Spatial correlation network, driving factors and planning application of land green use efficiency in Zhejiang Province, China

Exploring land green use efficiency (LGUE) is crucial for optimizing land allocation and spatial planning, especially in land-constrained regions. Using 2001–2023 panel data from 90 counties of Zhejiang Province, this study constructs a LGUE spatial correlation matrix using a modified gravity model, applies social network analysis (SNA) to examine LGUE spatial correlation network (SCN) evolution, and uses spatial Durbin model to identify drivers. The findings reveal that: (1) LGUE exhibits a "first rise then decline" trend with spatial polarization. (2) The SCN optimized, correlation pairs increasing from 70 in 2001 to 254 in 2023, forming a Hangzhou-Ningbo-Zhoushan radial pattern. (3) Network structure featured "stable connectivity but loose cohesion": connectivity remained 1, density 0.12–0.17, efficiency down to 0.7794 from 2001 to 2023, while hierarchical degree up to 0.3045 post-2015, (4) 90 counties were classified into four functional blocks. (5) Core areas concentrated in four metropolitan zones (Hangzhou, Ningbo, Jinhua, Yiwu), aligning with the "dual-core, quad-polar" plan, but key areas like Yiwu and Ninghai remained peripheral, with "multiple clusters" yet to form. (6) Econometric results show economic development, government intervention, openness, and population density positively impacted LGUE, while energy consumption intensity, industrial structure advancement, and rationalization had negative effects, with significant heterogeneity between core and peripheral zones. This study deepens the understanding of LGUE's spatiotemporal and network traits, constructs a systematic measurement framework, and provides targeted support for enhancing LGUE and optimizing land planning in Zhejiang Province and other similar regions.