Anthony Haberman

Before Matplotlib, VisIt, VAPOR, ParaView, there was this (even though it was before my time), so i wanted to go back in time for a little bit. I remember seeing it being used heavily in old weather documentaries. And yup! All through WSL toolkit, essentially Linux in a box for Windows.

What y'all know about Vis5D?

A 01Z forecast sounding also at Kellerville bears a striking resemblance the one synthesized from ELDORA data in Dowell and Bluestein (2002).

I recently ran a 12 km MPAS simulation with ERA-5 data of the 06/08/1995 TX tornado outbreak. Wanted get a better idea of the mesoscale environment the storms were working with. Here is a point sounding near Kellerville TX at 00Z!

Ah I see now, we're on v2.4

cc: @drbwx.bsky.social

Looks like we're on version 2.103.

Comparison from today's launch here at MU, focused sounding is from the Vaisala MW51 (RS41-SGP) system, the blue profile is from the Windsond (S1H2) system. Both sondes were attached to the same balloon. We were curious of sensor differences between the 2 systems.

Looks like we got 3 new NSSL MPAS configs! I'm most excited to see results with RN3 because its utilizing the NSSL triple moment microphysics scheme which tmk, is the 1st time a triple moment scheme has been used operationally in a model! Props to @ksquared812.bsky.social @tedmwx.bsky.social and co!

Same plot, but following the tornado cyclone.

Its time for some pseudo-RHI animations! Using the AIR dataset from the 2013 El Reno supercell.

@stormscale.io

I have completely remade sharppy in javascript. This is for an upcoming model viewing program and im super happy with the current result of it. sounding from 5/31/2013, at Powers OK. (from twitter)

Using gradient boosting regression, here are the importance of parameters for EF-rating (left), and sig hail size (right). Still figuring out this ML stuff though.

In case you wanted to know what a covariance matrix of 25 parameters from over 11k soundings of significant RM supercells looks like, well here you go! Got inspired by one of the plots from @streamwise.bsky.social's PhD defense.

Finally able to do proper CM1 runs. (876 x 876 x 275, ~211 million gridpoints)

Attempt at recreating Fig 1 from Houston and Wilhelmson 2007a, wasn't able to find old RUC data, so used 21Z ERA-5 instead to fill rest of Calvert sounding.

Managed to cobble together a May 6th composite sounding w/ 22Z HRRR, RAP and our 22:16Z balloon launch.

Man, why does SLURM gotta be so complicated 😭

Trying to figure out the best way to process 410 (def more than this though) soundings that our Met department has launched🫠

Lead the way sir!

Still figuring out how to use this, seems pretty easy so far though!