Mini tornado alley in IL?

[rdale]

"I'm having trouble understanding the connection between a river, and storm relative shear."

I'm not from that area but I've always heard it referred to as the Irish Hills Valley, as there is no river that I know of.

In any case there are more than 50 lakes and it is VERY hilly terrain - certainly not 'mountainous' but nothing like the flatlands of Indiana or the gentle rolls of Ohio. I don't think it's the water but the dramatic topographical change.

http://tinyurl.com/3nhck

- Rob

 

[Skip Talbot]

This is a pretty interesting thread, and indeed there are conflicting views even among the experts. At the weather seminar in Plainfield, IL, Tom Skilling referred to the mini tornado alley that is affecting the southwest subrubs, and how it used to be further north, and that is probably in the process of moving once again.

Meanwhile, during an earlier presentation at the seminar, one of the NWS forecasters said that there was no such thing as a mini alley and its all just probability and coincidence.

As far the Hill Valley in Michigan, could there be a higher amount of vegetation here possibly leading to locally higher dewpoints, allowing for more a condensation funnel to drop?

 

[rdale]

"Tom Skilling referred to the mini tornado alley that is affecting the southwest subrubs, and how it used to be further north, and that is probably in the process of moving once again."

I have a hard time believing you can pinpoint the change in location of something like that... If it moved to St Louis that's one thing, but we don't have enough data in the last 50 years to say it's shifted from O'Hare to DuPage.

"As far the Hill Valley in Michigan, could there be a higher amount of vegetation here possibly leading to locally higher dewpoints, allowing for more a condensation funnel to drop?"

No, the higher dewpoints are usually in southcentral/SW Mich either due to being closer to the source region and/or more farming. Since many of our tornadoes do not come from "true" supercells, but from squall lines (excellent article in this month's W&F) it wouldn't surprise me if low-level wind changes from the orography were an influence.

 

[Justin Turcotte]

Two thoughts on southest MI tornado max. There is a large pop density in that portion of the state which would lend to more reports and more stuff for twisters to munch on creating higher F-numbers. Also, stuff crossing the lake from the west is undoubtably effected by cooler surface temps. Then move a ways inland and bang.

 

[nickgrillo]

Since many of our tornadoes do not come from "true" supercells, but from squall lines (excellent article in this month's W&F) it wouldn't surprise me if low-level wind changes from the orography were an influence.

I somewhat disagree. I recall most tornadoes from the past couple years coming from isolated cells (besides the August 25 event...). Notable tornado days in 2004 for example.

May 13th - Unwarned Isolated Cell Produces F0 in Shiawassee Co.
May 14th - Isolated HP's devolope and one drops a F1 with no warning in Saginaw
May 22nd - Widespread supercells (?)
May 23rd - Isolated Cell breaks off main squall and produced F0 in St. Clair Co. and earlier that day, a mini-supercell tracks thru Ingham to Genesse; produced F0
May 31st - Isolated unwarned cell produced F1 in St. Clair Co.
June 13th - Cyclic Supercell tracks throughout Central MI; produced three and another supercell devolops in the Gaylor CWA; produced a tornado

..Nick..

 

[rdale]

The study compared supercells to squall lines - May 13 was clearly not a supercell. In any case about 50% of MI tor's are from squall lines, in the Alley the number was closer to 85%.

I can't believe that the population is that big of a deal - I don't have figures offhand for comparison but I wouldn't believe St Joseph Co & Lenawee Co have that much difference. In any case spotting in the Hills is MUCH more difficult that spotting in the more open areas of Mich so I would think the numbers would be less than actual for SEMI.

As for lake temps - the influence stops pretty close to shoreline, it wouldn't wait 100 miles to "bang"

- Rob

 

[Justin Turcotte]

I can't believe that the population is that big of a deal

I have been crunching numbers for a long time and I think I have enough data to show that population has a significant impact on revealing the true intensity of tornadoes. I'm convinced the number of "potential" (if there was something there for the tornado to eat) F1+ and F2+ tornadoes is on the order of 200% higher than currently observed. Michigan is not in my area of study so I can only comment on the general findings and not for that specific location. Pull up my plots and you will see that Iowa (and even southern MN for later periods) have a higher frquency of F1+ and F2+ tornadoes than does Kansas. How could this be if Kansas is in the heart of the Alley?... Population density. Kansas gets struck with lots of powerful tornadoes that hit sunflowers and get the low grade of F0. Maybe a lucky F1 for the telephone pole. This effect appears most dramatic for counties with fewer than 40 persons per square km.

http://snrs.unl.edu/amet898/turcotte/primary/climo.htm[/url]

 

[Jeremy Lemanski]

Here is something else that might explain it, county size. Look at a map of southern michigan, the counties in the SE part of the state, such as jackson, washtenaw, lenawee,oakland and even livingston counties. They are large, covering a much greater area then those counties to the west like barry, eaton, branch, etc. It could be more tornados are reported over SE michigan because there is more area for them to form in one county.

