EAMxx: Allow to save a dictionary of fields in the FieldManager

[bartgol]

The option (off by default) allows to save a yaml file containing a dictionary of the fields on the "Physics" grid, with some of their details.

---

As an example, this is a snippet of the file content I get in one of our units tests:

ps:
  layout: [ncol]
  standard_name: surface_air_pressure
  units: Pa
pseudo_density:
  layout: [ncol, lev]
  standard_name: pseudo_density
  units: Pa
pseudo_density_dry:
  layout: [ncol, lev]
  standard_name: pseudo_density_dry
  units: Pa
qc:
  layout: [ncol, lev]
  standard_name: cloud_liquid_water_mixing_ratio
  units: kg/kg

I may want to add something else, so if you have suggestions, fire away. E.g., I want to save the list of atm processes in the EAMxx configuration that generated the yaml file.

[github-actions]

PR Preview Action v1.4.8
🚀 Deployed preview to https://E3SM-Project.github.io/E3SM/pr-preview/pr-6842/
on branch gh-pages at 2024-12-12 00:44 UTC

[AaronDonahue]

@bartgol This issue lists a number of "wishlist" items for run logging: E3SM-Project/scream#1372 . There might be a few things there that we can add to this PR. We can also move this Issue over to E3SM too.

[bartgol]

@bartgol This issue lists a number of "wishlist" items for run logging: E3SM-Project/scream#1372 . There might be a few things there that we can add to this PR. We can also move this Issue over to E3SM too.

Of all those items, I think the only one that can make sense here is to list the providers/customers of each field. It's not going to tell the user the order in which things happen, but at least it's going to tell the user why the field is in the FM...

[bartgol]

@AaronDonahue @mahf708 I changed a bit the formatting:

• the filename is more descriptive now
• I added a better description entry for the file
• I added the run provenance info to the output
• I added customers/providers for each list

Snippet:

%YAML 1.1
---
description: |
  content of the EAMxx FieldManager corresponding to the 'Physics' grid.
  The dict keys are the field names as used in EAMxx.
  For each field, we add the following entries:
    - standard_name: the name commonly used to refer to this field in atm sciences (if applicable)
    - units: the units for this field used in EAMxx
    - layout: the names of the dimensions for this field (time excluded)
    - providers: the atm processes that update/compute this field
    - customers: the atm processes that require this field as an input
  
fields:
  LW_clnclrsky_flux_dn:
    customers: []
    layout: [ncol, ilev]
    providers: [rrtmgp]
    standard_name: LW_clnclrsky_flux_dn
    units: W/m2
...
  SW_flux_up:
    customers: []
    layout: [ncol, ilev]
    providers: [rrtmgp]
    standard_name: SW_flux_up
    units: W/m2
  T_mid:
    customers: [homme, shoc, p3, rrtmgp]
    layout: [ncol, lev]
    providers: [homme, shoc, p3, rrtmgp]
    standard_name: air_temperature
    units: K
...

I would like to merge an EKAT pr first, update ekat, then merge this, since it would help simplifying the format of the output file (e.g., output inline blah: [a, b] sequences rather than multiline ones).

[mahf708]

Looks good!