7/30/2019 Aurorae Australis in Radio

http://slidepdf.com/reader/full/aurorae-australis-in-radio 1/2

NO. 4627 July 5, 1958 NATURE 43

B. P. SANDFORD

OCT08ER

the visual observations tend to be concentrated. Sofar the evidencefor the simultaneity of the occurrenceshas been largely of a statistical naturel• The RoyalSociety's Expedition to Halley Bay has provided anopportunity for a simultaneous study of the auroraaustralis and borealis by the radio-echo technique,and a preliminary account of the results is given inthis communication.The continuous radio-echo observations of the

aurora australis have been made since May 1957 atHalley Bay, Antarctica (long.75 ,5° S., lat. 26 ·6°W.),using a wave-length of 4 metres. The apparatus hasa rotating aerial system and in principle is identicalwith one which has been in use at J odrell Bank(long. 53·2° N., lat. 2·3° W.). Full details of theequipment are given in the 1.G.Y. Annals'. Preliminary results on the occurrence of auroral echoes forthe period May-October 1957 have been comparedwith the J odrell Bank results for the same period.Echoes from the aurora australis have been

detected at Halley Bay on 134 out of 164 days ofobservation during the period, whereas on only

13days wereechoesfrom the aurora borealis obtainedat Jodrell Bank. Considered on an hourly basis,observations at Halley Bay were made for 93 percent of the time, and echoes were present for 26 percent of the observing time, compared with 2 per centof the time at Jodrell Bank. A study of the echogeometry at both locations shows that the echoesare obtained from the region where the line of sightis normal to the local magnetic field at a height ofabout 110· km. rather than from along a line ofgeomagnetic latitude as originally proposed byBullough and KaiserS. From Halley Bay the locusof specular reflecting points crosses the southern

region of maximum visual auroral activity. as givenby Vestine and Snyder', whereas for J odrell Bankthe locus of reflecting points lies about 9° south ofthe northern zone of maximum auroral frequency.The disparity in rates of occurrence is therefore tobe expected and is, in fact, a fortunate circumstancewhen considering the simultaneity of occurrence,since only the occasional aurorre coinciding withmoderate to great magnetic storms are detected atJodrell Bank.The relation between these widespread northern

aurorre and the Halley Bay results is shown in Fig. 1.The Halley Bay results show a close correlation

between the number ofhours a day containing echoes(irrespective of duration or amplitude) and thecorresponding daily sum of the planetary magneticK index, Kp• It can be seen that the world-wide

Fig. 1. Number of hourly intervals a day of auroral radio echoesdetected at Halley Bay and Jodrell Bank compared with the daily

sum of the planetary K index. Kp

HALL.EY lAY AURORAL E.CHOES

ItS7

9

R

G7

Amberley sample size14 45 44 62 31 17 7

MocquarieIs. sample size,.---"--- ,~m ~ ~ ~ ~ ~ ~ ~

Total Sample" 220 in t'ac~

oo

1·0

3 4 5 6

'Magnetic K·index

Fig. 1. Relative intensity of the 01 lines at 6300 A. and 5577 A.as a function of K-index. The heavy curves are the relativeintensities for the southern sky plotted as a function of the K-indexat Macquarie Island, the thin curves are the relative intensities forthe northern sky as a function of Macquarie Island K-index andthe ;lotted curves the relative intensities for the northern skyplotted as a function of the K-index at Amheriey. The sample sizesfor Amberiey and Macquarie Island are given at the top of the

flgnre .

Probo b~ity .6

of theIntensity of .7~\

6300> 5577 (R)

and .66300< 5577(G)

·5

·9

·1

International GeophysicalYear Station,DominionPhysical Laboratory,

Awarua Radio,Invercargill,New Zealand.

A Comparison of Radio Echoes from theAurora Australis and Aurora Borealis

THEproblem of the relation of the occurrence ofaurorre in the northern and southern hemispheres isone of considerable importance to the theories ofauroral formation. Hitherto, the assessment of theseoccurrences has been handicapped by the difficultiesof the collectionof simultaneousvisual observations inthe arctic and antarctic regions, and by the fact thatin any event aurorre occur on at least 75 per cent ofnights in the zones of maximum activity in which

When the northern sky intensities are correlatedwith the magnetic K-index from Amberley (north of

Inver~il1), as shown by the dotted lines, then thedistribution of probability for the northern sky curveis shifted into a position nearly coincident with thesouthern sky curve, which is plotted against theMacquarie Island K-index. The agreement between~ curves (the heavy and dashed curves) over therange of K from 1 to 4 where the samples are largesuggests that the relative intensity is a function ofthe local K-index.About 20 per cent of the spectra fall in the group

of equal intensity, which has a probability approximately independent of K-index. Of the 220 spectraobtained, aurorrewere recognizedas being present in

about seventy cases, either from the presence ofnitrogen bands in the spectra or from direct photography.

7/30/2019 Aurorae Australis in Radio

http://slidepdf.com/reader/full/aurorae-australis-in-radio 2/2

44 NATURE July 5, 1958 VOL. 182

C. D. WATKINS.

J odrell Bank Experimental Station,University of Manchester.

