- Email: [email protected]
An analysis of the upper atmospheric wind observed by LOGACS
1036
R E S E A R ( ' H NOTES
at lower latitudes, as much as the above-mentioned density of aerosol particles agrees with the observatmnai data on noctilucent clouds: according to Donahue et aL (1972), the particle concentration reaches 40 cm -~. It should be added that noctilucent clouds are only a partial case, when the aerosol layer at altitudes of about 80 km becomes sharply pronounced. The considerable aerosol content at these altitudes is retained during the whole year, which is seen, in particular, from variations in OH and Na emissions (Shefov, 1968 ~. Thus, the effect observed by Hays and Roble (1973) can be explained by the existence of night-time noctilucent clouds without assuming the existence of an additional ozone layer. In the region of about 2400 A an aerosol particle about 10 5 cm in size has a scattering cross-section close to the absorption cross-section of 5"107 ozone molecules, but this cross-section is much less selective. This makes it possible, through measurements in two regions of the spectrum, to differentiate between these effects experimentally. Unfortunately, the papers by Hays and Roble (1968, 1973) do not contain any data necessary for thi',. It appears that there are as yet no grounds for attributing low-latitude effects observed by Hays and Roble (1973) precisely to the ozone effect and even less so to regard them as single invariable global standards (Evans and Llewellyn, 1973). At the same time, the necessity of simultaneous investigations of ozone and aerosol in the upper atmosphere at different latitudes and in different seasons is becoming obvious, One should also bear in mind the influence of the aerosol presence in the lower atmosphere, in particaJar the aerosol layer at an altitude of 10-30 km and 40-60 km, on the results of measuring ozone and O~ 1-27 micron emission. Institute o f Physics of the Atmosphere, G, V. ROSENBER~: U.S.S.R. Academy of Sciences, A . I . SEMENOV Moscow, U.S.S.R. N. IN. SHEFOV REFERENCES BAKER, D. C. and HALE, L. C. (1970). D-region parameters from blunt probe measurements during solar eclipse. Space Research, Vol. X, p. 712. North-Holland, Amsterdam. BRONSHTEN, V. A. and GRISHIN, N. I. (1970). Noctilucent Clouds. U.S.S.R. Nauka, Moscow. DONAHUE, T. M., GUENTER, B. and BLAMONT, J. E. (1972). Noctilucent clouds in daytime: circumpolar particulate layers near the summer mesopause. J. atmos. Sci. 29, 1205. DONAHUE, T. M. (1971). Ionosphere, D- and E-regions. U.S. National report, XV General Assembly I.U.G.G. (Moscow, 1971), p. 513. A.G.U., Washington, D. C. EVANS, W. F. J. and LLEWELLYN,E. J. (1973). Atomic hydrogen concentration in the mesosphere and the hydroxyl emissions. J. geophys. Res. 78, 323. HAYS, P. B. and ROBLE, R. G. (1968). Stellar spectra and atmospheric composition. J. atmos. Sci. 25, 1141. HAYS, P. B. and ROBLE, R. G. (1973). Observation of mesospheric ozone at low latitudes. Planet. Space Sci. 21, 273. KAI'LAN, S. A. and PIKELNER,S. B. (1973). Interstellar Medium, p. 268. U.S.S.R. Fizmatgiz, Moscow. KHVOSTtKOV, I. A. and WITT, G. (Eds.) (1967). Noetilucent Clouds. Proc. Int. Symp. Tallinn, 1966, pp. 28-94. ROSENBERG, G. V. (1966a). On the visibility of the noctilucent and nacreous clouds from the spaceships. Phys. Atmos. Ocean, U.S.S.R. 2, 39. ROSENBERG, G. V. (1966b). On the choice of conditions of observations of noctilucent clouds from the Earth's surface and spaceships. Meteorological Researches, No. 12, p. 30. U.S.S.R. Nauka, Moscow. SnEFOV, N. N. (1968). The behaviour of upper atmosphere emissions during high meteoric activity. Planet. Space Sci. 16, 134. WmrTEN, R. C. and PovPorr, L G. (1965). Physics of Lower Ionosphere. Prentice-Hall, Englewood Cliffs, New Jersey.
Planet. Space Sci. 1974, Vol. 22, pp. 1036 to 1041. Pergamon Press. Printed in Nortl~rn Ireland
AN
ANALYSIS
OF THE
UPPER ATMOSPHERIC BY LOGACS
WIND
OBSERVED
(Received 17 December 1973) A l ~ r a c t - - W i n d velocities at 140-200 km altitude were observed by a Low-G Accelerometer Calibration System (LOGACS) flown on an Agena satellite during a geomagnetic storm. An interesting wind reversal observed by the satellite at auroral latitudes is satisfactorily explained by the neutral air motion caused by the E × B drift deduced from the ground-based geomagnetic data recorded at stations near the meridian of the satellite orbit.
