JPL Notebook

[smk0033]

Adding the JPL notebook and supporting submodule file for Science Support Team to review. Github ticket

[review-notebook-app]

Check out this pull request on 

See visual diffs & provide feedback on Jupyter Notebooks.

---

Powered by ReviewNB

[github-actions]

PR Preview Action v1.6.3
🚀 View preview at https://US-GHG-Center.github.io/ghgc-docs/pr-preview/pr-162/
Built to branch gh-pages at 2025-12-16 18:50 UTC. Preview will be ready when the GitHub Pages deployment is complete.

[wildintellect]

@Jeanne-le-Roux I have concerns with the Copyright selected. Is there a reason they can't use something like CC--BY-NC

As is the licensing makes it unclear that anyone can use the code or modify it.

I'll also note the Quarto cite is CC-BY, and Code repo is currently Apache 2.0

[review-notebook-app]

View / edit / reply to this conversation on ReviewNB

wildintellect commented on 2025-04-24T17:31:54Z
----------------------------------------------------------------

No mention of which compute environment or minimum/suggested size is needed to execute this on the hub. Guessing we want Pangeo, but unclear on Ram selection.

[review-notebook-app]

View / edit / reply to this conversation on ReviewNB

wildintellect commented on 2025-04-24T17:31:55Z
----------------------------------------------------------------

Can we suggest a reference at least which field of study this is. I'm guessing this is Inverse Modeling in statistics https://www.sciencedirect.com/science/article/pii/S0168169924000152 based on mention in https://dus.jpl.nasa.gov/projects/flux-inversion/

More background https://www.sciencedirect.com/topics/earth-and-planetary-sciences/atmospheric-inversion

[review-notebook-app]

View / edit / reply to this conversation on ReviewNB

wildintellect commented on 2025-04-24T17:31:55Z
----------------------------------------------------------------

Line #9.    from inventory_evaluation_helper import *  # Import all functions from the inventory evaluation helper module

This should be the last import as it's local.

Reminds me that we should enable nbqa https://github.com/nbQA-dev/nbQA like VEDA-docs does to fix general python linting issues.

Linting will also catch the use of Wildcard imports * which are always discouraged due to security concerns.

[review-notebook-app]

View / edit / reply to this conversation on ReviewNB

wildintellect commented on 2025-04-24T17:31:56Z
----------------------------------------------------------------

Eliminate the cause of this error "Country United States not found in the shapefile."

[review-notebook-app]

View / edit / reply to this conversation on ReviewNB

wildintellect commented on 2025-04-24T17:31:57Z
----------------------------------------------------------------

Consider applying a Projection more appropriate to the display of US data. Is Equal Area or Shape Distortion more important?

[wildintellect]

I reviewed the general text and code, comments above.

To be reviewed:

• Exectuting the code
• The scientific text for legibility

[smk0033]

Bringing @sudshu in as well for collaboration and to track progress. I added a cell to bring in the data files from Zenodo (I think it can be optimized which other members of Science Support can weigh in on, but I was able to run the entire notebook with it), and also added engine = "netcdf4" to the helper files for xarray. In addition, I formatted a raw cell at the top to match formatting with our other notebooks, and added packages for the Zenodo cell.

[HarshiniGirish]

Only notebooks can be reviewed on ReviewNB so just pasting all the comments for
user_data_notebooks/user_notebooks/inventory_evaluation_helper.py here

1. Using from pylab import * at the top of the module clutters the namespace and makes it unclear which functions come from NumPy/matplotlib.
   Replace it with explicit imports like import numpy as np and import matplotlib.pyplot as plt, and use np.zeros, np.arange, etc., to keep the module cleaner and safer to import.

2.In get_country_geometry, the "not found" message is inside the loop, so it prints once for every non-matching country, creating noisy and misleading output.
Only handle the match inside the loop, then after the loop either print a single "not found" message or raise a ValueError if no match was found.

3.In give GOSATdata, emiss_values is computed inside the loop but never used, which is confusing (please check again before removing it)

4. In makeGeoPandasDataFrame / makeGeoPandasDataFrameFOG, USA_area_fraction is now calculated as (intersection with USA / total cell area), but the comment still describes it as “1 minus the fraction in Canada and Mexico,” which is no longer true.
   Update the comment to reflect the current calculation (e.g., “fraction of each grid cell inside the U.S. boundary”) or re-implement the Canada/Mexico subtraction explicitly and keep the comment consistent.

apart form the code looks really good @smk0033 . Thank you

[review-notebook-app]

View / edit / reply to this conversation on ReviewNB

HarshiniGirish commented on 2025-11-18T22:02:58Z
----------------------------------------------------------------

Add a simple check at the start of the cell to skip the download when data_folder_path is already populated, and include a short comment explaining that this cell downloads the Zenodo benchmark data and only needs to be run once. This avoids repeated downloads and makes rerunning the notebook much smoother.

[HarshiniGirish]

@smk0033 I've run the notebook in the ghg hub too, apart from the above the code looks good to me. Thank you

[smk0033]

Only notebooks can be reviewed on ReviewNB so just pasting all the comments for user_data_notebooks/user_notebooks/inventory_evaluation_helper.py here

1. Using from pylab import * at the top of the module clutters the namespace and makes it unclear which functions come from NumPy/matplotlib.
   Replace it with explicit imports like import numpy as np and import matplotlib.pyplot as plt, and use np.zeros, np.arange, etc., to keep the module cleaner and safer to import.

2.In get_country_geometry, the "not found" message is inside the loop, so it prints once for every non-matching country, creating noisy and misleading output. Only handle the match inside the loop, then after the loop either print a single "not found" message or raise a ValueError if no match was found.

3.In give GOSATdata, emiss_values is computed inside the loop but never used, which is confusing (please check again before removing it)

4. In makeGeoPandasDataFrame / makeGeoPandasDataFrameFOG, USA_area_fraction is now calculated as (intersection with USA / total cell area), but the comment still describes it as “1 minus the fraction in Canada and Mexico,” which is no longer true.
   Update the comment to reflect the current calculation (e.g., “fraction of each grid cell inside the U.S. boundary”) or re-implement the Canada/Mexico subtraction explicitly and keep the comment consistent.

apart form the code looks really good @smk0033 . Thank you

Hi @sudshu! I've made a few updates to pull in the Zenodo data. Our Science Support team aided in reviewing, and I've addressed comment 2 here. Would you like to weigh in on comments 1, 3, and 4 and have me address those as well? You are also more than welcome to push updates to this branch if you wish at any time! I think this is about ready to send to Emily for her review.

[sudshu]

@smk0033 your suggestions look good to me. Feel free to implement those.

[HarshiniGirish]

Thanks @smk0033. I ended up overlooking this bit

1. EPA scrollable maps issue : Under # EPA inventory, the code cell that starts “Create a GeoPandas DataFrame … for the "epa" inventory” (gdf_epa = ...). It does not use with out: and the for-loop indentation is off, so the out widget ends up empty and nothing is actually scrollable.

2. add imports Output

[HarshiniGirish]

Thanks @smk0033. I ended up overlooking this bit

1. EPA scrollable maps issue : Under # EPA inventory, the code cell that starts “Create a GeoPandas DataFrame … for the "epa" inventory” (gdf_epa = ...). It does not use with out: and the for-loop indentation is off, so the out widget ends up empty and nothing is actually scrollable.

2. add imports Output