Transcript Slide 1

Evidence for short correlation lengths of the noontime equatorial electrojet – inferred from a
comparison of satellite and ground magnetic data.
C. Manoj
National Geophysical Research Institute, Hyderabad, India.
H. Lühr
GeoForschungsZentrum – Potsdam, Germany
S. Maus
CIRES, University of Colorado, USA
N. Nagarajan
National Geophysical Research Institute , Hyderabad, India.
03.07.2007 | 9:30 am | IUGG | page 1/25
Manoj et al, Evidence for short ....
Equatorial Electrojet - generation
Solar tidal effects causes current flow in
the day time ionosphere E region (Sq)
Sq current system sustains an eastward
directed electrified from dawn-dusk at
low latitude.
A Hall current is then generated, carried
by the upward moving electrons.
The non-conductive boundaries above
and below the lower ionosphere causes
large vertical electric field build up.
This vertical electric field (about 5 to 10
times stronger than the eastward electric
field that produced it.
(Figure from Anderson et al, 2002)
This vertical field generates an eastward
current called equatorial electrojet (EEJ)
in noon-time ionosphere
03.07.2007 | 9:30 am | IUGG | page 2/25
Manoj et al, Evidence for short ....
Equatorial Electrojet – magnetic fields
The equatorial electrojet produces strong
enhancement of horizontal magnetic
intensity within ±3° of the magnetic
equator.
EEJ has been studied using
magnetometer array, rockets,
radar, satellites… etc. etc..
Simulated horizontal magnetic anomaly at ground
due to ionospheric currents (from CM4). Unit - nT
03.07.2007 | 9:30 am | IUGG | page 3/25
Manoj et al, Evidence for short ....
Equatorial Electrojet – magnetic fields
60
50
ETT H - HYBH
A unique way of studying the EEJ is by
using the differences in horizontal
magnetic variations at an equatorial
observatory from another observatory
separated by 10°-15° in latitude.
40
30
20
10
0
-10
0
5
10
15
20
25
LT
EEJ was also studied by satellite
missions like POGO, Magsat, Oersted
and CHAMP. LEO satellites, which flies
above the ionosphere senses EEJ as
negative signal at dip equator.
Lühr et al, 2004
03.07.2007 | 9:30 am | IUGG | page 4/25
Manoj et al, Evidence for short ....
Some open questions on EEJ
Lühr et al, 2004 reports uncorrelated current
strength between successive CHAMP passes
over EEJ. These passes are separated in
space by ~23º and in time by ~93 minutes.
03.07.2007 | 9:30 am | IUGG | page 5/25
Manoj et al, Evidence for short ....
Some open questions on EEJ
90 N
180
0  W
Is the observed variability in EEJ
current strength due to spatial
(23º) or temporal (93 minutes)
effects ?
135 W
90 W
45 W
0
45 E
90 E
135 E
180 E
UT 6
45 S
90 S
-40
-30
-20
-10
0
10
20
30
40
23º West and 93 minutes later
90 N
180
0  W
135 W
90 W
45 W
0
45 E
90 E
135 E
45 S
90 S
03.07.2007 | 9:30 am | IUGG | page 6/25
Manoj et al, Evidence for short ....
UT 7:30
180 E
Some open questions on EEJ
Are Sq and EEJ current systems coupled ?
EEJ is often modeled as an equatorial enhancement of
a coherent, large scale Sq current system (for eg.
MacDaugall, 1979, CM4, Sabaka et al, 2004 ). Forbes
(1981) concludes that EEJ and Sq are coupled current
systems. This finding is also supported by Hesse
(1982).
However studies by Mann & Schlapp (1988) and Okeke
(2006) shows poor correlation of horizontal magnetic
fields between observatories within the equatorial region
and outside of it. Also studies by Raghavarao &
Anandarao (1987) finds that Sq and EEJ are decoupled.
03.07.2007 | 9:30 am | IUGG | page 7/25
Manoj et al, Evidence for short ....
