Studies of the Magnetospheric Substorm, 3, Concept of the ...

JOURNAL OF GEOPHYSICAL RESEARCH, SPACE PHYSICS VOL. 73, NO. 5, MARCH 1, 1958Studies of the Magnetospheric SubstormConcept of the Magnetospheric Substorm and its Relation to ElectronPrecipitation and MicropulsationsF. V. CORONITI, R. L. MCPHERRON, AND G. K. PARKS 1Department o• Physics and Space Sciences LaboratoryUniversity o• California, Berkeley, California 94720The concept of the auroral substorm is reviewed in terms of recent satellite measurementsof the magnetospheric effects of a substorm, and a more general and inclusive terminology, themagnetospheric substorm, is suggested. In support of this suggestion, it is shown that thevarious forms of geomagnetic micropulsations and energetic electron precipitation that havebeen discussed in papers I and 2 occur at their respective local times during a substorm.Balloon X-ray and micropulsation data obtained by the Berkeley group, as well as riometerabsorption, earth currents, and standard magnetograms from several auroral-zone stations,have been used.INTRODUCTIONIn papers I and 2 [Parks et al., 1968; Mc-Pherron et al., 1968, respectively], it was shownthat the forms and temporal characteristics ofenergetic electron precipitation an'd micropulsationactivity are determined by the local timeof occurrence. The purpose of this paper is topresent evidence that these various types ofactivity do not occur randomly but occur primarflyat times when an auroral substorm[AkasoJ'u, 1964] is in progress. Balloon X-rayand micropulsation data obtained by the Berkeleygroup during parts of August 1965, and September1966 will be used to show that wheneveran electron precipitation and micropulsationevent occurred, regardless of the local time ofobservation, an auroral substorm is in someo•bservations [Anderson et al., 1966; Parks et al.,1966], we suggesthat the concept of the auroralphase of its development on the night side of substorm be generalized to, include all of thethe earth. Standard magnetograms from various associate magnetospheric phenomena. To emauroral-zonestations are used to ascertain the phssize this point, we will review some of theoccurrence of an auroral substorm. No attempt magnetospheric observations that relate t.o auwillbe made to relate a particular form of ener- roral substorms.getic electron precipitation and micropulsation Several correlation studies between magnetotoa specific phase of an auroral substorm, nor spheric fields and particles measurements andwill the problem of possible systematic time auroral disturbances have appeared in the recentdelays between stations at different local timesbe discussed.x Present address: School of Physics and Astronomy,University of Minnesota, Minneapolis,Minnesota 55455.1715MAGNETOSPHERIC SUB STORMBefore presenting the observational results,the question arises as to the appropriate terminologyto describe the complicated physicalprocesses occurring in the magnetosphere thatproduce an auroral substorm. The concept ofthe auroral substorm was introduced by Chapmanand Akasofu [Akaso)•u, 1964] to describethe characteristic disturbed forms of the aurorasat different times and different longitudesafter the breakup of the auroral arcs aroundlocal midnight. The term substorm indicatesthat auroral disturbances are a fundamentalpart of the long-lived geomagnetic main phasestorms. In view of recent satellite and balloonliterature. Anderson [1965] observed that the40-kev electron islands in the tail of the magnetosphereoccurred more frequently duringtimes of high Ap. Behannon and Ness [1966]found that the magnitude of the magnetic fieldin the taft of the magnetosphere generally in-

1716 CORONITI, McPHERRON, AND PARKScreases during the main phase storm and thatit decreased during a substorm. Heppner et al.[1967] also found that the magnetic field in the[1965]. Recent observations by Heppner ei al.[1967], however, indicate that the physicalprocesses involved in a substorm may be moreregion 10-16 R• decreased after the occurrence complicated.of a negative bay. Consistent with this evidence,Anderson and Ness [1966] correlated an in-The various forms of energetic electron precipitationand geomagnetic micropulsations werecrease in the electron island fluxes with a de- shown in papers I and 2 to depend very stronglyerease in the tail magnetic field.Ness and Williams [1966] and Williams andNess [1966] observed that the collapse of thehigh-latitude trapping boundary during magneticstorms was correlated with an increase inthe magnetospheric tail field. Cahill [1966]showed the inner magnetosphere to be inflatedduring a magnetic storm, and