Add Complex Sounding Layout Example

[kylejgillett]

Description Of Changes

Adds a more advanced Skew-T layout example to examples\. Closes issue #2460. Uses the existing Advanced_Sounding.py example to build into a more complex plot. The same imports needed for the aforementioned existing example are used in this new example. Some simple adjustments and additions, mostly through Matplotlib manipulation, help improve the readability and complexity of the base MetPy functionality.

This example produces this plot from MetPy test data:

Checklist

• Closes Improve SkewT layout examples #2460
• Fully documented

[CLAassistant]

All committers have signed the CLA.

[kgoebber]

Thanks for the addition, @kylejgillett! Just a few things to clean up and I love that the example uses so many of the sounding calculations and of course the layout structure is great!

[kgoebber]

All looks good to me. @dcamron any other feedback?

[kylejgillett]

Hey guys,

I just check out this example under the dev release and it is showing up kinda wonky.

https://unidata.github.io/MetPy/dev/examples/Complex-Sounding-Plot.html

I think this may be due to the use of the these lines:

###########################################

A few of them need to be removed. Should a fix this with a new commit?

[kgoebber]

Yes, apologies for not catching that on the initial review. While were at it, let’s add Advanced to the beginning of the example title. And, the sigtor and supercell composite give back a list value, so you’ll want to call `signor[0].m` and `super_comp[0].m` to future proof the example due to the way Numpy will handle passing those to the `int()` function. When I make on the docs on my machine, the figure seems to be still not quite right (portions of the left and bottom of the figure are cut off and I can’t seem to figure out what might be causing that. So, go ahead and put in a new PR and I’ll be sure to create the docs locally for review and ensure the final image is appearing as desired.

…

-- Kevin Goebbert, Ph.D. Professor of Meteorology Kallay-Christopher Hall, 201-F 1809 Chapel Drive Valparaiso, IN 46383 (219) 464-5517 fax: (219) 548-7738 https://kevingoebbert.youcanbook.me/

On Sep 15, 2023 at 3:04 PM -0500, Kyle J Gillett ***@***.***>, wrote: Hey guys, I just check out this example under the dev release and it is showing up kinda wonky. https://unidata.github.io/MetPy/dev/examples/Complex-Sounding-Plot.html I think this may be due to the use of the these lines: ########################################### A few of them need to be removed. Should a fix this with a new commit? — Reply to this email directly, view it on GitHub, or unsubscribe. You are receiving this because you modified the open/close state.Message ID: ***@***.***>

[kylejgillett]

Hey @kgoebber,

I may have recreated the plotting issue you found. Is this what you saw?

I could only recreate this when using plt.savefig(), and not with plt.show(). Using plt.show() in a Jupyter Notebook creates the correct-sized plot for me on my machine. However, I corrected this issue with plt.savefig() by adding the bbox_inches='tight' kwarg. That created this:

Could it be some sort of issue with the plt.figtext() and the way I am calling new axes (ex: plt.axes((0.43, 0.40, 0.5, 0.5))), that may appear differently on varying machines due to different aspect ratios/resolutions or something along those lines? I can't seem to find any other discussion on an issue like this.

How would you feel about just including the plt.savefig('filename', bbox_inches='tight') instead of plt.show()?

[kgoebber]

Interesting. I don't think we want to introduce a plt.savefig() here. I've been attempting to run down a few things, most of which did not bear fruit. However, if in setting up the skew axes and setting the rect anchor point (left, bottom) to 0.05, 0.05 instead of 0.0, 0.0 and amending all of the plt.figtext() calls to add 0.05 to both the x and y positions, then the figure appears with the axes in the documentation (with still a bit of extra white space on the right side of the figure). The hodograph would also need to be reposted by adding 0.05 to both the left, bottom values. If those changes are made, how does the output from a Jupyter Notebook look?

[kylejgillett]

Making those changes did not change how the plot looks in the Notebook, so that's good.

[kgoebber]

@kylejgillett are you planning on pushing a new PR to make the corrections? If not, just let me know and I can do that.

[kylejgillett]

Hey @kgoebber, I created a new pull request with the updated code.