PROJECT RAND: INQUIRY INTO THE FEASIBILITY OF WEATHER RECONNAISSANCE FROM A SATELLITE VEHICLE

• Document Type: CREST
• Collection: General CIA Records
• Document Number (FOIA) /ESDN (CREST): CIA-RDP89B00708R000500090002-4
• Release Decision: RIFPUB
• Original Classification: U
• Document Page Count: 54
• Document Creation Date: December 22, 2016
• Document Release Date: June 17, 2009
• Sequence Number: 2
• Publication Date: April 1, 1951
• Content Type: REPORT

Approved For Release 2009/06/17: CIA-RDP89B00708R000500090002-4
(,Ii~.$:ii~~r~71! VZfIGUiGU _ I
J
wet
Q1IflD.
INQUIRY INTO THE FEASIBILITY OF WEATHER
RECONNAISSANCE FROM A
SATELLITE VEHICLE
April, 1951
REPRODUCIBLE COPY
4-
ULI:L.\: 12 YLAIC.l..
000 tit) un.[3
eaoouttle
I
Approved For Release 2009/06/17: CIA-RDP89B00708R000500090002-4
 Approved For Release 2009/06/17: CIA-RDP89B00708R000500090002-4
PROJECT RAND
(USAF PROJECT RR 1)
INQUIRY INTO THE FEASIBILITY OF WEATHER
RECONNAISSANCE FROM A
SATELLITE VEHICLE
S. M. Greenfield and W. W. Kellogg
FOR OFFICIAL USE ONLY
7e ^T 1 1 D eosfie:atce.r
1500 /OUITN ST . SANTA MONICA ? CA1I OSNIA
Approved For Release 2009/06/17: CIA-RDP89B00708R000500090002-4
 Approved For Release 2009/06/17: CIA-RDP89B00708R000500090002-4
CONTENTS
SUMMARY ........................................................ v
INTRODUCTION .................................................. 1
THE SATELLITE VEHICLE .................................. :....... 2
WHAT CAN BE SEEN ............................................... 3
Area Coverage .................................................... 3
Resolution and Limiting Contrasts .................................... 8
LIMITATIONS OF THE ANALYSIS ................................... 13
WHAT CAN BE DETERMINED FROM HIGH-ALTITUDE OBSERVATIONS 14
Cloud Identification ................................................ 14
The Analysis ....................................................... 15
SUGGESTED METHODS OF ATTACK ON THE PROBLEM
OF DETERMINING THE SYNOPTIC SITUATION ................... 22
RESULTS OF THREE ATTEMPTS AT ANALYSIS ....................... 24
Viking No. 3 (NRL) .............................................. 24
V-2 No. 28 (AFCRL) ............................................. 27
Aerobee A-7 and V-2 No. 40 ......................................... 27
Comments on the Preceding Analyses ................................. 30
CONCLUSION ...................................................... 31
APPENDIXES
I. DETAILED ANALYSIS OF SYNOPTIC WEATHER AS OBSERVED
FROM PHOTOGRAPHS TAKEN ON TWO ROCKET FLIGHTS
OVER WHITE SANDS, NEW MEXICO, JULY 26, 1948 .............. 33
II. FILMS AVAILABLE FOR THIS STUDY ............................ 43
Approved For Release 2009/06/17: CIA-RDP89B00708R000500090002-4
 Approved For Release 2009/06/17: CIA-RDP89B00708R000500090002-4
1. Viewing system ................................................. 3
2. Available contrast with varying cloud and background albedos........... 11
3. Vertical wind shear-temperature gradient relationship .................. 18
4. Graphical representation of degrees of stability as given by lapse rate of
temperature ................................................... 19
5. Schematic representation of viewing area from rocket camera ............. 25
6. Synoptic situation as estimated from Viking No. 3 photographs,
February 9, 1950 .............................................. 26
7. Actual weather situation from weather map for February 9, 1950, 1330 EST.. 26
8. Estimated synoptic situation from V-2 No. 28, December 8, 1947 ......... 28
9. Actual synoptic situation from surface map for December 8, 1947, 1330 EST.. 28
10. Estimated synoptic picture from V-2 No. 40 and Aerobee A-7, July 26, 1948. 30
11. Surface weather map for July 26, 1948, 1130 MST .................... 33
12. Selected upper-air wind maps ...................................... 35
13. Composite photographs taken from V-2 and Aerobee rocket flights on
July 26, 1948 ................................................. 36
14. Profile along 320N .............................................. 38
1.. Resolution of clouds by sunlight ................................... 4
2. Resolution of clouds by second- and third-quarter moonlight .............. 5
3. Resolution of clouds by light of the' night sky ........................ 5
4. Possible performance capabilities for various image orthicon cameras, taking
into account illumination sources ................................. 6
5. Surface albedo and scene contrast of clouds against various background
surfaces ...................................................... 12
Approved For Release 2009/06/17: CIA-RDP89B00708R000500090002-4
 Approved For Release 2009/06/17: CIA-RDP89B00708R000500090002-4
SUMMARY
It is assumed that, in the event of armed conflict, aerial weather reconnaissance over
enemy territory, similar to that obtained in World War II, will be extremely difficult
if not impossible. An alternative method of obtaining this information, however, is
thought to lie in the use of the proposed satellite vehicle. It is obvious that any meteoro-
logical reconnaissance utilizing only observations from such a high-altitude "eye" cannot
provide quantitative values for the parameters normally associated with standard weather
observation and forecasting techniques. In determining the feasibility of such a system,
therefore, the questions that must be answered are: (1) What extent of coverage can
be expected from a satellite viewing system? (2) In terms of resolution and contrast,
what can be seen from the satellite? (3) Given proper coverage and resolution, what
can actually be determined regarding the synoptic weather situation from this information?
A current technical report on the satellite' proposes several possible flight altitudes
between 350 and 500 mi. For the purpose of the present study, however, only the 350-
mi altitude was considered to any extent. At this altitude, a vehicle would have an
orbital velocity of about 24,870 ft/sec and would make one complete circuit of its orbit
in 1.6 hr. Assuming that any regressive motion of the satellite's orbit owing to the
spatial motion and oblate shape of the earth is corrected for, and that the area it is
desired to observe is in daylight during the vehicle passage for an extended period, this
area will be covered and televised in a grid fashion once every 24 hr. It is visualized
that, by means of mechanical scanning transverse to the path of the satellite, a continuous
strip whose width is equal in order of magnitude to the altitude of the vehicle will be
viewed. Utilizing a map of Russia with projections of trajectories for a satellite at 354.6
mi and 453.3 mi altitude, direct measurements were made of the area covered by differ-
ent strip widths. An area of 105,000 sq mi between latitude 45? and 50?N was used.
It was found that in the case of the 354.6-mi satellite, one-third of this area was covered
grid fashion by a 100-mi-wide strip in 24 hr, and two-thirds of this area was covered
grid fashion by a 100-mi-wide strip in 48 hr by the 453.3-mi satellite. Both cases natu-
rally indicated some overlapping, which increased as the original tangent latitude
was approached.
Utilizing photographs from recent vertically fired rockets (V-2), an estimate of the
dimensions of the smallest increment necessary for proper cloud identification was made.
This was found to be approximately 500 ft and is termed the "usable resolution" in
this report. Entering Tables 1, 2, and 3, which give resolution versus contrast for various
values of frame speed, .aperture size, and various types of illumination, showed that it
was possible to obtain this value of resolution in sunlight illumination with contrast
1 R. M. Salter and J. E. Lipp, A Satellite Rocket for Reconnaissance, Project RAND, The RAND
Corporation, Report R-217, April, 1951 (Secret).
Approved For Release 2009/06/17: CIA-RDP89B00708R000500090002-4
 Approved For Release 2009/06/17: CIA-RDP89B00708R000500090002-4
between cloud and background of less than 10 per cent. An examination of the albedos
from typical background objects, as presented in Fig. 2, page 11, compared with cloud
albedos seems to indicate that 10 per cent contrast is available over a wide range of
possible cloud-ground and cloud-cloud combinations. This, therefore, appears to establish
the feasibility of cloud identification from high altitudes, at least from the standpoint of
contrast and resolution.
Owing to the lack of quantitative measurements, the clouds must be utilized to their
utmost in determining the synoptic weather picture. Experience and statistical climato-
logical values play their part in forming this picture, and the process involves a "hunting
technique" that oscillates between the three main tools at the analyst's command. Some
detailed estimates of various parameters are possible from the visual cloud characteristics.
Items, such as moisture content, temperature gradient, stability, magnitude or direction
of vertical pressure gradient, wind shear, and wind direction all show promise of yield-
ing good estimates of the actual values to this type of analysis and of helping to clarify
the final estimated synoptic picture. For any future operational use, this study has shown
that such things as a cloud atlas of clouds viewed from above, complete climatological
material on the area in question (including a possible statistical survey of fluctuations
from the normal of the various parameters as attributed to synoptic systems and broken
down into small regions of similar climate and topography), and perimeter weather
will immeasurably help the job of the observer and analyst. As an aid to getting a "feel"
for the problem involved, photographs from three rocket flights were analyzed and the
synoptic situation was estimated. These results and the actual weather for the correspond-
ing times are presented in the section entitled "Results of Three Attempts at Analysis,"
page 24. In an attempt to correlate further the rocket photographs with the actual syn-
optic picture, Dr. J. Bjerknes, of U.C.L.A., independently made an analysis of photo-
graphs taken on a flight on July 26, 1948. In this analysis, all other synoptic meteoro-
logical data available for that date were utilized. The results of this analysis are presented
in Appendix I.
Approved For Release 2009/06/17: CIA-RDP89B00708R000500090002-4
 Approved For Release 2009/06/17: CIA-RDP89B00708R000500090002-4
R-218
v-a
The value of observing the weather over inaccessible areas by aerial
weather reconnaissance has been recognized for many years. An alternative
method of obtaining broad coverage of the weather, however, is thought to
lie in the use of a special satellite vehicle which could observe cloud
patterns. It is obvious that .........
