Skip to main content

Full text of "NASA Technical Reports Server (NTRS) 19750022659: Characteristics of anvil-top associated with the Poplar Bluff tornado of 7 May 1973"

See other formats

General Disclaimer 

One or more of the Following Statements may affect this Document 

• This document has been reproduced from the best copy furnished by the 
organizational source. It is being released in the interest of making available as 
much information as possible. 

• This document may contain data, which exceeds the sheet parameters. It was 
furnished in this condition by the organizational source and is the best copy 

• This document may contain tone-on-tone or color graphs, charts and/or pictures, 
which have been reproduced in black and white. 

• This document is paginated as submitted by the original source. 

• Portions of this document are not fully legible due to the historical nature of some 
of the material. However, it is the best reproduction available from the original 

Produced by the NASA Center for Aerospace Information (CASI) 



DBpartmmtU of the GeophyUxd Sciences 
The University of Chicago 


Edward W. Pearl 


rpcvAro OF 7 1AY 1‘<73 (Ctiica^o 'iniv.) 
ir C 



Edward W. Pearl 

Department of the Geoplayslcal Sciences 
Tlae University of Chicago 

SMRP Research Paper Number 119 
January 1974 

TTie research reported in this paper has been sponsored by the National Aercmautlcs 
and Space Administration under grant NCR 14 001-008 and by the National Oceanic 
and Atmospheric Administration under grant 04-4-158-1. 


Edward W. Pearl 

Department of the Geophysical Sciences 
Tlie University of Chicago 


Investigation of potential tornado-producing thunderstorms was performed 
during part of the 1972 and 1973 tornado seasons. Participants for the May 7, 

1973 mission included William E. Shenk of NASA and T. Theodore Fujita and the 
author of the University of Chicago. On this date twenty- one tornadoes were 
confirmed over southern Missouri, northern Arkansas, and south v^es tern Illinois. 

The region was surveyed by liigh altitude photography performed on a Learjet over 
the region of reported tornadoes . 

Each of the two storms in this report were chosen from aircraft observation 
with the guidance of ground and radar reports, A series of photographs were taken 
of a tornado producing cloud. An analysis of the activity before and during the 
tornado is made. Most noteworthy were changes detected in the growth and collapse 
of overshooting domes above the anvil. Suggestions are included for a comprehensive 


Inferences have been made as to the thunderstorm characteristics which 
signify tornadic activity. In the early sixties Fujita (1962) discussed a possibility 
that a 45-minute interval existed between consecutive tornado touchdowns, Darkow 
(1971) further supported this concept by his study on long-lived parent thunder- 
storms. Once again the periodicity was found to exist. 


Research was performed to recognize elements of tlie thunderstorm tliat 
may cause such an effect, Bonner and Kemi>er (1971) showed that in the central 
U. S, tliere was a significant dependence of hail production with die hel^t of the 
echo top and yet little dependence was found In relatlcMi to tornadoes. Fujita 
(1972) discovered the relationship that tornadoes may be occurring during die 
pausing or declining stage of the cloud top. 

Tlie motion of tlie overshooting top of a thunderstorm was analyzed by Newton 
(1963). The overshooting dome or turret (Fujita 1974) travels into an area with a 
stable lapse rate and hence will collapse back into the cloud. Newton computed that 
if the overshooting top were to reach 1500m then a downdraft velocity of 30 m sec ^ 
could be obtained at the anvil level. 

Many of the overshooting domes ivere greater in size tlian one kilometer. On 
previous fligjits, hicluding tlie one of May 7, it has not been uncommon to find domes 
greater than sever.’! miles in diameter. Individual turrets usually compose the dome 
struchire. Battan and Theiss (1969) suggest tliat the most common eddy sizes are in 
the order of one kilometer. Tliese turrets approach the Brunt- Vaisalla frequency of 
oscillation as described by Fujita (1974), When many of tliese turrets are close 
together and form the dome structure it appears as though the environment may be 
disturbed. Evidence to that effect has been found 1^ the extreme cold near tlie anvil 

The flights of the past two years have been directed to understanding die 
cloud-top motions. It has been found that in order to get an accurate correlation 
between cloud-top activity and surface phenomena one must include a damage-trudi 
survey and on-time observations at tlie cloud base. Hopefully, this will be included 
in future flights. 

