RADIO SCIENCE Journal of Research NBS j USNC-URSI
Vol. 68D, No.8, August 1964
Measurement of the Attenuation of Radio Signals by
Jungles
Jack W. Herbstreit and W. Q. Crichlow
Contribution From the Central Radio Propagation Laboratory, National Bureau of Standards, Boulder, Colo.
(Received March 27, 1964; revised April 10, 1964)
Recent interest in jungle communications has indicated the desirability of publishing
quantitative field strength m ea surements made in jungles by the authors during World
War II. The jungle attenuation of radio signals is so great that for satisfactory communications over distances greater than one mile, skywave propagation or elevated antennas
should be employed.
1. Introduction
A part of World War II was conducted in jungle
territory in the Southwest P acific area. In 1943, the
authors were m embers of Dr. W. L . Everitt's Operational R esearch Staff in the Office of the Chief Signal
Officer, Department of the Army, Washington , D.C .
In this capacity they conducted a communications
research study and quantitative field measurements
of radio propagation through jungles in the rain
forests of Panama and New Guinea. The results
obtained are given in Signal Corps reports [Herbstreit and Crichlow, 1943; Bateman, H erbs treit, and
Zechiel, 1944; War Department, 1944] but are not
generally available. Portions of these results are
published at this time because of recent interest in
radio propagation through jungles . It is believed
that t he conclusions r eached are still valid. The
original Signal Corps reports contain measurements
of atmospheric noise levels in jungle areas and applications of both the attenuation and nc,ise measurem ents to the evaluation of the performance of
particular transmitting-receiving set performa nce.
. A more comprehensive atmospheric noise study bas
been made following these original results and is
available [CCIR, 1963a; Crichlow, Disney, and
Jenkins, 1957].
2 . Jungle Attenuation Measurements
In order to determine the magnitude of the attenuation of radio waves caused by jungle growth,
measurements were made in Panama and New
Guinea of field strength versus distance at several
frequencies . The jungles of both Panama and New
Guinea where the measurements were conducted
were dense rain forests with many tall trees and
thick, almost impenetrable low level foliage of palms,
bamboo, and other plant life interspersed. In Panama the average tree height was approximately 50 ft,
and the thick foliage 10 to 30 ft. Generally, in order
to install the equipment at measuring points in the
Panama jungles it was necessary to cut paths through
the undergrowth perpendicular to the propagation
path. In the jungles selected for measurements in
New Guinea, the trees were appreciably taller, ranging upwards to 90 ft ; however, the undergrowth was
not quite as thick and inaccessible as that in Panama.
Low power transportable Signal Corps transmitters on frequencies of 2005 kc/s, 3010 kc/s, 5880)w/s,
5975 k c/s, 44 1/1c/s, and 98.8 Mc/s were used with
vertical whip antennas near the ground on all frequencies except 98.8 Mc/s. At 98.8 Mc/s, a halfwave dipole mounted on an 18 ft mast was used with
both horizontal and vertical polarization. C alibrated field strength meters were used at all frequencies for the m easurements.
In order to obtain a m easure of the radiated
power of each transmitter-antenna combination, field
strength versus distan ce m eaS Ul"em.ents were made in
Panama over flat open ground (golf course) . The
ground condu ctivity and dielectric constant of the
ground were estimated by comparing the measurem ents with theoretical curves of field strength versus
distance computed for several electrical ground
constants.
N ext, the transmitters were taken into the jungle
and measurements of field strength ver sus distance
were made at a number of distances along a radial
path from the transmitters. On 44 11c/s, a continuous recording of field strength versus distance,
which gives a picture of the large variations of the
field strength encountered in the jungle, was made as
the transmitter was carried along a narrow jungle
trail. Similar measurements of field strength versus
distance were made on frequencies of 3390 and 6070
kc/s in the New Guinea jungle usin g Signal Corps
transmitters similar to those used in Panama for
these frequencies. FigUTes 1, 2, and 3 show the
results of the Panama measurements, and figure 4
the New Guinea results . These results are shown
in terms of the field strengths meas UTed for the particular Signal Corps target tr ansmitters used and in
terms of transmission loss [CCIR, 1963b] which is
independent of transmitting and r eceiving equipment.
