Hi Russ,
The SINGLE qualifier in the LIST command documentation is a typo.
Ah! Thanks!
But, then, what determines that the variable SST should be written in
double precision? In the example I gave, the variable SST in the
original data (coads_climatology) is in single precision. I haven't
found a description of this point.
I've just recalled the OUTTYPE qualifier for LIST but the
documentation says it applies only to netCDF. I wish it applied to
"stream" and "unformatted" outputs, too.
One (not very good) solution would be to write the file to netcdf in single
precision and use "ncks -b binary.dat -C -v SST tmp.nc -o dummy.nc"
but that creates 2 temporary files with the variable.
Thanks!
Ryo