308 RADIO WAVE PROPAGATION EXPERIMENTS 
Ground Clutter 
The ordinary ground clutter consists of echoes from 
a variety of types of targets: earth, rocks, trees, 
branches, bushes, leaves, grass. Our present conception 
is that the fluctuation in ordinary ground clutter arises 
from the motion of leaves and branches in the wind, 
changing the phase patterns in a manner somewhat 
similar to that for random scatterers. There will in 
addition be a relatively steady signal from fixed ob- 
jects such as rocks and tree trunks. 
We have obtained much qualitative evidence for 
this picture, but it is difficult to obtain quantitative 
fe) 5 10 15 20 25 30 35 
WIND SPEED IN MPH—= 
Ficur 3. Fluctuation in signal from Blue Hills versus 
wind speed. S band, 10 a.m. April 24, 1944 to 11 a.m. 
April 25, 1944. 
° 0.5 10 eA) 2.0 25 3.0 3.5 
=-— 
To 
Figure 4. First probability distribution, Baker Hill; 
Maine. Wind speed 25 mph. Curve: theoretical, fixed 
to random signal =1 db. J) = average intensity. 
Histogram: experimental results. Film 103, 3,000 
pulses, S band. 
data because the wind speed at the target is not usually 
known. Fortunately, in the Boston area the largest 
ground signal is due to the Great Blue Hills, which 
is the site of the Blue Hill Observatory. It is thus 
possible to obtain data on the wind speed at the target. 
We monitored the signal from Blue Hills for a 24-hr 
period in April of 1944. Movies were taken of the 
A scope at regular intervals. During the period of 
observation the wind speed varied between 30 mph 
and dead calm. To interpret the data a somewhat crude 
parameter was defined as a measure of the amount of 
fluctuation. The change in the signal strength from 
one frame to the next (0.06 sec) was measured and 
averaged over 200 frames. 
to) 0.5 1.0 15 20 2.5 30 
Ficure 5. First probability distribution, Mt. Penobscot, 
Maine. Wind speed 10 mph. Curve: theoretical, fixed 
to random signal = + 7.2 db. I) = average intensity. 
Histogram: experimental results. Film 82 (16 fr per sec), 
400 frames, S band. 
This parameter was then plotted against the wind 
speed as shown in Figure 3. There is a quite good 
correlation between the amount of fluctuation as meas- 
ured by this parameter and the speed of the wind. The 
fluctuation is of the order of 0.2 db at 0 mph, which 
is almost as good as our steadiest signals. At the other 
extreme the fluctuation is about 3.4 db at 30 mph. 
There appears to be a rather sudden jump in the fluc- 
tuation at a wind speed in the neighborhood of 20 
mph. This jump has been observed at other seasons of 
the year and is believed to be rather general. It is sig- 
nificant that the wind speed at which the jump occurs is 
roughly that at which large branches and small trees 
begin to move as a whole. 
The theoretical description for a simple picture of 
ground clutter consisting of an assembly of random 
scatterers (leaves, grass, etc.) plus a fixed signal 
(rocks, trees, trunks) is not difficult to work out. When 
the proportion of steady signal is small, the first prob- 
ability distribution closely resembles that for purely 
random scatterers. For a large ratio of fixed-to- 
tandom signal the amount of fluctuation is greatly 
reduced, and the first probability distribution tends 
to a Gaussian curve about the average intensity. This 
is illustrated in Figures 4 and 5. Figure 4 is a plot of 
the experimentally determined first probability dis- 
tribution for a signal from heavily wooded terrain on 
S band at 25-mph wind speed. This has been fitted 
by a theoretical curve for a ratio of fixed to random 
signal of —0.1 db. 
Figure 5 shows the distribution for a similar type 
of terrain but for a wind speed of 10 mph. Here the 
results are fitted to a curve for a ratio of fixed to 
