Scaling between melt and circulation strength

[claireyung]

Scatter plots of all time for each model shows similarity across models

For Ocean1 and Ocean2

[claireyung]

Empty notebook here: https://github.com/misomip/isomip-plus/blob/main/notebooks/melt-circ-strength/melt-circ-strength.ipynb

Draft figure from Xylar's presentation:

[claireyung]

@nicojourdain at the second hackathon we discussed this plot - are you planning to add a fit and statistical tests to see how these compare to theory?

[nicojourdain]

Hi Claire,
yes about the fit and statistical test. And I'll check if theory can predict such a value, but not sure about the result...

[nicojourdain]

Hi @claireyung
What version of "MiTgcm BAS Coupled" should I use, is it for example Ocean1_COM_MITgcm_BAS_CoupledV3.nc or Ocean1_COM_MITgcm_BAS_Coupled.nc ?
Whatever the version, this model has wrong values for the "totalMeltFlux" variable (several orders of magnitude difference compared to all the other models). Has it been identified?

[nicojourdain]

I've updated the notebook in a pull request. I've made the plot vs mean melt rate to be consistent with other plots and avoid the aforementioned issue with MiTgcm BAS Coupled. Let me know if I need to improve anything.
circulation_vs_melt_Ocean12.pdf

[claireyung]

Thanks @nicojourdain, this looks great! Perhaps if the legend dots could be made bigger that would help with visibility!

We have been using V3 of the coupled BAS MITgcm results, i.e. Ocean1_COM_MITgcm_BAS_CoupledV3.nc. Does that improve the fit with mean melt rate vs circulation? Looking at the streamfunction plots, BAS Coupled doesn't seem weak compared to others (#6) so it being an outlier surprised me.

I don't think anyone has been using the totalMeltFlux variable in our analysis. However, it is included in the general plotting script (modified from Xylar's drive) and in those plots, totalMeltFlux looks sensible for BAS Coupled v3? Here is the link: https://github.com/misomip/isomip-plus/blob/main/notebooks/general-plotting/ScalarTimeseries.ipynb