488 
A typical procedure was to align the ship at a point 
about 6 miles off shore (closer ranges were impossible 
because of reefs lying off the northeastern coast of 
the+ island) and commence a run on a prescribed 
bearing away from the tower. This bearing was pre- 
determined by ship observations of the current wind 
and sea direction. The receiving antennas were 
aligned to maximum signal strengths recorded by the 
Teceivers and secured in this position by clamping to 
the deck. The ship operating speed was usually around 
10 knots, depending on the current sea conditions. 
While the ship was moving on the course, antenna 
changes on the receivers were made every 15 minutes 
for some runs, while antenna heights on the trans- 
mitting end were changed every 2 hours. After making 
several runs using this procedure, results showed that 
there was no discernible diurnal variation of signal 
strength. Therefore, later runs were made using an- 
tenna changes on the transmitting end only at the 
conclusion of the run out. Periodic changes of the 
receiving antenna heights were made in order to 
obtain a complete record of all possible antenna com- 
binations during each run. 
One of the main difficulties encountered in this 
type of operation was keeping the ship on the sched- 
uled course. Deviations from this course were detected 
by means of sudden drops in signal strength on the 
X-band receivers. When this occurred, one of the 
X-band antennas was realigned to give maximum 
signal return and the change in ship’s bearing noted 
by use of a bearing marker attached to the antenna. 
This change was then applied to the remaining an- 
tennas and the ship’s course changed accordingly. 
Additional checks on the ship’s course were obtained 
by means of the radio direction-finding station. By 
using this information, it was possible to detect de- 
viations in the ship’s course without losing any part 
of the record. The ranges of these runs extended up 
to.a maximum of 190 miles. Signals were- usually 
detected out to this range on the lowest X-band com- 
bination and the highest S-band combination of trans- 
mitting and receiving heights. 
Figures 1, 2, 3, and 4 show the plots for one com- 
plete run. It is apparent that the lower antenna com- 
binations on X band produced the highest signal level. 
Signal strengths from higher antenna combinations 
declined proportionately with height. On S band the 
reverse appeared to be true, the 46- to 94-ft antenna 
combination giving the highest average signal level. 
Figure 5 shows a composite presentation of 16-ft 
transmitting antenna to 14-ft receiving antenna. The 
average received signal with this antenna combina- 
tion is 5 to 10 db below the 8- to 6-ft X-band antenna 
combination. 
Figure 6 is a record of all the runs on the 40-1t 
transmitting and 94-ft receiving antenna combina- 
tion. This clearly shows that the highest combination 
available with this setup produced the best results. 
Tt can also be seen that the signal level is considerably 
further below the free space value than is the X-band 
signal for these ranges. 
In order to determine the effect on the signal 
strength of moving the antenna inland, a mobile unit 
consisting of an X-band receiver, test set, recorder, 
and 18-in. parabolic dish were mounted in a truck 
and operated from a gasoline-driven generator. Meas- 
urements during several runs were recorded 14, 1, 
and 1 mile inland from the tower. The antenna heights 
above the sea surface were 25, 50, and 100 ft, respec- 
tively. In one instance, the unit was placed behind a 
hill with the antenna several feet below the top to 
see if transmission over the hill was possible. There 
was a noticeable decrease in signal strength, approxi- 
mately 13 db, but some signal was still recorded. 
Meteorological measurements were taken simul- 
taneously with the inland radio measurements. Kite 
soundings at several points at increasing distances 
inland from the water’s edge were made, and detailed 
soundings on a 50-ft windmill tower about 14 mile 
inland were recorded over a 12-hr period. 
APPENDIX 
Additional meteorological measurements from the 
ship on the leeward side of the island were made to 
determine if duct conditions existed in this area. 
Measurements taken from 2 miles out to approxi- 
mately 20 miles off shore showed that duct conditions 
similar to those found on the windward side of the 
island existed. 
During the final phases of the project, an X-band 
tadar was installed at the base of the tower with an 
antenna height of 6 ft. Measurements of echo strength 
versus range were made on the PC boat to evaluate the 
effect of the duct on X-band radar. Antenna heights 
of the radar were varied from 6 ft to approximately 
90 ft by placing the installation on the truck in much 
the same manner as was done with the receiver in the 
one-way experiment. This was then set up on sites 
overlooking the coastline to sea. The heights at which 
signal strength versus range measurements were made 
were 6, 15, 50, and 90 ft. The variation in the range 
of sea clutter for these heights was also observed. 
Measurements on the leeward side of the island were 
also made with this radar with approximate antenna 
heights of 6, 10, and 75 ft above sea level. 
The maximum range obtained using the PC boat 
as a target with a broadside aspect was 47 miles. This 
Tange was observed with the radar antenna at the 
6-ft level. The maximum range obtained on the ship 
from the 90-ft level was 26 miles. Sea clutter war 
found to vary with the antenna height and wind speec 
Maximum return of 15 miles on sea clutter was ot 
served at the 6-ft level with wind speeds of 20 to 30 
knots. The maximum range at which sea return was 
obtained varied proportionately with height up to 
the 90-ft level. This range was decreased 50 per cent 
with lower wind speeds of 10 to 15 knots. The most 
significant radar datum obtained to leeward of the 
island was the detection of a ship at 45 miles from 
a 75-ft site. 
