Datasets

Author

Affiliation

Published

2026/06/14

This page is a reference point for every dataset used in BCB743. First, it explains where the data live and how the code reaches them, because the module relies on three types of sources. Second, it lists and describes the datasets, with links, so you can read the original documentation before you analyse anything. Third, it records which dataset is used where (in which chapter and in which Task) so that when a method needs a worked example you can find the matching data quickly.

Almost every mistake in multivariate ecology begins before any model is fitted, for example, a species matrix is confused with an environmental matrix, sites and samples are transposed, or double zeros are left to mislead a Euclidean method. Knowing the shape and origin of a dataset is therefore the first analytical step.

How the data are organised

Make the datasets available to your R session in one of three ways.

In-package data ship inside an R package and load with data(). After library(vegan), the call data(dune) places the dune meadow species matrix in your workspace, and data(dune.env) its matching environmental table. No file path is involved. The vegan and ade4 packages supply most of the teaching datasets used here.
Repository data are files stored in the course repository under data/BCB743/. The code reads them with here::here(), which builds a path from the project root so the same line works on any machine:

spp <- read.csv(here::here("data", "BCB743", "seaweed", "SeaweedSpp.csv"))
load(here::here(
  "data",
  "BCB743",
  "NEwR-2ed_code_data",
  "NEwR2-Data",
  "Doubs.RData"
))

External data are hosted elsewhere — most often on David Zelený’s Analysis of community ecology data in R site, or at original sources such as the WHO. The download links below point to those origins.

A note on the download links

The 💾 links point to the file as it sits in the course repository on GitHub. Package datasets (vegan, ade4) are not files you download; you obtain them by installing the package and calling data(). Their links lead to the package help page that documents the variables.

Datasets used in BCB743

The table below is the quick index: each dataset, where it comes from, and every place it is used. Chapter numbers follow the sidebar; Task letters follow the Tasks. A dataset that appears in both a chapter and a Task is doing double duty as a worked example and as an exercise.

Dataset	Source / package	Used in chapters	Used in Tasks
Doubs River fish (`spe`, `env`, `spa`)	NEwR / repo; also ade4 `doubs`	5 Correlations, 7 Ordination, 8a PCA, 9a CA, 9b DCA, 10 PCoA, 11a nMDS, 14 Cluster, 17 Multiple regression, 18 GLM, 19 GAM	A, B, D, E, F, H, I
Oribatid mite (`mite`, `mite.env`)	vegan; also ade4 `oribatid`	7 Ordination, 9a CA, 13b Two Oceans, 15 Model building	G, H, L, M
Dune meadow (`dune`, `dune.env`)	vegan	9b DCA	E, G
Lichen pastures (`varespec`, `varechem`)	vegan	—	J
Sibbo/Sipoo birds (`sipoo`, `sipoo.map`)	vegan	—	L
Pyrifos mesocosms (`pyrifos`)	vegan	—	N
Barro Colorado Island (`BCI`)	vegan	listed in module reading	—
Iris flowers (`iris`)	base datasets	8b PCA examples, 14 Cluster	—
WHO SDG indicators	repo `WHO/`; source WHO	8c PCA SDG, 14 Cluster	C, I
South African seaweeds	repo `seaweed/`; Smit et al. 2017	10 PCoA, 13a db-RDA (+ revised), 13b Two Oceans, 14 Cluster, 15 Model building, 16 Deep dive, 17 Multiple regression	K
Mayombo diatoms	repo `diatoms/`; Mayombo et al. 2019	11b nMDS PERMANOVA, 20 Mixed models	—
Yushan birds (`ybirds`)	repo + Zelený	—	B, D, E
Aravo alpine plants (`aravo`)	ade4; Zelený	—	B, D

R package datasets

vegan

The vegan package is the workhorse of the module, and its built-in datasets supply most of the teaching examples. They are not files; install the package and load each with data(). The links lead to the vegan reference manual, which documents the variables in each table.

Dataset	Description	Used in BCB743
`BCI`, `BCI.env`	Barro Colorado Island tree counts: a 50-ha tropical forest plot, 50 sites × 225 species, with site environment	Listed in module reading
`dune`, `dune.env`	Vegetation and environment in Dutch dune meadows: 20 sites × 30 species, plus five environmental descriptors	9b; Tasks E, G
`dune.taxon`, `dune.phylodis`	Taxonomic classification and phylogeny of the dune meadow species	—
`mite`, `mite.env`, `mite.pcnm`, `mite.xy`	Oribatid mites from a peat bog: 70 cores × 35 species, with substrate variables and spatial coordinates	7, 9a, 13b, 15; Tasks G, H, L, M
`pyrifos`	Response of aquatic invertebrates to the insecticide chlorpyrifos: 12 mesocosm ditches sampled over 11 weeks	Task N
`sipoo`, `sipoo.map`	Birds on the islands of the Sipoo (Sibbo) archipelago: presence/absence by island, with island areas	Task L
`varespec`, `varechem`	Lichen pastures (reindeer grazing): 24 sites × 44 species cover, plus 14 soil-chemistry variables	Task J

ade4 and related packages

ade4 (Analysis of Ecological Data) is the second major multivariate toolbox in R. It is less central to BCB743 than vegan, but it ships its own copy of the Doubs data (doubs) and a copy of the oribatid mite data (oribatid), and it is the original home of the aravo alpine-plant dataset used in Tasks B and D. The package carries over a hundred data objects; the table below is a selection of the ecologically relevant ones. For the complete catalogue, see the ade4 reference index.

