Read meridional heat transport lat bins from either of 2 streams

[xylar]

To save the bother of writing out the MHT stream just for the latitude bins, these bins can now (optionally) be read from the timeSeriesStatsMonthlyOutut stream instead.

The refZMid, refBottomDepth and refThickness are read or computed from a restart file instead of the MHT stream.

[xylar]

@mark-petersen, I know we don't have any output with the latitude bins in timeSeriesStatsMonthlyOutput to test this on yet. Please point me to it once we do.

In the meantime, you can make sure that an existing run of your choice still produces the MHT plot as before.

[xylar]

Testing

I tested on the QU240 run on my laptop. Here is the standard MHT plot from this branch:

and from develop:

There is no appreciable difference.

Here is the plot including depth from this branch (turned off in my run with develop):

[mark-petersen]

Yes, this is perfect. Since it is optional it is best to put it in now. Thanks for your quick action and testing here. Please merge.

[milenaveneziani]

We should have the test run with the new output sometime today, so we can test this then.

[xylar]

Yes, we need to test this on the new output once it's available. Merging now an fixing later doesn't really save us any time.

[xylar]

@milenaveneziani, do you remember what the status of testing this PR is? I lost track of it.

[milenaveneziani]

We do have the new data to test it with.
This can go in independently from #294, right?

[xylar]

Testing Continued

For the first 2 years of a QU240 run, everything seems to work both for MPAS-Analysis/develop and for this branch.

develop: http://portal.nersc.gov/project/acme/xylar/a24w_develop/

new branch: http://portal.nersc.gov/project/acme/xylar/a24w_update_source/

I also ran a test where I manually made sure the code doesn't use the MHT stream and that also worked fine. So I think MPAS-Analysis should work fine with these changes.

[xylar]

This can go in independently from #294, right?

Yes, the 2 PRs are independent so we can merge as soon as we're confident in the results. Which I think we are.