Abstract
Recent studies have tested the sensitivity of Microbial-Based Indicators (MBIs), such as Arbuscular Mycorrhizal Fungi (AMF), for monitoring changes in soil properties across a wide range of environments. However, the direction and magnitude of AMF responses depend on contextual factors such as land use, vegetation type, geography, and environmental variables. Thus, there is still no consensus about whether AMF show a consistent response to changes in certain soil properties. Likewise, a better understanding of how interactions among different aspects of the microbial community can modify the influence of soil properties on AMF responses is needed. Based on data compiled across a wide geographic range, this study analyzes the responses observed in several aspects of AMF to soil properties across different land uses. A Dependency Network Analysis (DEPNA) was performed within a correlation network constructed using average correlation coefficients to evaluate the strength of relationships between soil properties and different MBIs while controlling the effect of another MBI. Average correlation coefficients were estimated via meta-analysis to account for experimental heterogeneity. The total Influence Degree (TID), computed from partial correlations, suggests strong dependencies between MBIs (related to AMF diversity, mycorrhizal colonization rate, and Glomalin Soil Proteins) and soil properties (Pb concentrations, soil structural features, and nutrient stocks). The results suggest that AMF emerge as robust microbial indicators of soil condition, reflecting both fertility enhancement and degradation. Partial correlation and dependency network analyses show that soil effects on AMF and GRSP are largely mediated by microbial biomass, respiration, and diversity, explaining responses across land uses and stress gradients.