large periodic magneticfluctuations were associated with a substorm.Lin and Anderson [1966] also. observedthat hydromagnetic waves attained their deepestpenetration of the mqgnetosphere during timesof high Kp. The dynamics of the inner magnetospherehave been further described by Carpenter[1966], who showed that the plasmapausemoved inward when magnetic activity increased.Recently, Carpenter [1967] observed that theinward motion of the whistler duets startedabout 12 minutes before the onset of a negativebay and interpreted this motion in terms of anazimuthal electric field.It is also becoming evident. that substorms areessential in understanding the over-all dynamicalprocesses of the magnetosphere. Fairfieldand Cahill [1966] related the north-south componentof the interplanetary an'd transition regionmagnetic field to the occurrence of auroralzonedisturbances. When a sudden switch to asouthward-directed component of the magneticfield occurred, an increase in auroral-zone disturbancesfollowed. These results have beensubstantiated by Schatten and Wilcox [1967],who showed that Kp was generally higher whenthe interplanetary magnetic field had a southward-directedcomponent. It is interesting tonote that the polar station used by Fairfieldand Cahill in their analysis showed an almostimmediate increase in the disturbance level afterthe switch to the southward directed magneticfield component and that the associated substormat auroral-zone latitude occurred one totwo hours afterward. This evidence lends supportto the field line reeonneetion model for theconvective magnetosphere proposed by Dungey[1963], Levy et al. [1964], and Axyord et al.on the local time at which they were observed.The results to be presented below give strongevidence for the fact that these two phenomenaoccur primafly during auroral substorms. It isclear, then, that the auroral substorm is part ofa worldwide disturbance, and that it is the dynamicalprocesses occurring throughout themagnetosphere that determine the local-timecharacteristics of the substorm. To generalize theconcept of the auroral substorm to include allof the magnetospheric effects, we propose thatthe term magnetospheric substorm be used. Theneed for such a generalization was first pointedout by Jelley and Brice [1967], when they suggeste'dthe term elementary magnetosphericstorm. More recently, McPherron et al. [1967],Brice [1967], and Parks et al. [1967] suggestedslight modifications of this terminology. Brieesuggests elementary magnetospheric substorm;however, because of its greater convenience, weurge the adoption of the shorter term, magnetosphericsubstorm.MrTaODOF A•ALYS•STo study the relation of electron precipitationand magnetic micropulsation activity to themagnetospheric substorm, we have examinedevery electron precipitation and micropulsationevent that occurred during five days of continuousdata coverage at Flin Flon, Manitoba,Canada, in August 1965. Energetic electron precipitationdata from balloons flown by theBerkeley and Sparme> groups an'd riometers,micropulsation data from induction coils andearth currents, and magnetometer data frommagnetograms have been plotted to the sametime scale, and combined into correlation plotsfor a single station. Sequences of such plotsfrom three widely separated stations were thenexamined to determine the simultaneous, localtimecharacteristics of the electron precipitationand micropulsations and their relation tomagnetospheric substorms. The three stationsand their geomagnetic coordinates are: FlinFlon, Manitoba (63.6øN, 314.8 ø), College, Alaska

STUDIES OF MAGNETOSPHERIC SUBSTORM, 3 1717(64.6øN, 256.5ø), Kiruna, Sweden (65.3øN,115.6ø).Chart records from Flin Flon containingmicropulsation and X-ray data from September1-9, 1966, and magnetograms from a number ofauroral-zone stations were also examined todetermine the occurrence of substorms and thelocal-time characteristics of the two phenomenaduring substorms.RESULTSFigures 1 and 2 contain energetic electronprecipitation, micropulsation, and magnetometerdata from August 19 to 20, 1965, which was themost active part of the five-day period. In theupper portion of each figure, large increases inriometer absorption of X-ray fluxes indicatethe occurrence of electron precipitation events.In the center, the amplitude and frequency ofBALLOONzz. K I;5 / 65FLIGHT•- z I FLIN FLON•0, n- • RIOMETER • II FLIGHT 514L• • (CHURCHILL)zo I -I t/EeUENCII• • I J • I ; •t,", • ./I•L AMPLITUDE0 ',•'-5 " ' ,I •'• "I• FF A} •COL KIR FF• 0600 I• 18•UNIVERSALTIME•ig. 1. •]eetron precipitation, micropu]sation, gn• mggnetometet dg{g at {hree aurorg]-zone•tions, August 19, 1965.

