Abstract
Biodiversity is important for supporting ecosystem functioning. To evaluate the factors contributing to the strength of microbial diversity-function relationships in complex terrestrial ecosystems, we conducted a soil survey over different habitats, including an agricultural field, forest, wetland, grassland, and desert. Soil microbial multidiversity was estimated by the combination of bacterial and fungal diversity. Soil ecosystem functions were evaluated using a multinutrient cycling index (MNC) in relation to carbon, nitrate, phosphorus, and potassium cycling. Significant positive relationships between soil multidiversity and multinutrient cycling were observed in all habitats, except the grassland and desert. Specifically, community compositions showed stronger correlations with multinutrient cycling than α-diversity, indicating the crucial role of microbial community composition differences on soil nutrient cycling. Importantly, we revealed that changes in both the neutral processes (Sloan neutral modeling) and the proportion of negative bacterial-fungal associations were linked to the magnitude and direction of the diversity-MNC relationships. The habitats less governed by neutral processes and dominated by negative bacterial-fungal associations exhibited stronger negative microbial α-diversity-MNC relationships. Our findings suggested that the balance between positive and negative bacterial-fungal associations was connected to the link between soil biodiversity and ecosystem function in complex terrestrial ecosystems. This study elucidates the potential factors influencing diversity-function relationships, thereby enabling future studies to forecast the effects of belowground biodiversity on ecosystem function.IMPORTANCE The relationships between soil biodiversity and ecosystem functions are an important yet poorly understood topic in microbial ecology. This study presents an exploratory effort to gain predictive understanding of the factors driving the relationships between microbial diversity and potential soil nutrient cycling in complex terrestrial ecosystems. Our structural equation modeling and random forest analysis revealed that the balance between positive and negative bacterial-fungal associations was clearly linked to the strength of the relationships between soil microbial diversity and multiple nutrients cycling across different habitats. This study revealed the potential factors underpinning diversity-function relationships in terrestrial ecosystems and thus helps us to manage soil microbial communities for better provisioning of key ecosystem services.