Polarization pattern of low-frequency geomagnetic field fluctuations (0.8-3.6 mHz) at high and low latitude

Low frequency geomagnetic field fluctuations at cap and low

latitude during October 29-31, 2003

S. Lepidi, L. Cafarella, L. Santarelli

Istituto Nazionale di Geofisica e Vulcanologia, Rome, Italy

ABSTRACT

On October-November 2003 complex interplanetary structures, originated by a series of

solar eruptions, hit the Earth, triggering violent Sun-Earth connection events. In this paper we

analyze the low frequency geomagnetic field fluctuations detected on the ground during Oct.

29-31, 2003, a time period characterized by extremely high solar wind speed values and by

out-of-ecliptic interplanetary magnetic field orientation for intervals of several hours. We

analyze geomagnetic field measurements at four high latitude stations located in the polar cap,

three in the southern and one in the northern hemisphere. From a comparison with

simultaneous measurements at low latitude, we address the question of the global character of

the observed phenomena. The results show, for selected time intervals, the occurrence of

simultaneous fluctuations at all the stations, with high coherence even between high and low

latitude; it is interesting that these fluctuations are detected during open magnetospheric

conditions, when the high latitude stations are situated well within the polar cap, i.e. far from

closed field lines.

Key words: Magnetospheric ULF waves, Sun-Earth connections, polar cap.

INTRODUCTION

Several studies analyzed the interaction between coronal ejecta, characterized by long

periods of out-of-ecliptic interplanetary magnetic field (IMF) orientation, and the Earth’s

magnetosphere. While southward IMF conditions are associated with major geomagnetic

storms (Farrugia et al., 1996; Lepping et al., 1991; Gopalswamy et al., 2005), during

northward IMF conditions solar wind (SW) density enhancements can compress the

magnetosphere and trigger geomagnetic field fluctuations (Francia et al., 1999).

In previous papers (Villante et al., 1998; Lepidi et al., 1999) we analyzed the geomagnetic

field fluctuations at low and cap latitude triggered by magnetic clouds. In addition to storm

signatures, we found at both sites fluctuations following SW pressure variations; we also

found, during closed magnetospheric conditions, fluctuations at discrete frequencies (few

mHz) with high coherence between the stations. These results have been interpreted in terms

of global cavity/waveguide modes of the whole magnetosphere which acts as a resonant cavity

when excited by external stimulations, such as SW pressure pulses or Kelvin-Helmholtz

instability on the magnetopause (Walker et al., 1992; Harrold and Samson, 1992). Evidence

for such global magnetospheric oscillations had already been found at auroral latitudes

(Samson et al., 1992; Ziesolleck and McDiarmid, 1995), at low latitude (Ziesolleck and

Chamalaun, 1993) and even within the polar cap (Villante et al., 1997).

In the auroral region, i.e. the region separating open and closed field lines, intense

ionospheric currents can give rise to local variations of the geomagnetic field (Campbell,

1997); around midnight these features are mostly related to substorm occurrence (Olson,

1986). During daytime, typical polar cusp signals are observed in the Pc5 frequency band but

they find no correspondence in higher latitude observations (Lanzerotti et al., 1999).

Recent studies (Yagova et al., 2002, 2004) have shown the occurrence of geomagnetic

pulsations with frequency of few mHz specific to the polar cap; these cap pulsations are

decoupled from simultaneous auroral activity in the same magnetic local time (MLT) sector.

In this study we focus on the low frequency geomagnetic field fluctuations (~1-5 mHz)

observed at high latitude on October 29-31, 2003 when complex interplanetary structures,

with anomalous IMF and SW conditions, hit the Earth.

We analyze the geomagnetic field variations at the three Antarctic stations Mario Zucchelli

Station (formerly Terra Nova Bay, international geomagnetic observatory code TNB), Scott

Base (SBA) and Dumont D'Urville (DRV), at the same geomagnetic latitude but different

MLT, and at the Canadian station Cambridge Bay (CBB), at the same MLT and almost

opposite corrected geomagnetic latitude as TNB (Table 1). The Antarctic stations are located

in the southern polar cap, at the footprint of open geomagnetic field lines and around local

geomagnetic noon they approach the polar cusp. The Canadian station is located at a slightly

equatorward latitude, so during daytime hours it is typically under the polar cusp and in

particular magnetospheric conditions it could be even situated at the footprint of closed field

lines (Zhou et al., 2000).

