JUNE, iga$
Length from head kite (meters)
0 to 800 ___._____________________________
800 to 1,800 ________________ ______________ 1,W to 1,900 ____________________________ 1,oOO to 2,075 ____________________________ 2,076 to 2,360 __._________________________
a,m to 3,600 ____________________________ 3,Boo to 3,880 ____________________________ 3,W to 3.800 __________ ____ - _____________ 3,800 to 4,000
MONTHLY WEATHER REVIEW 219
Diameter
meters)
Condition
0.8 Destroyed. L 0 Brfttle like glass.
1.0 Dwk blue. 1.0 1.0 Very dark blue.
::! %?brown. 1.11 Dark brown. 1.11 Dark brown to dnrk blue.
Yellowish brown and dark blue.
ently not affected.
5. Drexel: Voltmeter showed in excess 50,000 volts for altitude Steady stream of brilliant sparks jumping 10 centi- A t 9.23 flash of light- Effect on wire as
1,600 meters. meters. ning and thunder; 4,000 meters of wire out.
follows:
Thunder first heard at 8.33 a. m.
The string attaching head kite to wire was burned. That portion
of the wire within the lower stratus cloud (below 1,000 meters)
showed no ill effectm from lightning, whereas that portion between base of cloud and earth (650 meters) was considerably affeched in
spite of the fact that i t was wire of larger diameter, and therefore less resistance. The wire in the dry air (2,300 to 1,300 meters) between the two clouds layers was either entirely destroyed or rendered unfit for use. It is evident that the electric charge originated in the upper cloud layer and much of i t passed along
the wire into the lower cloud. A portion continued to earth but
did not affect the wire because of the moisture on it, but did injure the wire in the drier air below. Thus an airplane might form part
of the path of discharge. (See Supplement No. 10, M. R. R.,
1918, pp. 5-6.) 6. Broken Arrow: 1,800 meters out; three kites. Stratus cloud
400 meters high. Lightning strikes head kite and coinnletelv
destroys wire from kit; to reel house leaving along the pat6 a dis- charge a streak of thick yellowish brown smoke. If this discharge occurred in 0.001 second, the voltage is not far from 3,000,000, or about that of the artificial near lightning of Peek. Compare this with the next case. 7. Drexel: 3,535 meters of wire, except 20 or 30 near the reel,
vaporized. The lower portion fused. 8. While not a kite wire record, i t map be mentioned that on April 16, 1936, an airplane carrying eight pamengers going from
Paris to London was struck near Beauvais. A large patch of fabric was torn out. the compass demagnetized, one of the main spars
scorched, all bondings fused, and one aileron badly damaged.
Doctor Dorsey has advanced the theory that there are
electronic darts, or localized stream lines of electrons and that a positive stroke advances by a series of steps depending upon the occurrence of free electrons. Branch-
ing is to be expected; while in a negative stroke the elec- trons advance in a mighty rush. He objects to Doctor
Simpson’s deductions from the preponderance of negative
polarity in side-split branches, as shown in many photo- graphs. Inspection of the 3,600,000 volt flash herewith
shows, curiously enough, split-off discharges in both
directions from the same flash. Humphreys has calculated (Physics of the Air, p. 396)
in the case of a hollow tubular conductor crushed by
lightning and assuming certain temperatures, an amper- age ranging from 19,470 to a maximum of 100,000. With
the latter value and assuming a megadyne pressure on the inner tube, there results a pressure of 2,638 by l o 4 dynes per square centimeter or roughly 26 atmospheres. He
wains, however, that these are rough estimates and “that this particular discharge presumably was exceptionally heavy since it produced an exceptional effect.” He also
quotes Pockels estimate of 10,000 amperes. Mr. S. A.
Korff, of the General Electric Co., has called my attention
to Steinmetz’s estimate of the energy as 10’ watt-seconds
or 2.8 kilowatt-hours which is oiily a thousandth of
Wilson’s value. Larnlor has estimated the energy as 28
kilowatt hours. (Proc. Roy. SOC. 1924, Vol. 90, p. 312.)
Since the voltage breakdown of air is 9 by 10 volts it
seems likely that estimates exceeding this are too high;
and as We breakdown is probably progressive, values of
1.2 by 10 volts are ample, thus bringing the energy value to approximately 28 kilowatt-hours.
For the benefit of the lay reader then we may say that in our opinion the energy of an average flash of lightning
does not run much over 10 kilowatt-hours or, let us say,
enough to operate three ordinary toasters (300 watts)
for 10 hours.
PHENOMENA PRECEDING LIGHTNING
By ALEXANDER MCADIE
(Blue Hill Observatory, Mans.)
In the Meteorological Magazine June, 1928, p. 113,
Mr. R. S. Breton, writing from Tung Sung, Southern Siam, states that on a number of occasions he has noticed
a sharp “vit” or “click” accompanying lightning that
has struck something in the immediate neighborhood, preceding the thunder by a perceptible fraction of a
second.
