Spatial distribution of Pi2 micropulsations

Spatial distribution of Pi2 micropulsations

JournalofAtmospheric acidTerrestrialPhysics, 1969,Vol.31, pp. 1301to 1309.PergamonPresa. PrintedinKorthemIrelmd Spatial distribution of Pi2 micropuls...

651KB Sizes 3 Downloads 13 Views

JournalofAtmospheric acidTerrestrialPhysics, 1969,Vol.31, pp. 1301to 1309.PergamonPresa. PrintedinKorthemIrelmd

Spatial distribution of Pi2 micropulsations H. N. JHA* and W. G. V. ROSSER Department

of Physics, Exeter University,

(Received 3 April


Devon, England

in r&-sed form 9 May 1969)


of Pi2 micropulsations show that one can obtain very good correlations over distances of about 1000 km in the North-South direction between geomagnetic latitudes of 62’30’N and 54’20’N, and yet obtain much poorer correlations over distances of about 140 km. These differences over distances of about 140 km, which are particularly marked in the vertical component, may be associated with coastline or geological effects.

1. INTRODUCTION TEE Pi2 (or irregular) type of geomagnetic micropulsations is an impulsive type disturbance of period in the range 40-150 set and lasting about 5-20 min. Typical examples are shown in Figs. 2, 4 and 5. Pi2 micropulsations are generally observed between 18.00 and 02.00 L.T., as illustrated in Fig. 1. Pi2 micropulsations are generally associated with the occurrence of magnetic bays in the aurora1 zone, which are themselves generally associated with the break up phase of aurorae. It has been suggested, for example by ATKINSON (1966) and ROSTOKER (1966), that aurora1 break up may be associated with the interchange of geomagnetic field lines between the closed region of the magnetosphere and the open tail. This movement of magnetic field lines may induce eigen oscillations in the geomagnetic field lines, possibly leading directly to Pi2 type geomagnetic micropulsations. When the interchange of field lines takes place, magnetic field energy may be converted into charged particle energy, leading to charged particle precipitation in the aurora1 zone. CAMPBELL (1961) and CAMPBELLand MATSUSHITA (1962) observed increases in fmin and f,E, and the sudden enhancement of bremsstrahlung X-rays associated with irregular pulsations in the aurora1 zone, suggesting that beams of high energy NISRIDA (1964) suggested that such a electrons were injected into the ionosphere. precipitstion of electrons of energy ~10 keV could lead to hydromagnetic instabilities in the magnetospheric plasma, when the electron beam passed through the magnetospheric plasma, resulting in the generation of irregular geomagnetic micropulsations in the magnetosphere, in the region above the ionosphere. The present work is concerned with the properties of Pi2 events at sub-aurora1 latitudes, where the geomagnetic field lines are not connected directly to the geomagnetic tail, and where there is no significant precipitation of aurora1 particles. If the primary causes of Pi2 micropulsations are associated with magnetic field lines anchored in the aurora1 zone, then Pi2 micropulsations must be propagated to lower latitudes, possibly ss hydromagnetic waves crossing the geomagnetic field lines or as leakage currents in the ionosphere, associated with the generation of * Present address: 1


Singh College, University 1301

of Bihar, Mazaffarpur,

Bihar, India.

