Closed climbfuji closed 2 years ago
@climbfuji Has this issue been solved via https://github.com/ESCOMP/CCPPStandardNames/pull/21?
@gold2718 @grantfirl @ligiabernardet I am working on this now, and in addition to the mismatch for ccpp_error_message
, where capgen will adopt the existing unit none
in ccpp-physics/SCM/UFS instead of 1
, there is another mismatch:
ccpp-physics/SCM/UFS use
[errflg]
standard_name = ccpp_error_flag
long_name = error flag for error handling in CCPP
units = flag
dimensions = ()
type = integer
while capgen uses
[ errflg ]
standard_name = ccpp_error_code
long_name = Error flag for error handling in CCPP
units = 1
dimensions = ()
type = integer
intent = out
I think we agreed on using the capgen version for errflg
, correct? This will require changing 120+ metadata files in ccpp-physics/SCM/UFS, but at least this is a straightforward change.
I kind of liked flag
but 1
is fine as well and more CF-like.
We need to revisit the units of dimensionless variables in the dictionary of standard names. For this, we first need to agree on what the units should be for each of the variables, create or update the documentation, and update the dictionary itself.
At this time, we have the following units for dimensionless variables:
For each dimensionless variable, we need to determine which unit is appropriate, if an additional unit is required.
Currently, there are mismatches between the units in the dictionary of standard names for variables with unit
1
, and the ccpp-physics, fv3atm and scm repositories, which are mostly usingnone
for these variables.