. p '

1

ct)

w 0 - a n. 0 0 (3

! I

A Study of Earth Radar Returns from Alouette Satellite

bY

R. C. Chia H. H. Doemland

R. K. Moore

CRES Report 37-1

he Remote Sensing Laboratory

Supported by NASA Grant No. NSG-47

CRES

. - _-

I THE UNIVERSITY OF KANSAS CENTER FOR RESEARCH INC Illlmlll ENGINEERING SCIENCE DIVISION LAWRENCE, KANSAS

https://ntrs.nasa.gov/search.jsp?R=19680012575 2018-07-27T10:09:35+00:00Z

CENTER ‘‘I1 THE UN FOR VERS

RESEARCH IN ENGINEERING SCIENCE TY OF KANSAS LAWRENCE KANSAS

Technical Report

37-1

A STUDY OF EARTH RADAR RETURNS FROM ALOUETTE SATELLITE

R. C. Chia H. H. Doemland

R. K. Moore

April 1967

A STUDY OF EARTH RADAR RETURNS FROM ALOUETTE SATELLITE

Table of Contents

I . INTRODUCTION . . . . . . . . . . . . . . . . . . . p . 1

I1 . THEORY OF EFFECTIVE REFLECTION COEFFICIENT . p . 2

I11 . . DATA ANALYSIS . . . . . . . . . . . . . . . . . . . p . 5

IV . EXPERIMENTAL RESULTS . . . . . . . . . . . . . . . . p . 9 A . Kg/Ki Correction . . . . . . . . . . . . . . . p . 1 2

B . Attenuation of R . F . i n Ionosphere by Collision Frequency . . . . . . . . . . . . p . 1 2

C . The "Quiet Day Curve" and Its Use . . . . . p . 1 3

D . Average Diurnal Attenuation . . . . . . . . . p . 15 E . Double Ground Bounce Experiment . . . . . . . p . 1 9

(brief d i scuss ion and expected resul ts)

V . DISCUSSIONOF RESULTS . . . . . . . . . . . . . . . p . 20

VI . CONCLUSIONS . . . . . . . . . . . . . . . . . . . p . 21

VI1 . APPENDIX . . . . . . . . . . . . . . . . . . . . . . p . 22

VI11 . BIBLIOGRAPHY . . . . p . 23

I

A STUDY OF EARTH RADAR RETURNS FROM ALOUETTE SATELLITE

I. INTRODUCTION

The Alouette satellite I , designated as the 1 9 6 2 pa 1 , was constructed

by the Defense Research Telecommunications Establishment (DRTE) of the Defense Research Board, Ottawa, Ontario, Canada, and launched on September 29, 1962 into orbit by the U. S. National Aeronautic and Space Administration (NASA). experiment. apparatus for counting cosmic ray particles, for observing very low frequency radio waves , and for monitoring engineering performance. The sounder was operated on command from any one of the 13 ground s ta t ions around the world. Once s ta r ted , it functions for 1 0 minutes, generating 30 transmissions in which the radio frequency changes with time. unti l commanded on again. Table 1 [Chia, et.al. , 19651 .

Ionosphere topside sounding was the principal The Alouette satellite contains , in addition to the sounder,

It then shuts off automatically Characterist ics of the sounder are shown in

A l l the information collected by the Alouette sa te l l i t e is telemetered back to ground and recorded onto magnetic tapes .

Since its launch the Alouette operated successful ly . Initial s tud ies

of the topside sounder ionograms revealed tha t , i n addition to returns from the ionosphere, many ionograms contained good returns from the earth at frequencies above the cr i t ical frequency of the ionosphere. These ear th

re turns represented the f i r s t radar returns from the earth from satellite al t i tudes. The Center for Research in Engineering Science of The University of Kansas

undertook the study of these earth returns.

-1-

Table 1

Alouette Sounder System Parameters

System - __

TX

RX

Antenna

_ _ -- -- Parameter - - - --

Frequency sweep Pulse width Pulse repetition frequency Peak pulse power

Frequency sweep Noise figure Minimum signal detect ion

through antenna matching networks

- Remark

-. . __- - -I_

0.45 to 11.8 Mc/s in approx. 1 2 sec. 100 p sec. 67 pps. Approx. 100 W into 400 ohm load.

0.45 to 11.8 Mc/s. 8 dB.

1 9 dB above kTB.

150 f t . t ip to t i p for 0.45 - 4.8 Mc/s. 75 f t . t i p to t i p for 4.8 - 11.8 Mc/s. __. . __ _ _ _ _ _ - .

Radar observations of the earth from satellite a l t i tudes a r e potentially valuable as a tool for studying the earth itself, as a technique for evaluating

radar observations from the moon and planets and as a n aid to the des ign of

radar alt imeters for rocket vehicles landing on the ear th and planets.

