Supplementary information
Article: Contextualized niche shifts upon independent invasions by the dung beetleOnthophagus taurus
List of institutions holding Onthophagus taurus occurrences
· Instituto de Investigación CIBIO, Universidad de Alicante: CEUA
· Texas A&M University Insect Collection
· BANDASCA, BAse de DAtos sobre SCArabaeidae
· IMEDEA-INSECTA
· Banc de Dades de Biodiversitat de Catalunya-ArtroCat
· Lund Museum of Zoology - Insect collections (MZLU)
· Museo Nacional de Ciencias Naturales, Madrid. Colección de Tejidos y AND
· Entomological Specimens of Museum of Nature and Human Activities, Hyogo Pref., Japan
· Natural History Museum Rotterdam
· Computarización de la Colección Nacional de Insectos Dr. Alfredo Barrera Marín del Museo de Historia Natural de la Ciudad de México. Coleóptera
· RBINS collections
· Queen Victoria Museum Art Gallery provider for OZCAM
· Australian Museum provider for OZCAM
· California Beetle Project
· Inventaire National du Patrimoine Naturel (I001): Insectes Coléoptères Scarabéidés Laparostici de France
· Naturkundemuseum im Ottoneum Kassel, Entomological Collection
· Biologiezentrum Linz
· ATBI in GemerArea (Slovakia)
· Nottinghamshire Biological and Geological Records Centre - UK abstract from Nottingham City Museums & Galleries (NCMG) Insect Collection Baseline database
· Natural England - Invertebrate Site Register – England
· Coleoptera identified from the Borax Works, Norman Road site, Bexley, London
List of papers holding Onthophagus taurus occurrences
Bertone M, Green J, Washburn S, Poore M, Sorenson C, Watson DW, Bertone M (2005) Seasonal activity and species composition of dung beetles (Coleoptera: Scarabaeidae and Geotrupidae) inhabiting cattle pastures in North Carolina. Annals of the Entomological Society of America 98:309–321.
Edwards P (2007) Introduced Dung Beetles in Australia 1967-2007: Current Status and Future Directions. 79.
Fiene JG, Connior MB, Androw R, Baldwin B, Mckay T (2011) Surveys of Arkansas dung beetles (Coleoptera: Scarabaeidae and Geotrupidae): Phenologies, mass occurrences, state and distributional records. American Midland Naturalist 165:319–337.
Fincher GT, Stewart TB, Hunter III JS (1983) Distribution of Onthophagus gazella Fabricius from releases in Texas and Onthophagus taurus Schreber from an unknown release in Florida (Coleoptera: Scarabaeidae). Coleopterists Bulletin 37:159–163.
Fincher GT, Woodruff RE (1975) A European dung beetle, Onthophagus taurus Schreber, new to the U.S. (Coleoptera: Scarabaeidae). Coleopterists Bulletin 29:349–350.
Guichard D (1999) Scarabaeides de la frange littorale vendéenne. Doctorate thesis; Université de Nantes; p 22.
House CM, Simmons LW, Kotiaho JS, Tomkins JL, Hunt J (2010) Sex ratio bias in the dung beetle Onthophagus taurus: adaptive allocation or sex-specific offspring mortality? Evolutionary Ecology 25:363–372.
Kaufman PE, Wood LA (2012) Indigenous and exotic dung beetles (Coleoptera: Scarabaeidae and Geotrupidae) collected in Florida cattle pastures. Annals of the Entomological Society of America 105:225–231.
Lobo JM, Guéorguiev B, Chehlarov E (2007) Convergences and divergences between two European mountain dung beetle assemblages (Coleoptera , Scarabaeoidea ). Animal Biodiversity and Conservation 30.1:83–96.
MacRae TC, Penn SR (2012) Additional records of adventive Onthophagus Latreille (Coleptera: Scarabaeidae) in North America. Coleopterists Bulletin 55:49–50.
Moczek AP (2003) The behavioral ecology of threshold evolution in a polyphenic beetle. Behavioral Ecology 14:841–854.
Moczek AP, Nijhout HF (2003) Rapid evolution of a polyphenic threshold. Evolution & Development 5:259–68.
Palestrini C, Rolando A (2001) Body size and paternal investment in the genus Onthophagus (Coleoptera, Scarabaeoidea). Journal of Zoology 255:405–412.
Rounds RJ, Floate KD (2012) Diversity and seasonal phenology of Coprophagous beetles at Lake City, Michigan, USA, with a new state record for Onthophagus taurus(Schreber) (Coleoptera: Scarabaeidae). Coleopterists Bulletin 66:169–172.
Vulinec K, Eudy SP (2012) A southern range extension for the introduced dung beetle Onthophagus taurus Schreber (Coleoptera: Scarabaeidae). Coleopterists Bulletin 47:129–130.
Supplementary tables
Table S1 – Pair-wise niche overlap comparisons (D) between each of the invaded and natural ranges of O. taurus.
