A Multi-scale Green Landscape Coordination Index for Characterizing Seasonal Dynamics across Urban Functional Zones

As China’s megacities shift from expansion-driven growth to stock-based renewal, urban green space planning is increasingly oriented toward quality improvement rather than quantity expansion. Conventional metrics based on green space ratios remain insufficient to capture the spatial and aesthetic coordination of urban landscapes. This study develops a function-adaptive Green Landscape Coordination Index (GLCI) that integrates color richness, color balance, and natural rhythm. Based on Sentinel-2 imagery, HSV-based hue metrics and differentiated weighting schemes are used to assess seasonal landscape coordination across three functional zones in Beijing (Natural-Recreational Zone (NRZ), Urban Comprehensive Functional Zone (UCFZ), Transport-Infrastructure Zone (TIZ)). Spatial clustering and an XGBoost-SHAP framework are then applied to characterize spatial heterogeneity and quantify the relative contributions of climatic, vegetation, thermal, anthropogenic, and socio-economic drivers. The results show that: 1) GLCI showed a stable functional gradient (NRZ > UCFZ > TIZ), with summer mean values of 0.62, 0.55, and 0.22, respectively. 2) Seasonal dynamics differed among functional zones, with stronger fluctuations in the UCFZ (0.25-0.55) and persistently low values in the TIZ (<0.22). 3) The NRZ exhibited relatively continuous high-value patterns (>90% medium- and high-value areas in summer), whereas the UCFZ was fragmented and the TIZ was dominated by low values. 4) SHAP results indicated that climatic and vegetation-related factors dominated the NRZ, while the UCFZ reflected coupled environmental and socio-economic influences. These findings provide quantitative support for refined urban green space management and resilience-oriented planning.