Dataset	Description	Dimensions
`aravo`	Distribution of alpine plants in Aravo (Valloire, France) — species, traits and environment	75 plots × 82 species, 6 environmental variables, 8 plant traits
`doubs`	The Doubs river fish data as a pair of ecological tables (species and environment)	30 sites × 27 fish species, 11 environmental variables, plus spatial coordinates
`oribatid`	Oribatid mite community, substrate and spatial data (Borcard)	70 cores × 35 species, 5 environmental variables, plus coordinates
`mafragh`	Phyto-ecological survey of the Mafragh plain, Algeria	97 sites × 56 plant species, 11 environmental variables, plus coordinates, traits and phylogeny
`meaudret`	Meaudret river: sites × environment and sites × species, across seasons and locations	20 sites × 13 mayfly species, 9 environmental variables, in a 5-site × 4-season design
`macroloire`	Macroinvertebrate assemblages in the Loire river, France	38 sites × 40 species, fuzzy-coded traits, 6 environmental descriptors
`rhone`	Physico-chemistry of the Rhône river	39 water samples × 15 variables, plus dates and tributary discharge
`trichometeo`	Caddisfly (Trichoptera) catches paired with meteorological data	49 trapping nights × 17 species, 11 meteorological variables
`rpjdl`	Avifauna and vegetation structure (Prodon’s bird data)	182 sites × 51 bird species, 8 environmental variables
`avimedi`	Mediterranean bird fauna for constrained ordination	302 sites × 51 species, with region and vegetation-structure factors
`aviurba`	Urban birds: a triplet of species, traits and environment tables	51 sites × 40 species, 11 environmental variables, 4 traits
`tarentaise`	Mountain avifauna of the Tarentaise valley	376 sites × 98 species, 14 environmental variables, 29 traits
`steppe`	Vegetation transect across a steppe gradient	512 relevés × 37 species (presence/absence)
`woangers`	Plant assemblages in woodlands around Angers, France	Quadrat presence/absence with 13 functional traits (21 columns); site/species counts not stated
`piosphere`	Plant-trait response to a grazing gradient (the “piosphere”)	Species cover, traits and environment across sites; exact counts not stated
`atlas`	A small ecological dataset (Corsican birds, climate and geography)	23 regions, 15 bird species, 7 meteorological variables, plus coordinates and maps
`butterfly`	Genetics–ecology–environment triple for Euphydryas butterflies	16 colonies, 4 environmental variables, 6 allele frequencies, plus coordinates
`ecomor`	Ecomorphological convergence: bird morphology, diet and foraging	129 bird species, 8 morphological variables, plus diet, foraging and taxonomy

A few related packages extend ade4 and appear in the more advanced chapters:

adespatial — spatial multivariate analysis: Moran’s eigenvector maps (MEM), asymmetric eigenvector maps (AEM), and forward selection (forward.sel(), which absorbed the older packfor). Used in the gradient and model-building material. See its vignettes for worked examples.
adegraphics — the modern graphics layer for ade4 ordination diagrams.
betapart — partitions β-diversity into turnover and nestedness components; used alongside the seaweed analyses.

Tangled Bank repository datasets

These datasets are stored as files in the course repository under data/BCB743/. They include the multivariate teaching sets that are not packaged in R, plus several datasets reserved for the integrative assignment and self-study.

The seaweed data

The seaweed data are the backbone of the constrained-ordination half of the module, and they are described in their own section below. The files live in data/BCB743/seaweed/.

File	Contents
💾 `SeaweedSpp.csv`	Species (Y) matrix: presence of macroalgal species in each of 58 coastal sections
💾 `SeaweedEnv.RData`	Environmental (E) matrix: in situ seawater temperature descriptors
💾 `SeaweedEnv.csv`	The environmental matrix as CSV
💾 `bioregions.csv`	Bioregion assignment (BMP, B-ATZ, AMP, ECTZ) for each section
💾 `SeaweedSites.csv`, 💾 `sites.csv`	Section coordinates and identifiers
💾 `Seaweed_geodist.csv`	Geographic (coastal) distances between sections
💾 `macroalgae.csv`	The full macroalgal records

Doubs River (Numerical Ecology with R)

The Doubs data are the recurring example for the unconstrained methods. They come bundled with the Numerical Ecology with R (2nd ed.) code, stored under data/BCB743/NEwR-2ed_code_data/. The Doubs.RData file loads three objects: spe (fish abundances), env (environment), and spa (spatial coordinates).

File	Contents
💾 `Doubs.RData`	`spe`, `env`, `spa` for the Doubs river
💾 `DoubsSpe.csv`	Fish species abundances (27 species × 30 sites)
💾 `DoubsEnv.csv`	Environmental variables (11 variables × 30 sites)
💾 `DoubsSpa.csv`	Spatial (x, y) coordinates of the sites

The same folder also ships the NEwR helper functions, including 💾 cleanplot.pca.R used in the PCA chapter.

Mayombo diatoms

Serge Mayombo’s diatom data record epiphytic diatom communities on two kelp host species (Mayombo et al. 2019). They are used for the PERMANOVA/nMDS example and again for the mixed-models chapter. Files are in data/BCB743/diatoms/.

File	Contents
💾 `PB_data_matrix_abrev.csv`	Diatom species × sample abundance matrix
💾 `PB_diat_env.csv`	Sample metadata: host species, host size, and so on

WHO Sustainable Development Goals

The WHO SDG indicators drive the PCA and cluster-analysis examples on socio-economic data. The full collection — dozens of indicator CSVs plus an assembled matrix — lives in data/BCB743/WHO/.

File	Contents
💾 `SDG_complete.csv`	The assembled country × indicator matrix used in the analyses
💾 `SDG_description.csv`	A key linking indicator codes to their descriptions
💾 `WHO.zip`	The complete set of individual SDG indicator files

Yushan birds (external example)

The Yushan Mountain bird data (Ding, via Zelený) give the Tasks a second community-ecology example beyond the Doubs. They are supplied both as repository files and on Zelený’s site.

File	Contents
💾 `ybirds_spe.txt`	Bird species abundances along the elevation gradient
💾 `ybirds_env.txt`	Environmental variables for each plot
💾 `ybirds.xlsx`	The same data as a spreadsheet

Further repository datasets (assignment and self-study)

These are available in the repository for the integrative assignment and for independent practice; they are not part of a numbered worked example.

