Multivariate Assessment of Soil Physical Quality under Crop-Livestock-Forest Integration in a Semiarid Planosol
摘要
This study evaluated the long-term effects of contrasting crop systems on soil physical properties in a tropical Planosol after 12 years of continuous management. The experimental design included a Crop–Livestock–Forest Integration (CLFI) system, two monocropping systems (MFS1 and MFS2), and a natural ecosystem (as reference), with soil samples collected at three depths (0–10, 10–20, and 20–30 cm) to assess the interaction between management and profile position. A comprehensive set of physical attributes was measured, including texture, bulk density, compaction indicators, macroporosity, total porosity, water-filled pore space, mean weight diameter, and water-stable aggregates. The results demonstrated strong interactive effects between crop system and soil depth, with the surface layer being the most responsive to management-induced changes. The CLFI improved porosity and aggregation while reducing bulk density relative to the conventional system, particularly in the upper 10 cm. Subsurface layers exhibited more conservative behavior, with weaker structural development and limited response to management. Multivariate analyses (PCA and Canonical Discriminant Analysis) revealed clear separation among systems, with native vegetation forming a distinct structural signature characterized by higher macroporosity and aggregate stability. A Soil Physical Quality Index (SPQI) synthesized multiple indicators and confirmed the superior structural condition of CLFI when compared with monocropping system, while both the MFS1 and MFS2 exhibited the lowest physical quality across all depths. Diversified systems promoted improvements in soil physical functioning by enhancing biogenic porosity and aggregation processes. These findings highlight the importance of integrated crop–livestock–forest strategies for sustaining soil physical quality in semiarid tropical environments and demonstrate the value of multivariate approaches for detecting complex management effects.