What is the feature/what would you like to discuss?
Instead of the current implementation where the initial data for constituents is read in in phys_run2, this data should be read in by the dycore. As part of this, there will be a new metadata field "default_value" for when the constituent is not found on the file.
Is there anyone in particular you want to be part of this conversation?
No response
Will this change (regression test) answers?
No
Will you be implementing this enhancement yourself?
This is a high priority issue given that it appears the SE-dycore isn't currently working properly (i.e. it is unable to use its default ncdata files even with the necessary variable name changes).
What is the feature/what would you like to discuss?
Instead of the current implementation where the initial data for constituents is read in in phys_run2, this data should be read in by the dycore. As part of this, there will be a new metadata field "default_value" for when the constituent is not found on the file.
Is there anyone in particular you want to be part of this conversation?
No response
Will this change (regression test) answers?
No
Will you be implementing this enhancement yourself?
Yes