R E S E A R C H NOTES
1037
1. I N T R O D U C T I O N High-resolution wind observations in the lower thermosphere were made by an Agena satellite with an experiment designed to measure very small accelerations in the orbital motion. Data from this experiment designated LOGACS (Low-G Accelerometer Calibration System) were analyzed to determine the magnitude and direction of the wind component perpendicular to the orbital plane of the satellite (DeVries, 1972). Observations of the lateral wind component at 140-200 km altitude on 26 May 1967 for different orbits of the satellite are shown in Fig. 1. As seen for the 54th and 56th orbits, a reversal of the lateral wind direction occurred at about 65°N lat. In order to find a possible cause of this wind reversal, geomagnetic data obtained at a chain of stations near the satellite orbit were examined. As shown in Fig. 1, Dixon Island (DI; 73°33'N, 80°34'E), Sverdlovsk (SV; 56°44'N, 61°04'E), Tehran (TP; 35°44'N, 51°23'E) and Quetta (QU; 3 0 ° l l ' N , 66°57'E)are located near the meridional plane of the 56th orbit. The electromagnetic drift speeds were estimated from the geomagnetic variations at these stations, and were compared with the LOGACS wind measurement for the 56th orbit. Since the geomagnetic D component during disturbed periods is strongly controlled by field-aligned electric currents, the H component has mainly been studied to estimate the electric current intensity in the ionosphere assuming that a significant part of this component arises from ionospheric sources. Whether this assumption is justifiable as a first approximation is one of the purposes of the present research note. 2. M E T H O D As given by Chapman and Bartels (1940), the sheet electric current intensity in the ionosphere, I, can be estimated approximately by the geomagnetic variation field AB as follows Ix = AB,(f/2~.) I, = --ABx(f]Zzr) wherefis a parameter used to eliminate the induction current effect (commonly assumed to be 0'6), and the subscripts x andy represent the southward and the eastward component, respectively. When the geomagnetic H component only is considered, we have Iw ~- A H ( f / 2 ~ ) = 0.0955 AH. (1) As discussed by many scientists (e.g. Maeda and Kato, 1966; Matsushita, 1967), the southward electric field E~ can be calculated by the following relation when Ix is ignored E, = --[, ~.~,/(Z~ + 2 , + T.x,') --0"0955 A H ~ , / ( E , ~ E,, + Z , , 2) (2) I
90 deg. 53
180 deg.
56
0 deg.
1.5 krn/sec WIND
FIG. ] . DISTRIBUTIONS OF WIND VELOCITIES NORMAL TO THE SATELLITE PATH AT ABOUT 1 50 k m
ALTITUDEMEASUREDBY LOGACS FOR ITS 53RD, 54TH AND 56TH ORBITSON 26 MAY 1967 ARE SHOWN IN GEOGRAPHIC COORDINATES WITH THREE GREENWICH LONGITUDES AND THE EQUATORIAL SEMI-CIRCLE.
The 56th orbit was along the 60°-70°E meridian at approx. 6 hr UT. Locations of four geomagnetic stations (Dixon Island, Sverdlovsk, Tehran and Quetta) near the 56th orbit are shown by solid circles.
1038
RESEARCH
NOTES
where 52 is the height-integrated electric conductivity. The eastward ion speed V, due to E X B drifts aoo~c 140 k m altitude can be obtained by P~ ..... (IZ./B) cosec (I, i~ ) where B is the E a r t h ' s m a i n magnetic field and 09 is the magnetic dip angle. The neutral wind velocity V~ can be approximately determined from the ion drift velocity V~ as follows dV,,
_I V • P, __ V ~ _ _ -V,, -
- o T + p.
.
.
.
.
(4)
,
where P . is the stress tensor, p. is the neutral density, and r . , is the n e u t r a l - i o n eollisional time. If *~e neglect viscosity a n d assume that the neutral pressure gradient can be expressed in terms of neutral velocity as
_ vf,
:::
<5)
fl 1,
then (4) becomes dVn Vi -- (1 - g)V. -dt r~
(6)
with ~ = flr.~; ct a n d fl being constants indicating the order of magnitude of the pressure gradient. T h e solution of (6) is V,(t) =
exp
I/-
.....% I
(t' - t)
dt'.
Tni
0
(7)
g~z i
GEOMAGNETIC
H COMPONENT
! P~.%...---
"-,__.L- " "
I-oq
I 6
f
lO0
12
18
2:*
HOUF U T
xz
18
12
18
It*
[2
18
24
HOUR, UT
FIG. 2. GEOMAGNETIC / ir VARIATIOYqS (POSITIVE UPWARD) RECORDED AT FOUR STATIONS ON 25-26 MAY 1967. T h e zero level is assumed to be the average value of 6-7 hr U T on 25 May, a n d shown by the horizontal straight line. T h e b r o k e n curves indicate ADst-cos 0 measured f r o m the zero level, where ADst is the deviation from the Dst value (+30'?') averaged for 6--7 hr U T on 25 M a y and 0 is the latitude angle of each station.
