PRELIMINARY MEASUREMENTS OF ATMOSPHERIC TURBULENCE AT HIGH ALTITUDE AS DETERMINED FROM ACCELERATION MEASUREMENTS ON LOCKHEED U-2 AIRPLANE

• Document Type: CREST
• Collection: General CIA Records
• Document Number (FOIA) /ESDN (CREST): CIA-RDP33-02415A000200350103-6
• Release Decision: RIPPUB
• Original Classification: K
• Document Page Count: 16
• Document Creation Date: December 12, 2016
• Document Release Date: October 7, 1998
• Sequence Number: 103
• Content Type: REPORT

Approved For Release 2001/09/07 : CIA-RDP33-02415A000200350103-6
NACA RM L57All
NATIONAL ADVISORY COMMITTEE FOR AERONAUTICS
PRELIMINARY MEASUREMENTS
OF ATMOSPHERIC TURBULENCE AT HIGH ALTITUDE AS
DETERMINED FROM ACCELERATION MEASUREMENTS ON
25X9A5 LOCKHEED U-2 AIRPLANE
An analysis of a sample of turbulence data obtained from VGH records
taken on Lockheed U-2 airplanes during research flight up to 55,000 feet
over England and Western Europe has indicated substantial reductions in
the number and intensity of gusts with increasing altitude. These results
on the variation of atmospheric turbulence over England and Western Europe
were found to be in overall agreement with previous turbulence data obtained
from airplane- and balloon-borne instruments over the United States.
The collection of detailed information at high altitudes on atmos-
pheric turbulence and other meteorological conditions has, in the past,
been largely dependent upon the availability of operational airplanes.
As a consequence, the collection of such data has, in general, lagged
behind the development of airplane altitude capabilities. To the present
time, the available airplane measurements of atmospheric turbulence have
been almost entirely limited to flight altitudes below approximately
45,000 feet. For higher altitudes, the only information available is
some measurements obtained by means of balloon-borne instruments. (See
ref. 1.) These measurements have provided limited information on tur-
bulence variations with altitudes to approximately 60,000 feet. In addi-
tion to the altitude limitations of the foregoing investigations, the meas-
urements were also limited in regard to geographic areas and were largely
confined to continental United States. This lack of information on high-
altitude turbulence conditions around the world has been a handicap in
aircraft design studies and in operational analysis, both in regard to
aircraft load problems and in regard to stability and control problems.
Approved For Release 2001/09/07 : CIA-RDP33-02415A000200350103-6
 Approved For Release 2001/09/07 : CIA-RDP33-02415A000200350103-6
2 - NACA RM L57A11
During the early part of 1956, the National Advisory Committee for
Aeronautics, in cooperation with the Air Weather Service of the United
States Air Force, initiated a research program aimed at providing detailed
meteorological information both at higher altitudes than those covered by
present-day operating airplanes and for various geographic areas of the
world. The availability of Lockheed U-2 airplanes at the initiation of
this program permitted the coverage of flight altitudes to approximately
55,000 feet. The NACA activity in this program has so far been primarily
aimed at obtaining information on the amount and intensity of atmospheric
turbulence at these higher altitudes. The Air Weather Service has simul-
taneously provided instrumentation to collect data on humidity, pressure
variations, and winds.
The initial research flights of the U-2 airplane were undertaken to
cover two geographic areas, the United States and Western Europe. Because
of operational difficulties, a statistically significant sample was ini-
tially obtained only from the operations over England and Western Europe.
For these operations, measurements covering a total of 22,000 flight
miles were made and evaluated. Although this sample is small, the initial
results appear to be of sufficient interest to warrant publication. Accord-
ingly, these results are presented herein and are compared with the earlier
estimates on the variation of atmospheric turbulence with altitude given
in reference 1.
INSTRUMENTATION AND SCOPE OF DATA
The flight measurements were obtained during flights of several
Lockheed U-2 airplanes for the high-altitude meteorological research
program of the NACA in cooperation with the Air Weather Service. The
Lockheed U-2 is a subsonic, straight wing, single-engine airplane
originally designed for use as a high-altitude test medium for engine
and aircraft-component testing. The high-altitude performance and economy
of operation of the U-2 airplane were the prime factors affecting its
selection for use in the present research program.
