Cars, tornadoes, and bad decisions: 2014 edition

   A "storm chaser"--I use the term very loosely, as he seems more like a local attempting suicide--came within a football field or two of getting himself killed in the 4/28/14 Tupelo, MS tornado.  This in itself is not good.  What's ten times worse is posting a dramatic video to Youtube and encouraging other people to follow suit:

Let me be the first of pretty much everyone in the meteorological community to say:

• This guy is incredibly lucky and incredibly stupid.
• At no point was he "in the tornado."  If what he experienced looks scary, realize that he was only getting inflow jets and possibly the outermost fringe of the circulation.  Had he been fully inside the tornado, he would probably have encountered winds twice as strong.  He and a crumpled wad of metal sort of resembling his car would have been found out in that field, if not in two different fields. This, or worse, is what happens to vehicles directly impacted by strong tornadoes:

(photo from NSSL)

• Vehicles are still not safe in tornadoes!  El Reno proved that point, when several died on the roads and none died in structures.  The Arkansas tornado that threw cars off I-40 the day before this storm proved that point again.  Drivers will keep becoming examples until we learn this.  Thousands of people have been pulled alive from homes and businesses directly hit by violent tornadoes in the past few years.  I don't know of very many who survived direct hits in cars, except the fortunate TWC team in their huge SUV.

   As hard as it is to say, I don't think this guy did quite everything wrong.  He could have made one worse mistake by driving straight into the tornado.  Stopping and letting it pass always beats racing it to a crossing point (ask Mike Bettes).  The idea is to stop away from the tornado, not as close as possible to certain death.

4/26-27/14 severe weather

   Flow associated with a developing surface low in the High Plains in advance of an upper trough ejecting from the Southwest will result in rapid low-level moisture advection into the Southern and Central Plains on Saturday.  Dewpoints in the low to mid 60s will result in considerable instability (2000-3000 J/kg CAPE), but a stout cap will remain in place through most of the day as the upper-level forcing to weaken it will lag to the west.  Convergence and convective mixing along the dryline will be nearly enough to overcome the cap as stronger flow aloft/differential vorticity advection finally arrives in the evening.  Models differ on timing and location of convective initiation, but the most likely scenario is a storm or two in southwestern Oklahoma or western North Texas just around sunset.  Large low-level hodographs and increasing deep-layer shear will support supercells capable of producing very large hail.  LCLs will initially be too high for an appreciable tornado threat.  However, as the boundary layer cools and moistens with loss of diurnal heating, cloud bases may lower.  Model consensus is that a ~50 kt low-level jet will emerge over northwestern Texas and western/central Oklahoma in this time frame.  Extreme low-level vertical shear combined with any cloud cover that results from distant anvils/synoptic lift/moist advection in the LLJ should prevent rapid stabilization of the boundary layer in the 00-04z period.  Remaining instability up to 2000 J/kg CAPE and large values of SRH owing to the LLJ would mean a risk of a few tornadoes, possibly significant, if mature surface-based convection continues beyond sunset.  The temporal window for tornadoes will be short.  The most likely area for this tornado threat, if it materializes, would be southwestern to west-central Oklahoma.  Given modest storm motion, storms do not appear likely to reach the OKC metro before boundary layer decoupling ends tornado chances for the night.

   Sunday's primary severe weather setup will be to the east of I-35 after the dryline surges through in the morning hours, with southeastern Oklahoma into the Arklatex seeing the highest risk.  Current trends in the NAM suggest against ruling out morning convection in the I-35 corridor, though, and will need to be monitored.

4/23/14 lightning show

   After an awards ceremony for the College of Atmospheric and Geographic Sciences, I headed southwest to just east of Chickasha to catch the one surviving thunderstorm from the evening's low-end severe event out west.  There were quite a few CGs over Chickasha, but activity dropped off a little once I headed back to Norman and met up with a dozen or so met students on the OU parking garage roof.

