Replication Data for "Global disparities in urban green space equity and the transformative potential of targeted pocket parks

Urban green spaces (UGS) are essential for public health and climate resilience, yet their distribution remains one of the most visible markers of spatial injustice. While localized studies have documented green space shortages, a standardized, global-scale assessment of accessibility equity remains elusive. Here we evaluate walking-distance green space accessibility for 118 major cities across all inhabited continents, covering a combined population of approximately 700 million. Using a 2-step floating catchment area (2SFCA) method integrated with high-resolution satellite vegetation indices and human settlement data, we find a profound "green divide" between the Global North and Global South. Our machine learning models (R squared = 0.91) reveal that rapid population growth and economic constraints are primary drivers of inequality. Furthermore, through scenario simulations, we demonstrate that a targeted 10% expansion of green space via "pocket parks" in underserved areas is up to ten times more effective at reducing Gini coefficients than traditional large-scale park planning. These findings provide a universal roadmap for city planners to transition from "quantity-driven" to "equity-led" urban greening.