What determines which way a tornado rotates?

[Mike Gauldin]

Forgive me if this is a stupid question, but what determines which way a tornado rotates (clockwise/counterclockwise)

 

[Pat Lawrence]

That's a good question, one I don't think any person knows the answer to for certain! I'll say the Coriolis "Force/Effect", although one must wonder if it has that large of an effect on a tiny little system like a tornado

Pat

 

[Mikey Gribble]

I think it is just because of winds veering with height in the Northern hemisphere when you have warm air advection, which is usually associated with tornadoes since you need bouyant air in the boundary layer. This means you usually have cyclonically rotating storms in the northern hemisphere and anticyclonically rotating storms in the southern hemisphere. Of course there are exceptions and we get tornadoes rotating both ways in both hemispheres. I could be wrong, but I don't think I am.

 

[Jeff Snyder]

Originally posted by Michael Gribble
I think it is just because of winds veering with height in the Northern hemisphere when you have warm air advection, which is usually associated with tornadoes since you need bouyant air in the boundary layer. This means you usually have cyclonically rotating storms in the northern hemisphere and anticyclonically rotating storms in the southern hemisphere. Of course there are exceptions and we get tornadoes rotating both ways in both hemispheres. I could be wrong, but I don't think I am.

That seems about right. Most tornadoes in the US occur in environment of mesoscale/synoptic-scale low-level warm-air advection and resultant veering vertical wind profile. This yields rotation in storms, and, depending upon the shape of the hodograph, cyclonically-rotating storms are often preferred. Cyclonic mesocycles often produce cyclonic tornadoes... Now, there are anticyclonic mesos (i.e. from a left-split supercell), but again, those would only be preferred (over cyclonic mesos) in an environment characterized by counterclockwise looping hodographs (which is often associated with cold-air advection). Before a supercell splits (particularly in a straight-line hodograph environment that favors splitting supercells), there is often a set of cyclonic - anticyclonic couplets across the updraft. There have also been observed anticyclonic storm-scale rotation on the south side of an RFD (right side of a hook)... I think this is how the 5-29-04 OK anticyclonic tornadoes developed, though I'm not sure. Regardless, it's also important to remember that there can be shallow, non-mesocyclone tornadoes that derive their rotation primarily from intense horizontal shear or previous vertical vorticity (or, say, a misocyclone), like a gustnado or landspout.

EDIT: The Coriolis force doesn't really affect the rotation of a tornado. The Coriolis force only significantly affects flow on the larger time and spatial scale... For short-term and small vortices like tornadoes, the Coriolis force is negligible.

 

[Pat Lawrence]

Yes, this is all very interesting regarding misos and mesos and the like, but what causes them to spin in the way which they seemingly choose to spin? I don't know that there is a concrete answer to that question, to say nothing about why a tornado spins one way or another. It does seem that there is a difference between whether or not one is north or south of the equator, however.

Pat

 

[Andrew Khan]

The Coriolis Effect, I believe is more contolling of a larger, massive space of planetary existance. However, since it affects such a large environment and area, it would seem likely for those affects to create a "Domino Effect", and affect smaller masses.....and consequently affect meso and synoptic scale systems.

 

[Pat Lawrence]

Originally posted by Andrew Khan
The Coriolis Effect, I believe is more contolling of a larger, massive space of planetary existance. However, since it affects such a large environment and area, it would seem likely for those affects to create a \"Domino Effect\", and affect smaller masses.....and consequently affect meso and synoptic scale systems.

Coriolis may have some small effect on torndoes, but as Jeff said, the effect is likely not that large. To state the matter simply, nobody knows the answer to the original question....it is nothing more than "guess ""work"" at present.

 

[Scott A. Kampas]

I agree with Pat in the uncertainity, we just don't know, it's partly a combination of things as Jeff (and Michael) alluded to, and partly things we just don't know yet. This paper may be of interest, and the lead author has a local copy so AMS journal access isn't required:

Monteverdi, J. P., W. Blier, G. Stumpf, W. Pi and K. Anderson, 2001: First WSR-88D documentation of a an anticyclonic supercell with anticyclonic tornadoes: the Sunnyvale/Los Altos tornadoes of 4 May 1998, Mon.Wea.Rev., 129, 2805-2814.

Scott

 

[Aaron Kennedy]

Tornado spin is pretty simple.... just look at what sort of shear the tornado forms in... cyclonic or anticyclonic.

Overly simplified explanation for a typical severe weather day in the US.

Look at the figure below... height increases with each line so the bottom arrow is say at the sfc, and the top is 500mb.

So you end up with horzizontal vorticity. When an updraft interesects one of these vortex lines, it creates cyclonic vertical vorticity on the S side and anticyclonic on the N. For a bunch of reasons I won't get into here, the right split is usually favored and the cyclonic mesocyclone becomes dominant, while the left (anticyclonic split) dies. Sometimes, the leftsplit can be favored and you end up with an anticyclonic meso.

