cdo remap from ERA5 grid (lonlat) to gx1 grid (generic)

Hello,
I'm trying to make model forcing data using CDO and need to regrid ERA5 data (lonlat) into a gx1 grid(generic).
First, I ran

cdo griddes target_grid.nc  > mygrid
cdo remapbil,mygrid ERA5.nc gx1_forcing.nc

But when I checked the generated data using panoply, I realized that there are missing values on Greenland which should not be existed.

And the CDO version is 1.8.2
I have attached the relevant files.
How can I fix it？I have spent quite a bit of time on this and would be most grateful for any help.
Wilma

Download all files

target_grid.nc (481 KB) target_grid.nc
mygrid (2.36 MB) mygrid
gx1_forcing.nc (3.75 MB) gx1_forcing.nc	generated forcing data
ERA5.nc (23.8 MB) ERA5.nc
clipboard-202404151104-xnmzd.png (352 KB) clipboard-202404151104-xnmzd.png

Replies (3)

RE: cdo remap from ERA5 grid (lonlat) to gx1 grid (generic) - Added by Wilma Wang about 1 month ago

I rechecked the distribution of variables on the target grid, and it seems that there are missing values in Greenland using panoply, but they don't show up using ncview.So I don't think these missing values are caused by CDO.

RE: cdo remap from ERA5 grid (lonlat) to gx1 grid (generic) - Added by Karin Meier-Fleischer about 1 month ago

Hi Wilma,

your target grid is a curvilinear grid where a pole is defined over greenland. So the remapped data is correct.

Plot with Python shows the same as panoply:

py_gx1_forcing.nc.png (251 KB) py_gx1_forcing.nc.png

RE: cdo remap from ERA5 grid (lonlat) to gx1 grid (generic) - Added by Wilma Wang about 1 month ago

Hi,Karin

Thanks for your check with python!

Best regards,
Wilma

(1-3/3)

Project

General

Profile

CDO