(dose ¼ 0) and the remaining eight were treated once with the insecticide chlorpyrifos, with dose levels of 0.1, 0.9, 6 and 44 μg L
À1 in two mesocosms each. Therefore,
the explanatory factor has four levels (the four dose levels, excluding control) and
the covariable “week” has 11 levels. The code below is extracted from the prc()
documentation file, with added comments.
# Code from the prc() help file, with additional comments
# Chlorpyrifos experiment and experimental design: Pesticide
# treatment in ditches (replicated) and followed over, from 4 weeks
# before, to 24 weeks after exposure
# Extract the data (available in vegan)
data(pyrifos)
# Create factors for time (week) and treatment (dose). Create an
# additional factor "ditch" representing the mesocosm, for
# testing purposes
week labels = c(-4, -1, 0.1, 1, 2, 4, 8, 12, 15, 19, 24))
dose 11))
ditch <- gl(12, 1, length = 132)
# PRC
mod <- prc(pyrifos, dose, week)
mod
# Modified RDA
summary(mod)
# Results formatted as PRC
# PRC plot; at the right of it, only species with large total
# (log-transformed) abundances are reported
logabu <- colSums(pyrifos)
plot(mod, select = logabu > 200)
# Statistical test
# Ditches are randomized, we have a time series, and are only
# interested in the first axis
ctrl within = Within(type = "series"), nperm = 999)
anova(mod, permutations = ctrl, first = TRUE)
The analysis results in the following plot (Fig. 6.14).
Figure 6.14 shows that the intensity of community response to the treatment
clearly depends on the chlorpyrifos dose. At higher doses, the response is greater and
lasts longer. Note, however, that between doses 6 and 44 μg L
À1 the peak intensities
of the responses are comparable; the difference is that dose 44 μg L
À1 produces a
longer effect.
270
6 Canonical Ordination