In order to ascertain the possible global character of the pulsation trains simultaneously

observed at the Antarctic stations, for selected time periods we extended the analysis also to a

latitudinal chain of European low latitude stations: Gibilmanna (GIB), L'Aquila (AQU),

Castello Tesino (CTS) and Furstenfeldenbruck (FUR), located at the footprint of closed

geomagnetic field lines (Table 1).

The analysis is based on 1-min values of the horizontal H component. Stations TNB, GIB,

AQU and CTS are run by INGV; data from SBA, DRV, CBB and FUR have been

downloaded from INTERMAGNET web site.

Interplanetary data are from ACE spacecraft; regarding SW measurements, standard SWI

mode ion data, collected every 64 sec, were recorded only from Oct. 31, 0051UT; for

preceding period, only STI mode ion data, collected approximately every 33 min, were

available (Skoug et al., 2004).

EXPERIMENTAL OBSERVATIONS AND DISCUSSION

In Fig. 1 we show interplanetary data on Oct. 29-31, 2003, together with Dst index. It can

be seen that in the time interval of interest there are several periods with strongly southward

IMF: approximately 06-09 UT and 14-03 UT on Oct. 29-30 and 17-01 UT on Oct. 30-31;

these periods correspond to strong geomagnetic storms, with Dst index reaching ~-180 nT, -

360 nT and -400 nT, respectively. There are also periods with strongly northward IMF:

approximately 03-10 UT on Oct. 30 and 01-11 UT on Oct. 31. Plasma data show

exceptionally high SW speed; indeed, it mostly exceeds 1000 km/s and, after the

interplanetary shocks on Oct. 29, 0558 UT and on Oct. 31, 1619 UT (Skoug et al., 2004), it

reaches values around 1700 km/s. The SW density shows several variations, especially during

Oct. 31.

The variation of the geomagnetic field H component at the eight stations in the time

interval of interest is shown in Fig. 2. It is evident that the geomagnetic activity is intense,

especially during the main phase of the geomagnetic storms. It is also evident that the

geomagnetic variations are always very similar between the low latitude stations; in some time

intervals there is a strict similarity between the geomagnetic variations at TNB and at the

closest station SBA and sometimes also between TNB and DRV.

Fig. 3 shows, in the upper panels, the dynamic power spectra at the four high-latitude

stations and at AQU, taken as reference for the low latitude stations; the spectra are computed

from the differenced data (in order to make more evident higher frequency variations) by

means of maximum entropy method, at order 20 of the prediction error filter, over 2-hour

intervals with a time shift of 30 min. The figure also shows, in the lower panels, the dynamic

coherence between couple of stations, computed over 2-hour intervals (with 8 degrees of

freedom) with a time shift of 30 min; in the plots the coherence is shown only where the

spectral power at high latitude exceeds 10
4
nT

2
/Hz.

From the dynamic spectra is evident a strong similarity at the high latitude stations in the time

sequence of the broadband major power enhancements, i.e. for several hours from ~06 UT on

Oct. 29 and from ~17 UT on Oct. 30, corresponding to the onset of geomagnetic storms;

however, in both cases the fluctuations are not coherent, even between the two closest stations

TNB and SBA. At low latitude, only the onset of the first storm, occurring during local

daytime hours, emerges from the power spectrum.

The power spectra show also fluctuations which are observed only at some of the stations,

and then can be considered as local phenomena. For example, we note the power peaks at

discrete frequencies at 20-21 UT on Oct. 29, only at Antarctic stations and not at the northern

high latitude station CBB nor at low latitude, and the broader power enhancement around 18

UT on Oct. 31, very evident only at CBB.

More interestingly, there are also fluctuations which are simultaneously observed over a wide

spatial separation, with high coherence between stations, and then are not just local

phenomena. In this sense, we selected for a more detailed analysis three fluctuation events.

