doi: 10.22564/rbgf.v3i2.

1024
ftevista Brasileira de Geof ísica;1985, Vol. 3,103-106

EFFECTS OF IONOSPHERIC BUBBLES ON F.REGION

PLASMA DRIFTS AT THE EOUATOR

ENIVALDO BONELLI

Grupo de Geof ísica

Dept9 de Física Teórica e Experimental - Centro de Ciências Exatas
Universidade Federal do Rio Grande do Norte
'
59.000 - NATAL - RN - Brasil

A model for the ionospheric electrodynamics that takes into account the affects of
both E-region and F-region dynamos has been developed in this research. The effect
of ionospheric plasma bubbles (plasma depletions) on the equatorial F-region zonal
plasma drifts have been estudied in this work. Some of the nightime results are: 1. The
plasma drift inside the bubble lags behind that of the background plasma; 2.f he presence
of a bubble might affect measurements of plasma drifts based on ionospheric irregula-
rities;3. The field aligned currents inside a bubble are reduced in magnitude, and may
even reverse direction for large plasma depletions (depletions larger than 80%).

Desenvolveu-se um modelo para a eletrodinámica da ionosfera, que considera os efeitos

da região E e da região F. Estudou-se o efeito das bolhas no plasma ionosférico (deple
cões no plasma) nas derivas zonais do plasma da região F equatorial. Alguns dos resulta-
dos correspondentes ao penodo noturno são: 1) O plasma dentro da bolha tem uma ve-
locidade de deriva menor que a do plasma de fundo; 2) a presença da bolha pode afetar
as medicões de derivas de plasma que são baseadas em irregularidades ionosféricas; 3) as
correntes paralelas ao campo magnético, dentro da bolha, sao reduzidas em magnitude,
poderrdo até mudar de sentido para grandes depleções no plasma (depleções maiores que
80%).

INTRODUCTION from 90 to 140 km, the ionospheric E-region. where the

transverse (to the magnetic field) conductivities are larger
The sun, through heating, and the moon, through than at any other height. The electrical conductivity along
gravitation, cause the earth's upper atmosphere to develop the magnetic field lines, though, is always large, so that
pressure gradients. These gradients, in turn, cause air mo-
the E-region polarization electric fields are "mapped"
tion with periods of solar and lunar days, and their harmo- into the F-region around 400 km. There, these f ields cause
nics. These winds are called "tidal" winds, by analogy the plasma to exhibt an E X B drift. These f ields are not
with the ocean tides. and flow mostly horizontally at shorted out due to the fact that the F-region perpendicular
heights above 60 km.
conductivities are much smaller than the E-region ones.
Tidal winds blow the ionospheric ionization along
and, if there is a component of motion perpendicular to From the work of Rishbeth (1971a, b) , we now
the geomagnetic field, polarization electric fields may know that the electric fields in the F-region are not due
appear. This process is usually referred to as the "ionosphe-
only to the E-region dynamo: they have a contribution
ric dynamo" and is the basis of this work. from the F-region dynamo, driven by thermospheric
The conduct¡vity in the ionosphere depends stron- winds, also of tidal nature. During the day, the F-dynamo
gly on the neutral and charged particle densities and on the electric f ield is shorted, in the E-region, due to the high
direction of the magnetic field. For this reason, it changes conductivity there. At night, however, the E-region con-
w¡th he¡ght, latitude, and longitude. These spatial grad¡ents ductivities are much smaller (Rishbeth & Garriott, 1969),
are important for the dynamo theory because by preventing
and the F-region can generate electric fields that are larger
a uniform flow of electric currents, driven by the winds, or comparable to the ones due to the E-dynamo.
they cause polarization electric fields to build up. These Models based on the above description of the dynamo
fields, in turn, cause the initial currents to change, in a process have been carried out with success (Heelis et al.,
self-consistent way. The current network so established i974; Bonelli et al,, 1984) . Now we consider the effects
causes the fluctuations of the magnetic f ield, as observed of ionospheric bubbles on the dynamo electric fields.
on the ground. lonospheric bubbles are plasma depletions (enhancements)
Until a decade ago (Rishbeth,197 1a, b), the dynamo originated due to Rayleigh-Taylor instabilities at the
process was supposed to take place only in the range bottom of the F-region (around 300 km). The east-west
t04 Effects of ianosphet'ic bubbles

