zotero/storage/BPXVD5HI/.zotero-ft-cache

5 February 1960, Volume 131, Number 3397

SCI}:NCE

transmission and reception consisted of

four rhombic antennas in a broadside

array, covering a rectangular area of

800 by 725-feet. The antenna gain is
Radar Echoes frotn the Sun t estimated to be 25 decibels relative to that of an onlnidirectional antenna.

The principal antenna beam is directed

Man's first direct contact with the sun opens new r approximately east at an elevation of

approaches for the study of solar events.

10 degrees. The sun is in the antenna

* beam only for about 30 minutes soon

after sunrise on a few days near each

V. R. Eshleman, R. C. Barthle, P. B. Gallagher r equinoctial period. T4e travel time of a radar pulse tcy

the sun (approximately 93 million miles

away) and back to the earth is about

1000 seconds. At the end of the trans-

On a number of mornings in September 195S, April 1959, and September 1959, attempts were made at Stanford University to obtain radar echoes from the sun. The data obtained on three days in April have been intensively analyzed with the aid of a digital electronic - computer. It appears that solar echoes were obtained on each of these days. This experiment was possible only because of the availability of facilities built up at Stanford for several different research programs (l).
In 1952 Kerr (2) discussed the scientific information that might be gained from radar studies of the sun and planets, if sufficiently sensitive radar systems could be built. Others (3) have also discussed the importance of radar studies of the solar system and the magnitude of the required installations. But continuing advances in large antennas, high-powered transmitters, low-noise receivers, and data-processing techniques will soon make it possible to conduct important radar investigations out to distances which include essentially all of the solar system. The dramatic beginning of radar probing beyond the moon was announced last year when scientists at the Lincoln Laboratory of the Massachusetts Institute of Technology described the first radar detection of Venus (4).
The equipment and techniques for radar detection of the sun differ in several ways from those required for detection of the nearerplanets. As-Kerrhas pointed out (2) a relatively low radar

frequency is needed to avo tid extensive mitting period of 900 seconds, the an-

absorption in the solar co srona above tenna was connected directly to the

the reflecting points. He es ,timates that receiving system, and the transmitter

the optimum frequency i is near 30 and pulsing circuits were turned- off.

megacycles per second. M [uch higher The receiver and its preamplifier are of

frequencies can be used to detect conventional design. An intermediate

planetsS so higher antenna gain can be frequency bandwidth of 2000 cycles per

obtained for a given antenr la size. The second was used, and this band was

new low-noise receiving dez vices, which translated with the receiver beat-fre-

are so important for radar s ensitivity at quency oscillator so that its lower end

the higher frequencies, are of no value was at zero frequency. The receiver was

at the lower frequencies w] here cosmic tuned to the transmitted frequency since

and solar noise limit the 4 detectability the computed Doppler shift was less

of weak signals. In additiol n, the char- than 25 cycles per second. The output

acteristics of solar noise difT Ser from the was recorded on magnetic tape for later

better understood features of random detection and analysis.

.

.

.

recelver and cosmlc nolsl ,e, so care Trials were scheduled for each

must be exercised in 4 determining morning from 5 to 13 April, inclusive

whether a solar echo has bes en obtained. Because of various ditEculties (for ex-

These special features were expected to ample, equipment failures, timing am-

make radar detection of tk e sun very biguity, and radio interference), re-

difficult, even though, becs ruse of the cordings suitable for intensive analysis

sun's size and despite its distance, a were obtained only on 7, 10, and 12

solar echo would be 100 times more April.

intense than an echo from Venus. The test procedure for September

1958 differed in several respects -fronz

that described above. The recorded

Equipment and Test Procel dure data have not yet been analyzed in detail. In Septemlber 1959, changes were

lFor the April 1959 sun-e zcho tests a made in the coding of the transmitted

transmitter (Collins PRT- 22) having waves and in the antenna. The antenna

about 40 kilowatts' aver; age output modification was made for the need of

power was operated at; about 25.6 the program for which the antenna was-

megacycles per second. The transmitter first constructed and was designed for

was pulsed on and off alternately short-pulse work only. A risk was taken

throughout a time interval of 15 min- in operating the modified antenna with

utes, each on and each ofT pe oriod lastin(l

30 seconds. The antenna system use

The authors are afliliated with the Space
Radioscience Laboratory of Stanford University
d for both Stanford, California.

