Skip to content

Vertical cross-section #145

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Open
nguy opened this issue May 17, 2016 · 7 comments
Open

Vertical cross-section #145

nguy opened this issue May 17, 2016 · 7 comments

Comments

@nguy
Copy link
Owner

nguy commented May 17, 2016

Create a vertical cross-section by adding a line to plot. Can use the existing index retrieval method in select region and then have the cross-section plot pop up in additional tab.

@gamaanderson
Copy link
Contributor

I agree with with the functionality, but I don't think we should use the the existing method, but rather just drawing the line in the axis and making the conversion from position in index to plot in a component for it self. This is mainly for performance and more flexibility. I already worked with cross section in an other software.

Should such a function of plotting a vertical section be make part of pyart? @jjhelmus?

@jjhelmus
Copy link
Contributor

I'd be open to including a method for extracting a vertical section from a radar volume to Py-ART. Using a KD-Tree for gate lookup would be recommended.

@swnesbitt
Copy link
Contributor

swnesbitt commented May 19, 2016

Essentially, wouldn't this just require refactoring the pyart grid methodology, for cross sections with arbitrary cardinal direction orientations? I suppose that great circle calculations would technically be appropriate, especially for longer cross sections.

This functionality is available in NCAR's REORDER software.

@jjhelmus
Copy link
Contributor

@swnesbitt Yes, although I think it might be desired to do nearest neighbor interpolation instead of the weighted interpolation that the grid method currently uses. Having a generic function in Py-ART which takes gates from one or more radars and a list of point and returns the fields values for those points using a weighted/nearest/k-nearest interpolator would be really useful in a number of cases. The current gridding function does some of that but some work would be needed to expand the functionality.

@nguy
Copy link
Owner Author

nguy commented May 20, 2016

@gamaanderson @jjhelmus I've been working on a KD-Tree for another application and I agree that this probably would be a good implementation, especially for a polar coordinate data space. I'm finding that the KD-Tree can take a long time to build, just as a heads up (or I'm just messing up the programming).

Let's discuss implementation a little more. Building this in Py-ART and then calling in ARTview seems the idea. I agree with that plan. Where should this go @jjhelmus? If I can get my code working nicely, I can help get this up and running.

@jjhelmus
Copy link
Contributor

I'd put it in the pyart/map directory for the time being. Once the details get fleshed out it may get moved but that seems reasonable for now.

@gamaanderson
Copy link
Contributor

gamaanderson commented May 20, 2016

So here we are again: good vs fast. A reformulation of the current map_to_grid or map_gates_to_grid should do a nice work. This would be fast enought if you follow the principe of drawing a line and then pressing a button for drawing. But not fast enought if you want to make the drawing in real time (while drawing), this is not only something that would look nice, but also the fast response create a 3 dimensional feelling.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

No branches or pull requests

4 participants