RESEARCH NOTES
1039
This expression indicates that the neutral wind velocity approaches the ion drift velocity with a delay time of approximately *,d(1 -- ~), which is on the order of 5 × 103 sec for I~1 < 1 (e.g. Banks, 1966; Fedder and Banks, 1972). In other words, through ion-neutral collisions, V, --~ V~ = Yv may occur within a few hours. When V~ lasts for a few hours, then a neutral wind flow which has approximately the same velocity as Vy may be generated. Accordingly, V~ estimated from A H using Equations (1)-(3) can be compared with the LOGACS neutral-wind measurements, taking into consideration a few hours time lag. 3. RESULT As presented in Fig. 2, DI shows a positive variation in the H component, while the three other stations (SV, TP and QU) in lower latitudes than DI show a negative variation, at about 6 hr UT on 26 May 1967 corresponding to the time of the 56th satellite orbit. To estimate AH, the zero level is assumed to be the average value of 6-7 hr UT on 25 May as shown by the horizontal straight line in Fig. 2. Since the AH due solely to ionospheric electric currents is needed to discuss ionospheric winds, ring-current effects must be MAY I
oN
1967
26
i
'
i
DI
AH
70
¢,
50
•
SV
\ \ \
TP QU
t
30 -400 ON
1
I I 1 . 1 200
-200 '
I
I
[
l
DI
:.-c---,
70
x
400 Y
l
• SV
50 l Ill
/
Ex
~
Tp Qu
30
i -60
I
i
-40
I
i
i
I
i
20
-20
oN
'
I
J
I 40
~
mV/m
l
DI
70
-----.-~°
50 I 30
I -1,6
FIG. 3.
(TOP.)
i
I -0.8
•
TP
•
QU
i 0.8
1.6 km/s
DEVIATIONS OF H VALUES FROM THE ZERO LEVEL (SOLID CIRCLES) AND THOSE
FROM ADst'cos 0 (CROSSES) AT 6 hr U T ON 26 MAY 1967. (MIDDLE.) SOUTHWARD ELECTRIC FIELD, E=, CALCULATED FROM H DEVIATIONS. (BOTTOM.) OBSERVED NEUTRAL WIND SPEED BY LOGACS (SMALL SOLID CIRCLES WITH A FITTED SMOOTH CURVE OBTAINED FROM THE WIND SPEEDS FOR THE 56TH ORBIT IN FIG. I ) ARE COMPARED WITH CALCULATED EASTWARD DRIFT SPEED Vu (LARGE SOLID CIRCLES AND CROSSES).
1040
RESEARCtt NOTES
eliminated. From the equatorial Dst values obtained by Sugiura and Cain (1970), ADst.cos 0 was computed (broken curves in Fig. 2), where 0 is the latitude angle of each station and ADst is the deviation from the Ost value averaged for 6-7 hr UT on 25 May. Deviations of H values from the zero level (solid circles) and those from ADst.cos 0 (crosses) for 01e four stations at 6 hr UT on 26 May are presented in the top diagram of Fig. 3. Solid circles indicate the extreme case with no Dst contribution. It can be noticed that the sign of A H changes at 60-70°N l:~t~ roughly corresponding to the change-over location of the LOGACS wind direction shown in Fig. 1~ For simplicity the scaling of AH is made for 6 hr UT on 26 May corresponding to the 56th orbit. As discussed at the end of the preceding section, a few hours are required to produce a neutral wind velocity t~f the same order as the ion drift velocity. Since the AH values remain generally the same during a few hotJrs before 6 hi" UT on 26 May (see Fig. 2), AH at 6 hr UT is acceptable as a representative value of the geomagnetic variation for the present discussion. To obtain E~ from Equation (2), values of the height-integrated electric conductivity ~ are necessary. Since these values during geomagnetic storms are difficult to estimate accurately, the Z values during quiet periods (Tarpley, 1969) are substituted, bearing in mind that the resulting [E,] must be close to the maximum value. The adopted conductivities at four geomagnetic stations are listed in Table 1. Calculated E~ distributions with (crosses) and without ~circles) Dst effect are shown in the middle of Fig. 3. TABLE 1. HEIGHT-INTEGRATEDELECTRIC CONDUCTIV1TIES (in 10 a emu)
DI SV TP QU
4.27 5-79 I 1.5 15.5
5.38 7.42 14.0 17.8
4.19 5.24 78.3 89.0
With E, from Equation (3), the eastward drift speed Vu can easily be obtained. These Vv are compared with the observed neutral wind speed by LOGACS in the bottom diagram of Fig. 3. Agreement between the two is satisfactory as a first order approximation. However, the absolute values of the LOGACS wind speeds at latitudes higher than 50°N are slightly larger than the calculated [ V~I values which are close to the maximum since they are obtained from the near maximum IE, I values. One possible explanation of this discrepancy is that IAHI recorded at the ground is smaller than that which can be attributed to an electric current of purely ionospheric origin. Field-aligned electric currents and some magnetospheric variations may cause a reduction of [AHI at the ground. Although various data and careful examination are needed to discuss the details, an interesting conclusion is that the reversal of the LOGACS wind direction near 65°N is understandable in terms of the reversed ionospheric drift directions between Dixon Island and Sverdlovsk inferred simply from the geomagnetic H records at these stations.
