Tip:
Highlight text to annotate it
X
On May 31, 2013 an exceptionally large and powerful tornado formed near El Reno, Oklahoma.
Up to 2.6 miles in diameter, this tornado produced winds near 300 mph, ranking with
some of the most intense tornadoes in history.
This tornado exhibited an unusual penchant for changing both speed and direction: Forward speeds ranged from nearly stationary
to over 50 mph while the direction of movement spanned over 360 degrees, looping over Interstate
40. This made safe observing at close range almost impossible.
The high-precipitation character of the parent thunderstorm made viewing very difficult for
storm spotters and chasers. All of these factors combined to produce an incredibly dangerous
situation in which many storm observers were forced to flee for their lives. Unfortunately,
not everyone made it out.
Among the eight victims this storm claimed were well-known storm chasers Tim Samaras,
Paul Samaras, and Carl Young - as well as at least one other storm observer.
In view of this tragedy, we deem it important to document the lessons that this storm has
pressed upon the chasing and spotting community. We hope that this documentation will reduce
the risk of another such tragedy.
Friday, May 31st
Severe weather has been in the forecast
for days. But where? And would a tornado threat exist?
It is early morning. Forecasters are convinced that a potent combination of atmospheric ingredients
will set up over central Oklahoma. Tornadoes look likely - in an area that has already
seen more than its fair share of wicked weather.
On May 20th, a rare EF-5 tornado ravaged Moore, Oklahoma, a suburb of Oklahoma City. This
tornado took the lives of over 20 people, and left hundreds injured.
Similar to the 20th, a large, slow-moving, upper-tropospheric trough is positioned over
the Plains. The southern periphery of this trough features
an intense mid-upper tropospheric jet, which has initiated strong southerly low-level flow.
Very rich moisture is moving toward the Plains. In the middle troposphere, strong westerly
winds are transporting a cool, dry layer of air over the warm, moist layer to the east.
The overlap of these two air masses is creating a zone of conditionally unstable air over
the Southern Plains and the Upper Midwest.
Conceptually, conditional instability is similar
to the hot air balloon. Hot air is less dense than cold air at the same pressure level,
so the balloon rises. This force is known as buoyancy. Gravity acts in the opposite
direction. If buoyancy is greater than gravity, an upward force is created
Instability can be measured from weather balloon data. These balloons rise through the troposphere,
creating a vertical profile of temperature, humidity and winds.
During the early evening, a weather balloon is launched from Norman, Oklahoma, revealing
a vertical profile of the troposphere. The environmental temperature with height is in
red. The parcel temperature, analogous to the hot air balloon, is the dotted line. Notice
that the parcel temperature is much greater than the environmental temperature. This means
that air originating from the ground will rise with great force once the environmental
temperature is cool enough. The total difference in the two temperature traces is known as
the Convective Available Potential Energy, or CAPE. The area between the two temperature
traces determines the value of CAPE. All of this energy is contained by a warm,
dry layer of air known as the "the cap". This layer of air originates in the Desert
Southwest. It prevents thunderstorm development through most of the day, generally until peak
heating. When it breaks, it breaks explosively - similar to the effect of removing a lid
from a boiling pot of water.
The combination of rich, Gulf moisture with cool, dry air above has led to the development
of extreme instability. CAPE values reach as high as 6000 j/kg in SW OK.
A surface low pressure has formed over southwestern Oklahoma, backing the flow in central Oklahoma
from south to southeast. In addition to strengthening convergence in southwestern Oklahoma that
may lead to the initiation of thunderstorms, this is also increasing the turning with height
in the troposphere, leading to stronger vertical wind shear.
In the middle troposphere, winds are from the west and southwest - generally above
40 kts over the Southern Plains.
Combined with the backed surface flow, this has led to the development of deep layer shear
more than 40 kts. Given the extreme instability, conditions have become more than sufficient
for rotating thunderstorms. To illustrate, we turn to the paddle wheel.
If water at the top a paddle wheel is flowing
quickly, and water at the bottom is flowing slowly, the paddle wheel will tend to rotate.
This effect also occurs in the atmosphere, since air is a fluid. Here is an image from
a developing thunderstorm on a day when tornadoes occurred nearby. The winds at the top of this
towering cumulus are much stronger than at the bottom. This causes the developing storm
to lean downshear. Unseen by the eye are numerous horizontal circulations.
The formation of supercell thunderstorms - from which tornadoes develop - begins with
these horizontal circulations, created by differences in wind speed and direction with
height.