Figure 1 (a) shows measurements of field stren gth
versus distance at 2005 k c/s over the golf COUl"se and
also in the jungle. At this location the jungle ended
903
at 1.6 miles, and the land was open and clear beyond
this point . It may be noticed t h at the field stren gth
did not fall off very rapidly after the end of the jungle
was reached. The presence of skywave is not indicated since no fading was noticed at any of the dis-
tances at which measW'ements were made. The
dashed line on the jungle cW've of figW'e 1 (a) is an
estimate of the way the field stren gth would have
fallen off at greater distan ces if the jungle had not
ended.
(NOMINAL TRANS M ITTER POWER OUTPUT = 25 WATTS)
(NOM INAL TRANSMITTER POWER OUTPUT = 1.0 WATT)
o
w
~
(0) 2005 Kc/s
120
(a) 44Mc/s
100 i - ; ; : l - i - i-;:===:I=:==:I=:==:r::====;l
SCR 300 - 44 MCIS
VE RT ICA L PO L ARIZAT IO N
GROUNDED HALF WAVE ANTENNA
CALCUL ATED POWER
RADIATED ~ 0. 4 WATTS
o
<t
-
0:
~
~
Goz
~ !"cou
100
~Sc
I,~~(4ft:
C\J
o
0:
60 f---+~~
-
~s
60
ol.L
..... ~ l'I)i)o CO c:
(<"
~
<t
q.~
20
f-
(f)
o
W
o
0.03 0.05
1
1
0.1
0. 2
fZ
60 <t
f-
-
0:
60 w
>
-
0
w
\
0
Z
\
10 0 0::J
a:
\
(!)
1.0
0.5
2.0
5.0
DI STA NC E I N MILE S
w
~
o<t
"
0:
(f)
f-
Goz!"
J
~
100
~
'i:
Q
0:
o
l.L
~
u.,~~ ~.
I
<t
40
'0
I
I
f(!)
20
(f)
a
...J
w
o
l.L.
SC R 69 4 - 5975 K cis
VER T ICAL POLARIZATION
16' WH IP ANTENNA
CALCULATED POWE R
RADIATED ~ 10 WATTS
0.03 0.05
0.1
0.2
DISTANCE
\
r ~~ --- 0'"
z ~
i'\c W~E_
,
~~
1.0
<t
130
~
:;:
- I Owf-
...J
z
'i:
Z
~w
¢
-=- 400 ~
l.L
E
=
1
l.L
...J
l.L
150 I<t
~~
- 300 w
0.1
0.2
0.5
1.0
2.0
0
5.0
DISTAN CE IN M ILE S
a:
(!)
100
a:
0
SCR 300 - 44 Mc/s
VERTICAL POLAR I ZAT ION
GROUNDED HALF WAVE ANTENNA
CALCULATED POWER
RADIATED ~ 0.4 WAT TS
CD
I
60
l.L.
50
...J
Z
(!)
70
z
w
0:
(f)
(f)
(f)
f-
:::<
(f)
Z
<t
0:
f-
60
0
(f)
(f)
:::<
(f)
z
0
90
...J
w
l.L
(f)
(f)
0
I
f-
0
z
::J
( b ) 44 M C/S
'0
w
a
0
::l
CD
W
>
<{
- 500
0
a:
<t
~
w
<{
f-
aJ
<{
z
z
fZ
(f)
W
(f)
40
<t
0:
f-
110
90
20
SKY
", 'T
0.5
~
' '~o
' '.,
'9&~\
CD
,,- 70
' , -~ vi'
~
w
W
0:
f-
~~ 0
'
OC
G(<" ~.
>
o
Z
~O'~<I
<' ;-~
"
,,
20
a:
'0
0
- 50
0
>
0
(f)
(f)
'k
60
aJ
- 30
' ~s '"C<I<c
l'.
