Microbial Network Complexity Emerges as a Key Predictor of Soil Multifunctionality Across Alfalfa Fields of Different Planting Years

Long-term alfalfa monoculture can lead to soil degradation, yet how microbial community attributes drive changes in soil multifunctionality (SMF) remains poorly understood. Here, we examined SMF and microbial community attributes across alfalfa fields over three planting year periods (3, 7, and 11 years) and used hierarchical partitioning and partial least squares path modeling to identify the key predictors of SMF and their influence pathways. SMF declined progressively with increasing planting years, decreasing by 26.3% from the 3-year to the 11-year fields, accompanied by decreases in alpha diversity and network complexity. Hierarchical partitioning analysis identified fungal and bacterial network complexity as the two most important predictors. PLS-PM analysis further revealed that planting years were associated with soil properties, triggering distinct inferred pathways for bacteria and fungi. In bacterial communities, alpha diversity enhanced soil multifunctionality through both direct positive effects and an indirect pathway, wherein suppressing beta diversity led to increased network complexity. In fungal communities, alpha diversity directly benefited SMF but simultaneously suppressed beta diversity, which in turn reduced network complexity. These findings indicate that microbial network properties play a central role in mediating soil multifunctionality under continuous alfalfa cropping, providing a network-based perspective for evaluating soil health in agroecosystems.