ionospheric currents in the amoral zone. HERROX (1966) has observed finite propagation times for Pi2 micropulsations. These results were interpreted by JACOBS, ROSTOKER and WATANABE (1965) and ROSTOKER (I 965) as the propagation of electromagnetic waves in the ionosphere, across the field lines with a phase where 0 is the conductivity and T the period. For Pi2 microvelocity ~(aT)-r’~ pulsations propagating in the E-region of the ionosphere, this phase velocity should Such electromagnetic waves should be linearly polarized. be ~40 km set-l. (ROSTOKER, 1967a). If the propagation of Pi2 micropulsations to lower latitudes takes place through the ionosphere, then disturbances and fluctuations in the ionosphere may affect the properties of Pi2 micropulsations observed at subauroral latitudes. It will be shown later that the local geology and coastline may also affect the properties of Pi2 events observed at sub-aurora1 latitudes. Of particular interest is the polarization of Pi2 events. For a plasma in which the magnetic energy density is much greater than the kinetic energ_y density, there are two main types of hydromagnetic waves, namely the pure Alfven wave (or slow mode) in which the energy is propagated along the magnetic field lines, and the fast-mode (or modified Alfven, or isotoropic or magnetosonic mode) in which the energy can propagate in all directions. For propagation in the magnetic field direction, the fast mode has right handed (clockwise) circular polarization, whereas the Alfven mode has left handed (anticlockwise) circular polarization (OBAYASHI, 1965). 2. EXPERIMENTAL, ARRANUEN~XT A two component fluxgate magnetometer was operated at Sidmouth, which is on the South Coast of England and has geographic co-ordinates 50°40’N, 3”15’W and geomagnetic co-ordinates 54”20’N and 80’6%. (In addition results were available at Sidmouth from a three component induction magnetometer and a two component earth current detector.) Similar fluxgates were operated by S.R.D.E. at Christchurch, which is also on the South Coast of England, geographic co-ordinates 50”44’N, 1”48’W, geomagnetic co-ordinates 54”10’N, 8 I “40’E, and subsequently by R.S.R.S. at Lerwick in the Shetland Islands, geographic coordinates 6O”lO’N, l”lO’W, geomagnetic co-ordinates 62’30’N, 88”36’E. The chart speeds were generally 12 in./hr. Pi2 events were selected for detailed investigation, since they are easily identifiable and normally appear in the evenings when conditions are otherwise fairly quiet. The selection of events for detailed investigation is not entirely random. but biased somewhat towards events occurring at times of quiet, background. 3. COMPARISON 0r Pi2 MICR~PUL~ATI~NS ( 1) Sidmouth and Lerwick The occurrence

frequencies of Pi2 micropulsations in August and September and Sidmouth are shown in Figs. I(a) and I(b) respectively. There is a close similarity in the occurrence frequencies of Pi2 micropulsations at, the two stations. The ratio of the amplitudes of Pi2 micropulsations at Lerwick 1965 at Lerwick


Spatial distribution of Pi2 micropulsations 1

I Sidmouth



- ‘35

- 30 - 25

- 20


- I5 - 10









Time ,G.hl.T. Fig.



,.rr,.,,,:,,,.t?l 04










Time, G.M.T.

1. Occurrence frequency of Pi2 micropulsations in August 1965 (a)at Lerwick (b) at Sidmouth.

and September

and Sidmoufh is generally about unity, showing that there are generally no large differences in the amplitudes of Pi2 micropulsations over a distance ~1000 km in the North-South direction between geomagnetic latitudes of 62”30’N and 54’20’N. The maximum values of the cross correlation coefficients between the NorthSouth and East-West components of Pi2 micropulsations at Lerwick, between the North-South and East-West components at Sidmouth, between the North-South components at Lerwick and Sidmouth and between the East-West components at Lerwick and Sidmouth are shown in Table 1 for 23 Pi2 micropulsations. It can be seen that the cross correlation coefficients between Lerwick and Sidmouth was 20.7 for both components in 15 cases, showing that the correlation of Pi2 micropulsations is generally high. A typical example of a high correlation Pi2 micropulsation is shown in Fig. 2. This is event 10 of Table 1. All four cross correlation coefficients were >0*9 in this example. At Lerwick the polarization of this event in the horizontal plane was circular, the vector rotating in the anti-clockwise direction as shown in Fig. 3(a). The North-South and vertical components at Lerwick were in phase as shown in Fig. 3(b). At Sidmouth the polarization of the same event, in the horizontal plane, was approximately linear as shown in Fig. 3(c). The power spectra at Lerwick and Sidmouth were very similar and consisted mainly of one component of frequency -96 sec. The Pi2 micropulsations showing high correlation coefficients (20.7 for both horizontal components) between Lerwick and Sidmouth showed the following general features. The polarization in the horizontal plane was fairly regular, (circular, elliptic or plane) at both stations, that is the polarization did not change its direction of rotation during the event. The corresponding components at The maximum Lerwick and Sidmouth generally had the same dominant frequency. cross-correlation coefficient between the North-South and East-West components at the same station was also generally high (20.7) for these events, as can be seen

