Download Properties and the origin of Almost Monoenergetic Ion (AMI) beams

Ann. Geophys., 29, 1439–1454, 2011
www.ann-geophys.net/29/1439/2011/
doi:10.5194/angeo-29-1439-2011
© Author(s) 2011. CC Attribution 3.0 License.
Annales
Geophysicae
Properties and the origin of Almost Monoenergetic Ion (AMI)
beams observed near the Earth’s bow shock
V. N. Lutsenko and E. A. Gavrilova
Space Research Institute RAS, Profsoyuznaya 84/32, Moscow, Russia
Received: 26 November 2010 – Revised: 15 July 2011 – Accepted: 1 August 2011 – Published: 29 August 2011
Abstract. Beams of Almost Monoenergetic Ions (AMI) in
the energy range from 20 to 800 keV were discovered in the
DOK-2 experiment (Interball project) in the magnetosheath
and upstream of the Earth’s bow shock. This work summarizes the analysis results of ∼730 AMI events registered in
1995–2000. Statistics of AMI properties, their nature and
origin are considered. The analysis of a large array of new
data confirmed our earlier suggested ideas on the AMI nature, origin, and their acceleration model. These ideas were
further developed and refined. According to this model, AMI
are a result of solar wind ions acceleration in small regions
with a potential electric field arising due to disruptions of the
bow shock current sheet filaments. It has been found that the
reason of the current filaments disruptions in most cases was
the Hot Flow Anomaly phenomenon (HFA) caused by an interaction of a tangential discontinuity in the solar wind with
the Earth’s bow shock. It is shown that the study of AMI
can provide new information on large-scale properties and
dynamics of the bow shock current sheet.
Keywords. Magnetospheric physics (Current systems; Solar wind-magnetosphere interactions) – Space plasma
physics (Charged particle motion and acceleration)
1
Introduction
Ions and electrons accelerated in the near-Earth plasma in
the energy range of 20 keV to 1000 keV usually have smooth
spectra with a negative slope. Particle acceleration is commonly obliged to inductive electric fields emerging as a consequence of magnetic fields variations in the magnetosphere
and on its borders. Broad humps can arise in spectra when
Correspondence to: V. N. Lutsenko
([email protected])
magnetospheric energetic particles gain additional acceleration near the quasi-perpendicular bow shock (Anderson,
1981). Rather narrow peaks can be observed by particles
propagation due to time-of-flight or gradient-drift dispersions. Energies of such peaks which were cut out from the
original smooth spectrum vary with time and an observation
site (Lutsenko et al., 2005, 2008). Almost monoenergetic
electrons with energies of 1 to 10 keV were often observed
in the auroral zone. They possibly have obtained their energy after passing electrostatic double layers (see e.g. Alfvén,
1977, and references therein). However, in the very sources
of energetic ions and electrons in the energy range of 20–
1000 keV, we always had to deal with smooth spectra, often
having power-law or exponential forms.
In 1995–2000, the DOK-2 experiment (Lutsenko et al.,
1998) was carried out within the framework of the Interball
project on both Tail (Interball-1) and Auroral (Interball-2)
spacecrafts (S/C). Interball-1 was launched on 8 March 1995
and worked until 16 October 2000; its orbit had an apogee of
31.14 RE , a perigee of 1.16 RE , an orbital plane inclination to
the equator of 62.8◦ , and a period of 4 days. The distinctive
characteristics of the DOK-2 spectrometer were record high
energy and time resolutions. The instrument used 56-channel
logarithmic analyzers to measure ion spectra in the range of
20–800 keV and electron spectra in the range of 25–400 keV.
It allowed to discover in the Earth’s magnetosphere and near
its borders the beams of Almost Monoenergetic Ions (AMI)
(Lutsenko and Kudela, 1999; Lutsenko, 2001), Fine Dispersion Structures (FDS) (Lutsenko et al., 2005, 2008), and a
number of other phenomena (see e.g. Lutsenko et al., 2002),
which could not be discovered earlier because of an insufficient energy resolution of the used instruments.
AMI were named so because their spectra contain 2 to 3
lines having the Gaussian shape, with relative values of full
width at half maximum (FWHW) for the first line dE1/E1
from 0.11 to 0.39 with a mean value of 0.25. In the most
general case we have observed events with 3 lines having
Published by Copernicus Publications on behalf of the European Geosciences Union.
1440
V. N. Lutsenko and E. A. Gavrilova: Properties and the origin of AMI beams
energy ratios of 1:2:(5–6). In events with lower intensities
or with a higher background, only the first two lines with
the energy ratio 1:2 were observed. Note that in energetic
electron spectra measured with the same resolution narrow
lines were observed only in the dispersion events (Lutsenko
et al., 2005, 2008) associated with the acceleration processes
in the plasma sheet of the magnetospheric tail. Except for
two of our papers (Lutsenko and Kudela, 1999; Lutsenko,
2001) there have been no other publications on similar ion
spectra until 2009. The first such publication appeared in
2009 (Klassen et al., 2009). The authors have used data from
SEPT and SIT instruments (project STEREO). AMI were observed not only upstream and close to the Earth’s bow shock,
but also at distances up to 1900 RE .
In our previous papers we have published results based on
the analysis of the limited number of AMI events and suggested a hypothesis about their nature and origin. According
to this hypothesis AMI are a result of solar wind ions acceleration by bursts of strong potential electric fields in small parts
of current sheets in the magnetosphere and on its boundaries.
