Evaluation of the intensive use of construction land based on multisource remote sensing: a case study of the Yellow River Basin

To address issues in previous intensive land use studies – such as data distortion, strong subjectivity in weight determination, and traditional models ignoring temporal dynamic correlations and spatial unit relevance – this study takes the Yellow River Basin as the research area, integrates multi-source remote sensing data, constructs a time-lag allometric growth model for intensive construction land use considering time lag, and couples it with a spatial autoregressive model. The time-lag model introduces time-lag parameters to accurately capture delayed response relationships with significantly improved goodness of fit; the spatial autoregressive model compensates for the neglect of adjacent spatial unit relevance. Results show: (1) Intensive use exhibits significant multi-scale temporal and spatial differences, with higher levels in the middle and lower reaches (allometric growth coefficient range [0.3, 2.3]) than the upper reaches ([–8.9, 1.9]); (2) Moran’s I indices are 0.4379 (prefecture-level) and 0.3856 (county-level); high-high agglomeration is in middle and lower reaches’ urban agglomerations, low-low in Lanzhou-Xining Urban Agglomeration, with more significant county-level high-high agglomeration; (3) Intensive use is positively correlated with three factors including primary industry proportion, negatively with three including hospital beds, and neighboring units have significant spatial spillover effects.