solar activity and terrestrial disturbances

Solar Activity and Terrestrial DisturbancesAuthor(s): Donald H. MenzelSource: Proceedings of the National Academy of Sciences of the United States of America,Vol. 40, No. 10 (Oct. 15, 1954), pp. 973-978Published by: National Academy of SciencesStable URL: http://www.jstor.org/stable/89357 .

Accessed: 03/05/2014 14:24

Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at .http://www.jstor.org/page/info/about/policies/terms.jsp

.JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range ofcontent in a trusted digital archive. We use information technology and tools to increase productivity and facilitate new formsof scholarship. For more information about JSTOR, please contact [email protected].

.

National Academy of Sciences is collaborating with JSTOR to digitize, preserve and extend access toProceedings of the National Academy of Sciences of the United States of America.

http://www.jstor.org

This content downloaded from 130.132.123.28 on Sat, 3 May 2014 14:24:33 PMAll use subject to JSTOR Terms and Conditions

http://www.jstor.org/action/showPublisher?publisherCode=nas

http://www.jstor.org/stable/89357?origin=JSTOR-pdf

http://www.jstor.org/page/info/about/policies/terms.jsp


BY DONALD H. nENZEL

HARVARD COLLEGE OBSERVATORY

Scientists have lolng recognized that solar activity is somehow responsible for, closely connected with, magnetic disturbances, the aurora polaris, ionospheric irms, vagaries of radio propagation, and certain other terrestrial phenomena. le research program for the International Geophysical Year would be incomplete it omitted studies of variable solar features. The terrestrial effects must arise the result of radiations, electromagnetic or corpuscular, emitted by active solar ,ions. The common statement that "sunspots cause magnetic storms" tells us thing. It is an expression of our ignorance of the detailed physical processes it must govern the interaction between sun and earth. However, attempts to connect specific solar events with specific terrestrial ents have been disappointingly inconclusive. We are still unable to designate ne particular area of the solar surface as an "M region," or area emitting the liations that cause the magnetic and other disturbances. The fact that sunspot numbers follow closely the curves of terrestrial magnetic ;ivity in no way signifies that spots, per se, are the basic phenomena. Of course, still do not know precisely what a sunspot actually is, though we recognize

ne of its main attributes. Detailed discussions of sunspot characteristics appear elsewhere.l A brief sur- y of the primary features of solar activity will be useful. Each spot contains a rk core or center, the so-called "umbra," surrounded by the less dark "penum- t," whose fibrous outer structure sometimes curves so as to suggest vortex )tion. Clear-cut cases of this phenomenon are so rare, however, that one may iclude that rotatory circulation is not essential for spot development. Spot ibras range in size from small "pores," of the order of 1,000-km. diameter, up really large spots, whose umbras occasionally attain diameters of 40,000 km. or )re. rhe umbra is dark compared with the surrounding photosphere, because of its atively low temperature. The spot spectrum shows spectral lines, split into nponents and possessing polarizations characteristic of the Zeeman effect. e must unavoidably conclude that magnetic fields of intensity up to 4,000 iss are present in the larger spots. Spots tend to occur in pairs, of opposite magnetic polarity -so-called "bipolar )ups." In the majority of large spots many smaller spots attend the dominant mbers. Associated with the spot group, and evidently a sensitive indicator of : solar disturbance, we find the patches of bright faculae, covering an area ap- ,ciably greater than that of the spot. These faculae precede the formation of a )t and persist after the spot has disappeared. The bright calcium flocculi, which mupy essentially the same regions as the faculae, are even more sensitive in out- ing the region of disturbance. Spots tend to differ in various ways. Speaking of the larger groups, we may ride them into two main categories: active and inactive. There is no simple y of distinguishing between the two. Furthermore, a given spot may. be active times and inactive at others. The most evident feature that enables us to pick