These bursts arise due to disruptions of filaments forming
these electric current sheets. After such disruption the electromotive force of the circuit (EMF) will be applied to a narrow disruption interval, creating a potential electric field with
a magnitude up to 0.1 V m−1 . For the neutral current sheet
in the magnetotail, this process was observed directly at the
acceleration place (Lutsenko et al., 2008). As to the magnetosheath and the region upstream of the bow shock where we
had most of our AMI observations, a much greater number
of events should be analyzed.
This article summarizes the results of study of the ∼730
AMI events observed during the 5 years (1995–2000) in the
solar wind (SW) and in the magnetosheath (MSH) near the
Earth’s bow shock (BS). Their properties, nature and origin
are clarified. The earlier proposed model of AMI acceleration should be confirmed by the results of all the AMI events
analysis, further developed and refined. In particular, the
aims are:
1. To build the statistics of AMI line properties using the
full set of our data, in particular for the ratio of numbers
of accelerated protons and alpha-particles.
2. To estimate angular distributions of AMI beams.
3. To find a confirmation of our assumptions regarding nature of AMI lines in the results of other experiments.
4. To check our assumption on the mechanism of AMI acceleration by comparison of our observations with some
consequences of the current layer filaments disruption
model.
5. To make a calculation of ion trajectories in the acceleration region for estimation of its dimensions.
Ann. Geophys., 29, 1439–1454, 2011
6. To find which current sheet is responsible for generation
of AMI observed in the magnetosheath and upstream of
the bow shock.
7. To find the reasons for current disruptions.
2
Special features of DOK-2 that made possible the discovery and study of AMI. Confirmation of our hypothesis regarding nature of the lines in AMI spectra
As it was noted in Sect. 1, the discovery and study of AMI became possible owing to the record high energy and time resolutions of the DOK-2 instrument (Lutsenko et al., 1998). The
device was equipped with four telescopes for measuring parameters of energetic ions in the range of 20–800 keV (1p and
2p) and energetic electrons in the range of 25–400 keV (1e
and 2e). All telescopes used passively cooled Si-detectors
with instrumental energy resolution dE1ins = 7–8 keV and
logarithmic pulse height analysers with 56 channels. The ion
telescopes had aperture angles of 12.5◦ , the electron ones of
27◦ . Geometric factors of the telescopes were of 0.015 cm2 sr
for ion telescopes and of 0.066 cm2 sr for electron ones. The
axes of 1p and 1e telescopes were directed along the spin axis
of the S/C in the antisolar direction. The 2p and 2e telescopes
were oriented in the solar hemisphere at an angle of 62◦ to
the S/C’s spin axis directed toward the Sun. A spacecraft spin
period was ∼2 min.
The discovery and study of AMI was made possible not
only because of the high energy resolution, but also through
the use of a special adaptive measurement algorithm. The
basis for the formation of the DOK-2 output data were Basic Measurements (BM) – complete 56-channel spectra measured every second. However, their transfer to ground receiving stations was possible only during the rare and short
sessions of direct transmission since capacity of the onboard
storage device was insufficient. To overcome this problem,
two types of output information complementing each other
were formed from BMs for each telescope:
1. Complete 56-channel spectra obtained by summing up
individual BMs until any of the channels gets a certain number of particles Nmax or accumulation time dT
reaches a certain value Tmax . These parameters could be
changed by commands, but most often Tmax = 1280 ◦ C
and Nmax = 10 240. As a result, at high particle fluxes
the spectrum accumulation time was reduced to 2–5 s,
and at low fluxes it increased, but not more than Tmax.
2. Three parameters of temporal variations: TP1, TP2, and
TP3 – each representing a sum of 4 consecutive spectrum channels. The accumulation time for TPs varied
from 1 to 260 s and was determined by a special algorithm which ensured a reduction of the output information volume up to 10–100 times, and at the same time
retaining the ability to record all statistically significant
variations in particle numbers in the TP’s energy ranges.
www.ann-geophys.net/29/1439/2011/
V. N. Lutsenko and E. A. Gavrilova: Properties and the origin of AMI beams
1441
Fig. 2. Positions of Interball-1, Geotail and Wind S/C during the
AMI event on 16 April 1996. Arrows show magnetic field directions on first two S/C.
Fig. 1. An example of mutual complementarity of the full ion spectrum and the time profiles parameters TP. At the top – a full ion
spectrum from 1p-telescope, measured in the solar wind and containing 2 AMI lines. Color markings show energy intervals of 3
TP-parameters. Here and below, Te corresponds to the end of the
spectrum accumulation interval, dT to the accumulation duration.
At the bottom – graphs of TP-parameters and the values of corresponding energies. For comparison, the spectrum accumulation
time interval is also shown: dT = 353 s.
Both of these types of output data are mutually complementary and proved to be very useful for searching and analyzing the data on AMI. Figure 1 gives an example of the
AMI spectrum and corresponding TP-parameters measured
in the solar wind on 12 March 1996 (1p-telescope). Three
parts of the spectrum highlighted in blue, green, and red correspond to TP1, TP2, and TP3, respectively. The intensity
of the AMI containing 2 lines was rather weak at a relatively
high level of background ions with a usual smooth spectrum.
