Merging Tornadoes: Fact or Fiction

Scott Currens

The KC/Pleasant Hill NWS noted a rare occurence. Apparently two tornadoes merged north of Arrow Rock, MO around 9PM. I observed these tornadoes from HW 41 NW of Arrow Rock and collected the following images.

As this tornado passed south of Marshall along US-65, a second tornado formed approximately 3 miles south-southeast of the first tornado. Both were visible from the I-70 and US-65 interchange. Both tornadoes moved northeast and paralleled each other for several miles! A third tornado in Saline County formed briefly as the first two approached the Missouri River. Storm chaser accounts describe this third tornado as short-lived, as it quickly merged with the second tornado (similar to what occurred on March 13, 1990 east of Hesston, KS) and quickly led ot the second tornado becoming over a quarter mile wide and intensifying to F3 strength, while the first tornado finally occluded and dissipated.[/b]

Here is an image of the second tornado as it crossed HW 41.

This image shows tornado #3 infront on the left and tornado #2 on the right behind the other one.

Does any one know of any images that show daytime torndoes merging?

Also, can someone post the base reflectivity and SR velocities from 8:35 to 9:10 PM? I know we have some radar experts on here.

I have more images from these tornadoes here.

Scott Currens
I wonder if this was a satellite vortex or a separate circulation.

There is an old classic video that I've seen on some of the early National Geographic/NOVA specials of a large, leaning stovepipe tornado crossing a road left to right. Just before it crossed, a strong, solid satellite vortex with full condensation formed to the right of the main vortex and merged into it.

The May 3, 1999 Chickasha tornado also had a well-defined satellite that rotated around the big tornado a couple of times before being absorbed into it. Neal Rasmussen has some great video of this.
Some recent observations from mobile Doppler radars as well as high-resolution numerical simulations show that at least some tornadoes may form from the merger of several smaller vorticies. It is not inconceivable that an established tornado may absorb smaller-scale circulations during its lifetime, some of which may be intense enough to be considered tornadoes in their own right.

This of course, is different from a multiple vortex tornado, where the individual vortices are embedded in an owe their existence to (through shear instabilities and other dynamic effects) the main tornadic circulation. Probably many past eyewitness accounts of tornadoes merging (and splitting) were simply examples of people viewing multiple vortex circulations from far away without perceiving the depth of the scene properly, so that tornadoes that appeared to be splitting and merging were actually multiple vortices rotating around each other.

Also, I should point out that satellite tornadoes appear to be formed differently from classic multiple vortices, maybe instead forming on regions of locally enhanced shear well removed from the main tornadic circulation.
During the 6/23/04 Wisconsin Outbreak, two F3 tornadoes merged in western Fond du Lac County at ~845pm. I'm not aware of any photos of this event, but the track map on the MKX website sure as hell makes me happy I don't live to the NW of Waupun.

On one the Tornado Video Classics they have the Hesston/Goessel, KS merge and also video that Gene Rhoden shot of 2 tornadoes merging. Not sure where that was shot, but I believe that was in Kansas as well.
On 3/16/06, I conducted a detailed survey of the damage tracks near Marshall, MO. My main focus was accurately plotting each tornadoes path. Unfortunately, I didn't have time to cross the river to see if the paralleling tornadoes paths came together. The line segments are based off of the NWS survey maps.


Scott Currens