timmean overwrites the mask from IMD data

Added by Dhirendra Kumar over 5 years ago

Hii Folks

I am using timmean operator on a file having data over land only. But the timmean operator somehow overwrites the mask and display values over ocean as well once plotted.
Does anyone have any idea to solve the problem.!!

See the two attached files for the reference. yearmean file has the mask but vanishes after applying timmean. any lead will is appreciated.

Thanks

Download all files

timmean.JPG (40.7 KB) timmean.JPG	plot after timmean
yearmean.JPG (42.9 KB) yearmean.JPG	plot after yearmean

Replies (3)

RE: timmean overwrites the mask from IMD data - Added by Karin Meier-Fleischer over 5 years ago

Hi Dhirendra,

please upload the data files (not JPEG) and tell us what CDO call and which CDO version you use.

-Karin

RE: timmean overwrites the mask from IMD data - Added by Dhirendra Kumar over 5 years ago

Hii Karin

Thanks for the prompt reply. I am using cdo-1.8.2. The variable of interest was precipitation (mm). While doing timmean the values over ocean were being set to 0. I solved the issue by setting the vrange from 0.0001 to 100 on the timmean file. Now the mask is the same as in the yearmean file.

Although I solved the issue for now. I would like to know the reason for such behavior. I am attaching the files here.

---------
Dhirendra

Download all files

timmean_yearmean_JJAS_rain_imd_1975-2005.nc (20.3 KB) timmean_yearmean_JJAS_rain_imd_1975-2005.nc	Timmean File
yearmean_JJAS_1975-2005.nc (546 KB) yearmean_JJAS_1975-2005.nc	Yearmean File

RE: timmean overwrites the mask from IMD data - Added by Uwe Schulzweida over 5 years ago

Thanks for the data, that helps a lot! Here is the result of the first timesteps from infon:

cdo infon yearmean_JJAS_1975-2005.nc 
    -1 :       Date     Time   Level Gridsize    Miss :     Minimum        Mean     Maximum : Parameter name
     1 : 1975-07-31 12:00:00       0     4485    3236 :     0.18673      8.3013      61.808 : rf            
     2 : 1976-07-31 12:00:00       0     4485    3234 :    0.066406      7.3346      40.890 : rf            
     3 : 1977-07-31 12:00:00       0     4485    3234 :     0.12227      7.3624      49.208 : rf            
     4 : 1978-07-31 12:00:00       0     4485    3234 :    0.026274      7.7532      68.295 : rf            
     5 : 1979-07-31 12:00:00       0     4485       0 :      0.0000      1.5754      49.190 : rf            
     6 : 1980-07-31 12:00:00       0     4485    3234 :     0.29706      7.6072      63.809 : rf            
     7 : 1981-07-31 12:00:00       0     4485    3234 :     0.16475      7.1165      52.384 : rf            
....

The yearly mean from 1979 doesn't have missing values and this minimum of that timestep is zero. Thats the reason why the minimum of timmean is zero. Try the operator timavg this could make more sense for this dataset. timavg handles the missing values different, see section Mean and average for more information.

Cheers,
Uwe

(1-3/3)

Project

General

Profile

CDO