Description Usage Arguments Details Value Author(s) See Also Examples
eco.scape
is a modified version of the Helmus et al. method
implemented in scape
. It produces phylogenetically
structured communities. It allows phylogenetic signals in niche
optima, but unlike scape
, does not include the
ability to specify niche optima signal type (attraction/repulsion)
or phylogenetic signal in range size. Instead, the focus is on
having more control over the macroecological characteristics of the
resulting landscapes. In particular, eco.scape produces landscapes
with fixed mean range sizes, reasonable range size and abundance
distributions, and control over whether species present on a tree
must be present in the landscape.
1 2 3 4 
tree 

scape.size 
edge dimension of square landscape 
g.center 
strength of phylogenetic signal in species range
centers. See 
wd.all 
niche width, larger values simulate broader range sizes 
signal.center 
simulate with phylosignal in range centers 
center.scale 
adjust strength of phylogenetic attraction in range centers independent of g.center 
site.stoch.scale 
adjust strength of random variation in species richness across sites 
sd.center 
sd in 
sd.range 
sd in rnorm for the range sizes, increase to get more variation in range sizes across gradients 
K 
carrying capacity of a site in terms of maximum individuals that can be present. Currently a constant value. Used to scale the presenceabsence matrix to include abundances. 
extinction 
TRUE/FALSE can species on the tree go extinct on the landscape? If the number of species present on the landscape should equal the number of tips on the tree, choose FALSE. See Details. 
rho 
Grafen branch adjustment of phylogenetic tree see

Simulates a landscape with species (i.e., tree tips) distributions
dependent on a supplied phylogenetic tree. The amount and type of
structure is determened by the signal parameter
g.center
. Parameters are based on an OrnsteinUhlenbeck
model of evolution with stabilizing selection. Values of g=1
indicate no stabilizing selection and correspond to the Brownian
motion model of evolution; 01 corresponds to disruptive selection
where phylogenetic signal for the supplied tree is amplified. See
corBlomberg. Communities are simulated along two gradients where
the positions along those gradients, g.center
, can exhibit
phylogenetic signal.
The function returns a landscape where the average range size is equivalent to the wd.all parameter  in the scape function, this parameter is not necessarily returned in the resulting landscape. To do this, the probability of presence (th) that returns the wd.all parameter is solved for. If there is no solution that can produce the wd.all given, the error "Error in uniroot(f, lower = 0, upper = max(X.), tol = 10^200): f() values at end points not of opposite sign" will occur. This seems to mostly arise for extreme or unlikely parameter values (small species pools, low carrying capacities). Try adjusting parameter values first.
The extinction
parameter specifies whether all of the
species on the tree should be present in the final landscape. Some
species will have probabilities of presence less than those
required for presence. If extinctions is TRUE
, these species
will not be present. If FALSE
, these species will be present
in 1 site, that in which they have the highest probability of
presence.
cc 

Y 
presence/absence matrix 
Yab 
abundance matrix 
index 
spatial coordinates for X and Y (stacked columns) 
X.joint 
full probabilities of species at sites, used to construct Y 
X1 
probabilities of species along gradient 1 
X2 
probabilities of species along gradient 2 
gradient1, gradient2 
environmental gradient values 
nichewd 
average niche width of the assemblage 
K 
carrying capacity of each cell 
environ 
matrix depicting environmental values over the 2D landscape 
sppXs 
full probabilities of each species as an array arranged in a scape.size X scape.size matr ix 
V.phylo 
initial phylogenetic covariance matrix from tree, output of vcv.phylo(tree, corr=T) 
V.phylo.rho 
phylogenetic covariance matrix from tree scaled by Grafen if rho is provided, other wise just an output of vcv.phylo(tree, corr=T) 
V.center 
scaled (by g.center) phylo covariance matrix used in the simulations 
bspp1 
species optima for gradient 1 
bspp2 
pecies optima for gradient 2 
Matt Helmus, Caroline Tucker, cosmetic edits by Will Pearse
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34  # Simulations
tree < rcoal(64)
scape1 < eco.scape(tree, scape.size=25, g.center=1,
signal.center=FALSE, K=100, extinction=TRUE)
scape2 < eco.scape(tree, scape.size=16, g.center=0.2,
signal.center=TRUE, K=100, extinction=FALSE)
scape3 < eco.scape(tree, scape.size=16, g.center=20,
signal.center=TRUE, K=100, extinction=TRUE)
# Plotting distributions and landscape patterns
original_landscape < scape1
abundmax < original_landscape$K
PA_mat < as.matrix(original_landscape$Y)
abund_mat < original_landscape$Yab
site.size < nrow(PA_mat)
species < ncol(PA_mat)
mx < original_landscape$gradient
env < original_landscape$environ$env.gradient
par(mfrow=c(2,2), oma=c(0,0,2,0))
heatcol < (colorRampPalette(c("yellow","red")))
image(matrix(env,sqrt(site.size),sqrt(site.size),byrow=TRUE),
col=heatcol(max(env)),xaxt="n",yaxt="n",main="Env gradient")
image(matrix(rowSums(PA_mat),sqrt(site.size),sqrt(site.size),byrow=TRUE),
col=heatcol(16),xaxt="n",yaxt="n",main="Species Richness")
hist(colSums(PA_mat),ylab="Number of species",xlab="Number of sites",
main="Species Area Relationship",col="lightgrey")
hist(colSums(abund_mat),ylab="Number of species",xlab="Number of individuals",
main="Species Abundance Relationship",col="lightgrey")
mtext("Env random, phy.signal=0.2, 32 species", outer=TRUE, side=3, cex=1.25)

Add the following code to your website.
For more information on customizing the embed code, read Embedding Snippets.