General considerations of ease of satellite launching and photographic
coverage suggest an orbiting altitude of 350 to 500 mi.1 For the purpose
..... viewed. As an example of the sort of coverage which could be pro-
vided at middle latitudes, with a satellite at 354.6 mi altitude the
fraction of the area between 450 and 500 latitude which can be covered
grid-fashion with a 100 mi wide scanning path in 24 hours is one-third,
and if the width of the path is increased to 450 mi the 24 hours coverage
is complete.
1For more concerning problems of satellite operation, reference is made
to R. M. Salter and J. E. Lipp, Project RAND R-217 (classified), The RAND
Corporation, April, 1951.
1
Approved For Release 2009/06/17: CIA-RDP89B00708R000500090002-4
 Approved For Release 2009/06/17: CIA-RDP89B00708R000500090002-4
In the event of armed conflict there will be large regions of the world from which it
will be impossible to obtain weather information by normal means. Owing to the fact
that the success of any aerial reconnaissance depends, to a large extent, on a knowledge
of the weather conditions over the target, the lack of this information will be felt more
and more as any planned air offensive progresses. Systematic weather reconnaissance by
some unconventional means must therefore be undertaken.
Although weather reconnaissance by means of aircraft is now a routine matter, war-
time defensive measures installed by the enemy might seriously curtail the successful
completion of this type of mission over hostile territory. A further solution to this prob-
lem, however, consists in weather observations made by means of a television camera
placed in an unmanned vehicle flying above the normal range of defense weapons. Such
a method has the double advantage of providing both a means of observation having a
high degree of safety and reasonable probability of success and an over-all picture of the
wide-scale weather situation that is lacking in normal daily station weather observations.
It is obvious that in observing the weather through the "eye" of a high-altitude robot
almost all of the regular quantitative measurements usually associated with meteorology
must fall by the wayside. It is impossible to make more than an intelligent guess at the
values of temperature, pressure, humidity, and the remaining quantitative meteorological
parameters. Because of this, the analyst must rely on the visible components of meteor-
ology to ascertain to some usable degree the synoptic weather situation.
Clouds, being the objects most easily discernible from extreme altitudes, become the
important item and must be utilized to the utmost in forming a synoptic picture. It is
apparent that from clouds alone it will be impossible to tell everything about the current
synoptic situation. Combined, however, with both theoretical knowledge and that gained
through experience, an accurate cloud analysis can produce surprisingly good results.
The purpose of this report is to present methods of attack on the above problem, to
show what may be actually seen from high-altitude photographs (primarily a discussion
on'necessary resolution and area coverages), to discuss what may be determined from
these photographs (both directly and indirectly), and to give some results obtained.
Although all the present analysis is based on data obtained from vertically fired rockets,
the experience gained therefrom permits recommendations on possible methods of
forming a synoptic picture from satellite-missile photographs.
Approved For Release 2009/06/17: CIA-RDP89B00708R000500090002-4
 Approved For Release 2009/06/17: CIA-RDP89B00708R000500090002-4
R-218
Is.
The foundation of all meteorological forecasting systems is the
weather observing network. Whether the forecast is "local" or for the en-
tire Northern Hemisphere, the starting point must be an appraisal of the
synoptic weather picture. Since storm systems at middle latitudes generally
move from west to east, a meteorologist who does not have good observations
from a rather wide area (particularly to the west) is at a disadvantage;
and such is often the case for coastal regions, since weather reporting
over the oceans is often inadequate.
Although ship reports and weather reconnaissance by aircraft help to
some extent to fill the gap, there has long been a need for extending
weather observations over the oceans and inaccessible areas. A solution
to this problem may lie in weather observations made by means of a television
camera placed in an unmanned satellite vehicle. Such a method has the
advantage of providing a means of observation of the over-all picture of
the wide-scale weather situation that is lacking in normal daily weather
observations, and should give new insight into the behavior of the atmos-
phere.
Approved For Release 2009/06/17: CIA-RDP89B00708R000500090002-4
 Approved For Release 2009/06/17: CIA-RDP89B00708R000500090002-4
Owing to the ever-changing pattern of the atmosphere, the need for almost constant
surveillance must be foremost in any plan to trace synoptic weather situations. Any vehicle
designed for such a purpose must therefore have the ability to make many trips over
the area in question. These traverses, moreover, must be made in such a fashion that
they not only cover a representative portion of the area, but also complete their cycle
often enough to enable an observer to notice any significant change or shift in the
cloud systems.
Such a vehicle is the satellite missile. Flying high above the uppermost reaches of the
known enemy defenses, the satellite becomes an unparalleled instrument for weather
reconnaissance when scope of view is considered. In the current RAND study on the
satellite2 various operational altitudes ranging between 350 and 500 mi are proposed.
For the purpose of simplicity, all calculations and performance considerations in this
report will be based on a satellite assumed to be circling the earth at an altitude of about
350 mi. At this altitude such a vehicles would have an orbital velocity equal to 24,870
ft/sec+ and would make one complete circuit of its orbit in 1.6 hr. Also, because of the
fact that this missile is theoretically moving in a stable orbit around the earth, the globe
turning under the vehicle causes the trajectory of the satellite to appear to "creep" over
the face of the earth, thereby increasing the area observed.' Depending on the efficiency
of the powerplant, the order of magnitude of the time period for which the vehicle
could be kept operating is thought to be 1 yr. However, in attempting to decide the
satellite's full worth for weather reconnaissance, the questions that must be considered
are as follows: Can enough be seen from such altitude to enable an intelligent, usable,
weather (cloud) observation to be made, and what can be determined from these
observations ?
2 Ibid.
3 The actual altitude to which these figures apply is 354.6 mi.
4 The actual velocity of a projection of the satellite's image over the face of the globe is really
a variable resulting from the change in angular velocity from latitude circle to latitude circle.
s It should be noted that the concept of "repetitive traverses" is in itself complicated in that,
regardless of the stability of the satellite orbit, the spatial movement and the oblate shape of the
earth impart a regressive motion to the vehicle relative to fixed points on the earth. For a satellite
at approximately 350 mi altitude in an orbit set tangent to a latitude of 56?, 78 days will be
required for it to appear twice over the same point on earth at exactly the same time. This regres-
sive motion can be partially corrected by an adjustment of the speed (through altitude change) of
the satellite. It further imparts a limitation on successful viewing in that for approximately half of
the 78-day period (assuming 12 hr of photographable time out of every 24) the desired area will
have night at the time of the satellite's passage. For a complete discussion on regression of the orbit,
the interested reader is referred to RAND Report R-217 (see footnote I, p. v).
Approved For Release 2009/06/17: CIA-RDP89B00708R000500090002-4
 Approved For Release 2009/06/17: CIA-RDP89B00708R000500090002-4
-218
2a
Owing to the ever-changing .........
cloud systems.
Such a vehicle is the satellite. Flying high above the sensible
part of the atmosphere, so that atmospheric drag becanes negligible, the
satellite becomes an unparalleled instrument for weather reconnaissance
when scope of view is considered. For the purpose ........
Approved For Release 2009/06/17: CIA-RDP89B00708R000500090002-4
 Approved For Release 2009/06/17: CIA-RDP89B00708R000500090002-4
Naturally, any estimate of the amount that can be seen from an extreme altitude must
be a function of both the resolving power of the camera system and the area that can be
scanned and recorded (or televised) and still retain usable detail. Much of the discussion
and most of the figures in this section are the result of previous RAND studies conducted
by Dr. R. S. Wehner.
-- - ------ -------
Television camera
Wf Sensitlve
optical ~t Nement
system-.
(aperture) 'At
// t \
W ? width of sensitive element // l \
area viewed by one frame /~ 11 \
W=2 tan
\ e - 2 ton e
/`at or
W .
/ t
I` L
Fig. 1-Viewing system
Using the relation (see Fig. 1)
and using Tables 1, 2, and 3, it is possible to compute the width of square surface
viewed and the angle of view for any given camera and aperture. This has been done
and is summarized in Table 4. As can be seen, if a limiting resolutions of 500 ft is
F = d = 2tan 2
where W = sensitive element width, in.
w = width of surface pictured per frame, mi
F = focal length of camera, in.
d = optical range, mi
a = angle of view, deg,
a The term "limiting resolution," as used in the television field, refers to the greatest possible
resolution attainable by a given TV pick-up tube and is wholly dependent on the structural make-
up of the tube itself. As used in this report, limiting, minimum, or usable resolution is a quantity
depending on scene contrast, signal-to-noise ratio, aperture, f number of camera, etc., and is chosen
to pick up the smallest object that it is desired to view.
Approved For Release 2009/06/17: CIA-RDP89B00708R000500090002-4
 Approved For Release 2009/06/17: CIA-RDP89B00708R000500090002-4
set, it is still possible to obtain this resolving power under sunlight illumination with a
contrast as low as 2.5 per cent (with a 5.0-in. aperture). Under moonlight, however,
this resolution is possible only with 100 per cent contrast, a very fast f/1.4 lens, and a
minimum exposure time of 0.25 sec; under light of the night sky illumination it is not
possible at all. Assuming, then, that the chosen limiting resolution is correct, the proba-
bility of obtaining identifiable cloud photographs under any but sunlight illumination
appears to be small.
Table 1 *
RESOLUTION OF CLOUDS BY SUNLIGHT
Ilmage orthicon f/10 camera operated at 20:1;t signal-to-noise ratio at rotes of
40, 10, and 4 exposure frames/sec; and a satellite height of 350 mil
Minimum Resolvable Surface Dimension
Contrast
Aperture
40 Frames/Sec
10 Frames/Sec
4 Frames/Sec
(%)
(in.)