When a developing thunderstorm updraft encounters a circulation, the circulation is tilted upward
by the updraft into the shape of a horseshoe. This creates circulations spinning opposite
directions. The northern circulation spins clockwise, and the southern spins counterclockwise.
The wind pattern often favors strengthening of the southern circulation, while the northern
circulation tends to dissipate. The southern updraft continues to strengthen as an adjacent
downdraft forms -created (in part) by precipitation. This precipitation is wrapped around the updraft,
creating the characteristic hook echo. A supercell is born.
The stage is set for intense supercells. A stationary boundary - generated by previous
convection is positioned over Interstate 40 in central Oklahoma. It is adding fuel to
the proverbial fire: it is simultaneously enhancing the low-level shear while creating
a focus for convective initiation at its intersection with the dryline.
Visible satellite imagery shows partly cloudy skies in central Oklahoma, warming the air
near the stationary boundary, providing power for explosive updrafts.
It is 4 p.m. The atmosphere is primed for the development of intense thunderstorms.
Radar shows a number of attempts at thunderstorm initiation west of Oklahoma City, but as of
yet, nothing has developed.
However, on the ground, cumulus clouds are
seen towering high just west of Oklahoma City, indicating that thunderstorms will initiate
soon.
Storm chasers are beginning to converge in the small town of El Reno, as
the location of key boundaries strongly hints that it will be ground zero.
Around 4:30, the capping inversion - the layer of warm air that often suppresses storm
development in the Plains - breaks in a northeast to southwest line about 50 miles west of Oklahoma
City.
Several storms rapidly form, reaching heights
well above 50,000 feet. Still, considerable uncertainty remains concerning which storm,
or storms, will dominate.
This storm, approximately 15 miles west-southwest of Calumet, quickly develops mid-level rotation.
Radial velocity shows a weak - but classic - rotational signature, about 10 miles west
of El Reno.
At the ground, storm observers note a rapidly rotating wall cloud just south of the interstate.
The National Weather Service issues a tornado warning. Unconfirmed reports of a large tornado
are hitting the TV airways.
Just as this storm is approaching maturity, another storm develops to its south, merges
and begins to rotate.
Shortly after 6 p.m., a vigorous circulation develops rapidly 10 miles southwest of El
Reno.
From the ground this circulation is seen as another large, rapidly rotating wall cloud.
Within two minutes of its development, a ground circulation develops.
The El Reno tornado has begun.
The tornado immediately shows strong multiple-vortex structure.
The Tempest Tour group is located southeast of the tornado - a traditionally safe spot
- if a tornado is moving northeast.
The tornado does not initially appear to deviate from the expected motion - to the east. It
also doesnÃt appear very close: an illusion created by the low-hanging cloud base. Subvortices
appear as well. But instead of moving left to right, they appear to move closer to
the photographer. The tour leader begins to sense danger, and signals his tour members
to come back to the vehicle. The tornado does not appear very large, but the entire circulation
is almost a half-mile across. The tourists frantically make their way back to the vehicle
as the tornado rapidly closes on their location. The tourists enter their vehicle, and are
barely able to escape - the widening tornado can be seen from a rear-view camera.
It is about 6:06 p.m. The Tempest Tour group is making their escape just south of the tornado.
Meanwhile, Brandon Sullivan and Brett Wright are stopped along Chiles Road, just south
of Reno Street, filming the developing tornado to their west.
The tornado appears to be a half-mile away, which should leave enough time for a safe
escape. What they don't realize is that the tornado has expanded to 3/4 of a mile wide,
and the forward speed has increased to 35 mph. The edge of the tornadic wind field is
close - and closing. Realizing the danger, Sullivan and Wright begin to pack up their
gear.
Sullivan and Wright head south. They realize the tornado is much closer and they make a
frantic escape. They are overtaken by the outer edge of the tornado. They survive without
injury. Meanwhile, Dave Demko and Heidi Farrar are observing the tornado a few miles to the
north.
Demko and Farrar are in the notch - the space in between the large hail to the north and
the tornado to the south. The storm is high precipitation in character, and so it is difficult
to see the tornado - especially from the north. Not knowing the exact size and movement
of the tornado, they begin to worry about where to go. They agree to bail west.
While Demko and Farrar are observing the widening tornado from the north, Skip Talbot and Jenn
Brindley are observing the approaching tornado from the east.
The time is 6:13 p.m. From their location, the tornado is difficult to see because of
the rain. Talbot notices very fast-moving rain curtains - very close - and decides
it's time to bail east.
