482 
urements, and (2) forecasting and analysis. 
All meteorological measurements were made with 
varying versions of the psychrograph, earlier models 
of which are completely described in reference 2. 
This instrument measures wet and dry bulb tempera- 
tures as a function of height, using the electrical re- 
sistanee thermometer principle. From these measure- 
ments the M curve is constructed. 
The meteorological soundings were made in the 
Massachusetts Bay area with psychrographs carried 
by two aircraft, by captive balloons operating from a 
boat and from two fixed land stations. The boat 
operated in the Bay and out to about 100 miles off- 
shore, while the aircraft operated as far as 170 miles 
offshore. It should be mentioned that aircraft sound- 
ings of this type are rather hazardous, since they in- 
volve descending to altitudes of approximately 20 fi 
at large distances from land. 
The fixed meteorological stations were located at 
Duxbury and Race Point. The Duxbury location was 
chosen to place the sounding station near enough to 
the shore to obtain a representative sample of the air 
just leaving the land. 
The Race Point meteorological station was located 
at a position to allow soundings at the water’s edge 
on the westernmost extremity of the top of Cape Cod. 
The primary purpose of this station was to measure 
the characteristics of the air after it had been sub- 
jected to the influence of the ocean surface between 
the mainland and the station. This location allowed 
measurements over a range of wind directions of ap- 
proximately 180°, but no relevant soundings could be 
made when the wind had an easterly component, since 
the air would have had a land trajectory for at least 
a short period. It was necessary to take all soundings 
very close to the water’s edge to prevent solar heating 
of the beach from influencing the bottom of the meas- 
ured M curve. At both Duxbury and Provincetown 
soundings were taken on a prearranged schedule 
which, when possible, involved both day and night 
operation. Soundings, surface wind velocities and 
hourly observations of sky conditions, etc., were made 
at both Duxbury and Race Point. The water tempera- 
ture was also measured at the Provincetown station. 
A 60-ft pole was erected at Race Point carrying 
four anemometers, four psychrographs, and a wind 
direction indicator. The original scheme involved con- 
tinuous recording of temperature, humidity, and wind 
speed at four levels by means of the instruments on 
the pole, but unfortunately a large sand bar formed 
in front of the pole soon after it was erected and 
caused sufficient disturbance of the air in the lowest 
levels that such measurements were not feasible. The 
psychrograph, anemometer, and wind direction in- 
dicator at the top of the pole continued to be useful, 
however, and provided the continuous information 
recorded by the station. 
A 50-ft boat, the Wanderer, was used for making 
measurements in Massachusetts Bay. The psychro- 
graph used for measurements from 2- to 48-ft eleva- 
tion is attached to a cable running between a boom 
extending outward from the side of the ship and an 
extension to the top of the mast. A similar psychro- 
graph was used for soundings to higher levels, operat- 
ing from the winch at the rear of the boat. Further 
essential meteorological information was provided by 
the boat in frequent measurements of surface water 
temperatures in Massachusetts Bay. A direction-find- 
ing loop was used to determine position at great dis- 
tances off shore 
RADIO AND RADAR TRANSMISSION 
MEASUREMENTS? 
The purpose of this paper is to describe the results 
of a rough preliminary analysis of the transmission 
experiment. A thorough analysis must await the com- 
pletion of the meteorological study, since the trans- 
*By Pearl Rubenstein, Radiation Lahoratory, MIT. 
APPENDIX 
mission depends directly upon the meteorological 
conditions over the path of the radiation. The em- 
phasis here will therefore be mainly on the strictly 
radio data with only qualitative reference to the 
meteorological information. 
One-Way Transmission 
The values of the transmitted powers, antenna 
gains, and receiver characteristics were chosen so as 
to make the standard signal level, as computed for the 
receivers at the top of the tower, well above the mini- 
mum detectable level and the minimum level at 
which the automatic frequency control [AFC] and 
AFC search are effective. Sufficient compression was 
used to give a range of about 60 db for useful recep- 
tion, which had been expected to be enough for the 
variations due to atmospheric conditions. It turned 
out, however, that additional range was needed, es- 
pecially in the direction of greater signal strengths; 
to accommodate additional received power, attenua- 
tors were inserted in the lines. Thus the actual range 
of values observed is at least 90 db at the microwave 
frequencies and 40 db at 117 me. 
Siena Tyres 
Figure 2 shows that the types of signal observed 
at the microwave frequencies (S and X) are not 
essentially different from those observed in previous 
tests on a shorter path. The first type is high signal 
on the average, well above the standard level, with 
roller fades which may go down to the minimum de- 
tectable level and with periods of 2 minutes to an 
hour or so. These periods are generally shorter at 
any time on X than on S band. When this type of 
signal is present on S band it is almost invariably 
present on X band also and on both paths. It always 
occurs simultaneously on the high and low receivers 
at any frequency. 
The second type is high and steady. Its level may 
be anywhere from 5 to about 30 db above the stand- 
ard, generally higher on X band than on S band. 