973



t the active spots is the tendency for flares to occur in the vicinity and particu- ly between the components of a bipolar group. We detect flares most readily with birefringent-filter or spectroheliograph records ken in the light of Ha. A sudden brightening, which can happen in an interval from two to ten minutes, occurs in the plage area of hydrogen flocculi near the At. Motion pictures, obtained when spots are tangent to the limb, often disclose e presence of intense activity in the chromosphere and higher regions of the solar nosphere. Pulsing surges reach to varying altitudes; luminous matter streams ;o the spot region through loops that straddle the area or through funnel-shaped fices that discharge the luminous gas in long curving filaments. We find prom- rnces elsewhere, great filamentous gas clouds, but their behavior differs markedly ,m that of those found associated with spots. It is difficult, indeed, to say whether the spot prominences, so photographed, have y relationship to flares. Many astronomers seem to hold the view that flares . photospheric phenomena and hence probably invisible at the limb. On the

l r hand, I have generally regarded the plage area associated with spots as some ?t of prominence phenomenon. Numerous minor brightenings occur above the ib-tangent prominence, and patches of luminosity form high in the atmosphere, liating intensely as the spreading patch develops into a loop or other form. casionally, we note an especially active and brilliant surge, which may attain tat height and move with high velocity.2 Intense outbursts of radio noise on frequencies of 200 me/sec and higher ac- npany the major fraction of bright flares.3 The reason for the association of the o, phenomena is not yet clear. Probably some form of magnetohydrodynamic plasma oscillations is responsible. In any event, the radio phenomenon pro- les us with another useful index of solar activity. These brightest (class 3) flares provide the most significant relationship between ar activity and a terrestrial event. For, as Dellinger showed many years ago d as many others have substantiated, radio fadeouts occur synchronously bh the intense flares. A fadeout or sudden ionospheric disturbance (SID) curs when a radio signal along some path in the sunlit hemisphere of the earth Idenly fades or disappears completely.4 The phenomenon probably occurs 3ause far ultraviolet emission (perhaps Lyman a) from the flare increases ioniza- n in the lower (D) region of the ionosphere. This excess ionization greatly ,ensifies the absorption of radio signals, especially in the lower frequencies of the ?called "HF" range. A minor oscillation (crochet) occurs in the magnetic records, simultaneously bh the other phenomena. But the duration of the entire event, flare, SID, and , is measured in minutes and rarely exceeds more than an hour or so. In passing, e may note that small increases in the intensity of the low-energy component5 of smic rays may also be associated with bright flares. The phenomenon of SID's just described is very different from that of a great ignetic storm,6 with its high-amplitude fluctuations of the compass needle, a onset of auroral displays, and the partial or complete disruption of the iono- nere and HF radio communication dependent thereon. The simultaneity of SID and flare and the confinement of the disturbance to the nlit hemisphere of the earth almost certainly identify the disturbing radiation



ultraviolet light. Although certain investigators7 have suggested a similar

igin for the magnetic storms, the preponderance of opinion appears to stand in vor of Chapman'ss proposal that a neutral stream consisting of equal numbers of ,sitive ions and electrons is responsible for the effects. Since hydrogen is the ief chemical constituent of the solar atmosphere, protons probably are the minant ions. These streams, as Chapman and others have shown, can account quantitatively r the principal features of magnetic storms in terms of electric currents induced in e conductive cloud by dynamo action caused by the earth's magnetic field. In

dition, this theory places a cloud of ions in the vicinity of the earth, ions that

nceivably could flow along magnetic lines of force into the auroral zone and pro- .ce the luminous effects of the aurora. But where do these ion clouds originate? In the explosive spot areas that pro- ce the flares? There is some evidence to support such a conclusion. A few of e most intense flares, especially those occurring not far from the central meridian the sun, appear to have expelled the ions, which encounter the earth some 24 urs later. This apparent coincidence has lent weight to a commonly accepted 3w that the ions travel through space with velocities of some 2,000 km/sec. lis view would further identify the M regions with flare areas near active spots. t us examine this hypothesis in the light of past experience. Very large spot groups are statistically associated with very intense magnetic )rms, in the sense that the disturbances tend to occur from one to four days er the spot passes the meridian of the rotating sun. Conversely, a statistical