The measurements pertinent to this report consisted in time-history
records of airspeed, acceleration, and pressure altitude taken with NACA
VGH recorders (ref. 2). The time histories were recorded on photographic
paper moving at 8 inches per minute. Records were obtained on 17 flights
during operations from bases at Lakenheath, England, and Wiesbaden, Germany,
between May and September 1956. The flight plans used in the operations
consisted in climbing to an altitude of approximately 45,000 feet, cruising
initially at approximately 45,000 feet with the altitude gradually increasing
to approximately 55,000 as the fuel load decreased, and then descending to
the home base. For these initial operations, the flight schedules were
based primarily on airplane availability, and no attempt was made to schedule
Approved For Release 2001/09/07 : CIA-RDP33-02415A000200350103-6
 Approved For Release 2001/09/07 : CIA-RDP33-02415A000200350103-6
NACA RM L57A11 3
flights to sample turbulence associated with specific meteorological
conditions. The turbulence encountered during the present operations
was, for the most part, in clear air, and no turbulence in heavy cumulus
clouds or thunderstorms is represented in these data.
The scope of the data in terms of miles flown within different alti-
tude intervals is summarized in the following table:
Pressure
altitude,
ft
Flight miles
20,000
to 30,000
1,548
30,000
to 40,000
3,492
1+0,000
to 50,000
6,021
50,000
to 55,000
10,461
Total
21,522
Because of the operational procedures used, only a small amount of infor-
mation was obtained at the lower altitudes; consequently, only the data
above an altitude of 20,000 feet are included in this paper.
The VGH records were evaluated to obtain the vertical gust veloci-
ties, the percent of rough air at various altitudes, and the horizontal
extent of the turbulent areas encountered. The evaluation procedures
are similar to the procedures used in reference 1 and are discussed
briefly in the following paragraphs.
The vertical gust velocities were derived from simultaneous readings
of acceleration, airspeed, and altitude through the use of the gust equa-
tion which is given in reference 3 as
Ude -
2anW
mpoKgVeS
where
Ude derived gust velocity, fps
an peak normal acceleration, g units
W airplane weight, lb
Approved For Release 2001/09/07 : CIA-RDP33-02415A000200350103-6
 Approved For Release 2001/09/07 : CIA-RDP33-02415A000200350103-6
4 NACA RM L57A11
S wing area, sq ft
Kg gust factor
Ve equivalent airspeed, fps
m wing lift-curve slope, per radian
Po air density at sea level, slugs/cu ft
In evaluating the records, the accelerations were read to a threshold
sufficiently low to yield complete frequency counts of all gust velocities
greater than 2 feet per second. Values of airspeed and pressure altitude
were obtained from the records for each acceleration evaluated. The in-
flight weight loss was accounted for in determining the values of wing
loading W/S for use in the equation. Appropriate values of the gust
factor Kg were computed for each part of the record where rough air was
encountered. The values of the lift-curve slope m used in deriving the
gust velocities were based on data from the airplane manufacturer.
The gust-velocity values presented herein may be open to some question
because of effects of airplane flexibility and stability on the accelera-
tions from which the gust velocities were computed. The magnitude of these
effects is not known at present, and additional work is required before
their influence on the gust-velocity values can be assessed.
In addition to determining the gust velocities, the VGH records were
evaluated to obtain the horizontal extent of turbulent areas and the per-
cent of rough air at the different altitudes. For the purpose of deter-
mining the horizontal extent of the turbulent areas, the airplane was
considered to be in rough air whenever the accelerometer trace was con-
tinuously disturbed and contained accelerations corresponding to gust
velocities greater than 2 feet per second; this threshold corresponds to
that used in previous gust studies such as reference 1. The length of
each turbulent area was found simply by multiplying the true airspeed by
the time spent in the rough air. The summation of the lengths of the
individual areas of rough air was divided by the total flight distance for
given altitude intervals in order to obtain the percent of rough air for
that interval.