1718 CORONITI, McPHERRON, AND PARKS.,1:•KIRUNABALLOON FLIGHT j•J/JI• J J•, IVALO.BALLOONFINLANDFLIGHTl,-- I I I I InZ •,J'• FLIN FLON BALLOON• ,.FLIGHT 514 L• I I'' I I I ' Iz , , ,-,,,,.,. ,.,,, k l,",l'"""•,.-•,i.1,,,.,, ,..,",: ,.,.,, ,,v, .',,,.:,•,-.-g7',,o........,,. ,?v¾,"" /7'TM......",,:•.,,,•,,,".,'.,I I I I I I ,,,. T+HF• COLI I0000 0600 1200 1800 2400UNIVERSALTIMEFig. 2. Electron precipitation, micropulsation, and magnetometer data at three auroral-zonestations, August 20, 1965.micropulsations at Flin Flon are plotted. Themicropulsation data from College and Kirunaconsisted of slow-speed earth current recordings.Therefore, it was not possible to make quantitativescalings for these records. However, theoccurrence and general character of micropulsationevents at these stations can be recognizedin the original data and were found to be inaccordance with Figure 13 of paper 2. In thebottom of each figure, magnetograms from thethree stations are superimposed to show theoccurrence of substorms.A. 1500 UT, August 19, 1965. The significanceof the following observations is betterunderstood when the geographic positions ofthe three stations are considered. Figure 3 showstheir locations at 1500 UT; College was near0500, Flin Flon at 0800, and Kiruna at 1600local time.Figure I shows a strong negative bay at Col-

STUDIES OF MAGNETOSPHERIC SUBSTORM, 31719NOON!i!!!iiii!iiiiiiiiiiiiiii!:'i::::iii::iiiiZO..NEiiiiiii:.. ........................MIDNIGHT(LOCAL TIME)Fig. 3. Position in local time of various auroralzonestations at 1500, UT.begin to rise about this time. Energetic electronprecipitation began sharply at 0515 UT andhad no evident structure. At 0640 UT an impulsiveelectron precipitation event (first largespike in Flin Flon balloon data) was accompaniedby a weak noise. A second impulsiveelectron precipitation event occurred (see secondspike in Flirt Flon Balloon data) at 0735UT, in conjunction with a very strong noiseburst. Sudden changes in both the Flirt Flonand Collage magnetograms at this instant indicatethis is the beginning of the expansionphase of the associated auroral substorm. Thisinterpretation is verified .by an examination ofthe magnetogram from a station at Baker Lake,which is considerably north of Flin Flon. Thismagnetogram (not shown) recorded the firstmagnetic disturbance at 0740 UT. In particular,the characteristic feature of this disturbancewas a rapid increase in the vertical component,which was just the disturbance expected for awestward line current moving northward towardBaker Lake. Coincident with noise burst at0735 UT, band-limited pulsations began to berecorded at Flin Flon. This was evident fromlege at 1400 UT indicating that an auroralbreakup occurred over central Russia. No significantdisturbance was evident on the FlinFlon or Kiruna magnetogram. The electron precipitationassociated with this substorm occurredfirst at College, then .at Flin Flon, and later atKiruna. The electron precipitation at Flirt Flonwas modulated with a 10-sec perio'd and hadoccasional microburst groups. An examinationof the riometer records from College and riometerand X-ray data from Kiruna shows thatthe rapid increase in frequency of the Flin Flonmicropulsation that began at this time. At 0820UT, the micropulsations reached a peak in frequencycorresponding to a 10-sec period, andmodulated electron precipitation began withessentially the same perio'd. About 90'00 UT athe electron precipitation was of worldwide extent.Somewhat earlier at 1400 UT a large-NOONamplitude, high-frequency