WIP: First draft of self-generating cover art on landing page

[brian-rose]

Addresses #27

Here I've replaced landing-page.md (which was just a placeholder) with a new landing-page.ipynb notebook.

The notebook contains a code cell that generates an image. It's a placeholder image for now (just a map of the world built with Cartopy).

The key is that I've set the hide-input cell tag on the code cell, so JupyterBook renders the image without showing the code. But also gives a little plus sign button to expand the code cell.

Self-reproducing and teachable cover art!

Now we need to actually design some good artwork.

[review-notebook-app]

Check out this pull request on 

See visual diffs & provide feedback on Jupyter Notebooks.

---

Powered by ReviewNB

[github-actions]

🚀 📚 Preview for git commit SHA: f18a50b at: https://60b13179f657163ea3f0a1ef--pythia-foundations.netlify.app

[brian-rose]

We'll have to figure out how to suppress those Cartopy download warnings....

[github-actions]

🚀 📚 Preview for git commit SHA: 76dc9f2 at: https://60b13ca6f3fc720482158058--pythia-foundations.netlify.app

[brian-rose]

@ProjectPythia/education please add here any of your cool thoughts about possible cover art images.

[clyne]

I like the idea of using the whole earth, and I like the Mollweide projection. It would be nice to have some compelling data mapped over it. Perhaps the GeoCAT team can help. CC @erogluorhan

[ktyle]

A global satellite mosaic, perhaps including an overlay of the lightning strikes via the GLM, would be cool. Let me see if I can get something whipped up ...

[ktyle]

I haven't found a global mosaic yet, but here is a GOES mosaic that covers the western hemisphere. One issue though is that the source GeoTIFF is quite large (17000x17000x4) and thus would not be an ideal candidate for a notebook that self-generates a cover image. But this one looks pretty :)

[github-actions]

🚀 📚 Preview for git commit SHA: c15899d at: https://60be7947336542108fde925e--pythia-foundations.netlify.app

[erogluorhan]

I like the idea of using the whole earth, and I like the Mollweide projection. It would be nice to have some compelling data mapped over it. Perhaps the GeoCAT team can help. CC @erogluorhan

Looking into it.

[github-actions]

🚀 📚 Preview for git commit SHA: 25e8b71 at: https://60c99a085639d97012a00ffb--pythia-foundations.netlify.app

[michaelavs]

Hi everyone,
Orhan asked Anissa and I to look into this a bit, so I've recently been working on putting together a data overlay on the pretty-earth.png picture mentioned. I have a couple of comments/ideas I wanted to run past you all now that I have a semi-working example.

For the graphic, I wanted to make something that incorporated a partial data overlay on the image so we could kind of have the best of both worlds with respect to what was mentioned above. I found that the image does not like to work very nicely with cartopy because it isn't a geotiff file (or at least that is my understanding on why there was some difficulty with setting extent and central lats wrt getting the cartopy map to line up with the image). So, after a lot of brute force, this is the image I have created with a wrf data overlay:

and to put it bluntly, it's kind of boring. However, it does show some functionality that may be useful for users so I feel we may want to incorporate it in some fashion, but not as a solo image.

With that being said, I did look into how NOAA and NREL have set up graphics for some of their pages to get some ideas on where we may take this graphic. I specifically like how the NOAA climate landing page showcases a bunch of data projections using their datasets. I think we should do something similar to that. A couple of ideas how we could accomplish that are:

1. Get a robust netcdf file (i.e. many variables) create a subplot showing a couple of the variables being plotted and use that. The subplot notebook would be uploaded with this.
2. Make a separate png file that shows a couple different projections in the image, then add the separate notebooks on how to create the projections to the website in a separate section (something like "landing page graphics") so users can find and use the code that created the projections.
3. Maybe finding a smaller geotiff file over the United States (or Colorado if needed to be small enough for upload) and essentially repeat what I had originally but with different data, and then again combining them into one subplot and using that.

I am completely open to other's ideas in this, so please feel free to critique or modify any of the ideas I have above. I also recognize it has been almost a month since this issue/PR was talked about, so if we've moved on from it all together, also ok, just let me know!

[dopplershift]

I'm happy to dig into any Cartopy challenges if that would help. Just need the code and any data files.

[michaelavs]

I'm happy to dig into any Cartopy challenges if that would help. Just need the code and any data files.

I can open a separate pull request with the notebook in it and links to the datafile if that would work for you?

[dopplershift]

That makes sense if the notebook is supposed to go into this repo (no idea if so). Otherwise, you could always create a Gist (https://gist.github.com) with the multiple files and link it here.

I'm fine either way though.

[michaelavs]

That makes sense if the notebook is supposed to go into this repo (no idea if so). Otherwise, you could always create a Gist (https://gist.github.com) with the multiple files and link it here.

I'm fine either way though.

I'll go ahead and make a separate PR and tag you once it's ready for you to take a look at!