Thus, the third line is almost invisible. As a result, the accumulation time for the spectrum turned out to be rather long
and reached 353 s. The energies E and intensities J at the
maxima of the lines as well as their Full Width at Half Maximum (FWHM) dE were determined by the Gaussian fitting
(blue dashed lines and legends). The figure shows that the
TP1 and TP2 intervals fall on the smooth background spectrum, while TP3 falls upon the first AMI line. Figure 1 (botwww.ann-geophys.net/29/1439/2011/
tom graphs) shows time variations of TP1–TP3, the interval of the spectrum accumulation time and the energy values for TP1–TP3. Around 01:28:19 UT the TP3 intensity increases sharply and during the next 13.7 s exceeds the initial
level by 46 times. Then, through 81.5 s after the increase it
falls to approximately the initial level corresponding to the
normal spectrum (TP3 is below TP2 and TP1). Thus, TPparameters allow to determine within 1 s accuracy the start
time (01:28:19 UT) and duration (81.5 s) of the AMI event,
which cannot be found only from the given spectrum in this
case.
Although DOK-2 measured the energy spectra of all ions,
doing so without charge and mass separation, the noted
above relations of line energies allowed us to suppose that
we deal with solar wind ions H+1 , He+2 , and the ion group
(C, N, O)+(5−6) accelerated in some spatially confined potential electric field to energies proportional to their charges
(Lutsenko and Kudela, 1999). We obtained a confirmation
of this hypothesis in 2002, after a private communication of
A. Lui and S. Nylund who sent us the detailed data from
the EPIC instrument (Geotail) for several time intervals when
Interball-1 and Geotail had been in the same physical region
close enough to each other. The EPIC device (Williams et
al., 1994) had ∼8 times lower energy resolution but was able
to measure separately the spectra of protons, alpha-particles,
and ions of the (C, N, O) group. On 16 April 1996, two
S/C were in the MSH near the quasi-perpendicular BS (see
Ann. Geophys., 29, 1439–1454, 2011
1442
V. N. Lutsenko and E. A. Gavrilova: Properties and the origin of AMI beams
Table 1. Distribution of a total number of AMI events with 2–3
lines among DOK-2 telescopes and measurement regions.
3
Fig. 3. Comparison of ion spectra measured on Interball-1 and Geotail in the MSH close to the BS during the AMI event on 16 April
1996. For DOK-2 (Interball-1) spectrum blue dashed lines and legends give results of Gaussian AMI lines fitting.
Fig. 2). The directions of the magnetic field on both satellites
were approximately the same, so that both S/C were close to
a magnetic connection.
Figure 3 compares the spectrum of all ions measured by
DOK-2 (INTERBALL-1) with spectra of protons, alphaparticles, and the (C, N, O) group measured by EPIC (Geotail). It is evident that 3 lines in the DOK-2 spectrum correspond to the peaks of protons, alpha-particles, and of the
(C, N, O) group in EPIC spectra. Ratios of the energies and
intensities of 3 peaks in EPIC spectra are very close to the
corresponding ratios in the DOK-2 spectrum.
After our first publications (Lutsenko and Kudela, 1999;
Lutsenko, 2001), AMI were also found by Klassen et al.
(2009) in the SEPT experiment (project STEREO) at distances up to 1100–1900 RE from the Earth. They present
one more example of the identification of the first two AMI
lines as ions of H+1 and He+2 .
Ann. Geophys., 29, 1439–1454, 2011
Telescope
SW
MSH
Total
1p
2p
Total
314
187
501
38
183
221
352
370
722
Where AMI events were observed
AMI were observed by DOK-2 (Interball-1) mainly in the
solar wind (SW) and magnetosheath (MSH) near the Earth’s
bowshock (BS), in all about 730 events. For our detailed
analysis, the 722 most distinct AMI events with 2–3 lines in
the spectrum were selected.
Table 1 shows how the number of observations is distributed between the 2 regions and 2 telescopes. Figure 4
shows the distribution of observation points in these regions.
The extent of the observation region in the solar wind was
limited by the S/C apogee. Red dots correspond to the measurements, with the 1p-telescope looking in the antisolar direction. Blue dots correspond to the measurements with the
2p-telescope directed toward the solar hemisphere at an angle of 62◦ to the spin axis of the S/C. Distributions in Fig. 4
and Table 1 show that AMI propagating towards the Sun
(1p) predominate in the SW, while in the MSH predominate AMI propagating from the BS (2p). It indicates that the
AMI sources are located most probably on the BS, but not on
the magnetopause, as it was supposed previously (Lutsenko,
2001).
Over the 5 years of Interball-1 operation, the total DOK-2
observation time reached 18 958 h in areas where these 722
AMI-events were observed. Thus, the average frequency of
AMI-events observations in the SW and MSH was 0.038
per hour or 0.91 AMI-events per day. Since the observations at any moment were carried out only at one point, and
AMI propagate as narrow beams, as we will shown further in
Sect. 5, the real frequency of the AMI-events must be much
higher.
AMI were also observed in the magnetotail plasma sheet
where observations were complicated due to high-intensity
background of energetic particles with the usual smooth
spectra. These particles were generated together with AMI
by the geotail current sheet filaments disruptions (Lutsenko
et al., 2008). These AMI are not considered here.
4
Statistics of AMI properties
Figure 5 presents a histogram of AMI events durations. The
exact determination of some specific event duration is not
always possible, because sometimes two events overlap or
follow directly one after another. When the data are obtained
www.ann-geophys.net/29/1439/2011/
V. N. Lutsenko and E. A. Gavrilova: Properties and the origin of AMI beams
1443
Fig. 5. The histogram of AMI event durations 1T .