Two of them, around 23 UT on Oct. 29 and 20 UT on Oct. 30, are observed during southward

IMF conditions, in the main phase of strong geomagnetic storms, in correspondence to Dst~-

350 nT and Dst~-300 nT, respectively. The last one, around 11 UT on Oct. 31, is observed

during northward IMF conditions, at the end of the recovery phase of a strong geomagnetic

storm, and can be related to the SW pressure pulses observed from ACE around 1030 UT.

Fig. 4 shows the analysis of the pulsation event around 23 UT on Oct. 29: the filtered (2.5-5

mHz) data (left panels), the power spectra from differenced data (center panels) and the

coherence between selected couple of stations (right panels). This event occurs during

southward IMF conditions, in the main phase of a strong geomagnetic storm (Dst~-350 nT),

and during high SW speed conditions (V~1000 km/s). Stations TNB, SBA and CBB are in the

early magnetic local afternoon sector, DRV in the prenoon sector and the European stations

around magnetic midnight.

In the filtered data emerge at all stations a main wave packet between ~2250-2315 UT (less

clear at CBB) and a strong damping at ~0015 UT. From the spectral and coherence analysis

we can see that a major power peak emerges at all stations around 2.9-3 mHz; the

corresponding fluctuations are highly coherent all over Antarctica and between TNB and low

latitude, while at CBB they are completely decoupled. Observations at CBB are coherent with

other observatories only at frequencies higher than 3.5-4 mHz, but in this frequency range

there is a power peak common only to the high latitude, and not also to the European,

stations.

Also the event around 20 UT on Oct. 30 (Fig. 5) is observed during southward IMF

conditions, in the main phase of a strong geomagnetic storm (Dst~-300 nT), and during

extremely high SW speed conditions (V~1500-1300 km/s); it could be related to the small SW

pressure pulse observed by ACE at ~1930 UT. TNB, SBA and CBB are around magnetic local

noon, DRV in the morning and the European stations in the evening sector.

The filtered data show at all stations a signal intensification just before 20 UT, a phase jump

around 2020 UT, a strong damping around 2045 UT and a smaller signal intensification just

before 21 UT; the peak-to-peak amplitude of these pulsations exceeds 100 nT at the Antarctic

stations and reaches 30-40 nT at the low latitude stations The spectral and coherence analysis

shows that two major power peaks emerge at all stations, around 3.2 and 4.2 mHz. The

corresponding fluctuations are highly coherent between high latitude stations and also

between high and low latitude.

The event around 11 UT on Oct. 31 (Fig. 6) is observed during northward IMF and high SW

speed conditions, at the end of the recovery phase of the preceding strong geomagnetic storm;

it could be related to the strong SW pressure pulse observed by ACE at ~1030 UT. TNB, SBA

and CBB are in the early magnetic morning, DRV in the premidnight sector and the European

stations just after local noon. It is interesting to note that in this case the pulsation amplitude at

low latitude is comparable, even higher, than at high latitude.

At the Antarctic and low latitude stations a main wave packet starts at 1100 UT; in Antarctica

it stops at 1130 UT, while at low latitude it continues, with increased frequency, for further 15

min. The common Antarctic and low latitude waves between 11-12 UT (Fig. 6, solid lines in

the spectral and coherence analysis plots), at ~2.8 mHz, are highly coherent all over Antarctica

but less coherent between TNB and low latitude. The time sequence of pulsations at CBB is

different: a main wave packet appears at 1130-1145 UT; it seems to correspond to the final

part of the pulsation train at low latitude: indeed, between 11-12 UT there is at all northern

hemisphere stations a broader power enhancement with frequency ~3.5 mHz, and the

corresponding coherence between CBB and AQU is very high. In the following hour, between

12-13 UT, a similarity between the wave packets observed at low latitude and CBB emerges;

also the power spectra are very similar, with clear power peaks at ~2.5 and 3.6 mHz, and the

corresponding coherence is very high (Fig. 6, dashed lines in the spectral and coherence

analysis plots); conversely, in this time period fluctuations in Antarctica are completely

decoupled not only from those the northern hemisphere, but also all over Antarctica.

