干旱区生态环境知识资源中心

Soil fungi and plants are tightly linked in pathogenic, commensal, and mutualistic ways. These interactions play a critical role in terrestrial ecosystem decomposition, nutrient cycling, and maintenance of plant productivity and diversity. A comprehensive understanding whether this fundamental plant-fungi relationship persists in ecosystems with increased N input is, however, still lacking. In this study, we investigated the relationships between plant and soil fungal diversity in a 6-year multi-level nitrogen addition experiment in a semi-arid grassland in northern China, using Illumina Miseq sequencing of the ITS1 barcode region for fungal identification. We hypothesized that N-induced changes in plant communities would be positively associated with those in soil fungal communities, and that the corresponding changes in both communities can be explained by the direct functional associations between plant and soil fungal communities and their shared environmental drivers. Our results showed that N-induced changes in plant alpha diversity, i.e., Shannon diversity, showed no significant relationship with those in soil fungal alpha diversity. The lack in significance of the relationship was primarily due to their contrasting correlates with N-induced soil physicochemical variables (i.e., soil available P (AP), inorganic N (NH4+ _N and NO3-_N) and organic C content), and weakened plant-ftmgi functional associations. In contrast to the lack of relationship between plant and soil alpha diversity, we did find a significant positive relationship between plant and fungal beta diversity (compositional dissimilarity between plots) under N enrichment. Our results reveal that the same soil physicochemical variables, including soil AP, inorganic N, moisture, pH, and associated extractable cations, correlated with compositional changes in plant and fungi following N enrichment. The strong coupling of plant and fungal beta diversity could largely be driven by their consistent responses to these shared edaphic factors. As such, our results suggest that information on N-induced changes in plant diversity can be used to predict beta, but not alpha diversity of soil fungal communities in semi-arid grassland ecosystems.