[R] Re place NA values in matrix with the value of Nearest Neighbour

[RJLewis]

I am extracting climate data for coastal areas from 5km grid data for
specific xy coordinates that relates to individual study sites (SQ). Code so
far:



setwd("/Users/roblewis/Documents/PhD/Climate Data")

fnamestemp=read.table("Path_Directory_Maxt")

fnames=paste(fnamestemp[,1],fnamestemp[,2])

nfiles = length(fnames)


xy=as.matrix(read.table("xy_.txt"))
sq<-xy[,1]
x<-xy[,2]
y<-xy[,3]
xy<-cbind(x,y)
xy<-xy/5000
xy<- xy+41
xy<- cbind(xy[,2],xy[,1])


nfiles=2

for( imap in 2:nfiles)
 {
 
  print(fnames[imap])
  
  x = as.matrix(read.table(fnames[imap],skip=6))
  
  x[x = -99999] = NA
  
  x=flipud(x)
  
  siteclim <- x[ xy]

 }

This all works fine, however as many of the sites are so close to the sea,
many NA's are returned, as teh climate information has not been extrapolated
to include detailed coastal areas. Thus i would like to know how i can
modify my script to instruct R to replace the NA values of 'siteclim' with
the value of its nearest neighbour in the 5km grid data on the condition
that this is also not NULL.

Any suggestions would be gratefully received

Thanks
-- 
View this message in context: http://old.nabble.com/Replace-NA-values-in-matrix-with-the-value-of-Nearest-Neighbour-tp26389519p26389519.html
Sent from the R help mailing list archive at Nabble.com.