Usually, we see a hook echo at the sfc when the meso becomes sfc based... precip/hydrometeors wrap around. If the conditions are ok, a tornado can form with the parent storm's rotation. However, if you have a strong downdraft in the RFD, you can get anticyclonic shear on the backside and sometimes you'll get an anticyclonic tornado. THis was the case on 5/29/04 in OK. Other times, an anticyclonic supercell can produce a anticyclonic tornado, but that is pretty rare in the northern hemi.

So in summary... we absolutely know why tornadoes spin either cyclonically or anticyclonically. No mystery!

Aaron

 

[David Schuttler]

On 5-29-04 with the Harper County cell you had this
an anticyclonic funnel on the s. side

and cyclonic circulation on the n. side

 

[Glen Romine]

Originally posted by Aaron Kennedy
Tornado spin is pretty simple.... just look at what sort of shear the tornado forms in... cyclonic or anticyclonic.

I think this is a bit misleading - as what Aaron describes is what gives rise to cyclonic rotation in a typical thunderstorm in the plains. Whether you are looking at streamwise (favored with a 'looping' hodograph) or crosswise vorticity (straight line hodograph) ingested into a storm updraft - it can favor the production of rotation through tilting of the horizontal shear (change in wind speed or direction with height). Most storms, it turns out, have a combination of both cyclonic and anti-cyclonic rotation within updrafts. The stronger of the two is typically cyclonic - favored both by the long life (which allows Coriolis to eventually become important after ~ 2 hours of life) and by the more common warm-advection pattern that provides cyclonic curvature in the hodograph (therefore a streamwise cyclonic vorticity source). In the central plains - the environment where severe storms are favored to develop is typified by a warm advection pattern - though not always. In places like California, anti-cyclonic supercells are not all that uncommon with hodographs showing increasing cold air advection with height. Anyhow - supercell tornadoes are favored by what the dominant rotation is in the parent updraft and probably through momentum transport by precip and downdrafts, which will be in the same sense of rotation as the parent updraft, the same sense of rotation is transported to the surface. Many supercells support anti-cyclonic rotation at the surface as well as aloft - though I think the numbers that lead to an anti-cyclonic tornado are fewer than 1/500.

The typical argument against the tilting mechanism for tornadogenesis is that you can't break vortex lines and there is no way to get vertical motions to the ground sufficient to tilt enough helicity (dominated by the horizontal component) into the vertical to have a tornado scale vortex. The process of how this happens is simply not known yet with any certainty.

Since non-supercell tornadoes do not benefit from this relationship with favored rotation derived from the parent updraft, why aren't more non-supercell tornadoes anticyclonic? Again, I think you have may have to look at Coriolis as a player with regard to favoring boundaries with cyclonic shear, as well as the typical environments that storms develop in (in the plains, that usually means a clockwise turning hodograph with height), such that the warm advection side of the boundary is on the south side, and not the north.

It's important to also keep in mind that in the southern hemisphere, tornadoes there are also typically cyclonic, but that means they rotate the opposite direction to northen hemisphere tornadoes. So again, you can't exactly throw out the Earth's rotation as a player in the preferred tornado sense of rotation.

Glen

 

[Aaron Kennedy]

Aaron Kennedy wrote:
Tornado spin is pretty simple.... just look at what sort of shear the tornado forms in... cyclonic or anticyclonic.

Earlier, someone implied we couldn't tell which way a tornado rotates... well if there is cyclonic shear in the surface flow (in an x-y domain) I can promise you there isn't a anticyclonic tornado in there. I don't think the original question was pertaining to actual tornadogenesis, rather the equivelent of the "which way does the toilet spin" question.


Aaron

 

[Jeff Snyder]

Originally posted by Aaron Kennedy

Aaron Kennedy wrote:
Tornado spin is pretty simple.... just look at what sort of shear the tornado forms in... cyclonic or anticyclonic.

Earlier, someone implied we couldn't tell which way a tornado rotates... well if there is cyclonic shear in the surface flow (in an x-y domain) I can promise you there isn't a anticyclonic tornado in there. I don't think the original question was pertaining to actual tornadogenesis, rather the equivelent of the "which way does the toilet spin" question.


Aaron

I agree in that it's really isn't a "mystery" why most tornadoes rotate the way they do... However, I don't think I'd say that we can "promise that there isn't an anticyclonic tornado in" an environment of cyclonic low-level shear ... In this thread, we've only really looked at mesocyclone/supercell tornadoes, not some of the more shallow, "spin-up" varieties that may not derive their rotational energy from the mesocyclone. In addition, shallow anticyclonic mesocyclones may still develop, I believe, in environments that otherwise favor cyclonic mesos (clockwise looping hodographs), due to storm-scale processes... I think Wurman documented this in a May 15th 2003 TX supercell (with anticyclonic rotation between two cyclonic mesos). In addition, I think the anticyclonic tornadoes from the central OK supercell on 5-29-04 may have developed from a shallow mesocyclone to the immediate south of the intense RFD.

 

[Aaron Kennedy]

I'm just relating the wind flow pattern to the spin of the tornado... the anticyclonic tornadoes typically occur on the backside of the hook echo where there is anticyclonic shear.

----> *cyclonic*
----------->
---------------->
----------->
-----> *anticyclonic*

Aaron