摘要:在应用景观连接度研究景观格局变化时，选择适宜的距离阈值是关键。适宜的距离阈值有利于识别景观中的关键斑块并及时发现连接脆弱的区域，对景观格局动态分析、生态恢复研究等均具有重要意义。以天津市滨海新区湿地生态系统为研究对象，将研究区域划分为不同的焦点景观，焦点景观是依据研究区域栖息候鸟的生境面积而划分的面积大小相同的区域，并通过ArcGIS中的泰森多边形分析实现，焦点景观分析可为面积更大、结构更加复杂的景观分析提供新的思路。文章选取斑块间链接数（NL）、组分数（NC）、等效连接面积指数（ECA）、景观巧合概率指数（LCP）来表征景观连接度，设定100、200、400、600、800、1000、1500、2000、3000、5000、8000 m共11个距离阈值，探讨不同距离阈值下各指数的变化规律，对比分析研究区域整体景观斑块和焦点景观内各斑块的连接度指数随距离阈值变化的异同，最终确定研究区域最佳距离阈值。结果表明：（1）各焦点景观内斑块的景观连接度指数随距离阈值的变化趋势与研究区域整体景观斑块相似且对应的最佳距离阈值相同；（2）天津市滨海新区湿地景观连接度距离阈值的适宜范围是400-800 m，该区间内进行景观格局分析研究能够较为真实的反映出研究区域实际景观状况；（3）通过进一步研究，最终确定400 m为天津滨海新区湿地景观连接度最佳距离阈值。研究结果可为天津市滨海新区湿地景观研究提供参考，在实际研究中距离阈值的选择可根据具体研究目标、研究层次需求等进行适当调整。



Abstract:When applying landscape connectivity to the study of landscape pattern changes, it is important to choose the appropriate distance threshold, which affects the analysis of landscape connectivity in the study area. If the distance threshold is set too low, the patches in the area will remain unconnected, and fragmentation will be severe, whereas if the distance threshold is set too high, the patches in the area are interconnected and grouped together. However, neither of these scenarios is likely to truly reflect the actual condition of the landscape in the study area. Using an appropriate distance threshold will facilitate the identification of key patches in the landscape and the discovery of vulnerable areas, both of which are important for the dynamic analysis of landscape patterns and ecological restoration studies. The present study investigated this principle using Landsat 8-OLI remote sensing images of a wetland ecosystem of the Tianjin Binhai New Area, taken on April 14, 2017. The research area was divided into different focal landscapes, in order to prevent software stalling and reductions in data processing speed that could have resulted from the large study area and provide a new analytical idea for this field. The focal landscape was the same size area, based on the habitat area of the migratory birds in the study area, and was analysed by Thiessen polygon analysis in ArcGIS. The number of links, number of components, equivalent connectivity area, and landscape coincidence probability were selected to characterize the landscape connectivity, and a total of 11 distance thresholds (100, 200, 400, 600, 800, 1000, 1500, 2000, 3000, 5000, and 8000 m) were selected to draw the ‘Landscape Connectivity Index-Distance Threshold’ curve, which was used to study the trend of each index under different distance thresholds. By comparing the overall landscape patch and the focal landscape connectivity index with distance thresholds, the optimal distance threshold for the study area was determined. (1) The optimal distance thresholds selected by the two methods of focal landscape and overall landscape patch were the same, and the proposed method provides new ideas for future regional landscape analysis and more complex ecological landscape pattern research. (2) The range of 400-800 m can be used as suitable distance threshold ranges for the dynamic change of landscape pattern and restoration of ecosystems in the Binhai New Area. In this interval, the connectivity of the habitat patch in the study area was relatively stable, which clearly illustrates the distribution of the patches and helps to identify the important patches in the landscape. (3) Through further research, 400 m was finally selected as the optimal distance threshold for the connectivity of the wetland ecosystem in Tianjin Binhai New Area. These findings provide a reference for studying wetland landscapes in the Tianjin Binhai New Area. In addition, the selection of the distance threshold in practical research can be adjusted according to specific research objectives and research level requirements.





PDF全文下载地址:

https://www.ecologica.cn/stxb/article/pdf/stxb201804190897