SITING AND COVERAGE OF GROUND RADARS 99 
The presence of some distinctive echo is of great 
assistance in orientation of the picture, but scope 
distortions and the nature of the echoes cause much 
confusion. It is therefore desirable to be able to 
correct the distortions and to be able to prepare a 
radar map which shows the terrain features likely 
to contribute to the observed pattern. 
The PPI distortions are due to the beamwidth, 
range marker errors, and nonlinear sweep. The 
width of the beam causes objects to appear wider 
than they are, as discussed in preceding sections. 
The range marker errors may be determined by 
calibration with a precision-type calibrator. By 
preparing a cardboard scale to line up with the range 
pips the correct ranges of echoes may be obtained. 
Because the sweep usually takes about 15 usec to 
attain a steady speed the pattern is displaced inward 
with respect to the map. This may be compensated 
in part by adjusting the centering control so that 
at least one of the range markers is moved out 
radially to its true range. The pattern will then 
show a central hole, and the first half mile will be 
displaced from its true position, but the pattern 
as a whole will be more accurate. 
For construction of the radar map it is desirable 
to have topographic sheets of a scale of 1 to 20,000 
which show modern structures. Aerial photographs 
are also useful. Map matching is done by adjusting 
the sweep length and centering controls with major 
changes in scale made photographically. To eliminate 
detail of little interest it is desirable to ink in only 
those contours which correspond to equal increments 
of radar range based on the curved surface of the 
earth. That is, the retraced contour intervals should 
form a sequence of squared numbers (1, 4, 9, 16, 
25 - - - n®), for example, 20, 80, 180, 320, and 500 ft. 
The amount of distortion to introduce into the radar 
map is obtained from the range correction scale and 
the shift of the PPI center. For each azimuth 
considered the map is shifted to compensate for 
the centering error, and the corrected range scale 
is used to lay off distance. 
THE CALCULATION 
OF VERTICAL COVERAGE 
Introduction a 
The computation of vertical coverage diagrams in 
the optical region consists essentially of adding two 
vectors, the contributions of the direct and reflected 
waves, which have been modified by earth curva- 
ture, antenna directivity, etc. The actual computa- 
tion of the contours of constant field strength tends 
to be laborious because of the implicit nature of the 
parameters. The problem may be formulated in a 
rigorous, general manner, but the solution is likely 
to be unwieldy. 
For field purposes where high accuracy is not 
required, a method of computing vertical lobe 
patterns is desired that is direct, does not require 
excessive calculations, provides a simple physical 
interpretation of terrain effects, and is flexible. The 
methods presented here are designed to meet these 
requirements, and the computer may readily accom- 
modate the labor of calculations to the required 
accuracy and the complexity of the problem. 
The path difference of the direct and reflected 
rays, the distance of the reflection point, and the 
vertical angle are functions of each other, while the 
reflection coefficient, the divergence factor, and other 
factors depend on the vertical angle. It is therefore 
desirable to examine the problem in a general way 
to determine what simplifications may be introduced. 
With microwaves the reflecting surface must be 
quite smooth to be effective. Thus by equation (16) 
for the S band and an angle of 1 degree the roughness 
must be less than 15 in. if the reflection is to be of 
much assistance. The rolling character of sea waves 
makes a substantial variation in signal strength so 
that the reliable range is only slightly greater than 
that of the direct wave alone. Also highly directive 
antennas are commonly used with microwave radars. 
These factors reduce the magnitude of the interfer- 
ence effects. The fineness of the structure of a micro- 
wave pattern and the relatively weak reflection 
effects commonly encountered therefore render it a 
useful approximation to deal with the direct wave 
pattern only for most purposes. 
Fire control and searchlight radars normally 
operate at high angles so that they also are mainly 
concerned with the direct wave. The GCE and other 
low-sited radars have their reflection areas within a 
mile of the antenna so that earth curvature may be 
ignored, which means a considerable simplification. 
The case which requires the most careful considera- 
tion is early warning, VHF, high-sited radar which 
is dependent on the reflected wave for much of its 
performance. A careful analysis of all factors involved 
is therefore usually required. Prepared diagrams for 
various heights and wavelengths must be considered 
carefully before being used, as local terrain features 
may radically alter the lobe pattern. 
The accuracy and detail desired and the type of 
site influence the amount of calculation involved. 
With a low-sited VHF radar only a few lobes are 
formed so that the shape and location of the lobes 
‘is of interest. With a high-sited VHF radar the lobes 
are numerous and the gaps are small so that there 
is little likelihood of losing a target in a null area or 
of being able to associate an echo with a particular 
lobe. In this case the envelope of the lobes is of 
particular interest. 
The high-power microwave radars are best suited 
for vital areas with high traffic density. However, 
for most purposes the basic long-range, early warning 
radars used by the ground forces operate in the VHF 
band. They are normally sited high, that is several 
hundred feet and up, in order to secure low lobe 
