Viewing archived soundings in SHARPpy

[James Hammett]

I'd like to get archived model output into SHARPpy for analysis. Looks like http://mtarchive.geol.iastate.edu/ might have the source data I need. How do I get it into a format that SHARPpy can read? It's supposed to look like this: https://github.com/sharppy/SHARPpy/blob/master/tutorials/14061619.OAX

 

[Jeff Duda]

I doubt the ISU archive has model soundings archived. I guess I don't know exactly where in the directory structure to look, but sounding data I looked at looked like observations.

Okay, looking closer I see they have gempak data stored. Probably horizontal model cross sections that you'd have to parse through to construct a sounding manually. I don't read GEMPAK data so I can't help you with that. Perhaps Rob Dale can.

 

[rdale]

I've never seen (or tried) a way to convert model soundings. You might play with this UWyo observed soundings converter?

https://www.facebook.com/SHARPpy.wx...=1821120834812199&comment_tracking={"tn":"R"}

 

[Greg Blumberg]

Hi James. SHARPpy (the app) can load in archived Bufkit soundings from IEM. I've used this functionality several times in my own research. Download a .buf file and use the menu bar when the SHARPpy sounding picker window opens up to select "Open..." Navigate to the Bufkit file and you should be set!

If you don't have Bufkit data, then you'll have to make your own files straight from the model data.

 

[Brian G]

Hmm...I've never been able to get SHARPpy to open a buf file. It doesn't matter if it comes from the PSU Bufkit Data Distribution, IEM, or any other source. I always get the error "Could not figure out the format of '<filename>'!"

 

[GPhillips]

Hi Greg. Does this work for the latest compiled SHARPpy (Xenia), or only in the very latest build? Is there any needed naming convention for the bufkit file? I get the same errors as Brian using Xenia.

 

[Tim Supinie]

I was able to load a Bufkit file from IEM into the latest version of Xenia, both using the binary and when running from the source code. However, there is currently a bug that shows up after loading in which you can't switch between the forecast hours.

You might try to update to the latest version of Xenia. I've fixed a few bugs since the release that might affect the ability to load in Bufkit soundings from disk. Loading bufkit files from disk should also work in the development version. I'll see if I can fix the bug with the forecast hours.

 

[James Hammett]

I'm getting the same error as Brian running the v1.3.0 Xenia binary using http://mtarchive.geol.iastate.edu/2016/04/26/bufkit/15/rap/rap_kmhk.buf

 

[Brian G]

Hmm...no luck. I just got the latest version from the master branch. It runs fine and everything works except the opening bufkit files. I noticed that I had pyside 1.2.1 so just in case I updated it to 1.2.4 (which apparently updates a lot of other packages) and now sharppy won't run at all. I'll try uninstalling Anaconda and then reinstalling everything from scratch when I get time.

 

[Tim Supinie]

Are you guys on Windows, by chance? It works fine on my Mac, but when I tried it on my wife's Windows computer, I got the same error as you guys.

Also, @Brian G, that is a known issue with PySide. You'll have to downgrade back to v1.2.1 or install from the conda-forge channel, which has a working version of v1.2.4.

 

[Brian G]

Are you guys on Windows, by chance? It works fine on my Mac, but when I tried it on my wife's Windows computer, I got the same error as you guys.

Also, @Brian G, that is a known issue with PySide. You'll have to downgrade back to v1.2.1 or install from the conda-forge channel, which has a working version of v1.2.4.

Thanks Tim. Yes, I'm on Windows. Good to know on the pyside. I'll just downgrade it.

 

[Tim Supinie]

Okay, I was able to fix the error on a Windows computer. @Brian G (and @James Hammett, if you can), would you try downloading the latest version of the code from the master branch and try loading in bufkit files from the disk? It should work this time.

 

[Brian G]

Okay, I was able to fix the error on a Windows computer. @Brian G (and @James Hammett, if you can), would you try downloading the latest version of the code from the master branch and try loading in bufkit files from the disk? It should work this time.

Works like champ! Thanks for your help Tim!

 

[JF Massicotte]

Hi James. SHARPpy (the app) can load in archived Bufkit soundings from IEM. I've used this functionality several times in my own research. Download a .buf file and use the menu bar when the SHARPpy sounding picker window opens up to select "Open..." Navigate to the Bufkit file and you should be set!

If you don't have Bufkit data, then you'll have to make your own files straight from the model data.

I have a small utility that I built myself in python to build sounding data out from gridded model data. Not very fast and still needs some work to guarantee portability to windows but if there is enough demand I may try to fix it quickly and make it availlable to everyone.