S FEB1tUARY 1960

329

the long pulse needed for the solar experiment. Data analysis was proceeding when it was discovered that certain antenna components had failed during the test}ng so that antenna performance was seriously impaired. Therefore interest was diverted to the April 1959 test results, and only these are discussed further.
Data Analysis and Results
Data analysis was conducted with an I BM 797 computer and associated equipment. The recordings made on 7, l0 and 12 April at 0 to 2000 cycles per second were sampled electronically

4000 times each secondn and thus the data were converted from analog to digital form. The absolute values of the samples were summed for periods of one second, and these one-second sums were stored for further analysis. (By taking absolute values, the receiver output was detected in an ideal linear detector. ) The lolagest usable echo time common to the three days is 12 minutes, or 720 one-second sums.
A square wave of period 1 minute, representing the transmitted wave, was cross-correlated with the 720 one-second sums for each day. That is, the sums 1 through 30 were added, 31 through 60 were subtracted, 61 through 90 were added . . ., and 691 through

720 were subtracted, to obtain one point of the cross-correlation curve. The next point was obtained by adding the sums 2 through 31, subtracting those from 32 through 61, and so on, the first one-second sum being included with the subtraction of the sums 692 through 720. After 60 crosscorrelation points were obtained in this way, the values would start to repeat, so only 60 points (seconds) were computed The cross-correlation curves that result are shown in Fig. 1.
Curves a, b, and c in Fig. l show the cross-correlation curves for 7 10 and 12 AprilS respectively,- and curve d is a composite curve for the three days.

200

3 2 1 o SOLAR RADI I

150

loo -

so-

I \

I W

-

, 1, 1 Tt , '

S

f

l

v

otS

s

}

- | ^ ^ -

CL

20 . 40\ 6( O TIME

- -
: -5O-

-loo -

-150
-200-

(a) 7 APRIL 1959

_\\

200-
150 100-

3 2 1 0 SO ILAR RADI I

c;n _ llJ Qv J c:}
J L
ca: -50
-100

40 C l ,,} ,t, l , l , v

\
t

I -

I -

O TIME

/20

_S,/

-150

-200

(b) 10 APRIL 1959

3 2 1 o SOLAR RADII
, | , I

RADI I

llJ cM

-

-

t

L

TIME

J

Tl ME

(c) 12 APRIL 1959 (d) 7, 10712 APRIL 1959
Fig. 1. Processed data (cross-correlation of 12 minutes of transmitted code with 12 minutes of received signal), showing evidence of radar echoes from the sun. (a) Results for 7 April; (b) results for 10 April; (c) results for 12 April; (d) combination of al1 data compared with an ideal echo curve. The ordinate is relative amplitude in arbitrary units, aTld the abscissa is relatlve time in
secollds.
SCIENCE, VOL. 131