SUMMARY

In this paper we analyze the low frequency geomagnetic field fluctuations detected during

Oct. 29-31, 2003 at four high latitude stations within the polar cap, three in the southern and

one in the northern hemisphere. From a comparison with simultaneous measurements at low

latitude, we address the question of the global character of the observed phenomena. The time

period of interest is characterized by extreme values of the SW speed and by several rotations

of the IMF, which is strongly southward or strongly northward during time intervals of several

hours.

Fluctuations with different characteristics are observed:

- broadband fluctuations at the onset of strong geomagnetic storms, which are observed at all

high latitude stations but are not spatially coherent;

- fluctuations observed only at some of the stations, corresponding to local phenomena;

- fluctuations simultaneously observed at all stations which are spatially coherent, even

between high and low latitude.

We focus on the spatially coherent fluctuations, selecting three wave packets: two occurring

during southward IMF conditions, in the main phase of strong storms, and one during

northward IMF conditions, in the recovery phase of a storm.

For both pulsation events occurring during geomagnetic storms, in correspondence to open

magnetospheric conditions, the high latitude stations are in the dayside sector, between 07

MLT and 16 MLT. We found that these events are characterized by simultaneous wave

packets at discrete frequencies, the same at all the stations; moreover the pulsations are

generally highly coherent not only between Antarctic stations, but also between high and low

latitude. This result is interesting in that for previous pulsation events triggered by magnetic

clouds, we had found that the coherence between TNB and AQU attains high values only

during closed magnetospheric conditions and when TNB is located close to the local

geomagnetic noon (Villante et al., 1998; Lepidi et al., 1999); we had speculated that in a

similar situation TNB, which is usually located in the polar cap, reaches minimum distances

from closed field lines (Carbary and Meng, 1986), and this is the most favorable condition for

a cap station to observe the same phenomena as a low latitude station. Conversely in this

particular case, which is characterized by extreme SW speed values, we found evidence for

oscillations extending to a major portion of the Earth’s magnetosphere, even deep in the polar

cap, during open magnetospheric conditions.

On the other hand, some differences emerge, for these two pulsation events occurring

during open magnetospheric conditions, in the observations at CBB, which is located close to

the northern dayside cusp; in this sense, it is well known that, expecially during open

magnetospheric conditions, intense ionospheric currents are present in the auroral oval, giving

rise to local geomagnetic field variations (Campbell, 1997).

In the event occurring during closed geomagnetic conditions, the fluctuations are still

highly coherent among the Antarctic stations, but not between high latitude opposite

hemispheres. In this case, the observations at European stations are definitely more coherent

with CBB than with Antarctic stations. We note that in this case all the high latitude stations

are far from the local geomagnetic noon, then the Antarctic stations are located deep in the

polar cap. As to CBB, taking into account that the latitude of the auroral oval moves poleward

for closed magnetospheric conditions, it could be located in a different magnetospheric region

with respect to the Antarctic stations, being surely closer to the auroral oval and to closed field

lines. In this sense, our result indicates the occurrence of phenomena extending, in the

northern hemisphere, from auroral to low latitudes and the simultaneous occurrence of

different phenomena deep in the southern polar cap.

Acknowledgements.

The research activity at TNB is supported by Italian PNRA. Authors thank Dr. John Steinberg

(LANL, USA) for providing STI mode ion data from ACE.

REFERENCES

CAMPBELL, W.H. (1997): Introduction to geomagnetic fields, Cambridge Univ. Press, 135-

142.

CARBARY, J.F. and C.I. MENG (1986): Correlation of cusp latitude with Bz and AE (12)

using nearly one years’ data, J. GEOPHYS. RES., 91, 10147-1054.

FARRUGIA, C.J., R.P. LEPPING, L.F. BURLAGA, A. SZABO, D. VASILLIADIS, P.

STAUNING and M.P. FREEMAN (1996): The WIND magnetic cloud of October 18-20,

1995: implications for the magnetosphere, EOS, Transactions of the American Geophysical

Union, 77, 17, S241.