;ociationl occurs between severe storms and large spots. But the correspondence far from one to one. Many large spots produce no magnetic activity whatever. Id a few intense storms have occurred when the sun possessed neither spots nor a bright patches of faculae that ordinarily precede or follow the development of a ot and which are particularly sensitive indicators of the presence of disturbed sions. A further statistical relationship exists involving the tendency of weaker mag- tic storms to repeat at intervals of approximately 27 days, corresponding to the nodic period of solar rotation. This result led Chapman and Ferraro to con- de that the ion streams, previously mentioned, were like streams of water from

ilowly rotating fireboat. Such localization in space and permanence in time pose many theoretical diffi- ties. Yet the outer solar corona exhibits occasional beams, rays, or streamers it strongly suggest the action of some sort of focusing. Here solar magnetic ds may play a part. Observations made with the coronagraph include records of the intensity distri- tion, around the solar limb, of several coronal lines. The green line, X 5303, m 13-times ionized iron, and the red line, X 6374, from 10-times ionized iron, are 'haps the most significant. Since these atoms indicate the presence of high 'itation, we originally hoped that the bright patches of coronae might outline M regions. In this respect, the results to date have been disappointing.

Statistical analyses by Waldmeier,9 by Shapley and Roberts,l1 and by others," the method of superposed epochs, indicated a correlation between the east-limb ssage of bright corona and the Kp indices of magnetic activity. However, since



e regions of bright corona are quasi-permanent and thus display the 27-day rota- n cycle, the interval between the appearance of a bright coronal area and the set of a storm may be purely coincidental. The discrepancy of this interval at ferent times of the spot cycle indicates a secondary rather than a casual relation-

ip. Bell and Glazer,'2 re-examining the coronal data, have applied appropriate tests - physical significance of the various relationships. I am indebted to them for rmission to quote the most significant correlation they have found to date, viz., at the east-limb passage of a corona (X 5303) minimum tends to be followed nine

ys later by a maximum of magnetic activity (Kp). In other words, Kp is high To days after the central meridian passage of a zone of low coronal excitation. The foregoing examples are selected from a large number of similar studies. ie relationships are primarily statistical; none of them yields an unambiguous termination of the location of M regions. However, if a corpuscular beam is to originate in an M region, we should expect see some indication of matter being driven from the sun. There are three dif- -ent zones, where we encounter appreciable upward motion of the solar gases. ie type of motion to be found in each is distinctive. First of all, we find sunspots, with violent surges accompanying the most active ots. Many of the surges are oscillatory, though intense eruptions may occur

igly. A large fraction of the ejected matter appears to fall back into the sun.

trely, if ever, do we obtain clear-cut evidence supporting the idea that matter has

tually left the sun. Quite the contrary! Most of it appears to cascade back, d the peak velocities usually are well under the velocity of escape. Even so, ots and their associated activity provide the most likely areas for the ejection corpuscular material. The second region of significance occurs at intermediate latitudes, between the ot zones and the polar regions. Here we usually find the long hydrogen fila-

ents, which we associate with the arching or hedgerow prominences. The minant motion through such gaseous structures is down, toward the solar surface.

casionally, however, one of these vast structures may ascend to great heights. it the ascent slows at the uppermost levels, and most of the material appears to

turn to the sun. The third, and in some respects the most promising, area of rising material lies

the regions near the poles. As Robertsl3 has shown, polar spicules do exhibit a

pe of jet activity. The material fades as it continues to rise, a behavior that

ggests the possibility of the spicules as significant sources of coronal matter and

rpuscular streams. To indicate the relative importance of the corpuscular and radial streams, we

ay estimate the energy flux of the former in terms of its perturbing effects on the 'restrial magnetic field. The dipole magnetic field of the earth possesses intensity H at radial distance

nd angle 0 from the geomagnetic pole:

2 = -

(t) (1 + 3 cos2 ), (1) 4 V/



LU. ILu, 1Ui1't u'JJI 11 . xi .- o. L. -L. YiviJd i1 oi Uf

lere Ho is the intensity at the pole, 0.7 gauss, and ro is the radius. Chapman and rraro'4 have shown that the earlier stages of a magnetic storm are roughly tat one would expect if a plane, conductive cloud of neutral ions were to approach

earth at distance d from its center, tangent to the geomagnetic equator. Since a earth's field will tend not to penetrate the conductive surface, the resulting itortions represent the disturbed terrestrial field. The closer the cloud approaches e earth, the greater the disturbance. A detailed analysis of the approach of such an ionic cloud is inherently difficult. It, for an order-of-magnitude estimate, we may use the magnetohydrodynamic ndition that penetration occurs until the energy density of the terrestrial mag- tic field equals the kinetic energy density of the moving cloud, or