micropulsation eventibegan at Flirt Flon and was of the band-limitedirregular type described in paper 2. The Col-Il•ge earth currents show a high-frequencymicropulsation event that coincided in time withthe one at Flin Flon. No micropulsation activitywas evident in the Kiruna earth currents.B. 0500-1500 UT, Augusi 20, 1965. Thegeographic location o.f the three stations interms of local time for this event is shown'inDUSKNFigure 4. The arrow indicates the region of local0500 UTtime through which each station rotates duringthe event.An examination of the Flirt Flon data (Figure2) shows a close relationship between all threerecords. The ionospheric currents eventuallyresponsible for a large negative bay first causedMIDNIGHT( LOCAL TIME)a magnetic disturbance at Flirt Flon zbout 0500. Fig. 4. Position in local time of various auroral-The micropulsation amplitude and frequenc. yzone stations at 0500 UT.

1720 CORONITI, McPYIERRON, AND PARKSsecond substorm apparently began before thefirst substorm was completely recovered. Theexpansion phase of this substorm appeared tooccur between 1015 and 1045 UT during whichthe Flin Flon magnetometer reached a maximumnegative excursion in H component.Again, this interpretation was verified by anincrease in the vertical component at BakerLake. During this second substorm, only occasionalfluxes of electrons were observed atFlin Flon, and a second band-limited irregularevent occurred. About 1130 to 1145 UT a noisewith large sequences of very regular 10-see pulsations.Band-limited irregular micropulsationsof similar period were still present.Simultaneous balloon X-ray data for thisevent from Kiruna have been published by theSparmo group [Eh•ert et al., 1966]. The X-raydata for August 19-20, when available, havebeen plotted in place of the Kiruna riometer.At 0400 UT, August 20, a silght increase inX-ray flux was observed at Kiruna and reacheda peak at 0500 UT. It should be noted that theonset time and the time of maximum X-ray intensityat Flin Flon and Kiruna were the same.From 0800-1000 UT, a large flux of X rays weredetected at Kiruna. This period corresponded tothe recovery phase of the first substorm andcoincided with the detection of large X-rayfluxes at Flin Flon. No indication of electronprecipitation was recorded in the College riometeruntil the beginniag of the second substorm,during which low-level electron fluxeswere observed simultaneously at all three stations.The most striking confirmation of the worldwideextent of the electron precipitation isdemonstrated by the third substorm. Large electronfluxes were observed simultaneously over12 hours of longitude beginning at about 1200UT. Electron precipitation begins first at Collegeand then almost simultaneously at FlinFlon and Kiruna. Band-limited irregular micropulsationswere recorded at Flin Flon throughoutthe entire period 0800-1400 UT (2-8 localtime). They were also observed at College duringthe third substorm, when College finallyrotated into the local time interval 0200-1000.This type of micropulsation was not recordedat Kiruna, since Kiruna had passed beyon'd theappropriate local-time interval by the time thefirst substorm reached maximum development.OBSERVATIONS FROM SEPTEMBER 1-9, 1966burst was recorded that preceded the beginningof a third substorm. It appears that noise burstsare a characteristic feature in the development ofan auroral substorm. (This concept is beingexamined in more detail, and the results will bereported in a later publication). The expansionphase of this third substorm began