(ft)
(ft)
(ft)
100
0.5
200+
100`+
64+
1.0
100+
50+
32++
2.0
50+
25+
16+
5.0
20+
10+
6+
25
0.5
800
400+
250+
1.0
400
200+
125+
2.0
200
100+
64+
5.0
80
40++
25+
10
0.5
2,000
1,000
640
1.0
1,000
500
320
2.0
500
250
160
5.0
200
100
64
2.5
0.5
8,000
4,000
2,500
1.0
4,000
2,000
1,250
2.0
2,000
1,000
640
5.0
800
400
250
1
0.5
20,000
10,000
6,400
1.0
10,000
5,000
3,200
2.0
5,000
2,500
1,600
5.0
2,000
1,000
640
* The material contained in this table was prepared by Dr. R. S. Wehner
and is included in RAND Report R-217 (see footnote 1, p. v).
t It should be noted that this table (and also Tables 2 and 3) is un-
realistic in that the 20:1 signal-to-noise ratio is applicable only to 25 per
cent contrast. For 10 per cent contrast, a signal-to-noise ratio of 50:1 is
required. This would mean a required transmitter power increase by a factor
of 2.5 (assuming a 2-in. aperture, 1000 TV lines, and a frame frequency of
10/sec). This is still not prohibitive but does become so with a substantial
increase in either the number of TV lines or the frame frequency.
+ Values of computed resolution smaller than realizable with present
commercial image orthicons.
Approved For Release 2009/06/17: CIA-RDP89B00708R000500090002-4
 Approved For Release 2009/06/17: CIA-RDP89B00708R000500090002-4
Table 2*
RESOLUTION OF CLOUDS BY SECOND- AND THIRD-QUARTER
MOONLIGHT
(Imago orthicon f/1.4 camera operated at 20:1; signal-to-noise ratio at rates of
40, 10, and 4 exposure frames/sec; and a satellite height of 350 mill
Minimum Resolvable Surface Dimension
Contrast
Aperture
40 Frames/Sec
10 Frames/Sec
4 Frames/Sec
(%)
(in.)
(mi)
(mi)
(mi)
100
5
1.08
0.54
0.34
10
.54
.27
0.17
20
.27
.14
0.09
25
5
4.32
2.16
1.36
10
2.16
1.08
0.68
20
1.08
0.54
0.34
10
5
10.8
5.4
3.4
10
5.4
2.7
1.7
20
2.7
1.35
0.85
* The material contained in this table is included in RAND Report R-217
(see footnote 1, p. v).
Table 3*
RESOLUTION OF CLOUDS BY LIGHT OF THE NIGHT SKY
(Imago orthicon f/0.7 camera operated at 20:1; signal-to-noise ratio at rates of
40, 10, and 4 exposure frames/sec; and a satellite height of 350 mil
Minimum Resolvable Surface Dimension
Contrast
Aperture
40 Frames/Sec
10 Frames/Sec
4 Frames/Sec
(%)
(in.)
(mi)
(mi)
(mi)
100
10
4.3
2.15
1.36
20
2.15
1.08
0.68
40
1.08
.54
.34
25
10
17.2
8.6
5.4
20
8.6
4.3
2.7
40
4.3
2.15
1.36
10
10
43.0
21.5
13.6
20
21.5
10.8
6.8
40
10.8
5.4
3.4
* The material contained in this table is included in RAND Report R-217
(see footnote 1, p. v).
Approved For Release 2009/06/17: CIA-RDP89B00708R000500090002-4
 Approved For Release 2009/06/17: CIA-RDP89B00708R000500090002-4
L Q
- u E
Y
1/~
V V
O. p W W
n
en
vQ ~ ?
C o
L
.~y
h
N
N
E w
U
E
E a
o
o
o
% E
Zw
E
7 N w O
CIA
~
2
Y w
c
U 3
a0
vim.
.a a
~^
o
0
N
0
h
~
v
~
m
N
<
v
a
C
r
o
0
N
0
V1
V1
O
O
d v
N
?
?c
r
?J7 eEa ?'
? H m
C
?~ eCa C
O
s o
1I
~
;!
V '~
A T L
A 't^ i
E
o -o
E u
E E
E _'
b
4
'c
~ ?c `
V
n y
a
Q
o
E
E ..,
c <
Approved For Release 2009/06/17: CIA-RDP89B00708R000500090002-4
 Approved For Release 2009/06/17: CIA-RDP89B00708R000500090002-4
C
Calculations must also be performed to arrive at the possible area coverage. Since it is
apparent that cloud observations, to be at all useful, have to be made over a wide-enough
strip (at least as wide as the height of the satellite), it should be considered that the
camera will be mechanically scanned. This may be accomplished by means of a 450
plane mirror rotatable about the axis of the camera. The mirror actually does the "look-
ing" and scanning for the camera, which is mounted horizontally, its axis being parallel
to the axis of the missile. Taking a sequence of 20 nonoverlapping frames will produce
a strip 350 mi long, transverse to the trajectory of the satellite, and 17.5 mi wide. If the
camera is set to take 5 frames/sec and the rotatable mirror is fixed with a fast snap-back
device, the system will then be in position to take a second strip by the time the satellite
has moved ahead approximately 17.5 mi relative to the earth. (The speed of the missile
relative to the earth's surface is about 4.4 mi/sec.) This will produce a continuous 350-mi
wide strip around the earth with each complete traverse of the missile.
The daylight camera with an f/10 lens and an image orthicon television tube and
whose performance is summarized in Table 4, should have a 2-in. objective to give the
proper ground coverage per frame. This combination allows a 500-ft object to be resolved
with only 10 per cent contrast,-, which is reasonably small. It should be emphasized that
these figures are presented here merely to give some examples of performance of viewing
systems and not as a description of the optimum system performance.
Some calculations of total area coverage were also made by direct measurements (assum-
ing different strip widths) on a scale map of Russia. On this map were projected com-
plete cycles" of traverses for two proposed satellite trajectories. Once again, curvature
of the earth was neglected. The results obtained are as follows:
1. For a satellite with 24-hr complete cycle (354.6 mi altitude, angular velocity 15
times that of earth, and trajectory tangent to lat. 56?N. (Moscow)
Assuming a 100-mi-wide scanning band (50 mi on either side of path): In the
vicinity of lat. 45?-50?N., taking an area of approximately 105,000 sq mi, we
find that in 24 hr the surface has been covered in a grid fashion such that about
one-third (32,500 sq mi) of its area has been scanned and presumably televised.
Asruming a 200-mi-wide scanning band (100 mi on either side of path): As
may be expected, doubling the scanning band does not quite double the area
covered. This. is owing to some overlapping of the bands. (It can be shown that,
T This resolution and contrast represent the maximum needs of satellite weather observation. As
pointed out in the general report on the satellite, 200-ft resolutions are available for reconnaissance.
This is obtainable with 25 per cent contrast when using an f/10 lens with a 2-in. objective in
sunlight illumination (see footnote to Table I marked (t) ).
"Initially, the trajectory of the satellite is set tangent to a given latitude. Owing to the relativ.
difference in the angular velocities between the satellite and the earth and to the relative stability
of the orbit of the missile, the vehicle's trajectory appears to "creep" over the surface of the globe.
A complete cycle is the time it takes for the trajectory of the satellite once again to become tangent
to the original point. (This "creep" causes the traverses to become more widely dispersed as the
trajectory approaches the equator.)
Approved For Release 2009/06/17: CIA-RDP89B00708R000500090002-4
 Approved For Release 2009/06/17: CIA-RDP89B00708R000500090002-4
R-218
7a
Some calculations of the efficiency of coverage of an inaccessible
area such as an ocean were also made by direct measurements (assuming
different strip widths) on a grid map. On this map were projected com-
plete cycles8 of traverses.......
(Note deletions on page 7)
& 8.
Approved For Release 2009/06/17: CIA-RDP89B00708R000500090002-4
 Approved For Release 2009/06/17: CIA-RDP89B00708R000500090002-4
to cover the area completely, a scanning band approximately 450 mi in width
is needed.)
As a result of the grid-like coverage, the 100-mi-wide band, at its worst, should
pick up at least portions of the largest, most active weather disturbances and
enough of the remaining cloud coverage to orient the system in relation to
the ground.
2. For a 48-hr complete cycle (altitude 453.3 mi, angular velocity 14.5 times that
of the earth, and trajectory tangent to lat. 56?N. (Moscow)
Assuming a 100-mi-wide scanning band (50 mi on either side of path): In the
vicinity of lat. 45?-50?N., using a region of approximately 231,000 sq mi, we
find that in 48 hr it has been covered, grid fashion, so that two-thirds of its
area (148,400 sq mi) has been scanned and presumably televised. (It can be
shown that to cover the area completely in a 48-hr cycle a scanning band approxi-
mately 250-300 mi in width would be required.
The results so obtained give an idea of the areas which can be covered (or scanned)
from a vehicle in an orbit 350 mi above the surface of the earth. The 350-mi-wide strip
discussed in the first part of this section will therefore cover in 24 hr a large percentage
of the area between 45?N. and 56?N. over Russia9 with considerable overlapping of
scanning, particularly around the 56th parallel. In any event, the coverage, as mentioned
here, if achieved with any measure of success, should produce good weather recon-
naissance results.
RESOLUTION AND LIMITING CONTRASTS
Since it is now obvious that clouds will be the chief meteorological element directly
observable from high-altitude photographs, it must be ascertained how closely these
clouds may be identified and what may be determined from them, either directly
or indirectly.
As can be seen from Tables 1, 2, and 3, when a set of conditions such as aperture,
illumination, exposure time, and focal length-aperture diameter ratio of a given camera
have been established, the remaining factor for determination of the minimum resolution
attainable is the contrast value. In cloud photography of the type to be attempted from
the satellite, one is unable to choose the surrounding photographic conditions. Features
such as background, lighting at time of observation, etc., are examples of the uncontrolled
variables, and, as a consequence, any system of data gathering by photographic means
must be flexible enough to give adequate results over a wide range of limiting factors.
The question is: If the camera and optical system are chosen,10 and if the various con-
ditions of lighting, background, etc., are assumed to remain within the limits providing
9 A large percentage of the area should be covered in the 24-hr trajectory, and almost all should
be scanned in the 48-hr trajectory.
10 Previous studies at RAND have shown that one of the best available television cameras for
use in the satellite would be one employing an image orthicon pica-up tube. The characteristics of
this tube approach those of the human eye over part of its operating range, it has a greater sensi-
Approved For Release 2009/06/17: CIA-RDP89B00708R000500090002-4
 Approved For Release 2009/06/17: CIA-RDP89B00708R000500090002-4
usable resolution, will the resulting limiting contrast values still enable one to observe
the weather under a wide-enough range of actual conditions?
