Grazing livestock species composition influences community assembly and determines scale-dependent plant diversity
Abstract
Grazing by domestic herbivores such as cattle and sheep is a major driver of grassland biodiversity. However, it remains unclear how livestock composition influences community dynamics and consequently shapes plant diversity across scales. Here, we conducted a five-year livestock grazing experiment with continuous in situ vegetation monitoring by manipulating the grazing ratio of cattle and sheep in a meadow steppe of northern China. We found that all mixed grazing of cattle and sheep can continuously improve the α and β diversity of plants. Analysis of species dynamics within permanent quadrats showed that the number of newly colonizing species gradually increased as the proportion of grazing cattle increased, leading to higher plant α diversity under cattle dominated grazing (CCS). By contrast, β diversity was higher under sheep-dominated grazing (CSS). Grazing effects on α diversity were mainly driven by shifts in community dominance, whereas β diversity responded to both community dominance and soil nitrogen heterogeneity, the latter playing a stronger role. Cattle have more potential to produce competitive release and increase plant α diversity, while sheep are more conducive to creating soil resource heterogeneity and enhancing environmental filtration, thus leading to higher plant β diversity. Our study provides important mechanistic insights into how livestock species composition affects plant diversity at different scales. This presents a novel perspective on the need for precision control of grazing livestock composition to optimize grassland management for diversity conservation.
Dataset DOI: 10.5061/dryad.j3tx95xvk
Description of the data and file structure (column and explanation)
File: data.xlsx
Description: We have submitted the raw data. The data were obtained from four different livestock species composition treatments: CK (no grazing); CCS (cattle grazing intensity is twice that of sheep); CS (the grazing intensity of cattle and sheep is the same); CSS (sheep grazing intensity is twice that of cattle). Plant measurements include Richness, β diversity, community dominance, number of species colonization, number of species extinction, and community assembly process. Soil measurements include soil nitrogen, the quantity, and heterogeneity of soil-available nitrogen.
Key Variables
- Richness:number of species in quadrat
- β diversity:The average pairwise dissimilarity between quadrats within each plot (using the abundance-based Bray–Curtis)
- Dominance:Berger–Parker Dominance Index(%)
- Relative changes in dominance(%)
- Quantity of soil available N(mg/kg)
- Heterogeneity of soil available N(%)
- RC.bray: The null modeling outcomes of the community assembly process
Tabs
data_diversity: The calculated values of plant diversity during the experimental process of this study (2020-2024).
data_compared: The values of dominance, soil nutrients, and community assembly process were calculated at the beginning and end of the experiment in this study (2020, 2024).
data_relative change: The relative changes in dominance, colonization, and extinction numbers between the beginning and end of this study (2020, 2024).
Here is a guide to each tab:
Tab: data_diversity
Column Explanation
Year: Year of data collection from 2021 to 2024
Block: Experimental block within which grazing treatments were embedded
Treatment: Grazing treatments
- CK: no grazing
- CCS: cattle grazing intensity is twice that of sheep
- CS: the grazing intensity of cattle and sheep is the same
- CSS: sheep grazing intensity is twice that of cattle
Richness: The number of plant species in each quadrat
Beta diversity: The average pairwise dissimilarity between quadrats within each plot (using the abundance-based Bray–Curtis)
Tab: data_compared
Column Explanation
Year: Years of experimental starting and ending data collection: 2020 is the first year of data collection, and 2024 is the last year of data collection
Block: Experimental block within which grazing treatments were embedded
Treatment: Grazing treatments (see data_diversity)
Dominance: This index quantifies the relative cover of the most abundant species (%)( useing the Berger–Parker Dominance Index)
Quantity of soil available N: The content of soil nitrogen availability within quadrats(mg/kg)
Heterogeneity of soil available N: The coefficient of soil nitrogen availability variation (CV) across all five sampling quadrats within each plot(%)
RC.bray: The null modeling outcomes of the community assembly process
Tab: data_relative change
Column Explanation
Block: Experimental block within which grazing treatments were embedded
Treatment: Grazing treatments (same as above)
Number of colonization_2020-2022: The number of newly appearing plant species in the quadrat from 2020 to 2022
Number of extinction_2020-2022: The number of disappearing plant species in the quadrat from 2020 to 2022
Net_2020-2022: The differences in numbers between newly appearing species and old disappearing species in the quadrat from 2020 to 2022
Number of colonization_2020-2024: The number of newly appearing plant species in the quadrat from 2020 to 2024
Number of extinction_2020-2024: The number of disappearing plant species in the quadrat from 2020 to 2024
Net_2020-2024: The differences in numbers between newly appearing species and old disappearing species in the quadrat from 2020 to 2024
Change of dominance_2020-2024: Relative changes in dominance from 2020 to 2024(%)
Code/Software
All data were analyzed using the open-source software R version 3.6.2.