Acknowledgements--We wish to thank Drs. A. D. Richmond, K. Kawasaki and D. N. Anderson for their fruitful discussions throughout this study and Mrs. Robin C. Lyons for her assistance in the preparation of the present research note. One of the authors (STW) wishes to express his gratitude to the National Center for Atmospheric Research for sponsoring his visit to the High Altitude Observatory and also the support from MSFC/NASA through a contract (NAS8-25750) with the University of Alabama in Huntsville. The National Center for Atmospheric Research is sponsored by the National Science Foundation. High Altitude Observatory, National Center for Atmospheric Research, Boulder, Colorado 80302, U.S.A.
S . T . Wu* S. MATSUSmTA
Aerospace Environment Division, Marshall Space Flight Center[NASA, Huntsville, Alabama 35812, U.S.A.
L . L . DEVR~s
REFERENCES
BANKS, P. (1966). Collision frequencies and energy transfer--Ions. Planet. Space Sci. 14, 1105-1122. CHAPMAN,S. and BARTELS,J. (1940). Geomagnetism. Oxford University Press, London. * Permanent address: School of Science and Engineering, The University of Alabama in Huntsville, Huntsville, Alabama 35807, U.S.A.
R E S E A R C H NOTES
1041
DEVR1ES, L. L. (1972). Analysis and interpretation of density data from the low-G accelerometer calibration system (LOGACS). Space Research, Vol. XII, pp. 777-789. Akademie, Berlin. FEDDER, J. A. and BANKS, P. M. (1972). Convection electric fields and polar thermospheric winds. J. geophys. Res. 77, 2328-2340. MAEDA, K. and KATO, S. (1966). Electrodynamics of the ionosphere. Space Sci. Rev. 5, 57-79. MATSUSHITA,S. (1967). Solar quiet and lunar daily variation fields. Physics of Geomagnetic Phenomena (Eds. S. Matsushita and W. H. Campbell), p. 301. Academic Press, New York. SUGIURA, M. and CAIN, S. (1970). Provisional hourly values of equatorial Dst for 1964, 1965, 1966 and 1967. NASA TN D-5748. TARPLEY, J. D. (1969). The ionospheric wind dynamo. Ph.D. Thesis, Univ. of Colorado, Boulder.
Planet. Space Sci. 1974, Vol. 22, pp. 1041 to 1042. Pergamon Press. Printed in Northern Ireland
CORRELATION
OF INTERPLANETARY PLASMA
DENSITY
SCINTILLATION
AND
SPACECRAFT
MEASUREMENTS
(Received 2 January 1974) Abstract--Interplanetary scintillation and spacecraft measurements of proton density are observed to be strongly correlated. This result confirms the association between corotating sectors of enhanced scintillation and compression regions located at the interface between fast and slow solar wind streams; it also suggests that scintillation observations provide a means of predicting, several days in advance, the arrival at the Earth, of solar plasma of enhanced density. Extended observations of interplanetary scintillation on a grid of radio sources suitably disposed about the Sun have shown that enhanced scintillation sectors are strongly correlated with the velocity structure of the solar wind. This association suggests that enhanced scintillation is caused by the existence of regions of compressed plasma resulting from the interaction of fast and slow velocity streams (Houminer and Hewish, 1972). The scintillation index F varies as F oc ANPZ2, where AN is the r.m.s, fluctuation of electron density and l the scale-size of the irregularities, and it has been shown that enhanced scintillation is caused by an increase of AN (Houminer, 1971). Hitherto there has, however, been no direct evidence that the mean density N also increases in the enhanced scintillation regions due to insufficient spacecraft data. Proton density data from Pioneer 6 and 7 have recently become available (Lazarus, private communication) and some results for the period January-April 1971 are shown in Fig. 1, together with scintillation 20
10
0 0"15 0-1 0"05 0 /-,0 20
0 20
25 JAN
30 J
5
10
15 FEB
20
25
J
5
10
15
20 MAR
25-
30 J
5
10 APR
FIG. I. OBSERVATIONSOF INTERPLANETARYSCINTILLATIONF, SPACECRAFTPROTONDENSITYN (PIONEER 6 AND 7) AND GEOMAGNETICINDEXAp DURINGJANUARY-APRIL1971. The values of F and N have been corotated to the Earth.