NAw / NAe / Native / AUSw / AUSe1à2 / NAw / NA / NA / NA / NA / NA
NAe / 0.161 / NA / NA / NA / NA
Native / 0.442 / 0.180 / NA / NA / NA
AUSw / 0.218 / 0.150 / 0.182 / NA / NA
AUSe / 0.251 / 0.252 / 0.299 / 0.214 / NA
Native: Native Mediterranean Range; AUSe: Eastern Australian Range; AUSw: Western Australian Range; NAe: Eastern North American Range; NAw: Western North American Range; IRs: Invaded Ranges;
Table S2 – Pair-wise niche similarity comparisons (p-values) between each of the invaded and natural ranges of O. taurus.
2à1NAw / NAe / Native / AUSw / AUSe
1à2 / NAw / NA / 0.465 / 0.01 / 0.010 / 0.010
NAe / 0.139 / NA / 0.01 / 0.119 / 0.010
Native / 0.050 / 0.119 / NA / 0.020 / 0.317
AUSw / 0.406 / 0.010 / 0.01 / NA / 0.010
AUSe / 0.059 / 0.010 / 0.01 / 0.040 / NA
Native: Native Mediterranean Range; AUSe: Eastern Australian Range; AUSw: Western Australian Range; NAe: Eastern North American Range; NAw: Western North American Range; IRs: Invaded Ranges;
Table S3 – Pair-wise niche unfilling proportions between each of the invaded and natural ranges of O. taurus.
2à1NAw / NAe / Native / AUSw / AUSe
1à2 / NAw / NA / 0 / 0.069 / 0.018 / 0.015
NAe / 0 / NA / 0.052 / 0.036 / 0.027
Native / 0 / 0 / NA / 0.022 / 0.016
AUSw / 0 / 0 / 0.037 / NA / 0.000
AUSe / 0 / 0 / 0.020 / 0.000 / NA
Native: Native Mediterranean Range; AUSe: Eastern Australian Range; AUSw: Western Australian Range; NAe: Eastern North American Range; NAw: Western North American Range; IRs: Invaded Ranges;
Table S4 – Pair-wise niche expansion proportions between each of the invaded and natural ranges of O. taurus.
2à1NAw / NAe / Native / AUSw / AUSe
1à2 / NAw / NA / 0.000 / 0.000 / 0.000 / 0.00
NAe / 0.000 / NA / 0.000 / 0.000 / 0.00
Native / 0.069 / 0.052 / NA / 0.037 / 0.02
AUSw / 0.018 / 0.036 / 0.022 / NA / 0.00
AUSe / 0.015 / 0.027 / 0.016 / 0.000 / NA
Native: Native Mediterranean Range; AUSe: Eastern Australian Range; AUSw: Western Australian Range; NAe: Eastern North American Range; NAw: Western North American Range; IRs: Invaded Ranges;
Table S5 – Pair-wise niche stability proportions between each of the invaded and natural ranges of O. taurus.
2à1NAw / NAe / Native / AUSw / AUSe
1à2 / NAw / NA / 1.000 / 1.000 / 1.000 / 1.00
NAe / 1.000 / NA / 1.000 / 1.000 / 1.00
Native / 0.931 / 0.948 / NA / 0.963 / 0.98
AUSw / 0.982 / 0.964 / 0.978 / NA / 1.00
AUSe / 0.985 / 0.973 / 0.984 / 1.000 / NA
Native: Native Mediterranean Range; AUSe: Eastern Australian Range; AUSw: Western Australian Range; NAe: Eastern North American Range; NAw: Western North American Range; IRs: Invaded Ranges;
Supplementary figures
Figure S1 – PCA-env results and the relationship of each environmental variable related to the distribution of Onthophagus taurus in all of its native and introduced ranges. A) The first two PCA axes and the orientation of each environmental variable in each axis. B) Individual contribution of each environmental variable to the first PCA axis. C) Individual contribution of each environmental variable to the first PCA axis.
Onthophagus taurus invasion: Code description
About
This document serves as a guide to reproduce the analyses and results presented in the manuscript: "Contextualized niche shifts upon independent invasions by the dung beetle Onthophagus taurus" by Daniel P. Silva, Bruno Vilela, Bruno A. Buzatto, Armin P. Moczek and Joaquín Hortal.
The objective of this document is to record the code used to generate the results and allow the readers to better explore the possibilities of the research.
This document was written in R markdown format, which allows the use of easy-formatting plain text with R code chunks. For more information see the package knitr (http://yihui.name/knitr/).
Before starting
Prior to the analyses, we recommend all users to check the latest version of R at http://www.r-project.org/ and to make sure that they are using the updated versions of their installed R packages. Users can automatically update their installed packages with the following code:
# Update installed packages
update.packages(checkBuilt = TRUE, ask = FALSE)
The analysis presented here makes use of the following R packages available at CRAN. Use the following code to install them.
# Install packages
install.packages("knitr")
install.packages("spThin")
install.packages("rgeos")
install.packages("sp")
install.packages("maptools")
install.packages("raster")
install.packages("ecospat")
Once installed, load them.
