ice_in variables undefined in chapter 5 of docs

[rgrumbine]

https://cice-consortium-cice.readthedocs.io/en/main/cice_index.html

As we discussed at the meeting, not all variables are expected to be defined here. Still, these are variables from sections of ice_in I was looking at for setting loose an evolutionary optimization program and which don't have definitions in the main docs:

&dynamics_nml
evp_algorithm 'standard_2d'
Ktens 0.
capping_method 'max'
k1 7.5
k2 15.
alphab 20.
threshold_hw 30.
coriolis 'latitude'
kridge 1
ktransport 1
ssh_stress 'geostrophic'
maxits_nonlin 10
precond 'pgmres'
dim_fgmres 50
dim_pgmres 5
maxits_fgmres 1
maxits_pgmres 1
monitor_nonlin .false.
monitor_fgmres .false.
monitor_pgmres .false.
ortho_type 'mgs'
reltol_nonlin 1e-8
reltol_fgmres 1e-1
reltol_pgmres 1e-6
algo_nonlin 'picard'
use_mean_vrel .true.
/

&shortwave_nml
snw_ssp_table 'test'
sw_redist .false.
sw_frac 0.9d0
sw_dtemp 0.02d0
/

&forcing_nml
fe_data_type 'default'
/

[apcraig]

I believe all of these are documented (I just checked 5 of them and they were all there), https://cice-consortium-cice.readthedocs.io/en/main/user_guide/ug_case_settings.html. The documentation is grouped by different name namelist groups, is that where the confusion is? Should we change the table so they are all alphabetical and we add a column indicating which namelist group they are in?

[rgrumbine]

Ah. That's good. But perhaps there should be a link from section 5 over here? It was section 5 I was looking at, which I agree isn't the best place but one a new user may well look.

Or maybe all variables in section 3 ug_case_settings could automatically be dragged in to section 5? https://cice-consortium-cice.readthedocs.io/en/main/cice_index.html

[apcraig]

OK, I understand now. Section 5 (the code variable index) is sort of interesting. It's not comprehensive (and it probably can't/shouldn't be) and it's not supposed to cover the namelist/model inputs completely either. There already is a link to the namelist/model inputs section at the top of section 5, see the top paragraph in section 5 (I guess that's easy to overlook). I don't think we should duplicate the namelist inputs into section 5. About the only thing I think we could do is place the link to the namelist settings in a separate line at the top of section 5, taking it out of the paragraph so it's more obvious? Maybe we should also rename section 3.4 from "Case Settings" to "Model Namelist Inputs, Model CPPs, and Other Settings"?

[rgrumbine]

I like your new name for section 3.4

Even knowing the link was in that short paragraph, it was almost invisible to me. Something about the color scheme and wording, I guess.

So yes, I like your solution of splitting the text to something like:

• Below are variable descriptions for those variables which are ... (including describing why certain classes of variables aren't present here)
• For parameters involved in controlling a model run see section 3.4 -- Model Namelist Inputs, Model CPPs, and Other Settings