about 1225UT as shown 'by the sudden enhancement of anegative bay at College. The electron precipi-Simultaneous balloon X-ray and micropulsationdata obtained at Flin Flon from September1-9, 1966, were compared with standard magnetogramsfrom several stations to furtherexamine the relationship. of electron precipitationand micropulsations to the magnetosphericsubstorm. The results of this analysis show thatevery micropulsation and electron precipitationevent duriag this time occurred when a magtationstarted at 1230 UT and was modulated netospheric substorm was in progress somewhereon the night side and confirm the relationshipsfound in Figures 1 and 2 to holdgenerally. The temporal characteristics of theelectron precipitation and micropulsations andthe local time at which they were observed atFlin Flon are noted and agree with the localtimeoccurrence patterns presented in papers1 and2.CONCLLISIONHaving shown in papers 1 and 2 that whenelectron precipitation and micropulsations occurat a given local time, they possess distinct temporalcharacteristics for each local time, evidencewas presented in this paper that thesetwo phenomena occur almost simultaneouslythroughout the auroral zone during an auroralsubstorm. Aka•o]u [1964] showed that at differentphases of the substorm and at differentlongitudes, the physical appearance and temporalcharacteristics of the auroral forms possessedwell-defined structures and occurrencepatterns. Even though simultaneous balloon X-ray and micropulsation experiments from manydifferent stations have not been performed fromthe established local time occurrence patternand the relationship to the substorm that wasdemonstrated above, it can be concluded thateach specific form of energetic electron precipitationand micropulsation is probably occurring

STUDIES OF MAGNETOSPHERIC SUBSTORM, 3 1721at its respective local time during some phaseof the substorm development.The fact that the occurrence of an auroralsubstorm is present, not only in auroral disturbanees,but also in electron precipitation andmicropulsation activity throughout the auroralzone, indicates that it is the dynamical processesoccurring in the magnetosphere that determinethe local-time characteristics of the substorm.Recent satellite observations have shown thatmany magnetospheric phenomena are correlatedwith substorms. To generalize the concept of theauroral substorm to. include the worldwide disturbaneecharacteristics and to emphasize theimportance of the magnetosphere in auroralzoneobservations, we have suggested the term,magnetopsherie substorm.Acknowledgments. We would like to thankboth Professor K. A. Anderson and ProfessorR. R. Brown for their interest and advice givenin many helpful discussions.This research was supported by the NationalAeronautics and Space Administration undergrants NsG-243 and NsG-387; the National ScienceFoundation under grants GP-5583, GP-5331,and GA-734, and the Office of Naval Researchgrant Nonr 222(89).REFEREI•CESAkasofu, S.-I., The development of the auroralsubstorm, Planetary Space Sci., 12, 273, 1964.Anderson, K. A., Energetic electron fluxes in thetail of the geomagnetic field, J. Geophys. Res.,70, 4741, 1965.Anderson, K. A., L. M. Chase, H. S. Hudson, M.Lampton, D. W. Milton, and G. K. Parks,Balloon and rocket observations of auroral-zonemicrobursts, J. Geophys. Res., 71, 4617, 1966.Anderson, K. A., and N. F. Ness, Correlation ofmagnetic fields and energetic electrons on theIMP I satellite, j. Geophys. Res., 71, 3705, 1966.Axford, W. I., H. E. Petschek, and G. L. Siscoe,Tail