:Sa j V ' WTlMEo V V \v

.mA

burst

burst

1minule 1 9 s *2
Fig. 2. Processed data (cross-correlation of 1 minute of transmitted code with 12 minutes of received signal), showing individua) echo returns and disturbing noise bursts compared with ideal echo curve for 7 April 19S9. Curve a in Fig. l is derived from these data. The ordinate is relative amplitude in arbitrary units and t:he abscissa is relative time in seconds.
wcccyrreecorabirsfzoscbtatltcsetetdbsasss-eFo(ticdccaf[nwdhootvleefnscouhaatofheraolGeaem netrtiteeacposihheapoueleufoerrnalcrasdrrgfshtnierirodheerlievtloeW omrcifnahwwtoaansteIStntoohclate.ogcgrrrpvvdduroicler1iteeaelniuer.esnthenvecehetegwirceropra1nuneeemoylruhlrtnewenFs-aseciroscoonuqcrtiuerta.iwesaaesm tlducerdusada)eesetggovrshbbiaithgereobyurutia,deeo,wstirttculssgtotiihhnithonnufvsehndoPsgetereehstunasaahpihhoetelaetc.niaareedenlhhieendsroawlashcom orsm etocm hnetcsfils-1eantttissah)nwtoaeyafetcieratsre.i-perwppegtstnevvooddnistobrtrhdecnirlpapeIhorruecfeeeseeiistrlyos£teoeormearenn,po.oeiorselrf.ium rirkiiepaekssem an.tcm eretsgsnocTutbIpreckreecptcavathexeenoohrIbleonutqlehlrrsurdfrsm peztialeodeFioaiefilstduetcosysearoestervnrfafsranforsum ita,iotnbrgeaisvergtacseaiftenatuucdihotadunheloen-iawtgriacraehnt.adhernyngm osnedaclwnsbttaotsl,hattbendeleaewnnsloeshpatwplddoetadderaurheicoaatre.rpieeorarnasshnsbvuehridfr.ciaregs.trrsirTaefupndingwdceuoteArsieddeaunhwuIngeoTeocenihoeFrt1atm tlerrvem ipsm tfeegdatanaraliha6aita-tievcsieftshrrtngkhplredgtsnnuhs.m ueedAm ueeepil5oes.wayhreoieeeswwddocturcpnisnum sudonr.iostaeaes1naotvecdceorsetosasAesbaetrx,hddrcdptsm itSueneureessaiaoumajropseauphioucuoietpoarcm deeraltm bellfwaenutfetm rdssndicaiclvandrspnvptploabhaoclestrereAttptfodeeuielayoebrneeiriettshevooeelu.colrrm iaooliadtlihediesm stfeeylnraglcsnnpbldnkissoedisililuedeog,o-ssaa,lotlt(clnIwssrrItccppbasticsvcnr2tanssiisintoim difoiTuifsyuhm tn-s-hiotnnt)rcoeoteotenhhnorhoeeeosva2tr.oirfm oe4eqh3cw4rhrnlm phm niege-etl5osaaogetm sevu)eteeiacesh3tu3y)ia1edisak)thues-nuieetobetcleeathnn.Ssraptre4wte-aIsdt.lyhclLe8inahSp,ptiesm em e7rdyiiizgd4oiowoftldctiiooejdniposotoueo-eanrcngnusnoim rditfosapiipooiotntenesdngbusfbibrdtnhiseroau7roerbrrbourootawksoFitterauusreycm idteda.m ccenltore-wtyfleoinericlevsAuottrtelgdiaannTaivraietiiit,aflasbKhteerwtsrieriei.crtnbom p-agm eaohrhsnfihcfesothslnrewyueevt1etgbcmelnpiriiisroehgierylraen.osngoirtlleefsosnsyshcirsworahcidsryeosrtop3-etlhsuaeviudoass.tiaom teanitnfcasnrgy0rnns.nlnraietiiivsrlv(uw idhetTm fueioooa2nm cdnthtepsenhm ntW iouoirisop(ueai)haitrefdsvruoasohtoeA,losauineooinutnm cheegpeetlovageocnentoefua.epchcfw ethsrtnna)tlayvethhrrhueplerahotwelhesigocAcorsrttpctnoesooasebpdn.aelpndniieelaehtatyocheqnbxebefnhtdoierieodsriItgriartdbhm ounssrireapoaibestodm iabsioohtcsaam em euesa1fctee1pscanuorcdvneiweeexptahbtntvS,f.oelo/etvce.reetnaufelhvittoopaiatolcrnOorelsAeaio.aepaloIm oyceviestshstihyeucduees-u1nenearssxdergirTossehoncy2castuot5lrpintedfoatosealfapsleiadiodrh2tsteafeacsc5,aeigyotatbhliesm fdt-nrg.htioereltnoaie,ooiprinwcehhseottdecufcm sidum rencbotsmoop,fdyiotroeahaonoeaecm rsdeeitneblsnvullhm soiranrsssnicrcerei-eeiepdru-erasrtteytoelpfnevoitf.aec.t,aepnyadmdorhbelrtlriedwneiesootardsaosem fhoistoseeetanO.ete-enrrsKrifpfeabfehlrkaesle.s-h-o,-s-ysaeorrecrodroisanebitceanapfaem csctn-w tfftotFoeAawtnswatfstpedonntf-parhnttaonhreeatchoh4chbA enefeiuhroidoroedouithiipoapglouahndagihIagtrahaadrtSvictes4loariiidresnlhitserrsiieuorstsipdnehvreesnzie.ate3rttssrno-esoo.seeu,eievlcipnrstafiecem i-be8tapuesSvapxa2tlyeesdrsnfphtec1trhe.linraa,anrhsm eerietisuloaidioesnshrleoeemdogo0wpnSrpcadatw ieoabpaenirrpnedscantsisabagkegaehutsfdtntntttafsysdrsueheticonora,nrisehanitrschiea-aeaossnedirttrarpitcohosenbtw eontociesuthm scficalm naeoohem verswnprltiaeoaim ihpgltnurgtotm fayinudavntobbeelts.eegerAdl.btbnidaoirneothSirdtf]etatngatsFirtrepiotplcm happeTirvuoso1tobreiehgla;sryeofsntontplfraeoer1dnu.nuhesrbfou2ulm ofpornehm soairnddlrssiasstcer2eetm 6aalyirrluFctf,eei1etoethaotdm arstsopcett0tawnfufhsvm thnpil79iyrbrwe.eeedict-aostlrbihererhiA gttteo.aatea5eisdaforkntuhsnsio.vehcviesrtreeentcxhoidiam uT9eatloadtinm sep,Teoceetiteenafaaspisoossnhephnerceronss?itihnnw.ctlppwparrfoat)tluagronuehoteageedttielafbsncbeilaphaaerlisbim sedbiTfhehlthrwdnatym sssiluoaonpisleeivnoohccd2iiheeasocpgereebecm ternipooarstpaebuestootetptllhoennnndyfaossehiinutseeonlitaeausnecseedrilisi1ro1ancttrahss-dloertasiiutricvlhlgyeeia.od,0l.1otso.-inim itet7deoeism rhtarrcoostuthpu3dnsnersionbIcf
331 S FEBRUARY 1960