Dataset	Files	Source
Barents Sea fish	💾 `BarentsFish_spp.csv`, 💾 `BarentsFish_env.csv`	Fish catches and environment in the Barents Sea
Macintyre kwongan/woodland	💾 folder	Plant community and environment matrices (Macintyre et al. 2018)
Thomsen biodiversity–stress	💾 folder	Long-term community data (Thomsen, Jørgensen et al. 1992–2009)
Data dictionary	💾 `data-dictionary.csv`	A key to the variables across the BCB743 datasets

David Zelený’s example data

David Zelený’s Analysis of community ecology data in R hosts a large, well-documented collection of example datasets. Several of them coincide with the package datasets above (the same Doubs, mite and dune data), and his pages are an excellent companion to the methods chapters. The complete catalogue follows; datasets used in BCB743 are flagged.

Vegetation data

Dataset	Description	Dimensions	Used in BCB743
Aravo	Alpine plant communities, Aravo, France (Choler 2005; Dray et al. 2014)	75 plots × 82 species, 6 environmental variables, 8 traits	Tasks B, D
Barley	Barley field weed community (Pyšek & Lepš)	122 plots × weed species, 3 environmental variables (species count not stated)	—
BCI	Barro Colorado Island forest permanent plot (Condit et al.)	50 1-ha plots × tree species, 7 environmental + 13 soil variables	Module reading
Bryce Canyon	Bryce Canyon vegetation (Dave Roberts)	160 plots × 169 species, 11 environmental variables	—
Carpathian wetlands	Spring-fen wetlands (Hájek, Hekera & Hájková)	70 plots × species (plants and bryophytes), 15 environmental variables	—
Danube meadow	Danube floodplain meadow (Ellenberg 1956)	≈25 plots × 94 species (48 × 171 in the download), plus Ellenberg values	—
Dune meadow	Dutch dune meadows (Jongman et al. 1995)	20 plots × 30 species, 5 environmental variables	9b; Tasks E, G
Taiwan MQU forests	Forest plots along a Taiwan elevation gradient (Li & Zelený)	9 plots × 89 woody species, 2 environmental variables, 5 leaf traits	—
Gentry’s transects	Gentry’s global forest transects	197 localities × species, with latitude, longitude, elevation, precipitation	—
Lalashan transect	Forest vegetation plots (Zelený et al.)	18 plots × woody species, ~40 environmental variables, ~16 traits	—
Nanjenshan	Nanjenshan forest dynamics plot (Sun & Hsieh)	Page restricted; dimensions not stated	—
Němčičky	Forest understory permanent plot (Chudomelová et al.)	97 samples × 274 species, plus Ellenberg values	—
Ohrazení	Wet-meadow experiment (Jan Lepš)	96 samples × 86 species, with experimental factors	—
Seedlings (removal)	Randomised-block seedling experiment (Špačková & Lepš)	16 plots (4 blocks × 4 treatments) × seedling species	—
Taiwan GBIF	Taiwan GBIF records (Liao & Chen)	Page restricted; dimensions not stated	—
Taiwan 1-ha plots	Seven 1-ha forest plots (Zelený & Li)	Page restricted; 7 plots, further dimensions not stated	—
Třebíč grasslands	Dry grasslands (David Zelený)	48 samples × 171 species, ~18 environmental variables	—
Vltava valley	Vltava river valley vegetation (David Zelený)	97 plots × 274 species, ~28 environmental variables, plus traits	—

Zoological data

Dataset	Description	Dimensions	Used in BCB743
Yushan birds	Bird communities along an elevation gradient, Taiwan (Ding)	50 sites × 59 bird species, 20 environmental variables, plus traits	Tasks B, D, E
Carabids (Finland)	Boreal-forest carabid beetles (Niemelä et al.)	Species × 5 habitat types (counts not stated)	—
Carabids (Canada)	Carabid beetles from Canada (Bergeron, Blanchet)	194 sites × species, with vegetation-structure predictors	—
Coral reefs	Indonesian coral-reef data (Warwick, Anderson)	10 transects over 6 years × 75 coral species	—
Doubs fish	Doubs river fish (Verneaux)	30 sites × 27 fish species, 11 environmental variables, plus coordinates	Extensively (5–19); Tasks A, B, D, E, F, H, I
Spring-fen molluscs	Molluscs, vegetation and water chemistry (Horsák & Hájek)	43 localities × species, 14 water-chemistry variables	—
Oribatid mites	Oribatid mites (Borcard et al.)	70 cores × 35 morphospecies, 5 environmental variables, plus coordinates	7, 9a, 13b, 15; Tasks G, H, L, M

Simulated data

Dataset	Description	Dimensions
Simulated data	Simulated community along a gradient (Minchin 1987; Fridley et al. 2007)	Up to 500 samples × ~280–296 species, 1–2 gradients (several variants)
Spatial simulated data	Spatially structured simulated community (Smith & Lundholm 2010)	144 habitats × 50 species, elevation plus coordinates

Other

Dataset	Description	Dimensions
Cookies / pastries / pizza	A light teaching example (everest4ever)	1931 recipes × 133 ingredients (presence/absence), 3 food types
Normal pig	A teaching curiosity	Stub page; dimensions not stated
Morse codes	Rothkopf’s Morse-code confusion experiment	36 × 36 confusion matrix, plus 2 code attributes
What’s Cooking?	Recipe-ingredient data (Kaggle)	~40,000 recipes × ~6,700 ingredients, 20 cuisines

The seaweed data

Because the seaweed data anchor the constrained-ordination chapters, they deserve a fuller account.

The analyses rest on two matrices. The first, Y, holds distribution records of 846 macroalgal species across 58 coastal sections, each 50 km long, spanning the South African coast. This represents about 90% of the known seaweed flora of the country, assembled from verifiable literature and from John Bolton and Rob Anderson’s own collections over three decades (Bolton 1986; Stegenga et al. 1997; Bolton and Stegenga 2002; De Clerck et al. 2005). The second, E, is a dataset of in situ coastal seawater temperatures (Smit et al. 2013), derived from daily measurements over up to 40 years.

