Floral trait functional diversity is related to soil characteristics and positively influences pollination function in semi-natural grasslands

2020 
Abstract Semi-natural grasslands are threatened habitats providing many ecosystem services, such as climate regulation or water erosion control. Factors affecting the vegetative trait diversity of their plant communities are well studied, but those affecting their floral diversity are not. Local factors, such as land-use intensification and soil characteristics, can greatly influence floral trait diversity in semi-natural grasslands, which in turn can influence pollination function. We selected 16 semi-natural grasslands in France along a local land-use intensification gradient. Floral functional diversity indices were calculated from measurements of five floral traits known to influence interactions with pollinators: flower area, flowering height, floral reflectance, nectar tube depth and nectar sugar production per floral unit. Using linear mixed models, we found a negative influence of phosphorus soil content, and a positive influence of a composite soil variable including soil texture, organic carbon and total nitrogen soil content, on floral functional diversity. However, local land-use intensification did not influence floral functional diversity. Moreover, our results showed a positive relation between floral functional diversity and plant-pollinator interaction frequency as a proxy of pollination function. More specifically, we found a positive influence of the nectar resource diversity (i.e. the variability in production of nectar sugar by flowers) on plant-pollinator interaction frequency. These results, based on a correlative approach and functional diversity indices as measure of functional complementarity, suggest that the diversity of floral traits may promote niche partitioning of pollinators and pollination function approximated by plant-pollinator interaction frequency, and highlight the main role of nectar in this process.
    • Correction
    • Source
    • Cite
    • Save
    • Machine Reading By IdeaReader
    80
    References
    4
    Citations
    NaN
    KQI
    []
    Baidu
    map