15 1971
R E S E A R ( ' H NOTES
at lower latitudes, as much as the above-mentioned density of aerosol particles agrees with the observatmnai data on noctilucent clouds: according to Donahue et aL (1972), the particle concentration reaches 40 cm -~. It should be added that noctilucent clouds are only a partial case, when the aerosol layer at altitudes of about 80 km becomes sharply pronounced. The considerable aerosol content at these altitudes is retained during the whole year, which is seen, in particular, from variations in OH and Na emissions (Shefov, 1968 ~. Thus, the effect observed by Hays and Roble (1973) can be explained by the existence of night-time noctilucent clouds without assuming the existence of an additional ozone layer. In the region of about 2400 A an aerosol particle about 10 5 cm in size has a scattering cross-section close to the absorption cross-section of 5"107 ozone molecules, but this cross-section is much less selective. This makes it possible, through measurements in two regions of the spectrum, to differentiate between these effects experimentally. Unfortunately, the papers by Hays and Roble (1968, 1973) do not contain any data necessary for thi',. It appears that there are as yet no grounds for attributing low-latitude effects observed by Hays and Roble (1973) precisely to the ozone effect and even less so to regard them as single invariable global standards (Evans and Llewellyn, 1973). At the same time, the necessity of simultaneous investigations of ozone and aerosol in the upper atmosphere at different latitudes and in different seasons is becoming obvious, One should also bear in mind the influence of the aerosol presence in the lower atmosphere, in particaJar the aerosol layer at an altitude of 10-30 km and 40-60 km, on the results of measuring ozone and O~ 1-27 micron emission. Institute o f Physics of the Atmosphere, G, V. ROSENBER~: U.S.S.R. Academy of Sciences, A . I . SEMENOV Moscow, U.S.S.R. N. IN. SHEFOV REFERENCES BAKER, D. C. and HALE, L. C. (1970). D-region parameters from blunt probe measurements during solar eclipse. Space Research, Vol. X, p. 712. North-Holland, Amsterdam. BRONSHTEN, V. A. and GRISHIN, N. I. (1970). Noctilucent Clouds. U.S.S.R. Nauka, Moscow. DONAHUE, T. M., GUENTER, B. and BLAMONT, J. E. (1972). Noctilucent clouds in daytime: circumpolar particulate layers near the summer mesopause. J. atmos. Sci. 29, 1205. DONAHUE, T. M. (1971). Ionosphere, D- and E-regions. U.S. National report, XV General Assembly I.U.G.G. (Moscow, 1971), p. 513. A.G.U., Washington, D. C. EVANS, W. F. J. and LLEWELLYN,E. J. (1973). Atomic hydrogen concentration in the mesosphere and the hydroxyl emissions. J. geophys. Res. 78, 323. HAYS, P. B. and ROBLE, R. G. (1968). Stellar spectra and atmospheric composition. J. atmos. Sci. 25, 1141. HAYS, P. B. and ROBLE, R. G. (1973). Observation of mesospheric ozone at low latitudes. Planet. Space Sci. 21, 273. KAI'LAN, S. A. and PIKELNER,S. B. (1973). Interstellar Medium, p. 268. U.S.S.R. Fizmatgiz, Moscow. KHVOSTtKOV, I. A. and WITT, G. (Eds.) (1967). Noetilucent Clouds. Proc. Int. Symp. Tallinn, 1966, pp. 28-94. ROSENBERG, G. V. (1966a). On the visibility of the noctilucent and nacreous clouds from the spaceships. Phys. Atmos. Ocean, U.S.S.R. 2, 39. ROSENBERG, G. V. (1966b). On the choice of conditions of observations of noctilucent clouds from the Earth's surface and spaceships. Meteorological Researches, No. 12, p. 30. U.S.S.R. Nauka, Moscow. SnEFOV, N. N. (1968). The behaviour of upper atmosphere emissions during high meteoric activity. Planet. Space Sci. 16, 134. WmrTEN, R. C. and PovPorr, L G. (1965). Physics of Lower Ionosphere. Prentice-Hall, Englewood Cliffs, New Jersey.
Planet. Space Sci. 1974, Vol. 22, pp. 1036 to 1041. Pergamon Press. Printed in Nortl~rn Ireland
AN
ANALYSIS
OF THE
UPPER ATMOSPHERIC BY LOGACS
WIND
OBSERVED
(Received 17 December 1973) A l ~ r a c t - - W i n d velocities at 140-200 km altitude were observed by a Low-G Accelerometer Calibration System (LOGACS) flown on an Agena satellite during a geomagnetic storm. An interesting wind reversal observed by the satellite at auroral latitudes is satisfactorily explained by the neutral air motion caused by the E × B drift deduced from the ground-based geomagnetic data recorded at stations near the meridian of the satellite orbit.
