Theory, Examples & Exercises
The data originates from the research of Batterink & Wijffels (1983), published as a report in Dutch. Data were used as training example in Jongman et al. (1995) and are part of the vegan package.
This dataset was collected as a part of the project focused on the effect of management on dune meadows. It contains 20 plots of 2×2 m2, sampled in 1982 following Braun-Blanquet method estimating plant cover in each plot (using 9-grade ordinal cover scale). Total of 30 species (28 vascular and 2 bryophytes) were recorded.
The data provided here are slightly modified: two bryophyte species were removed from the original species composition dataset (namely Brachythecium rutabulum and Calliergonella cuspidata). Additionally, species attributes, namely traits (5 variables, Jamil et al. 2013) and Ellenberg indicator values (6 variables, Ellenberg et al. 1992) have been added. Plant traits from Jamil et al. (2013) were compiled from LEDA database (Kleyer et al. 2008) and from Lienin & Kleyer (2011).
To avoid confusion with the original (complete)
dune dataset in
vegan, I call the dataset
The following is the description of the dataset from Jongman et al. (1995): This set of data stems from a research project on the Dutch island of Terschelling (Batterink & Wijffels 1983). The objective of this project was to detect a possible relation between vegetation and management in dune meadows. Sampling was done in 1982. Data collection was done by the Braun-Blanquet method; the data are recorded according to the ordinal scale of van der Maarel (1979). In each parcel usually one site was selected; only in cases of great variability within the parcel were more sites used to describe the parcel. The sites were selected by throwing and object into a parcel. The point where the object landed was fixed as one corner of the site. The sites measure 2x2m2. The sites were considered to be representative of the whole parcel. From the total of 80 sites, 20 have been selected to be used in this book. This selection expresses the variation in the complete set of data. The names of the species conform with the nomenclature in van der Meijden et al. (1983) and Tutin et al. (1964-1980).
Terschelling, Dutch Island.
|A1||Thickness of the A1 horizon [cm]1)|
|Moisture||Moisture content of the soil (ordinal variable 1-5)|
|Management||Grassland management type (SF - standard farming, BF - biological farming, HF - hobby-farming, NM - nature conservation management)|
|Use||Agricultural grassland use (hayfields, pastures or combination of both; ranked nominal variable)|
|Manure||Quantity of manure applied (ordinal variable 0-4)|
|SLA||Specific leaf area [mm2 mg-1]|
|Height||Plant height [cm]|
|LDMC||Leaf dry matter content [mg g-1]|
|Seedmass||Seed mass [g]|
|Lifespan||Lifespan [annual, perennial]|
|Light||Ellenberg indicator values for light|
|Temp||Ellenberg indicator values for temperature|
|Cont||Ellenberg indicator values for continentality|
|Moist||Ellenberg indicator values for moisture|
|React||Ellenberg indicator values for soil reaction|
|Nutr||Ellenberg indicator values for nutrients|
# Species and environmental data dune2.spe <- read.delim ('https://raw.githubusercontent.com/zdealveindy/anadat-r/master/data/dune2.spe.txt', row.names = 1) dune2.env <- read.delim ('https://raw.githubusercontent.com/zdealveindy/anadat-r/master/data/dune2.env.txt', row.names = 1) # Species attributes dune2.traits <- read.delim ('https://raw.githubusercontent.com/zdealveindy/anadat-r/master/data/dune2.traits.txt', row.names = 1) dune2.ell <- read.delim ('https://raw.githubusercontent.com/zdealveindy/anadat-r/master/data/dune2.ell.txt', row.names = 1) # Original dune dataset from vegan: # If you need to install vegan package, use the following: # install.packages ('vegan') library (vegan) data (dune) # matrix with species data (20 samples in rows and 30 species in columns) data (dune.env)# matix of environmental variables (20 samples in rows and 5 environmental variables in columns)