Fig. 4. Observation positions for 722 AMI events with 2–3 lines
upstream of the BS and in the MSH (August 1995–August 2000).
Table 2. Statistical parameters of AMI beams (see Figs. 5–7).
Parameter,
units
Duration, s
E1, keV
E2/E1
E3/E1
dE1corr /E1
Sα /Sp
Nα /Np
Total numb.
of values
Min.
value
Max.
value
Mean
value
Std.
deviation
204
722
722
114
231
154
124
17
25.8
1.52
3.68
0.107
0.34
12.95
294
409.9
2.91
7.13
0.393
66.77
91.98
104.7
94.6
2.034
5.25
0.253
12.35
37.26
61.12
59.56
0.158
0.698
0.058
13.24
18.77
from the 2p-telescope, the AMI beam often leaves the telescope field of view before the event ends because of the S/C
rotation. Therefore, we used only the data from 1p-telescope
and solely for those 204 AMI-events whose durations could
be determined accurately enough. The statistical parameters
of this and other distributions presented in this section are
listed in Table 2.
www.ann-geophys.net/29/1439/2011/
Fig. 6. Histograms for E1 (a), E2/E1 (b), E3/E1 (c) and
dE1cor /E1 (d) (see text for details).
Figure 6 gives the histograms of values of the first line
energy E1, of line energies ratios E2/E1, E3/E1, and of
relative values of the first line FWHM dE1cor /E1. In the
last case, from the measured value dE1
q the instrument resolution dE1ins was excluded: dE1cor = dE12 − dE12ins . The
dE1ins values were regularly measured during the in-flight
instrument calibrations. The absolute intensity J1 in the maximum of the first line vary widely in different events, from
0.3 to 1.0 × 105 cm−2 s−1 sr−1 keV−1 .
Ann. Geophys., 29, 1439–1454, 2011
6
1444
V.N. Lutsenko and E. A. Gavrilova:
Propertiesand
andE.the
of Almost
Monoenergetic
Ionof(AMI)
beams
V. N. Lutsenko
A. origin
Gavrilova:
Properties
and the origin
AMI beams
and corrected FWHM as E p , J p , and dE p , and similarly, for
the 2nd line (alpha-particles). Figure 7a shows the distribution of line areas ratio of protons S p to alpha-particles S α
in the AMI spectra. Since the shapes of all the lines do not
practically differ (Gaussians), we accept that
S p J p · dE p
=
,
S α Jα · dEα
Figure 7b gives the distribution for ratio of densities protons and alpha particles N p /Nα in the solar wind plasma during the AMI measurements (data from the Wind S/C).
7. (a) Histogram for the ratio of proton and alpha-particles
ToFig.
compare
these ratios let’s calculate S p /S α starting from
line areas (Sp /Sα ) in AMI spectra. (b) Histogram
for the ratio of
the Nprotons
p /Nα :and alpha-particles densities in the solar wind (N /N ).
p
α
N p · V p · dE p
(S p /S α )calc =
,
· Vα ·section,
dEα we will denote the parameters of
Further N
inα this
first
line
energy
and intensity
the maximum
herethe
V p /V
the(protons):
ratio of ion
velocities
after theatacceleration.
α is
√ similarly for
and
corrected
FWHM
as
E
,
J
,
and
dE
and
p V p /Vα =
p
As E p /Eα =0.5, dE p /dEα =0.5p and
2 we obtain
the calculated
2nd line (alpha-particles).
for the
value of S p /S α :Figure 7a shows the distribution of line areas ratio of protons Sp to alpha-particles Sα in
√
the AMI spectra.
N p · 2Since the shapes
N p of all the lines (Gaussians)
(S p /S
=
= 0.707
.
(1)
α )calcdiffer,
hardly
we accept
that·
Nα · 2
Nα
Sp
Jp · dEp
The average
value
=
, for N p /Nα is equal to 37.26 (TaSα Thus,
Jα · dEfor
α
ble 2).
the ratio (S /S )
we should have
p
α calc
(S p /S
=37.26
0.707
= 26.34.for ratio
In reality
we protons
have
Figure
gives ·the
distribution
of densities
α )calc7b
alpha
particles
N
/N
in
the
solar
wind
plasma
during
S p /Sand
=12.35
(Table
2),
i.e.
it
is
2.1
times
lower.
This
difp
α
α
the AMI
measurements
the Wind model,
S/C). which
ference
is explained
by our(data
AMIfrom
acceleration
To advantage
compare these
ratios let’s calculate
/Sα starting
from
gives an
to alpha-particles
overSpprotons
by mothe
N
/N
:
p lines
α
noenergetic
creation (see Sec. 7).
Np · Vp · dEp
(Sp /Sα )calc =
,
5 Angular
distributions
Nα · Vα · dEof
α AMI
where
Vp /Vtelescopes
of ion
accelerα is the ratio
While
DOK-2
could
notvelocities
examineafter
the the
whole
4π
√
ation.
As
E
/E
=
0.5,
dE
/dE
=
0.5,
and
V
/V
= 2,
p
α
p
α
p
α
solid angle, in some cases we were able to estimate angular
we obtainofforAMI
the calculated
value
of Swith
:
p /Sαdistributions
distributions
and compare
them
of
√
ions with usual, smooth
spectra.
