Fix q_vector units w/ default static stability #3690
Conversation
dopplershift
added
Type: Bug
There unfortunately is, as Holton (2013) does not include the 1/sigma factor, which was unfortunately the (undocumented except in the issue tracker) intended behavior when it was originally written. Though, I'm not sure if that difference of definition is enough to stick with the old behavior or not! EDIT: See #3689 (comment) for a much better discussion on this. Based on this insight, I would concur that the best move is likely moving away from the prior behavior and do something like this PR. |
dopplershift
had a problem deploying
to
CodeCov
GitHub Actions
Failure
There's no reason the units should change in this case, just assign proper units to the default unity value of static stability.
dopplershift
force-pushed
the
qvector-units
branch
from
e1e3ae8
to
19065c5
Compare
dopplershift
had a problem deploying
to
CodeCov
GitHub Actions
Failure
| The static stability at the pressure level. Defaults to 1 J/ kg / Pa^2 if not given | ||
| to calculate the Q-vector without factoring in static stability. |
There was a problem hiding this comment.
With the insight of @kgoebber's comment here, I think it would be helpful to include a bit more detail here (or elsewhere in the docstring) on how static stability is important for this calculation, and perhaps recommend, when possible, to not rely on this default?
There was a problem hiding this comment.
I think that's reasonable, and I'm also inclined to at least pick a default that has a sensible scale, even if it does change results.
There's no reason the units should change in this case, just assign proper units to the default unity value of static stability.
Description Of Changes
Checklist