Vv~'\
60
0
4fc
"
4su
!'(
'?c4t,
..... , "1-;-
~
110
'" S1?"<"
:;;
0:
.1
w
I
l.L.
~ufi'sc
'" , <1;-('0
0
a:
0
5 9 7 5 Kc/s
(b )
120
, 4:
° - 200.03 0.05
f-
(f)
o
C<;
, <c:
40
w
<t
u
...J
l.L
.....>20
,
...J
\
90
<t
Z
_ 0<:5
SCR 694 - 2005 Kcls
VERTICAL POLARIZATION
16' WHIP AN TE NNA
CALCULATED POW ER
RADIATED ~ 0.2 WAT TS
(!)
(f)
z
- ....
w...,
J'~
I
0:
"
40
,, -
<;'
l.L W
1--",
40
'0
w
=----
<;,J'v-j> ~
>
-
u~"
... s/I'I)
~
w
Z
4zc:
0-
'VG'
I
f-
70
60 I - - - - t - - t - - - - t --
,
' .... olO'2
uZ 4r
vG<
60
aJ
20
. -
.........
50
2.0
- 600
110
~
-
130
:: Iz
-= 500 t;:j
- If-Q
W
~l.L
~ 400 'j ~
130
150
w
-20
0.03 0.05
5.0
IN MILES
: 300
0.1
0.2
DI STANCE
1. Measurements of field strength or transmission
loss versus distance over flat ground (golf course) and through
jungles of Panama.
FIGURE
0.5
1.0
2.0
5.0
IN M IL ES
M easurements of fi eld strength or transmission loss
versus distance over flat ground (golf couT.~e) and through
jungles of Panama.
FIGU RE 2.
904
J
THR OUG H JUNGLES OF NEW GUINEA
(NOMINAL TRAN SMITTER POWER OUTPUT = 25 WA TTS)
(NO M INAL TRANSMITTER POWER OUTPUT = 50 WATTS )
12 0 ,---(:..::O~)-=9-=8~.8=-.M_e~/~s_-r
V.::.
E_
RT_I--=C_A.;::
L_P_0:...:L=-ArR_I.::.
Z _ATTI-"O_N_--,
o
w
~
<1
Co
a:
-
<::",
I-I--
120 ,..----....0---.----.----.---'-(a",)-,3""3r9,,,,0,---kr c,--l_s-,----,--,----,
AN /T RC 1- 98.8 Mel.
CALCULATED POWER RADIATED ~ I WATT
Ci
40
C'o
100
100 I---"tt-+--t-='~
u-?
<1
~
S(-
1,.(-
-
<1s
v-?,
a:
o
.... ~;,
'1/.
0.' 0'
't-G'
,, "" ....
'('
/0"
~-
60
<;
w
<;,
ro
<1
.... C'v-
<'1-
'
' , ."
40
"0
1
601--_+--
80
"'1. . ."'~o
~ IT~~ foeo
o>
CD
60
("1,.~
80
LL
4 01--_+-_+-~
100
' <1,>,
, ("
I
I-t?
Z
-
20
W
a:
--
I 0- >--w
-
4 00 §! LL
I--
If)
o
-'
w
-
120
::>
.::L 20
o
6
w
w
~
>--w
:
~
- I~~
140
OW
0
0.03
LL
"
-= 500 z
E
iii
w
o-'
"-
.... '?l<
LL
oz
I
a:
o
a:
I-
~ 1001---+--r--t-='~
If)
If)
120 ,------,----,-------,-----,---,------,-------,
o
o
-'
w
z
LL
-'
w
~
o
SO
<1
80
0
If)
If)
100 1-__1---"-'.....