As
a
first
example
here
are
Np · 2
Np
(Sresults
= 0.707
.
given
for=the AMI event
on ·23 April
1999 in the MSH,(1)
p /Sα )calc
Nα · 2
N
close to the BS. Figure
8 presents anα ion spectrum from the
The average
value
for Nintervals
equal to 37.26
(Table
p /Nα is correspond
2p-telescope,
where
3 color
to the
3 TP-2).
Thus,
for
the
ratio
(S
/S
)
we
should
have
(S
/S
)calc =
p
α
p
α
calc
parameters. One can see that TP1 is a part of a usual smooth
37.26
·
0.707
=
26.34.
In
reality
we
have
S
/S
=
12.35
(Tap
α
background spectrum, while TP2 and TP3 correspond to the
ble
2),
i.e.
it
is
2.1
times
lower.
This
difference
is
explained
first two AMI lines. As mentioned in Sect. 2, time variations
by our AMI acceleration
model,with
which
gives an advantage
of TP-parameters
were measured
a variable
time reso-to
alpha-particles
over
protons
by
monoenergetic
lines
lution depending on variability of counts in 1-second creation
BM’s
(see Sect. 7).
in the TP1 and TP3 energy intervals. Energy values for the
TP intervals are shown in Fig.9 (upper panel). When the variability
high this
resolutionof
reached
5 was
Angular
distributions
AMI 1 s. The 2p-telescope
with the angular aperture of 12.5◦ was directed at the angle
While
DOK-2
could
notSun.
examine
theS/C
whole
4π
62◦ to
the S/C
spintelescopes
axis pointed
to the
By the
rotasolid
angle,
in
some
cases
we
were
able
to
estimate
angular
tion the telescope scans the corresponding cone what allows
distributions
of AMI
and compare
to estimate
the angular
distributions
for them
AMI with
(TP2,distributions
TP3) and
of
ions
with
usual,
smooth
spectra.
As
a
first example, we
for a background spectrum (TP1).
Figure 9 shows time profiles for the three TP-parameters,
as well
for magnetic
field and angles
Ann.asGeophys.,
29, 1439–1454,
2011 of an ion flux unit
Journalname
Fig. 8.8.The
Figure
TheDOK-2
DOK-2ion
ionspectrum
spectrumininthe
theMSH
MSHonon2323April
April1999,
1999,
11:33:21–11:35:21
11:33:21
- 11:35:21 UT.
UT.
present here results for the AMI event on 23 April 1999 in
the MSH, close to the BS. Figure 8 presents an ion spectrum
from the 2p-telescope, where 3 color intervals correspond to
the 3 TP-parameters. One can see that TP1 is a part of a
usual smooth background spectrum, while TP2 and TP3 correspond to the first two AMI lines. As mentioned in Sect. 2,
time variations of TP-parameters were measured with a variable time resolution depending on variability of counts in 1-s
BMs in the TP1 and TP3 energy intervals. Energy values for
the TP intervals are shown in Fig. 9 (upper panel). When
the variability was high, this resolution reached 1 s. The 2ptelescope with an angular aperture of 12.5◦ was directed at
the angle 62◦ to the S/C spin axis pointed toward the Sun.
By the S/C rotation, the telescope scans the corresponding
cone, which allows to estimate the angular distributions for
AMI (TP2, TP3) and for a background spectrum (TP1).
Figure 9 shows time profiles for the three TP-parameters,
as well as for magnetic field and angles of an ion flux unit
vector in a local magnetic field (LMF) frame: a pitch angle
(PA) and an azimuth angle (AA). LMF is a rectangular coordinate system in which Z LMF = B/|B|, Y LMF = (Z LMF ×
XGSE ), and X LMF = (Y LMF × Z LMF ). The pitch angle PA is
the angle between the ion flux direction and the axis Z LMF ,
the azimuth angle (AA) corresponds to the rotation around
the ZLMF axis. All the data were averaged in time intervals
of TP measurements. The time interval for the spectrum from
Fig. 8 is shown in the upper panel. The vertical lines in Fig. 9
correspond
to variations
the times of
of TP-parameters,
maximum intensity
of AMI
Figure
9. Time
magnetic
field (solid
and ion
lines)
and
of
ions
with
the
usual
spectrum
(dashed
lines).
The
flux angles in LMF frame for the AMI event in the MSH near
the
values
of pitch
azimuth
angles
these times are shown
Bow
Shock
on 23and
April
1999 (see
text for details).
in the lower frame. Time intervals between these maxima
correspond to ∼2 min spin period of the S/C. The intensity
vector in a local magnetic field (LMF) frame: a pitch angle (PA) and an azimuth
angle (AA). LMF is a rectanwww.ann-geophys.net/29/1439/2011/
www.jn.net
ation.
btain
this value includes also the telescope aperture (12.5 ), the in
trinsic FWHM of the AMI beam must be no more than 10-15
Figure 8. The DOK-2 ion spectrum in the MSH on 23◦April
. 1999,
V. N. Lutsenko and E. A. Gavrilova: Properties and the origin of AMI beams
1445
11:33:21 - 11:35:21 UT.