The gust velocities derived from the acceleration and airspeed data
are presented as frequency distributions in table I for four altitude
Approved For Release 2001/09/07 : CIA-RDP33-02415A000200350103-6
 Approved For Release 2001/09/07 : CIA-RDP33-02415A000200350103-6
NACA RM L57A11 5
intervals between 20,000 and 55,000 feet and are shown in figure 1 as
cumulative frequency distributions per mile of total flight for each
altitude interval. The cumulative distributions give the average num-
ber of gusts per mile of total flight which exceeded given values of
gust velocity. Inspection of the results of this investigation in fig-
ure 1 shows that, over the altitude range covered, variations on the
order of 10 to 1 exist in the frequency with which given gust velocities
were encountered. In general, the distributions depict a consistent
decrease in gust frequency with increasing altitude. The data for the
altitudes of 30,000 to 1+0,000 feet deviate from this pattern to some
extent, however, and indicate a lower frequency for the higher gust
velocities than would be expected from the general pattern. Because of
the limited sample size, this deviation from the general pattern may not
be real, but due to a sampling error.
For comparison with the present data, the cumulative frequency
distributions of gust velocity were determined from the results of ref-
erence 1 and these distributions are also shown in figure 1. These dis-
tributions are based on the basic distribution of non-thunderstorm tur-
bulence and the variation of gust intensity with altitude given in fig-
ures 5 and 1, respectively, of reference 1. Comparison of the two sets
of distributions in figure 1 shows that the slopes of the two sets of
distributions are about the same but that, in general, the data of this
investigation indicate somewhat lower gust frequencies than those obtained
from reference 1. One possible reason for the lighter level of turbulence
for the present data is the limited seasonal coverage, the present data
covering only the summer months; whereas the results from reference 1
represent average turbulence conditions for operations throughout the
year. More severe clear-air turbulence may be anticipated during seasons
of the year other than summer and particularly during the winter months
when the winds at high altitude are likely to be the strongest.
Intensity of Turbulence
In order to compare in further detail the gust measurements of this
investigation with the results from reference 1, it is helpful to con-
sider separately the variations in the intensity and in the amount of
rough air with altitude. As an indication of the severity of the rough
air alone, the cumulative frequency distributions of gust velocity per
mile of flight in rough air are plotted in figure 2 for the different
altitude intervals. For comparison, the corresponding gust distributions
for non-thunderstorm turbulence were calculated from reference 1 and are
also shown in the figure. Inspection of the present data in figure 2
shows that, in general, the gust frequency per mile of rough air also
decreased with increasing altitude. However, again, the data for the
altitudes of 30,000 to 1+0,000 feet do not follow this pattern and, in
this case, indicate a lower frequency than is shown for the other
Approved For Release 2001/09/07 : CIA-RDP33-02415A000200350103-6
 Approved For Release 2001/09/07 : CIA-RDP33-02415A000200350103-6
6 NACA RM L57A11
altitudes. Examination of the time-history records revealed that this
low gust frequency resulted mainly from two long (30 to 50 miles) areas
of turbulence of low intensity. In view of the peculiarity of the data
for the altitudes of 30,000 to )+0,000 feet, the present indication of a
low gust frequency for this altitude interval is open to question.
Comparison of the results of this investigation on the gust inten-
sity, based on miles of flight in rough air, with the corresponding
results from reference 1 shows not only that the two sets of distribu-
tions have approximately the same slopes but also that the gust inten-
sity is considerably lower for the present data. (See fig. 2.) For a
given frequency of occurrence, the gust velocities shown by the present
data are, in general, only about 75 percent as large as the gust velocities
given by the results from reference 1 for the same altitude interval.
For the altitude of 30,000 to 40,000 feet, the percent is even lower,
it being about 50 percent. Thus, in general, the intensity of the rough
air for the data of this investigation is substantially lower than that
given by reference 1.