11. THEORY OF EFFECTIVE REFLECTION COEFFICIENT

Radar observations at wavelengths of s l ight ly under a meter have indi- cated tha t surfaces smooth to t h i s order a re very difficult to find i n nature [Edison, et.al. , 1960 and Taylor, 19593 . surface is smooth enough to give specular reflection only i f the mean square fluctuation about the mean height is less than about a tenth wavelength [LMoore, 1957 and Hayre, 19621 . Over the s e a , t h i s condition may frequently be m e t at t he wavelength of about 30 m involved i n th i s experiment, land surfaces wil l be th i s smooth over a d i s t ance as grea t a s tha t indicated by the diameter (347 km) of the illuminated circle direct ly beneath the satellite. Consequently, it is to be expected that land echoes wi l l be primarily due to scat ter ing rather than Fresnel reflection, and also sea echoes may of ten be scat tered.

Various theories indicate tha t a

I t is unlikely that

- 2-

The power received from a rough scat ter ing surface is expressed

by

Illuminated Area

where Wt = Transmitted power

G = Antenna gain

R = Distance from the transmitter to a point in the illuminated a rea

u0 = Scattering c ros s section per unit a rea

A = Illuminated area

A = Wavelength.

For pulse-length limitation of the illuminated a rea on the ground with rectangular pu l se s and essent ia l ly constant antenna gain, t h i s integral reduces for the "altitude s igna l I' to

WtA 2 2 s," +? oodR

Wr = 2 (47r 1 R3

where h = Altitude

c = Velocity of l ight T = Pulse length.

We may express t h i s integral in terms of the angle rather than the range in which case it is

C T 2 2 -1 -ji- WAX G ptan L CJ (e) s i n 0 c o s 0 d e

2 2 J o W r =

2(4n) h

where 0 is the angle made by an element on the ground with the vertical.

-3-

Experiment and theory both indicate tha t for surfaces that are relatively smooth with respect to a wavelength, a0 decreases rapidly with increasing 0

i n the vicinity of the vertical. It is likely tha t m o s t grounds (at least outs ide

of mountain areas) w i l l sa t i s fy this criterion for 30 m wavelength. quently, most of the contributidn to the integral normally comes from angles considerably under 10 deg.

Conse-

This means

When th i s situation prevails, the upper l i m i t determined by pulse length is immaterial. That is, the radiation is essent ia l ly limited by the beam angle of the backscattering rather than by the pulse length. When th is is true an

arbitrary upper l i m i t such as 7 may be placed on the integral without significant error:

IT

2 2 - 7r WtX G lo2 0,(0) s in 2 0 d 0 wr ( 4 ~ ) ~ ( 2 h ) ~

This equation is the same as tha t for a specular reflection where the integral is the square of the Fresnel reflection coefficient. Thus, we may def ine

IT K i f f = yoT ao( 0) s in 2 0 d 0

as the square of the effective Fresnel reflection coefficient, Keff. Beam-width-limited illumination is character ized by the expression

where 0, is the l i m i t of the beam, in which the variation with height is the same as that for a specular reflection. Pulse-length limitation, on the other

hand, corresponds t o a different variation with height [Moore et. al. , 1 9 5 7 1 . -4-

Thus , use of th i s effective reflection coefficient implies beam-width l imi-

tat ions. In th i s case, the beam is se t by the scattering coefficient rather than by the width of the antenna pattern as it is in some microwave vertical-

looking radars. The use of the effect ive reflection coefficient may or may not imply tha t

Fresnel reflection does , i n fact, occur. If Fresnel reflection occurs , the s igna l should not fade i n amplitude. If scattering occurs , the s ignal should fade, as contributions from the different scatterers add success ive ly in and out of phase.

111. DATA ANALYSIS

The upper frequency l i m i t of the sounder is lower than one would prefer Because of a large number of interferring s ignals in for ground echo s tudies .

the top 2 m c of the sweep range, it w a s decided to concentrate on the frequency

range from 8 .5 to 9.5 Mc/s for ground echo studied. Figure 1 shows a simplified block diagram of the sounder. Computation

of system sensi t ivi ty can be understood better by referring to th i s diagram. Unfortunately the calibrations had to be obtained on the ground somewhat piecemeal because the antennae were not self-supporting in t h e presence of gravity. for:

Calibration curves obtained before launch were supplied by DRTE

1) Radiated power 2) Transmitter output 3) Mismatch loss 4) AGC vs. frequency 5) AGC variation with supply voltage 6) AGC vs. input.

These curves may be found in the Appendix.

19 65) the power received at the antenna for a given receiver output. Based on these calibration curves, it w a s possible to calculate [Chia,f%, al. ,

-5-

Y

a, tr 5

-

C 0

E

a, P c ro p:

a, N

2

c Q: v1

C

c

c

;?