a:
If)
:;:
I--
If)
~
70 ~
z
~
~
120,---~-c--c--.~~~~~~-,---,--,--,
w
LL
( b ) 98 .8 MC/S - HOR IZONTAL POLARIZATION
DISTANCE IN MILE S
1
I
<1
5.0
00L02==0:r.0=5=:r0.1==oI.2==:J0~5- JI.-=-O- J2.0=-----:c5LO--cILo-..)20
'0
-'
DISTAN CE IN MILES
• ~ TEST OVER JUNGLE ROAD
o ~ TEST OV ER JUNGLE PATH
USI NG 16' WHIP AND LO NG
WIRE (AN 160) ANTE NNA
6 01---+--r--r~~
-1---j---1--~-- r-~
I--
80 1---1--~.-_+~~__1--~-_+--~
a:
40
o
LL
90
E
1
60 1-- + - - - + -'
20
w
o>
ro
<1
110
0.05
40 1---1- - 1 - - _ + - -
"0
1
I
w
130
0.2
0.5
LO
2.0
50
10
20
4. M easurements of field strength versus distance
through jungles oj N ew Gui nea.
::_ I i5>j::w
a:
I--
t:
- 400 '$
I w- - !z-
=
o
LL
F I G URE
20 I----II--+--t---+----'~ - 500
If)
-'
w
0.1
DISTANCE IN MI LES
CD
I-t?
Z
AN /TRC 2 - 6070 k c/s
• ~ TEST No lOVER JUNGLE ROA D
o , TEST No.2 OVER JUNGLE ROAD
USING LONG WIRE
( AN 16 0) ANTENNA
0L-~~_
_ L_ _~_ _~_ _~_-~3~0~0_w
0.03
0.1
0.05
0.2
OS
1.0
ISO
_ ___!
2,0
5.0
DISTANCE IN MILES
TRANSMITTER ANTENNA HEIGHT ~ 18 FEET
RECEIVER ANTENNA HEIGHT ~ 10 FEET
3. Measurements of field strength or transmission loss
versus distance over flat ground (golf course) and through
j ungles of Panama.
FIGURE
FigUl'e 1 (b) gives the r esults of meaSUl'ements on
597.'5 kc/s. At this frequency, fading was noticed at
a little over 1 mile from the transmitter, and the field
strength did not fall off at distances greater than 1
mile. This indicates that skywave was becoming an
appreciable part of the received signal even though
a short vertical antenna was being used on the
transmitter.
The m eaSUl'ements at 44 Mc/s show that the elevation of the intervening terrain as well as the jungle
growth have a marked effect on the strength of the
r eceived signal. Profiles of the actual terrain elevation, without regard to the tree heights, are plo tted
on figure 2 below the corresponding field strength
m easurements. Figure 2 (a) shows the point to
point measurements made in dense jungle, and figure
905
2(b) shows data taken along a jungle trail. It may
be seen that the field strength roughly follows the
ground profile. Both the maximum and minimum
field strength measurements at each location are
plotted on figure 2, and the solid line drawn through
the measurements is through the maximum values.
The dashed lines through the jungle measurements
on figure 2 (a) al1d (b) are estimates of the field
strength versus distance if the jungle were level.
It is interesting to note that the same dashed line
can be drawn through the point-to-point measurements and those made along the jungle trail. It is
found that in a particular location a variation of as
much as 4 to 1 in field strength is obtained simply
by moving the receiving antenna a few feet.
On 98.8 Mc/s the effect of the terrain elevation is
also very marked, and the field strength again
roughly shows the jungle profile. Considerable difference exists in both attenuation and field strength
variation between horizontal and vertical polarization in the jungl e. The vertically polarized fields
are attenuated more rapidly and show more variation
in strength than do the horizontally polarized fields .
Fio-ure 3 shows that the horizontally polarized fields
ar; approximately 15 dB less than vertically polarized
fields at a distance of 1 mile. Because of the magnitude of jungle attenuation found at low antenna
heio-hts the effect of elevating the antennas above
theOjun'gle growth was investigated for a frequency
of 44 Mc/s and a distance of 0.4 mile. These
measurements indicate that the received field strength
increases approximately 12 dB with the transmitting
antenna raised to an elevation of 50 ft, the average
height of intervening jungle growth being approximately 50 ft . When both antennas are raised into
trees one 50 ft and the other 70 ft above the jungle
floor ' an increase of approximately 23 dB is obtained
over' that when both antennas are essentially at
ground level.
