Closed AndyHoggANU closed 4 years ago
I think all the grid variables that are in the grid.nc
file should be excluded, so adding
-x TLON -x TLAT -x ULON -x ULAT -x NCAT
to the splitvar
call should avoid this.
OK, so there were heaps of -x
variables - I had to delete them because they weren't defined in every case. Have added these ones back in and will re-test.
Note that there are some other variables in grid.nc, but am hoping this are not used?
OK, this works.
There is a remaining error with aicen-m
, because it wasn't saved for the first couple of years. Not much we can do about this - have removed this diagnostic from the list of variables for the 0.1° simulation.
Splitvar has a minor issue with some of the ice files at 0.1°. These are the files from 1985-2004 which seemed to have a slightly different ice grid. Our intention was to override these with
-a grid.nc
but it seems to have picked up an incompatibility via this error message:I think this is something that needs to be altered in splitvar itself -- unless we can pass that argument through?