E

c c c a !?

-r

: -

I 0

The video s ignal contains the reference line sync , the zero t i m e

reference, and the echo returns. The height of the returned echo is measured in terms of percentage of the l ine sync unit. Then using AGC information the echo height is converted to microvolts at the receiver input.

The sounder receiver AGC is controlled by the leve l of externally

impressed noise. At each imposed frequency marker, AGC returns to zero volts. deviation from the reference leve l and comparing it with the calibrated value.

The calibrated chart reveals the input magnitude through the AGC and the echo height readings.

Thus the magnitude of the AGC can be decoded by measuring the

The t i m e code can be used to check on the satellite map, the ephemeris,

prepared by NASA, so tha t the actual target area c a n be correctly located. Signal t apes for a total of 53 passes of the satellite were obtained

from DRTE in two different groups. i n October, 1963. It contains 31 passes mainly over the Southern hemisphere. The second group of 11 t apes w a s received in June, 1965. p a s s e s are over the Northern hemisphere.

The first group of 31 t apes was received

M o s t of its 2 2

In many locations the ground signal is too weak to be measured accurately. This is due in part to the scattering properties of the ground, i n part t o the ionosphere, and in large part t o the presence of many interfering s ignals at various frequencies within the range of interest , cular ly bad in the Northern hemisphere. are those ref lected from the continent of Australia. hemisphere often contain only a f e w pulses in the appropriate range. ana lys i s of the da t a , 70 points were used for the f inal s tudies .

Interfering s ignals are parti- The bes t ground echoes so far obtained

Ground returns from Northern After

Variations in the performance of the telemetry link do not affect the r e su l t because a “line sync un i t ” i s the standard amplitude measure. This

is determined by a calibration pulse (transmitted once for each transmitter pulse) which accurately determines the transfer gain for the telemetry recording

and playing-back system. Table 2 i l lust rates the anticipated errors i n the various quantit ies for

both ionosphere and ground echoes, obtained for these . It is ? 3.35 dB.

Maximum overall RMS error has been

-7-

Table 2

Summary of Errors

Parameter

Signal reading Line Sync reading AGC reading AGC input curves Reading errors Antenna gain

For lower frequency: Frequency Transmitted power A1 titud e RCVR input ZR Mismatch loss

For higher frequency: Frequency Transmitted power

Altitude RCVR input ZR Mismatch loss

Variation Range

2 4 percent -4 3 percent i 1 percent

4.9-5.5 MC/S 7-9 dB below 10( 600-800 km

8.7-9.3 MC/S 6. 6-8.2 dB belot

1 oow 1004-1 050 km

Selected Reference

1.31

5.2 Mc/s 7 dB below lOOW

700 km 366-j99 ohm

2.86 dB

9.0 Mc/s 7 dB below lOOW

1000 km 311-j66 ohm

1.7 dB

Max. error i n dB

1.50 2 1.50 2 1.50

2 0.25 i- 1.20 2 1.00 -f. 1.00

t 1 .00

2 0.15 'r 1.20

- 0.20 : 1.00

0.50

Maximum possible overall RMS error . . , . . . . . . . . . . . . . 3.35

- 8-

IV. EXPERIMENTAL RESULTS

Table 3 shows the effect ive reflection coefficients calculated for

the se lec ted terrains on both the absolute and relative bas i s .

t ions of the targets were obtained from examination of detailed vegetation m a p s of the a r e a s involved. The s t a t e s of the sea were not known exactly

at the sampled t i m e . Only the wind speed could be obtained from the nearby weather s ta t ions. It was insufficient to correlate the echo strength with the

sea states at the particular ins tan ts ,

The descrip-

- 9-

2 v a, a m I3

1

a m 4 w O L n I I

. . m C D a C D m N

m v m I l l

. . .

4

2 a,

8

-

c rb a, 0 0

Fs, W

a, rb E-,

a

m co v a, a m I3 -

co u3

m I

a, a I F

o v m m

m v 1 1

. .

v o 4 0 )

m c o I I

. .

u 3 w d m o v

I

. .

c m a,

8

I N h

Ln I

v u3

u3 I

__

CD O

0 I

. . . $ 6 1 d o 0 I I l l

c -rl

; a, I3

h a, cn rb a ?? a, c a a, 3 E:

c 0 0

-rl CI

v

-10-

u c m

..-I CJ

4

2 s 3 0 cn

V

c m

.rl u

4

2 5 3 0 cn

c 0

m u 0 I4

..-I CI

cn 0

m u3 u3

__

3 0

h

0 N

cn 0

0

w Ln

m c n a 0 C g O

N m v v ) . .

3 3 0 0

u 3 - l

N O m W

. . 3 3 3 3

0

I-I

h h

3 3 O D

,-IO

a m v v . .