How do we go about it ?
While, from the ground, a continuous record of the current-induced
magnetic field is obtained, polar orbiting satellites take just a snapshot of
the latitudinal current distribution while passing over the equatorial region.
The temporal variations recorded by a ground station can either be
caused by a change in current strength or by a displacement of the
current axis. Satellite measurements on the other hand give no
information on the temporal variation of the EEJ but a good picture of the
current geometry.
By combining both data sets the advantages can be used to eliminate
several ambiguities and answer the questions we discussed.
03.07.2007 | 9:30 am | IUGG | page 8/25
Manoj et al, Evidence for short ....
Roadmap
Observatory and satellite data.
Data processing
Correlation analysis
Results
03.07.2007 | 9:30 am | IUGG | page 9/25
Manoj et al, Evidence for short ....
Observatory data
Distribution of the
geomagnetic observatories
used for the study.
GUI
QSB
ABG
MBO
FUQ
Hourly means of the horizontal
intensities from 13 observatories.
CBI
AAE
TIR
HYB
PND
ETT
GUA
HUA
Period: Sep 2000 – Dec. 2002
Screened for Kp ≤ 2 to limit the
analysis to magnetically quiet
days.
number of hourly means
30000
25000
20000
15000
10000
5000
0
03.07.2007 | 9:30 am | IUGG | page 10/25
ABG ETT HYB TIR HUA FUQ MBO GUI AAE QSB GUA CBI PND
Manoj et al, Evidence for short ....
EEJ signals from ground data
60
ΔHEEJ – ΔHNon-EEJ
50
ETT H - HYBH
ΔH is the variation from
midnight level.
Average daily variation of the
horizontal components of
geomagnetic field observed at ETT
with respect to the station HYB.
40
30
20
10
0
Typically, the EEJ signal reaches
up to 53 nT. The solid line
represents a polynomial fit to the
data.
-10
0
5
10
15
20
25
LT
EEJ signals for 2000-2002
UT (Hours)
0
100
10
0
20
2000
-100
2001
2002
Time (Years)
03.07.2007 | 9:30 am | IUGG | page 11/25
Manoj et al, Evidence for short ....
2003
Satellite data
Scalar magnetic field data from 2000 to 2002
Local Time : 10 to 13
Kp index ≤ 2
Total 1653 crossings
L
R
Data reduction
Main field (Pomme 1.4, Maus et al, 2005)
Lithospheric field (MF2, Maus et al, 2002)
Diamagnetic effect (Lühr et al, 2003)
Large-scale magnetospheric fields by polynomial
fitting
Re-drawn from Lühr et al, 2004
Current density distribution was modeled by series
of EW oriented current lines separated by 0.5º in
latitude and located at an altitude of 108 km.
Induction effect conductosphere at depth of 200 km
03.07.2007 | 9:30 am | IUGG | page 12/25
Manoj et al, Evidence for short ....
Magnetic profile from CHAMP
100
Predicted ground magnetic field
profile due to the noon time
equatorial electrojet from the
CHAMP average current
profile.
TIR
HUA
AAE
ETT
80
60
B
ABG
20
MBO
PND
0
FUQ
GUA
HYB
-20
-40
-20
03.07.2007 | 9:30 am | IUGG | page 13/25
Bz
40
nT
The locations of geomagnetic
observatories are plotted with
respect to the dip-equator
along the magnetic field profile
Bx
-15
-10
-5
0
5
10
15
Degrees Latitude about dip-equator
Manoj et al, Evidence for short ....
20
LT correction
60
A degree-9 polynomial was used to find
The ratio of expected EEJ strength at
observatory and satellite local time
03.07.2007 | 9:30 am | IUGG | page 14/25
50
ETT H - HYBH
Since the satellite crosses the dipequator at a certain LT and the
corresponding observatory data may
have a different LT, a correction needs
to be applied to make the data set
comparable.
40
30
20
10
0
-10
0
5
10
15
20
LT
Manoj et al, Evidence for short ....
