Key points are not available for this paper at this time.
Many regions on the globe are at risk of soil erosion, especially in the ecologically fragile areas of the Qinghai-Tibet Plateau. Enhancing the soil conservation capacity of regional vegetation is one of the main ways to address soil erosion. However, most studies ignore the relationship between the spatial distribution structure of vegetation and soil conservation capacity, and the construction of ecological spatial networks is an effective means to study the spatial distribution structure of vegetation. In this study, we extracted the ecological spatial network of Southeast Tibet by the minimum resistance model (MCR) and combined the ecological network with complex network theory to calculate the topological indices of the ecological spatial network and ecological nodes. Subsequently, the soil conservation capacity of ecological nodes is calculated based on the USLE model, and the relationship between it and topological indices is analyzed. It was shown that the degree, closeness centrality, betweenness centrality and eigenvector centrality of forest nodes showed a positive linear correlation with their soil conservation capacity, and the clustering coefficient of grassland nodes showed a positive linear correlation with their soil conservation capacity. This study calculates habitat quality based on the Invest model. We propose a Integrated Soil Conservation and Habitat Quality Evaluation (ISHE) Model to determine the direction of network optimization and propose an optimization strategy. We analyze the network robustness and soil conservation capacity and find that the optimized network robustness and soil conservation capacity are significantly improved. This study provides strategies and methods to protect ecologically vulnerable areas, optimize the spatial distribution of vegetation, and enhance soil conservation capacity. In addition, this can provide a theoretical reference for planning national ecological projects. We use the ISHE model to filter out three types of nodes that need to be focused on optimization and propose different optimization strategies.
Wang et al. (Wed,) studied this question.