SOUND PROPAGATION IN THE ATMOSPHERE 
OBSERVATIONS AND THEIR INTERPRETATION 
Natural Sound. Microbarographs of sufficient sensi- 
tivity record a background which consists of the effect 
of air currents and of natural pressure waves which are 
sometimes especially clear on calm winter days. There 
are several types of natural sound waves. At Pasadena, 
California, the most frequent are sinusoidal waves with 
variable amplitudes and periods of about 14 to 5 see 
[3, 18] (Fig. 7), corresponding to wave lengths of about 
Prine umn Nanette re ieee i 
a ana anil oat KA monic cyA y A antler 
ian neo ane Ll ceehdhn triateeiedinaieasalil ant RON ie M 
eSeerncsent aivaa ra enerentlliNne menitelimatnnabatnd bse rptP hae fehl tan 
bog 1 MIN. 
5 
f 
| sicoanaaemaegcies cae etal cerca Rann cain” ra Se cont ia clteenirtnen aneertecilaraad 
= ————————— 
Reerie eee , 
pb 10 SEG 
. [ny 
SNe ee oe nonin reer eh aaron seater endinenmmmemammianad rescore 
Se "aaeme mnmaarnt ten 
eS eS set ate TANS Permanent 
ie MIN. 
Fig. 7.—Pressure waves recorded by Benioff microbaro- 
graphs at Pasadena. (a) January 6, 1939; these coincided with 
the largest surf waves in several years near Scripps Institute 
of Oceanography at La Jolla; no meteorological element had 
unusually great values within several hundred miles; the two 
records correspond to two instruments about 30 m apart. (6) 
Pressure waves recorded on December 27, 1940, with three 
microbarographs forming a triangle with sides of 290, 320, and 
264 m, respectively; these waves are frequent during the winter 
and arrive usually from the SW. (c) Pressure waves with longer 
periods, recorded on January 8, 1939, by two instruments 120 
m apart. (After Benioff and Gutenberg [3].) 
100 m to 1500 m. In general, they are largest in winter. 
During the three winter months of 1940/41, three micro- 
barographs were operating (Fig. 7b) which permitted 
the calculation of the direction from which the waves 
arrived; all came from southwest to west (azimuths 
between west and 40° south of west). In all instances 
of large amplitudes a low-pressure area was situated off 
the coast of southern California; as soon as the low- 
pressure area passed the coast, the amplitudes decreased 
rather rapidly. The waves arrived almost horizontally, 
but this is to be expected (even if the source is rather 
high in the atmosphere) due to the curvature of the 
rays. No connection with microseisms could be found. 
Similar air-pressure oscillations have been recorded in 
Christchurch, New Zealand [2]. Their periods were be- 
tween 4 and 10 sec, and they showed similarity to 
microseisms, but a time lag of pressure oscillations, up 
to 200 sec, was suspected. Baird and Banwell believed 
that the two types of waves are independent phenomena 
but possibly have the same cause. Polli (1949) found 
similar results in Trieste. Observations of these pressure 
waves at other locations are very desirable. 
Occasionally, short groups of pressure oscillations 
were recorded at Pasadena with periods of about 14 to 
1 sec and occurring at intervals of about 20 see (Fig. 
7a). Their direction (from SW), their regular coinci- 
ByAl 
dence with high ocean waves at the coast, and the lack 
of other unusual phenomena make it likely that they 
are “‘sound” from surf. The causes of longer pressure 
waves which were recorded occasionally (Fig. 7c) could 
not be found due to the scarcity of the observations. 
Sound Waves through the Troposphere. When un- 
usually strong sound waves from distant explosions 
were first heard, attempts were made to explain the 
observed zones of audibility and of silence by com- 
binations of temperature, lapse rates, and change of 
wind with elevation [20]. This possibility was dis- 
proved by the occasional occurrence of instances in 
which the abnormal zone formed a full ring around the 
source of sound, separated from the normal zone of 
sound by a zone of silence. However, there are instances, 
not infrequent, of strong sound in relatively small 
areas, which are due to meteorological conditions in 
the troposphere. It follows from equations (20) to (22) 
that either a temperature inversion, or a certain in- 
crease of wind with elevation (depending on the lapse 
rate), or both jointly, may produce zones of strong 
sound. It is of interest to note that in the attempted 
explanations mentioned above an increase of wind with 
elevation by 4 or 5 m sect km was usually assumed 
together with the typical temperature lapse rate, after 
considering a variety of conditions; such requirements 
follow now from equation (22). Figure 11a shows rec- 
ords with clear dispersion. Schulze [27] explained this 
dispersion as a consequence of the fact that the velocity 
changes with elevation, and because the energy of 
longer waves is propagated in a thicker layer than that 
of shorter waves. 
Strong audibility of certain sounds (e.g., of trains) is 
occasionally used by laymen for short-range weather 
forecasting. This possibility is based on the fact that 
the repeated occurrence of a certain unusually strong 
sound from the same source will frequently be a con- 
sequence of similar meteorological conditions at a given 
place. 
Sound Waves Through the Stratosphere and Ab- 
normal Audibility Zones. In 1903 an accidental ex- 
plosion of dynamite in Westphalia resulted in sound 
waves which were heard far away, beyond a zone of 
silence. The observations were investigated by von dem 
Borne who believed that an increase in the percentage 
of light gases with elevation in the atmosphere causes 
an increase in sound velocity. (For historical data and 
bibliographies see [28, 10].) The details of these abnor- 
mal zones were studied in instances of artificial ex- 
plosions, especially in Germany [16], of gun fire, and of 
accidental explosions, using ear observations (Fig. 8) 
as well as instrumental records. There are instances 
where the abnormal zone forms a complete ring around 
the source, and others in which only parts of a ring 
with audible sound were developed. Frequently, parts 
of a second or third ring of abnormal audibility were 
found (Figs. 8 and 9). 
In Europe and in Japan, the radius of the ring as 
well as the distance of the largest sound intensity 
shows a yearly period (Fig. 10) with a minimum dis- 
tance of about 100 km late in winter or early in spring, 
