The Tornadoes of March 2, 2012

From Birth to final hours the cells travelled from Missouri into West Virginia. Individual
cells could be identified on radar for more than 550 miles. Times are USA EST (UTC -5 hrs).

The Crittenden-Piner-Morning View EF4 tornado just after touchdown before it struck Piner, KY. Taken from south of Crittenden.

The chaos that ensues with the touchdown of a major tornado or three of them in the case of March 2, 2012 is a heartbreaking testament to the power of Nature. Chaos pretty much says it all, nearly total disorder and the end of daily routines until lives and homes can be rebuilt.

That chaos ripples down to the weather office. Unlike the victims of the storms, for us of course, it is a matter of a minor inconvenience and catching up with tasks put off for storm coverage. This is one of  those tasks.

EF3, 4 and 5 tornadoes are classified as major tornadoes. There was one EF4, two EF3s in the Cincinnati tristate region, 2 - EF2s, 2 - EF1s and 5 - EF0s for a total of 12.

High resolution Doppler radars like the TDWR (Terminal Doppler Weather Radar) often incorrectly called  - Turbulence Detecting Doppler Radar - were located near major airports to detect wind shear, wind shifts and turbulence so aircraft can avoid the dangers associated with these  on takeoff and landing. There is a TDWR in Kenton County south of CVG.

Here is what TCVG,  the Terminal  Doppler Weather Radar near CVG in Kenton County, KY at 4:21:26:57 PM EST (21:26:57 UTC) Friday 2 March 2012.

The well defined hook echo is attenuated by heavy rain at the radar site (blank circle north of the hook). A debris ball is visible near the end of the hook.  Just what is a debris ball? It is parts of houses, trees, peoples memories and prized possesions abducted and pirated away by the 160 mph winds of the EF4 Crittenden-Piner-Fiskburg-Morning View tornado.

Tornadoes With Hook Echoes

Take a look at the 6 radar images below. They are from the EF4 and EF3 tornadoes that struck the  FOX19 viewing area 2 March 2012. The top image is from the Crittenden-Piner-Fiskburg-Morning View tornado, the second is from the Peach Grove-Moscow-Hamersville tornado and the third through sixth are from the Holton, IN tornado.

The supercell that casued the Holton, IN EF3 tornado has no obvious hook at the lowest elevation angle scan on the KIND NEXRAD.

One scan elevation up the hook is visible but not pronounced. The tornado looks to have formed as an RFD inflow jet formed (notice the notch in reflectivity at the tip of the arrow) and began to invade the mesocyclone causing the break in reflectivity southwest of the tornado icon.

The base velocity (slightly wider view than above) shows the inflow jet as a darker shade of pale red flowing into the mesocyclone. Velocities near Westport are typically 20-25 kts. As the air converges and accelerates into the mesocyclone veloities are typically 40 - 50 kts near the tip of the arrow.

GRLevel 2 Analyst display of NROT (normalized rotation at the same scale as the image above). The role of the RFD Inflow Jet is not completely understood but it appears to be an important factor in intensifying the spin of a tornado cyclone by increasing the available air than can be evacuated up through the funnel and by "squeezing" the funnel into a smaller radius thus increasing the rate of spin.

The lesson is that here, 1 EF4 and 2 EF3s touched down. The Holton, IN tornado lacked a well defined hook echo at the lowest elevation scan but at the next higher antenna elevation (.9°) the hook is visible. Using normalized rotation the tornado is evident.

Tornadoes Without Hook Echoes

Now take a look at the Berlin, KY EF0 radar images. The red square dot northwest of Brooksville is the location of the Berlin, KY tornado.

Radar reflectivity of  flanking line thunderstorms entering and moving through northwest Bracken Co. KY at

5:05:15 PM (22:05:15z) March 2, 2012. the black dot west of Brooksville is Berlin, KY and the approximate location

of the brief touchdown of an EF0 tornado. 

This is a leading edge vortex sometimes called a gustnado due to its location along the gust front. These are mostly EF0 and EF1 funnels and many meteorologists feel tornado warnings should NOT be issued for these small, insignificant, short-lived funnels.  

Since 1950, the beginning of the Storm Prediction Center tornado database no one has been killed by an EF0 tornado and only 1 person has been killed by and EF1 tornado in what is now the County Warning Area of NWS Wilmington, OH. This calls into question the wisdom of NWS tornado warning policy which does nothing to distinguish between monster tornadoes and weak, insignificant leading whirlwinds.,

Close up of above, note the "cupping" of the leading edge east of the area of heavy rain in red. The small square dot represents the location of the tornado.

De-aliased velocity display. Because the storms were moving to the northeast at 30 - 40 kts the green everywhere indicates flow towards KILN NEXRAD. The strong lovlevel jet supplies rapidly moving air and induces cyclonic rotation. As it gets lifted over the gustfront and gets pulled into the updraft it is tilted to nearly vertical and is str4etched increasing therate of rotation.

Same as above but with base reflectivity in the background.  The cupping at the leading edge of the moderate to heavy rain (yellow) indicates the circulation creating the leading edge whirlwind.

Normalized rotation from GRLevel2 Analyst.

Close up of above, tornado at small red square dot.

Spectrum width, the range of velocities in a sample volume (specifically the standard deviation of the velocity distribution in a sample volume) is more easily understood as turbulence. 

Close up of the above image. The small square red dot is the location of the tornado. The spectrum width is greatest along the gust front.


The Berlin, KY tornado, an EF0, was a different type of tornado than the EF3s and EF4 which were mesocyclone tornadoes.

Berlin, KY was a leading edge tornado. Leading edge tornadoes are difficult to detect because almost all leading edge vorticies are weak, small and are usually around for only a minute or two, maybe less. They form from air that gains spin outside the thunderstorm. The rotating air is pulled into the thunderstorm and tilted to nearly vertical and can cause a small tornado.

There is mounting evidence from Doppler on Wheels investigations that many large tornadoes may be caused by continual importation of rotation which is tilted to the vertical and stretched by the updraft. Stretching a rotating "tube" of air has the same effect as an ice skater pulling his or her arms towards the body, the rotation is concentrated in a smaller area and the rotation rate increases. When the tube is stretched the diameter shrinks and the rotation rate increases.

Take a look at the images I have produced to explain the importation of rotation:

Development of horizontal rotation from vertical wind shear

Rotation caused by vertical wind shear increases as air streams towards the thunderstorm through the

environment with large vertical shear values.

Wide view of rotation being imported into a thunderstorm. Rotation can be imported over a large area.

An additional source of rotation involves the cold pool that develops under a thunderstorm. As air is lifted up and over the cold pool horizontal rotation is converted to rotation around a vertical axis.

When the rotating tubes of air are tilted to the vertical by the thunderstorm updraft and stretched vertically they can spin up enough to cause a leading edge tornado.

Leading edge tornadoes only live a very short time and the radar signatures are weak so they are much harder to warn for than a mesocyclone tornado.

These elements can also become a significant source of rotation that help maintain large tornadoes as the super cell travels across the landscape.