Tuesday Nov. 22, 2011
click here to download today's notes in a more printer friendly format

A long chain of events lead to this morning's music selection.  First I've been reading lots of Foucault Pendulum reports.  Several students mentioned they would like to go to Paris to see the replica of Foucault's original pendulum (an excellent idea).  Over the weekend I went to watch one of our graduate students play at a Southern Arizona Symphony Orchestra which featured a composition by George Gershwin.  That made me thing of "Rhapsody in Blue" so I downloaded a couple of shorter versions of it.  The one you heard in class featured Pearl Kaufman on piano.  I couldn't find that online, but here's a pretty good substitute.  While looking online I found a pretty amazing version of "Bumble Boogie" that was played by Pearl Kaufman.

The Foucault Pendulum reports and the Stability Analysis have been graded and were returned in class today.  Once a topic is graded, we update the list of students that have earned 45 1S1P pts.

I have found many of the tornadoes that I tried to show in class last Thursday online.  I replayed a couple of them this morning (the McConnell AFB tornado and the Kansas Turnpike tornado.  The turnpike video also has a warning that a highway underpass is actually a very dangerous place to take shelter from a tornado.  Here is some additional information from the Norman OK office of the National Weather Service.  Slide 6 lists some of the reasons why underpasses are so dangerous.
The figure below (p. 162 in the ClassNotes) illustrates the life cycle of a tornado.  Have a close look at the next tornado you see on video and see if you can determine whether it is in one of the early or late stages of its development.


Tornadoes begin in and descend from a thunderstorm.  You would usually see a funnel cloud dropping from the base of the thunderstorm.  Spinning winds will probably be present between the cloud and ground before the tornado cloud becomes visible.  The spinning winds can stir up dust at ground level.  The spinning winds might also be strong enough at this point to produce some minor damage.  Here is video of a tornado in Laverne Oklahoma that shows the initial dust swirl stage very well.

In Stage 2, moist air moves horizontally toward the low pressure in the core of the tornado.  This sideways moving air will expand and cool just as rising air does (see figure below).  Once the air cools enough (to the dew point temperature) a cloud will form. 

Tornadoes can go from Stage 2 to Stage 3 (this is what the strongest tornadoes do) or directly from stage 2 to stage 5.  Note a strong tornado is usually vertical and thick as shown in Stage 3.  "Wedge tornadoes" actually appear wider than they are tall.

The thunderstorm and the top of the tornado will move faster than the surface winds and the bottom of the tornado.  This will tilt and stretch the tornado.  The rope like appearance in Stage 5 is usually a sign of a weakening (though still a dangerous) tornado.





A tornado cloud forms is mostly the same way that ordinary clouds do.  Moist air moves into lower pressure surroundings and expands.  The expansion cools the air.  When the air cools to its dew point a cloud forms. 

This seemed like a good place to briefly discuss supercell thunderstorms.


Here is a relatively simple drawing showing some of the key features on a supercell thunderstorm.  In a supercell the rotating updraft (shown in red above) is strong enough to penetrate into the stratosphere.  This produces the overshooting top or dome feature above.  A wall cloud and a tornado are shown at the bottom of the mesocyclone.  In an ordinary thunderstorm the updraft is unable to penetrate into the very stable air in the stratosphere and the upward moving air just flattens out and forms an anvil.  The flanking line is a line of new cells trying to form alongside the supercell thunderstorm.


Here is a second slightly more complicated drawing of a supercell thunderstorm.  A typical air mass thunderstorm (purple) has been drawn in for comparison.

A short segment of video was shown at this point (again without sound).  It showed a distant supercell thunderstorm and photographs of the bases of nearby supercell thunderstorms.  Here you could see the spectacular wall cloud that often forms at the base of these storms.  Finally a computer simluation showed some of the complex motions that form inside supercell thunderstorms, particularly the tilted rotating updraft.  I haven't been able to find the video online.




A wall cloud can form a little bit below the rest of the base of the thunderstorm.   Clouds normally form when air rises, expands, and cools as shown above at left.  The rising air expands because it is moving into lower pressure surroundings at higher altitude. 

At right the air doesn't have to rise to as high an altitude to experience the same amount of expansion and cooling.  This is because it is moving into the core of the rotating updraft where the pressure is a little lower than normal for this altitude.  Cloud forms a little bit closer to the ground.