typical dimension of a bubble is of order of hundreds effects on the vertical electric fields, which depend mostly
of kilometers (e.g., Kelley & McClure, 1981)" From the on the vertical gradients in the plasma density.
above discussion on dynamo theory, we see that extra
density gradients may alter the electric f ields. lt is the
purpose of this work to show that this effect actually RESULTS AND CONCLUSION
occu rs.
We calculated winds, currents and electric fields
for both the no-bubble case and for the case irrcludirrg a
THE MODEL bubble. ln t¡oth cases the ionospheric and solar parameters
were the same so that the difference between the results
'The present would be due only to the inclusion of the plasma depletions
model is a modif ication of the model of
llellis et al, 11974],. lt considers a thin (negligible width) E- in the calcr.rlations.
region and a realistic F-region, where we take into account
lreigirt profiles o'f the wind field, of tlre electron and neu- II
trals density. The modcl atmosplrere used is that of Jacchia
.í"
Ë wlTH BUBBLE
(1977], , while the experinrental parameters are from Heelis
et al. (1974).
--*
-. NO BUEBLE
A brief description of the model is as follows: in tlre h- 2.
r'\
I
t¡J /\
F-region, therrnospheric winds generate polarization electric /\
I'ields, perpenclicular to the geomagnetic field. Because : ,t
tlre magnetic f ield lines can be considered as equipotential t
þ
(n t
lines (Farley, 1960) , these F-region generated electric f ielcls t
can be partially shorted by the Ë-region below, through
zrd
Õ
field-aliç.¡ned currents. ln the E-region, the model considers l*,
a height integrated conductivity tensor and an "effective" zIJ
wirrd, which f lows in a tl-rin layer. ln this region, atr electros- E"
(Y
tatic potential, which represents the polarization electric :)
field, is calcr"¡lated taking into account the f low of current
(.) 4 68 to l?. 14

into (and out of) the upper boundary of the layer. Frorrr LATI UDE (des. )
this potential, we calculate the Ë-region electric fields.
l"hese fields, in turn, rnap into the F-region affecting the Figura 1 -' Field aligne<J current density at the 2000 L"['meridian,
plasrna rnotion and consequently, changing the f ield-aligned as a function of lal¡lude at the top of the E-region.
currents. For this reason. the model is solved self-consistetr-
tly. For details about the numerical computations, the ln Fig. 1 we show the f ield aligned current density
reader is referreci to the paper of Heelis et al. (1974). flowing out of the E-region for 2000 local time, as a func-
Now we preserlt a simple model for an iclnospheric tion of the latitude where the field lines enter the E-region.
bubble. ln the vertical direction, the bubble density is 'fhe solid lirre represents results for a bubble at 350 km,
given by with a rnaximun depletion of 80%. The dashed line repre-
sents currents for the nobubble case. The effect of the
nB: N{ I ¿ exp [-0.5 (R-HA\' lZg'1¡ field aligned depletion is to reduce the f ield aligned current
inside a given flux tube, being able to even reverse it inside
where Zg is a vertical characteristic size of the bubble, the tr-rbe where the depletion is maximum (flux tube at
Hg is the height of its center, R is the height where we Bo in Fig. 1).
want to calculate the plasma density. Here d is the bubble Fig. 2 shows the eastward (west to east) plasma drifts
depletion, given by d: 1-nBg/N, where nB0 is the as a function of height at 2000 local time. As for Fig. 1.
plasma density at the center of the bubble, and N is the the sr¡lid line corresponds to the case including a depletion,
background plasma density. ln the approximation used and the dashed line corresponds to the no-bubble case.
in this work, the motion of the bubble is not taken into lnside the bubble (arouncl 350 km), the plasma drifts more
account, this is because our calculation is done for a single slowly than the neighboring plasma. ln the figure we see
universal time. The results for the different local times that the center of the bubble actually drifts westwardsl
represent the point of view of observers at different longi- This is not surprising, since around this local time the
tudes. E-region dynamo contributes with an upward electric
Our f inite difference solution has resolution of 10 f ield. Since the F-region does not drive the bubble strongly,
km in altitude and of 500 km in longitude. The later is the E-region may dominate drifts in the case of strong
too large when we compare it to the size of a typical plasma depletions. The reversal of the drifts at around
bubble, which is of 100 km (see, for example, Kelley 600 km is not related to the problem of bubt¡les and
& McClure, 1981). For this reason we will only consider will be discussed elsewhere.
 Ënivaldo tlonelli 105

700
t
600
600
E
:
UJ

I
E 500 o
f
L
trJ
c)
) )
t--

þ 400 I
I
Ë
_l
200
300 -50 0 50 I50
WITH
-- BUBBLE EASTW\RD ION DRIFT (m/s)
NO
200 BUBSLE Figura 3 - Eastward ion (plasma) drift at the equator, as a
function of height, for 19O0 LT (dotted lines),2000
LT (solid lines). and 2100 LT (dashed lines). For
-50 0 50 ræ each line style, the heavier one indicates results
EASTWARD ION DRIFT (m/s )
including a bubble.

