Config Cleanup

[uwagura]

This PR starts the process of making the config.yaml file a bit easier use now that it has settings for each of the physics diagnostic scripts. So far, I have just regrouped some of the entries so that similar settings are grouped together and the comments are clearer, but let me know if there are more changes we should make

[yichengt900]

@uwagura, thanks for submitting this cleanup PR. I appreciate your initiative! I think we could take the refactoring a step further by grouping standalone keys that are used within the same script into cohesive dictionaries (or mappings). For instance, rather than accessing figures_dir directly, we could include it within a general section, which would then be accessed as general:figures_dir.

Below is a possible layout for the new config.yaml:

# General settings for dataset locations and figures
general:
    figures_dir: 'figures/'
    model_grid: '../data/geography/ocean_static.nc'

# Datasets paths
datasets:
    glorys: '/work/acr/mom6/diagnostics/glorys/glorys_sfc.nc'
    glorys_zos: '/work/acr/glorys/GLOBAL_MULTIYEAR_PHY_001_030/monthly/glorys_monthly_z_fine_*.nc'
    oisst: '/work/acr/oisstv2/'

# Variables renaming map
rename_map:
    geolon: lon
    geolat: lat
    geolon_c: lon_b
    geolat_c: lat_b

# Domain and general plotting details
plotting:
    domain: ocean_monthly
    lat_range:
        south: 0
        north: 60
    lon_range:
        west: 260
        east: 330
    xlim:
        min: -99
        max: -35
    ylim:
        min: 4
        max: 59
    projection:
        grid: 'PlateCarree'
        data: 'PlateCarree'
    figure_size:
        width: 11
        height: 14
    skill_score_text:
        x: -98.5
        y: 54
        xint: 4
        yint: 4
    plot_lat: False

# SST Evaluation and Trend settings
sst_eval:
    colorbar:
        levels_min: 2
        levels_max: 31
        levels_step: 2
    colorbar_diff:
        bias_min: -2
        bias_max: 2.1
        bias_step: 0.25
        ticks: [-2, -1, 0, 1, 2]

# SST Trends settings
sst_trends:
    start_year: "2005"
    end_year: "2019"
    colorbar_diff:
        bias_min: -2
        bias_max: 2.1
        bias_step: 0.25
        bias_min_trends: -1.5
        bias_max_trends: 1.51
        ticks: [-2, -1, 0, 1, 2]

# SSS Evaluation settings
sss_eval:
    rcdir: '/net2/acr/regional_climatologies/'
    rc_vars:
        - swa
        - gom
        - nwa

# SSH Evaluation settings
ssh_eval:
    colorbar:
        levels_min: -1.1
        levels_max: 0.8
        levels_step: 0.1

This structure could help by consolidating settings relevant to each script, improving readability and maintainability. These are just some initial thoughts—I might have overlooked something. Also, this will require some changes in the Python source code to adapt to the new structure. We can discuss further tomorrow. Thanks again for getting the ball rolling on this!

[yichengt900]

@uwagura, as discussed this morning, I'll go ahead and approve this PR for now and hold off on further config.yaml refactoring until we complete the primary code cleanup. I still think that consolidating the configuration file using a dictionary structure would enhance clarity, but it’s not a priority at this stage. Let’s plan to revisit this improvement later.