Lol if this isn't me

That was a gorgeous halo!!

Thanks

Best I can do is

Tornado
Hail
Wind
Unified
Mapping
Badaboom
Omg

(THWUMBO)

LOL you guys 🤣🤣🤣

Thanks man!! 😄 I appreciate you

Here you go!!

The long tracks here I've found are usually created by some robust, distinctive mesovortex/embedded supercell with an obvious persistent track

So that derecho could be made up of 3 dozen or more smaller swaths

We can chat about this next time we chat!! (sorry, haven't been able to check out the radar stuff yet, but that's definitely on my short-term list). Essentially though, this is a more storm scale report clustering. It won't pick out massive MCS-scale swaths but smaller swaths within it

Oh heck yes this is awesome!! 😄

Yup!!

Ooohh I haven't seen this yet, thanks for sharing this! Yes, 100%, my *only* goal with using MESH is to detect a reasonable portion of observed hail events, such that it may determine if reports were connected. As of this time, no intention of using the actual values except as a threshold!

Yes! I'm hoping that even just the existence of MESH data in these rural areas (without even assessing its "expected size") could help us with the population issue. I'm currently exploring ways to incorporate these late reports, as they're obviously important!

It's possible! I think the biggest issue here though is that the threshold I use for swath-detection (0.75") is simply too high to capture many events (especially in the cool season Oct-Feb). I'm exploring something as simple as using a variable threshold by month to capture these.

I agree wholeheartedly. My near-term goal is to simply get to a point where I have a reasonably high detection rate and MESH swath for each observed hail event, and will *not* be using actual MESH values for any analysis except for fun. Just for object detection!