Yann Chemin wrote:
Result is a flipped image (North is South and East is West)
I seem to remember something about this.. at least the north-
south flipped part. You might try asking Frankie on the DebianGIS
list, very vague memory that he might have known about it too.
(also wiki & ml archive)
I also seem to recall there is a module (not built by default)
in the grass5 source code which will take care of the n-s flip.
probably the easiest way to fix it is to export with r.out.mat
then in Octave or Matlab load in the data and run
output_array = flipud(fliplr(input_array));
and then resave to a v4 .mat file and re-import with r.in.mat.
just for the archives, rotating raster array 90deg is trivial
there too: output_array = input_array';
just curious-
may I ask about the data? is it long-term climatological u,v
wind conditions? Hellerman's dataset?
I have had good luck lately with importing 10 years of QuikSCAT
observations and 6 years of US Navy & NARR met model historical
0-hour forecasts as GRIB files*. [*r.region needed to fix gdal
grid registration bug]
Hamish