Bikem Ekberzade

To follow up on this thread (instead of opening a new one), Charlie, despite using your above mentioned command (I also utilize an alternative: ncks -C -O -x -v time_bnds ....) the variable time still has a residue in metadata left (CMIP6 dataset) : double time(time) ; time:standard_name = "time" ; time:long_name = "time" ; time:bounds = "time_bnds" ; time:units = "hours since 1900-01-01 00:00:00.0" ; time:calendar = "gregorian" ; time:axis = "T" ; This -- time:bounds = "time_bnds" -- is interfering...

Yes, it does indeed! Thank you!

Thanks, just wrote a similar code in CDO, will try to convert it to NCO to see if it works. Just wasn't sure if NCO took e as a value (CDO, much like C++ doesn't)

I have a netcdf file with temp and dp as two variables over an area for 38 years. I need to calculate relative humidity with the equation below: RH=100× ((e (exp) (17.625×DP / 243.04+DP)) / ((e (exp) (17.625×T / 243.04+T)) I can write out the simple arithmetic (except I do not know how to write out e (exp) ), but defining a new variable to an existing NetCDF is where I'm not sure if I can do that with NCO. Any assistance in this matter would be greatly appreciated, including help with the equation....

I see, thank you! will try that now.

Charlie hi, I have one more multiplication question if possible. I couldn't find the answer anywhere and my trials with ncmultiply (or ncbo --op_typ='*') gave no result. I have two netcdf files (GCMs) and after doing separate arithmetics on both, I want to multiply these two (same calendar days). In cdo I have cdo mul in.nc out.nc but my results are not coming out correct (in parts where the sea should be masked, CDO is creating results, which is then messing up my spatial average) Is there any way...

Oh, I see now. Thank you for the explanation! I invented another, rather round about way meanwhile (further splitting the year to months, constructing February 29, and stiching back up)

Charlie, thanks for the suggestion, I had tried something similar to your suggestion before, and now I tried your suggestion too, but with this I end up with a single timestep... What I am trying to do is rather extend the file, insert a day to the end of February for a leap year (say 2016) and have the value be the same as the day before. I can do this in R, but the file is large (many grids) and R cannot call it into memory.. I am rather a novice with the NCO codes...