Project

General

Profile

uv2dv only processes one level

Added by Claudia Stephan over 3 years ago

Hello,

I have data on sigma levels:

cdo info IG_all_202001200000287.nc | head -7
-1 : Date Time Level Gridsize Miss : Minimum Mean Maximum : Parameter ID
1 : 2020-01-21 13:00:00 0.00968932 524288 0 : -16.361 0.49311 17.942 : -1
2 : 2020-01-21 13:00:00 0.0120526 524288 0 : -20.049 -0.83703 13.233 : -1
3 : 2020-01-21 13:00:00 0.0147811 524288 0 : -17.054 -0.75813 10.774 : -1
4 : 2020-01-21 13:00:00 0.0178977 524288 0 : -14.427 -0.66065 10.239 : -1
5 : 2020-01-21 13:00:00 0.0214237 524288 0 : -12.571 -0.55527 11.227 : -1
6 : 2020-01-21 13:00:00 0.0253793 524288 0 : -13.945 -0.43074 11.966 : -1

When I apply
cdo -P 24 sp2gp -uv2dv IG_all_202001200000001.nc IG_all_202001200000001_dv.nc

I get the correct results only at level 1 and zero at all other levels.

If, however, I select a different level, i.e.,
cdo -P 24 sp2gp -uv2dv -sellevidx,7 IG_all_202001200000001.nc IG_all_202001200000001_dv.nc

then I also get the correct result. Thus, the problem seems to simply be looping over the levels. How do I fix it?

Best regards,
Claudia


Replies (5)

RE: uv2dv only processes one level - Added by Ralf Mueller over 3 years ago

hi Claudia!

do you have a path for some sample input? I'd like to test this with other CDO versions

cheers
ralf

RE: uv2dv only processes one level - Added by Uwe Schulzweida over 3 years ago

I can confirm the error for CDO version 1.9.8 The problem is fixed in the latest version 1.9.9.
A test version is available at mistral:/home/mpim/m214003/local/bin/cdo
This version will be installed next month on the mistral.

RE: uv2dv only processes one level - Added by Ralf Mueller over 3 years ago

in case you use conda, the 1.9.9 release is already available in the conda-forge channel

RE: uv2dv only processes one level - Added by Claudia Stephan over 3 years ago

I tried and now I get a different error message:
"cdo(1) uv2dv (Abort): U and V fields on Gaussian grid are not global!"

Here the file:
/work/mh1119/m300685/MODES/nwp2.5winter/inverse/IG_all_202001200000001.nc

Claudia

RE: uv2dv only processes one level - Added by Ralf Mueller over 3 years ago

there is a problem with the grid of /work/mh1119/m300685/MODES/nwp2.5winter/inverse/IG_all_202001200000001.nc

# gridID 1
#
gridtype  = gaussian
gridsize  = 524288
xsize     = 1024
ysize     = 512
xname     = lon
xlongname = "longitude" 
xunits    = "degrees_east" 
yname     = lat
ylongname = "latitude" 
yunits    = "degrees_north" 
numLPE    = 0
xfirst    = 0
xinc      = 0.3515625
yvals     = 89.7311486184091 89.3828738963335 89.0325424237899 88.6817462435914 88.3307737888069 
            87.979716034326 87.6286106094846 87.2774758672234 86.9263218176456 86.5751543820948 
            86.2239772863457 85.8727929914673 85.5216031882814 85.170409076734 84.8192115319312 
            84.4680112070663 84.1168....

whereas another file, which works with uv2dv (/work/mh1119/m300685/MODES/nwp2.5winter/infile_202001200000207.nc)
#
# gridID 1
#
gridtype  = gaussian
gridsize  = 524288
xsize     = 1024
ysize     = 512
xname     = lon
xlongname = "longitude" 
xunits    = "degrees_east" 
yname     = lat
ylongname = "latitude" 
yunits    = "degrees_north" 
numLPE    = 256
xfirst    = 0
xinc      = 0.3515625
yvals     = 89.7311486184145 89.3828738963346 89.0325424237903 88.6817462435912 88.3307737888073 
            87.979716034326 87.6286106094843 87.2774758672239 86.9263218176459 86.5751543820948 
            86.2239772863455 85.8727929914671 85.5216031882812 85.1704090767342 84.8192115319314 
            84.4680112070663 84.116808599553 83.7656040948441 83.4143979962481 83.0631905457019 
            82.7119819385429 82.3607723342125 82.0095618641375 81.6583...

Both grids are not identical. I fixed the missing numLPE and set it to 256. then the processing works.

you can make use of it with setgrid like this

/home/mpim/m214003/local/bin/cdo -sinfov -uv2dv -setgrid,/home/dkrz/k202125/gdFixed /work/mh1119/m300685/MODES/nwp2.5winter/inverse/IG_all_202001200000001.nc

but please note, that I created this fix from a single data file. the coordinates must be checked for other file again.

cheers
ralf

    (1-5/5)