CDO

I manage to figure it out. Used cdo ifthen infile infile outfile.

Sondre,
could you prove some sample data and one or two CDO calls? this would be very helpful to understand your solution.

thx in advance
ralf

I was not able to reduce the gridsize, but only adding more missing data.
I did this:

cdo ifthen northsea.nc northsea.nc nonzeronorthsea.nc

This made all the 0 values to NaN and where not added when I plotted the file in panoply(which is a mapping program).

Hope this is helpful!

Kind regards,
Sondre

didn't look into your data, yet. But I think the operators reducegrid might be what your are looking for. There is a special post about it: https://code.mpimet.mpg.de/boards/53/topics/3895

hth
ralf

Hi Sondre, hi Ralf,

just to mention that you can use

cdo -f nc -setctomiss,0 northsea.nc nonzeronorthsea_miss.nc

The output of reducegrid is an unstructured grid which can't be diplayed by Panoply.

-Karin

A more general remark: If you want to reduce the gridsize, you should be aware of the fact, that the grid is constant for all timesteps. In case of a time-varying number of missing values (or zeros) this will not be possible. And CDO does not support time dependent grids - nor do I know any other tool, that does.

So it depends what your plan was: reducing file size? avoid handling of missing values in your program?