11处障碍点与夹点的空间分布位置较一致,是生态网络优化中需重点修复的区域。3种优化策略均显著提高了酒泉市生态网络的稳定性,按度增边优化策略的生态网络在随机攻击和蓄意攻击模式下最大连通子图相对大小和边连通率均最稳定,是酒泉市生态网络的最佳优化方案。.
Keywords: Jiuquan City; complex network theory; ecological network; optimize the strategy; robustness.