W. K. JHA and

1304 Table

Event _._._. 1. 2. 3. 4. 5. ii. ”I

x. !I. LO. 1I. 12.

13. 15. 15. Iti. ii. is. IM. 20. 21. 22. 23.

1. Coefficient

of Pi2 between ~~xirnllrn

Time of occu~%ncG (G.M.T.)

1.)&c .-_--__23.8.6.5 23.8.65 23.8.65 24.8.65 35.8.65 45.8.65 26.8.65 26.8.65 37.8.65 30.X.65 30.8.65 16.9.65 i 7.9.65 19.9.65 28 .L.I Y 6i “8.9.65 f 9.2.66 _C.Y. *9 * 66 33.2.66 X2.66 -05.?.66 L “7 .& ‘?.66

of correlstion

FI’. (1. \:. ROSREH

LerWi& I?:--H/E-V: .--..--.---.-.---__

20.03-20.16 20.4G20.62 23.45-24.00 22.00.-22.07 00.~3-00.1~ 00.04--00. iH 01.27-01.37 21.10--21.24 22.05 -22.19 19.59--20.04 23.22 -23.28 23.55-00.08 23.20.-23.30 05.26~-05.31 15.19.--15.26 19.25-19 . * .32 I9.99~-19.57 19.08 -19.30 %2.50-~ 23.00 19.52--20.03 22.40~_22.47 01.01 -01.06 00.53 01 .0”_

Sidmouttl N-S/E-U __


and Sidtnoutb

c(~~~l~tioI~ coeff’ieiettfs _..-...-___ fA‘?P. (E--W) Ler. (N-K) & Sid. (N--S) & $itl. (E.-W)


~_..~_. __._..--

0.85 0.95 0.52 0.85 0.x O-73 0.76 (),!):I Ii.9 0.95 lB.9-c O-85 0.9 0.81 0.86 0.86 0.82 0.82 0.5 (I,.? O-7 ().!):I 0.x


South - North

West - East

Fig. 2. Example between 19.59

of‘ a f’i.2 event at Leuwkk and 20.04 L.T. on 27 .411&&


Vertical I Downwards




Spat&l distribution of Pi2 micropulsdions




(upwards) (a)



-d Fig.

3. Polarization



of the Pi2 event in the time interval 27 August 1965.




from Table 1. The maximum correlation between Lerwick and Sidmouth was generally displaced from zero time shift. This time shift was generally different for the two horizontal components and, the difference between the time shift for the two components varied from event to event, due possibly to different propagation times from the source of Pi2 micropulsations for different events and different changes in polarization for different events, possibly due to disturbances along the propagation paths. The poorer correlation events, (that is one component ~0.7) showed the following general features. The polarization in the horizontal plane tended to be irregular, particularly at Sidmouth, that is tended to change its direction of rotation during the event. There was also generally poor correlation between the North-South and East-West components at the same station. Power spectra analyses showed that for these events corresponding components at Lerwick and Sidmouth had somewhat different power spectra, though there was no systematic During these events K,-index was variation of power spectra with latitude.