N W 0 0 u3a 0 0

N u3

co N

-

r; 0

o a o v o v

0 W Ln 0

N u3

03 N

-

l n h c o N 0 N - l m d 4 N 0 0 4

m u 3 ( D C D u 3 w a 0 0 0 0 0 0 0

w W

h N

CI

8

4 ';p CD a, a rrl bl

rn v W a, a (d rn

cu W a, a a I3

c D W W o u 3 N O W N

I I I I

. . . d m m m

W O m W m m a o m v v w I I I I

. . . . m m o c o v m c u m w d h m ( D h m W m W

1 1 1 1 1 1

. . . . . .

m N o N a m m m . . . . . . .

m m m q W m h m 1 1 1 1 I I I I

m m m d m m c u m v N o . - l o o d 0 0 0 0 0 0 . . . . . . . I I I I I

m W v v 0 0 m c o c o c u m m . - l . - 4 N O O O

1 1 1 1 I

. . . . . . . . . . . . . 7 9 7 7 ' 1 O d c u d I l l

2

5 a 0 6

3 3 e- 0

h

a, tn rd a 2 E

2 E a a,

0 0 Y

c 0

(d u 0 d

-4 +I

c n c n m c n c n c n c n

m N r t o o 0 0 0 0 0 . 0 . .

m c n c n c n

N r r v m

m c o . - l o m m m m

0 0 0 0 . . . . m c n c n m 0 0 0 0

N @ a @ a d

o d m m m m m q . . . .

c n c n m c n m t n

. - t o o r t o

v o c t o c o v V v m m N C U

0 0 0 0 0 0

9 . . . . . . . . . . . . W h O W W h r n d r l @ a N N N N

w w w w w w w w w w w w w w w m m m c o 0 0 0 0 0 0 0 0

m b m d O O d N m m m m d d d d

. . . .

m W

N N

rrl

-

6' k

m W

h cu

rd

-

2 k

0

a, a m I 3

d -3

2 m E

CV a a, a rd &

d

a, a rd b

a, a m b

a m m a

m v I I

c o d b m o a . . . . . . . . . . .

m a m m m m m m I I I I I I I I

c o w 0 I l l

00 m v m a m N m a m m a

1 I l l

. . . O C V N O O b

m m N I l l

. . . . . . . . . . . m a m m a m m m I 1 I I I I I I

N U 3 m a 4 0

I I

N d m w o o . . . O d N I l l

. . . . . . . . 0 0 0 0 0 0 0 0 I I I I I

c m a,

8 c z

c

t C c

Q .r t

- cc a c

L K E-

c u

z 2 2 2 0 0 0 0

d - m a m o m m m a m a

. . . 0 0 0 0 0 0 0 0

O d d . - I C V C V C V C

C D v m C V d o c O u m m L n m m m v . F . . . . . . .

3 3 3 3 3 3 3 : 0 0 0 0 0 0 0 0

d v o a m o w o ~ o o m m o m r h h h a w a a u

. . . . . . . d d d d d d d F

3 3 5 3 3 5 3 0 0 0

a m - v o m m m b . . . 0 0 0 0

b m m o h m o m 0 m m m

. . .

m m m m m o a w h 0 0 0

m a

m

h a a, 2 c c 0 u m a,

ru E-r

-4 c,

v

2

. . . . O O d O O

I I I.1 w m m o o o m r . c v o o m I I I I

. . . .

1

3 3 3 3 3 1 3 3 3 3

a, a ru E-r

o o d o v O m m h m

v m h l m o I I I I I

. . . . .

0 0 0 0 0

N N m m v b C D r n N C n N N N m N . . . . .

3 3 3 3 3 0 0 0 0 0

m v o o u 3 v v v m a d d d 0 0

. . . . . d d d d d

Attenuation in the ionosphere at 9 Mc/s has been studied. It was found that a n accurate instant correction would be too much involved. A rough correction by the Quiet Day curve method [Mitra and Shain, 1953; Steiger and Warwick, 19 61 ] was attempted for correction factors ranging from 0. 65 to 6.5 dB.

Table 4 compares the results of the power reflection coefficients with comparable data from other observers at other frequencies.

Table 4

Power Reflection Coefficients at Several Different Frequencies

Observer

Sandia' Corp.

d Auburn University

Alouette

Frequency

415 Mc/s

3800 Mc/s

1600 Mc/s

9 Mc/s

Altitude

12,000 ft .

8 , 000 ft.

7 , 000 ft.

12,000 ft.

7 ,000 ft.

2,.000 to 70,000 ft.

1000 km

Forest

- 1 2 to -21 dB

-13 to -16 dB

-7 to -14 dB

Terrain

Desert

-13 dB

-14 dB

-3.5 to -10 dB

Sand

-6 dB

Water

b -1 dB

+la dB

+2b dB

+la dB b -3a to -5

dB

-6 to 0 d5

a Smooth water. bRough water.