minute period, have Poisson distributions, as would be expected. If a Poisson distribution of the noise peaks is assumed, the positions of the measured peaks on 7, 10, and 12 April, ^ expressed in solar radii between 0 and 13, are bunched to such a degree that the probabilities are on the order of 10-4 that they could be caused by solar noise alone. Since the position of the bunched peaks for each day corresponds to the expected range of the reflecting regions, the probabilities that solar echoes were obtained on each day are very nearly unity, being about 1 to 10 5. This result is for each day considered independently. Since similar results were obtained on all three days, the total probability of success is even much nearer unity.
Prom a preliminary analysis of the spectrum received it appears that the echo energy is spread over at least 2000 cycles per second. Solar rotation alone could account for much of this Doppler broadening, but gross motions in the solar corona would also be expected to produce a wide echo spectrum.

Collelusions
There is a growing interest in the potentialities of probing the solar system with man-made radio waves. An obvious name for this Seld of investigation is "radar astronomy.' With the added versatility inherent in the control oJi the transmitted waves, it is expected that much will be learned which will complement and extend knowledge gained from passive visual and radio astronomy, and from rocket probe measurements.
The scientific information about the sun gained from the first radar experiments, described above, is very limited. However, it is now possible to plan with confidence the systems and test procedures needed for more meaningful radar studies of our dynamic sun. Prom the time variability of the echo strength, delay, polarization, and spectrum, much will surely be learned about the constantly changing solar phenomena which affect so vitally the earth and its surroundings. A4ore sensitive installations that will be suitable for solar and other studies in radar as-

