Skip pob=0.0 in ACARS height fix

[gmao-msienkie]

Check that pressure is greater than zero before calculating new standard atmosphere height for ACARS data
(pob=0.0 is an incorrect obs anyway)

[github-actions]

Label error. Requires at least 1 of: 0 diff, 0 diff trivial, Non 0-diff, 0 diff structural, 0-diff trivial, Not 0-diff, 0-diff, automatic, 0-diff uncoupled. Found:

[gmao-msienkie]

I think that my branch is up to date with GEOSadas-5.29.4-p4 (for sweeper and GEOS-IT).
The sweeper needs this change now and GEOSIT will need it really soon.
If you want/need to update for 'develop' that's fine by me.

I have tested to see that the code change gives zero-diff running a test case.
If it is run with a file with ACARS data with pob=0 it will not be zero diff because it will not crash = but all of the good observations will have the same correction as before the modification.

[rtodling]

Meta - thanks for the change here. I have to say that I am not sure how changes to GEOS-IT, and much less, to M21C (and sweeper) are being handled. I am under the impression that GEOS-IT has already suffered multiple updates/changes and nothing has made it into the repo.

As for M21C ... I imagine this can be coordinated w/ Amal who's still finalizing changes for that purpose.

What I can do is:

1. make sure this gets to develop - easy - I will approve this as soon as I finish typing this.
2. and I have have FP patched to incorporate this as well - though I have just issued a patch to FP and much prefer not to have to do another one ...I'll see w/ Rob and Mark if it's till time to push the tag.

[gmao-msienkie]

I just checked in the pull request "R21C DAS Candidate #268" and it looks like this change (as well as many others) is already in that branch. I believe I created the change when the sweeper encountered the problem and crashed and Rob recompiled with the patch prior to GEOSIT reaching the problem at 20150304 18z. At the time both Rob and Amal agreed it was better to modify the code and recompile rather than just fix the problem input file.
I don't think it is necessary to modify FP since it doesn't use the preprocessing QC. It is also not needed for earlier systems like MERRA2 since the code to recalculate standard atmosphere height was not included in that version. Recalculating standard atmosphere height was only needed after the NRL aircraft QC was adopted to correct the heights in aircraft data processed by an earlier version of the PREPDATA processing.
This can be added now to develop unless you're going to bring in the fix with other changes from #268 - either way we can close out this ticket.

[gmao-msienkie]

There is a Trac ticket (2304) for GEOS-IT where Rob implemented this fix.

[rtodling]

There is a Trac ticket (2304) for GEOS-IT where Rob implemented this fix.

I understand Meta - I know about the trak tickets, but things still need to make their way into the git repo.

[gmao-msienkie]

The change was downloaded into the GEOSadas-5.29.4-p4 build in ~dao_ops and recompiled.
I don't know if there is a process where Rob is updating a Github branch with patches for GEOSIT the way we used to have branches and tags in CVS. It should be easy enough to do.

[rtodling]

The change was downloaded into the GEOSadas-5.29.4-p4 build in ~dao_ops and recompiled. I don't know if there is a process where Rob is updating a Github branch with patches for GEOSIT the way we used to have branches and tags in CVS. It should be easy enough to do.

There is no process other than create a PR w/ the pointer to each branch we mean the change to go into. I change like this would have at least two (of three) PRs, one pointing to develop, another to GEOS-IT and perhaps a third to R21C - the last two are existing branches in the analysis and GEOSadas repos ... sorry - it's cumbersome but git is not very smart about put the same change in multiple branches!

[rtodling]

@mathomp4 it's not clear why the CI keeps failing here for such a simple change ... could it be related to the fact that GEOSadas still doesn't have a changelog?

[mathomp4]

@mathomp4 it's not clear why the CI keeps failing here for such a simple change ... could it be related to the fact that GEOSadas still doesn't have a changelog?

@rtodling Until #297 is folded into develop (and any branch based on develop), the CI for GEOSadas will fail. This is due to the problems with f2py and our CI environment.

[rtodling]

This looks good to me.