Population Growth and Shrinkage Reshape Urban Land Use Structure: Evidence from Multi-Entropy Analysis

Population change is closely associated with urban land use structure, yet their nonlinear relationship and distributional heterogeneity remain poorly understood. Drawing on panel data for 2000–2020 from urban districts across 284 Chinese prefecture-level cities, this study applies generalized additive models, segmented regression and quantile regression to investigate the nonlinear association, phase features, and heterogeneity between population change and land use structure. The results indicate that: (1) Urban district land use structures across China became more balanced during 2000–2020; population shrinkage cities underwent larger structural shifts alongside more marked fluctuations. (2) The population change rate shows a significant U-shaped association with land use structure entropy (LUSE), turning at approximately −1% annually; growth cities follow an inverted U-shaped pattern, and shrinkage cities exhibit distinct phase-dependent effects. (3) Cities with higher LUSE demonstrate more substantial land use proportion adjustments, a tendency most salient among population shrinkage cities. These findings offer empirical support for differentiated land use governance and optimized territorial spatial planning for both growth and shrinkage cities.