Delineating and Mapping Green Infrastructure Systems: A Case Study of Central Wuhan, China

Green infrastructure (GI) is widely recognized as an essential solution for addressing urban environmental issues such as pollution, flooding, urban heat islands, and climate change. However, despite its increasing importance, a unified definition and classification system for green infrastructure has yet to be established. This study selected the central urban area of Wuhan as the study area, where extensive urban development coexists with diverse ecological spaces. Based on web map AOI data, high-resolution remote sensing imagery (approximately meter-level spatial resolution), the Third National Land Survey, and urban planning databases, a green infrastructure classification system encompassing natural, semi-natural and artificial elements was identified and constructed through multisource spatial data integration and deep learning-assisted image interpretation. Landscape indices were used to measure the connectivity and fragmentation of green infrastructure landscapes within the city, including metrics such as contagion, Shannon diversity index, and edge density. Finally, the Minimum Cumulative Resistance (MCR) model and the gravity model were employed to construct the green infrastructure network. The results identified 6974 green infrastructure elements, which were classified into two categories and 15 subcategories. The green infrastructure system was found to be dominated by natural green infrastructure, particularly rivers and lakes, while semi-natural and artificial green infrastructure showed a relatively high degree of fragmentation. The constructed green infrastructure network identified 813 ecological source patches as key nodes, 45 ecological corridors, 24 ecological pinch points, and 37 ecological barrier points, revealing key areas for improving ecological connectivity. By evaluating ecological corridors, key green infrastructure nodes and corridors were identified to help determine intervention areas for improving ecological flows and connectivity. The study emphasizes the importance of maintaining and expanding these corridors to support biodiversity, climate adaptation, and the overall well-being of urban residents. This research contributes to improving the management efficiency, resource allocation, and urban ecological restoration of green infrastructure. It also addresses issues related to data quality and granularity, proposing methods to enhance the accuracy of green infrastructure identification through multi-source data. By establishing a comprehensive green infrastructure classification and network system, it provides valuable insights to inform more effective management and planning of urban green spaces in the context of rapid urbanization.