Äîêóìåíò âçÿò èç êýøà ïîèñêîâîé ìàøèíû. Àäðåñ îðèãèíàëüíîãî äîêóìåíòà : http://www.izmiran.rssi.ru/magnetism/Feldstein_2012/PDF/355.pdf Äàòà èçìåíåíèÿ: Wed Mar 11 16:45:08 2009 Äàòà èíäåêñèðîâàíèÿ: Mon Oct 1 21:08:37 2012 Êîäèðîâêà: ISO8859-5 Ïîèñêîâûå ñëîâà: ï ï ï ï ï ï ï ï ï ï ï ï ï ï ï ï ï ï ï

GEOPHYSICAL RESEARCH LETTERS, VOL. 21, NO. 19, PAGES 2095--2098, SEPTEMBER 15, 1994

Electromagnetic weatherat 100km altitudeon3 August 1986
Y. I. Feldstein,A. E. Levitin, L. I. Gromova,andL. A. Dremuhina
IZMIRAN, Troitsk,MoscowRegion,Russia

L. G. Blomberg, P.-A. Lindqvist,andG. T. Marklund
Alfv6n Laboratory, Royal Institute Technology, of Stockholm, Sweden

Abstract. The electromagnetic weatherat high altitudes Model description

above Earth's the surface determined the transport is by of ionospheric plasma,whichin turn is governed the by magnitudewell asthedirection theelectric magas of and netic fields. Differentmodels[Levitin al., 1984; Friiset Christensen 19815; etal., Mishin, 1990] havebeen proposed that allowan estimation the electromagnetic of parametersof the upperatmosphere, given a knowledgeof the magnitude orientation theinterplanetary and of magnetic field.Here we useone suchmodelto estimate global the convection pattern and its temporalevolutionduring a pass theSwedish of satellite Vikingoverthenorthern polarcap. The modelpredictions shownto agreewell are withthe electricandmagnetic fieldsmeasured alongthe satellite trajectory.The goodagreement impliesthat the
model can be used to reconstruct, with reasonable con-

Thelarge-scale distribution convection theupper of in atmosphere (ionosphere)conveniently is represented theelecby trostatic potential, At highlatitudes ismainlydetermined
tion part, IMF BÇtandB, components and determine pan the of convection whichis related merging to [Friis-Christensen
et aI., 19815; Feldstein and Levitin, 1986; Mishin, 1990 and

references therein].TheIMF BÇt andB, components vary
morein time(overseveral hours) thanthesolar windvelocity and density. Therefore themodel Levitin al. [1984] in by et theelectric potential attheposition field Çm (ILAT,MLT) and time(t) is represented therelation by

fidence, large-scale the distribution electricandmagof
netic fields and their time-variation in the entire auroral

(I,m (ILAT, MLT, t) =

ionosphere.
Introduction

*Ç(ILAT, MLT)+ *Ç(ILAT, MLT)-BÇ(t)+
*Ç(ILAT, MLT). BÇ(t)
where Ç(ILAT, MLT) describes "quasiviscous" the convectionpan, depending n andv only. In the first approxion mation termrepresents "quasiviscous" this the convection for

In this paper, brieflydescribe model estimating we a for the electric magnetic and fieldsin theionosphere, themeawith sured interplanetary magnetic field(IMF) asinput.Theresults ofthis modelling compared actual are to measurements in the

average values n andv, namely, = 5 cm andv = 450 of n -a

km/s. The othertwo termsrepresent contribution the the to ionosphere. thecomparison havechosen pass For we a of convection arising frommerging, assumed to be linearly here theSwedish satellite Vikingon3 August 1986. On thisdate dependent the By andBz components. a resultthe on As anumber geophysically of interesting phenomena place model took gives spario-temporal the distribution ï in thehighof inthe ionosphere-magnetosphereThese system. phenomena, latitude region bothquietandweakly for disturbed intervals as and particular response changes theinterplane- wellasformore in their to in disturbed intervals which arising for the electary medium, havebeendescribed discussed and extensively tricfields controlled the"driven" are by mechanism energy of intheliterature. inter-hemispheric The conjugacy polar transfer of from the solarwind to the ionosphere relation (the cap aurorae investigated was [Craven al., 1991]based et on between andIMF maybe considered linear). ï as

simultaneous fromtheViking theDynamics images and ExThe coefficients
Copyright by theAmerican 1994 Geophysical Union.
Paper number 94GL01762
0094-8534/94/94GL-01762503.00

rents, well aswith the ionospheric as electric field (E) andits potential (Ç). Eventually system equations the of reduces a to second orderdifferential equation ï whichis thensolved for numerically. unknown The quantity thisequation thepoin is tentialï , whiletheionospheric conductivity the ground and magnetic variations knownquantities. field are
2095