Before endeavoring to answer this question it is desirable to define the term "usable
resolution." It was thought that details of cloud structure as small as several hundred
feet in diameter might possess significance when an attempt was made to form a synoptic
picture by means of cloud analysis. This was borne out when high-altitude rocket photo-
graphs were examined. Further reasons for assuming this to be the approximate mini-
mum size to be resolved were found when a simple test was conducted on these photo-
graphs. (The heights at which these pictures were taken varied between 50 and 70 mi.)
Using an adjustable viewer, the photograph was taken slowly out of focus until it was
impossible to identify definitely the forms of clouds other than by saying that they were
widespread or were in small clusters. For example, beyond this point it was impossible to
distinguish between closely packed cumulus and a deck of altocumulus, and also between
a dense layer of stratus or altostratus and the fibrous texture of cirrostratus. A study of
other parts of the photograph, where recognizable or measurable objects were located
at ranges about equal to those of the clouds, showed that the limiting resolution at which
the clouds lost their distinguishability was from 500 to 1000 ft. This is what is meant
by "usable resolution." As may be imagined, this is at best only a rough approximation,
but because of its apparent agreement with previously estimated values it should serve
very well as a working basis.
It was mentioned above that in order to obtain a known, usable resolution, once the
camera and lighting conditions are chosen, the limiting contrast value must also be
specified. It is obvious that if, for various combinations of cloud-ground and cloud-cloud,
albedo differences are such that their contrast values fall below the limiting contrast,
these combinations cannot be observed by high-altitude weather reconnaissance.
Hewson, in an article in a meteorological journal" and in his book (written in collabo-
ration with Longley) on theoretical and applied meteorology,12 calculated and tabulated
diffuse-reflection coefficients for clouds of various thicknesses. In doing so, as a result
of the extensive variation of cloud liquid-water densities and cloud droplet radii, he was
forced to choose one set of values for these two parameters. Those on which his figures
are based are a density of 1.0 gm of liquid water per cubic meter of cloud and a droplet
radius of 5 X 10-1cm. Owing to the fact that these values probably apply to a large
percentage of the usable clouds observable from extreme altitudes, they may be reasonably
tivity than earlier types, and it is capable of stable operation without adjustment over a wide range
of illumination intensity. Since it is not the purpose of this report to delve too deeply into the
technical aspects of the problems of a television viewing system, only results of resolution compu-
tations for the image orthicon tube are presented here. These are summed up in Tables 1, 2, and 3.
For technical information and equations involved, see R. B. James, R. E. Johnson, and R. S. Moore,
RCA Review, Vol.. 10, July, 1949, pp. 191-223; and A. Rose, "Television Pickup Tubes and the
Problem of Vision," Advances in Electronics, Vol. 1, Academic Press, New York, 1948.
11 E. W. Hewson, Quart. J. Roy. Met. Soc., Vol. 69 (1943), p. 47.
12 E. W. Hewson and R. W. Longley, Meteorology, Theoretical and Applied, John Wiley &
Sons, Inc., New York, 1944, pp. 73-75.
Approved For Release 2009/06/17: CIA-RDP89B00708R000500090002-4
 Approved For Release 2009/06/17: CIA-RDP89B00708R000500090002-4
employed in making estimates for this study. These values are plotted in Fig. 2, the
ordinate and abscissa being contrast and background albedo, respectively. Each curve
represents a particular albedo applicable to a particular cloud thickness. According to the
definition of contrast,
C = Pb - Pd
Pb
where Pb = brightness (albedo) of the brightest thing viewed
(either object or background)
pd = brightness of darkest object viewed (albedo)
C = contrast between the two.
From the above definition, each curve may be represented by the following relation:
1 - Ab , for Ab < Ar
A,
C = 0, for Ab = At
1 - At for Ab > At ,
b
where C = contrast between object and background
Ab = albedo of background
At = albedo of clouds of various thicknesses.
It is therefore seen that, except for the small range of albedo combinations around the
point of discontinuity on the curves, a large majority of possible cloud-background albedo
combinations fall within the range of at least 10 per cent contrast. As can be seen from
Table 1, assuming at least a 2.0-in, aperture and sunlight illumination, an f/10 camera
will permit at least 10 per cent contrast for approximately 500-ft resolutions.13 Table 5
(page 12) gives the albedos for various ground covers. Applying these values to Fig. 2,
it can be seen that, except for the case of newly fallen snow combined with clouds thicker
than 600 meters and the case in which the background albedo approaches very close to
cloud albedo, 500-ft resolutions are obtainable over a wide range of conditions.
There is one other factor that might limit contrast and, therefore, resolution. This is
aerial haze between the camera and the ground. As has recently been shown in several
to Using the equation (for altitude of 350 mi)
c = 1600
8a
where C = contrast
a = aperture
r = exposure time (or time of one frame)
8 =minimum resolvable surface dimension,
it is possible to calculate the contrast (minimum) needed to obtain at least 500 ft resolution under
the conditions given in the example on ground coverage which assumed full daylight illumination.
This value turns out to be 3.56 per cent. Owing to the unrealistic power requirements necessary to
transmit 3.56 per cent contrast, this value has been raised to 10 per cent.
Approved For Release 2009/06/17: CIA-RDP89B00708R000500090002-4
 Approved For Release 2009/06/17: CIA-RDP89B00708R000500090002-4
Dashed line represents contrast necessary
to transmit about 500 ft resolution with f/10
image orthicon camera, 2.0 In.operature,
50:1 signal to noise ratio, and frome speed
of 40/sec. Satellite at 350 miles altitude.
Thickness of cloud
(meters)
20
10
oo,
60
100
200 10
00
000
40 60
Albedo of background (percent)
V-2 photographs, this problem is almost completely solved by use of an infrared filter
in the optical system.
From the foregoing section we may conclude that, from the standpoint of area cover-
age and resolution, weather observations from a satellite are a definite possibility.
Approved For Release 2009/06/17: CIA-RDP89B00708R000500090002-4
 Approved For Release 2009/06/17: CIA-RDP89B00708R000500090002-4
Table 5*
SURFACE ALBEDO AND SCENE CONTRAST OF CLOUDS
AGAINST VARIOUS BACKGROUND SURFACES
Ground Surface
Albedot
References$
Fresh snow
.80-.93
1, 3, 4
Old snow, sea ice
.40-.60
3, 4
Brown soil
.32
1
Grass
.10-33
4
Green leaves
.25
1
Sandy loam
.24
2
Sand
.13-.18
3
Asphalt paving
.15
2
Dry earth
.14
4
Rock
.12-.15
4
Moist earth
.08-.09
2, 4
Cultivated soil, vegetable
.07-.09
3
Smooth sea surface
Solar elev 5 deg
.40
3
Solar elev 10 deg
.25
Solar elev 20 deg
.12
Solar elev 30 deg
.06
Solar elev 40 deg
.04
Solar elev 50-90 deg
.03
* This table was prepared by Dr. R. S. Wehner and is included
in the RAND general report on the satellite (see footnote 1, p. v).
t Values of albedo apply to illumination by "white" light
or sunlight.
*+ References:
1. International Critical Tables, 1929 ed., Vol. 5, p. 262.
2. Handbook of Chemistry and Physics, 1942 ed., pp.
2147-2148.
3. H. Landsberg, Handbook of Meteorology, McGraw-Hill
Book Company, Inc., New York, 1945, p. 932.
4. J. Charney, Handbook of Meterology, McGraw-Hill Book
Company, Inc., New York, 1945, p. 296.
Approved For Release 2009/06/17: CIA-RDP89B00708R000500090002-4
 Approved For Release 2009/06/17: CIA-RDP89B00708R000500090002-4
LIMITATIONS OF THE ANALYSIS
It is a known fact that the reliability of any form of synoptic meteorological analysis
depends on the experience of the analyst. An analysis of the type dealt with in this
report is no exception. If anything, it is even more dependent on analytical experience
because of the sparseness of data and the difficulties in interpretation. To date, the
meteorological cloud atlas has been built up almost entirely from ground observations.
The change-over to "looking down" upon the clouds means that the dominant features
which, served to identify types of clouds when observed from the ground are no longer
to be seen. The halo and corona that served so well to classify cirrostratus and altostratus,
respectively, are absent. Also, the upper surface of large-scale cloud decks is, for the most
part, completely different in appearance from the lower surface. Therefore a completely
new concept of cloud-identification features must be formed, and only those experienced
in these new concepts will be able to make an intelligent analysis.
There is also danger of an incorrect interpretation of the cause for the clouds, which
might lead to a completely erroneous analysis. Take, for example, the case in which the
entire picture under consideration exhibits one complete deck of clouds. In this case the
deck of clouds might be stratus caused by radiational cooling and so might constitute an
entirely local phenomenon. An analyst looking at this situation might jump to the con-
clusion that the clouds in question were of frontal origin, possibly altostratus, and might
forecast accordingly. It is evident that a forecast made from such an erroneous assump-
tion of the cause would be completely incorrect. (Methods of attack on this problem of
analysis are treated more fully in a later section of this report.)
There are also many definite advantages to be gained in the analysis of weather by
this method; chief among these is the fact that extremely large areas may be visually
observed in a relatively short period of time. The disadvantages of large gaps (between
stations) on the usual weather map and the comparatively limited field of view of each
ground observer are eliminated. What is obtained is, in effect, the cloud pattern inte.
grated over a wide area. From many points of view this is highly desirable, owing to the
fact that, for the first time in the history of synoptic meteorology, the classical models of
various weather situations may be examined in toto.'?
24 This idea of "the over-all look" was first described by Major D. L. Crowson, USAF, in a
recent article in the B. Amer. Meteorol. Soc., Vol. 30, No. 1, January, 1949. His primary object
was the use of vertically fired rockets in conjunction with the regular meteorological observations
as a supplement rather than as a possible replacement. In this regard, his analysis of rocket photo-
graphs is very similar to that presented by Dr. J. Bjerknes in Appendix I.
