This function plots the distribution of planning units and the distribution of demand points for a particular species in an attribute space. Note that this function only works for attribute spaces with one, two, or three dimensions.
space.plot(x, species, space, ...)
# S3 method for RapData
space.plot(
x,
species,
space = 1,
pu.color.palette = c("#4D4D4D4D", "#00FF0080", "#FFFF0080", "#FF00004D"),
main = NULL,
...
)
# S3 method for RapUnsolved
space.plot(
x,
species,
space = 1,
pu.color.palette = c("#4D4D4D4D", "#00FF0080", "#FFFF0080", "#FF00004D"),
main = NULL,
...
)
# S3 method for RapSolved
space.plot(
x,
species,
space = 1,
y = 0,
pu.color.palette = c("#4D4D4D4D", "#00FF0080", "#FFFF0080", "#FF00004D"),
main = NULL,
...
)
RapData()
, RapUnsolved()
, or RapSolved()
object.
character
name of species, or integer
index for
species.
integer
index of attribute space.
not used.
character
vector of colors indicate planning unit
statuses. Defaults to c("grey30", "green", "black", "red")
which
indicate not selected, selected, locked in, and locked out (respectively).
character
title for the plot. Defaults to NULL
and
a default title is used.
integer
number specifying the solution to be plotted. The
value 0
can be used to plot the best solution.
# \dontrun{
# load RapSolved objects
data(sim_ru, sim_rs)
# plot first species in first attribute space
space.plot(sim_ru, 1, 1)
# plot distribution of solutions for first species in first attribute space
space.plot(sim_rs, 1, 1)
# }