Processing of CORDEX data using CDO
Added by Gunavathi S over 3 years ago
Hello everyone,
I am doing research on climate change modeling. So, I am in need to calculate seasonal mean from the CORDEX data. I am following the below procedures but I am not getting any results.
1.Data downloaded: pr_WAS-44_MPI-M-MPI-ESM-LR_historical_r1i1p1_MPI-CSC-REMO2009_v1_day_19960101-20001231.nc
2.Data Merging (30 years): cdo merge infile1 infile2 outfile
3.Conversion from rotated to regular coordinate system:
cdo remapbil,r720x360 input.nc output_file.nc
4.Setting of missing value:
cdo -setctomiss,-999.9 infile outfile
5.Conversion of Unit(to mm):
cdo mulc,86400 input.nc ouput.nc
6.Calculations of seasonal mean (JJAS):
cdo seasmean infile outfile
7.Conversion from Netcdf to ASCII and Extraction:
cdo -outputtab,date,lon,lat,value -remapnn,"lon=78.8_lat=10.8" infile.nc > output.csv
are the above procedures and commands correct? what should I do to calculate seasonal mean? Thanks in advance.
| cdo sinfo pr.docx (12.3 KB) cdo sinfo pr.docx | meta data |
Replies (6)
RE: Processing of CORDEX data using CDO - Added by Karin Meier-Fleischer over 3 years ago
Hi Gunavathi,
I would say it looks good. You already computed the seasonal mean in step 6.
-Karin
RE: Processing of CORDEX data using CDO - Added by Gunavathi S almost 3 years ago
Dear Sir,
I have a doubt at points 3 and 7. In point 3, I have used bilinear interpolation to convert from rotated to regular coordinates with the resolution of r720x360 (0.5 degrees). But in point 7, I have used the nearest neighbour interpolation to extract values for particular lat/long at 1 degree resolution. Is that correct? Can I proceed?
RE: Processing of CORDEX data using CDO - Added by Karin Meier-Fleischer almost 3 years ago
Hi Gunavathi,
I don't know what you mean when you say you remap to 0.5 degree but do the outputtab from 1 degree resolution file (which input file with 1 deg res?).
If you decide to remap to 0.5 degrees resolution than be strictly consistent.
The following example uses the file you mentioned above:
cdo -remapbil,global_0.5 pr_WAS-44_MPI-M-MPI-ESM-LR_historical_r1i1p1_MPI-CSC-REMO2009_v1_day_19960101-20001231.nc pr_0.5.nc cdo -setctomiss,-999.9 pr_0.5.nc tmp1.nc cdo -mulc,86400 tmp1.nc tmp2.nc cdo -seasmean tmp2.nc seas.nc cdo -outputtab,date,lon,lat,value -remapnn,"lon=78.8_lat=10.8" seas.nc > outfile.csv
Or in one step
cdo -outputtab,date,lon,lat,value -remapnn,"lon=78.8_lat=10.8" -seasmean -mulc,86400 -setctomiss,-999.9 -remapbil,global_0.5 pr_WAS-44_MPI-M-MPI-ESM-LR_historical_r1i1p1_MPI-CSC-REMO2009_v1_day_19960101-20001231.nc > outfile.csv
RE: Processing of CORDEX data using CDO - Added by Gunavathi S almost 3 years ago
Dear Sir,
Thank you for your reply. I have attached an image with CDO comment to extract the values for a particular lat/long in which, what does line-3 mean?
line 3: cdo(2) remapnn: Nearest neighbor weights from lonlat (1440x720) to lonlat (1x1) grid , with source mask (69551).
The input file resolution is 0.25 degree. is it again remapped to 1 degree?
| screen.PNG (13.7 KB) screen.PNG | cdo file |
RE: Processing of CORDEX data using CDO - Added by Karin Meier-Fleischer almost 3 years ago
No, it is not remapped to 1 degree.
Do you know what -remapnn,"lon=78.8_lat=10.8 does? It extracts the data value closest to the point coordinates (lon=78.8, lat=10.8), thus only 1 data value per timestep.
In the remapnn output message from cdo the (1x1) means the output is 1 grid point with 1 lon value and 1 lat value.
For comparison global (360x180)
360 lon values, which means that the increment is 1 degree
180 lat values, which means that the increment is 1 degree
RE: Processing of CORDEX data using CDO - Added by Gunavathi S almost 3 years ago
Thank you so much sir..