How to track a tornado - Karen Kosiba

View full lesson: http://ed.ted.com/lessons/how-to-track-a-tornado-karen-kosiba

Atmospheric scientist Karen Kosiba studies how tornadoes form and do damage. Getting measurements near the surface of these twisters is difficult, though, and driving into them is a practice mostly reserved for the big screen. In this TEDYouth Talk, Kosiba describes how she and her team use observations and modeling to track these super storms, while sharing some incredible footage from the field.

Talk by Karen Kosiba.

So, I think all good
tornado talks need to start

with an awesome tornado shot.

And this is not that awesome tornado shot.

That was the first tornado I ever saw,
it was really cool, really scary,

and I'm showing it to you guys

because that's why I got
into the field in the first place.

So even though it's a bad photograph,

it was really cool
to be out there the first time.

But now I'm taking real tornado footage.

Fast forward a few years.

This is a few years ago,
during a field project called VORTEX2,

where myself and a bunch
of other scientists were out there,

surrounding tornadoes
with different types of instrumentation

and trying to figure out
how tornadoes form.

It's a big question
we're trying to answer.

It sounds like a very basic one,

but it's something
we're still trying to figure out.

We're also still trying to figure out
what the winds are like near the surface.

We know what the winds are like
above building level,

but we really don't know
what they're like at the surface

and how that relates
to what we're seeing above building level.

Most tornadoes form from what we call
supercell thunderstorms.

Supercell thunderstorms
are what you commonly think of

as tornado-raising storms.

They're big, rotating thunderstorms

that happen a lot of times
in the midsection of the United States.

But the problem is that even though
they're rotating up above,

it doesn't mean they're rotating
at the surface.

And when we look at these storms

and at these pictures
and at the data we have,

they all kind of look the same.

And it's really problematic

if we're trying to make
tornado forecasts or warnings,

because we only want to warn
or forecast about the storms

that are going to actually make a tornado.

One of the big, critical distinguishing
features, we think, between these storms,

is something about
the rear-flank downdraft.

So these big rotating thunderstorms
have this downdraft

that wraps around the rear edge of it,

hence the "rear-flank" downdraft.

But we think how warm that is,
how buoyant that air is,

and then also how strong
the updraft it's wrapping into,

makes a big difference on whether or not
it's going to make a tornado.

There's a lot more that goes into it --

I'll tell you about that in a second.

Once you actually get a tornado,
again, the problem that we have

is getting measurements near the surface.

It's really hard to get measurements
near the surface --

most people don't want
to drive into tornadoes.

There are a few exceptions;
you might have seen them on TV shows.

But most people don't want to do that.

Even getting instrumentation in the path
of the tornado is pretty tricky, too.

Because, again, you don't want
to be that close to a tornado

because sometimes the winds
around the tornado are strong as well.

So getting information,
that critical location,

is key for us because,
again, we don't know

if the winds that we're seeing
above ground level,

way above building level,

actually map to the surface,

if they're stronger, weaker,
or about the same

as what we're seeing above buildings.

The way we get at answering
a lot of these questions --

and I'm an observationalist;
I love to get out in the field,

and collect data on tornadoes --

we compile a lot of observations.

I work with this group
who operates mobile radars,

and they're exactly
what they say -- basically, a radar

on the back of a big blue truck,

and we drive up really close
to tornadoes to map out the winds.

We map out the precipitation.

We map out all these
different things that are going on

in order to better understand
the processes in these storms.

And that bottom there,
that's what a tornado looks like

when you're looking at it
with a mobile radar, and really close.

Also, what we do is a lot of modeling,

so we do a lot of computer
models and simulations,

because the atmosphere
is governed by the laws of physics.

So we can model the laws of physics

and see where the tornado might go,

where the storm might go,

how strong the winds are near the surface

and not actually have
to go out in the field.

But of course, we want to have
both observations and modeling

to move forward with the science.

So, I showed you that video earlier
that went real quick, too.

This is what it looks like,
looking at it with a radar.

So you saw it visually,

but this is what I get really excited
about when I see now in the field,

stuff that looks like this.

The really exciting thing
about looking at stuff like this

is that we caught this storm
from when it didn't make a tornado

to when it made a tornado and intensified

and when it dissipated.

This is the one of the rare data sets
that we have out there

that were able to study
the entire life cycle of a tornado.

I talked about how we think
that rear-flank downdraft is important

because it tilts, there's a lot
of spin in the atmosphere,

but the problem with
all this spin in the atmosphere

is it needs to be oriented vertically,

because that's what tornadoes are doing,

and it needs to orientated
vertically near the ground.

So we think this rear-flank
downdraft just pulses.

And these pulses in this
rear-flank downdraft, we think,

are very important
for converging that rotation,

but also getting that rotation
into the right place.

Other things we've learned

is that we have gotten
a bunch of fortuitous measurements

in the path of the tornadoes
and very near the surface.

And we found out

that the winds near the surface
are actually pretty comparable

to what we're seeing 30, 40 meters
above ground level.

So there's not a big reduction
in what we're seeing above the surface

to what we're seeing at house level.

And that was a pretty
surprising finding for us,

because we kind of assumed
that the winds decrease

pretty substantially near the surface.

I'm going to end with this real quick.

And this is not my last
tornado I ever saw,

but I really like this image,

because this was taken with one of those
mobile radars I was talking about.

This is a tornado, not a hurricane,

and this is what it looks like
when you're really close to it.

And I find this amazing,

that we can actually take technology
this close to these types of storms

and see these inner workings.

And for those of you who look
at tornado images often,

you can see there's a lot going on --
there's rain spiraling,

and you can actually see the debris cloud
associated with this tornado.

I look forward to the future
and future technologies

and being able to learn
a lot more about these storms,

as the world advances,

as you guys contribute to the science

and we're able to really learn
more about how tornadoes form.

Thank you.

(Applause)