Edison et. al. , 1960. dSummer, 19 63. C

-11-

A. Kg/Ki Correction

A s stated above, the effective reflection coefficient, Ki, of the ionosphere was calculated using data at approximately 5 MHz. reflection coefficient of the ground, Kg, was calculated from da ta at 8.5 to 9.5 MHz. These figures are given in columns A and B respectively of Table 3. Column C represents a "corrected " Kg.

The effect ive

The ratio of Kg/Ki i n dB is

K (absolute) 2o log Kiqabsolute) = 20 log Kg(abs) - 20 log Ki(abs) .

If Ki(abs)< 1 then t h i s ra t io i n dB is the effective reflection coefficient of the ground, K that would resu l t i f there had been no ionosphere present. This correction suffers from the fact that Ki is at 5 MHz and K is at 8. 5 to 9.5 MHz. However, t h i s is the bes t tha t can be done for th i s type of correction.

9'

g

B. Attenuation of R. F. i n Ionosphere by Collision Frequency

Attenuation a wave suffers in the ionosphere due t o e lectrons coll iding with heavy par t ic les has been d i scussed by many papers. of the theoretical expressions for the absorption coefficient K given by Ratcliffe [19 has the form

One

2

1 ON nepers/m 2 K = -

where c = the velocity of light i n f ree space

p = the rea l part of the refraction index v = the average frequency of col l is ion between electron

w = the wave s ignal frequency o L = the longitudinal gyro-frequency

and heavy particles

- the ionosphere plasma frequency N- -1 2-

This equation is derived by introducing into the bas i c equation of

motion of a s ingle electron a "viscous damping force I' - m v t with m being

the mass and f the velocity of the electron. It is applicable to quasi- longitudinal propagation. In th i s equation, we can see that only c and w

are independent of position. Other parameters vary from point t o point along a vertical path of interest at a particular t ime; t hus , i f the instantaneous attenuation is to be found, the correct values for t hese parameters must be found first.

Besides finding the magnitude of t he col l is ion frequency v and its or u H (see Appendix), we need furthermore to know the relation with w

condition of the bottom s ide of t h e ionosphere in addition t o the simultan- eously measured ionospheric conditions. Detailed s tudies of the instant- aneous absorption require de ta i l s which are difficult t o obtain for the locat ions of the Alouette.

C. The Quiet Day Curve and Its Use

Since the instantaneous absorption could not be found, we turned to another method, using the quiet day curve to try to find a reasonable

average approximate correction factor.

Warwick, 19611 is basical ly as follows: Every kind of radio wave which travels through the ionosphere is

at tenuated, as is cosmic radio noise at the frequency of interest . It was

assumed that the minimum ionosphere attenuation occurs between 01 00 and 0600 hours local t ime. These cosmic-noise levels were plotted aga ins t the corresponding s iderea l t i m e for a n entire solar year, or a complete s idereal day. The envelope, referred as the quiet day curve, is drawn through the maximum of the sca t te r distribution. It is assumed to represent incoming

s igna l s tha t have not suffered any attenuation.

The idea of the quiet day curve [Mitra and Shain, 1953; Steiger and

Fortunately for our purpose, a 9 m c antenna was constructed by the High Altitude Observatory , University of Colorado, near Boulder, Colorado.

A set of recorded cosmic noise for nine months of 1964 was reduced at CRES and t h e quiet day curve determined (as shown i n figure 2).

-1 3-

~~ - . . . . .

. . . . . . . . . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . ................. . . .

. . . . . . . . . . . . . . . . . . . . . . ..................

...............

. . . . . . . . . . . .

. . . . . .

. . ...

..

.......... ............

.........

. . . . . .

. . . . .

. .

..

.......... . . . . . . . . . . . . . . . . . . . . . .

. *:-I:............... .\. ..............

............ . . : . ...........

. . a

. . . . . .

..

9 .

. . ... ... .... . . .

.... . . . . .

. . . . . . . . . . . . . . . . . .

. . . . . . . . .. . . . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . . .

. . ....... . . . . ......

.. . . . . . . . . . . . .

. . . . . .

CJ

0

In order to apply the idea of the quiet day curve t o the correction

of the Alouette sa te l l i t e da ta , several assumptions were made, namely:

1) the ionosphere is symmetrical about the ear th ' s geomagnetic axis, so tha t the first order correction can thus be made i n both hemispheres

from observations i n the northern hemisphere;

2) the quiet day curve is invariant with t i m e , so tha t the curve reduced from the 19 64 period can be used for any other t ime;

3) the same correction factor may be applied to 2 10" on each s ide of the geomagnetic lati tude over Boulder, Colorado (135" W 40" N geographic location) and over the same region i n the southern hemisphere;

4) the correction factor is taken as the ratio of the quie t day curve reading t o the attenuated record at the s ideral t i m e of interest .