(1)
(Tahave
have
s difwhich
mo-
le 4π
gular
ns of
e are
MSH,
m the
3 TPmooth
o the
tions
resoBM’s
or the
variscope
angle
rotallows
) and
eters,
x unit
Fig. 9.9.Time
ofof
TP-parameters,
magnetic
field,
and
ionion
Figure
Timevariations
variations
TP-parameters,
magnetic
field
and
fluxangles
anglesinin LMF
LMF frame
frame for the
flux
the AMI
AMIevent
eventininthe
theMSH
MSHnear
nearthethe
BowShock
Shockon
on23
23 April
April 1999 (see
Bow
(see text
textfor
fordetails).
details).
Fig. 10. Angular distributions for the AMI event in the MSH on 23
Figure 10. Angular distributions for the AMI event in the MSH on
peaks in
foraAMI
narrowerfield
than (LMF)
for the usual
spectrum.
vector
localaremagnetic
frame:
a pitch A
an- April 1999 at 11:31:00–11:37:00 UT.
more detailed analysis showed that the FWHM of these
peaks
23.04.1999
gle (PA) and an azimuth angle (AA). LMF is ◦a rectan- at 11:31:00-11:37:00 UT.
for AMI corresponds to intervals of PA and AA < 20 . Since
this value also includes the telescope aperture (12.5◦ ), the
period. In the upper panel the time intervals for spectra ]2–
intrinsic FWHM of the AMI beam must be no more
than
]7 are
shown. Thin
vertical lines inof
Fig.
12 correspond
to the distribu
www.jn.net
Figure
10
presents
a summary
relative
angular
10–15◦ .
flux maxima of ions with a usual spectrum (TP1, TP2, dashed
tions for 3 lines)
TP-parameters
in the
considered
time inter
and for AMI (TP3,
solidwhole
lines). They
allow to find
Figure 10 presents a summary of relative angular distrithe pitch
azimuth
for these moments
(seeusual
bottomspectrum
butions for 3 TP-parameters in the whole considered
val. time
One can
seeand
that
the angles
distribution
for the
panel). The maxima of TP3 (AMI) fall in the pitch angles
interval. One can see that the distribution for the usual spec(the upper from
diagram)
is broader and the azimuth angles for th
63◦ to 66◦ and in two ranges of the azimuth angles:
trum (upper diagram) is broader and the azimuth angles for
◦
◦
◦ to 234◦ . The maxima for TP1
◦
◦
and 227
133 to 140differ
strongly
for AMI (133◦ toand140◦ ) and
the intensity maxima differ strongly for AMI (133intensity
to 140 ) maxima
◦
◦
TP2 (usual spectrum) fall◦ at the pitch
◦ angles 107◦ to 116◦
and for the usual spectrum (243 to 249 ).
for the usual
spectrum
(243
to
249
).
◦
and azimuth angles from 247 to 248◦ . The difference in
In Figs. 11–13, one more example of AMI observation is
In Fig. these
11-13,
more
ofillustrated
AMI observation
i
anglesone
for both
typesexample
of spectra is
by the
presented, which took place on 16 April 1996 at 18:35:43–
diagrams in Fig. 13.
18:45:35 UT in the MSH close to the BS. In Fig. 11presented
as well as
which
took place on 16 April 1996 at 18:35:43
It follows from the discussed examples that the angular
in Fig. 8, three colors mark energy intervals corresponding to
in the for
MSH
tonarrow
the BS.
In Fig.
distributions
AMI close
beams are
(FWHM
< 15◦ )11
andas well a
the three TP-parameters. Here the first lines of AMI18:45:35
(with the UT
differ
strongly
from
the
distributions
for
ions
with
usual
specexception of the spectrum ]3) fall into the energy range
of
the
in Fig. 8 three
colors mark energy intervals corresponding to
tra. These facts point to small dimensions of the acceleration
TP3. It also allowed the comparison of angular distributions
the(TP1,
three TP-parameters.
Here
the
first lines of AMI (with th
region and short duration
of the
process.
for AMI and for ions with the usual smooth spectrum
TP2). Periodical (2 min) increases of ion intensities (Fig. 12,
upper panel), as in the previous case, correspond to the spin
www.jn.net
www.ann-geophys.net/29/1439/2011/
Ann. Geophys., 29, 1439–1454, 2011
differ strongly from the distributions for ions with usual spectra. These facts point to small dimensions of the acceleration
region and short duration of the process.
1446
V. N. Lutsenko8and E. A. Gavrilova:V.N.
Properties
and and
the origin
AMI beams
Lutsenko
E. A. of
Gavrilova:
Properties and
SH on
ribuntertrum
or the
) and
on is
:43 ell as
ng to
h the
Figur
16.04
Time
variations
of TP-parameters,
magnetic
and
Fig. 12.12.Time
variations
of TP-parameters,
magnetic
field field
and anFig. 11.
spectra
(2p-telescope)
in the the
MSH
nearnear
the BS
Figure
11.AMI
AMI
Spectra
(2p-telescope)in
MSH
the on
BS16
on Figure
gles of
ion
fluxes
ininLMF-frame
forfor
AMI
event
on on
16 16
April
1996.
angles
of
ion
fluxes
LMFframe
AMI
event
April
1996.