Four bioregions structure the coast (Bolton and Anderson 2004): the Benguela Marine Province (BMP, sections 1–17), the Benguela–Agulhas Transition Zone (B-ATZ, 18–22), the Agulhas Marine Province (AMP, 19–43/44), and the East Coast Transition Zone (ECTZ, 44/45–58). The plotting functions partition and colour-code the data by bioregion so that regional patterns in β-diversity become visible.

The data and the full analysis are described in Smit et al. (2017) and worked through in the Two Oceans appendices. Background reading is provided in Smit et al. 2017 and a description of the seaweed data. The complete code and data are also on the Seaweed-beta GitHub repository.

References

Bolton JJ (1986) Marine phytogeography of the Benguela upwelling region on the west coast of southern Africa: A temperature dependent approach. Botanica Marina 29:251–256.

Bolton JJ, Anderson RJ (2004) Marine Vegetation. In: Cowling RM, Richardson DM, Pierce SM (eds) Vegetation of southern africa. Cambridge University Press, pp 348–370

Bolton JJ, Stegenga H (2002) Seaweed species diversity in South Africa. South African Journal of Marine Science 24:9–18.

De Clerck O, Bolton JJ, Anderson RJ, Coppejans E (2005) Guide to the seaweeds of KwaZulu-Natal. Scripta Botanica Belgica 33:294 pp.

Smit AJ, Roberts M, Anderson RJ, Dufois F, Dudley SFJ, Bornman TG, Olbers J, Bolton JJ (2013) A coastal seawater temperature dataset for biogeographical studies: large biases between in situ and remotely-sensed data sets around the coast of South Africa. PLOS ONE 8:e81944.

Smit AJ, Bolton JJ, Anderson RJ (2017) Seaweeds in two oceans: Beta-diversity. Frontiers in Marine Science 4:404.

Stegenga H, Bolton JJ, Anderson RJ (1997) Seaweeds of the South African west coast. Contributions of the Bolus Herbarium 18:3–637.

Reuse

CC BY-NC-SA 4.0

Citation

BibTeX citation:

@online{smit2026,
  author = {Smit, A. J.},
  title = {Datasets},
  date = {2026-06-14},
  url = {https://tangledbank.netlify.app/BCB743/datasets.html},
  langid = {en}
}

For attribution, please cite this work as:

Smit AJ (2026) Datasets. https://tangledbank.netlify.app/BCB743/datasets.html.