In addition to the direct comparison of the gust intensities given
in the preceding paragraph, it is of interest to compare the relative
variations of the gust intensities with altitude for the two sets of gust
distributions. In figure 3, the variations of the gust intensity with
altitude for the present results and the results based on reference 1
are given in terms of the ratio of the maximum gust velocity expected
in a given flight distance in rough air at the lower altitude interval
(20,000 to 30,000 feet) to the maximum gust velocity for the same flight
distance in rough air at higher altitudes. The data in figure 3 were
obtained by determining from the gust distributions of figure 2 the maxi-
mum gust velocity expected in a given flight distance in rough air at
different altitudes for each sample of data. For this purpose, a gust
frequency of 0.05 per mile in figure 2 was selected as being through the
more reliable range of the present distributions. The values of maximum
gust velocity obtained from figure 2 for each set of distributions were
then normalized to the gust velocity for the lower altitude interval in
order to obtain the gust-velocity ratios plotted in figure 3.
Figure 3 shows that the relative variation of the gust intensity
with altitude for the present data is in good agreement with the results
from reference 1 except for the data for the altitudes of 30,000
to -I-0,000 feet. As previously noted, however, the reliability of the
data at this altitude is open to some question. The comparison presented
in figure 3 implies that the variation in gust intensity with altitude
over Western Europe, at least for the summer season, is similar to that
indicated in reference 1 for this country.
Approved For Release 2001/09/07 : CIA-RDP33-02415A000200350103-6
 Approved For Release 2001/09/07 : CIA-RDP33-02415A000200350103-6
NACA RM L57A11 7
The percent of the flight distance within each altitude interval
which was in rough air is presented in figure 4. Similar results, based
on airplane and telemeter data from reference 1, are also given in the
figure for comparison. Figure 4 shows that the results of this investi-
gation are in good agreement with the results from reference 1 for alti-
tudes above 35,000 feet. Below this altitude, the present data indicate
slightly higher percentages of rough air than do the data from refer-
ence 1. The high percentage of rough air for the altitudes of 30,000
to 35,000 feet resulted from the two areas of 30 to 50 miles of turbulence
of low intensity that were mentioned previously. Thus, in general, the
amount of rough air at the various altitudes is fairly close to that
given in reference 1, with the principal difference between the present
results and those from reference 1 being associated with the less severe
intensity of the turbulence noted previously.
The distribution of the horizontal extents or lengths of the turbu-
lent areas, determined from the acceleration records, is given in fig-
ure 5 as the percent of the total number of areas within class intervals
of 10 miles. The distribution is based on 115 turbulent areas encountered
during the present operations between 20,000 and 55,000 feet. The dis-
tribution of turbulent areas given in reference 1 is also shown in the
figure for comparison. For the present data, approximately 75 percent
of the turbulent areas were less than 10 miles in length and less than
2 percent of the rough areas exceeded a length of 40 miles. Additional
analysis of the data indicated that the distributions of the lengths of
turbulent areas did not vary significantly with altitude for altitudes
between 20,000 and 55,000 feet. Figure 5 shows that, on the average, the
sizes of the turbulent areas for the present data appear to be somewhat
smaller than those given by the results of reference 1. The smaller sizes
for the present turbulent areas may also be associated with the limited
seasonal coverage of the data of this investigation.
An analysis of a small sample of turbulence data obtained from NACA
VGH recorders during research flights of the Lockheed U-2 airplanes to
attitudes of 55,000 feet over England and Western Europe during the summer
of 1956 has indicated substantial reductions in the number and intensity
of gusts with increasing altitude. The results indicated that the length
of turbulent areas was less than 10 miles for about 75 percent of the
areas encountered and that less than about 2 percent of the areas of rough
Approved For Release 2001/09/07 : CIA-RDP33-02415A000200350103-6
 Approved For Release 2001/09/07 : CIA-RDP33-02415A000200350103-6
8 NACA RM L57A11
air extended more than 40 miles. These results on the intensity, amount,
and extent of atmospheric turbulence were found to be in overall agree-
ment with the results given in NACA RM L53G15a for operations over the
United States. The principal difference between the results of the two
investigations was the indication that, on the whole, the turbulence
encountered during the surveys over England and Western Europe was about
25 percent less severe than that for operations over the United States.
Langley Aeronautical Laboratory,
National Advisory Committee for Aeronautics,
Langley Field, Va., December 17, 1956.