April
1996
at
18:35:53–18:45:35
UT
April 16, 1996 at 18:35:53 - 18:45:35 UT
6
The change of energies and intensities of proton lines
after the AMI event start
Journalname
For the AMI events with high time resolution in spectra measurements (2 to 10 s), we have plotted and analyzed the time
dependence of energies and intensities of proton lines after
the start of the acceleration process. The results for 6 such
AMI events are given in Fig. 14. Evidently, the proton line
energy E1 consists of two parts: a constant one (A) and a second one (B) exponentially decreasing with time. In a frame
of our hypothesis, the first one, obviously, corresponds to the
electric circuit electromotive force (EMF), the second one to
the self-inductance EMF, arising after the circuit disruption
and depending on the circuit inductance (length). The time
T0 corresponds to the events start (J1 ≈ 0). The time dependence of E1 was approximated by the expression:
E1 = A + B · exp(−C · (T − T 0))
(2)
The values obtained for the parameters A, B, C, and T 0
are given in Table 3. A very fast increase of intensities J1
(blue points and lines in Fig. 14) immediately after the current disruption can be explained by a fast filling of an almost
“empty” disruption region by the MSH plasma (see Sect. 8).
Ann. Geophys., 29, 1439–1454, 2011
6 The
change
of energies
and line
intensities
of cannot
proton
A smooth
decrease
of the proton
energy here
after
thetime-of-flight
AMI event start
belines
a result
of the
dispersion during the propagation of particles with the usual spectra from the site of their
For
the AMI events
with high
time resolution
in spectra
meaacceleration
to the points
of registration.
Firstly,
in this case,
surements
10 s) we
andhave
analyzed
the untime
the energy(2Etochange
withhave
timeplotted
T would
happened
2 /2/(T − T )2 , where M is the mass
der
the
law:
E
=
M
·
S
dependence of energies and intensities
of proton lines after
g
of start
the particle,
S is the distance
to theThe
source,
andfor
Tg is
the
the
of the acceleration
process.
results
6 such
time
of
particles
generation.
This
dependence,
in
contrast
to
AMI events are given in Fig. 14. Evidently, the proton line
has
no
constant
component
and
has
a
quite
difthe
Eq.
(2),
energy E1 consists of two parts: a constant one (A) and a secferent
Secondly, the
energy ratio
alpha-particles
ond
oneshape.
(B) exponentially
decreasing
withoftime.
In a frame
and
protons
would
have
been
equal
to
4,
not
2
as in ourtoobof our hypothesis, the first one, obviously, corresponds
the
servations.
Ofelectromotive
course, our spectra
measured
not right
electric
circuit
force were
(EMF),
the second
one acceleration, but
with
a delay
depending
on time
of
toafter
the the
self-inductance
EMF,
arising
after
the circuit
disrupflight of protons to the observation point. Rough estimate of
tion and depending on the circuit inductance (length). The
these delays for the 6 events showed that they ranged from
time T0 corresponds to the events start (J1≈ 0). The time
1.5 to 24 s, which cannot substantially change the results of
dependence
of E1 was approximated by the expression:
our analysis.
E1 = A + B · exp(−C · (T − T 0))
(2)
valuesofobtained
for the parameters A, B, C and T 0 are
7The
Model
AMI acceleration
given in Table 3. A very fast increase of intensities J1 (blue
points
and lines
in Fig.
after
the current
disPreviously
on the
basis14)
of immediately
our preliminary
analysis
of AMI
ruption
explained
a fast filling
of an almost
’empty’
events can
we be
proposed
a by
hypothesis
(Lutsenko
and Kudela,
disruption region by the MSH plasma (see Sect. 8).
A smooth decrease of
the proton line energy here cannot
www.ann-geophys.net/29/1439/2011/
be a result of the time-of-flight dispersion during the propaJournalname
Figur
AMI
gatio
accel
the e
der th
partic
gener
const
ondly
have
our s
but w
obser
event
V. N. Lutsenko and E. A. Gavrilova: Properties and the origin of AMI beams
1447
Table 3. Parameters of the time dependence of the first line energy E1 near the acceleration start for six AMI events (see Fig. 14). T0 is in
seconds of the day.
Date
6 Jun 1996
9 Apr 2000
12 Apr 1996
10 Aug 1997
23 Apr 1997
20 Apr 1999
UT interval
Tel.
12:24:30–12:25:00
03:08:00–03:10:00
22:22:20–22:23:20
23:13:35–23:14:54
20:38:30–20:39:45
13:43:20–13:44:25
1p
1p
2p
1p
1p
1p
T0,
A,
B,
C,
s
keV
keV
1/s
44 672
11290
80 552
83626
74 319
49 402
26.65
55.84
22.5
46.2
48.9
80.43
34.82
32.89
29.74
45.3
22.8
59.47
0.082
0.058
0.04
0.035
0.06
0.09
s and the origin of Almost Monoenergetic Ion (AMI) beams
tric field of the SW. After the circuit disruption, a sum of
the EMF and the self-inductance EMF (B in Fig. 15) will be
applied to the short disruption interval W as a voltage difference 1V = A + B.
This assumption was entirely confirmed by the results of
our full analysis of the ∼730 AMI events, in particular:
1. by the positions of AMI observation points,
2. by the energy ratios of AMI lines corresponding to the
ratios of H, He, and the CNO-group charges,
3. by the average energy of the proton line (see Table 2),
which is not inconsistent with the average potential drop
across the magnetosphere due to the motional electric
field of the solar wind: E sw = −(V sw × B sw ),
4. by short durations of AMI events (on average 104.7 s),
5. by the character of time variations of proton line energies and intensities at the event start. It is this character
of voltage 1V change that can be expected by the disruption of the electric circuit, shown in Fig. 15, and
6. by the total absence of monoenergetic electrons in the
AMI events.