# Load packages
library(knitr)
library(spThin)
library(rgeos)
library(sp)
library(maptools)
library(raster)
library(ecospat)
Data
Load the occurrence records
Place the file containing the occurrence records (file points.txt) in your work directory (use getwd() to check your work directory). The next step is to load the occurrence records into the R environment.
occ.points <- read.table("points.txt", sep = "\t", header = TRUE)
The loaded table includes 1272 occurrence records.
Thining occurrence records
The occurrence records gathered (see the methods section of the manuscript, for the description of how we obtained the data) are not free from geographical sample bias. To minimize this problem, we applied a thinning procedure using the spThin package to make sure that all the points have at least a minimum distance of 10 km from each other (see Aiello-Lammens et al. 2014 for the algorithm description).
occ.points.thin <- thin(occ.points, verbose = FALSE,
lat.col = "Latitude",
long.col = "Longitude",
spec.col = "Scientific.name",
thin.par = 10,
reps = 1,
write.files = FALSE,
write.log.file = FALSE,
locs.thinned.list.return = TRUE)
After the thinning procedure the number of occurrence points is reduced to n = 1059.
To check the distribution of the occurrence records we map them in a world context.
data(wrld_simpl)
plot(wrld_simpl)
points(occ.points.thin[[1]], col = "purple", pch = 20, cex = 0.7)
Define the regions to be tested
The niche comparisons can be done between any group of occurrence points defined. For example, the groups can be divided into major regions, e.g. native occurrences from the Native Mediterranean Range (Southern Europe and North Africa - Native), Australia and North America, or each region can be divided into more sub-regions, e.g. Western Australia range (AUSw), Eastern Australia (AUSe), Western North America (NAw), and Eastern North America (NAe). We left the option here for the readers to define their own regions and explore the results. In the following analyzes we decided to divide the occurrence records into five (5) groups: Native, NAw, NAe, WAUS, and EAUS. We choose these regions, as we believe that their invasion history are different and independent (see details in the manuscript).
The next step is to define the number of groups (regions) to be tested. In the follow case, we choose 5 groups.
n.groups <- 5
Now, it is necessary to define the longitude limits of the region. In this specific case, only the longitude is needed to separate the groups. Change the object’s limits to define other groups, note that the object must have the n.groups - 1 length.
limits <- c(-100, -50, 100, 125)
begin <- min(occ.points.thin[[1]][, 1]) - 10
end <- max(occ.points.thin[[1]][, 1]) + 10
group.long <- c(begin, limits, end)
Next, we checked the limits by plotting them.
plot(wrld_simpl)
points(occ.points.thin[[1]], col = "purple", pch = 20, cex = 0.7)
abline(v = group.long, lty = 2, col = "red", lwd = 2)
Now, we define the name of the groups, in the same geographical order of the groups, starting from the west to east. You can also define the codes to be used in the tables.
g.names <- c("NAw",
"NAe",
"Native",
"AUSw",
"AUSe")
g.codenames <- c("NAw", "NAe", "Native", "AUSw", "AUSe")
It is also necessary to set what colors will be used in the next plots for each group (using the same order as the names). Change the colors according to your preferences.
g.colors <- c("cyan", "darkblue", "red", "green", "darkgreen")
Background definition
An important step in the niche analyses is the definition of the background. Here, we applied a background based on a minimum convex polygon (MCP) made from the occurrence records of each group. Additionally to the MCP we add a buffer around it. The polygon buffer size for the background (in degrees) can be changed below. We chose one degree based on the published values of dispersion for Onthophagus taurus (Hanski & Cambefort, 1991).
buffer.size <- 1
We defined a minimum convex polygon (MCP) function below (this function was obtained from https://github.com/ndimhypervol/wallace).
mcp <- function (xy) {
xy <- as.data.frame(coordinates(xy))
coords.t <- chull(xy[, 1], xy[, 2])
xy.bord <- xy[coords.t, ]
xy.bord <- rbind(xy.bord[nrow(xy.bord), ], xy.bord)
return(SpatialPolygons(list(Polygons(list(Polygon(as.matrix(xy.bord))), 1))))
}
Environmental variables
The environmental variables used are available at the WorldClim website (http://www.worldclim.org). Download all the 19 bioclimatic ('Biolclim') variables for the current conditions (we used the resolution of 10 arc-min) with the code below. Note that you need to have the internet on. The download files are opened directed in the R environment, but they are also saved in your work directory (to see where it is, use getwd()).
variables <- getData('worldclim', var='bio', res=10)
In the manuscript, we used the all 19 bioclimatic variables as before the analysis we will reduce them to a two-dimensional space with a PCA. However, the readers can choose the number of variables to keep by changing the sequence 1:19 in the code below for the variable number you want to keep (to see the name sequence of the variables apply names(variables)).
variables <- subset(variables, 1:19)
You can also check the variables by mapping them.
plot(variables)
Group assigning
Once, we have the occurrence data, the environmental data, the defined groups and their background parameters chosen, we can prepare the data for the analysis. Below we use the occurrence points to generate the MCP plus a buffer defined by the user as the background (see above). Next, the variable values per group are extracted from the species occurrence points and from the background (defined above). Finally, we plot the resulting groups with their respective backgrounds.