cleaned up burial and organic matter remin, added options to toggle/scale burial

[charliestock]

This pull request responds to issue #70. It adds an option to toggle and scale burial with a parameter called "scale_burial". The default value is 0 to match prior simulations that had burial turned off. Changing the value to 1.0 turns on the full burial amount, and intermediate values produce intermediate levels of burial.

One can also modulate burial with the parameter z_burial, which provides a depth scale over which burial is ramped up to its full value. This parameter is useful to avoid excessive burial in nearshore regions. The default value of this is 10m.

In prior versions of the code, z_burial was also used to scale benthic denitrification in nearshore waters. I changed this to z_denit to avoid any confusion. The default value for this is also 10m, which is what we were using in prior versions of the code.

The section of the code handling the transformations of particulate organic matter in the benthos was also cleaned up for this pull request. This includes the addition of substantial comments. This addition should not change answers, but it should ease exploration of the impact of burial on the hypoxic zone in association with issue #70.

[yichengt900]

@charliestock,

Thanks for adding the new scale_burial option. Below, please find the test results with scale_burial = 1.0 (red line, z_burial = 10m for all the tests). As we expected, the hypoxic zone has shrunk and is now pretty close to the baseline (results from Andrew's paper).

[charliestock]

Great news @yichengt900. Would it be possible to do a short runs with z_burial = 20m and z_burial = 30m to quantify the general sensitivity. I would think it would then be safe to pull this in.

[yichengt900]

@charliestock, definitely. Below, please find the additional sensitivity tests for z_burial=20m (magenta) and z_burial=30m (yellow). As we expected, z_burial can reduce excessive burial in shallow waters (e.g., nearshore regions) and, as a consequence, increase the hypoxic zone.

[charliestock]

Thanks @yichengt900. This really helps to understand the burial sensitivity, though it seems like we would have more work to do to fully understand and calibrate the hypoxic zone dynamics. The challenge seems to be pushing the initiation earlier about 2 weeks to a month earlier in the year while also reducing it earlier. It will probably take a more focused look at the phys and BGC dynamics to trim up the dynamics further. However, even the "no burial" case gives a hypoxic zone of reasonable size. If I had to choose one of the options as the "best", it looks like scale_burial = 1 and z_burial = 20m.

We could perhaps do a quick complementary pair of simulations with z_denit = 0m and z_denit = 20m to assess this sensitivity, branching these from the scale_burial=1/z_burial = 20m case. Lower z_denit should reduce hypoxia by allocating more sedimentary remineralization to anaerobic pathways, and higher z_denit should do the opposite. I'd be particularly curious whether z_denit = 20m shifts the development of hypoxia earlier in the year. @yichengt900, could I cajole you into running two more quick sensitivity tests?

Once we have these in hand, perhaps @andrew-c-ross could start a longer simulation to understand the implications of any burial/denitrification changes done to improve the hypoxic zone on other aspects of the simulation? What do you guys think?

[charliestock]

Sounds good Yi-Cheng, thank you!

[yichengt900]

@charliestock, below is the plot for the two additional sensitivity experiments (z_denit=0m and z_denit=20m with scale_burial = 1 and z_burial = 20m). It appears that z_denit does not significantly affect the peak, although there is still a delay of a few weeks. However, it does slightly alter the shape after the July peak.

[andrew-c-ross]

@charliestock @yichengt900 I'd be happy to run a longer simulation if we've settled on a set of parameters.

I've also been working on a diagnostic for bottom dissolved oxygen at the A01 mooring in Massachusetts Bay. This could be a test for whether this PR has any relevant effects outside of the Mississippi area. The simulation I ran in April looks pretty decent, with a slight high oxygen bias mainly from missing a few of the observed low extremes. Most of the observed daily and monthly variability is not captured, though in addition to being a very nearshore point I think there are relevant processes like wastewater discharge that we don't model.

[yichengt900]

@jessluo, thanks for the comments! I've fixed the typos. @charliestock, please feel free to chime in if you think this PR is ready to go.

[andrew-c-ross]

I will run NWA12 with 10m/10m and 20m/20m

[yichengt900]

@andrew-c-ross and @amoebaliz,

I have merged this PR into the dev/cefi branch. I also updated our CEFI-regional-MOM6 repository, so you can use it directly for the scal_burial sensitivity experiments. Similar to MOM6, we can now define and override the default parameters using COBALT_input and COBALT_override respectively:

do_case2_mod = True       
scale_burial = 1.0
z_burial = 20.0
z_denit = 20.0