1. McDougal, Robert L., Coleman, Thomas L., and Smith, Philip L.: The
Variation of Atmospheric Turbulence With Altitude and Its Effect
on Airplane Gust Loads. NACA RM L53G15a, 1953-
2. Richardson, Norman R.: NACA VGH Recorder. NACA TN 2265, 1951.
3. Pratt, Kermit G., and Walker, Walter G.: A Revised Gust-Load Formula
and a Re-Evaluation of V-G Data Taken on Civil Transport Airplanes
From 1933 to 1950. NACA Rep. 1206, 1954. (Supersedes NACA TN's 2964
by Kermit G. Pratt and 3041 by Walter G. Walker.)
Approved For Release 2001/09/07 : CIA-RDP33-02415A000200350103-6
 Approved For Release 2001/09/07 : CIA-RDP33-02415A000200350103-6
NACA RM L57A11
TABLE I.- FREQUENCY DISTRIBUTIONS OF DERIVED GUST VELOCITY BY ALTITUDE
Frequency distribution for altitudes, ft, of -
Gust velocity,
Udef fps
20,000 to
30,000 to
40,000 to
50,000 to
30,000
40,000
50,000
55,000
2 to 2.9
94
84
70
74
3 to 3.9
44
42
51
28
4 to 4.9
21
10
12
12
5 to 5.9
8
4
5
3
6 to 6.9
4
1
4
3
7 to 7.9
2
--
--
--
8 to 8.9
2
--
1
1
9 to 9.9
--
--
--
--
10 to 10.9
1
--
1
--
Total
176
141
144
121
Approved For Release 2001/09/07 : CIA-RDP33-02415A000200350103-6
 Approved For Release 2001/09/07 : CIA-RDP33-02415A000200350103-6
NACA RM L57A11
4 p /2 20
6115 1 Ve /oC/75" 4'e
f VS
Figure 1.- Comparison of present results on the frequency of exceeding
given values of gust velocity per mile of total flight with results
from reference 1.
Presen/ Ref. / A/fi/ude, fl
0 20 fo 3o X/o3
^ ---- 30to40
O - 40/050
o -- 50to55
Approved For Release 2001/09/07 : CIA-RDP33-02415A000200350103-6
 Approved For Release 2001/09/07 : CIA-RDP33-02415A000200350103-6
NACA RM L57A11
/0
4 8 /2 /6
Gast velocity) aae f,os
20
Figure 2.- Comparison of present results on the frequency of exceeding
given values of gust velocity per mile of rough air with results from
reference 1.
Present Ref. / Altitude, ft
0 20 /o 30 x /03
o ---- .301040
40 to So
o - - .50 fo .SS
\
\
\
Approved For Release 2001/09/07 : CIA-RDP33-02415A000200350103-6
 Approved For Release 2001/09/07 : CIA-RDP33-02415A000200350103-6
L2 NACA RM L57All
o Pr~Seu /
Ref /
.6
Ude of o/z','/ de
5000 fi`
/.U
Figure 3.- Variation in relative gust velocities with altitude.
Approved For Release 2001/09/07 : CIA-RDP33-02415A000200350103-6
 Approved For Release 2001/09/07 : CIA-RDP33-02415A000200350103-6
NACA RM L57AII 13
0
O
U
130
Approved For Release 2001/09/07 : CIA-RDP33-02415A000200350103-6
 14
Approved For Release 2001/09/07 : CIA-RDP33-02415A000200350103-6
NACA RM L57A11
80
bo
40
20
f'reser7f da/a
Refere/7c~ /
2? 40 60 80 /00
L ellglh, rn//es
Figure 5.- Distribution of the lengths of turbulent areas for altitudes
between 20,000 and 55,000 feet.