The correction factors thus obtained under these assumptions range from 0. 65 dB t o 6.5 dB and were applied to se lec ted points in column B of Table 3.

The reduced points which do not fall within the regions s ta ted i n condition (3) have not been corrected. The corrected effective reflection

coeff ic ients are l is ted in column B of Table 5.

Do Average Diurnal Attenuation

Using the 9 MHz cosmic.radio noise da ta mentioned i n sect ion B

above , curves for the average diurnal attenuation of the ionosphere vs. Mountain Standard Time were plotted. 3 and 4.

t i m e for the satellite p a s s w a s u s e d t o obtain the correction.

These curves are shown in figures Figure 3 is for the summer and figure 4 for the winter. Local

These

corrections varied between . 5 dB and 3.4 dB. column C of Table 5.

The resu l t s are shown in

-1 5-

I I I

\

0

,

0 m

al N

W N

N N

N

,

m 0 u

N "

., N 0- 0

a 0 -

"

.,

01 e 01

10 *

n - N

m - " 0 N

0

E. Double Ground Bounce Experiment

In a n attempt to obtain a "bench mark" for comparison with the Alouette da ta a n experiment h a s been designed to obtain variations i n

the ground reflection coeff ic ient at a fixed location. This experiment wi l l give useful information regarding the effect of ground moisture, snow, vegetation, etc. , on the ground reflection coefficient.

Half-wave horizontal crossed-dipole antennas have been constructed. These antennas are placed a quarter wavelength about the ground. A 5.3 MHz carrier modulated by a 1 0 0 p s e c pulse at a rate of 100 pulses per second is directed vertically to the ionosphere. t h e ionosphere. ground, and the' ionosphere. Therefore th i s is equivalent to a satellite transmitter at the image in the ionosphere.

The first echo returns directly from The second echo returns by way of the ionosphere, the

If PR is the power received on the first return then 1

WTG kI - -

pR1 4 n (2h)2

where W G is the power radiated vertically by the antenna system, kI is the ionosphere reflection coefficient and h is the height of the ionosphere. The power received on the second echo is

T

2 h 2 2 2 = P (z) k kI.> pR2 R1 g

where k is the ground reflection coefficient. Thus, In order to obtain k radiated vertically from the antenna system, WTG, must be determined also. kept constant for all data by adjusting the transmitter output.

g we must first determine kI from (8).

This wil l be done by calculation and measurement. WTG wil l be

The power g '

To da te the experiment is not ye t operational.

-1 9-

V. DISCUSSION OF RESULTS

A. The velocity of the satellite, as specified on the da t e of September 29, 1962, at perigee was 16,467 m i l e s per hour, and that at apogee was 16 ,387 miles per hour; it w a s at a height of 618.7 s ta tu te m i l e s (995.7km) at perigee and 641.1 s ta tute m i l e s (1031.75 km) at apogee. During 7 mill iseconds, roughly the t i m e required for a radar pulse to travel twice the d is tance between the satellite and the earth (about 2,000 km) the satellite advanced only about 50 meters, or it turned a n angle 4 x degree around the center of the earth. With such a small movement, the satellite c a n be considered as stationary with respect to e a c h individual radar pulse. is considered a constant , even with the much slower satellite spin effect superimposed.

Therefore, for each pulse the antenna pattern

B. With the 1OOp sec pulse and the 1000 km orbit , the first pulse length over t he ground il luminates a circle with radius 173 km, corresponding to an illuminated cone with a n apex angle of 2 0 " . Because of th i s large angle and the relatively weak s igna ls , the shape of the received pulse g ives little information about the variation of scat ter ing as a function of angle.

C. The dipole antenna on the satellite h a s a broad pattern and the satellite axis is inclined with respec t to the radial direction by a maximum of around 20" . Consequently, no significant correction i n the returns is required because of antenna tilt. I t is assumed tha t the antenna pattern is the same as it would be i f the antenna were in space . pattern is based on the calculated values supplied by the designer , the Sinclair Radio Laboratories.

The antenna gain

D. were possible. A series of cal ibrat ions, however, w a s carried out previous

to launch. Errors may be introduced because t h e receiver sensi t ivi ty varies greatly for such s m a l l changes i n the AGC l eve l , and its measurement is not as precise a s would be desirable .