Photograph of the base of a thunderstorm showing part of the wall cloud and what looks like a small and weak tornado. (from the University Corporation for Atmospheric Research)
Thunderstorms with rotating updrafts often have a distinctive radar signature called a hook echo.
We haven't discussed weather radar in this class yet.  In some ways a radar image of a thunderstorm is like an X-ray photograph of a human body.  The Xrays pass through the flesh but are partially absorbed by bone.


Xrays pass through tissue but get absorbed by bone.  They reveal the skeletonal structure inside a body.  In some respects radar is similar.





The radio signals emitted by radar pass through the cloud itself but are reflected by the much larger precipitation particles.  The intensity of the reflected signal (the echo) is color coded.  Red means an intense reflected signal and lots of large precipitation particles.  The edge of the cloud isn't normally seen on the radar signal.

Here is an actual radar image with a prominent hook echo.





This is the radar image of a thunderstorm that produced a very strong tornado that hit Oklahoma City in May 1999
( http://www.spc.noaa.gov/faq/tornado/radscel.htm ).  The hook echo is visible near the lower left hand corner of the picture.  Winds in the tornado may have exceeded 300 MPH.  You can read more about this tornado here.  And here is some storm chase video of the tornado.

It is very hard to actually measure the speed of the rotating winds in a tornado. Researchers usually survey the damage caused by the tornado to come up with a Fujita Scale rating. 

Here is the link to the photos that were shown in class with some discussion from the National Oceanic and Atmospheric Administration Storm Prediction Center.

The photographs below weren't shown in class.



Roof damage is typical of an F1 tornado.  The buildings on the left suffered light roof damage.  The barn roof at right was more heavily damaged.  Barns present a larger crossection to the wind and often aren't built as sturdily as a house.

More severe damage to what appears to be a well built house roof but still an F1 tornado.


Even relatively weak winds can damage a mobile home.  F1 tornado winds can easily tip over a mobile home if it is not tied down (the caption states that an F1 tornado could blow a moving car off a highway).  F2 level winds (bottom photo above) can roll and completely destroy a mobile home.




Trees, if not uprooted, can suffer serious damage from F1 or F2 tornado winds.

F1 winds will damage a roof, F2 level winds can completely remove the roof.  The outside walls of the building are still standing.

The roof is gone and the outer walls of this house were knocked down in the photo above.  This is characteristic of F3 level damage.  In a house without a basement or storm cellar it would be best to seek shelter in an interior closet or bathroom (plumbing might help somewhat to keep the walls intact).

In some tornado prone areas, people construct a small closet or room inside their home made of reinforced concrete.
A better solution might be to have a storm cellar located underground.


An F4 tornado knocked down all of the walls in the top photo but the debris is left nearby.   All of the sheet metal in the car body has been removed in the bottom photo and the car chasis has been bent around a tree.  The tree has been stripped of all but the largest branches.



An F5 tornado completely destroyed the home in the photo above and removed most of the debris.  Only bricks and a few pieces of lumber are left.

I would have liked to have played another video tape of a tornado that struck Pampa Texas but decided not to because of the audio problems.  I did manage to find the tornado online even though it was missing some of the interesting commentary on the tape so we watched that.  Near the end of the segment, the video showed several vehicles (pick up trucks and a van) that had been lifted 100 feet or so off the ground and were being thrown around at 80 or 90 MPH by the tornado winds.  Winds speeds of about 250 MPH were estimated from the video photography (the wind speeds were measured above the ground and might not have extended all the way to the ground).


Several levels of damage (F1-F3) are visible in the photograph above.  It was puzzling initially how some homes could be nearly destroyed while a home nearby or in between was left with only light damage.  One possible explanation is shown below (from the bottom of p. 164 in the photocopied ClassNotes.



Some big strong tornadoes may have smaller more intense "suction vortices" that spin around the center of the tornado.  Tornado researchers have actually seen the pattern shown at right  scratched into the ground by the multiple vortices in a strong tornado.



The sketch above shows a tornado located SW of a neighborhood.  As the tornado sweeps through the neighborhood, the suction vortex will rotate around the core of the tornado.




The homes marked in red would be damaged severely.  The others would receive less damage.  Remember that there are multiple suction vortices in the tornado, but the tornado diameter is probably larger than shown here.