June 4.'Little, B. E., and Shrum, C. M., Trans. Roy. Soc. Oanada, 44 (III), 3

(1950).'Lovell, A. C. B., I.G.Y. Annals, 3, Part IV, 337 (1957).'Bullough, K. ,and Kaiser, T. R., J. Atmos. Terr. Phys., 5, 189 (1954).• Vestine, E. H., and Snyder, E. J., Terr. Mag., 50, 122 (1945).

Ismbestic point(mil)

469'509'480'485'500'481'350'

Indicator

Table 1

I pHrange

I

- ----Iethyl orange 2 '6-5'0

Bromocresol green 3 '4-5'2Neutral red 5 '6-8'4Thymol blue (acid) 1'2-3'4

Tbymol blue (alkaline) 7,4-10'2 I

Phenol red 6 ,2-9'2p-NitropheIOol 5 ,0-9'0

factors remain constant and balance out without

affecting the accuracy of the calibration.An incidental advantage in the method is that the

procedure involves the determination of a point ofintersection; this is normally more precise thandetermination of a maximum or minimum absorb

ency. In practical determinations, the equality ofabsorbency is far easier to establish than determ

ination of the wave-length corresponding to amaximum or minimum. We have calibrated the

spectrophotometer by this method and .measured theisosbestic point of bromocresol green and have beenable to reproduce the value reported in the literature.It is, however, necessary to add here that there mustbe no impurity in the indicator, the absorbency ofwhich depends on pH. Although we have usedmethyl orange and bromocresol green, many otherindicators would also serve the purpose.

D. P. HARRISON

Royal Society Base,Halley Bay, Antarctica.

magnetic activity maintains a strong control overthe occurrence of echoes at Halley Bay over thewhole range from quiet to severely disturbed conditions.On almost all the occasions when "'L.Kp exceeded

about 40, echoes were obtained from the auroraborealis at J odrell Bank, and these events coincidedwith the high peaks in the echo activity at Halley

Bay from the aurora australis. It therefore seemsreasonable to conclude that the great auroroo of thenorthern hemisphere are accompanied simultaneouslyby greatly enhanced activity of the aurora australis.It is hoped to make a more detailed study of thetime correlations when the full results from HalleyBay become available.

A Simple Technique for the Calibrationof the Wave-length Scale of Spectro

photometersIT is often essential to test the calibration, setting

and working of either old or new spectrophotometers.

Generally, this is carried out by the standard plastic,gelatine or glass filters supplied with the instruments.Standard solutions may also be used. We wererecently faced with such a problem, but found thatno solid standard filters were immediately availableor easily obtainable. Further, calibration andstandardization with solutions is always associatedwith numerous difficulties of purity of chemicals andsolvents and the actual preparation of standardsolutions. It is well known that even chemicalslabelled 'chemically pure' are somf:Jtimes contaminated. Further, departures from Beer's law andthe exact range of wave-length in the so-calledmonochromatic beam used are also common sources

of error. The problem is often complicated bysystematic errors such as dichroism, colloidal particles or temperature effects.An attempt was therefore made to find an easy

and reliable method of calibration relatively freefrom error. This has been achieved by taking

advantage of the transmittancy curves of solutionsof acid-base indicators such as methyl orange. It isknown1 that the transmittancy curves for the differentpH values for any particular indicator all intersectat a point known as the isosbestic point, which hasa value of 469 m[L for methyl orange. It is clear,therf:Jforf:J,hat if an acidic and a basic solution made

from equal amounts of the same stock solution ofmethyl orange are kept in two cells and the differencein their transmittancy observed at different wavelengths, then ATs would be zero at only one wavelength, namely! 469 m[L, ATs can be determinedby simply sliding the two cells containing the acidicand the ba.sic solutions' of the indicator, one by one,inthf:J path of the beam. As this method is verysimilar to a. null method, all the disturbing minor

The wave-length values for the isosbestic points ofa few indicators are given in Table 1. In addition,the transmittancy curves for bromophenol blue(pH range 2·4-5·6), methyl red (pH 3·4-7'0),

bromocresol purple (pH 4·8-7 ·6), cresol red (pH6·8-9·6) and bromothymol blue (pH 5·8-8·4) are

given in the literature", from' which the exact wavelength values for their isosbestic points may beascertained.

We are not aware of any previous reports of theuse of this principle for accurate calibration.We wish to thank Dr. K. S. G. Doss, director of the

Central Electrochemical Research Institute, for hisuseful suggestions.

N. V. PARTHASARATHYINDRA SANGHI

Central Electrochemical Research Institute,Karaikudi,South India.

'cf. ]'ortune and Mellon, J. Amer. Chern. Soc., 60, 2607 (1938).'Wallace, R. Brode, J. Amer. Chern. Soc., 46, 589 (1924).'Briggs, A. I., Trans. Farad. Soc., 50, 800 (1954).

Photovoltaic Pile

IN a previous communication!, a retinal type ofphotovoltaic cell is described in which many inert'point' electrodes are placed in a parallel mosaicarrangement in a small volume of photo-activesolution. To indicate that each electrode is a photoreceptor, a separate d.c. amplifier is required. Anattempt was then made to construct a multiple-pointphotovoltaic cell in which all the individual electrodesare connected in series, as shown in Fig. 1. One setof electrodes is kept dark with tape while the otheris irradiated as in a thermopile. Only one amplifier1is of course needed for this cell. As more light- anddark-electrode pairs are added in series, larger photopotentials are eventually generated, but at a muchslower rate than with the individual point electrodesin parallel; also, with certain photo-active solutes,