FRANCIA, P., S. LEPIDI, U. VILLANTE, P. DI GIUSEPPE and A.J. LAZARUS (1999):

Geomagnetic response at low latitude to continuous solar wind pressure variations during

northward interplanetary magnetic field, J. Geophys. Res., 104, 19923-19930.

GOPALSWAMY, N., L. BARBIERI, G. LU, S.P. PLUNKETT and R.M. SKOUG (2005):

Introduction to the special event: Violent Sun-Earth connection events of October-November

2003, Geophys. Res. Lett., 32, L03S01, doi: 10.1029/2005GL022348.

HARROLD B.G. and J.C. SAMSON (1992): Standing ULF modes of the magnetosphere: a

theory, Geophys. Res. Lett., 19, 1811-1814.

LANZEROTTI L.J,. A. SHONO, H. FUKUNISHI and C.G. MACLENNAN (1999): Long-

period hydromagnetic waves at very high geomagnetic latitudes, J. Geophys. Res., 104,

28423-28435.

LEPIDI S., P. FRANCIA, U. VILLANTE, A. MELONI, A.J. LAZARUS and R.P. LEPPING

(1999): The Earth’s passage of the April 11, 1997 coronal ejecta: geomagnetic field

fluctuations at high and low latitude during northward interplanetary magnetic field

conditions, Ann. Geophysicae, 17, 1245-1250.

LEPPING R.P., L.F. BURLAGA, K.W. OGILVIE, B.T. TSURUTANI and A.J. LAZARUS

(1991): The interaction of a very large interplanetary magnetic cloud with the magnetosphere

and with cosmic rays, J. Geophys. Res., 96, 9425-9438.

OLSON J.V. (1986): ULF signatures of the polar cusp, J. Geophys. Res., 91, 10055-10062.

SAMSON J. C., B.G. HARROLD, J.M. RUOHONIEMI, R.A. GREENWALD AND A.D.M.

WALKER (1992): Field line resonances associated with MHD waveguides in the

magnetosphere, Gewophys. Res. Lett., 19, 441-444.

SKOUG, R.M., J.T.GOSLING, J.T. STEINBERG, D.J.MCCOMAS, C.W.SMITH, N,F,

NESS, Q.HU and L.F. BURLAGA (2004): Extremely high speed solar wind: 29-30 October

http://cdsads.u-strasbg.fr/cgi-bin/author_form?author=Samson,+J&fullauthor=Samson,%20J.%20C.&charset=ISO-8859-1&db_key=AST
http://cdsads.u-strasbg.fr/cgi-bin/author_form?author=Harrold,+B&fullauthor=Harrold,%20B.%20G.&charset=ISO-8859-1&db_key=AST
http://cdsads.u-strasbg.fr/cgi-bin/author_form?author=Ruohoniemi,+J&fullauthor=Ruohoniemi,%20J.%20M.&charset=ISO-8859-1&db_key=AST
http://cdsads.u-strasbg.fr/cgi-bin/author_form?author=Greenwald,+R&fullauthor=Greenwald,%20R.%20A.&charset=ISO-8859-1&db_key=AST
http://cdsads.u-strasbg.fr/cgi-bin/author_form?author=Walker,+A&fullauthor=Walker,%20A.%20D.%20M.&charset=ISO-8859-1&db_key=AST

2003, J. Geophys. Res., 109, A09102, doi: 10.1029/2004JA010494.

VILLANTE U., S. LEPIDI, P. FRANCIA, A. MELONI and P. PALANGIO (1997): Long

period geomagnetic field fluctuations at Terra Nova Bay (Antarctica), Geophys. Res. Lett.,

24, 1443-1446.

VILLANTE, U., P. FRANCIA, S. LEPIDI, M. DE LAURETIS, E. PIETROPAOLO, L.

CAFARELLA, A. MELONI, A.J. LAZARUS, R.P. LEPPING and F. MARIANI (1998):

Geomagnetic field variations at low and high latitude during the January 10-11, 1997

magnetic cloud, Geophys. Res. Lett., 25, 2593-2596.