4/13/14 Oklahoma storms

   This all started as a trip back to Norman from a weekend with my family in Texas, just south of Dallas/Fort Worth.  Wind profiles were highly supportive of severe weather ahead of a dryline across central and southwestern Oklahoma, but a crashing cold front turned out to be the only adequate source of lift given some mixed-layer inhibition in place.  The trip north began in light rain after church services and lunch.  Skies quickly cleared to the north and west.  I stopped briefly in Sanger, TX and south of Thackerville, OK to investigate radar echoes that popped up near the Red River, but the showers appeared both shallow and elevated.  The low-level jet was evident by then in low cumulus fragments shooting north-northwestward.  Meanwhile, the OKC metro was catching the southern end of the QLCS resulting from the cold front overtaking the dryline.  As the QLCS expanded rapidly southward, it became obvious there would be no pure dryline supercells.

   I was about to venture northward on I-35 and aim for Norman through a gap in the line when the southern end of the QLCS began firing off short-lived supercell structures.  At I-35 and U.S. 70 near Ardmore, indecision set in.  I wanted the "tail-end Charlie," which would have been best intercepted along U.S. 77/I-35 to my north, but expected another cell to take over farther south on the front, making a westward play necessary.  After a couple false starts, I hedged my bet and scooted north on I-35 to head west on Highway 53.  Flying west through rolling hills in the middle of nowhere, I came up to the first sight of interest--a rainy lowering far to the northwest, close to where a brief tornado had been sighted earlier, with a little shelf cloud to its south.  At that distance in that terrain, there was no way to see what was happening near the ground.

   At first the plan was to get west to Highway 76 near Healdton and then position north or south, but the storms had other ideas, barreling southeastward and cutting off that route.  I met the gust front along Highway 53 northeast of Healdton.

   I stayed a little too long.  With a solid wall of precip maybe a mile out and closing, and winds whipping dust off the backroads, I got a new radar update and saw quite a bit of low-level shear on top of my location.  Visually, nothing was too threatening, but a tornado had been confirmed with the next cell up the line and I didn't want to find out what might be in the rain.  I zigzagged generally southward on backroads to Lone Grove.  (Thank you, AT&T, for somehow getting me maps and radar out there.)  This completed a loop, back to the earlier decision point near Ardmore.  The new tail-end cell exhibited very strong mid-level rotation, with low levels invisible so far from the NWS radars, and was tornado-warned.  Ultimately, if I'd stuck with my original thought of going west on U.S. 70 toward the newer cell an hour or more prior, I'd have missed the marginal structure shots on Highway 53 but probably could've had better down south near Ringling.  Lesson learned.

   As it was, the storm was all over U.S. 70 near Wilson at that time.  The number of chasers and locals alike driving into the teeth of a tornado-warned HP supercell on NWS Norman's first tornado day since El Reno was disappointing.  There was a feature of interest partially wrapped in rain, but it looked disorganized and was just a ragged gap between precip cores by the time I pulled over for a picture.

   Turning around, I planned to return to I-35 and go south to intercept the southeastward-moving storm near Marietta.  The cell to the north was passing close to Ardmore with a bit of rotation, but south was the better and easier play.  Rolling south on I-35, I began to see the huge bell shape of what I was chasing.  The structure was spectacular--so much so that many chasers, including a crew from KOCO, were stopping along I-35.  As a rule, I won't stop on a limited-access road unless I'm about to die or the tornado of the century is scouring a field by the highway.  Pointing the camera in the storm's general direction while still watching the road for chasers weaving on and off, I did manage to get some of the structure.

   I jumped off I-35 at Overbrook and cut over to U.S. 77 so that I'd be able to pull over and stop.  What looked like a funnel, in a similar shape to the tornado that had touched down earlier, appeared on the north side of the heaviest precip.  By the time I got clear of trees and stopped on the shoulder, it had all but vanished.  What was left of it is above and just to the right of the bare tree in this view:

In this photo from a few seconds later, contortions in the edge of the precip might be a small gustnado:

As everything on the other side of I-35 suddenly disappeared into rain and hail...

...I bailed southward to Marietta.  Though any possible circulation was embedded deep in heavy precip, the storm was still tornado-warned and sirens were wailing in town.  Daylight was nearly gone, and road options and suitable terrain both looked hard to come by east of Marietta.  I stopped to watch the core blast through.

Finding a gap in the line behind that cell, I squeezed through and finally finished the trip to Norman.