He adds that he has three times noticed that animals
show alarm immediately before a flash and that in one
case a dog walking on grass turned and began to bark
angrily in the direction of a very strong flash that came
one-fourth second after, striking several of a group of trees 200 yards awa . He mentions two occasions when
very near discharge actually took place. In each case
the discharge was a very powerful one, taking place on
dry soil before rain had fallen. He asks “if it may be that the sensitive feet of the dog could detect vibrations
before the discharge took place.” The editors of the magazine answer “that the ‘vit’
or ‘click’ accompanying lightning which has struck close
by appears to be new; no reference to any similar observa-
tion can be found in the literature and a t present it is not possible to offer any explanation.” Clicks preceding intense lightning flashes are common
at Blue Hill Observatory and undoubtedly can be heard
fowls rushed for she T ter from the open in alarm before a
elsewhere under certain conditions, when an insulated metallic conductor is exposed, in a strong electric field,
and a grounded conductor is close by. At Blue Hill
every intense flash within a radius of 1,000 meters gives this click preceding thunder by an interval which i s a
function of the distance of the flash. Thus for an iiiterval of 0.4 second (a frequent value), with mean temperature
of air column from ground to cloud 1,100 kilograds
(303’ A. or 86’ F.) relative humidity 90 per cent absolute
humidity 27 grams per cubic meter of space, wind direc-
tion 235’ (SW. by S.) velocity 7 meters per second, the
distance is
d = t (Vo J T /1 0 0 O ) + wind
=0.4 (332.11 s 1.05) X 7
= 142 meters
Intervals as large as six seconds indicating a flash
distant 2 kilometers or more have been noted. Regarding the behavior of the dog, it would seem to
be not so much a question of sensitive feet as a matter
of insulation and increasing electrification to a degree that the hairs, for instance, become discharging points. This bristling can be seen readily on animals caught in
thunderstorms near the top of a mountain. I recall
being near the summit of Mount Whitney (4,420 meters above sea level, 14,502 feet), during a thunderstorm.
The hairs of the burros (pack animals) stood out straight,
220 MONTHLY WEATHER REVIEW JUNE, 1988
and a faint hissing could be heard. A metal button on
my cap gave a tingling sensation. I kept wondering
how long it would be before a flash of lightning would
demolish the entire party of astronomers as they pro- ceeded in close formation to the summit. I think we had a narrow escape from disaster. During a week’s
stay at the summit we had several thunderstorms,
when the lightning seemed to be below us.
The feeling of uneasiness preceding lightning flashes may be due, aside from eflects of pressure, temperature,
and humidity, to the increasing electrical strain, as a
charged cloud comes over the position of the observer.
We know from our quadrant-electrometer measurements
that at such times the potential gradient increases steadily from 50 volts per meter to 10,000 or more. A
jet of water from an insulated collector exhibits many
interesting changes as the charged cloud approaches. In fact we can tell just about when the flash will occur.
We can also detect and record discharges which an observer fails to detect, if dependent on the eye alone. With each flash there is an instantaneous equalization
of potential and return of the needle to zero.
Prof. C. T. R. Wilson has shown (Phil. Trans. A Vol.
221, p. 112, 1921) that continuous currents carried by ions moving in the strong field below the cloud, exist,
these ions being produced as a result of point discharges frdm trees and bushes below the cloud. Prof. B. F. J. Schonland has estimated the magnitude of these cur-
rents. Using an insulated Acacia Karroo tree, a small thorn
tree, about 12 feet high with plenty of thorns, he meas- ured the field and current strength. With a field of
negative 16,000 v/m, the current as measured by a unipivot galvanometer with one terminal to tree and
other to earth, was 4 microamperes. His table shows
that during 230 minutes of strong negative field, the tree
discharged 0.0129 coulomb of positive electricity upward,
while during 10 minutes of strong positive fields it dis-
charged 0.0001 coulonib of negative electricity. The latter effect was due to a mammatiform cloud residue, the actual storm having receded far away. Confirming
previous measurements in 1926, some made in 1927 lead
him to estimate the quantity of electricity in an average vertical flash, whether to ground or in the cloud, of 3
kilometers, as 15 coulombs. We may quote from an earlier paper by the same author, in which the polarity of distant, intermediate, and near-thunder clouds is discussed. We give only
intermediate and near storins, although the distances
are much greater than where the “click” occurs with
the flash. But even a t these larger distances it is plain that the increase in the strength of the field is of a pro-
gressive character; and hence in case of very near light-
ning, there might easily be esperienced by insulated animals excitation of the fur or hair. Or again the dog
may have simply heard the hissing caused by point discharges from the leaves of the tree, which under the conditions given would be excessive. The following par-
ticulars have been excerpted from “The polarity of
thunderstorms,’, by B. F. J. Schonland, Proc. Roy. SOC. A Vol. 118, 1928:
Intermediate storms.-Figure 6 : Storm 25, 24/1127, 15 h. 7 m. to 15 h. 12.8 m., about 1% hours after Figure 1.