Figure 4 shows similar measurements of field
strength versus distance made. in the rain for~st
jungles near Oro Bay, New Gmnea, on frequencIes
of 3390 kc/s and 6060 kc/s. The jungles at this
location were said to be typical of the rain forest
type of jun~le found in most of New Guinea. The
undergrowth in this jungle was not as dense as the
undero-rowth in the Panama jungle through which the
previoous measurements were made; however the high
trees were considerably taller than those in Panama.
3 . Effective Range of Communications
To determine the effective range of particular
transmitting and receiving systems, it is necessary
to know the magnitude of the required field strengths
in each case; and these, in turn, depend upon the
levels of the received external noise and the noise
generated in ~he receiver, as well as t~e kinds of
signal modulatIOn used. Up to frequenCIes the order
of 30 Mc/s, atmospheric noise is expected to be the
controllin g factor at these low latitudes where
jungles are encountered, and at higher frequencies
galactic or set noise is expected to limit communica-
906
tions [CCIR 1963a, 1963c]. Both the received signal
level and the received atmospheric noise level are
expected to depend to a certain exten t on the
polarization and directivity of the receiving antennas, as well as their heights above or below the
height of the jungle growth. An analysis of these
factors in terms of the signal strength required for
satisfactory communication for the particular sets
used was included in the original Signal Corps
reports. This analysis indicated that with the
equipment used in the tests, communication via
groundwave through the jungle for distances greater
than approximately 1 mile is virtually impossible.
Because of the high rate of measured groundwave
attenuation with distance, considerable increase in
power would be required to extend this groundwave
range appreciably. However, it was demonstrated
that satisfactory communication by means of skywave transmission could be obtained at greater
distances by using long wire skywave antennas when
transmitting on frequencies reflected by the ionosphere. It should be noted that when using skywave
frequencies for jungle communications, interference
to and from other stations at considerable distances
is possible. Ionospheric predictions are useful for
obtaining optimum results.
4 . Conclusions
Attenuation due to dense jungle growth is so great
that for communication over distances greater than
approximately 1 mile the groundwave which is normally employed for these ranges is practically useless. Jungle communication greater than 1 mile
may be obtained by elevating antennas of VHF sets
either by raising them into trees or by using hilltop
sites. At VHF, horizontal polarization was found
to be preferable because of its lower absorption rate.
Ranges greater than 1 mile using medium or high
frequencies were found to require skywave transmission using antennas radiating energy almost
vertically on frequencies reflected by the ionosphere.
5. References
Bateman, Ross, Jack W. I-Ierbstreit, and Robert B. Zechiel
(1944), Measurement of factors affecting radio communication and loran navigation in SvVPA, Report No. ORB-2-4,
Office of the Chief Signal Officer, Washington, D.C.
CeIR Report 322 (1963a), World diRtri bution of characteristics of atmospheric radio noise, Documents of the Xth
Plenary Assembly, lTU, Geneva; revision of R eport 65,
Revision of atmospheric radio-noise data, meeting in Los
Angeles 1959.
CeIR Recommendation 341 (1963b), The concept of transmission loss in studies of radio systems, ITU, III, Geneva.
CeIR Recommendation 339 (1963c), Bandwidths and signalto-noise ratios in complete systems, ITU, III, Geneva.
Crichlow, W. Q., R. T. Disney, and M. A. J enkins (1957) ,
Quarterly radio noise data, NBS Technical Note 18-1 and
su pplements.
I-Ierbstreit, Jack W., and W. Q. Crichlow (1943), Measurement of factors affecting jungle radio communication,
Report No. ORB-2-3, Office of the Chief Signal Officer,
Washington, D.C.
War Department (1944), T echnical Bulletin SIG 4.
(Paper 68D8-388)