H2 pv2 P = 8 (2)

iere p is the density and v the velocity of the cloud. To obtain the flux, F, of ietic energy through a surface, we multiply both sides of the above by v, or

HI2 v (d)\ 6 v F = ro =

Lo2 . (3) 8 7r \d/ 32 w

r a representative storm we may take d/ro = 5 and set 0 = 7r/2. Then, numeri- ly,

F = 3.1 x 10-7 v erg/cm2 sec. (4)

Le solar constant or radiative flux is, in general, much larger,

Fr = 1.35 X 106 erg/cm2 sec. (5)

we take v = 1.7 X 108 cm/sec, corresponding to a lone-day lag, the ratio of the xes, F/Fr, equals 4 X 10-5. Even though this corpuscular flux appears to be all, it is far from negligible. The focusing action of the earth's magnetic field ty further increase the flux in the polar regions. And we must recall that the Losphere is largely transparent to the radiative flux, whereas the corpuscular nponent must be completely stopped. rhe number of particles, counted as protons moving with the velocity v, re- ired to produce this flux is of the order of 12 per cubic centimeter. The number, wrefore, is not excessive. Indeed, the ordinary diurnal variation of the solar d would require the continued presence of at least one fast-moving ion per cubic itimeter in the vicinity of the earth. 3ne final calculation indicates a possible mechanism for the ejection of ionized itter from the sun. Recent examination of many miles of prominence motion- 'ture records, made principally at Sacramento Peak and Climax, strongly sug- sts that the ejected matter tends to form loops or arches, with the lower ends ,ached to the surface of the sun. Phe very high temperatures found in certain regions may assist, however, in ex- ling the gases; dynamic and magnetic viscosities must play a role in the for- ,tion of the filaments.



Forces such as radiation pressure or simple convections appear to be incapable of clting the material. The only one I have discovered capable of performing the quired actions is the magnetohydrodynamic force associated with a current loop. One can show that such forces are entirely adequate to account for the elevation prominence material against the force of gravity. At the same time these cur- its can produce and maintain the observed filamentation of spots. Further, it' pears that the energy stored in the current system is of the order of that released

ring a solar flare. Thus, even though we still canulot define an M region un-

ibiguously, we have achieved some understanding of the force fields that may cur in the solar atmosphere and which may have some effect on the ionosphere.

* The research reported on was made possible in part through support and sponsorship ex- Lded by the Geophysics Research Division of the Air Force Cambridge Research Center, under ntract AF 19(604)-146 with Harvard University. It is published for technical information y and does not necessarily represent recommendations or conclusions of the sponsoring agency. 1 G. P. Kuiper (ed.), The San (Chicago: University of Chicago Press, 1953); G. Abetti, Sole; D. H. Meizel, Our Sun (Cambridge, Mass.: Harvard University Press, 1949). 2 Helen W. Dodson in Kuiper, op. cit. 3 Ibid.; J. L. Pawsey and S. F. Smerd, in Kuiper, op. cit. 4 D. H. Menzel, Elementary Manual of Radio Propagation. 5 J. A. Simpson, in Kuiper, op. cit. 6 S. Chapman and J. Bartels, Geomagnetism (2 vols.; Oxford, 1940); D. H. Menzel, Our Sun. 7 E. 0. Hulbert, Phys. Rev., 34, 344, 1929; 36, 1560, 1930. 8 S. Chapman and J. Bartels, op. cit. 9 M. Waldmeier, Z. Astrophys., 18, 241, 1939; 21, 275, 1942. 0 A. Shapley and W. 0. Roberts, Astrophys. J., 103, 257, 1946. L1 J. W. Lincoln and A. Shapley, Trans. Anm. Geophys. Union, 29, 849, 1946; A. Shapley, ,restrial Magnetism and Atm., Elec., 51, 257, 1946; Trans. Am. Geophys. Union, 28, 715, 1947. 2 B. Bell and H. Glazer (unpublished). 3 W. 0. Roberts, Astrophys. J., 101, 136, 1945. 4 S. Chapman and V. C. A. Ferraro, Terrestrial Magnetism and Atm. Elec., 43, 77, 1938. Cf. also apman and Bartels, op. cit.