R E S E A R C H NOTES
1037
1. I N T R O D U C T I O N High-resolution wind observations in the lower thermosphere were made by an Agena satellite with an experiment designed to measure very small accelerations in the orbital motion. Data from this experiment designated LOGACS (Low-G Accelerometer Calibration System) were analyzed to determine the magnitude and direction of the wind component perpendicular to the orbital plane of the satellite (DeVries, 1972). Observations of the lateral wind component at 140-200 km altitude on 26 May 1967 for different orbits of the satellite are shown in Fig. 1. As seen for the 54th and 56th orbits, a reversal of the lateral wind direction occurred at about 65°N lat. In order to find a possible cause of this wind reversal, geomagnetic data obtained at a chain of stations near the satellite orbit were examined. As shown in Fig. 1, Dixon Island (DI; 73°33'N, 80°34'E), Sverdlovsk (SV; 56°44'N, 61°04'E), Tehran (TP; 35°44'N, 51°23'E) and Quetta (QU; 3 0 ° l l ' N , 66°57'E)are located near the meridional plane of the 56th orbit. The electromagnetic drift speeds were estimated from the geomagnetic variations at these stations, and were compared with the LOGACS wind measurement for the 56th orbit. Since the geomagnetic D component during disturbed periods is strongly controlled by field-aligned electric currents, the H component has mainly been studied to estimate the electric current intensity in the ionosphere assuming that a significant part of this component arises from ionospheric sources. Whether this assumption is justifiable as a first approximation is one of the purposes of the present research note. 2. M E T H O D As given by Chapman and Bartels (1940), the sheet electric current intensity in the ionosphere, I, can be estimated approximately by the geomagnetic variation field AB as follows Ix = AB,(f/2~.) I, = --ABx(f]Zzr) wherefis a parameter used to eliminate the induction current effect (commonly assumed to be 0'6), and the subscripts x andy represent the southward and the eastward component, respectively. When the geomagnetic H component only is considered, we have Iw ~- A H ( f / 2 ~ ) = 0.0955 AH. (1) As discussed by many scientists (e.g. Maeda and Kato, 1966; Matsushita, 1967), the southward electric field E~ can be calculated by the following relation when Ix is ignored E, = --[, ~.~,/(Z~ + 2 , + T.x,') --0"0955 A H ~ , / ( E , ~ E,, + Z , , 2) (2) I
90 deg. 53
180 deg.
56
0 deg.
1.5 krn/sec WIND
FIG. ] . DISTRIBUTIONS OF WIND VELOCITIES NORMAL TO THE SATELLITE PATH AT ABOUT 1 50 k m
ALTITUDEMEASUREDBY LOGACS FOR ITS 53RD, 54TH AND 56TH ORBITSON 26 MAY 1967 ARE SHOWN IN GEOGRAPHIC COORDINATES WITH THREE GREENWICH LONGITUDES AND THE EQUATORIAL SEMI-CIRCLE.
The 56th orbit was along the 60°-70°E meridian at approx. 6 hr UT. Locations of four geomagnetic stations (Dixon Island, Sverdlovsk, Tehran and Quetta) near the 56th orbit are shown by solid circles.
1038
RESEARCH
NOTES
where 52 is the height-integrated electric conductivity. The eastward ion speed V, due to E X B drifts aoo~c 140 k m altitude can be obtained by P~ ..... (IZ./B) cosec (I, i~ ) where B is the E a r t h ' s m a i n magnetic field and 09 is the magnetic dip angle. The neutral wind velocity V~ can be approximately determined from the ion drift velocity V~ as follows dV,,
_I V • P, __ V ~ _ _ -V,, -
- o T + p.
.
.
.
.
(4)
,
where P . is the stress tensor, p. is the neutral density, and r . , is the n e u t r a l - i o n eollisional time. If *~e neglect viscosity a n d assume that the neutral pressure gradient can be expressed in terms of neutral velocity as
_ vf,
:::
<5)
fl 1,
then (4) becomes dVn Vi -- (1 - g)V. -dt r~
(6)
with ~ = flr.~; ct a n d fl being constants indicating the order of magnitude of the pressure gradient. T h e solution of (6) is V,(t) =
exp
I/-
.....% I
(t' - t)
dt'.
Tni
0
(7)
g~z i
GEOMAGNETIC
H COMPONENT
! P~.%...---
"-,__.L- " "
I-oq
I 6
f
lO0
12
18
2:*
HOUF U T
xz
18
12
18
It*
[2
18
24
HOUR, UT
FIG. 2. GEOMAGNETIC / ir VARIATIOYqS (POSITIVE UPWARD) RECORDED AT FOUR STATIONS ON 25-26 MAY 1967. T h e zero level is assumed to be the average value of 6-7 hr U T on 25 May, a n d shown by the horizontal straight line. T h e b r o k e n curves indicate ADst-cos 0 measured f r o m the zero level, where ADst is the deviation from the Dst value (+30'?') averaged for 6--7 hr U T on 25 M a y and 0 is the latitude angle of each station.