From Talbot and Brindley's location, the condensation funnel is faintly visible
The visible funnel is approximately 0.3 miles wide. The tornado width - based on mobile
radar data -
is about 1.6 miles in diameter, noted by the red circle. Thus, the area occupied by the
full tornadic windfield is more than 10 times larger than its condensation funnel.
In this modified version of Talbot's video, you can see how large the tornadic windfield
is compared to the condensation funnel.
Now realizing the imminent threat, Talbot and Brindley retreat east on 15th Street,
the tornado keeping pace at a blistering 45 mph.
While they are fleeing, Ray Bohac and his crew are following the tornado on Reno, just
a few miles northwest of Talbot and Brindley.
They watch as the large tornado intensifies in front of them. It is 6:15. The tornado
is difficult to see, enshrouded by rain and debris. Suddenly, two more tornadoes appear:
satellites. These tornadoes are 1/4 to 1/2 mile from the edge of the main tornado. These tornadoes
spin about the main funnel in a counter-clockwise orbit.
It is 6:18. A Weather Channel crew, led by Mike Bettes, is racing south down HW 81. Meanwhile,
Richard Henderson is trying to beat the tornado to the east, his progress delayed by a chaser
traffic jam.
Near the intersection of 15th Street with Highway 81, the tornado appears as a wall
of condensation.
Mikey Gribble shoots video to the west along 15th Street as the Weather Channel crew is
frantically attempting to get past the tornado a mile to his west. They do not make it.
While the Weather Channel is attempting to outrace the tornado, Richard Henderson continues
east on Reno, hoping to beat the tornado to the east. Hindered by the blinding rain and
powerful winds, Henderson stops. With brutal force, the tornado overtakes him. He does
not survive.
The lead car of the Weather Channel crew is hit by a sub-vortex within the larger tornado,
and their vehicle is rolled almost 200 yards. The car is badly damaged, and all airbags
are deployed. Amazingly, all of the team's crew members have survived with non-serious
injuries.
As the Weather Channel crew is getting hit, Dan Robinson is heading east on Reuter Road,
just west of Highway 81. He briefly considers heading south, but realizes there isn't enough
time. The tornado is moving fast - much faster than expected. Robinson decides to continue
heading east on Reuter.
Unknown to Robinson at the time, a white Chevy Cobalt is following him. In it are Tim Samaras,
Paul Samaras, and Carl Young.
As Robinson crosses HW 81, he approaches Alfadale Road and realizes that something is wrong
- terribly wrong.
The time is 6:20. Rain curtains - associated with the tornadic circulation - begin to
envelop his vehicle. The winds become much stronger, too
Though the main condensation funnel is still a mile away, Robinson is overtaken by
the invisible edge of the tornado. Robinson's 4-cylinder vehicle struggles
to go eastward against strong
east and northeasterly winds. The wind
slows down Robinson's attempt to escape the approaching tornado.
Visibility is very low, as dust and precipitation bands periodically hide the road in front
of him. The tornado is just to Robinson's south. To make matters worse, the traction
control on his Toyota Yaris engages,
reducing power to his wheels. Robinson struggles to stay on the road. Not long after passing
Radio Road, Robinson clears the future path of the core flow - with winds now approaching
300 mph.
Less than a half mile behind Robinson, a powerful subvortex is swinging northwest toward Reuter.
In the direct path, the TWISTEX crew is riding out the storm alongside the road near a creek.
The vortex briefly stalls over their Chevy Cobalt. It tumbles over 5 times. They do not
survive.
The tornado continues to head north. The forward speed of the tornado slows down to less than
10 mph. The tornado becomes increasingly wrapped in rain.
Meanwhile, at 6:41, Skip Talbot spots another tornado approximately 5 miles to the southeast
of the main tornado.
This tornado, though, is spinning clockwise, or anticyclonically. Though only a couple
hundred yards in diameter this tornado is powerful, with peak winds approaching 150
mph.
Finally, at 6:43 pm, the tornado dissipates near the intersection of I-40 and Banner Road.
Several lessons have re-emerged from this tragic event. It is our hope that these will
lower the probability of another chasing or spotting tragedy.
First, tornado motion is always unpredictable - even for big tornadoes. They don't move
in straight lines or at constant speed. And it's often difficult to tell where a tornado
is moving. The El Reno tornado changed directions over
360 degrees. Thus, if you were close, there was no safe spot, regardless of what direction
the tornado had been moving. Additionally, the range of speeds in the El
Reno tornado was enormous - from nearly stationary to over 55 mph. Near Highway 81, the tornado
doubled its speed - from 25 to 50 mph in 5 minutes.