GEOGRAPHIC BA SIS FOR ANTAIRCTIC SCIEN1TIFIC OBSERVATIONS

BY PAUL A. SIPLE

DEPARTMENT OF THE ARMY, RESEARCH AND DEVELOPMENT DIVISION

Antarctica has emerged from total terra incognita since the beginning of this

ntury. Although it was recognized as a probable continent more than one .ndred years ago, the contiguous coastal outline was rnot completed as a reliable etch until the past decade. The state of current antarctic geographic knowledge ight well be likened to that of the Americas in the sixteenth century. The task of classical exploration has virtually ended, and the period of detailed .entifie study has already begun. With the exceptions of large sections of the in-

rior, approximating the area of the United States, the major features have been etched in (see Fig. 1). In contrast to sixteenth-century exploration tech-

ques, those of the twentieth century, together with scientific interest, have oduced small samplinug of intensive studies in Antarctica comparable in quality

Forces such as radiation pressure or simple convections appear to be incapable of clting the material. The only one I have discovered capable of performing the quired actions is the magnetohydrodynamic force associated with a current loop. One can show that such forces are entirely adequate to account for the elevation prominence material against the force of gravity. At the same time these cur- its can produce and maintain the observed filamentation of spots. Further, it' pears that the energy stored in the current system is of the order of that released

ring a solar flare. Thus, even though we still canulot define an M region un-

ibiguously, we have achieved some understanding of the force fields that may cur in the solar atmosphere and which may have some effect on the ionosphere.

* The research reported on was made possible in part through support and sponsorship ex- Lded by the Geophysics Research Division of the Air Force Cambridge Research Center, under ntract AF 19(604)-146 with Harvard University. It is published for technical information y and does not necessarily represent recommendations or conclusions of the sponsoring agency. 1 G. P. Kuiper (ed.), The San (Chicago: University of Chicago Press, 1953); G. Abetti, Sole; D. H. Meizel, Our Sun (Cambridge, Mass.: Harvard University Press, 1949). 2 Helen W. Dodson in Kuiper, op. cit. 3 Ibid.; J. L. Pawsey and S. F. Smerd, in Kuiper, op. cit. 4 D. H. Menzel, Elementary Manual of Radio Propagation. 5 J. A. Simpson, in Kuiper, op. cit. 6 S. Chapman and J. Bartels, Geomagnetism (2 vols.; Oxford, 1940); D. H. Menzel, Our Sun. 7 E. 0. Hulbert, Phys. Rev., 34, 344, 1929; 36, 1560, 1930. 8 S. Chapman and J. Bartels, op. cit. 9 M. Waldmeier, Z. Astrophys., 18, 241, 1939; 21, 275, 1942. 0 A. Shapley and W. 0. Roberts, Astrophys. J., 103, 257, 1946. L1 J. W. Lincoln and A. Shapley, Trans. Anm. Geophys. Union, 29, 849, 1946; A. Shapley, ,restrial Magnetism and Atm., Elec., 51, 257, 1946; Trans. Am. Geophys. Union, 28, 715, 1947. 2 B. Bell and H. Glazer (unpublished). 3 W. 0. Roberts, Astrophys. J., 101, 136, 1945. 4 S. Chapman and V. C. A. Ferraro, Terrestrial Magnetism and Atm. Elec., 43, 77, 1938. Cf. also apman and Bartels, op. cit.

GEOGRAPHIC BA SIS FOR ANTAIRCTIC SCIEN1TIFIC OBSERVATIONS

BY PAUL A. SIPLE

DEPARTMENT OF THE ARMY, RESEARCH AND DEVELOPMENT DIVISION

Antarctica has emerged from total terra incognita since the beginning of this

ntury. Although it was recognized as a probable continent more than one .ndred years ago, the contiguous coastal outline was rnot completed as a reliable etch until the past decade. The state of current antarctic geographic knowledge ight well be likened to that of the Americas in the sixteenth century. The task of classical exploration has virtually ended, and the period of detailed .entifie study has already begun. With the exceptions of large sections of the in-

rior, approximating the area of the United States, the major features have been etched in (see Fig. 1). In contrast to sixteenth-century exploration tech-

ques, those of the twentieth century, together with scientific interest, have oduced small samplinug of intensive studies in Antarctica comparable in quality



solar activity and terrestrial disturbances

Documents