Hi users of raster:
When a point lands within a raster cell, the cell is correctly identified.
But when a point lands on the border between two adjacent raster cells,
cell identity seems to be inconsistent. We are wondering if this is an
error?
Example code follows to demonstrate our question:
require(raster)
xmn=-62
ymn=40
xmx=-60
ymx=42
res=.1
# simple raster with integers fill
r0 <- raster( xmn=xmn, ymn=ymn, xmx=xmx, ymx=ymx, res=res )
r0[] <- 1:ncell(r0)
# choose points along a horizontal band:
locx = seq(xmn, xmx, res)
xy1 <- cbind( locx + res/2, rep(ymn+res+res/2, dim(r)[1] ) ) # center of a
raster cell
xy2 <- cbind( locx, rep(ymn+res+res/2, dim(r)[1] ) ) # on x-boundary
xy3 <- cbind( locx + res/2, rep(ymn+res, dim(r)[1] ) ) # on y-boundary
xy4 <- cbind( locx, rep(ymn+res, dim(r)[1] ) ) # on a corner
# plots to visualize selected points
cid1 = cellFromXY(r, xy1); r = r0; r[cid1] <- NA; plot(r)
cid2 = cellFromXY(r, xy2); r = r0; r[cid2] <- NA; plot(r)
cid3 = cellFromXY(r, xy3); r = r0; r[cid3] <- NA; plot(r)
cid4 = cellFromXY(r, xy4); r = r0; r[cid4] <- NA; plot(r)
The expectation would have been to see no banding as seen in xy1 and xy3.
But xy2 and xy4 seem to skip a few cells. Not sure why?
Thanks!
[[alternative HTML version deleted]]
______________________________________________
R-help@r-project.org mailing list -- To UNSUBSCRIBE and more, see
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.