Most of the time this type of signal occurred simul- 
taneously on S and X, but there were some occasions 
when the S-band signal was of the high and steady 
type while the X-band signal became of the first 
type, high with roller fades. 
The third kind of signal is about standard and 
fairly steady. (This may be a limiting case of the 
high and steady variety.) It does not necessarily 
occur on both frequencies and on both high and low 
receivers at the same time. 
The fourth type is standard on the average, with 
scintillation of more than 10 db. The preliminary 
analysis has not revealed the reasons, or any correla- 
tions, for the difference between this and the preced- 
ing type; it is certainly nothing obvious, such as wind 
speed, for example; and it may occur on either fre- 
quency when the other is steady. 
The fifth type is the “blackout,” below standard 
and variable. This signal type is strongly scintillat- 
ing. It occurs simultaneously on both frequencies, 
both paths, and on high and low receivers (except 
possibly for low X. where the difficulty mentioned 
above of determining an average value of something 
very low on the scale is important). 
Figure 3 shows the signal types observed at 256 
em. These are distinct from those observed at the 
microwave frequencies not only in appearance but 
also in times of occurrence. In general no relation has 
been found to exist between the types at the two 
frequencies although on rare occasions such a relation 
is indicated ; indeed the type may remain constant on 
one and change on either of the others. Steady signal 
is most frequent at 256 cm, but the other types shown 
also occur fairly often. Variations of 30 to 40 db 
overall take place, and the vartatienssmay be fast or 
slow. 
STaTIsTics 
A fairly detailed statistical stuuy has been made of 
the S and X signals at the top level. These were 
chosen because they were available for the longest 
periods, and because they gave the most reliable re- 
sults (because of the receiver characteristics the re- 
lation of the standard to the minimum detectable 
level was most suitable). The other microwave records 
gave similar results. As for the 256-cm transmission, 
the most important result was that the signal level 
was above the minimum detectable very nearly 100 
per cent of ‘the time, although fades to this level were 
fairly frequent. If a choice had to be made of the 
most reliable frequency for transmission over the 
circuit, there would be no question in the choice of the 
longer wavelength. 
The statistics available on K band are very similar 
to those on X band as far as can be determined. 
Signal levels less than about 20: db above standard 
cannot be detected on the K band. 
The study was made of the average signal level on 
a weekly basis; it showed marked differences from 
week to week, depending upon the specific weather 
situation. For purposes of the statistics a range of 
values around the standard was included in the 
standard signal (allowance for scintillation, tides, 
ete.). This range was taken as +5 db on S band and 
+10 db on X band, values determined by inspection 
of the entire record and thought to give comparable 
results. 
The most interesting result of this analysis was 
the discovery that standard signal occurs extremely 
rarely over this path. High signal is most frequent; 
depending upon the wavelength and the season of the 
year, substandard and standard signal occur less fre- 
quently. In the summer no significant frequency de- 
pendence was observed in the statistics. Some typical 
weeks gave the figures shown in Table 1. 
As the season progressed to the fall, however, sev- 
eral related trends became apparent: (a) the increas- 
ing incidence of standard signal, especially on S 
band; (b) the increasing incidence of high, steady 
signal, especially on X band, with the level higher 
above the standard on X than on 8; (ce) the fre- 
quency effect on the incidence of above-standard 
signal indicated in (b); and (d) the decreasing oc- 
currence of substandard signal. These trends are illus- 
trated in Table 2. 
No diurnal effect was found in the signal except 
under some very special circumstances. Not only was 
no such trend apparent upon visual inspection, but 
also an analysis of the material by 6-hour intervals 
confirmed appearances. 
CorRELATIONS 
In addition to the statistical study, another type 
of analysis has been made to look for correlations 
between the variations of signal strength with fre- 
quency at a given location or with height at a given 
frequency. Figures 4 to 6 show some typical graphs 
of such correlations, each point representing average 
hourly values, for 1 week. Figure 4 shows the varia- 
tion of the high S- and X-band signal strengths. Ii 
is clear that in most cases the two wavelengths change 
together. This was the predominant behavior through- 
out, the summer. The notable exceptions are those 
points where X is high and S nearly standard; this 
is the frequency effect remarked in the discussion 
of the high and steady signal which became common 
in the fall. As will "be seen later, this occurs with 
very low modified index inversions, less than 20 ft 
high. 
Figure 5 shows the relation between S-band signal 
strengths for high and low receivers; the correlation 
is excellent in practically every case. A similar cor- 
‘relation exists for the high and low X-band signal, 
except tor the case of very low signal where the ap- 
parent average value of the signal strength on the 
low receiver is always relatively high. Whether this 
is caused by the lack of receiver sensitivity or is a 
real transmission phenomenon cannot be conclusively 
decided on the basis of the present information. 
Figure 6 shows the relation of signal strengths at 
117 me and § band; the difference between this figure 
