triangle.plot(as.data.frame(fish.pod.qJ.3[, c(3, 1, 2)]),
show = FALSE,
labeltriangle = FALSE,
addmean = TRUE
)
text(-0.45, 0.5, "AbDiff", cex = 1.5)
text(0.4, 0.5, "Repl", cex = 1.5)
text(0, -0.6, "S = 1 – Ružička D", cex = 1.5)
triangle.plot(as.data.frame(fish.pod.qS.3[, c(3, 1, 2)]),
show = FALSE,
labeltriangle = FALSE,
addmean = TRUE
)
text(-0.45, 0.5, "AbDiff", cex = 1.5)
text(0.4, 0.5, "Repl", cex = 1.5)
text(0, -0.6, "S = 1 – Percentage difference", cex = 1.5)
# Display values of the mean points in the triangular plots
colMeans(fish.pod.J.3[, c(3, 1, 2)])
colMeans(fish.pod.S.3[, c(3, 1, 2)])
colMeans(fish.pod.qJ.3[, c(3, 1, 2)])
colMeans(fish.pod.qS.3[, c(3, 1, 2)])
8.4.3.5 Explaining Replacement and Richness Difference
Is there a link between replacement or richness difference and the environmental
variables? These questions can be addressed by means of canonical ordination.
Indeed, replacement and richness difference matrices can be submitted, together
with the environmental variables, to distance-based RDA (db-RDA, Sect. 6.3.3). Let
us apply this method by using function dbrda() of package vegan.
# Replacement
repl.dbrda <- dbrda(fish.repl ~ ., data = env , add = "cailliez")
anova(repl.dbrda)
RsquareAdj(repl.dbrda)
# Richness difference
rich.dbrda <- dbrda(fish.rich ~ ., data = env , add = "cailliez")
anova(rich.dbrda)
RsquareAdj(rich.dbrda)
plot(
rich.dbrda,
scaling = 1,
display = c("lc", "cn"),
main = "Richness difference explained by environmental variables"
)
402
8 Community Diversity