25
Sq correction
50
GUAH-CBIH (nT)
By subtracting the data from non – equatorial
observatory, we remove a part of the Sq
variation at the equatorial observatory.
The unresolved part corresponds to the
latitudinal slope of the Sq between the
observatory pair.
40
30
20
10
0
Although none of the two stations is directly
below the EEJ a daily variation of more than
50 nT is seen here.
5
ETT-HYB 2000 /10 /7, 7:30 UT
ETT H
 H (nT)
50
HYB H
Sq at ETT
CM4 Sq
0
-50
50
03.07.2007 | 9:30 am | IUGG | page 15/25
20
100
PND H
CM4 model (Sabaka et al, 2004) was used to
obtain an estimate of the latitudinal slope of
the Sq signal between the observatory pairs.
10
15
LT (Hours)
30
10
-10
-30
Geographic Latitude
Manoj et al, Evidence for short ....
-50
Correlation Analysis
0
100
CC 0.94
= 299.6 * I + -9.75
80 H
60
H
-10
100
CC 0.83
10
100
CC 0.81
40
80
80
20
60
60
0
0
40
20
0
0
0.1
0.2 0.3
A/m
0.4
40
20
0.1
0.2
0.3
0
0
0.4
-20
100
0.1
0.2 0.3
A/m
0.4
CC 0.49
80
100
60
80
40
60
20
CC 0.56
1
20
0.1
0.2
0.3
0.4
-30
CC 0.15
80
H
60
40
Correlation Coefficient
20
0
0
100
40
0
0
0.1
0.5
0
0.1
0.2
0.3
0.4
03.07.2007 | 9:30 am | IUGG | page 16/25
0.4
With LT correction
Without LT correction
20
0
0
0.2 0.3
A/m
-0.5
-40 -30 -20 -10
0
10 20 30 40
Distance from Observatory in Degrees
Manoj et al, Evidence for short ....
Correlation Analysis
GUI
QSB
MBO
FUQ
CBI
ABG
AAE
TIR
HYB
PND
ETT
HUA
Correlation coefficients as function of
distance from the observatories.
The central bin gives a high
correlation between the satellite
and ground data. However, the
correlation decays very fast, when
the satellite passes further away
from the station longitude.
Statistically significant correlation
lengths of ~± 15º is observed in
Indian and American sectors.
03.07.2007 | 9:30 am | IUGG | page 17/25
GUA
1
0.8
0.6
0.4
0.2
0
-0.2
1
0.8
0.6
0.4
0.2
0
-0.2
1
0.8
0.6
0.4
0.2
0
-0.2
ETT-HYB
-40 -20
0
20
40
TIR-ABG
-40 -20
0
20
40
HUA-FUQ
-40 -20
0
20
40
1
0.8
0.6
0.4
0.2
0
-0.2
1
0.8
0.6
0.4
0.2
0
-0.2
1
0.8
0.6
0.4
0.2
0
-0.2
AAE-QSB
-40 -20
0
20
40
MBO-GUI
-40 -20
0
20
40
GUA-CBI
-40 -20
0
Without Sq correction
With Sq correction
Manoj et al, Evidence for short ....
20
40
Low correlation
Is the observed variability in EEJ current strength due to spatial (23º) or
temporal (93 minutes) effects ?
From our ground/satellite comparison performed at various longitude separations
we may conclude that this is primarily a spatial effect
Reason ?
The driving electric fields has large spatial scales (~ 30º)
Since we have excluded the electric field, the conductivity may be responsible for
the short-range coherence of the EEJ.
A promising candidate for local conductivity modulation is plasma instability within
the Cowling channel.
Implications ?
03.07.2007 | 9:30 am | IUGG | page 18/25
Manoj et al, Evidence for short ....