Meteorological Measurements? 
The description of the meteorological measure- 
ments in connection with the experiment at Antigua 
is divided into three parts, as follows: first, a brief 
general description of the West Indian climate; sec- 
ond, a survey of the low-level soundings; and third, 
a necessarily hurried analysis of the data, with certain 
tentative conclusions. 
The most noteworthy feature of the climate at 
Antigua during the late winter is the persistence of 
one type of weather. This weather condition is deter- 
mined largely by the position and strength of the 
Bermuda high, a large semipermanent high-pressure 
area covering much of the Atlantic from 10 to 30 
degrees north latitude. The northeast trades blow 
around and out of the high’s southern rim. With a 
few exceptions during the period of the experiment, 
the wind direction at Antigua was east-northeast. 
Once, for a period of 3 days, it went around to north- 
northeast and on two separate occasions blew from 
the east. Average daily surface wind speed was 16 
knots, with occasional variations between 8 and 27 
knots. Representative air temperatures varied between 
74 and 78 F, relative humidities between 60 and 80 
per cent. The sea water temperature was reasonably 
constant at 77.5 IF’, with occasional variations between 
76.5 and 78. No significant horizontal gradients of 
sea temperature were found. Precipitation was wholly 
im the form of showers with a maximum frequency of 
occurrence around sunrise. Periods of relatively dry 
weather followed by periods of relatively showery 
weather and accompanying transitions were experi- 
enced. It is felt that these variations were caused by 
fluctuations in the intensity and position of the Ber- 
muda high or by the trough effects ahead of dis- 
sipating cold fronts. 
During the entire period of observations, a simple 
surface duct was found to exist over the water. From 
the second week in February through the third week 
bBy Lt. W. Binnian, U. S. Naval Research Laboratory. 
in March, and again in the first week of April, duct 
conditions were essentially constant. This condition, 
which is called herein the normal condition, is shown 
in five figures. 
Figure 7 shows the average temperature and mixing 
ratio values for a 2-day period plotted against height. 
Curves of daytime and nighttime conditions are 
shown. Soundings were taken every 2 hours. The 
water surface values are derived from measurements 
made on the ship. Considerable difficulty was found 
in obtaining accurate soundings in the daytime due 
to radiation from the warm land in the case of the 
tower soundings and the warm ship in the case of 
ship soundings. As mentioned in Section 4.1, sound- 
ings were possible on the ship only when running with 
the wind. Thus, radiation effects of the ship were 
maximized, especially in the daytime. However, valu- 
able psychrometer measurements were made on the 
outbound runs which showed the air to be consistently 
cooler than the water. On this basis, absolute values 
of temperature in the daytime tower soundings have 
been arbitrarily adjusted. 
Figure 8 is the M curve computed from the tem- 
perature and mixing ratio curves just given. The sur- 
face duct and the small diurnal change in its proper- 
ties are readily seen. An interesting point is the exist- 
ence of a rather sharp discontinuity at the 1-ft level. 
Careful independent measurements were made using 
a number of locations and techniques. All these tests 
confirmed the failure of the sea surface values to fit 
to the smooth curve. It appeared possible that propa- 
gation results might be more dependent on the M 
deficit as computed using the 1-ft value than on that 
computed from the sea temperature. The terms “effec- 
tive surface values” of temperature, mixing ratio, and 
M were therefore established, these being defined as 
the values of these quantities at 1 ft above the water 
surface. Correspondingly, the effective value of M 
deficit is the difference between the value of M at 1 ft 
above the sea surface and the lowest value of M for a 
given sounding, and this effective value should not be 
confused with the total M deficit, which may be con- 
siderably different. This concept will be employed 
later in the paper. 
Another significant feature of the normal sounding 
is the fact that, although the minimum value of M 
is at a height of about 40 ft, the curve does not quite 
reach the slope corresponding to mixed air in the first 
100 ft. Due to the roughness of the few higher sound- 
ings obtained, it has been impossible to determine the 
exact height at which the air becomes mixed. It appears 
to be between 100 and 200 ft. 
The next four soundings show what happened to the 
duct under abnormal synoptic conditions. The major 
variations were (a) relatively low winds, (b) rela- 
tively high winds, (c) relatively dry air, and (d) rela- 
tively moist air. 
The figures which follow are mean or representative 
sample soundings made during each of the conditions 
described above. All were made on the tower and are 
chosen as best illustrating the effect on the Mf curve. 
Figure 9 is a mean curve for low winds. It shows a 
lowering of the top of the duct and a change in slope 
of that portion of the curve lying between 1 ft and 
the top of the duct. No marked change is found in the 
total M deficit. 
With wind speeds greater than normal, the duct 
thickness increased, the effective Af deficit decreased, 
and the total M deficit also decreased slightly. The 
average of 4 days’ soundings during a windy period 
is shown in Figure 10. 
At one time there was an influx of exceptionally 
dry air with winds of normal speed. Figure 11 shows 
the effect on the M curve. The major change is an in- 
crease in the total M deficit. 
Figure 12 is a sample sounding made during a pe- 
riod when the air was relatively moist. The significant 
deviation from the normal soundings is the decrease in 
the total M deficit and the lack of any change in the 
effective M deficit or in the duct height. 
