[R-sig-Geo] recommended workflow for non-Pacific grids

Michael Sumner mdsumner at utas.edu.au
Mon Mar 26 10:14:07 CEST 2007


Hello,
I have a fairly convoluted set of functions (using GridTopology, rgdal 
and some pretty ad-hoc conditionals . . .)
that allow me to easily read Pacific-centred regions from 
longlat-gridded datasets that
are Atlantic-centric. I basically revert to SpatialPixels and let it 
sort out the indexing after adding
360 to western-hemisphere coordinates.  I

I'd like to ask for suggestions as to the best way to approach this.

I've not found neat way to do it, and it seems the most efficient would 
be to read the two sections
as SGDF, then convert those with as.image.SpatialGridDataFrame, 
re-binding the western portion
to the eastern and resolving the grid topology and recreating the SGDF.

I've not actually tried this for multi- datasets, but it would involve a 
loop in the above
workflow for each band.

If there's nothing obvious that I'm missing I'll follow this post up 
with a real example and
go from there.

Cheers, Mike.




More information about the R-sig-Geo mailing list