Project

General

Profile

different number of longitude points when converting global vs longitudinal band data from reduced to regular Gaussian grid

Added by Guillaume S 5 months ago

Hello,

I use the "-setgridtype,regular" operator to convert ERA5 data from a reduced (N320) to a regular (F320) Gaussian grid.

When I use global data, everything works fine:

Grid coordinates :
1 : gaussian : points=819200 (1280x640) F320
lon : 0 to 359.7188 by 0.28125 degrees_east circular
lat : 89.78488 to -89.78488 degrees_north

But if I use a longitudinal band (but cyclic) extraction, I obtain:

Grid coordinates :
1 : gaussian : points=249405 (1279x195) F320
lon : 0 to 359.4375 by 0.28125 degrees_east
lat : 89.78488 to 35.26931 degrees_north

So the last longitudinal point is not interpolated.
I guess CDO behaves as it is a regional grid, while it should detect the grid is cyclic and interpolate the last longitudinal point.
Is there a way to change this behavior and to tell CDO the grid is cyclic ?

Thanks,

Guillaume


Replies (1)

RE: different number of longitude points when converting global vs longitudinal band data from reduced to regular Gaussian grid - Added by Uwe Schulzweida 5 months ago

Hello Guillaume,

In your longitudinal band the last longitude column is missing, there are only 1279 instead of 1280 longitudes. There is no way in CDO to define such a grid as cyclic, sorry.

Cheers,
Uwe

    (1-1/1)