Hair-like clouds have been observed on some of the previous flights. Further 
research may find the cause and the relation of this type of cloud element to tlie rest 
of the thunderstorm. 



A detailed discussion will be given here of a rather small, tornado- producing 
tijunderstorm with an oscil’acing top. Tliis cell also indicated inflow by Iwind-s of 
stratocumulus curved counterclockwise into the base of die cumulonimbus. Tic 
discussion would not be complete if it were not also mentioned that on die same 
day another anvil was photogra|ihed. In this case only a widespread, slowly changing 
overshooting top was observed. 

May 7 began with thunderstorms scattered from Louisiana and Mississippi 
to Iowa and Illinois. Tlie storms were along and ahead of a weak surface cold front. 
The mission began at San Antonio, Texas. At first, we were not sure of di 2 location 
for the most severe activity. After observing the thunderstorms over Louisiana and 
Mississippi, die decision was made to move northward. We positioned the aircraft 
in an area covering northern Arkansas and southern Missouri. The choice was 
excellent since there were twenty-one tornadoes confirmed in this area and a portion 
of southwestern Illinois. 

Fig. 1. Large dome showing little change in a twenty-one minute 
period. First flight of May 7, 1973. 


Tlie weak, widespread overshooting tiiunderstorm was approached on tliis 
first of two flig^its. We continuously i^iotographed the cell for a fifty minute period. 

It became obvious tliat daere was little variance in the amount or height of tlie over- 
shooting dome above what could be considered the anvil surface. One may question 
whether or not the entire dome may be considered a raised anvil top. Our position 
relative to the cloud changed considerably throughout tlae flight. Figure 1 shows 
this cell twaity-one minutes apart. The first photograph was taken at 2123 Z and 
tlie second at 2144 Z, indicating little change. 

The pictures indicated a massive overshooting dome. Tlie individual turrets, 
see Fujita (1974), indicated up and down motions hut the dome as a whole varied only 
slightly, A cloud of this magnitude would likely be seen from present day satellites, 
however it would probably be difficult to distinguish the actual anvil from tlie dome. 

It was interesting that there were not any confirmed tornadoes with this cell. 
We are, of course, basing this on the state-of-tlie-art of the present St^rm Data 
publication, A survey of the sight would have been desirable to confirm such a report. 

James Purdom was located at NSSFC in Kansas City for the purpose of 
supplying the valuable up-to-the-minute forecasts. Utilizing ATS photographs and 
ground based reports, he emphasiz. d the necessity for the second flight. He 
informed the research team of several unconfirmed reports of tornadoes. Therefore, 
although it was late in the day and the photographic mission would be limited, it was 
decided to go forward with the second flight. 

Shortly after takeoff for tiie second flight for May 7, the author and the other 
researchers observed a few cells surrounded by a clear, almost cloud-fret ■irea. 

One of the cells in particular indicated a pulsating growtii and collapse of - r .ull 
dome at the top of the anvil. The author was aware that bands of stratocumuius 
clouds were curved in a counterclockwise manner into the base of the storm as in 
figure 2. 

The photograpiiy canmenced shortly after 2326 Z and continued for several 
minute intervals until 0015 Z or about forty-nine minutes later. There were a number 
of times when no pictures were taken due to the difficult maneuvers that were 
necessary in order to pAiotograph. A position close to the cloud was disturbed by 






Fig. 2. Indication of cyclonic flow seen as bands of stratonimuli 
move toward base. 

the anvil of a nearby thunderstorm. The last fifteen minutes were used in a maneu- 
ver to fly over the cell in order to obtain a good view of the rear edge of the anvil. 


The tornado- producing cell indicated that the maximum heights of the over- 
shooting dome occurred at approximately 232945, 233830, and at 234300. The dome 
at 232945 is shown in figure 3. After the last significant overshooting dome of 234300 
there was a gradual collapse and the top became much less active as shown in figure 4. 
The author discovered later that during the slow cessation of overshooting activity a 
tornado was spawned by this thunderstorm. The tornado was confirmed in Storm Data. 
It was described as originating at Poplar Bluff at 2350 Z and continuing in a skipping 
path for fifty miles northeastward. The tornado was intense enough to designate the 
damage as F 3 according to the Fujita scale (Fujita 1971). oscillation of the dome could only be described in scalar quantities as 
used in figure 5. The reason for the scalar value is due to the great variation of 
distance and especially in the height of the aircraft in relation to the anvil top. The 


Fig. 3. Maximum dome height is reached. 