Sq and EEJ
Without Sq correction
1
Madras
PND
Correlation coefficient
HYB
0.8
0.6
0.4
0.2
PND-HYB
ETT-PND
ETT
SRI LANKA
80

0
-40
-20
0
20
40
Distance from the observatory in degrees
With Sq correction
E
Correlation coefficient
1
0.8
0.6
0.4
0.2
0
-40
03.07.2007 | 9:30 am | IUGG | page 19/25
ETT-PND
PND-HYB
-20
0
20
40
Distance from observatory in degrees
Manoj et al, Evidence for short ....
Sq and EEJ
GUI
1
0.8
0.6
0.4
0.2
0
-0.2
-40
MBO
-20
0
20
40
Distance from the observatory in degrees
10 W
03.07.2007 | 9:30 am | IUGG | page 20/25
Manoj et al, Evidence for short ....
Sq and EEJ
The uncorrelated variations in the Sq and EEJ signals show that the
temporal variations of EEJ and Sq are decoupled.
Reason ?
A possible cause for the latitudinally very confined variations of the EEJ can be the
penetrating electric field associated with DP2 fluctuations (e.g. Kikuchi et al., 1996,
2000).
The amplitude of these magnetic signatures is at dip-latitudes sometimes 10 times
larger than at stations outside the Cowling channel (see Kikuchi et al., 1996, Fig. 2).
The Sq system, on the other hand, is driven primarily by tidal winds which do not show
short-period variations
Implications ?
Monitoring of EEJ should be done with the reference observatory 4° to 5° apart
from the dip latitude
>> ExB drift monitoring
03.07.2007 | 9:30 am | IUGG | page 21/25
Manoj et al, Evidence for short ....
Summary of correlation analysis
Station Pair
CC without Sq
correction
CC with
Sq
correctio
n
Distance between
the station pair
(degrees)
ETT-HYB
0.93
0.94
10.26
TIR-ABG
0.94
0.94
13.4
HUA-FUQ
0.8
0.76
16.47
AAE-QSB
0.69
0.56
29.94
MBO-GUI
0.51
-0.02
18.45
GUA-CBI
0.163
-0.12
14.6
ETT-PND
0.97
0.97
3.35
PND-HYB
0.53
0.30
6.91
03.07.2007 | 9:30 am | IUGG | page 22/25
Manoj et al, Evidence for short ....
Conclusions
Combined analysis of satellite and ground magnetic data gave new insights on
the noon-time EEJ.
The uncorrelated EEJ current strengths observed by CHAMP in its successive
passes are caused by short longitudinal correlation lengths of EEJ. A suggested
reason is the conductivity discontinuities in the Cowling channel due to plasma
instabilities
The uncorrelated variations in the Sq and EEJ signals show that the temporal
variations of EEJ and Sq are decoupled. Possibly, the penetrating electric fields
from high latitude regions are responsible for the uncorrelated, short period
fluctuations of current strength in EEJ
Satellite data along with data from a dedicated, a dense NS magnetometer array
near geomagnetic dip-equator would be ideal to further probe EEJ
03.07.2007 | 9:30 am | IUGG | page 23/25
Manoj et al, Evidence for short ....
Satellite data.
The operational support of the CHAMP mission by the German Aerospace
Center (DLR) and the financial support for the data processing by the Federal
Ministry of Education and Research (BMBF) are gratefully acknowledged
Observatory data.
Organization / Institute
Country
Observatories
Instituto Geográfico Agustín Codazzi
COLOMBIA
FUQ
Addis Ababa University
ETHIOPIA
AAE
Institut Français de Recherche Scientifique pour le
Développement
FRANCE
MBO
Indian Institute of Geomagnetism
INDIA
ABG, PND, TIR
National Geophysical Research Institute
INDIA
ETT, HYB
Japan Meteorological Agency
JAPAN
CBI
National Centre for Geophysical Research
LEBANON
QSB
Instituto Geográfico Nacional
SPAIN
GUI
US Geological Survey
UNITED STATES
GUA
Instituto Geofisico del Peru
PERU
HUA
03.07.2007 | 9:30 am | IUGG | page 24/25
Manoj et al, Evidence for short ....