Dear all,
We are trying to clean multiple realizations of a pattern. Erroneous NODATA
and spurious DATA occur in the realizations. As we have to do a 1000
realizations for many patterns, efficiency of the code is important.
We need to correct the realizations with a 'mask' pattern of
DATA/NODATA.
We think an ifelse should do the job. Spurious DATA will be simply removed
using the mask. Erroneous NODATA should be filled in by averaging across
the cell's nearest neighbors.
The ifelse table is as follows:
IFELSE criteria to determine TARGET from realization and mask
Real Mask Target
10 10 10
10 1 0/1 determined through post processing function (average
across neighbors)
0/1 10 10
0/1 1 0/1
We think that an APPLY on a multi dimensional array is the way to go, with
each realization (2 dimensions) being a dimension of the array (like a
stack of maps)
This is where we have got so far:
*********************
library(abind)
#Sim: 0=Absent; 1=Present; 10=NODATA
vectSim <-
c(0,1,0,1,10,0,1,0,1,1,10,0,1,1,0,1,0,10,10,1,0,1,0,1,0,1,1,10,0,10,10,1,0,1,0,10)
#Mask: 1=DATA; 10=NODATA
vectMask <- c(10,1,10,1,1,10,1,10,1,10,10,1)
length(vectSim)
length(vectMask)
numRow<-3
numCol<-4
numReal<-3
Sim <- array(vectSim, c(numRow,numCol,numReal))
Sim
Mask <- array(vectMask, c(numRow,numCol))
Mask
SmoothSim <- apply(Sim, c(1,2), ifelse(MaskSim==10,33,99))
SmoothSim
********************
Any help is much appreciated!
Thanks,
Sander and Alessandro.
