Spatial scale-dependent dilution effects of biodiversity on plant diseases in grasslands
Xiang Liu,Yao Xiao,Ziyuan Lin,Xingxing Wang,Kui Hu,Mu Liu,Yimin Zhao,Yanwen Qi,Shurong Zhou
DOI: https://doi.org/10.1002/ecy.3944
IF: 4.8
2023-01-01
Ecology
Abstract:The rapid biodiversity losses of the Anthropocene have motivated ecologists to understand how biodiversity affects infectious diseases. Spatial scale is thought to moderate negative biodiversity-disease relationships (i.e., dilution effects) in zoonotic diseases, whereas evidence from plant communities for an effect of scale remains limited, especially at local scales where the mechanisms (e.g., encounter reduction) underlying dilution effects actually work. Here, we tested how spatial scale affects the direction and magnitude of biodiversity-disease relationships. We utilized a 10-year-old nitrogen addition experiment in a Tibetan alpine meadow, with 0, 5, 10, and 15 g/m(2) nitrogen addition treatments. Within the treatment plots, we arranged a total of 216 quadrats (of either 0.125 x 0.125 m, 0.25 x 0.25 m or 0.5 x 0.5 m size) to test how the sample area affects the relationship between plant species richness and foliar fungal disease severity. We found that the dilution effects were stronger in the 0.125 x 0.125 m and 0.25 x 0.25 m quadrats, compared with 0.5 x 0.5 m quadrats. There was a significant interaction between species richness and nitrogen addition in the 0.125 x 0.125 m and 0.25 x 0.25 m quadrats, indicating that a dilution effect was more easily observed under higher levels of nitrogen addition. Based on multigroup structural equation models, we found that even accounting for the direct impact of nitrogen addition (i.e., "nitrogen-disease hypothesis "), the dilution effect still worked at the 0.125 x 0.125 m scale. Overall, these findings suggest that spatial scale directly determines the occurrence of dilution effects, and can partly explain the observed variation in biodiversity-disease relationships in grasslands. Next-generation frameworks for predicting infectious diseases under rapid biodiversity loss scenarios need to incorporate spatial information.