generally 4 or 5. Our results confirmed those of ROSTOKER (lYtii), that the number of peaks in the power spectra of the Pi2 events increased with K,,. The polarization of 20 cases of Pi2 micropulsations in the horizontal plane was determined by plotting the North--South component as ordinate against the East-West component as abscissa. In estimating the degree of polarization of the horizontal vector, the criterion adopted was that cases with minor axis to major axis ratios of O-8 and above were considered as circular, with axes ratios between 0.2 and 0.8 as elliptical and with axes ratios less than 0.2 as plane polarized. At Lerwick, 9 of the 20 cases had circular polarization in the horizontal plane, seven had elliptical polarization, three were plane polarized and one was irregular. .\I1 the circularly and elliptically polarized Pi2 micropulsations at Lerwirk showed anticlockwise rotation. On the other hand, at Sidmouth, thirteen out. of the twent!y cases were plane polarized, four elliptically polarized, one circularly polarized and two were irregular. All elliptically and circularly polarized Pi2 micropulsat,ions at Sidmouth rotated predominantly in the anticlockwise direction. The sense of rotation of the disturbance vector at Lerwiok is consistent with the pure Alfv& (or slow) mode, in which the energy is propagated along the magnetic field lines. In the Northern Hemisphere, this mode is circularly polarized with anticlockwise rotation for propagation parallel to the field lines. In the Southern Hemisphere this mode should have clockwise rotation, for propagation anti-parallel to the magnetic field. The clockwise rotation in the Southern Hemisphere for Pi2 micropulsations was confirmed by CHRISTOFFEL and LINPORU (1966). If three dimensional polarization plots are made. the plane of the circular polarization can be determined. At Lerwick the normal to this plane was generally not, For example for the case illustrated in exactly in the magnetic field direction. Figs. 2 and 3, at Lerwick the normal to the plane of circular polarization points at an angle of about 60” to the horizontal but pointing in a southerly direction. whereas at Lerwick the geomagnetic dip angle is 75”, but pointing in the northerl? direction. Too much significance should not. be attached to t,hese discreganoic~s. since geological anomalies or coastal effects can have a marked cTffec%ton the 7%~ more linear type polarization at vertical component (WEAVER. 1963). Sidmouth is consistent with either hydromagnetic wave propagation across the field lines. or as propagation from the aurora1 zones as electromagnetic waves in the ionosphcrcb. (2) Skimouth and Christchurch (England) Both Sidmouth and Christchurch are on tlrtl South Coast, of England about 140 km apart. The comparison between Sidmouth and Christchurch was not carried out on the same events as discussed in Tablo 1. Only one tluxgatc. measuring the East-West magnetic variations, was operating at Sidmouth for most of this period of comparison with Christchurch. The results do show interesting differences between Christchurch, Sidmouth and Lerwick. The maximum vslues of the cross correlation coefficients between Sidmouth and Christchurch for 15 Pi2 micropulsations during March 1963; are shown in Table 2. The correlation coefficient, between the East-West components at, the two stations was generally high, being

Spatial Table

Event 1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14. 15.


of Pi2 micropulsations


2. Coefficient of correlation between component fields of Pi2 at Christchurch and the E-W component at Sidmouth

Date 5.3.63 5.3.63 5.3.63 6.3.63 6.3.63 7.3.63 7.3.63 7.3.63 8.3.63 8.3.63 8.3.63 8.3.63 8.3.63 9.3.63 11.3.63

Time of occurrence (G.M.T.) 20.24-20.51 20.24-21.09 20.55-21.09 17.52-18.09 18.44-19.00 03.15-03.25 20.42-20.57 21.55-22.23 18.45-18.55 20.00-20.21 20.20-20.35 20.35-20.51 20.00-20.51 21.15-22.00 22.42-22.51



Christchurch N-S and E-W 0.53 0.6 0.7 0.7 0.85 0.74 0.2 0.4 0.4 0.65 0.72 0.52 0.63 0.44 0.7


Christchurch & Sidmouth

(E-W) (E-W)

0.9 0.95 0.96 0.8 0.95 0.81 0.9 0.9 0.66 0.8 0.75 0.8 0.62 0.71 0.8

K, 1+ 1+ 1+ 2O 2O 2O 2+ 5O 4+ 4+ 4+ 4+ 4+ 4+ 3O

Vertical ti

East - West 0 Fig.