--- date: last-modified title: "Datasets" --- ```{r setup, echo=FALSE} knitr::opts_chunk$set( comment = "R>", warning = FALSE, message = FALSE ) ``` This page is a reference point for every dataset used in BCB743. First, it explains *where* the data live and how the code reaches them, because the module relies on three types of sources. Second, it lists and describes the datasets, with links, so you can read the original documentation before you analyse anything. Third, it records *which* dataset is used *where* (in which chapter and in which Task) so that when a method needs a worked example you can find the matching data quickly. Almost every mistake in multivariate ecology begins before any model is fitted, for example, a species matrix is confused with an environmental matrix, sites and samples are transposed, or double zeros are left to mislead a Euclidean method. Knowing the shape and origin of a dataset is therefore the first analytical step. ## How the data are organised Make the datasets available to your R session in one of three ways. 1. *In-package data* ship inside an R package and load with `data()`. After `library(vegan)`, the call `data(dune)` places the dune meadow species matrix in your workspace, and `data(dune.env)` its matching environmental table. No file path is involved. The **vegan** and **ade4** packages supply most of the teaching datasets used here. 2. *Repository data* are files stored in the course repository under `data/BCB743/`. The code reads them with `here::here()`, which builds a path from the project root so the same line works on any machine: ```{r eval=FALSE} spp <- read.csv(here::here("data", "BCB743", "seaweed", "SeaweedSpp.csv")) load(here::here( "data", "BCB743", "NEwR-2ed_code_data", "NEwR2-Data", "Doubs.RData" )) ``` 3. *External data* are hosted elsewhere — most often on David Zelený's [*Analysis of community ecology data in R*](https://www.davidzeleny.net/anadat-r/doku.php/en:data) site, or at original sources such as the WHO. The download links below point to those origins. ::: callout-note ## A note on the download links The 💾 links point to the file as it sits in the course repository on [GitHub](https://github.com/ajsmit/R_courses). Package datasets (**vegan**, **ade4**) are not files you download; you obtain them by installing the package and calling `data()`. Their links lead to the package help page that documents the variables. ::: ## Datasets used in BCB743 The table below is the quick index: each dataset, where it comes from, and every place it is used. Chapter numbers follow the sidebar; Task letters follow the [Tasks](tasks/Task_A.qmd). A dataset that appears in both a chapter and a Task is doing double duty as a worked example and as an exercise. | Dataset | Source / package | Used in chapters | Used in Tasks | | :--- | :--- | :--- | :--- | | **Doubs River fish** (`spe`, `env`, `spa`) | NEwR / repo; also **ade4** `doubs` | 5 Correlations, 7 Ordination, 8a PCA, 9a CA, 9b DCA, 10 PCoA, 11a nMDS, 14 Cluster, 17 Multiple regression, 18 GLM, 19 GAM | A, B, D, E, F, H, I | | **Oribatid mite** (`mite`, `mite.env`) | **vegan**; also **ade4** `oribatid` | 7 Ordination, 9a CA, 13b Two Oceans, 15 Model building | G, H, L, M | | **Dune meadow** (`dune`, `dune.env`) | **vegan** | 9b DCA | E, G | | **Lichen pastures** (`varespec`, `varechem`) | **vegan** | — | J | | **Sibbo/Sipoo birds** (`sipoo`, `sipoo.map`) | **vegan** | — | L | | **Pyrifos mesocosms** (`pyrifos`) | **vegan** | — | N | | **Barro Colorado Island** (`BCI`) | **vegan** | listed in module reading | — | | **Iris flowers** (`iris`) | base **datasets** | 8b PCA examples, 14 Cluster | — | | **WHO SDG indicators** | repo `WHO/`; source [WHO](https://www.who.int/data) | 8c PCA SDG, 14 Cluster | C, I | | **South African seaweeds** | repo `seaweed/`; Smit et al. 2017 | 10 PCoA, 13a db-RDA (+ revised), 13b Two Oceans, 14 Cluster, 15 Model building, 16 Deep dive, 17 Multiple regression | K | | **Mayombo diatoms** | repo `diatoms/`; Mayombo et al. 2019 | 11b nMDS PERMANOVA, 20 Mixed models | — | | **Yushan birds** (`ybirds`) | repo + [Zelený](https://www.davidzeleny.net/anadat-r/doku.php/en:data:ybirds) | — | B, D, E | | **Aravo alpine plants** (`aravo`) | **ade4**; [Zelený](https://www.davidzeleny.net/anadat-r/doku.php/en:data:aravo) | — | B, D | ## R package datasets ### vegan The **vegan** package is the workhorse of the module, and its built-in datasets supply most of the teaching examples. They are not files; install the package and load each with `data()`. The links lead to the **vegan** reference manual, which documents the variables in each table. | Dataset | Description | Used in BCB743 | | :--- | :--- | :--- | | [`BCI`](https://search.r-project.org/CRAN/refmans/vegan/html/BCI.html), `BCI.env` | Barro Colorado Island tree counts: a 50-ha tropical forest plot, 50 sites × 225 species, with site environment | Listed in module reading | | [`dune`](https://search.r-project.org/CRAN/refmans/vegan/html/dune.html), `dune.env` | Vegetation and environment in Dutch dune meadows: 20 sites × 30 species, plus five environmental descriptors | 9b; Tasks E, G | | [`dune.taxon`](https://search.r-project.org/CRAN/refmans/vegan/html/dune.taxon.html), `dune.phylodis` | Taxonomic classification and phylogeny of the dune meadow species | — | | [`mite`](https://search.r-project.org/CRAN/refmans/vegan/html/mite.html), `mite.env`, `mite.pcnm`, `mite.xy` | Oribatid mites from a peat bog: 70 cores × 35 species, with substrate variables and spatial coordinates | 7, 9a, 13b, 15; Tasks G, H, L, M | | [`pyrifos`](https://search.r-project.org/CRAN/refmans/vegan/html/pyrifos.html) | Response of aquatic invertebrates to the insecticide chlorpyrifos: 12 mesocosm ditches sampled over 11 weeks | Task N | | [`sipoo`](https://search.r-project.org/CRAN/refmans/vegan/html/sipoo.html), `sipoo.map` | Birds on the islands of the Sipoo (Sibbo) archipelago: presence/absence by island, with island areas | Task L | | [`varespec`](https://search.r-project.org/CRAN/refmans/vegan/html/varechem.html), [`varechem`](https://search.r-project.org/CRAN/refmans/vegan/html/varechem.html) | Lichen pastures (reindeer grazing): 24 sites × 44 species cover, plus 14 soil-chemistry variables | Task J | ### ade4 and related packages **ade4** (*Analysis of Ecological Data*) is the second major multivariate toolbox in R. It is less central to BCB743 than **vegan**, but it ships its own copy of the Doubs data (`doubs`) and a copy of the oribatid mite data (`oribatid`), and it is the original home of the `aravo` alpine-plant dataset used in Tasks B and D. The package carries **over a hundred** data objects; the table below is a selection of the ecologically relevant ones. For the complete catalogue, see the [ade4 reference index](https://search.r-project.org/CRAN/refmans/ade4/html/00Index.html). | Dataset | Description | Dimensions | | :--- | :--- | :--- | | [`aravo`](https://search.r-project.org/CRAN/refmans/ade4/html/aravo.html) | Distribution of alpine plants in Aravo (Valloire, France) — species, traits and environment | 75 plots × 82 species, 6 environmental variables, 8 plant traits | | [`doubs`](https://search.r-project.org/CRAN/refmans/ade4/html/doubs.html) | The Doubs river fish data as a pair of ecological tables (species and environment) | 30 sites × 27 fish species, 11 environmental variables, plus spatial coordinates | | [`oribatid`](https://search.r-project.org/CRAN/refmans/ade4/html/oribatid.html) | Oribatid mite community, substrate and spatial data (Borcard) | 70 cores × 35 species, 5 environmental variables, plus coordinates | | [`mafragh`](https://search.r-project.org/CRAN/refmans/ade4/html/mafragh.html) | Phyto-ecological survey of the Mafragh plain, Algeria | 97 sites × 56 plant species, 11 environmental variables, plus coordinates, traits and phylogeny | | [`meaudret`](https://search.r-project.org/CRAN/refmans/ade4/html/meaudret.html) | Meaudret river: sites × environment and sites × species, across seasons and locations | 20 sites × 13 mayfly species, 9 environmental variables, in a 5-site × 4-season design | | [`macroloire`](https://search.r-project.org/CRAN/refmans/ade4/html/macroloire.html) | Macroinvertebrate assemblages in the Loire river, France | 38 sites × 40 species, fuzzy-coded traits, 6 environmental descriptors | | [`rhone`](https://search.r-project.org/CRAN/refmans/ade4/html/rhone.html) | Physico-chemistry of the Rhône river | 39 water samples × 15 variables, plus dates and tributary discharge | | [`trichometeo`](https://search.r-project.org/CRAN/refmans/ade4/html/trichometeo.html) | Caddisfly (Trichoptera) catches paired with meteorological data | 49 trapping nights × 17 species, 11 meteorological variables | | [`rpjdl`](https://search.r-project.org/CRAN/refmans/ade4/html/rpjdl.html) | Avifauna and vegetation structure (Prodon's bird data) | 182 sites × 51 bird species, 8 environmental variables | | [`avimedi`](https://search.r-project.org/CRAN/refmans/ade4/html/avimedi.html) | Mediterranean bird fauna for constrained ordination | 302 sites × 51 species, with region and vegetation-structure factors | | [`aviurba`](https://search.r-project.org/CRAN/refmans/ade4/html/aviurba.html) | Urban birds: a triplet of species, traits and environment tables | 51 sites × 40 species, 11 environmental variables, 4 traits | | [`tarentaise`](https://search.r-project.org/CRAN/refmans/ade4/html/tarentaise.html) | Mountain avifauna of the Tarentaise valley | 376 sites × 98 species, 14 environmental variables, 29 traits | | [`steppe`](https://search.r-project.org/CRAN/refmans/ade4/html/steppe.html) | Vegetation transect across a steppe gradient | 512 relevés × 37 species (presence/absence) | | [`woangers`](https://search.r-project.org/CRAN/refmans/ade4/html/woangers.html) | Plant assemblages in woodlands around Angers, France | Quadrat presence/absence with 13 functional traits (21 columns); site/species counts not stated | | [`piosphere`](https://search.r-project.org/CRAN/refmans/ade4/html/piosphere.html) | Plant-trait response to a grazing gradient (the "piosphere") | Species cover, traits and environment across sites; exact counts not stated | | [`atlas`](https://search.r-project.org/CRAN/refmans/ade4/html/atlas.html) | A small ecological dataset (Corsican birds, climate and geography) | 23 regions, 15 bird species, 7 meteorological variables, plus coordinates and maps | | [`butterfly`](https://search.r-project.org/CRAN/refmans/ade4/html/butterfly.html) | Genetics–ecology–environment triple for *Euphydryas* butterflies | 16 colonies, 4 environmental variables, 6 allele frequencies, plus coordinates | | [`ecomor`](https://search.r-project.org/CRAN/refmans/ade4/html/ecomor.html) | Ecomorphological convergence: bird morphology, diet and foraging | 129 bird species, 8 morphological variables, plus diet, foraging and taxonomy | A few **related packages** extend **ade4** and appear in the more advanced chapters: - [**adespatial**](https://cran.r-project.org/package=adespatial) — spatial multivariate analysis: Moran's eigenvector maps (MEM), asymmetric eigenvector maps (AEM), and forward selection (`forward.sel()`, which absorbed the older **packfor**). Used in the gradient and model-building material. See its vignettes for worked examples. - [**adegraphics**](https://cran.r-project.org/package=adegraphics) — the modern graphics layer for **ade4** ordination diagrams. - [**betapart**](https://cran.r-project.org/package=betapart) — partitions β-diversity into turnover and nestedness components; used alongside the seaweed analyses. ## Tangled Bank repository datasets These datasets are stored as files in the course repository under `data/BCB743/`. They include the multivariate teaching sets that are not packaged in R, plus several datasets reserved for the integrative assignment and self-study. ### The seaweed data {#sec-seaweed} The seaweed data are the backbone of the constrained-ordination half of the module, and they are described in their own section below. The files live in `data/BCB743/seaweed/`. | File | Contents | | :--- | :--- | | [💾 `SeaweedSpp.csv`](../data/BCB743/seaweed/SeaweedSpp.csv) | Species (Y) matrix: presence of macroalgal species in each of 58 coastal sections | | [💾 `SeaweedEnv.RData`](../data/BCB743/seaweed/SeaweedEnv.RData) | Environmental (E) matrix: *in situ* seawater temperature descriptors | | [💾 `SeaweedEnv.csv`](../data/BCB743/seaweed/SeaweedEnv.csv) | The environmental matrix as CSV | | [💾 `bioregions.csv`](../data/BCB743/seaweed/bioregions.csv) | Bioregion assignment (BMP, B-ATZ, AMP, ECTZ) for each section | | [💾 `SeaweedSites.csv`](../data/BCB743/seaweed/SeaweedSites.csv), [💾 `sites.csv`](../data/BCB743/seaweed/sites.csv) | Section coordinates and identifiers | | [💾 `Seaweed_geodist.csv`](../data/BCB743/seaweed/Seaweed_geodist.csv) | Geographic (coastal) distances between sections | | [💾 `macroalgae.csv`](../data/BCB743/seaweed/macroalgae.csv) | The full macroalgal records | ### Doubs River (Numerical Ecology with R) The Doubs data are the recurring example for the unconstrained methods. They come bundled with the *Numerical Ecology with R* (2nd ed.) code, stored under `data/BCB743/NEwR-2ed_code_data/`. The `Doubs.RData` file loads three objects: `spe` (fish abundances), `env` (environment), and `spa` (spatial coordinates). | File | Contents | | :--- | :--- | | [💾 `Doubs.RData`](../data/BCB743/NEwR-2ed_code_data/NEwR2-Data/Doubs.RData) | `spe`, `env`, `spa` for the Doubs river | | [💾 `DoubsSpe.csv`](../data/BCB743/DoubsSpe.csv) | Fish species abundances (27 species × 30 sites) | | [💾 `DoubsEnv.csv`](../data/BCB743/DoubsEnv.csv) | Environmental variables (11 variables × 30 sites) | | [💾 `DoubsSpa.csv`](../data/BCB743/DoubsSpa.csv) | Spatial (x, y) coordinates of the sites | The same folder also ships the NEwR helper functions, including [💾 `cleanplot.pca.R`](../data/BCB743/NEwR-2ed_code_data/NEwR2-Functions/cleanplot.pca.R) used in the PCA chapter. ### Mayombo diatoms Serge Mayombo's diatom data record epiphytic diatom communities on two kelp host species (Mayombo et al. 2019). They are used for the PERMANOVA/nMDS example and again for the mixed-models chapter. Files are in `data/BCB743/diatoms/`. | File | Contents | | :--- | :--- | | [💾 `PB_data_matrix_abrev.csv`](../data/BCB743/diatoms/PB_data_matrix_abrev.csv) | Diatom species × sample abundance matrix | | [💾 `PB_diat_env.csv`](../data/BCB743/diatoms/PB_diat_env.csv) | Sample metadata: host species, host size, and so on | ### WHO Sustainable Development Goals The WHO SDG indicators drive the PCA and cluster-analysis examples on socio-economic data. The full collection — dozens of indicator CSVs plus an assembled matrix — lives in `data/BCB743/WHO/`. | File | Contents | | :--- | :--- | | [💾 `SDG_complete.csv`](../data/BCB743/WHO/SDG_complete.csv) | The assembled country × indicator matrix used in the analyses | | [💾 `SDG_description.csv`](../data/BCB743/WHO/SDG_description.csv) | A key linking indicator codes to their descriptions | | [💾 `WHO.zip`](../data/BCB743/WHO/WHO.zip) | The complete set of individual SDG indicator files | ### Yushan birds (external example) The Yushan Mountain bird data (Ding, via Zelený) give the Tasks a second community-ecology example beyond the Doubs. They are supplied both as repository files and on Zelený's site. | File | Contents | | :--- | :--- | | [💾 `ybirds_spe.txt`](../data/BCB743/ybirds_spe.txt) | Bird species abundances along the elevation gradient | | [💾 `ybirds_env.txt`](../data/BCB743/ybirds_env.txt) | Environmental variables for each plot | | [💾 `ybirds.xlsx`](../data/BCB743/ybirds.xlsx) | The same data as a spreadsheet | ### Further repository datasets (assignment and self-study) These are available in the repository for the integrative assignment and for independent practice; they are not part of a numbered worked example. | Dataset | Files | Source | | :--- | :--- | :--- | | **Barents Sea fish** | [💾 `BarentsFish_spp.csv`](../data/BCB743/BarentsFish_spp.csv), [💾 `BarentsFish_env.csv`](../data/BCB743/BarentsFish_env.csv) | Fish catches and environment in the Barents Sea | | **Macintyre kwongan/woodland** | [💾 folder](../data/BCB743/Macintyre%20et%20al%202019/) | Plant community and environment matrices (Macintyre et al. 2018) | | **Thomsen biodiversity–stress** | [💾 folder](../data/BCB743/Thomsen%20et%20al%202015/) | Long-term community data (Thomsen, Jørgensen et al. 1992–2009) | | **Data dictionary** | [💾 `data-dictionary.csv`](../data/BCB743/data-dictionary.csv) | A key to the variables across the BCB743 datasets | ## David Zelený's example data David Zelený's [*Analysis of community ecology data in R*](https://www.davidzeleny.net/anadat-r/doku.php/en:data) hosts a large, well-documented [collection of example datasets](https://www.davidzeleny.net/anadat-r/doku.php/en:data). Several of them coincide with the package datasets above (the same Doubs, mite and dune data), and his pages are an excellent companion to the methods chapters. The complete catalogue follows; datasets used in BCB743 are flagged. ### Vegetation data | Dataset | Description | Dimensions | Used in BCB743 | | :--- | :--- | :--- | :--- | | [Aravo](https://www.davidzeleny.net/anadat-r/doku.php/en:data:aravo) | Alpine plant communities, Aravo, France (Choler 2005; Dray et al. 2014) | 75 plots × 82 species, 6 environmental variables, 8 traits | Tasks B, D | | [Barley](https://www.davidzeleny.net/anadat-r/doku.php/en:data:barley) | Barley field weed community (Pyšek & Lepš) | 122 plots × weed species, 3 environmental variables (species count not stated) | — | | [BCI](https://www.davidzeleny.net/anadat-r/doku.php/en:data:bci) | Barro Colorado Island forest permanent plot (Condit et al.) | 50 1-ha plots × tree species, 7 environmental + 13 soil variables | Module reading | | [Bryce Canyon](https://www.davidzeleny.net/anadat-r/doku.php/en:data:bryce) | Bryce Canyon vegetation (Dave Roberts) | 160 plots × 169 species, 11 environmental variables | — | | [Carpathian wetlands](https://www.davidzeleny.net/anadat-r/doku.php/en:data:wetlands) | Spring-fen wetlands (Hájek, Hekera & Hájková) | 70 plots × species (plants and bryophytes), 15 environmental variables | — | | [Danube meadow](https://www.davidzeleny.net/anadat-r/doku.php/en:data:danube) | Danube floodplain meadow (Ellenberg 1956) | ≈25 plots × 94 species (48 × 171 in the download), plus Ellenberg values | — | | [Dune meadow](https://www.davidzeleny.net/anadat-r/doku.php/en:data:dune) | Dutch dune meadows (Jongman et al. 1995) | 20 plots × 30 species, 5 environmental variables | 9b; Tasks E, G | | [Taiwan MQU forests](https://www.davidzeleny.net/anadat-r/doku.php/en:data:mqu) | Forest plots along a Taiwan elevation gradient (Li & Zelený) | 9 plots × 89 woody species, 2 environmental variables, 5 leaf traits | — | | [Gentry's