Abstract
Urban ecological security (ES) is increasingly challenged by rapid economic development. Previous studies have primarily focused on the physical characteristics of the ecosystem, rarely analyzing the interdependencies and impacts of natural, social, and economic development on ecological security. To address this gap, we examined the Guangdong-Hong Kong-Macao Greater Bay Area (GBA) and proposed a novel comprehensive framework that includes ecological security assessment (ESA) and ecological infrastructure (EI) planning. The Driver-Pressure-State-Impact-Response-Structure (DPSIR-S) framework and obstacle degree model (ODM) were used to assess the ES level and analyze the impacts of natural, social, and economic development on it. Subsequently, leveraging natural language processing (NLP) technology and the "matrix-patch-corridor" method, we proposed an EI planning method that integrates the outcomes of the ESA and the responsive policy context to optimize urban ecological security. The results indicated that the response level was a significant factor in determining urban ES, with environmental protection investment share, GDP, population density, and GDP per capita identified as the main obstacle factors impeding ES in the GBA. Furthermore, the proposed EI network increased ecological space by 10.5%, incorporating 121 ecological nodes and 227 ecological corridors, which significantly improved the connectivity of fragmented ecological sources and optimized the urban landscape. The findings of this study contribute to the exploration of urban ES protection within the context of balanced natural, social, and economic development, providing a theoretical foundation and practical guidance for optimizing urban ES.