N o in-flight calibration of the satellite receiver system gain

- 20-

E. The quiet day curve correction factors obtained range from . 65 dB t o 6.5 dB. It is evident t ha t some of the large correction factors a re not quite reliable because they result i n K > 1 (line 2 , 3 ; Column B, Table 5). for using the quiet day curve are perhaps not fully justif ied.

t ions a re l is ted in paragraph B, Section IV.

g This may be contributed to the fact tha t some of the assumptions

These assump-

G. The average diurnal attenuation correction factors ranged from . 5 to 3.5 dB. They were obtained from figure 3 (summer: May through October inclusive) and figure 4 (winter: November through April inclusive). These corrections give quite consistent results. obviously wrong corrections as in Column B, Table 5 , l ines 2 and 3 which indicate Kg> 1.

indicate the correction was made using Ki> 1.

inclusive) may be due to the s e a state. Lines 6 through 1 0 inclusive are

returns from the Australian Continent and are certainly consis tent in that it is expected that K for land i s l e s s than Kg for salt water.

g for (line 1 2 ) 'the forested area i n Australia. A t t h i s frequency a forest can be expected to introduce considerable scatter.

That i s , there are no

Similarly, i n Column A, Table 5 l ines 2 , 5 , 1 0 , 15 , 16 , 1 9 , 21 and 22

The variation in K (Column C , Table 5) for the ocean (lines 1 through 5 g

Similarly

Lines 13 through 25 inclusive are points obtained over the North Pacific

and again are consis tent with theory. Table 5 a l s o lists the averages for the various a reas and the ranges

of the resu l t s using the 2 3.35 dB overall RMS error expected.

VI. CONCLUSIONS

The observation of effective reflection coefficients from the 1000 km al t i tude of the Alouette satel l i te have been made for 70 sample points over the ear th surface, at frequencies i n the order of 9 Mc/s. The absolute

ca lcu la ted ground effect ive reflection coefficient ranges from -1 2 dB to -2.8 d B for land and 09.2 dB t o -2.68 dB for the ocean. Estimated

-21 -

variability of the da ta is 2 3.35 dB. from the quiet day curve, t he ionosphere absorption ranges from 0. 65 dB to 6.5 dB at 9 Mc/s and the corrected values range from -10.5 dB for t h e forest to -0.02 dB for the ocean. The average diurnal attenuation correction

g ives ranges of -5.35 dB to -0.06 dB for the ocean and -10. 67 dB to -4.6 dB

for land.

By applying the first order correction

VIII. APPENDIX

The relation between o N2 = Ne2/e om and W H = p .HOe/m

where N = the electron densi ty of the ionosphere

the location i n the space , Ho= the total ear th magnetic field without concerning with

c a n be estimated by assuming

1. ordinary and the extraordinary wave from the vertical plane of incidence may be ignored,

2. as the point of reflection getting deeper into the space .

For t h e vertical incident case the horizontal deviation of the

The strength of the ear th ' s magnetic field dec reases monotonically

3. is negligibly smal l .

The col l is ion i n the ionosphere along the wave traveling path

4. wave at t h e electron dens i ty maximum, as we l l a s at the Alouette satel l i te level. t ion, the s ignal frequencies for the ordinary and the extraordinary wave are related through the equations

It is the same layer that ref lects the ordinary and the extraordinary

Therefore, at these two particular points of reflec-

-22-

.

5. The relation between the ordinary and the extraordinary wave frequencies is assumed to be f i t by the polynomial

i = u

After working over several examples it is found that the simplest f i t

w = a o + a o 0 1 E

is the best approximation. frequency w E of the extraordinary wave corresponding to the same reflecting layer may thus be determined, and the gyro-frequency, u H accordingly the

geomagnetic field strength, H , a t this layer can be calculated from the relation

For any given w 0 the ordinary wave, the

No te tha t these resu l t s wil l have nothing to do with altitude.

electron density profile must be solved by employing methods suggested, for example, by Titheridge [1961] , Unz (1 9641 , etc.

In case the absorption in the ionosphere cannot be ignored completely, the wave reflection index IJ. will no longer be reduced to zero when reflection occurs . As long as the coll ision frequency v is s m a l l comparing with the

s igna l frequency W , the relations engaged in the discussion a re held true approximately [Ratcliffe, 19591 . still be valid within acceptable error l i m i t s .

The

Relations among w0, w E , and wH should

- 23-

Alouette Calibration Curves

1.6

1.4

1 . 2

* 1 .0 CI -4

c. =I .8

cn . 6

u C x Q) d l-l .‘-I

. 4

.2 AGC Set With Noise

4 1 1

-70 -60 -50 -40 -30 -20 -10 0 +10 +20 db

O d b = 3 MV

S-27-6 RX 3 MHz

1 . 6 1 1 .4

1 . 2

u x cn C .8

Q) c: .6

.4

. 2

AGC = 0

A G C = 1 . 2 5 ~ AGC = 2 . 5 ~ AGC = 3 . 7 5 ~

AGC = 4 . 5 ~

AGC AGC AGC AGC

AGC Se t with CW Signal

= 1 . 2 5 ~ = 2 . 5 0 ~ = 3 .75v = 4 . 5 v

-70 -60 -50 -40 -30 -20 -10 0 +10 +20 db

Odb= MV

Figure 5.