transects](https://www.davidzeleny.net/anadat-r/doku.php/en:data:gentry) | Gentry's global forest transects | 197 localities × species, with latitude, longitude, elevation, precipitation | — | | [Lalashan transect](https://www.davidzeleny.net/anadat-r/doku.php/en:data:lalashan) | Forest vegetation plots (Zelený et al.) | 18 plots × woody species, ~40 environmental variables, ~16 traits | — | | [Nanjenshan](https://www.davidzeleny.net/anadat-r/doku.php/en:data:nanjenshan) | Nanjenshan forest dynamics plot (Sun & Hsieh) | Page restricted; dimensions not stated | — | | [Němčičky](https://www.davidzeleny.net/anadat-r/doku.php/en:data:nemcicky) | Forest understory permanent plot (Chudomelová et al.) | 97 samples × 274 species, plus Ellenberg values | — | | [Ohrazení](https://www.davidzeleny.net/anadat-r/doku.php/en:data:ohrazeni) | Wet-meadow experiment (Jan Lepš) | 96 samples × 86 species, with experimental factors | — | | [Seedlings (removal)](https://www.davidzeleny.net/anadat-r/doku.php/en:data:removal_experiment) | Randomised-block seedling experiment (Špačková & Lepš) | 16 plots (4 blocks × 4 treatments) × seedling species | — | | [Taiwan GBIF](https://www.davidzeleny.net/anadat-r/doku.php/en:data:taiwan-gbif) | Taiwan GBIF records (Liao & Chen) | Page restricted; dimensions not stated | — | | [Taiwan 1-ha plots](https://www.davidzeleny.net/anadat-r/doku.php/en:data:taiwan-hp) | Seven 1-ha forest plots (Zelený & Li) | Page restricted; 7 plots, further dimensions not stated | — | | [Třebíč grasslands](https://www.davidzeleny.net/anadat-r/doku.php/en:data:grasslands) | Dry grasslands (David Zelený) | 48 samples × 171 species, ~18 environmental variables | — | | [Vltava valley](https://www.davidzeleny.net/anadat-r/doku.php/en:data:vltava) | Vltava river valley vegetation (David Zelený) | 97 plots × 274 species, ~28 environmental variables, plus traits | — | ### Zoological data | Dataset | Description | Dimensions | Used in BCB743 | | :--- | :--- | :--- | :--- | | [Yushan birds](https://www.davidzeleny.net/anadat-r/doku.php/en:data:ybirds) | Bird communities along an elevation gradient, Taiwan (Ding) | 50 sites × 59 bird species, 20 environmental variables, plus traits | Tasks B, D, E | | [Carabids (Finland)](https://www.davidzeleny.net/anadat-r/doku.php/en:data:carabids) | Boreal-forest carabid beetles (Niemelä et al.) | Species × 5 habitat types (counts not stated) | — | | [Carabids (Canada)](https://www.davidzeleny.net/anadat-r/doku.php/en:data:beetles) | Carabid beetles from Canada (Bergeron, Blanchet) | 194 sites × species, with vegetation-structure predictors | — | | [Coral reefs](https://www.davidzeleny.net/anadat-r/doku.php/en:data:coral) | Indonesian coral-reef data (Warwick, Anderson) | 10 transects over 6 years × 75 coral species | — | | [Doubs fish](https://www.davidzeleny.net/anadat-r/doku.php/en:data:doubs) | Doubs river fish (Verneaux) | 30 sites × 27 fish species, 11 environmental variables, plus coordinates | Extensively (5–19); Tasks A, B, D, E, F, H, I | | [Spring-fen molluscs](https://www.davidzeleny.net/anadat-r/doku.php/en:data:spring-fens) | Molluscs, vegetation and water chemistry (Horsák & Hájek) | 43 localities × species, 14 water-chemistry variables | — | | [Oribatid mites](https://www.davidzeleny.net/anadat-r/doku.php/en:data:mite) | Oribatid mites (Borcard et al.) | 70 cores × 35 morphospecies, 5 environmental variables, plus coordinates | 7, 9a, 13b, 15; Tasks G, H, L, M | ### Simulated data | Dataset | Description | Dimensions | | :--- | :--- | :--- | | [Simulated data](https://www.davidzeleny.net/anadat-r/doku.php/en:data:simul) | Simulated community along a gradient (Minchin 1987; Fridley et al. 2007) | Up to 500 samples × ~280–296 species, 1–2 gradients (several variants) | | [Spatial simulated data](https://www.davidzeleny.net/anadat-r/doku.php/en:data:simul-spatial) | Spatially structured simulated community (Smith & Lundholm 2010) | 144 habitats × 50 species, elevation plus coordinates | ### Other | Dataset | Description | Dimensions | | :--- | :--- | :--- | | [Cookies / pastries / pizza](https://www.davidzeleny.net/anadat-r/doku.php/en:data:cookies-pastry-pizza) | A light teaching example (everest4ever) | 1931 recipes × 133 ingredients (presence/absence), 3 food types | | [Normal pig](https://www.davidzeleny.net/anadat-r/doku.php/en:data:normal_pig) | A teaching curiosity | Stub page; dimensions not stated | | [Morse codes](https://www.davidzeleny.net/anadat-r/doku.php/en:data:morse) | Rothkopf's Morse-code confusion experiment | 36 × 36 confusion matrix, plus 2 code attributes | | [What's Cooking?](https://www.davidzeleny.net/anadat-r/doku.php/en:data:recipes) | Recipe-ingredient data (Kaggle) | ~40,000 recipes × ~6,700 ingredients, 20 cuisines | ## The seaweed data {#sec-seaweed-detail} Because the seaweed data anchor the constrained-ordination chapters, they deserve a fuller account. The analyses rest on two matrices. The first, **Y**, holds distribution records of 846 macroalgal species across 58 coastal sections, each 50 km long, spanning the South African coast. This represents about 90% of the known seaweed flora of the country, assembled from verifiable literature and from John Bolton and Rob Anderson's own collections over three decades [@Bolton1986; @Stegenga1997; @Bolton2002; @DeClerck2005]. The second, **E**, is a dataset of *in situ* coastal seawater temperatures [@Smit2013], derived from daily measurements over up to 40 years. Four bioregions structure the coast [@Bolton2004]: the Benguela Marine Province (BMP, sections 1–17), the Benguela–Agulhas Transition Zone (B-ATZ, 18–22), the Agulhas Marine Province (AMP, 19–43/44), and the East Coast Transition Zone (ECTZ, 44/45–58). The plotting functions partition and colour-code the data by bioregion so that regional patterns in β-diversity become visible. The data and the full analysis are described in @smit2017seaweeds and worked through in the [Two Oceans appendices](two_oceans_appendices.qmd). Background reading is provided in [Smit et al. 2017](../docs/Smit_et_al_2017.pdf) and [a description of the seaweed data](../docs/Smit_the_seaweed_data.pdf). The complete code and data are also on the [Seaweed-beta GitHub repository](https://github.com/ajsmit/Seaweed-beta). ## References ::: {#refs} :::