Multi-Objective Optimization of Ecological Security Patterns in Intensive Agricultural Landscapes: Trade-Offs Among Network Structure, Connectivity, and Robustness

Against the backdrop of rapid urbanization and agricultural intensification, habitat fragmentation and declining ecological connectivity have become increasingly pronounced. To address the limitations of conventional ecological security pattern (ESP) optimization, which often prioritizes a single connectivity objective, this study selected Jingmen City in Hubei Province, China as the research area. Based on three aspects: comprehensive network structure, connection efficiency and anti-interference capability, an ESP optimization framework with multiple objectives was constructed. Ecological source areas were identified by integrating morphological spatial pattern analysis (MSPA) with landscape connectivity indices. A comprehensive resistance surface was then constructed using the analytic hierarchy process (AHP) and minimum cumulative resistance (MCR) model, and ecological corridors were delineated using circuit theory and Linkage Mapper. The non-dominated sorting genetic algorithm II (NSGA-II) was applied to optimize the ecological network under four scenarios: structure-maximized, connectivity-maximized, robustness-maximized, and balanced. From 2004 to 2024, Jingmen’s land-use structure remained relatively stable, but urban expansion and agricultural development reduced natural ecological space and intensified fragmentation. The initial network included 54 ecological source areas and 119 potential corridors. The reconstructed network contained 77 source areas and 183 candidate corridors. Under the scenarios of structure maximization and connectivity maximization, the structure and connectivity performed optimally, while the resilience-maximization scenario achieved the highest robustness at the expense of structural complexity and connection efficiency. The balance scenario achieved a favorable trade-off among the three objectives. Based on this scenario, ecological-conservation-priority zones, ecological-restoration zones, and graded-ecological-regulation zones were delineated. Low-impact measures, including restoration of corridor bottlenecks, enhancement of riparian vegetation along ditches and canals, and establishment of farmland shelterbelts, are proposed. These findings provide a scientific basis for ecological network optimization and territorial ecological restoration in intensively cultivated agricultural landscapes.