4. Example

of a Pi2 micropulsation at Sidmouth. The vertical is comparable to the East-West component.


H. 2% JHA and



C:. 1’. ROSSE~~





5. Example

at Christchurch of Pi2 micropulsation activity in the vertkal componwt.


I s~l~mw~gno

coefficient ‘>O-7 for all but two cases. However, the maximum cross-correlation between the North-South and East-West components at Christchurch was generally (O-7, in contrast to the results at Sidmouth and Lerwick shown in Table 1. The power spectra of the North-South and East-West components were generally different to each other at, Christchurch, again, in contrast to Lerwick and The most significant difference between Christchurch and Lerwick Sidmouth. and Sidmouth was in the magnitude of the vertical component. At Lerwick and Sidmouth the variations in the vertical direction were generally a substantial proportion of the North-South and East-West variations as shown in Figs. 1 and 4 whereas at Christchurch there appeared to be no vertical variations for either Pi2 or PC type activities as illustrated in Fig. 5. The marked differences between the vertica.1 components at Sidmouth and Christchurch may be due either to rrustal geological anomalies or to coastal effects.

Spatial distribution of Pi2 micropulsations


(3) Sidmouth and Green Hailey Fast charts were run on the East-West component at Sidmouth and Green 51”45’N, 0”50’W, geomagnetic co-ordinates Hailey (geographic co-ordinates Times were marked automatically 54”5O’N, 83’1O’E) at speeds of 1 in./min. Observations of the trace were taken at 2.4 set directly on to the waveforms. intervals. The two stations are on an East-West line about 300 km apart. The frequency responses and time delay characteristics of the equipment at the two stations were very similar. Analysis of the cross-correlation coefficients of Pi2 events showed that maximum cross correlation was generally obtained when Green Hailey lead Sidmouth by about 7-12 set suggesting that the Pi2 micropulsations might be propagated from East to West. However, since the measurements were on one component only, and in view of the changes in polarization that may be associated with local geological and coastal effects, it was felt that no definite conclusions could be drawn from the results with one component only. 4. SUMMARY The investigations of Pi2 micropulsations show that one can have very good correlations over distances about 1000 km in the North-South direction between geomagnetic latitudes of 62’3O’N and 54”2O’N, and yet get much poorer correlations over a distance about 140 km. These differences over distances about 140 km are particularly marked in the vertical component. Acknowledgements-The work described here was carried out at the Norman Lockyer Observatory, Sidmouth as part of the research programme of the Geophysics Section of the department of Physics, and we are very grateful to the Council of the Corporation of the Observatory for allowing us to use the Observatory as & field station. We would like to thank Dr. K. WEEKES for his advice and encouragement throughout, and Dr. A. NICHOL for developing the computer progrsmmes. We would like to thank Mr. P. J. STEVENS of S.R.D.E. for supplying the Christchurch data, and the Director, R.S.R.S., for supplying the Lerwick and Green Hailey data. REFERENCES ATKINSON G. CAMPBELL W. H. CAMPBELL W. H. and MATSUSHITA S. CHRISTOFFEL D. A. and LINFORD J. G. HERRON T. J. JACOBS J. A., ROSTOKER G. and WATANABE T. NISHIDA A. OBAYASHI T. ROSTOKER G. ROSTOKER G. ROSTOKER G. ROSTOKER G. WEAVER J. T.

1966 1961 1962

J. geophys. Res. 71,6167. J. geop%mJs. Rea. 66, 3699. J. geophys. Res. 67, 555.

1966 1966 1965

J. geophys. Re.s. 71, 891. J. geophya. Res. 71, 871. Nature, Lord. 205, 61.

1964 1965 1965

J. geophys. Res. 69, 947. J. geophya. Res. 70, 1069. J. geophys. Ree. 70, 4388. J. geophys. Rerr. 71, 79. Can. J. Phys. 45, 1319. J. geophys. Rea. 72, 2032. Can. J. Phys. 41, 484.


196’7a. 1967b 1963