Fig. 4. Activity subsides as tornado is produced. 




scale was made such that O indicates the anvil Itself; 5, tlie greatest helglit reached 
by one of tiie domes; and every otlier value between is a proportionate amount. In 
other words , if a value of one were given tlien the dome would be one-fiftli of die 
heiglit of the higliest dome observed on tliis flight, A negative value was given when 
a dome dropped below tlie anvil top. We were able to verify this since our altitude 
was such tiiat the surface of the anvil was observable; however, the amount of depar- 
ture below the anvil was unknown. 


-25 -20 -15 -10 -5 0 +5 +10 +15 +20 +25 


Fig. 5. Oscillation of dome as a function of time. 

Figure 5 shows the change in height of the dome as a function of time. There 
were three periods of overshooting prior to the tornado. It is evident that the tornado 
began during a quieting of the overshooting activity. Shortly after the tornado began 
the dome actually dropped below the anvil top. Only a small oscillation with a maxi- 
mum value slightly over two was observed over fourteen minutes after the tornado 



Railar iiiiliLaicd U»al tiic cells investigated l>y Uie Learjet team had significantly 
smaller echoes tJian a large group of cells located about twenty-five miles to die east. 
Hie crew of die Learjet were aware of diese cells and one is clearly vi.«ible during 
one |xiss at 2.1S(>Z as shown in figure b. Strangely enougli, diese larger cells did not 
produce any tornadws during die time of our flir it. 

Fig. 6. Larger storm seen twenty -five miles to die east. 

Tlie tornado- producing echo was not easily identified on radar. Tlie echoes, 
unfortunately, were almost at die limits of diree radar stations in die area: Little 
Rock (LIT), Mempliis (NQA), and St. Louis (STL). Tlierefore, diere was some 
difficulty due to attenuation and heigtit of die radar bc-am. Furdier problems were 
encountered, not die least of which was a broken timer on die St. Louis radar. 'Die 
Memphis radar indicated a much greater definition of die echoes and especially of die 
echo producing die tornado. 

Tlie echoes at 2350 Z are shown in figure 7 along with the flight track of die 
Learjet covering die entire period of piiotograpliy. Central Dayliglit Time in five 
minute intervals was used on die flight jiath. 



^ 1 

t I 

Fig. 7. Track of Lcarjet and echo 
l.Kalion -1 2350Z (1850GDT). 


The flight track In the last figure shows that the Learjet was not in a good 
position for photography all of the time. Although the flight was very successful there 
is room for improvement in the future. 

The flight path close to a thunderstorm may obscure interesting large scale 
details of a thunderstorm. It L' possible to deteimine that turret motion approaches 
the Brunt-Vaisalla frequency at a close distance (Fujita 1974). Although of interest 
scientifically, the fact may be impractical from a stand[x>int of satellite photography 

There is a likelihood that we were too close to the tornado-producing cell. 
Fortunately, we were able to observe the dome of the tornado- producing thunderstorm. 
Still it would have been desirable to have seen the relation of this cell to the odiers 
in the vicinity. Secondly, when studying the radar film, many problems are eliminated 
if a one-to-one relation can be easily found between the echo and die observed storms. 
Also, at a greater distance the researchers could have observed more than one cell at 
a time and thereby increase the probability of observing a tornado- producing thunder- 


llie autliur and tlie otlier riisearehers have decidtid tliat a greater distance is 
needed for applicability towards die future satellite programs. It is die audior's 
opinion tiiat a distance slightly greater Uian 100 nautical miles would be appropriate 
depending on die atmospheric coiiditions. At a greater distance from die cloud die 
aircraft would not likely be in die vicinity of odier severe storms. Tlierefore difficult 
maneuvers widi die aircraft could be avoided. Tlie maneuvers diat were required 
during die May 7 flight caused periods O: time where no photograi^is were taken. The 
continuity is dierefore scanewhat disturbed. 


The storm discussed in diis report is die first tornado-producing one that was 
photographed before and during a tornado by the use of a Learjet. Tliere may be 
great significance in the fact diat die top of the storm showed a decrease in activity 
associated with die production of die tornado. Furdier evidence must be gathered 
before any final conclusions can be made. 