Approved For Release 2009/06/17: CIA-RDP89B00708R000500090002-4
 Approved For Release 2009/06/17: CIA-RDP89B00708R000500090002-4
WHAT CAN BE DETERMINED FROM HIGH-
ALTITUDE OBSERVATIONS
Assuming, from the previous section, that loud shapes of the order of 500 ft or more
in diameter are distinguishable from an altitude of 350 mi, the problem of identifying
these clouds can be treated. As stated previously, attributes and/or phenomena that served
to establish the classification of clouds when viewed from the ground are almost com-
pletely different. when these same clouds are viewed from above. The question is, What
can actually be done to tell the various cloud forms apart?
The solution to this problem may lie in a new classification system formed by means
of close correlation of observations of clouds viewed from above with observations of
these same clouds viewed from below. In this manner, an atlas of identifying cloud
features as scanned from extreme altitudes might be built up. Using this information, a
trained observer should have little trouble in establishing the identity of almost any
visible cloud. The importance of such an atlas cannot be over-emphasized, because the
degree of confidence in a synoptic picture formed from this type of observation or in
the subsequent forecast becomes extremely small if the identity of the clouds cannot be
established. An attempt along these lines has been made, utilizing several series of photo-
graphs taken from V-2's fired at White Sands, New Mexico. It should be kept in mind,
however, that this attempt was made using data which were not originally gathered for
this purpose, and the necessary ground observations are therefore not available for posi-
tive identification purposes. Because of this, the results presented here are a classifica-
tion and identification based on the writers observational experience.
From a study of the above-mentioned photographs it was observed that two general
cloud forms stand out from each other under the usable resolution conditions. Since these
two forms are also two types of cloud formations, most other clouds can be considered
as being a special form or combination of these and may be so categorized. This is
partially attempted in the table on the opposite page.
It is noticed that certain formations of clouds very often assemble in over-all patterns
peculiar to these formations. Clouds, therefore, may also be partially categorized accord-
ing to pattern. In the case of clouds formed by globular masses joined together to
produce a single layer, the pattern is still apparent to an observer on the ground as
a result of the differences in light intensities caused by the variations in cloud thickness.
It is likely, therefore, from the section on cloud contrast, that these patterns will also be
visible to an observer stationed above the layer, owing to the difference in albedo values
caused by cloud sections of different thicknesses. These patterns are very useful in cutting
down the overlap present in the following table.
Approved For Release 2009/06/17: CIA-RDP89B00708R000500090002-4
 Approved For Release 2009/06/17: CIA-RDP89B00708R000500090002-4
A
I
Vertically Developed
Remarks
Stratiform
Remarks
1. Cumulus
Varying degrees of
1. Stratus
In some forms may be
2. Cumulonimbus }
vertical development
2. Altostratus
very similar.
3. Cirrostratus
May be distinguished be-
cause fibrous texture
is visible even when
viewed from above.
4. Nimbostratus
When there is no verti-
cal development on top
this form may appeal
to be very similar to
Nos. I and 2.
IAltocumulus 1
Cirrocumulus J
IAltostratus
Nimbostratus
Altocumulus cloud elements may exhibit vertical development, or there may
be just closely packed globular masses. In the first case, the altocumulus
may seem to be very similar to altostratus or nimbostratus that have
vertical development in their tops, although the layer may retain some
semblance of orderly pattern.
Very often these formations contain considerable -vertical development.
This seems to be especially true when these forms are associated with
the passage of a front.
It is clear that any attempt to formulate an atlas of cloud appearance as seen from high
altitudes is a major undertaking. The work done on the subject in this report represents,
at best, the beginning of the work that must be accomplished to make high-altitude cloud
photographs a usable weather tool.
THE ANALYSIS
Having once established the identity of almost all the clouds viewed, the formation of
the synoptic weather picture becomes the next problem. The following question arises:
Given an over-all cloud picture, what, in fact, can be determined, either directly or
indirectly?
According to conventional meteorological practice, the various parameters, such as
pressure, temperatures, etc., are plotted on a map, and the subsequent analysis of these
quantities produces the synoptic picture. Almost the reverse is true in the case at hand.
Here the synoptic situation must first be established, and the various parameters must be
estimated from it. Actually, it is not quite so straightforward a procedure. Rather, it
becomes a "hunting" technique, in which one makes a first approximation to the over-all
weather situation, using the clouds, and, from this, a first estimate of the value of tem-
perature, pressure, humidity, etc. This picture of the weather is then modified to fit
Approved For Release 2009/06/17: CIA-RDP89B00708R000500090002-4
 Approved For Release 2009/06/17: CIA-RDP89B00708R000500090002-4
obvious deviations of the estimated values from those indirectly observed. This process
continues until a satisfactory situation is evolved that appears to fit all existing condi-
tions (an attempt being made to satisfy both physical and theoretical considerations);
from this, final estimations of the various parameters are made. (Several possible ap-
proaches to the problem of approximating the synoptic picture are discussed in the section
entitled "Suggested Methods of Attack on the Problem of Determining the Synoptic
Situation," page 22.)
The normal observable meteorological parameters may be divided into two main cate-
gories, viz., those that may be estimated in some measure directly from observations of
the clouds and/or ground, and those that require a knowledge of the over-all weather
patterns before an estimate can be made. In the first category may be listed wind, hu-
midity, precipitation, and a variable not normally considered by itself as such-degree
of stability. In the second listing may be found pressure (and pressure tendency) and
temperature. Before an attempt at its analysis can be made, a considerable amount of
experience and general knowledge of the workings of the atmosphere is required con-
cerning each item, regardless of which category it comes under. It is found that this
estimation method is neither a quick nor a simple process, regardless of the qualifica-
tions of the analyst. Rather, each of the items requires a very careful study and the weigh-
ing of all the possible influencing conditions before approximate values can be assigned.
As a result of this pilot study, several suggested methods of estimating the various
meteorological parameters were evolved and are discussed as follows:
Wind
1. From the established meteorological models it is assumed that certain definite
weather situations will produce certain sequences of clouds preceding or following them.
This will therefore tend to orient the situation with respect to the ground. Once this
orientation has been established, the wind direction may be approximated through a.
knowledge of the theoretical circulation associated with a given synoptic weather situation.
2. It has been noticed in several photographs that, in the presence of strong upper
winds, cumulus clouds that have formed in mountainous country appear to form to the
lee of the mountains rather than to their windward side. In the presence of very light
winds, it was noticed that the cumulus tended to form on the peak of the mountain.
This phenomenon requires further study before its degree of usefulness as an observa-
tional tool can be determined.
3. Owing to the fact that cumulonimbus clouds extend from as low as 1600 ft up to
40,000 ft, their slope becomes a good indication of the vertical shear within the layer.
It was first thought that this direction of slope would be an indication of the direction
of the upper winds. However, although the wind velocity normally increases with alti-
tude, it is obvious that for any given case one should not disregard the possibility that
the wind velocity might decrease with height or that the direction and velocity distribution
in the vertical might be of such a nature as to cause the cumulonimbus to slope into the
upper wind. When this slope is combined with other factors that indicate wind direction
Approved For Release 2009/06/17: CIA-RDP89B00708R000500090002-4
 Approved For Release 2009/06/17: CIA-RDP89B00708R000500090002-4
at one particular level, it may be possible to construct a picture of the change of wind
direction with height in the layer under consideration.
4. A further indication of wind direction (in the lower levels) was observed when
small, detached clouds were seen to form in line, streaming from a mountain top. These
could be due to moist air being forced upward by the mountain and then moving down-
slope on the lee side, causing the formation of small "rotors" or individual cellular
eddies each capped by small cumulus clouds and extending for a considerable distance
downwind from the mountain. This phenomenon is known as a "standing wave" and is
often accompanied by other standing clouds at higher altitudes.
5. It has also been observed in a layer of stratus overlying mountainous terrain that
air funneling down a valley and spreading out in a relatively flat section produced lines
and swirls in the top of this cloud layer that closely matched the path the air must have
taken. This action may be very useful in determining wind direction in sections com-
pletely covered by sheet-type clouds and may be found to be of further use over areas
that are not particularly mountainous. Although photographs of large flat areas were
not available for analysis, it is thought that wind-direction determination in these sections
may still be accomplished in the lower levels. This may be done by utilizing many of
the above methods and several others that could be an outgrowth of such an analysis.
One such method might use the inherent uniform structure of a stratus sheet. In this
case it is thought that if a sheet passes over flat ground on which there are isolated
protuberances projecting into the sheet, a wake will be produced in the cloud that may
also show up when viewed from above and that will stretch downwind from the object.
Temperature
The starting point for any determination of temperature must be the statistical normal
for that time of the year. The first estimation may then be modified by t)le various
affecting conditions. The prevailing weather situation provides the first modifying
influence. This estimation is, of course, dependent on the analyst's ability to estimate
the synoptic conditions with a degree of accuracy that will answer the question, Is the
sector under observation being affected by relatively cold or warm air? Cloud systems,
wind-directions, and even forms of ground cover (snow, etc.) will help- in deciding
this. This is the first indication of the over-all complexity of this type of analysis and
serves as an actual illustration of the "hunting" technique mentioned above.
Upper air temperatures may be estimated in the same manner, clouds indicating the
boundaries between air masses (fronts). A further help in estimating this quantity is
the fact that, once having decided on a ground temperature, the degree of stability
(indicated by vertical development in clouds) and the presence or absence of intervening
fronts will enable one to construct an applicable temperature lapse rate. (The degree of
stability will determine the departure from an adiabatic lapse rate, while the degree of
cloudiness (moisture) will help an analyst to decide whether to use the moist or the dry
adiabatic lapse rate as the limiting one.)
Vertically developed cloud will also aid in determining the temperature gradient of
the surrounding area. This is true because of the fact that the vertical shear, as indicated
Approved For Release 2009/06/17: CIA-RDP89B00708R000500090002-4
 Approved For Release 2009/06/17: CIA-RDP89B00708R000500090002-4
Fig. 3-Vertical windshear- 4emperature gradient relationship
by the slope of towering cumulus and cumulonimbus clouds, orients the direction of the
higher and lower temperatures in the area. This method is employed by taking the direc-
tion of vertical shear as being from the low levels toward the high levels. If one then
faces in the direction of shear in the northern hemisphere, the lower temperature will be
on the observer's left and the higher on the observer's right (see Fig. 3).