Figura 2 *- [:astw¿rrd ion (plasma) drift at the equator, as a

functir:n of height, at 2000 LT.
This cliscussion would not be complete if we did
not merìtion the effect of plasma enhancements and of a
Kudeki et al, (1981) present some few drift measure- rnixture of plasnìa depletions and plasma enhancements,
rnents where the reversal clf the zonal drift occurs at higher as this actually is the case irr the ionosphere [e. 9., Yeh
heights for later tinres, between 1900 and 2000 hours, et al., 19791. A ¡rlasma enhancement would have the
local time. lrr Fig¡. l'l we see that the presence of a bubble opposite behavior of a plasma depletion: it would drift
coulcl explain this rest¡lt. Without the bubble, the reversal faster than the backgrouncl plasma. Furtherrrlore, a mixture
of the 1900 hours, local time, profi!e (light dotted line) of depletiorrs and enhancements creates a complicate
occr¡rs above the reversal of the 2000 LT profile (light pattern of vertical electric fields and consequently of
solid curve). After the inclusion c¡f a bubble in the calcula' eastward drifts. For this reason we should expect that
1ion, however, the situation is changed, and the 2000 LT drift measurements would show clscillations arotlnd a mean
reversal (heavy solid line) occurs at a higher height than valLre, of walelengths equal to the bubbles east-west charac-
for the 1900 LT profile (heavy dotted line). This suggests teristic sizes.
that Kudeki et al. (1981) might be observing the bottom-
side of a very large plasma depletion.
Another interesting result is that although the drifts ACKNOWLEDMENTS
and currents are reduced around the bubble' they are
enhanced outside it, as compared to the no-bubble results. Ithank Cornell University Space Plasma Physics
These enhancements occur several bubble-lengths away Group for letting me use their Harris H-800 computer to do
from the bubble. This suggests that, in experiments, we the calculations of the present work. Thanks are also due to
should look for long range effects of plasma depletions' Mr. José Pereira de Lima, who made the drawings.

REFERENCES

BONELLI, E., LARSEN, M, F., FEJER, B. G., & FARLEY, D' T. - JACCHIA, L. G. - 1977 - Thermospheric Temperature, density,
1984 - Numerical Modelinq of Low Latitude lonospheric and composition: new models. Smithsonian Astrophys. Obs.
Spec. Rep. N9 375.
Electric Fields. Eos, 65: 252,
FARLEY, D. T.. - 1960 - A Theory of Electrostatic Fields in KELLEY, M. C., & McCLURE, J. P. - 1981 - Equatorial Spread-F:
the lonosphere at non-polar Geomagnetic Latitudes, J. Geophys. a review of recent exper¡mental results, J. atmos. terr. Phys.,
Res..65: 869-877. 43: 427'435.
HEELIS, R. A.. KENDAL, P. C,, MOFFETT, R, J,, WINDLE, KUDEKI, E., FEJER, B. G., FARLEY, D, T., & IERKIC, H, M.-
D. W., & RISHBETH, H. - 1974 - Electrical Coupling of the 1981 - lnterferometer studies of equatorial F-region irregula-
E- and F-reqion drifts and winds. Planet. Space Sci,,22:743- rit¡e6 and drifts. Geophys, Res. Lett.,8: 377-380.
756. RISHBETH, H.. & GARRIOTT, O. K. - 1969 -- lntroduction to
 -
Í06 Effects of ionospheric bubbles

lonospheric Physics. New York, Academic, 136-'137 and 167. lonospheric bubbles observed by the Faraday rotation method
RISHBETH, H. - 1971a - The F-layer dynamo. Planet. Space at Natal, Brazil. Geophys. Res. Lett.,6: 473474.
Sci., 19: 263-267.
RISHBETH, H. - 1971b - Pr:larization fielcls produced by winds
in the equatorial F-region. Planet. Space Sci.,19: 357-369. Versão origìnal recebida em Jul./85;
YEH, K. C., SOICHER, H,, LIU, C. H. & BONELLI, E._ 1979- Versão final em Jun./86.