Fujita (1974) points out that diere appears to be a relation between the vertical 
mass transport and the spreading rate of the anvil. Purdom (1971) indicated diat it is 
when the spreading rate slows or temporarily stops diat tornadoes often occur, It 
could be deduced that when die vertical transport of mass slows or comes to a halt 
the chance of a tornado increases. 

Figure 8 shows die last picture taken of the cell. There is a definite anvil 
overhang on die upwind siue close .o where the overshooting is taking place, Fujita 
(1974) describes die likelihood of a meso-high structure at die top of die anvil wldi 
overshooting domes. Tlie high induces a modification in die tropopause and an 
outflow of cold air spreads outward from the vicinity of the cold dome. 

Damage-truth surveys and on-time observations should be run in order to 
augment the results in die future. Storm Data cannot possibly verify every tornado 
that occurs. For instance, if there is a sparse population in a particular area, die 
tornado may go completely undetected, A detailed investigation could be accomplished 
by die use of Cessna aircraft and automobiles. Also, similar vehicles could be used 
to photograph or at least record any peculiar activity at die base of the storm. 


Fig. 8. Anvil lip shown on upwind side. 

Foif^.^r research concerning the relation between u cessation of overshooting 
ccti'.'ity and tornado production is needed. Forecasting of tornadoes with the present 
skill of the Lcarjet aircraft or future use of satellite photography depends on the 
determination of a one-to-one relation between what occurs at the anvil top to what is 
occurring at the base of the cloud. 

All flights performed should have guidance as to where the most severe activity 
is to occur. Radar, as shown in this papier, may not be enough for propicr directives. 
It is a combination of surface and aircraft observations as well as an analysis of 
current ATS pihotograpihs tliai wv.ald lead to a smootlily controlled project. 

A greater distance between the storm and the Learjet has been found to be 
necessary. The enlarged field of view enhances the possibility of piliotographing 
a tornado -producing thunderstorm. Therefore tlie large details of tornado- producing 
storms should be visible. 

Ackncwleclgem ents : 

Tlie LearjcJt mission could not have been successfully completed without tlie 
full cooperation of those involved: Mr. Jaiiies Purdom who prepared the forecasts 
for and aided in tlie guidance of die aircraft throug^i the use of die facilities at 
NSSFC in Kansas City; Messrs, Vincent J. Oliver and Edward Ferguson who gave 
helpful advice when It was most needed; tlie {ilot, Clyde Alber, and the co-pilot, 

Ray Miller, who carefully took die aircraft wherever and whenever diey were asked: 
die researchers, William E. Shenk of NASA and T. Theodore Fujita from die Univer- 
sity of Chicago who jointly worked with the author. 

Special additional commendation should be given to the pilot and co-pilot for 
their careful not^takliig which allowed us to reconstruct our flight track. 


Battan, L. J. andj. B. Theiss (1969): Variable Structure of Thunderstorm Updrafts. 
6th Conference on Severe Local Storms, Chicago, 155-158, 

Bonner, W, D, andJ, E. Kemper (1971): Broad- Scale Relation Between Radar and 
Severe Weather Reports. 7th Conference on Severe Local Storms, Kansas 
City, 140-147, 

Darkow, G. L, (1971): Periodic Tornado Production by Long-Lived Parent Thunder- 
storms. 7di Conference on Severe Local Storms, Kansas City, 214-217. 

Fujita, T. T. (1962): A Review of Researches on Analytical Mesometeorology. 

SMRP Research Paper 8. 32 pp. 

Fujita, T. T. (1971): F P P Tornado Scale and Its Applications. SMRP Research 
Paper 98, 15 pp. 

Fujita, T, T. (1972): Tornado Occurrences Related to Overshooting Cloud- Top 

Helots as Determined from ATS Pictures. SMRP Research Paper '^7, 32 pp. 

Fujita, T. T. (1974): Ovei-shooting Thunderheads Observed From ATS and Learjet. 
SMRP Research Paper 117, 29 pp. 

Newton, C. W. (1963): Dynamics of Severe Convective Storms. AMS Meteorological 
Monographs, Vol. 5, No, 27, 33-55,