2096

FELDSTEIN ET AL.: ELECTROMAGNETIC WEATHER

is to the weather Oncethepotential determined, electric is the fieldis given model itscapacity determine electromagnetic at highlatitudes continuously, givingthedependencethe of by Ig = -V,I,% theheight-integrated ionospheric current by large-scale convection thecurrent and systems theIMF. on I = :g and field-aligned byill -- V. I, where Ig, the current 2E is theheight-integrated ionospheric conductivity tensor.For comparison measured with data,themagnetic fieldvariations Modelapplication andcomparison Viking with at theposition Vikingcaused themodelcurrent of by system observations (ionospheric well asfield-aligned as currents) werecalculated - (see Dremuhina al. [1985]). also et The electricfield instrument Viking measured two on the Whendeveloping modelof theabove a described type,inac- spin plane components electric [Block al.,1987; ofthe field et curacies arise,connected imperfections theiono- Lindqvist Marklund, will with in and 1990.] Oneof these components spheric conductivity modelmÇd Earth's the conductivity model is perpendicularthemagnetic to fieldandoftennearly anti(ionospheric currents induction effect)usedin the analysis. parallel thesatellite's to velocity theparticular used (for pass The validityof themodelbecomes evident only aftercompar- here,they are anti-aligned within 20 degrees).The other to isons withexperimental data.Possible masons discrepancy component oftennearlyparallelto themagnetic for is field. The between modelandthedatainclude assumptions the the that interplanetary magnetic field was monitored the iMP 8 by

thegeomagnetic lines equipotentials, theelectric field are that fieldis curl-free, etc. Comparisons themodelelectric of and magnetic fields themodel and currents satellite from with data TRIAD,MAGSAT COSMOS-184 and [Feldstein andLevitin,

satellite,which recordedsubstantial variations. As a conse-

quence, ionospheric the convection pattern changes considerablyduring course thisevent,asis demonstrated the of below. Figure1 shows modeled the potential patterns thebeginin 1986 andreferences therein]have shownthat the modelrea- ning andin the middleof a Viking passon 3 August1986. of sonably describes spatial well the dynamics thecurrent of sys- Plasmaflows alongthe isocontours the potential. During temin dependence the IMF for smooth on magnetospheric theearlypart of the passthewholepolarcapis covered plasma circulation. AftertheIMF disturbances. Apparently, during these intervals so-called by a cellwith clock-wise the a system, with antidrivenmechanism energy of transfer from the solarwindto hasturnedsouthward two-cell convection flow in the polar cap, is established.Thus, in the themagnetosphere prevails.The modelis, however, ap- sunward not of the plasma motion plicable events to characterized strong by magnetospheric beginning theVikingpass high-latitude disto trajectories poleward 75 degrees of geomrbances whennon-linear effectsandlocalionospheric con- is confined closed latitude.Thismeans thattheionospheric ionization ductivity variations animportant play role. The virtueof the magnetic

MODEL EQUIPOTENTIAL DISTRIBUTIONS
August !986 3,
1 8z=6nT,Sy=9nT

].,2 Bz=-6nT. By=-SnT

.,,...'-'

06
,,

i

o I'.ILT

o MLT

1715UT

1810UT

Figure Ionospheric 1. equipotential distributions atnorthern latitudes UT(left) at1810 high at1715 and UT(fight). contour The separation kV. polygonalshows projection Viking is!0 The line the ofthe satellite trajectory km to100 altitude. dots The indicate satellite the location 5minutes 1710 to1925 every from LIT LIT, the circle and open indicatesinstantaneous ofViking. model the position The results that show witlfin less one the than hour potential distribution drastically. coordinate isbased corrected changed The grid on
geomagnetic and latitude magnetic time. and are components. local BÇt Bz IMF


FELDSTEIN

ET AL.: ELECq'ROMAGNEÇC

WEATHER

2097

MODELLED ELECTRIC FIELDVECTORS ALONGTHE
VIKING TRAJECTORY

which carries with it the IMF, to travel from the !MP.8 satel-

lite to the magnetopause, well as the subsequent as time for the change the IMF at the magnetopause influencethe in to ionospheric convection pattern.The modelcalculations were performed different with delaytimes(AT), in therange10-30 minutes. Figure3 (toppanel)AT is 20 minutes.Fromthe In Figurewe notethat the agreement between modeledand the measured field is good. The correlation coefficient between
the curves is r = 0.88 4- 0.07 and the standard deviation is

o- = 15.5 mV/m.

