September 16 2006 southeastern SD chase



April in September is a rare treat!  My shift ended at 7am and I departed with a target of Madison, SD.  Began analyzing data by 1pm  from the intersection of I-29 and I-90

in north Sioux Falls.  A prefrontal windshift was nearly stationary over the James River valley.  If this was to serve as the westward bound of tornadic potential--or as

the initiating boundary altogether--I didn't dare go any farther west.  Perhaps north instead.  However, to my delight, the 2pm surface obs showed a rapid response

to the the incoming, unseasonably strong shortwave trough... with the surface boundary basically washing out as Mitchell, Chamberlain, and Winner all backed to the

southeast once again.  Convective initiation occurred soon after to my west-southwest, over Hutchinson county.  I headed west on I-90.  If storms could remain isolated,

a threat for significant tornadoes was certain due to favorable LCL heights and the evolving 1km SRH environment.






I reached the Salem exit on I-90 at 3:40pm and prepared for an intercept.

A large wall cloud developed.  A period of disorganization followed, as the entire updraft lowered its base.  A tornado warning was issued by NWS FSD, and they upped the storm motion to 45kts.  With the 70-80kt mid-level speed max nosing into the area, it was possible the storm could accelerate even further.  I took a chance and drove away from the storm... eastward, to the next I-90 exit. 
Here is the first image I took from the next exit (4:02pm).  The RFB was organizing nicely and sloping in toward the precip.  I plotted the future track of the storm on a map and prepared to potentially play the when-will-the-warp-speed-storm-drop-a-tornado-and-will-I-be-in-position-to-see-it? game.  (A game I usually lose).
Significant cloud-base rotation was underway, with rotating rain curtains near the center of this photo.
At 4:08pm, a nub funnel peeked out from the center of the dynamic merry-go-round meso.
During the next minute, the "Salem" tornado was born... with a single wild condensation vortex having swirled across the ground seconds before I snapped this photo.  The rainy RFD cleared locally, offering strong contrast.
Tornado transitioned back to a truncated cone within pronounced low-level meso.
Condensation reached for the earth.  The tornado kicked up a large amount of dust/debris over the next minute.
The tornado veered leftward on a more easterly course for several moments... then turned back to the northeast.
Tornado grew into a stovepipe.
Image of entire mesocyclone structure, framed by some local flora and the photographer's right foot.
Tornado approaches and rips through several stands of cottonwood trees.
Dramatic helical motions near ground.
The debris fan reminded me a lot of SLT's imagery of the Webb IA tornado.
At 4:16pm I snapped my final photograph of the Salem tornado, and abandoned my location for I-90 eastbound.  From the onramp, I filmed the tornado as it began to cross the area in which I'd previously been standing.  I then gathered some fairly remarkable footage as the strengthening tornado spread its debris fan across the interstate.  The tornado passed into a field, growing into a stout stovepipe while clear-slotting colored it ghostly white.  Debris began falling on my vehicle.  The tornado continued northeastward, shrunk, bended into a 90-degree angle at its midsection, and roped out ~4:22pm a couple miles north of the interestate.
Post-tornadic large & wet RFD.
I stairstepped northeastward toward the town of Colton, SD and in time noted significant rotating/cascading motions.  I stopped and observed these for a moment, beginning to think they were the sort of "hind-end" storm features I often stare at in awe while the primary meso is recycling/scooting away well ahead of me.  I continued northward in an effort to retain/improve my position on the storm, when tornadogenesis did in fact occur with those features.  The tornado began at 4:54pm as a narrow gray tube, then began to lean over as a dust cloud developed.  A bright shower of debris followed; a farmstead had been struck.  The tornado lasted two minutes.
I fought to keep up with the storm through Pipestone, MN.  The updraft looked fairly sheared-over and disorganized at cloud base.  Newly tornado-warned cells had developed back in southeastern SD.  I moved back to the southwest side of town for an intercept along the MN border
This was the tail-end tornado-warned cell as it approached Dell Rapids, SD.  It first sported a blocky wall cloud but quickly transitioned to a more highly-based P.O.S.  The warning expired.
I headed east on the same highway with which I'd pursued the previous storm.  Winds in the inflow sector were dead calm.  About this time, it earned a severe thunderstorm warning again.
I bid the storm and its cute horseshoe RFD farewell ~7:15pm.
Departing convection viewed from I-90 at sundown.  The ride home was pleasant until I encountered a wall of water south of Omaha.  A five-hour drive took nearly eight as a result.



Maps, weather data


observed supercells


Storm survey map showing long-track Salem F2 and Colton F1 tornadoes (NWSFSD)


Detailed survey map of Salem F2 tornado (NWSFSD)


Base reflectivity & storm-relative velocity one minute before Salem tornado touched down (NWSFSD)


representative hodograph for Colton SD tornado via Wood Lake profiler

storm motion: 237 deg @ 38 kts

0-1 km SRH: 231 m2/s2

0-3 km SRH: 241 m2/s2

0-1 km bulk shear: 22 kts

0-6 km bulk shear: 63 kts

7-10 km SR flow: 45 kts



Supercell train!  3 successive supercells take identical paths thru northeastern Minnehaha/southeastern Lake/Moody counties

2203z: tornado-warned supercell 1 (Colton tornado occurred ~10 minutes prior)
2300z: tornado-warned supercell 2
2356z: weakening supercell (had been tornado-warned earlier over McCook co.).  note this cell and previous cell are both undercut by windshift/front
ABR STP showing storm tracks (FSD radar was in and out during the event)