One other poss. is timing. Look at 7-2-97. The storms fired over over central mich. and moved east. The only reason they formed over that area and effected only the SE and east part of the state was timing. the cold front was already though many of the SW counties by the time everything began to fire. I can't tell you how many times i have seen Fronts move storm free in the early afternoon across western michigan only to light up as daytime heating works its wonders across the eastern half of the state.

of course the above discounts the idea of the Irish Hills doing anything sig. to a storm that would cause add. rotation. I'm going to do some searching and see what i can find about this.

 

[Robert Dewey]

I think it's better to look at the actual data, versus "per county" stuff. I have been messing around with the SVRPlot program from SPC, and have created a plot of all F4/F5 tornadoes since 1950, and even excluded the April of 1974 data, since it could be considered "anomalous"/once in a 300 year event...

Two main areas stick out like a sore thumb when it comes to numbers and path lengths - And the first area stretches from WI through northern TX, with the second big area being from the Gulf of Mexico states into southern lower MI (the south half). There is also a nice little area over NC/SC... The state of MO seems to be void of tornadoes for some reason, even though they have quite a few hail and wind damage reports - Indicating that there are/were people spotting in those areas...

This data DOES NOT include the recent 2003 and 2004 years, which would add quite a few lines on the map...

Of course, this has nothing to do with how chaseable the storms are, but none the less, there appears to be two main "alleys" across the United States when it comes to devastating tornadoes.

One other thing to note, is that the F scale is based upon damage. The "eastern alley" has a higher population, and thus more structures which may be damaged. Some tornadoes that would be considered F1 in the fields of OK, would be considered F4 if they were to move through Dayton, OH. Every year I see videos and photos of very large tornadoes in OK, with a rating of F1 or so... In contrast to RARELY seeing ANY video of a tornado from the "eastern alley'...

I'll put that graphic up a bit later, when I can upload it onto my server...

 

[Justin Turcotte]

I think it's better to look at the actual data, versus "per county" stuff.... The "eastern alley" has a higher population, and thus more structures which may be damaged.

Agree, but the "actual data" needs to be filtered with some statistics. Counties provide a nice framework as they have a fixed unit area. There are some population shifts to deal with but mainly confined to exurban areas. One cannot simply look at one or even a few counties as the sample size is too small. I have attempted to quantify your assimption using groups of counties within a CWA. An entire state in the Plains is too large given climate variation (east vs west KS for example), so this was a reasonable alternative as most CWA's have a fair number of counties to play with yet are small enough to limit the climate variation. Here is a plot of tornado frequency (1950-2001) versus population density (year 2000) by county for the Norman CWA. This is the "real data" filtered in a way to hopefully give meaningful information rather than generalizations.

[Broken External Image]:http://snrs.unl.edu/amet898/turcotte/misc/cwaplots/Norman.jpg

Here is another rural CWA. Imagine what would happen to the F1 and F2 plots if there was more tornado food out there in central Nebraska.

http://snrs.unl.edu/amet898/turcotte/misc/...ts/Hastings.jpg

Undoubtably, many Plains tornadoes would be higher rated if they simply hit something... anything. Doswell, McCarthy, Gruzulus, among others have discussed this thoroughly. Only a couple folks have actually tried to quantify the "bias".

One last graphic possibly further supports your hypothesis. Look at how the rural counties in the Plains have a very high ratio of F0 tornadoes to all tornadoes...

http://snrs.unl.edu/amet898/turcotte/misc/...ercentF0-76.gif

 

[Robert Dewey]

Good stuff...

Here are the tracks of all F4/F5 tornadoes between 1950 and 2002. The first dataset that I loaded was for tornadoes between 1950 to April 1st, 1974. The second dataset extends from April 5, 1974 to 2002, which is overlayed on top of the first set of tornado tracks. This removes the anomaly of the "Super Outbreak". (The time stamp on the bottom only shows the second dataset).

There appears to be two main areas of concentrated and long tracked significant tornadoes. I believe the western "alley" would have a higher number of strong tornadoes if the population and structural density were to be greater, since most tornadoes in that location tend to be rated F1 as they remain in an unpopulated area/field.

http://69.14.190.10/images/STAT/tornadoes.jpg