Pressure
It is apparent that no quantitative values of pressure are forthcoming from this analysis.
Furthermore, it is virtually impossible even to make a quantitative estimate other than to
state whether the area is thought to be under the influence of a high- or a low-pressure
system. Charts of average pressures for various times of the year in different areas of
the world are available. Using these and the weather situation at the time, trends of
pressure may be established. This information when applied in conjunction with known
weather on the perimeter of Russia may be a very useful tool for forecasting purposes.
Little work has been attempted on this subject in this pilot study, and the above should
serve only as a possible starting point for any detailed research along these liQes.
C. F. Brooks's points out some further pressure information that may be obtained
from clouds. He says, in effect, that, since in the presence of any constant vertical shear
the cumulus clouds will tend to lean or slope (the amount of departure from the vertical
being a resultant of the vertical velocity and the rate of change of wind velocity with
height), any cloud that has a uniform rate of vertical growth and a 90? slope through-
out is an indication of the "uniformity of wind velocity in all layers pierced." This indi-
cates a decrease of horizontal pressure gradient with height. (This.can be shown very
simply by an examination of the geostrophic wind equation
V, P(anlA,
` l
where V, = the geostrophic wind velocity
p = density of air
ap/an = horizontal pressure gradient
A = Coriolis parameter.
is C. F. Brooks, "Clouds in Aerology and Forecasting," B. Amer. Meteorol. Soc. J-22, Novem-
ber, 1941, pp. 335-345.
y. 18
This relationship holds for
the northern hemisphere.
In the southern hemisphere
the directions of decreasing
and Increasing temperature
In relation to vertical shear
are reversed.
Approved For Release 2009/06/17: CIA-RDP89B00708R000500090002-4
 Approved For Release 2009/06/17: CIA-RDP89B00708R000500090002-4
It can be seen that since a, which depends on the sine of the latitude, will remain constant
and p decreases with height, ap/8n must also decrease for V, to remain constant.) This
decrease turns out to be very small when actual values are used. In the case of a
uniformly growing cumulus that slopes in its lower layers and then straightens or even
bends back on itself with increasing height, the decrease of the horizontal pressure
gradient with height is (as Brooks also points out) much stronger than in the previous
case. If one assumes that the slope of vertically developed clouds may be observed from
350 mi altitude (at least at the edges of scanning strip), further pressure data may
be gathered.
Degree of Stability
As has been mentioned above, the degree of stability in a given layer may be esti-
mated by the amount of vertical development present in clouds. In any mechanism of
vertical development, the stability of the air plays a "major part. Convective, orographic,
or upslope lifting may produce clouds in the absence of instability, but, for any large-
scale vertical build-up of clouds, a great tendency for the atmosphere to "overturn"
must be present. (Absolute instability is
Cass 3 aCan 2 S ? saturated adiabatic lop" rats
taken to mean that the decrease of tem- a "\' t ? dry adiabatic laps@ rata
perature with height is greater than the rase "ab absolute stability
NP ow Y ? ability
bility
dry adiabatic lapse rate. In the presence f cet. I ~~ Cass 3. conditional instability
of unsaturated water vapor, the dry
adiabatic lapse rate is about 9.8?C/km,
whereas, in the presence of saturated
water vapor the smaller saturated adia-
batic lapse rate with a nonlinear variation Fig. 4-Graphical representation of
degrees of stability as given by
of temperature is used.) In the presence lapse rate of temperature
of water vapor, the latent heat (energy)
of condensation that is released when the air is forced to rise and its moisture forced to
condense may be sufficient to continue independently the upward motion. This motion
indicates a condition of instability where none may have existed at the beginning of the
process. Continuation of this motion, therefore, indicates the instability of the air in
the presence of saturated water vapor and is evidenced in towering cumulus or cumulo-
nimbus. If, on the other hand, condensation occurs but the ascending air is not provided
with a sufficiently large amount of heat so as to warm it to a higher temperature than
that of the surrounding air, the layer is considered absolutely stable and may be char-
acterized by smooth, flat-topped cloud forms, usually arranged in layers or sheets. This
is also true when a small layer of instability is "capped" by an inversion (increase of
temperature with height). This concept of absolute stability, absolute instability, and
conditional instability (unstable or stable depending on whether the water vapor present
condenses or not) is presented graphically in Fig. 4.
It may be said that, in the presence of vertically developed clouds, a dry adiabatic
lapse rate (or very close to it) exists below the base of the cloud, a relatively steep, lapse
rate exists within the cloud, and a relatively stable lapse rate exists above the cloud. In
Approved For Release 2009/06/17: CIA-RDP89B00708R000500090002-4
 Approved For Release 2009/06/17: CIA-RDP89B00708R000500090002-4
- ------------
the case of flat-topped or sheet-type douds, it may be that, although instability may exist
in a small laver comprising the loud, an inversion layer of very stable air exists imme-
diately above, causing the loud to stop its vertical growth.
In his paper on clouds, Brooks" suggests the following further refinements on this:
1. Detached, lumpy loud with a flat base and rounded top has (a) adiabatic lapse
rate below it, (b) greater than saturated-adiabatic lapse rate (unstable) within
the cloud, and (c) almost the same lapse rate as (b) (unstable) from its top
to the height that the cloud will grow.
2. Towering, sharply-bounded cumuliform cloud: The diameter of cloud at differ-
ent levels is an indication of the relative steepness of the lapse rate (except in
the presence of large wind shear). "The narrower such a cloud or doudlet is,
relative to its height, the greater the lapse rate of the surrounding air."
This provides one with very rough criteria for estimating the degree of stability of the air.
To sum up, water vapor in the- air is a latent source of heat energy. When moist air is
carried rapidly upward, the water vapor condenses in the form of liquid droplets and
the latent heat of condensation is released to the surrounding atmosphere. It is this source
of latent heat that feeds thunderstorms and other types of vertically developed clouds.
Cumulus clouds are an indication of moisture and relative instability, and, conversely,
when there is moisture in the air there will be a greater tendency toward convection
and turbulence.
Clouds, being composed of water droplets, naturally indicate the presence of moisture
in the atmosphere (see the above section). Resulting from the difference in formation
conditions, cloud types can give a further breakdown of moisture distribution. For ex-
ample, cumulus and cumuliform clouds of vertical development require the entrainment
of continuous supplies of moist air to prevent their complete evaporation shortly after
forming. It can therefore be said that with this type of cloud we may associate fairly
moist air'near the surface. In like manner, positioning of the moisture in the atmosphere
may be associated with other cloud forms, and an over-all estimate may be made from
visual observations. Once the synoptic picture has been established, closer estimates may
be made utilizing the other meteorological parameters, and the value of moisture content
may be worked into the "hunting" technique previously mentioned.
Precipitation
Although it will not be possible to observe any form of precipitation directly, it is
known that the largest amounts usually fall from two main types of clouds: cumulonimbus
(showers-hail, rain, snow, etc.) and nimbostratus (steady precipitation, sleet, etc.).
Furthermore, the probability of precipitation in one form or another, which arises when-
ever these types are present, is higher than for any other types of clouds. Further infor-
Approved For Release 2009/06/17: CIA-RDP89B00708R000500090002-4
 Approved For Release 2009/06/17: CIA-RDP89B00708R000500090002-4
mation may be obtained from the fact that it may be possible to distinguish between
newly fallen snow and old snow, owing to a difference in albedos (see Table 5,
page 12), and the new snow may then be connected with the proper form of cloud
observed downwind from it.
Approved For Release 2009/06/17: CIA-RDP89B00708R000500090002-4
 Approved For Release 2009/06/17: CIA-RDP89B00708R000500090002-4
SUGGESTED METHODS OF ATTACK ON THE
PROBLEM OF DETERMINING THE
SYNOPTIC SITUATION
From the above discussion it can be seen that the analysis is based primarily on cloud
observations. During the course of this study several systematic methods of accomplish-
ing these presented themselves. Although neither time nor proper data were available
for a complete study of these possibilities, the most promising were considered and are
presented herewith as a guide to any more intensive study.
1. It is suggested that a tying of clouds as to cause rather than appearance will greatly
facilitate the identification of the synoptic situation. Classification into two main cate-
gories would constitute a possible breakdown, as follows: (a) Regional clouds (those
caused by purely local conditions), and (b) migrating cloud systems (clouds that appear
to move as a unit). This breakdown might then be coupled with a knowledge of the.
clouds associated with various weather phenomena to complete the synoptic picture.
2. It is a recognized fact that similar synoptic situations occurring under different
climatic and/or topographic conditions may produce radically different weather. A
statistical analysis is therefore suggested, in which (a) the desired area is divided into
small regions of similar climate, geography, etc., and (b) a statistical survey of cloud
types and associated weather found with various weather situations (fronts, etc.) in each
region is made.
3. Owing to the fact that identification of fronts as fronts may be very difficult, it is
suggested that it may be possible to identify, air masses from high-altitude pictures and
to utilize them in the formulation of the synoptic picture. Since general classifications of
air masses include as integral identifying features the stability of the air, the moisture,
and the type of clouds produced in a given air mass, this should not be too difficult, in
many cases. An air-mass identification has the further advantage of establishing more
closely the possible limits of the various meteorological parameters.
4. A further possible method assumes the ability to secure synoptic, essentially
perimeter, weather information from all parts of the world other than Russia. On this
basis all weather systems entering and leaving the area will be known, and a measure of
continuity will be established. It is a relatively simple matter to identify a system once
it is known that such a system is present. Once a weather situation is so identified, it
can be earmarked from high-altitude pictures, and not only may it then be tracked
throughout the desired area, but any over-all changes or modifications that affect the
visible parameters may be almost immediately noticed. It is also likely that, having a
complete analysis of the surrounding territory and many satellite observations of the
unknown area (through which it is possible to get fixes on systems and. to examine
Approved For Release 2009/06/17: CIA-RDP89B00708R000500090002-4
 Approved For Release 2009/06/17: CIA-RDP89B00708R000500090002-4
.R-218
22a
SUGGESTM 1&ODS OF ATTACK ON THE
PROBLEM OF KENIM TEE
SYNOPTIC SITUATION
From the above.....
le 000* 0
2. .....