tronomw7 are now under construction at
several locations, including Stanford
University.
References and Notes
1. The work reported here was supported principally by the Electronics Research Directorate of the Air Force Cambridge Research Center, under contract AF-19(604)2193. We wish to acknc)wledge in particular the assistance of Philip Newman and the late Joseph P. Casey, Jr., of that center. The solar experiments would not have been possible without the existence at Stanford of facilities built up for several other research programs. The antenna system was constructed for ionospheric research under the direction of O. G. Villard, Jr., with support from the Office of Naval Research under contract Nonr-225(33). T. V. Huang, W. A. Long, and others provided valuable assistance in the transmitting-receiving phase of the tests. The data processing facility was organized through the efforts of A. M. Peterson, with the assistance of R. D. Egan and D. S. Pratt. The IBM 797 unit was a gift from the International Business Machines Corp. to Stanford University for use in electrical engineering and mathematical research programs.
2. F. J. Kerr, "On the possibility of obtaining radar echoes from the sun and planets," Proc. I.R.E. (Inst. Radio Engrs.) 40, 660 (1952).
3. F. G-. Bass and S. I. Braude, "On the question of reilecting radar signals from the sun," Ukraizz. J. Phys. 2, 149 ( l 957 ); J. L. Pawsey and R. N. Bracewell, Rad io Astronomy (Oxford Univ. Press, New York, 1955), chaps. 9, 10; J. Pfeiffer, Tke Ckanging lJnierse (Random House, New York, 1956), chap. 13; R. Hanbury Brown and A. C. B. Lovell, Exploratfon of Space by Radro (Wiley, New York, 1958 ), chaps. 8-10.
4. R. Price et al., "Radar echoes from Venus," Science 129, 751 (1959).

CostnicRayProduced Silicon-32 in Nature
Silicon-32, discovered in marine sponges, shows promise as a means for dating oceanographic phenomena.
Devel:ldra Lal, Edward D. GoldbergS Minoru Koide

The nuclear transmutations resulting from the interaction of cosmic rays with nuclear species in the atmosplhere have produced a variety of radioactive prod-
tS detectable on the surface of the earth. Such isotopestas Cl4, H3, Bet°, and P32 have been found, and their individual distributions and concentrations in the various geological domains have led to many significant concepts and contributions in geochemistry, geophysics, and geochronology (see, for example, 1).
This article (2) concerns still an,other
332

isotope produced by cosmic rays-Si32, which we have detected in the marine environment. It is thought th!at this isotope is produced from the nuclear spallations of argon by cosmic rays. It has a half-life of roughly 710 years (3). Any Si32 that reaches the earth from the atmosphere will be rapidly diluted with stable silicon, and the resulting specific activity of Si82 will be very small. However, Si82 decays by negatron (a-) emission to p32 which is a negatron emitter with a half-life of 14.3 days. This makes

it possible to detect SiS2 by milking and by counting the p32 daughter from large amounts of silicon.
The principal exchange reservoir for Si32 is probably the marine hydrosphere which most likely receives SiS° via oceanic rains. The small amounts of silicon in surface marine waters should yield a relatively high specific activity of Si32, whereas the fallout on land will be so diluted by exchange and other chemical interactions with the exposed crustal materials that the detection of this nuclide will be extremely difficult. We estimate the average concentration of Si39 to be 2.6 X 10-° disintegrations per minute, per liter of sea water, or 8 disintegrations per minute, per kilogram of silicon, for a hypothetical thoroughly mixed ocean. The amount of sea water required to yield 1 disintegration per minute, an activity conveniently detectable, is 3.8 X 104 liters. Since the handling of such an amount of water for the extraction of silicon presents mansJZ difficulties, SiS2 was sought initially in siliceous (opaline) sponges, which derive
The authors are affiliated with the Scripps Institution of Oceanographyt University of CaliforniaS La Jolla. Dr. Lal is on leave from the Tata lDnstitute of Fundamental Researchs Bombay, India.
SCIENCE, VOL. 131