Vertical spatial patterns and associated factors of soil physicochemical quality in Camellia oleifera plantations in the hilly red soil region of Southern China: A space-for-time substitution approach

The hilly red soil region of southern China is a critical area where economic forest expansion overlaps with ecological restoration. We selected a typical Camellia oleifera ( C. oleifera ) plantation site in Meizhou, eastern Guangdong, and used a space-for-time substitution design to compare three adjacent vegetation-management units: secondary shrubland (SL), 4-year-old C. oleifera plantation (CO4), and 9-year-old C. oleifera plantation (CO9). We measured 12 soil physicochemical indicators and Cu, Zn, Pb, and As concentrations in the 0–20, 20–40, 40–60, and 60–100 cm soil layers. For each layer, a five-indicator minimum data set (MDS) was constructed using functional screening, principal component analysis, Norm values, and correlation-based redundancy reduction. Cu, Zn, Pb, and As were treated as heavy-metal environmental quality indicators, evaluated against the soil environmental background values of Guangdong Province and the risk screening values of GB 15618–2018, and were not included in the fertility-focused MDS. Within the sampled layers, MDS-SQI was generally higher in the C. oleifera units than in SL, although the direction of the CO4 and CO9 difference varied among layers. CO4 had relatively higher MDS-SQI in all layers from 20 to 100 cm; CO9 was similar to CO4 at 0–20 cm, slightly higher at 20–40 cm, and lower at 60–100 cm. Across the four layers, MDS-SQI was strongly correlated with FULL-SQI (Pearson r = 0.870–0.968, P < 0.001), indicating internal ranking consistency within the present dataset. GeoDetector analysis showed that slope position had the strongest explanatory power for spatial variation in MDS-SQI (q = 0.447–0.580), followed by vegetation-management unit (q = 0.119–0.289), whereas aspect and slope gradient had lower explanatory power. Interactions of slope position with vegetation-management unit, slope gradient, or aspect were predominantly enhancing. Mean concentrations of Cu, Zn, Pb, and As in all soil layers were below the upper limits of soil environmental background values for Guangdong Province and the risk screening values of GB 15618–2018, indicating that heavy metals were generally at natural background levels. These findings reveal layer-specific patterns of soil quality among three adjacent vegetation-management units on the same slope and their association with joint topographic management stratification. They provide case-study evidence for depth-stratified soil assessment and management in C. oleifera plantations in the hilly red soil region of southern China.