Spatial Network Structure Characteristics and Resilience Evaluation of Urban Ice and Snow Activities in Cold Border Regions: A Case Study of Harbin City, China

Under the background of the diversification and increasingly frequent flow of ice and snow activities, revealing the characteristics of the spatial network of ice and snow activities and enhancing the resilience of the network to cope with multiple disturbances have become the core issues of the adaptive transformation and development of ice and snow tourism in cold border cities. Taking Harbin City, Heilongjiang Province, China as the research object, based on Python and Large Language Model (LLM), this paper obtained the ‘O (origin)-D (destination)’ space movement trajectory in Mafengwo Tourism Official Website online travel notes from 2023 to 2025. With the help of complex network analysis and scenario simulation, it analyzed the ‘overall-individual’ characteristics of the static ice and snow activity space network, and evaluated the resilience of the dynamic space network. The results show that: 1) the spatial network nodes of ice and snow activities show the characteristics of ‘overall aggregation-local dispersion’, and the moving trajectory shows the radial spatial characteristics of ‘center-surrounding’, the network structure shows the spatial imbalance state of unipolar aggregation, and the mobility between a large number of nodes is weak. 2) There are local dense areas and strong clusters in the spatial network of ice and snow activities, but the connections between nodes are ‘redundant-inefficient’ hierarchical connectivity, and a few hub nodes have very many connections, which combines the hub monopoly of scale-free network and the local clustering of small world network. 3) The relative positions of nodes in the network are quite different. Some nodes show a significant central superposition of ‘flow control-transit monopoly-network center-hub agglomeration’, and the network shows a significant ‘core-periphery’ feature. 4) The spatial network of ice and snow activities is strong and resilient when subjected to random attacks, but it shows significant vulnerability under intentional attacks, with high network crash rate and strong dependence on the connection between key nodes and the core. This study provides a scientific basis for decision-makers to accurately protect core activity nodes, cultivate secondary hubs and alternative routes, thereby reasonably promoting the continuous optimization of the spatial layout of urban activity networks.