New Forecast Soundings Online - SHARPpy

Suggestion: Use UWyo's upper air soundings page for the RAOB? It seems to be updated much faster than where you're using currently...
That's actually something I've wanted to do for a while. My hangup is with the interface: I want people to be able to go back for the entire archive, but the current interface (a list of available times) would make that a pain in the neck. I'm thinking of some kind of calendar widget, but that's for the future.

What's the easiest way to update the program without using GitHub? I have Anaconda and Windows; or would github be the easiest?
One option would be to rename the SHARPpy directory to something else (like "SHARPpy-old") and then use whatever procedure you used to obtain the code the first time. Make sure you run 'python setup.py install' on the new code, though. :) After verifying that your new version of the code works, you can delete SHARPpy-old, if you want. I'm a fan of the github solution that Rob suggested, but I also understand people not wanting to mess with git for only one thing.
 
Just wanted to provide people with a heads up: SHARPpy has just released another major update, adding some pretty cool features, including displaying multiple model forecasts/observed soundings at the same time. There are plenty other features too, so I recommend reading through the release notes to get acquainted with the new stuff.

RELEASE: https://github.com/sharppy/SHARPpy/releases/tag/v1.3.0-Xenia-beta

Additionally, the README document has been restructured to give clearer descriptions of features and how to use the program.
README: https://github.com/sharppy/SHARPpy/blob/master/README.md

Lastly, we now support Windows and Mac OS X binaries. No more difficult installations and updates - just download the executable, open it, and you're on your way!

Mac OS X 64bit
Windows 8.1 64bit
Windows 7 64bit
Windows 7 32bit
 
Impressive. Was literally able to download the zip folder, extract the files, and run it right out of my Downloads folder first time. Not a problem was had. Thanks!

I should add that I did this on a machine running 64-bit Win 7.
 
I thought I might go ahead and share an example of one of my personal favorites in the new features category: the ability to do dprog/dt analysis of forecast models.

There is no automatic function to do this yet, so you have to add in each sounding one at a time. Perhaps in the next update we'll change that. However, the idea being that you can add, say, all GFS forecasts for a particular location and time, and see how they vary from run to run. This image is of every GFS forecast available to the program for Yankton, SD, at 00Z tonight. In front is this morning's 12Z model forecast, and in the background are the 06Z, 00Z, and 18Z model forecasts for the same location and time. It's very useful for evaluating model consistency and certainty.

There's no reason to limit it to just GFS. You could also add all NAM and RAP forecasts for the same location and time (if the location is available in the dataset) and get a model-comprehensive look at uncertainty. Anyway, just thought it was pretty neat/powerful.

8cc07ea66768d93372d575b2dfc9eff3._.jpg
 
We are presenting SHARPpy to the attendees of the Annual National Weather Association meeting being held in Oklahoma City this October, and we would like to know more about who uses SHARPpy and why!

If you could please take the time to fill out this short survey, we would greatly appreciate it. The responses to this survey will be presented at the annual meeting, so by taking the survey we are assuming consent. The responses are anonymous, save for the affiliation of your organization (one of the form questions).

Please share this with other users of SHARPpy so that we can get a comprehensive sample!

http://goo.gl/forms/TLJFC022o2
 
For those people wondering how to find and/or display the SARS or hail analog soundings (Linux executables) within the SHARPpy database, here's a four piece tweet on how (should work for all platforms).

https://twitter.com/ahberrington/status/684242683062165505

By the way, can anyone confirm if this is also works with the binary download? Have had someone tell me that it doesn't.
 
Last edited:
Andy, it will not. Our binary has the SARS sounding files bundled within the binary, and it may be difficult to get at the files. Your method should only work as long as you have downloaded the source code from Github and also installed that same source.
 
Back
Top