Approved For Release 2001/09/07: CIA-RDP33-02415A0002'?d3501ED341d. V-
 prod T& Release 2001/09/07: CIA-RDP
a"~voo
U,
~--E aE~ ~a
a Oor zyy
- "~ N M M G FI
I N I C; a F W a al rOr 'O N
y N a C ?ay q +a+ 1
!q "o
Cc Iw
CD 0
C9 0'
o P,
e
d
~ a ala
ca a~i a W q q R co A
89 P. 8
FNL" ag,N ? z
u -.4 `~ D+ V '~ Mi a F q7 ilifl 'O O
ua w d O d 415 d a
a~y~~Q~Wi Cu Ua ~,vC pbgS~Tj
A adz ?~ ,~ p ~ a S. 0
zzWNOU tO+wW a w v
3-024100200350103-6
'-
o ti
a0
E~
tiv
~0 U
0 Oor4z
N N M??Ib~
1 d
(D 0 10 1.
p er' W p '0 O Cd ti
N a)Ob y
o
F G4
J7
- U C 0 00
~OE NV t?'. rt > 3
w d V Z Cdd w OOW 0 0/ ?
y ?. (OL) ? O 0
Aw a 0y y p ~"
U47WNkci? y aR. 10
r.) 2
d N ~0~7+D?~giagFy
yy
R-1 W W N :' ICJ d fem. -0 D 01
z aa cU ., o dv 3
M .0
zzaN 0V flijiflItfi $.
L
a W Fi x J7 q y O W o, 14
~~JsJ{1 (~~r a i c. O
aFZ y uF a Cd 0
mgwr lD 0 W o'?
~ A y~ ~~d?iaoMq~oaai
ccOd [a Ri N `~ w a :p ccd F1
r4J ID 'a
Wo ~E"rn g Na ~ ca P?Nd'~' a ?a
N N v c?. N
a m
a R
z
O h ca a) a W q a ?~ N
ti C O m~wO60t0w,S: w...Oi U
d ?y a y i d
U @ N G, En 7 a A w CI +~' > a A z
F, p U a O N ~ E1
D+t.7 'd a1 A a O
" y
uo
F
a
et 7Ud
i 0(
"~
a
i
(a
pr ~ Fie ea 2 ~ -RD 3-4
zza~o~
-w z
Olin I
p4Fx ?? y q o
w w
z MU c PM 9 v' y ay v
'
?
H 4.
d' ao a c
0003
`1~ 0 0
O i p 0 N
U
U ~ha+wW cdc2Uw!s
 pprliie&r Release 2001/09/07: CIA-RDIi
~
0 ~
P4 H
c7 6 Cdc~r
a c7 o?Uwz
.-1 N M 1-1 F?I M
W
EN
~fAFAE J1
Zj~~~UNwl~n
tiOa 0~ACCaa
zzP04 0
.R96u1
.M b
y r.,M Cw
t?Ia gCg C ? O v M"
p~7p y F '%a ~ `y
N O .:r 'C O +`7 C a~.
N d F G)
O D+ d '`" +"~ O A. C
Q O
.~ a O, dY f0 tlJ 'C~ R
dac 0 oa)'~ly
p+yA.Q
o bddp C CE+ o
C aFi`~ d Fb "
mca 'oa,'d
14
U O ?Y N O N
4'. FOi w W C cFd CC U w
x W . x y F c
P4 CD
ww >Fl ~~AO
C)Q a ~y m m a c d a
~~pp U N y
00
O W O1 d C C
pp
CD so.
. V] F R ^' N R~ C O
w~az V O~ y C'+7~'~ IV C4
0H a yd ?
++ ~ O~ to C y .+ O
? N ~
~ 0do,~8
Cc)
,~ F wUA U F eomo p ~E1 ~iy m
W >4 z~A1'] ,I[) mA 1 y
(D 16
33'-0241ff00200350103-6
~ ..1 F cC G
' c 0
aE-
t d CAS
~a d
.-i i ri 1-i Ci 0
0~wx~
~o?wHo~
m (D =0
C4 0
o Oar
C)) O A 'OC CF) i-j -
O it C H '~' .r?i
N b 'S a N O F G)
aw C o...~3 Ap
4) ca 00
d ~j y ? c?. C6
V~[~]4WNwIt aAw dP' a~ ti
~?1 a0 O
-. MO O N E- C O
O N
~
NpN 9 .+ 1 ...,
Lo 'o -4 A1" F ud) om
o
0G.
ee7ea2 t(Dj .
Fw a w xza
da ?)
z >?ati;" "c
?C Q035%1~elf,
.g O N
CU+~ dF+~wWC:3 S. cd Uw