Researchers Simulate Monster EF5 Tornado

New submitter Orp writes: I am the member of a research team that created a supercell thunderstorm simulation that is getting a lot of attention. Presented at the 27th Annual Severe Local Storms Conference in Madison, Wisconsin, Leigh Orf's talk was produced entirely as high def video and put on YouTube shortly after the presentation. In the simulation, the storm's updraft is so strong that it essentially peels rain-cooled air near the surface upward and into the storm's updraft, which appears to play a key role in maintaining the tornado. The simulation was based upon the environment that produced the May 24, 2011 outbreak which included a long-track EF5 tornado near El Reno Oklahoma (not to be confused with the May 31, 2013 EF5 tornado that killed three storm researchers).

Comments

Amazing

[JustShootMe]

This is pretty amazing. I've heard the theory before that tornadoes are formed from that same baroclinic horizontal vortex tilting upwards, but the mechanism for that has never made a whole lot of sense to me. The idea of it getting pulled up into the actual mesocyclone itself and powering it so that the tornado can form makes a lot more sense. It also makes it a lot more clear what role the RFD has in tornadogenesis. And that parade of vortices, I'd never heard of that before.

Hopefully this will help the weather people start to see clues that a tornado is trying to form even before the hook starts to become obvious on radar.

Presentation as seen on YouTube well done

[Maow]

That was a very well done presentation even if it was so far over my head that I understood little but, "oooh, pretty".

The pacing was fast, confident, and even had the audience laughing at times. Congratulations.

Now I feel an evil urge to make a joke about how, since your model didn't properly account for "hydrometeor centrifigal whatzits" then it is therefore worthless and you, Mr Orf, like those climate researchers, are in it for the big bucks in grant money to fund your lavish Toyotas and suburban middle class homes.

Or something. I've likely failed at humour. But you've succeeded in your research, kudos.

Re:Presentation as seen on YouTube well done

[JustShootMe]

There was a lot of jargon. Let's see if I can help.

baroclinic == energy created by movement of air because of differences in pressure.
forward flank downdraft == I believe the downdraft caused by the air cooled by the rain, also could be air forced down by the forward movement of the storm. Usually cool and wet.
rear flank downdraft == warm, dry air that hits the storm from the back and is forced downward.
mesocyclone == the rotating updraft in a supercell.

Basically, he's stating that the interface between the different types of air on the ground is creating a rolling tube of air, and the updraft of the storm is so powerful that it sucks that tube up - and the energy of the rotation helps to give the updraft an extra "kick"... which helps to power and maintain the rotation of a long-track tornado. That tube isn't the tornado itself, it just powers the updraft that spins the tornado.

Re:Presentation as seen on YouTube well done

[JustShootMe]

Learning how thunderstorms work is something I think everyone should do on a basic level - they are utterly fascinating. What is amazing to me is that these repeatable and frankly amazing structures are created out of nothing but air, water, and heat in varying combinations.

In layman's terms (and while I'm a layman I'm educated enough to be able to say that) a thunderstorm is simply caused by buoyant air rising. As it rises, the moisture inside of the air condenses, creating a cloud and releasing heat (latent heat). That warms the "parcel" of air more (another term he used frequently) and it rises faster.

Air that rises sucks up more air from below it because it creates a low pressure region. Eventually the air hits the top of the troposphere, which is stable (stable means that the air is warmer than the rising parcel and the rising air is no longer buoyant. Keep in mind that "warmer" is relative and can mean -60F.)

In conditions that cause a storm like this to form, vertical wind shear is important. In a pulse thunderstorm, the downdraft (the rain cooled air) gets in the way of the updraft and chokes it off. But the wind shear not only causes the updraft to rotate, it pushes the downdraft out of the way of the updraft, so nothing chokes it off. This is why when you look at a supercell, it is nearly always tilted to the direction of travel. (mesocyclone is another word for updraft in this case.)

Now that the storm is created, you have room for the other factors he was mentioning, such as the RFD, FFD, etc. Basically there is a certain combination of factors required to set the air at the ground to spinning. The interesting thing about this simulation was that the managed to find the sweet spot and get their simulation to create a long-tracked tornado. Much of his presentation was spent highlighting certain parcels of air and showing how they got ingested by either the meso or the tornadic circulation (which are related but not necessarily the same thing.)

Some technical info for slashdotters

[Orp]

I wanted to give some info on the technical aspect of getting this to work that might be appreciated by slashdotters.

You can read about the Blue Waters hardware profile here. Our simulation "only" utilized 20,000 of the approximately 700,000 processing cores on the machine. Blue Waters, like all major supercomputers, runs a Linux kernel tuned for HPC.

The cloud model, CM1, is a hybrid MPI/OpenMP model. Blue Waters has 16 cores (or 32 depending on how you look at it) per node. We have 16 MPI processes going and each MPI rank can access two OpenMP threads. Our decomposition is nothing special, and it works well enough at the scales we are running at.

The simulation produced on the order of 100 TB of raw data. It is easy to produce a lot of data with these simulations - data is saved as 3D floating point arrays and only compresses roughly 2:1 in aggregate form (some types of data compress better than others). I/O is a significant bottleneck for these types of simulations when you save data very frequently, which is necessary for these detailed simulations, and I've spent years working on getting I/O to work sufficiently well so that this kind of simulation and visualization was possible.

The CM1 model is written in Fortran 90/95. The code I wrote to get all the I/O and visualization stuff to work is a combination of C, C++, and Python. The model's raw output format is HDF5, and files are scattered about in a logical way, and I've written a set of tools to interface with the data in a way that greatly simplifies things through an API that accesses the data at a low level but does not require the user to do anything but request data bounded by Cartesian coordinates.

I would have to say the biggest challenge wasn't technical (and the technical challenges are significant), but was physical: Getting a storm to produce one of these types of tornadoes. They are very rare in nature, and this behavior is mirrored in the numerical world. We hope to model more of these so we can draw more general conclusions; a single simulation is compelling, but with sensitivity studies etc. you can really start to do some neat things.

We are now working on publishing the work, which seems to have "passed the sniff test" at the Severe Local Storms conference. It's exciting, and we look forward to really teasing apart some of these interesting processes that show up in the visualizations.