WALKER, A.D.M., J.M.RUOHONIEMI, K.B. BAKER, R.A. GREENWALD and J.C.

SAMSON (1992): Spatial and temporal behaviour of ULF pulsations observed by the Goose

Bay HF radar, J. Geophys. Res., 97, 12187-12202.

YAGOVA, N.V., L.J. LANZEROTTI, U. VILLANTE, V.A. PILIPENKO, S. LEPIDI, P.

FRANCIA. V.O. PAPITASHVILI and A.S. RODGER (2002): ULF Pc5-6 magnetic activity

in the polar cap as observed along a geomagnetic meridian in Antarctica, J. Geophys. Res.,

107, SMP22, doi: 10.1029/2001JA900143.

YAGOVA, N.V., V.A. PILIPENKO, L.J. LANZEROTTI, M.J. ENGEBRETSON, A.S.

RODGER, S. LEPIDI and V.O. PAPITASHVILI (2004): Two-dimensional structure of

long-period pulsations at polar latitudes in Antarctica, J. Geophys. Res, 109, A03222, doi:

10.1029/2003JA010166

ZIESOLLECK C.W.S and F.H. CHAMALAUN (1993): A two-dimensional array study of

low-latitude Pc 5 geomagnetic pulsations, J. Geophys. Res., 98, 13703-13713.

ZIESOLLECK C.W.S. and D.R. MCDIARMID (1995): Statistical survey of auroral latitude

Pc 5 spectral and polarization characteristics, J. Geophys. Res., 100, 19299-19312.

ZHOU X.-W., C.T. RUSSELL, G. LE, S.A. FUSELIER, and J.D. SCUDDER (2000): Solar

wind control of the polar cusp at high altitude , J. Geophys. Res., 105, 245-251.

http://dx.doi.org/10.1029/2003JA010166

Station Geographic Coord. Corr. Geom. Coord. MLT NN (UT)

TNB 74.7S 164.1E 80.0S 307.7E 20:11

SBA 77.8S 166.8E 80.0S 326.5E 19:01

DRV 66.7S 140.0E 80.4S 235.7E 00:55

CBB 69.2N 255.0E 77.2N 309.6E 19:54

FUR 48.2N 11.3E 43.4N 86.9E 10:28

CTS 46.0N 11.7E 40.8N 86.7E 10:28

AQU 42.4N 13.3E 36.3N 87.4E 10:24

GIB 37.9N 14.0E 30.6N 87.3E 10:24

Table 1. Geographic coordinates, IGRF2003 corrected geomagnetic coordinates and time in UT of the
magnetic local noon for the stations.

FIGURE CAPTIONS

Figure 1. ACE data (from top: IMF strength and north-south component, SW density and

speed) and Dst index.

Figure 2. Variations of the geomagnetic field H component at the eight stations.

Figure 3. Upper panels: dynamic power spectra from differenced data in logarithmic scale at

the four high-latitude stations and at AQU, taken as reference for the low latitude stations

(note the different color scale at AQU). Lower panels: dynamic coherence between TNB and

each of the other four stations and between CBB and AQU; the coherence is shown only when

the spectral power at high latitude exceeds 10
4
nT

2
/Hz.

Figure 4. Analysis of the pulsation event around 23 UT on Oct. 29. Upper panels show the

observations at high latitude, lower panels at low latitude. From left: filtered (2.5-5 mHz)

data, power spectra from differenced data (2230-0030 UT) and coherence (2230-0030 UT).

The coherence in the upper panel is between TNB and the other high latitude stations; in the

lower panel between TNB-AQU and CBB-AQU.

Figure 5. The same as Fig. 4, for the pulsation train observed around 20 UT on Oct. 30. Power

spectra and coherence are computed between 1930-2130 UT.

Figure 6. The same as Fig. 4, for the pulsation train observed around 11 UT on Oct. 31. Power

spectra and coherence are computed between 11-12 UT (solid lines) and between 12-13 UT

(dashed lines); 12-13 UT power spectra have been down shifted by two decades.

Figure 1

Figure 2.

Figure 3.

Figure 4.

Figure 5.

Figure 6.