of the magnetosphere, J. Geophys. Res., 70,1231, 1965.Behannon, K. W., and N. F. Ness, Magneticstorms in the earth's magnetic tail, J. Geophys.Res., 71, 2327, 1966.Brice, N., Morphology of elementary magnetospheriesubstorms, Conjugate Point Symposium,June 1967, Boulder, Colorado, p. IV-12-1, 20,ITSA Mem. 72, July 1967.Cahill, L. J., Jr., Inflation of the inner magnetosphereduring a magnetic storm, J. Geophys.Res., 71, 4505, 1966.Carpenter, D. L., Whistler studies of the plasmapausein the magnetosphere, 1, Temporal variationson the position of the knee and someevidence on plasma motions near the knee, J.Geophys. Res., 71, 693, 1966.Carpenter, D. L., Direct detection by a whistlermethod of the magnetospheric electric fieldassociated with a polar substorm, PlanetarySpace Sci., 15, 395, 1967.Dungey, J. W., Interactions of solar plasma withthe geomagnetic field, Planetary Space Sci., 10,233, 1963.Ehmert, A., G. Kremser, G. Pfotzer, K. H. Sarger,K. Wilhelm, W. Rideler, A. Brewersdorfi J.P.Legrand, M. Palous, J. Oksman, and P. Tanskanen,Simultaneous measurements of auroralX-rays at Kiruna (Sweden) and Ivalo/Sodankyl•i(Finland) from July to September, 1965,Sparmo Bull. 2, 19'66.Fairfield, D. H., and L. J. Cahill, Transitionregion magnetic field and polar magnetic disturbances,J. Geophys. Res., 71, 155, 1966.Heppner, J.P., M. Sugiura, T. L. Skillman, B. G.Ledley, and M. Campbell, OGO-A magneticfield observations, Goddard Space Flight Centerpubl. X-612 67-150, 1967.Jelley, D. H., and N. Brice, Association betweenincreased radiation in the outer belt and polarauroral substorms (abstract), Trans. Am. Geophys.Union, 48, 180, 1967.Levy, R. H., H. E. Petschek, and G. L. Siscoe,Aerodynamic aspects of the magnetosphericflow, AIAA J., 2, 12; 1964.Lin, R. P., and K. A. Anderson, Periodic modulationsof the energetic electron fluxes in thedistant radiation zone, J. Geophys. Res., 71,1827, 1966.McPherron, R. L., G. K. Parks, and F. V. Coroniti,Relation of Correlated magnetic micropulsationsand electron precipitation to theauroral substorm, Conjugate Point Symposium,June 1967, Boulder, Colorado, p. 111-4-1, 15,ITSA Mem. 72, July, 19'67.McPherron, R. L., G. K. Parks, F. V. Coroniti,and S. H. Ward, Studies of the magnetosphericsubstorm, 2, Correlated magnetic micropulsationsand electron precipitation occurring duringauroral substorms, J. Geophys. Res., 73(5), 1968.Ness, N. F., and D. J. Williams, Correlated mag-netic *• and r•rdiation belt o•scrvadons, •' •' •. •Geophys. Res., 71, 322, 1066.Parks, G. K., F. V. Coroniti, R. L. M ePherron,and K. A. Anderson, Stu. dies of the magnetospheriesubstorm, 1, Characteristics of modulatedenergetic electron precipitation during auroralsubstorms, J. Geophys. Res., 73(5), 1968.Parks, G. K., R. L. MePherron, and K. A. Ander-son, Relation of 5- to 40-second-period geomagneticmicropulsations and electron precipitationto the auroral substorm, J. Geophys. Res., 71,5743, 1960.Parks, G. K., R. L. MePherron, and F. V. Coroniti,Relation of energetic electron precipitationand geomagnetic micropulsations to auroral substorms,paper presented at the Birkland Sym-

1722 CORONITI, McPttERRON, AND PARKSposium, Norway, September 1967 (to be publishedin Ann. Phys.)Schatten, K. H., and J, M. Wilcox, Response ofthe geomagnetic activity index K• to the interplanetarymagnetic field, J. Geophys. Res., 72,•i185, 1967.Williams, D. J., and N. F. Ness, Simultaneoustrapped electron and magnetic tail field observations,J. Geophys. Res., 71, 5117, 1966.(Received August 4, 1967;revised November 29, 1967.)