3. .....
4. A major advantage of satellite weather observations is the re-
peated broad spatial coverage. Such broad coverage provides the meteoro-
logist with an essential element for his analysis, which is generally re-
ferred to as continuity. It permits him to follow a given system as it
moves and developes over a period of days. It is a relatively simple
matter to identify a system once it is known that such a system is present.
Once a weather situation is so identified, it can be earmarked from high-
altitude pictures, and not only may it then be tracked across an inaccess-
ible area like an ocean, but any over-all changes or modifications that
affect the visible parameters may be almost immediately noticed. It is
also likely that, having a complete analysis of the surrounding territory
on land, where observations are plentiful, and many satellite observations
of the unknown area (through which it is possible to get fixes on systems
and to examine ........
Approved For Release 2009/06/17: CIA-RDP89B00708R000500090002-4
 Approved For Release 2009/06/17: CIA-RDP89B00708R000500090002-4
visually the over-all weather picture), a complete analysis of the desired region will
become a much simpler thing to construct.
Each of the above suggestions affords excellent possibilities of providing the required
information. It should be kept in mind, however, that these suggestions appear to offer
the best solution when systematically used together.
Approved For Release 2009/06/17: CIA-RDP89B00708R000500090002-4
 Approved For Release 2009/06/17: CIA-RDP89B00708R000500090002-4
RESULTS OF THREE ATTEMPTS AT ANALYSIS
As discussed previously, an attempt at a weather analysis was made, in which aspect
photographs taken from vertically fired rockets were used. Very little information was
used other than that supplied by the applicable surface weather map with which to check
the results.
These results should not be construed as proving or disproving any of the previous
discussion, as the observations were not conducted as a test of the suggested methods.
Rather, these attempts were conducted in order to gain experience in handling the prob-
lem, and the results are presented merely for the background information.
VIKING NO. 3 (NRL)
Date of firing .................... Feb. 9, 1950
Time of firing ....................... -t444 MST
Peak altitude ..................... 49.5 mi
Time to peak..... ................ 169 sec
What Was Observed
Because of the tilt and rotation of the rocket near peak altitude, the pictures there are
divided into sequences of nine frames, each interspersed with shots of the sky. Since
each sequence views what is relatively the same area, only one set need be used. Foggi-
ness of the film prevents positive orientation or identification of clouds. From a knowl-
edge of past performance, however, the rocket is assumed to be oriented as shown in
Fig. 5, and cloud types are estimated from over-all appearance and contrast. The apparent
cloud picture is as follows:
Over the launching area there appears to be some cirrus and cirrostratus. A layer of
thicker clouds, apparently altostratus, spreads away from this area and back toward
what appears to be a line of convergence in the west and northwest. There is some
noticeable vertical development and possibly some nimbostratus closer to this line of
convergence. The vertical development observed appears both as detached masses (the
type that might be associated with orographic or convective lifting) and as that occurring
in the tops of stratiform clouds.
Synoptic Situation-Preliminary Estimate
The layer of what appears to be cirrostratus is indicative of convergence in the upper
layers and possibly indicates convergence and a low-pressure center at the surface (to the
west of the cirrostratus). In the case at hand, this is further verified by the presence of
a considerable deck of altostratus further in the distance. This type of cloud cover is
normally associated with upslope motion, and, judging by the prevailing winds aloft,
Approved For Release 2009/06/17: CIA-RDP89B00708R000500090002-4
 Approved For Release 2009/06/17: CIA-RDP89B00708R000500090002-4
Area enclosed by dotted line
thought to be approximately
'hot viewed from Viking No.3
near top of flight (49.5 mi)
?I.ounching
point
Fig. 5-Schematic representation of viewing
area from rocket camera
a line of convergence may lie in the west and northwest section of the photographic
area. As a result of the peculiar orientation of the cloud sequence, viz., cirriform clouds
in the foreground in the west and northwest with thicker clouds in the background, one
is led to believe that this line of convergence extends from the northwest and west down
toward the southern border of the United States.
Winds. Despite the fact that no direct observations of the winds (from clouds) were
available because of the poor photographic conditions, estimates can be made from the
synoptic situation.
. In the presence of a frontal condition as estimated above, the winds to the east of
this supposed line of convergence should be southwesterly to southeasterly, depending
on the distance from this line. The wind in the vertical should be primarily from the
same direction. To the west of the line of convergence, the winds would be expected
to be west or northwest in the lower layers.
Temperature. Taking into account the analyzed weather picture, estimates of the
direction of temperature gradients were made, and the results are plotted on Fig. 6.
Results of the Analysis. The results of this analysis are presented in Fig. 6, and
the actual conditions are shown in Fig. 7.
Approved For Release 2009/06/17: CIA-RDP89B00708R000500090002-4
 Approved For Release 2009/06/17: CIA-RDP89B00708R000500090002-4
cloudiness as seen from rocket
Exact geographical extent of
=- Estimated direction of
decreasing temperature
Direction of wind
assumed from
Fig. 6-Synoptic situation as estimated from Viking No. 3
photographs, February 9, 1950
Since time of flight was 1440 MST,
actual temperatures were token
to be approximately equal to
maximum temperatures for the day
Fig. 7-Actual weather situation from weather map for
February 9, 1950, 1330 EST
Approved For Release 2009/06/17: CIA-RDP89B00708R000500090002-4
 Approved For Release 2009/06/17: CIA-RDP89B00708R000500090002-4
V-2 NO. 28 (AFCRL)
Date of firing .................... Dec. 8, 1947
Time of firing .................... 1440 MST
Peak altitude ..................... 63 mi
What Was Observed
Films were very clear, and data were available from two cameras located in different
parts of the rocket. Details such as landmarks, roads, airfields, and towns were partially
identifiable, and the general tilt of the rocket and the scope of the photographic sweep
were ascertained. The over-all cloud picture is as follows:
To the west and west-northwest lies a very extensive layer of altostratus. To the east
of this layer and in the immediate west and northwest foreground can be seen patches
of cumulus and stratocumulus. This layer stretches completely to the horizon in the north-
west (approximately 500-700 mi) and either this or an adjoining sheet of clouds can
be seen on the horizon in north and north-northeast quadrants. The area within 75- to
100-mi radius of the launchinglife is almost completely clear of clouds. Various widely
separate cloud elements (cumulus, lenticular clouds) can be seen throughout the camera
sweep in other directions but they appear to be caused by local or topographic conditions.
Synoptic Situation-Preliminary Estimate
As in the previous example, it appears that an area or line of convergence or upslope
motion lies quite far to the west and northwest of the launching area (approximately
500 mi). The position of the area of complete clearness and the estimated position of
the line of convergence suggest that the low-level flow in the foreground region is from
the south and southwest. To the north and northeast, the layer of clouds that appears on
the horizon seems to be of a very thin nature (cirrostratus or thin altostratus).
Temperature. As in the previous case, directions of temperature gradients were esti-
mated to fit the existing situation and are plotted on Fig. 8.
Results of the Analysis. The estimated and actual situations appear in Figs. 8 and 9,
respectively (page 28).
AEROBEE A-7 AND V-2 NO. 40
On July 26, 1948, two rockets were fired approximately one hour apart. The unusual
quality and quantity of clear photographs obtained from these firings merit their mention
in this report. At first glance there appears to be no occurrence of great meteorological
interest, but on closer inspection a clear picture of synoptic interest becomes evident.
Date of firing ..................... July 26, 1948
Time of firing 0947 MST
Peak altitude ..................... 70 mi
Time to peak ..................... 189 sec
Approved For Release 2009/06/17: CIA-RDP89B00708R000500090002-4
 Approved For Release 2009/06/17: CIA-RDP89B00708R000500090002-4
? Area of major cloudiness
as seen from rocket
p Exact geographical extent of major
cloud Zone is not known
a- Estimated direction of decreasing
temperature
assumed from theo-
retical circulation
around system
Fig. 8--Estimated synoptic situation from V-2 No. 28, December 8, 1947
Fig. 9-Actual synoptic situation from surface map for
December 8, 1947, 1330 EST
Approved For Release 2009/06/17: CIA-RDP89B00708R000500090002-4
 Approved For Release 2009/06/17: CIA-RDP89B00708R000500090002-4
Date of firing .................... July 26, 1948
Time of firing .................... 1103 MST
Peak altitude ..................... 60.3 mi
Tian to peak ..................... 205 sec
Due to the dose proximity of the two flights their data are considered as one unit.
What Was Observed
To the north can be seen scattered cumulus and cumulonimbus over the mountains,
with a deck of altostratus or altocumulus in the distance, just south of Albuquerque and
spreading west as far as Flagstaff, Arizona (this appears to be a possible line of con-
vergence and may indicate the leading edge of a front at that level, 10,000-15,000 ft).
To the immediate east lies a relatively clear area, while to the northeast and background
east can be seen further evidence of an altostratus cloud deck. This appears to indicate
that this possible line of convergence may be discontinuous to the immediate east, appear-
ing again several hundred miles further east. To the southeast can be seen occasional
patches of stratus and altostratus,-indicating possible low-level flow from the Gulf of
Mexico. There is a widespread cloud cover to the south and southwest (Mexico) which
consists of altostratus and stratocumulus, cumulus and cumulonimbus being seen over
the mountains. To the west there is considerable cumulonimbus activity, with what appears
to be a large deck of cirrostratus covering part of Arizona. The towering cumulonimbus
in the west and southwest appear to slope toward the north and northwest, and those in
the south appear to slope toward the west. This, combined with the "torn-off" direction
of the tops, seems to indicate that the upper-wind direction is from the south, southeast,
and east, respectively. Over the mountain range immediately to the northeast of the
launching site, small cumulus clouds can be seen streaming from the peak tops in an
east-northeasterly direction, indicating that winds at that level (10,000 ft) are from
the west-southwest.