The values of r decrease and those of cr

increase for other values of AT.

Figure3 (bottompanel)presents variations the alongthe orbitof thedawnward component themagnetic of field,measured Viking (ABa), andcalculated by fromthemodeled currentsystems (ABp), respectively. ABÇ wasdetermined by subtraction a"baseline," of representing magnetic the effectof internal currents. This baseline wasdetermined the straight as
line from the measured field value at 1720 UT to that measured
10 mV/m

1930UT [Erlandson al., 1991]. Themagnetospheric et cur-

rents contributing theremaining to variations Ba aremainly in field-aligned. in thecaseof the electric As field, a delaytime 0b (AT) of 20 minutes wasused.The Figuredemonstrates that thereis goodagreement alsobetweenthe modeledand the magnetic fieldvariations alongtheViking orbit. Figure Horizontal 2. (perpendicular component toB) ofthe observed

model electric along Vikingtrajectory, field the plotted every 5 minutes. 20-minute A time delaybetween measured the IMF
and electric the fieldresponse assumed. was

Summary
In conclusion, modelhasbebnshown reasonably the to well

predict large-scale the electric magnetic and fieldvariations

inthe polar becomes small cap fairly because theabsence of ofplasma transport lower from latitudes. themiddle Near of the satellite convective pass plasma transport thepolar into cap auroral sub-auroral from and latitudes in. In this sets case, ahigher degree ionization thepolarcapionosphere of of is
expected.

MODEL-VIKhNG

COMPARISON

August3, 1986

[

",Ç.--x

ELECTRIC FIELD

Significant temporal changes theconvection in pattern during course thesatel!ite's the of traversal thehigh-latitude of region means theelectric measured thesatellite that field by changes aswell.Toverifythepredicting power themodel, of model potential patterns calculated each were for 5-minute
interval. These werethencompared thesatellite to measure-

30 [

/ f"X.. \',
/

E2 componcnt

/ !
..........

,",// Ç t/

\", \

ments each from correponding 5-minute interval.Figure 2
shows modeled the electric fieldalongthesatellite trajectory.

|

I a'r-=20Çl n

obsÍÇ,Ç,tions ......

r---0.88 [

|

The plasma convection velocity depends the magnitude on
(and direction) the electric of field. With a typicalmagnetic

field strength theauroral in region 60 t,T, theconvection of velocity 1 km/sfor anelectric of 60 mV/m. Thisis is field the speed which with plasma transported thepolarcap is into onthedayside. thenightsector, in where electric the field isweaker, about mV/m, theplasma 20 flowsat a speed of roughly m/s(ationospheric 300 altitudes).
Themodeled electricfield shownin Figure2 hasbeenpro-

loÇ
,,,,//

f'X,Ç
Ç...

MAGNETIC FIELD

}
i-----4--.--.t I I i,, : :

.

model

observations ......
,Ç, ,I I Ç---Ç ] : Ç', : ',

jected thesatellite's plane, allow direct onto spin to for compailson themeasurements. resultis foundin Figure with The

Ç, Ç,, --t----+,----+-,--4---I

Intercomparison modeled of (solidcurve)and (dashed curve)electric fieldandmagnetic fieldalong and measured the electric fields along Vikingtrajectory. the To measured The fieldcomponent plotted (top makequantitative a comparison possible measured the electric theVikingtrajectory. electric panel) theonewhichis perpendicular themagnetic is to field field "mapped" 100kmaltitude, theassumption was to using spinaxis. It is positiveroughly ofa vanishing magnetic-field-aligned fieldbetweenas well as to the satellite's electric dusk.The magnetic field component plotted(bottom the altitudes two (Viking's altitude approximately was 10(X)0 towards km). panel) theonewhichis perpendicular themagnetic is to field roughlytowards Since iMF is measured the up-stream theEarth, of account aswell asto the Sundirection.It is positive must takenfor the time it takesthe solarwind plasma, dawn. be

