spe.sc1
scaling = 1,
display = "sp"
)
arrows(0, 0,
spe.sc1[, 1] * 0.92,
spe.sc1[, 2] * 0.92,
length = 0,
lty = 1,
col = "red"
)
# Scaling 2
plot(spe.rda,
display = c("sp", "lc", "cn"),
main = "Triplot RDA spe.hel ~ env3 - scaling 2 - lc scores"
)
spe.sc2
display = "sp"
)
arrows(0, 0,
spe.sc2[, 1] * 0.92,
spe.sc2[, 2] * 0.92,
length = 0,
lty = 1,
col = "red"
)
Hints In the scores() function argument, choices= indicates which axes are to be
selected. Also, be careful to specify the scaling if it is different from 2.
The plot() function used above has an argument type that determines how
the sites must be displayed. For small data sets (as here), the default is
type="text" and so the site labels are printed, but for large data sets it
switches to "points". If you want to force site labels instead of points with a
large data set, write type="text", but be prepared to see a crowded plot.
See how to choose the elements to be plotted, using the argument
display=c(...). In this argument, "sp" stands for species, "lc" for fitted
site scores (linear combinations of explanatory variables), "wa" for site scores in
the species space (weighted averages in CCA or weighted sums in RDA), and
"cn" for constraints (i.e., the explanatory variables).
In the arrows() call, the scores are multiplied by 0.92 so that the arrows do not
cover the names of the variables. Adjust this factor by trial and error.
6.3 Redundancy Analysis (RDA)
215
scaling = 1,
display = "sp"
)
arrows(0, 0,
spe.sc1[, 1] * 0.92,
spe.sc1[, 2] * 0.92,
length = 0,
lty = 1,
col = "red"
)
# Scaling 2
plot(spe.rda,
display = c("sp", "lc", "cn"),
main = "Triplot RDA spe.hel ~ env3 - scaling 2 - lc scores"
)
spe.sc2
display = "sp"
)
arrows(0, 0,
spe.sc2[, 1] * 0.92,
spe.sc2[, 2] * 0.92,
length = 0,
lty = 1,
col = "red"
)
Hints In the scores() function argument, choices= indicates which axes are to be
selected. Also, be careful to specify the scaling if it is different from 2.
The plot() function used above has an argument type that determines how
the sites must be displayed. For small data sets (as here), the default is
type="text" and so the site labels are printed, but for large data sets it
switches to "points". If you want to force site labels instead of points with a
large data set, write type="text", but be prepared to see a crowded plot.
See how to choose the elements to be plotted, using the argument
display=c(...). In this argument, "sp" stands for species, "lc" for fitted
site scores (linear combinations of explanatory variables), "wa" for site scores in
the species space (weighted averages in CCA or weighted sums in RDA), and
"cn" for constraints (i.e., the explanatory variables).
In the arrows() call, the scores are multiplied by 0.92 so that the arrows do not
cover the names of the variables. Adjust this factor by trial and error.
6.3 Redundancy Analysis (RDA)
215