This is a portion of a record taken on the ball, which had to be lowered to a height of 1 meter above the ground, owing to the negative field of -4,000 v/m prevailing. The following outside
observations were made: 7% m. flash in cloud . . . 8.0 m. double
(Proc. Roy. SOC. A. 1‘01. 118, p. 252.)
flash to ground at 10.4 kms. . . . 8% m. flash in cloud . . . storm getting nearer . . . 9% m. flash to ground at 5 kms. . . . 10% ni. flash in cloud . . . 12 m. flash in cloud a t 5 kms. . . . The storm now came overhead and gave a strong negative field of
- 5,000 to - 10,000 v/m, accompanied by heavy rain, which con- veyed a positive charge to the test plate. The record shows 22
positive and 16 negative field changes. This record was taken on the ball a t the usual height. The storm moved over the station a t about 30 kms. per hour. When distant it gave
7 negative and 3 positive field changes. During this record it
approached from 11.3 kms. (flash a t 1S.8 m.) to less than 7 kms.
(at 20.8 m.) and a t 26% m. a flash to ground took place at a dis- tance of 2.8 kms. The field, which was -530 v/m a t 21.8 m.,
increased to -6,000 v/m as the storm came overhead. The field changes a t 20.8 in. and 21.4 in. were double, $650 and +1,080
followed by -360 and -760 v/m, respectively, at intervals of about
0.5 second. The record shows 7 positive and 3 negative field
changes. This record was taken on the ball
during the approach of the storm and shows 18 negative and 7 positive sudden field changes. Flashes to ground occurred at 18.4 and 20.4 m., which the record shows to have produced positive changes of field.
The ball was lowered to measure the field a t 17.3, 19.6, and 21.4
m. Initially, -109 v/m, the field rose to -420 v/m a the storm approached and later became SO strong as to drive the meniscus
out of the field of view. A t 32 m., after the close of the record, it
had reached - 10,000 v/m, and the nest record, which is shown in Figure 9, was obtained.
A record was made at 17 h. 10 in., when the storm was a t a distance of 11 knis., showed that the steady field was then +40 v/m. Near storms.-Figure 9: Storm 30. Test plate, 17 11. 34 m. to 17
h. 44% m. The distances of 9 of the flashes have been determined frum the thunder marks following the field changes and lie between 3 and 6.4 knis. The test plate was uncovered a t 34yi m. and covered for a few seconds at 37, 37%, 40, and 42% m., the field
varying from -8,800 t o - 10,600 v/m. The largest sudden change of field, at 35.5 m., amounted to f14,SOO v/m and was due to a flash in the cloud at a distance of 4.4 kms. It was followed 3 seconds later by a change of -3,000, r/m. The other negative change, -2,000 v/m, occurred nt 39.6 in. Rain started to fall at 35.5 m., was noted as heavy at 40 m. and
as very heavy from 4lf.4 m. to the end of the record. This is the cause of the upward slope of the record, which indicates an average current of 6 .5 x 10-14 amps./scl. em. from 3434 m. t o 42% m. The
peculiar hump at 43j4 m. is probably due to the heavy rainfall. Flashes to ground occurred a t 37.5 in. (5.4 kms.) and 38.8 m.
(3.0 knis.). The record shows 34 pusitive and 2 negative field changes. Figure 10. Storm 36, 9/2/27. Test plate, 14 h. 6.5 ni. to 14 h. 22 m. The distances of 13 of the flashes have been determined and lay between 2 aud 7. 2 kms. The test plate was covered for a
few secuiids at 7, 10, 11.7, 13.6, 15.1 and 19 m., the field varying
from -11,500 v/m t u -3,500 v/m. The largest sudden change of field amounted to + 11,800 vim and occurred a t 9.4 m., momen- tarily reducing the field to zero. Three small negative changes of field occurred a t 8.7,11.3,and 1.22
m., the first two being due to discharges at distances of 5.4 snd 11.2
kms., .respectively. Two larger negative changes occurred a t 18.6 m. at the eud of a half-minute interval during which a positive
field of the order qf 6,000 vim prevailed. They reduced the field by 14,000 v/m in two jumps of -4,700 and -9,300 v/m about 1%
seconds apart. Rain fell a t 12 m., becoming heavy a t 13 m., and the upward slope of the record indicates that this was positively charged and have
an average current of 4.9XlO-’‘ amps./sq. cm. between 11.7 and 15.2 m. The full history of this storm is as follows: At 12 h. 20 m. it waa approaching from a distance of 40 kms, and gave 90 negative and
4 positive field-changes. At 13 h. 12 m. it was 6 kms. away and
gave rise t u a field of -7,000 v/m which rose to - 16,000 v/m a t 13 h. 41 m. and continued to vary from -S,OOO to -12,000 v/m (ewept for the half-minute of positive field referred to) until 14 h.
27 m. During this half-hour the discharges were between 2.0 and 7.2 kms. off and 72 positive and 5 negative changes were observed. Another record was taken a t 14 h. 27 m., just after fig. 10, when
the active center was reported to be moving away and the flashes were at distances lying between 7 and 10 kms. This record shows 9 positive field changes followed by 8 negative ones and the
sign of the field-changes evidently reversed. The steady field, however, remained betweeu -5,500 and -8,100 v/m for another
31 minutes.
Figure 7: Storm 41, 16/2/27, 16 h. 18 m. to 16 h. 23 m.
Figure 8: Storm 30, 31/1/27.