268 RADIO WAVE PROPAGATION EXPERIMENTS 
propagation. These ridges simulated waves, and their 
wavelength D lay between 0.6 and 1.2 m, whereas the 
double amplitude h varied between 5 and 15 cm, and 
the wavelength of the radiation used wag 9 cm. 
Tables 8 and 9 give the reflection coefficients of 
the uneven ground when the direction of propagation 
is at 45° with the ridge and along or across the ridge 
system. The tables also give an estimation of the re- 
flection coefficient of level ground of the same moisture 
content as that under test. 
The results given in these tables show how a rela- 
tively small irregularity in the ground surface is 
sufficient to prevent regular reflection. The reflection 
coefficient becomes erratic when it has fallen below 
a value of about 0.1. The values given for level ground 
refer only very approximately to the state of the 
ground in the ridged condition. Since the measure- 
ments extended over several days, those relative to 
level ground may not correspond necessarily to the 
same degree of moisture as those referring to the 
ridged ground. The level reflection coefficients in the 
two preceding tables differ from each other because 
those of Table 8 refer to drier ground than: those of 
Table 9- 
MEASUREMENTS OF THE REFLECTION 
COEFFICIENT OF LAND AT CENTI- 
METER WAVELENGTHS, CARRIED 
OUT AT NATIONAL PHYSICAL 
LABORATORY: 
Experiments have been made on the reflection and 
absorption of radio waves in the S-band of wavelengths 
by workers in the National Physical Laboratory in 
England. The reflection coefficient has been measured 
at angles of incidence to the vertical, of 80°, 68°, 54°, 
and 44° on level ground, fresh water, sea water, uneven 
ground, ground covered with vegetation, and ground 
covered with 1 in. mesh wire netting. The absorption 
°By W. Ross, British Central Scientific Office, Washington, 
D.C. 
in soil, fresh water, sea water, and 1 in. mesh wire net- 
ting has also been measured by a-laboratory method. An 
interim report! gave some of the salient results 
obtained on ground reflection. 
The main conclusions which have been drawn from 
the results obtained are given below. 
1. Specular reflection can be obtained only from a 
very level surface, with little or no vegetation on it. 
The electrical constants of such surfaces are given in 
Table 10, from which the coefficient of specular reflec- 
tion can be deduced for the angle of incidence and 
state of polarization concerned. 
2. The reflection coefficient decreases with uneven 
ground and is reduced to a value of about 0.2 by in- 
equalities of level of about one wavelength. This con- 
clusion is based mainly on a series of experiments in 
which the ground was raked into a series of ridges 
resembling waves, which could be either in, across, or 
at an angle to, the direction of transmission. Similar 
results were obtained when a large sheet of wire net- 
ting was similarly disposed in a series of waves. 
3. Vegetation reduces the reflection coefficient, in 
general, and when about 2 ft high causes a reduction 
in reflection coefficient to a value of about 0.2. An in- 
teresting exception was found when level ground was 
covered with vegetation less than half a wavelength 
high (about 1% in.) when the reflection coefficient 
with vertical polarization increased slightly with high 
angles of incidence over that obtained with level 
ground. 
Tasie 10 
Dielectric Conductivity 
Nature of surface constant mhos /m 
Bare sandy loam, very dry 2 3 xX 10? 
Bare sandy loam, saturated with water 24 6 xX 1071 
Turf with grass very short (cricket 
wicket), dry 3 5 x 103 
Turf with grass very short (cricket 
wicket), wet 1 x 1071 
Fresh water 80 2 
Sea water (4% salt solution) 80 5 