3(top panel) which shows comparison a between modeled Ç'J?uÇe $. the


2098

FELDSTEIN

ET AL.: ELECTROMAGNETIC

WEATHER

alongthe -Viking Çajectory. The agreement suggests the Feldstein, I., V.G. Vorobjev, V. Leontyev, D. Elphinstone, that Y. S. R. modelcanbe usedto describe temporal the evolution the of I. I. Alexeev, E. S. Belenkaya, and Auroras thepolar in cap, IRFScientific Report 209, ISSN 0284-1703, 123-140,1992. p. electrostatic potential (convection), theelectricandmagand neticfieldsin tÇhe entirehigh-latitude region. Thus,the dec- Friis-Christensen,E., Y. Kamide, A.D. Richmond, and S. Matsushita, Interplanetary magnetic field control high-latitude of tromagnetic weather thehigh-latitude in ionosphere, whichis electric fieldsandcurrents determined Greenland from magnemainlydetermined thestateof theinterplanetary by medium, tometer data,J. Geophys. Res.,90, 1325-1338,1985. maybedescribed themodelof thelarge-scale by electric fields Levitin,A. E., R. G. Afonina,B. A. Belov,andY. I. Feldstein, and currents [Levitin al., 1984:] et presented thispaper. in Geomagnetic variation and field-aligned currents northern at high-latitudes their relations solarwind parameters, and to
Phil. Trans. R. Soc., A304, 253-301, 1982. Acknowledgments. This work wassupported the Russian by Foundation Fundamental of Research, grant93-05-8722, andby Levitin, A. E., Y. I. Feldstein,R. G. Afonina, B. A. Belov,L. A. Dremuhina, D. S. Faermark, Yu. Z. Demidora, M. Yu. the Swedish NationalSpace Board.

Markova, I. Avdushin, V. Mihnevich, S. V. P.M. Scidckiy, and
References
V. Yu. Gaidukov,The model large-scale electric field andcur-

rentsin high-latitude ionosphere, Gidrorneteoizdat, Moscow, parts1-6, 1984.

Block,L. P., C.-G. FÇilthammar, P.-A. Lindqvist,G. T. Marldun& Lindqvist, P.-A.andG. T. Marklund, statistical A study highof F. S. Mozer, andA. Pedersen, Measurement quasistatic of and latitude electric fields measured on the Viking satellite,J. low frequency electric fieldson theViking satellite, RoyalInst. Geophys. Res.,95, 5867-5875,1990. of Technology Rep.,TPdTA-EPP-87-02,Stockholm, 1987. Mishin, V. M., The magnetogram inversiontechnique some and Craven,J. D., J. S. Murphree,L. A. Frank, and L. L. Cogger, applications, Sci.Rev., 83-163, Space 53, 1990. Simultaneous opticalobservations transpolar in thetwo of arcs polar capswith DE-1 and Viking, Geophys. Res. Lett., 18,
2297-2300, 1991.

calculations currentsand magneticfields along a Magsat hinaIZMIRAN, Troitsk,MoscowRegion,142092,Russia. of L. G. Blomberg, P.-A. Lindqvist and G. T. Marklund, Ditrajectory, Geophys. J. Res.,90, 6657-6661, 1985. Erlandson, E., D. G. Sibeck,R. E. Lopez,L. J. Zanetti,and vision of Plasma Physics,Alfv6n Laboratory,Royal Institute R. T. A. Potemra, Observations solarwind pressure of initiated of Technology,S-10044 Stockholm, Sweden. (e-mail: blomlindqvistÇplasma.kth.se, marklund@plasfastmodewavesat geostationary andin thepolarcap,J. berg@plasma.kth.se, orbit ma.kth.se) Atmosph. Terr. Phys.,53,231-239, 1991.
Feldstein, Y. I. and A. E. Levitin, Solar wind control of electric

Dremuhina, L. A., Y. I. Feldstein, and A. E. Levitin, Model

Y. I. Feldstein, A. E. Levitin, L. I. Gromova, and L. A. Dremu-

6 1993;revised May 1994; accepted 20 24 fieldsandcurrents the ionosphere, Geomag. Geoelectr., (Received December in J.
38, 1143-1182, 1986.

June1994)