If you can't see the tornado - as was the case with Dan Robinson and the TWISTEX chase
team north of the tornado - you may be in mortal danger. Sudden turns can and do happen.
Tornadoes can expand rapidly. From 6:05 to 6:10 - when Brandon Sullivan and Brett
Wright stopped to shoot video along Chiles Road - the width of the tornado increased
from 6/10 of a mile to 1.2 miles wide! Making a close approach to a tornado can be very
dangerous - and potentially deadly.
A tornado is often larger than its condensation
funnel - in some cases, much larger. Skip Talbot's view of the tornado at 6:13 demonstrates
this quite well. The tornado appeared to be 1/3 of a mile wide, but the tornadic wind
field was well over a mile wide. In the case of the TWISTEX group, it is quite
likely that they thought they had more time to escape the tornado than they actually had,
since the outer edges of the tornado were not visible. However, the easterly winds inside
the uncondensed tornadic circulation were powerful enough to hinder their escape on
Reuter, resulting in tragedy.
When big tornadoes occur, they are often accompanied
by other tornadoes. These additional tornadoes present a big problem for those trying to
observe storms safely. The first type is the satellite tornado. At
6:15 p.m., multiple satellite tornadoes were observed on the west and south side of the
El Reno tornado. These tornadoes generally occur within a mile of the main tornado in
any direction. Close observers are particularly vulnerable to this type of tornado. The second
type is the anticyclonic tornado. This type of tornado spins in the opposite direction
of the main tornado. While the El Reno tornado was wrapped in rain near I-40, a powerful
anticyclonic tornado - with winds up to 150 mph - developed to the southeast of the main
tornado. Typically, these tornadoes form to the right of the hook echo, a fair distance
away from the main cyclonic tornado. The final type of danger comes from new tornadoes forming
in new circulations within the parent thunderstorm. They generally form downstream of the existing
tornado.
The "notch" of a high precipitation supercell
is extremely dangerous.
It is why "core punching" -
approaching the tornado from the rain and hail - is so perilous.
This is the area of the storm immediately to the left of the tornado, and just to the
right of the large hail.
Those in the notch are in danger of the sharp left turns that tornadoes often make when
they are dissipating.
If the tornado can be seen, successful escapes can be made. However, heavy rain may hide
the tornado. In that case, radar updates may be the only way of knowing where the tornado
is located. But in the case of the El Reno storm, the
tornado moved 2 miles north in less than 5 minutes - less than the interval of a WSR-88D
volume scan. Additionally, the radar cannot pinpoint the exact location of the tornado
with certainty.
So you should not depend on radar to know where the tornado is.
And even if the position of a tornado is known, strong inflow winds may hinder a quick escape.
This almost certainly was the case for the TWISTEX team on Reuter Road.
As mentioned previously, new tornadoes are always a danger in the notch
And, of course, there's the lesser threat of very large, glass-breaking hail in the
core of the storm.
In the path of an approaching tornado, a quick
escape may not be possible. A lack of good road options, poor road conditions, or even
traffic may hinder a safe escape. And in the case of the El Reno tornado, numerous traffic
jams were reported. It appears that these traffic jams may have resulted in the deaths
of at least 3 people in 3 different cars.
Based on these lessons, we suggest that storm spotters and chasers place more distance between
themselves and the tornado - especially on days when parameters are particularly volatile.
When the instability and shear combination is high, the storm evolution may occur more
quickly, decreasing the margin for error. Moreover, given the rarity of the ingredients
that produced the El Reno tornado, storm behavior may differ greatly from more "normal" tornado
days. For example, on June 8th, 1995, a very large tornado accelerated to nearly 60 mph
near Allison, TX, before slowing down to nearly stationary.
You may have seen videos of people who escaped
death and serious injury when their vehicles were hit by the El Reno tornado. But it's
critical to remember that in most of those cases, the vehicles in most of were not impacted
by the strongest winds in the tornado. It is possible that this has resulted in a false
sense of security within the storm observing community.
The most powerful winds in a tornado are located
in sub-vortices, which are smaller tornadoes within the larger circulation. These are the
vortices responsible for leveling one house, but leaving a house next door unscathed. Given
the small area they occupy, the probability of being hit is actually rather low. However,
as the number of close encounters increase, the odds increase that more chasers will encounter
these deadly winds. This is especially true now given the growing trend of "extreme"
chasing.
Remember that no footage, report, or data
is worth your life. Of the 8 deaths in the tornado, at least 4 were chasers. The number
of chasers that were killed, injured, or narrowly escaped was far larger than any other documented
tornado. There will always be more storms.