RESEARCH NOTES
1039
This expression indicates that the neutral wind velocity approaches the ion drift velocity with a delay time of approximately *,d(1 -- ~), which is on the order of 5 × 103 sec for I~1 < 1 (e.g. Banks, 1966; Fedder and Banks, 1972). In other words, through ion-neutral collisions, V, --~ V~ = Yv may occur within a few hours. When V~ lasts for a few hours, then a neutral wind flow which has approximately the same velocity as Vy may be generated. Accordingly, V~ estimated from A H using Equations (1)-(3) can be compared with the LOGACS neutral-wind measurements, taking into consideration a few hours time lag. 3. RESULT As presented in Fig. 2, DI shows a positive variation in the H component, while the three other stations (SV, TP and QU) in lower latitudes than DI show a negative variation, at about 6 hr UT on 26 May 1967 corresponding to the time of the 56th satellite orbit. To estimate AH, the zero level is assumed to be the average value of 6-7 hr UT on 25 May as shown by the horizontal straight line in Fig. 2. Since the AH due solely to ionospheric electric currents is needed to discuss ionospheric winds, ring-current effects must be MAY I
oN
1967
26
i
'
i
DI
AH
70
¢,
50
•
SV
\ \ \
TP QU
t
30 -400 ON
1
I I 1 . 1 200
-200 '
I
I
[
l
DI
:.-c---,
70
x
400 Y
l
• SV
50 l Ill
/
Ex
~
Tp Qu
30
i -60
I
i
-40
I
i
i
I
i
20
-20
oN
'
I
J
I 40
~
mV/m
l
DI
70
-----.-~°
50 I 30
I -1,6
FIG. 3.
(TOP.)
i
I -0.8
•
TP
•
QU
i 0.8
1.6 km/s
DEVIATIONS OF H VALUES FROM THE ZERO LEVEL (SOLID CIRCLES) AND THOSE
FROM ADst'cos 0 (CROSSES) AT 6 hr U T ON 26 MAY 1967. (MIDDLE.) SOUTHWARD ELECTRIC FIELD, E=, CALCULATED FROM H DEVIATIONS. (BOTTOM.) OBSERVED NEUTRAL WIND SPEED BY LOGACS (SMALL SOLID CIRCLES WITH A FITTED SMOOTH CURVE OBTAINED FROM THE WIND SPEEDS FOR THE 56TH ORBIT IN FIG. I ) ARE COMPARED WITH CALCULATED EASTWARD DRIFT SPEED Vu (LARGE SOLID CIRCLES AND CROSSES).
1040
RESEARCtt NOTES
eliminated. From the equatorial Dst values obtained by Sugiura and Cain (1970), ADst.cos 0 was computed (broken curves in Fig. 2), where 0 is the latitude angle of each station and ADst is the deviation from the Ost value averaged for 6-7 hr UT on 25 May. Deviations of H values from the zero level (solid circles) and those from ADst.cos 0 (crosses) for 01e four stations at 6 hr UT on 26 May are presented in the top diagram of Fig. 3. Solid circles indicate the extreme case with no Dst contribution. It can be noticed that the sign of A H changes at 60-70°N l:~t~ roughly corresponding to the change-over location of the LOGACS wind direction shown in Fig. 1~ For simplicity the scaling of AH is made for 6 hr UT on 26 May corresponding to the 56th orbit. As discussed at the end of the preceding section, a few hours are required to produce a neutral wind velocity t~f the same order as the ion drift velocity. Since the AH values remain generally the same during a few hotJrs before 6 hi" UT on 26 May (see Fig. 2), AH at 6 hr UT is acceptable as a representative value of the geomagnetic variation for the present discussion. To obtain E~ from Equation (2), values of the height-integrated electric conductivity ~ are necessary. Since these values during geomagnetic storms are difficult to estimate accurately, the Z values during quiet periods (Tarpley, 1969) are substituted, bearing in mind that the resulting [E,] must be close to the maximum value. The adopted conductivities at four geomagnetic stations are listed in Table 1. Calculated E~ distributions with (crosses) and without ~circles) Dst effect are shown in the middle of Fig. 3. TABLE 1. HEIGHT-INTEGRATEDELECTRIC CONDUCTIV1TIES (in 10 a emu)
DI SV TP QU
4.27 5-79 I 1.5 15.5
5.38 7.42 14.0 17.8
4.19 5.24 78.3 89.0
With E, from Equation (3), the eastward drift speed Vu can easily be obtained. These Vv are compared with the observed neutral wind speed by LOGACS in the bottom diagram of Fig. 3. Agreement between the two is satisfactory as a first order approximation. However, the absolute values of the LOGACS wind speeds at latitudes higher than 50°N are slightly larger than the calculated [ V~I values which are close to the maximum since they are obtained from the near maximum IE, I values. One possible explanation of this discrepancy is that IAHI recorded at the ground is smaller than that which can be attributed to an electric current of purely ionospheric origin. Field-aligned electric currents and some magnetospheric variations may cause a reduction of [AHI at the ground. Although various data and careful examination are needed to discuss the details, an interesting conclusion is that the reversal of the LOGACS wind direction near 65°N is understandable in terms of the reversed ionospheric drift directions between Dixon Island and Sverdlovsk inferred simply from the geomagnetic H records at these stations.