Synoptic Situation Analysis
There appears to be no major low-pressure center, except possibly to the northeast out
of range of the photographs and associated with the apparent line of convergence to the
north of the launching site. From the apparent directions of the wind (deduced from
clouds), one possible analysis of the situation is that a large high-pressure area covers
most of the southwest (Arizona, New Mexico, etc.) in the upper layers (20,000 ft),
while in the lower layers (4000-10,000 ft) the center of this high is shifted east. From
the amount of cloudiness to the south and southwest and the apparent direction of the
wind in that quadrant, there is a possible indication of a warm, moist tongue of air
advancing from the southeast and recurving back to the north and east. Directions of
temperature gradients are estimated by using the shear as indicated by the slope of the
cumulonimbus clouds (this method has been explained previously). The results of this
analysis are presented in Fig. 10, and the actual synoptic picture is shown in Appendix I.
As indicated previously, only the photographs were used in performing this analysis.
In an effort to associate weather parameters and features more closely with those visible
Approved For Release 2009/06/17: CIA-RDP89B00708R000500090002-4
 Approved For Release 2009/06/17: CIA-RDP89B00708R000500090002-4
Exact geographical extent of
major cloud zone is not known
~. Estimated direction of
decreasing temperature
Direction of wind assumed from
theoretical circulation around system
Fig. 10-Estimated synoptic picture from V-2 No. 40 and
Aerobee A-7, July 26, 1948
in a rocket photograph, Professor J. Bjerknes, of the University of California at Los
Angeles, has reanalyzed the flights of July 26, 1948, using all the synoptic information
available. His results are presented in Appendix I. Also included in Appendix I is an
illustrative mosaic of the actual photographs taken on these two flights.
COMMENTS ON THE PRECEDING ANALYSES
Although some of the discrepancies in the analyses may be explained by the difficulty
usually encountered in locating fronts in mountainous terrain, it is obvious that much
more than a simple look at a loud photograph is needed. It is not necessary to elaborate
on these needs, as the suggested methods of satisfying them have been discussed in the
previous sections. However, one can easily see how applications of such items as
perimeter weather, etc., to these attempted analyses would have greatly improved the
final results. The reader should also keep in mind the fact that each observation was
made from a vertically fired rocket, and its field of view was, therefore, comparatively
limited. In the case of the satellite this limitation would be partially removed.
Several interesting things are to be observed from these analyses, such as the follow-
ing: (1) The fact that by simply using the analyzed circulation (both theoretical and
observed) and a knowledge of regional climatology, one was able to obtain good estimates
Approved For Release 2009/06/17: CIA-RDP89B00708R000500090002-4
 Approved For Release 2009/06/17: CIA-RDP89B00708R000500090002-4
of the actual temperature-gradient direction; and (2), that these photographs would be
of considerable assistance to normal means of forecasting. It was noticed, for example,
that in photographs from the three latter rocket flights one was able to see a layer of
cirrus clouds in the distance (denoting convergence aloft) that later (within 12 to 24 hr)
resulted in regions of widespread precipitation.
Obviously many more photographs of rocket flights exist than have been analyzed
here. However, for various reasons it was not possible to perform a full analysis on a
great majority of them. All the available photographs are listed in Appendix II, show-
ing their degree of usefulness to this study.
CONCLUSION
In the section entitled "What Can Be Seen," page 3, it was shown that, given at
least 500-ft resolution, it was possible to differentiate between the various types of clouds.
Under "Limitations of the Analysis," page 13, the possible limitations to the type of
analysis to be studied were indicated. Given the identity of virtually all the cloud forms
viewed, it was further shown, in the section entitled "What Can Be Determined from
High-Altitude Observations," page 14, that it may be possible to estimate the various
meteorological parameters under certain conditions and assumptions. The main assump-
tion was that some estimate of the over-all synoptic situation could be made initially and
a "hunting" technique could be applied. Several suggested methods of estimating the
synoptic picture were presented and discussed.
This report has attempted to show what is thought to be necessary in the making of
such an analysis. It is obvious, however, that, with the limited data available, many
important points may inadvertently have been overlooked. An inquiry of this type can
therefore serve only as a guide to a full-scale study of the subject, in which every sug-
gestion and method is put to a full test and is either accepted, modified, or discarded.
The development of all the suggested methods mentioned in this report appears to
be feasible. As any analysis depends on its integral parts for its accomplishment, from
this standpoint, if from no other, the analysis of synoptic weather from satellite obser-
vations is also feasible. -
Approved For Release 2009/06/17: CIA-RDP89B00708R000500090002-4
 Approved For Release 2009/06/17: CIA-RDP89B00708R000500090002-4
DETAILED ANALYSIS OF SYNOPTIC WEATHER AS OBSERVED
FROM PHOTOGRAPHS TAKEN ON TWO ROCKET FLIGHTS
OVER WHITE SANDS, NEW MEXICO,
JULY 26, 1948
BY J. BJERKNES
Department of Meteorology, University of California
at Los Angeles
THE FRONTAL CLOUDS
In. the composite picture from the highest altitude of V-2 No. 40, a strip of flat loud
(actually altostratus and altocumulus) can be seen in the northwest quadrant, extending
from the Rio Grande 300 mi west-to Flagstaff, Arizona. That cloud strip can be identi-
fied as being the remnant cloud of the trailing end of a cold front from the cyclone,
north of the Great Lakes in Fig. 11. The outline of the main areas of alto cloud and
cirrus cloud as seen from the highest levels of the rocket flights has been entered on
the map in this figure.
Rocket site 1012.5 1012.5 1015 1017.5 .?'
ACu -Altocumulus
Fig. I1-Surface weather map for July 26, 1948, 1130 MST
Approved For Release 2009/06/17: CIA-RDP89B00708R000500090002-4
 Approved For Release 2009/06/17: CIA-RDP89B00708R000500090002-4
In the Rio Grande valley, the southern sharp edge of the alto cloud sheet is located
near Socorro, New Mexico, while the more irregular northern edge lies about 50 mi
farther upriver in the region south of Albuquerque, New Mexico. The front cloud is
thin, shows holes and cracks at many places, and does not seem to give any frontal
precipitation. Nevertheless, the strip of frontal cloud marks an important air-mass limit
between subsident polar air to the north and partly thundery tropical air to the south.
Presumably also of frontal origin are the cirrus clouds, which can be seen to extend
in a zone near the Mexican border parallel to the strip of altostratus-altocumulus. The
zone of cirrus seems to come to an end over southwestern Arizona.
For the reader with access to the complete sequence of pictures taken from the Aerobee
(firing time 9:47 A.M., MST) and the V-2 (11:03 A.M., MST) the detailed description
of the frontal-cloud cirrus, altocumulus, and altostratus on pages 40 through 42 may
be useful. The study of the pictures from low altitude in conjunction with those from
the higher altitude is a help toward the right interpretation of the unfamiliar look of
clouds from 60 to 70 miles elevation.
The distribution of the frontal cloud, viz., altocumulus, altostratus, and cirrus, can be
understood from an inspection of the maps of upper winds in Fig. 12. The line of
convergence, marking the front, goes through New Mexico .on-the 14,000-ft map just
where the strip of alto cloud is seen on the composite photograph in Fig. 13. While
there is a general afflux of tropical air to the line of convergence at 8000 ft, the flow
picture is more complicated south of the front at 14,000 ft, owing -to the splitting up
of the subtropical high into two cells, one over northern Mexico and one over the
Mexican Gulf. West of the Mexican high, the tropical air continues to flow toward the
front, thus keeping the strip of alto cloud, from Flagstaff, Arizona, to the Rio Grande,
narrow and well defined. Farther east, the tropical air partly branches off southward
around the Mexican high, and what there is available of frontal cloud will be thinning out
and spreading over a large area (such as that shown by the northeast-quadrant pictures).
The cirrus zone through southern Arizona and New Mexico lies along a line of con-
vergence between northerly and easterly winds on the 30,000-ft map. This line of con-
vergence at 30,000 ft is not the same as that shown at 14,000 ft between the cold high
over Utah and the warm high over northern Mexico. Already, at 18,000 ft (not repro-
duced), the cold high has disappeared, and, farther up, the warm high is found to tilt
northwestward so as to be centered over northwestern New Mexico at 30,000 ft. The
origin of the cirrus probably must be sought over the front a day or two earlier, when
it must have had the usual high build-up of cloud typical of rainy fronts. The light
northerly winds to the east of the 30,000-ft high-pressure center must have separated
the cirrus from the stagnating frontal altocumulus strip over northern New Mexico and
herded them into the zone of convergence between northerlies and easterlies at the
Mexican border. The westward views therefore show the cirrus and altocumulus in two
parallel zones separated by a 150-to-200-mile-wide cloudless space. The eastward views,
on the other hand, cover an area where, in the, distance, the altocumulus is drifting
southward as "prefrontal cloud" with about the same forward edge as that of the cirrus.
Approved For Release 2009/06/17: CIA-RDP89B00708R000500090002-4
 Approved For Release 2009/06/17: CIA-RDP89B00708R000500090002-4
Fig. 12-Seleded
upper-air wind
maps
Approved For Release 2009/06/17: CIA-RDP89B00708R000500090002-4
 Approved For Release 2009/06/17: CIA-RDP89B00708R000500090002-4
gco~E ;t i.5 g
S9=z
c
N S: c N Q
20 u X,3
FS 2Z~~9 aim
N it C? i $ C Z<
cc e
1-0 0
~ ?O too ` V ?. ~ ~
h2 Z0^?'uc pa
i. o E 89 -.2 0
ap ^ 2 .'. c
C _ > pro
p ? v c E
cmg a
c ?`J ? C r
d ... 4 c 0 w 0
? p P (~ c
C6o0
? ? < S ? O ? c i <
y w
O o O c O
? p, a..
C i .O,.H ~N O.
O 1 O ? C =
..~ . 0 '< -
Ow 9_$
O C ? O/._9
h : - yj - 5' Ec
p i e
`
epi'>3 V 7 r- C ; C Z O_
0 0 We 0
E
~?n0A >d GZ ? O
r ? ??= E - _ E - ?
p' c c w
Z 6o`2 `p