Aim: Microbial diversity is one of the most important factors for maintaining the performance of multiple functions in soils (multifunctionality). However, existing studies typically consider taxonomic richness or Shannon diversity of the entire community. We know little about the connection network of taxonomic and phylogenetic diversity facets and their collective impact on multifunctionality. In this study, the linkages of diversity effects within functional groups were disentangled in drylands.

Location: The central Tibetan Plateau.

Time period: Present.

Major taxa studied: Phototrophs and diazotrophs (mainly cyanobacteria).

Methods: Given the carbon- and nitrogen-limited nature of drylands, we conducted a high throughput sequencing of C/N-fixing functional groups from biocrusts and evaluated multiple diversity facets (i.e., richness, evenness, and phylogeny-related trait dissimilarity), together with seven crucial variables of soil functioning to calculate multifunctionality. The relations between multifaceted diversity and abundance with individual functions and multifunctionality were validated by a set of solid statistical analyses.

Results: We found that the integrated biodiversity index was a stronger predictor of multifunctionality than richness. The divergent performance of different diversity facets determined the idiosyncratic effect of each functional group on soil multifunctionality. Moreover, the evaluation of functional significance at the species level gave important clues on the trade-offs and redundancy in each functional group, explaining the distinct patterns of diversity effects. Namely, richness was the dominant factor for diazotrophs to maximize multifunctionality, whereas phylogenetic dissimilarity was the essential one for phototrophs.

Main conclusions: our study demonstrated that multiple diversity facets should be considered when grasping the biodiversity effects. In contrast with the community level, within-functional group measures may adequately capture the features of diversity that are most correlated with soil multifunctionality. Our results provided a perspective to bridge the gap between taxonomic and trait-based approaches for elucidating the biodiversity-ecosystem function relationship.