Acknowledgements--We wish to thank Drs. A. D. Richmond, K. Kawasaki and D. N. Anderson for their fruitful discussions throughout this study and Mrs. Robin C. Lyons for her assistance in the preparation of the present research note. One of the authors (STW) wishes to express his gratitude to the National Center for Atmospheric Research for sponsoring his visit to the High Altitude Observatory and also the support from MSFC/NASA through a contract (NAS8-25750) with the University of Alabama in Huntsville. The National Center for Atmospheric Research is sponsored by the National Science Foundation. High Altitude Observatory, National Center for Atmospheric Research, Boulder, Colorado 80302, U.S.A.
S . T . Wu* S. MATSUSmTA
Aerospace Environment Division, Marshall Space Flight Center[NASA, Huntsville, Alabama 35812, U.S.A.
L . L . DEVR~s
REFERENCES
BANKS, P. (1966). Collision frequencies and energy transfer--Ions. Planet. Space Sci. 14, 1105-1122. CHAPMAN,S. and BARTELS,J. (1940). Geomagnetism. Oxford University Press, London. * Permanent address: School of Science and Engineering, The University of Alabama in Huntsville, Huntsville, Alabama 35807, U.S.A.
R E S E A R C H NOTES
1041
DEVR1ES, L. L. (1972). Analysis and interpretation of density data from the low-G accelerometer calibration system (LOGACS). Space Research, Vol. XII, pp. 777-789. Akademie, Berlin. FEDDER, J. A. and BANKS, P. M. (1972). Convection electric fields and polar thermospheric winds. J. geophys. Res. 77, 2328-2340. MAEDA, K. and KATO, S. (1966). Electrodynamics of the ionosphere. Space Sci. Rev. 5, 57-79. MATSUSHITA,S. (1967). Solar quiet and lunar daily variation fields. Physics of Geomagnetic Phenomena (Eds. S. Matsushita and W. H. Campbell), p. 301. Academic Press, New York. SUGIURA, M. and CAIN, S. (1970). Provisional hourly values of equatorial Dst for 1964, 1965, 1966 and 1967. NASA TN D-5748. TARPLEY, J. D. (1969). The ionospheric wind dynamo. Ph.D. Thesis, Univ. of Colorado, Boulder.
Planet. Space Sci. 1974, Vol. 22, pp. 1041 to 1042. Pergamon Press. Printed in Northern Ireland
CORRELATION
OF INTERPLANETARY PLASMA
DENSITY
SCINTILLATION
AND
SPACECRAFT
MEASUREMENTS
(Received 2 January 1974) Abstract--Interplanetary scintillation and spacecraft measurements of proton density are observed to be strongly correlated. This result confirms the association between corotating sectors of enhanced scintillation and compression regions located at the interface between fast and slow solar wind streams; it also suggests that scintillation observations provide a means of predicting, several days in advance, the arrival at the Earth, of solar plasma of enhanced density. Extended observations of interplanetary scintillation on a grid of radio sources suitably disposed about the Sun have shown that enhanced scintillation sectors are strongly correlated with the velocity structure of the solar wind. This association suggests that enhanced scintillation is caused by the existence of regions of compressed plasma resulting from the interaction of fast and slow velocity streams (Houminer and Hewish, 1972). The scintillation index F varies as F oc ANPZ2, where AN is the r.m.s, fluctuation of electron density and l the scale-size of the irregularities, and it has been shown that enhanced scintillation is caused by an increase of AN (Houminer, 1971). Hitherto there has, however, been no direct evidence that the mean density N also increases in the enhanced scintillation regions due to insufficient spacecraft data. Proton density data from Pioneer 6 and 7 have recently become available (Lazarus, private communication) and some results for the period January-April 1971 are shown in Fig. 1, together with scintillation 20
10
0 0"15 0-1 0"05 0 /-,0 20
0 20
25 JAN
30 J
5
10
15 FEB
20
25
J
5
10
15
20 MAR
25-
30 J
5
10 APR
FIG. I. OBSERVATIONSOF INTERPLANETARYSCINTILLATIONF, SPACECRAFTPROTONDENSITYN (PIONEER 6 AND 7) AND GEOMAGNETICINDEXAp DURINGJANUARY-APRIL1971. The values of F and N have been corotated to the Earth.
15 1971