[Please note: this is a special episode of Earth, and runs longer than the usual length. Below is the original script. But a few changes may have been made during the recording of this audio podcast.]

Mother nature is known for her irony.  And last week’s tornadoes that ripped through the American heartland are a case in point.

Just days after the VORTEX2 scientists, part of largest study of tornadoes in history, packed up their instruments after only snagging one single twister—the Midwest got hit with more than 75 tornadoes, across seven states.

The National Oceanic & Atmospheric Administration reports 873 twisters this year, while last year had nearly double, at 1691.  Not great news for scientists wanting to study twisters this summer. But there’s always next year.

VORTEX2 received nearly $12 million in funding from the National Science Foundation, NOAA and various universities and organizations, to go out in the field for two seasons.

Nearly 100 researchers literally surrounded tornadic storms with mobile radar and wind, humidity, pressure and temperature instruments, in an attempt to answer a deceptively simple question:  how to accurately predict when a tornado will form.

“The predictability is still an unknown,” that’s Don Burgess, retired chief of the Warning Research and Development Division NOAA, and now at the Cooperative Institute for Mesoscale Meteorological Studies at the University of Oklahoma. “Whether the processes are so complicated and such small scales dominate—whether we can really forecast them in the future an hour or more is something about which right now we don’t know.”

The seemingly unlimited changes in variables like wind direction, temperature, or moisture, and having such few sample cases (they only have one sample from VORTEX2), make it a near impossible challenge to precisely nail down why one storm produces a tornado and another very similar storm does not.

“But, the way that I look at it is that this research project, and I think the same analogy is true in a lot of research, it’s like an onion.  And that onion has lots of layers. In VORTEX1 we peeled back some of those layers. Since then we’ve learned some more, we have better technology we maybe peeled back another layer or two.  Now we’re looking at that onion and I know in VORTEX2 we’re going to peel back more layers but do we get to the core or not—my guess, based on being around this for forty years—is we’re not going to get to the core.  We’re going to push back the horizons, we’re going to better and we’re going to get more understanding, we’re going to get more predictability, we’re going to get better applications for warnings to the public. But are we going to solve it soon?  No.”

—Christie Nicholson