Dear r-help members,
Could any of you help me with this model, please?
This model gives error when some value touch whatever border and I do not know
how to correct it. The 80% of the seeds produced by a plant will fell into the
parent cell, the 15% in the first ring according to the king movement (in
chess), and a 5% in the second ring defined by the queen2 matrix. Someone said
me the functions that order the position of the seed dispersed, they are:
neighb <-
aperm(apply(aperm(outer(king,IJ,'+'),c(2,4,1,3)),3:4,diag),c(2,1,3)) #Seed
going back into source cell???
for(i in 1:nrow(IJ)) SB[matrix(neighb[,,i],nc=2)]<- SR1[matrix(IJ[i,],1,2)]/8
+ SB[neighb[,,i]] #Seed going to anuli 1 neighbor cells from all possible
sources
neighb <-
aperm(apply(aperm(outer(queen2,IJ,'+'),c(2,4,1,3)),3:4,diag),c(2,1,3))
for(i in 1:nrow(IJ)) SB[matrix(neighb[,,i],nc=2)]<- SR2[matrix(IJ[i,],1,2)]/8
+ SB[neighb[,,i]] #Seed going to anuli 2 neighbor cells from all possible
sources
But they produce that when a dispersed seed goes into a border, it stops the
model and gives an error. I guess I should put a buffer in the border to get
the value 0 each time some seed reach that buffer but I do not how to manage
the model with those complicated dispersal functions.
I do not understand how these functions disperse the seeds, Do they it in a
particular order? The queen2 ring should be 16 cells, why in the 'SR2[]'
expression divides by 8 and not by 16?
Do anyone know if the dispersed seed are properly included in the SB for the
next iteration?
Here I attach you the full model to be this message understandable:
width = 10
height= 10
endt = 6
nPop = 1
SBLL = 1
SBUL = 10
LrD <- matrix(0, width, height)
x <- as.integer(runif(nPop,min=1,max=10)) #uniform random x ccordinates for
occupied cells
y <- as.integer(runif(nPop,min=1,max=10)) #uniform random y ccordinates for
occupied cells
z <- cbind(x,y)
LrD[z] = runif(nPop,min=SBLL,max=SBUL) #uniform random SB densities
assigned to random locations
par(mfrow=c(1,1))
image(x=1:width,y=1:height,z=LrD, col=terrain.colors(6))
#Parameter values for population simulation
sl1 = matrix(0.93, ncol=width, nrow=height)
gl1 = matrix(0.64, ncol=width, nrow=height)
vB = matrix(0.76, ncol=width, nrow=height)
f = matrix(100, ncol=width, nrow=height)
a = matrix(0.34, ncol=width, nrow=height)
#Initial population values at time = 0
SB <- matrix(LrD, ncol=width, nrow=height)
SDL <- matrix(gl1*LrD, ncol=width, nrow=height)
AP <- matrix(vB*SDL, ncol=width, nrow=height)
SP <- matrix((f*AP)/(1+a*AP), ncol=width, nrow=height)
gN <- matrix(0, ncol=5, nrow=endt)
#Define dispersal anuli at radius 1 (king) and 2 (queen2).
king<- matrix(c(1,1,1,-1,-1,1,-1,-1,0,1,0,-1,1,0,-1,0),nc=2,byrow=T) #first
anuli
queen2<-
as.matrix(rbind(expand.grid(seq(-2,2,1),c(-2,2)),expand.grid(c(-1,0,1),c(-2,2))))
#second anuli
# Main population simulation loop
for (t in 1:endt){
SB<- round(SB)
SDL = SB*gl1
AP = SDL*vB
SP = (f*AP)/(1+a*AP)
SR1 = 0.15 * SP #Total Seed dispersed to radius 1
SR2 = 0.05 * SP #Total Seed dispersed to radius 2
IJ = which(SP>0,arr.ind=T)
SB[IJ] = 0.8 * SP[IJ]
neighb <-
aperm(apply(aperm(outer(king,IJ,'+'),c(2,4,1,3)),3:4,diag),c(2,1,3)) #Seed
going back into source cell???
for(i in 1:nrow(IJ)) SB[matrix(neighb[,,i],nc=2)]<- SR1[matrix(IJ[i,],1,2)]/8
+ SB[neighb[,,i]] #Seed going to anuli 1 neighbor cells from all possible
sources
neighb <-
aperm(apply(aperm(outer(queen2,IJ,'+'),c(2,4,1,3)),3:4,diag),c(2,1,3))
for(i in 1:nrow(IJ)) SB[matrix(neighb[,,i],nc=2)]<- SR2[matrix(IJ[i,],1,2)]/8
+ SB[neighb[,,i]] #Seed going to anuli 2 neighbor cells from all possible
sources
image(1:width,1:height, SB, col=terrain.colors(6), xlab="", ylab="")
gN[t,1:5] <- cbind(t, mean(SB), mean(SDL), mean(AP), mean(SP))
}
gN
Thank you very much in advance, and congratulations for this magnificent help
that all of you offer,
Judit Barroso
[[alternative HTML version deleted]]
______________________________________________
R-help@r-project.org mailing list
https://stat.ethz.ch/mailman/listinfo/r-help
PLEASE do read the posting guide http://www.R-project.org/posting-guide.html
and provide commented, minimal, self-contained, reproducible code.
