Data from: Long-term nutrient addition and grazing exclusion determine flower abundance, diversity and community composition in high-latitude grasslands
Data files
Jul 23, 2025 version files 60.57 KB
-
flower_unit_data.csv
19.88 KB
-
flower_units_2020_2023_corrected.csv
38.98 KB
-
README.md
1.71 KB
Abstract
The abundance and composition of flowers within plant communities shape the resources available to flower-visiting insects. However, it remains unclear how nutrient enrichment and changes in grazing impact plant communities and their floral resources, and whether these effects contribute to ongoing pollinator decline. We investigated how the abundance, diversity, and species composition of flowers (excluding graminoids) respond to long-term nutrient additions and grazing exclusion in two high-latitude grasslands (tundra and montane). We used two factorial experimental designs: 1) a nutrient experiment with additions of nitrogen (N), phosphorus (P), and potassium (K with micronutrients, and 2) a grazing experiment that included combined NPK addition with and without grazing exclusion. Flower abundance was assessed in 5 × 5 m experimental plots after 6–10 years of treatments. N and P interactively affected flower abundance, with N counteracting the positive effects of P. N reduced flower abundance in the montane grassland, whereas P and K increased it. Grazing exclusion increased flower abundance, with its effect amplified under NPK addition, but this joint treatment reduced flower species richness. Grazing exclusion reduced flower species diversity, with contrasting site-specific effects depending on NPK. N, P, and K addition, as well as grazing exclusion and NPK addition, changed the composition of flower assemblages. At the montane grassland, joint NPK addition and grazing exclusion led to strong dominance of forbs, such as Anthriscus sylvestris and Geranium sylvaticum, and grazing exclusion led to homogenisation of the flower community at the montane grassland. Our study demonstrates the importance of grazing and nutrients as key determinants of floral resources, which are vital for flower-visiting insects. We suggest that quantifying floral resources could contribute to ecosystem status assessment. Our findings offer insights for managing and conserving pollinator habitats in the face of environmental change.
Metadata of flower_unit_data.csv:
- Site = factor with two levels: the tundra (kilp.fi) and the montane (saana.fi) site
- Block = factor with four levels: the number of the block
- Blocks = factor with four levels: the number of the block separated for kilp.fi (k) and saana.fi(s) sites
- Block_Plot = plot code
- Plot_nr = plot number within the site
- N = factor of addition, levels '1=yes' or '0=no'
- P = factor of P addition, levels '1=yes' or '0=no'
- K = factor of K addition with micronutrients, levels '1=yes' or '0=no'
- Fence = factor grazing exclusion, levels '1=yes' or '0=no'
- NPK = factor NPK addition, levels '1=yes' or '0=no'
- treatm = treatment code
- treatm4 = factor to subset data with NPK-Fence treatments
- treatm8 = factor to subset data with N, P, K addition treatments
- species columns (N-BL) mean flower abundance per 1 sq. m (mean of 2020 and 2023)
- family columns (BM - BU) family-wise sums of flower abundance per 1 sq. m (2020 and 2023)
- other = collected values for the other families
- invsimp inverse Simpson's index values
- fu.tot = total of flower units (sum of all species)
- fu.rich = flower species richness (flower species per 25 sq. m)
Metadata of flower_units_2020_2023_corrected.csv:
As above but additionally;
- nr = a sequential number
- Site = factor with two levels: the tundra (kilp.fi) and the montane (saana.fi) site
- Year = sampling year, levels 2020 or 2023
- trts.started = year when experimental treatments started
- columns E-CC correspond to C-BZ in flower_unit_data