.

1.6-

1.4.

1 .2 -

1 .0 -

. 8 -

. 6 -

. 4 -

. 2 - AGC Set With Noise

1.6-

1.4.

1 .2 -

1 .0 -

. 8 -

. 6 -

. 4 -

. 2 - AGC Set With Noise

1 1 1 1 1 1 1

S-27-6 RX 9 MHz

-70 -60 - 5 b -40 -30 -20 -10 0 +10 +20 -70

Figure 6.

I 1 1 1 1 1 1 1

-60 - 5 b -40 -30 -20 -10 0 +10 +20

1.4-

1 . 2 - rn + .rl

(3 's 1.0-

u c tA a c

h - 8 -

.6-

. 4 -

. 2 -

r 1 1 1 1 1

8

1.6 -

1.4 -

1.2 -

rn 1.0 - +J -4

c D u c x v3 a, c cl

. .a -

- 6 - -4

.4 -

.2 ~ AGC S e t With Noise

AGC

AGC

AGC

AGC

1 I I I

-70 -60 -50 -40 -30 -20 -10 0 +10 +20 db

Odb= 3 MV

S-27-6 RX 6MHz

1.6

1.4 1

1,25v. 2.5 v 3 75v

4.5v

1 . 2 5 ~

2.5v 3.75v 4.5v

db Odb = 3 MV

Figure 7.

.

-2 1.0- s z ' .8- h rn 2 .6-

.4-

r(

I4

.2-

1 .6

1.4

1.2

rn -2 1.0 s u .8 c x rn 2 .6 -+-I I 4

.4

. 2

1 I 1 1 1 1 1 1

AGC = 0 A G C = 1 . 2 5 ~ AGC = 2.5 v AGC = 3 . 7 5 ~

AGC = 4 . 5 ~

AGC Set with Noise

1 1 1 1 1 1

0 +10 +20 -70 -60 -50 -40 -30 -20 -10 db

O d b = 3 MV.

S-27-6 RX 12 MHz

1.6

1 .4

1 . 2

= o = 1 . 2 5 ~ = 2.5v = 2.75v = 4.5 v

AGC Set with CW. Signal

Figure 8.

m a m Q a a

+ + + 0 0 0

rn

m Q 0 N +

m a 0

+ d

?i rn

5 - U L :

0 0 0 a, 0 I

I

0 0 C D b d

00 0 0 0 0 0 -4 N m w m 0

I I I I I I I I

ap asuodsetl:

3

.

D

References

Ch ia , R. C . , A . K. Fung and R . K. Moore, "High Frequency Backscatter from the Earth Measured at 1000 Km Altitude, 'I Radio Sc ience , Journal of Research NBS/USNC-URSI, vol. 69D, .no. 4 , April 1965, pp. 641-649.

Edison, A. R . , R. K . Moore and B. D. Warner, "Radar Terrain Return Measured at Near-Vertical Incidence , I' IRE Trans. Ant. Prop. , vol. 8 , no. 3 , May 1960, pp. 246-254.

Hayre, H. S . , "Radar Backscatter Theories for Near-Vertical Incidence and Their Application to a n E s t i m a t e of the Lunar Surface Roughness, University of New Mexico, Doctoral disser ta t ion, 1962.

Mitra, A. P . and C. A. Shain, "The Measurement of Ionospheric Absorption Using Observations of 18 .3 Mc/s Cosmic Radio No i se , " J . A t m o s . Terr. Phys. , V O ~ . 4, 1953, pp. 204-218.

Moore, R. K . , "Resolution of Vertical Incidence Radar Return into Specular Components, 'I University of New Mexico, Engr. Expt. Sta . Tech. Rept. EE-6, July 1957.

Ratcliffe, J. A. , Magneto-Ionic Theory, Cambridge University P res s , 1959.

Taylor, R. C . , "Terrain Return Measurement at X , K u , and Ka Band, IRE N a t . Conv. Record, vol. 7 , part 1 , 1959, pp. 19-26.

Titheridge, J . E . , "A New Method for the Analysis of h ' ( f ) Record, I'

J. A t m o s . Terr. Phys. , vol . 2 1 , no. 1 , April 1961.

TJnz, H . , "The Reduction of Ionograms to Electron Densi ty Profile, I'

J. A t m o s . Terr. Phys. , vol . 26, no. 1 , Jan. 1964.

CRES LABORATORIES

Chemical Engineering Low Temperature Laboratory

Remote Sensing Laboratory

Electronics Research Laboratory

Chemical Engineering Heat Transfer Laboratory

Nuclear Engineering Laboratory

Environmental Health Engineering Laboratory

Digital Computer Technology Laboratory

Water Resources Institute