Scale-Dependent Variation in the Coupling Coordination of Production–Living–Ecological Spaces: Evidence from a Multi-Level Analysis in Shandong Province, China

The coupling coordination of production, living, and ecological (PLE) spaces is central to territorial spatial optimization, yet its assessment may vary substantially with spatial scale. Using 2020 land use/land cover data for Shandong Province, China, this study evaluated PLE functions at four levels: 2 km grid, township, county, and city. The coupling coordination degree model is then used to measure the coupling coordination level of PLE spaces at these scales, and Global Moran’s I and local spatial autocorrelation analysis are further employed to examine spatial associations. The results showed clear scale-dependent variation in PLE functions, coupling coordination, and spatial clustering. Although the relative ordering of PLE functions remained consistent across all four analytical levels, their mean values varied across scales. Production and living functions reached their highest mean values at the township level, whereas ecological function was lowest at this level and highest at the city level. Spatial variability also differed across levels, with the city level showing the lowest variability for all three functions. Mean coupling degrees remained high across the four levels (0.878–0.968), while mean coupling coordination degrees (0.623–0.670) remained within the moderate-coordination category. Both indicators were lowest at the township level and highest at the city level, although coupling coordination was most variable among townships. Significant positive spatial autocorrelation was observed at all four levels, with Global Moran’s I ranging from 0.285 to 0.732 (p < 0.05); spatial dependence was strongest at the township level and weakest at the city level. Local clusters and spatial outliers also became less detectable at broader levels. These findings show that analytical-unit selection influences estimates of PLE functions, coordination, and spatial dependence. A multi-level framework can therefore support hierarchical and spatially differentiated territorial governance, thereby contributing to regional sustainable development.