Local loss and spatial homogenization of plant diversity reduce ecosystem multifunctionality

Journal article

Biodiversity is declining in many local communities1 while also becoming increasingly homogenized across space2-4. Experimental studies show that local plant species loss reduces ecosystem functioning and services5-10, but the role of spatial homogenization of community composition and the potential interaction between diversity at different scales in maintaining ecosystem functioning remains unclear, especially when many functions are considered (ecosystem multifunctionality)11-14. We present a global analysis of eight ecosystem functions measured in 65 grasslands worldwide. We find that more diverse grasslands – those with both species-rich local communities (alpha diversity) and large compositional differences among localities (beta diversity) – had higher levels of multifunctionality. Moreover, alpha and beta diversity synergistically affected multifunctionality, with higher levels of diversity at one scale amplifying the contribution to ecological functions at the other scale. The identity of species influencing ecosystem functioning differed among functions and across local communities, explaining why more diverse grasslands maintained greater functionality when more functions and localities were considered. These results were general across continents and robust to variation in environmental drivers. Our findings reveal that plant diversity, at both local and landscape scales, contributes to the maintenance of multiple ecosystem services provided by grasslands. Preserving ecosystem functioning therefore requires conservation of biodiversity both within and among ecological communities.

Authors: Hautier, Y; Isbell, F; Borer, E; Hector, A; al., E

• Publication status: Published
• Peer review status: Peer reviewed
• Publisher: Springer Nature
• Journal: Nature Ecology and Evolution
• Volume: 2
• Pages: 50–56
• Publication date: 2017-12-04
• Acceptance date: 2017-10-25
• DOI: 10.1038/s41559-017-0395-0
• ISSN: 2397-334X