Fig. 13. Angular distributions for the AMI event in the MSH on 16
Figure 13.
Angular distributions for the AMI event in the
April 1996 at 18:36:12–18:44:33 UT.
16.04.1996 at 18:36:12-18:44:33 UT.
1999; Lutsenko, 2001) in which generation of AMI observed
near BS occurred in spikes of potential electric field arising as a result of BS current sheet filaments disruptions (see
Fig. 15). Like each current circuit, these filaments have an
electromotive force EMF (A in Fig. 15), a resistance R, and
an inductance L. The EMF is distributed along that part of
the circuit which is exposed to an action of a motional elecwww.ann-geophys.net/29/1439/2011/
Let’s examine now in more detail the acceleration process
and limitations on dimensions and other characteristics of the
acceleration region resulting from our model. Let us assume
that the Y-axis is directed along the current sheet filament
(along the potential electric field E), the X-axis is perpenMSH
on
dicular to the filament, and the Z-axis is directed along the
magnetic field. An ion trajectory in the crossed E and B
fields (E⊥B) is a cycloid, and acceleration of ions up to the
same energy level will be possible if the cycloid width Ymax
in the Y-direction is notably larger than the acceleration region dimension W in this direction. Let’s find the resulting
limitations for the W -value. The equation of ions motion in
the crossed electric E and magnetic B fields:
M·
dv Z · e
=
(v × B) + Z · e · E
dt
c
Ann. Geophys., 29, 1439–1454, 2011
ratios of H, He and the CNO-group charges,
Date
Figure
13. Angular
distributions
for the AMI event in the MSH on
UT interval
Tel. T0,
A,
B,
C,
3. by the average energy of the proton line (see Table 2),
is notProperties
inconsistent
the average
V. N. Lutsenko and E. A.which
Gavrilova:
andwith
the origin
of AMIpotential
beams drop
1p 44672 26.65 34.82 0.082
across the magnetosphere due to the motional electric
1p 11290 55.84 32.89 0.058
field of the solar wind: : E sw = −(V sw × B sw ) ,
s
keV keV 1/sUT.
16.04.1996 at 18:36:12-18:44:33
1448
6.06.1996
9.04.2000
12.04.1996
10.08.1997
23.04.1997
20.04.1999
12:24:30 - 12:25:00
03:08:00 - 03:10:00
22:22:20 - 22:23:20
23:13:35 - 23:14:54
20:38:30 - 20:39:45
13:43:20 - 13:44:25
2p
1p
1p
1p
80552
83626
74319
49402
22.5
46.2
48.9
80.43
field and
ril 1996.
change substantially the results of our analysis.
7 Model of AMI acceleration
29.74
45.3
22.8
59.47
0.04
0.035
0.06
0.09
4. by short durations of AMI events (on average 104.7 s),
5. by the character of time variations of proton line energies and intensities at the event start. It is this character
of voltage ∆V change can be expected by the disruption
of the electric circuit, shown in Fig. 15,
6. by the total absence of monoenergetic electrons in the
AMI events.
proton
Previously, on the basis of our preliminary analysis of AMI
events we proposed a hypothesis (Lutsenko and Kudela,
1999; Lutsenko, 2001) in which generation of AMI observed
near BS occurred in spikes of potential electric field arising
as a result of BS current sheet filaments disruptions (see Fig.
15). Like each current circuit these filaments have an electromotive force EMF (A in Fig. 15), the resistance R and
inductance L. The EMF is distributed along that part of the
circuit which is exposed to an action of a motional electric
field of the SW. After the circuit disruption a sum of the
EMF and the self-inductance EMF (B in Fig. 15) will be
applied to the short disruption interval W as a voltage differFig.∆V
14. =
Time
of energies E1 and intensities J 1 of AMI proton lines near the events start.
ence
A +variations
B.
ra meahe time
es after
6 such
ton line
d a seca frame
ds to the
nd one disruph). The
he time
on:
Figure 14. Time variations of energies E1 and intensities J1 of
AMI proton lines near the events start.
Here:
ωc is the cyclotron frequency, ωc = (Z · e · B)/M =
9.579 × 10−2 · Z · B/A in rad s−1 , Z · e is the ion charge in C,
M = A·mp is the ion mass in kg, B is the magnetic induction
in nT, the ratio e/mp = 9.579 × 107 C kg−1 . The solution of
Eq. (3) for an initial velocity v̄0 = 0:
gation of particles with the usual spectra from the site of their
Z ·e·E
Z ·e·E
x=
· sin(ω
· t −Firstly,
acceleration to the points of registration.
inc · t),
this case,
M · ωc
M · ωc2
the energy E change with time T would have happened unZ ·e·E
· (1 − cos(ω
c · t)) mass of the
der the law: E = M ·S 2 /2/(T −T yg)=2M, where
M-the
· ωc2
Figure 16. a) Proton trajectories in the acceleration region. D particle, S -the distance to the source,
T Kg-the
time
ofcurrent
particles
or the
denoting
=of
1.044
107 · (A
· E)/(Z
· B 2 ), in km:
thickness
the ×
disrupted
filament.
W - the width of
(2)
the
region
with
the
potential
electric
field.
D
effective
thickness
generation. This dependence in xcontrast
to the
Eq. 2, has no
= K · (ωc · t − sin(ω
(4)
c t)), y = K · (1 − sin(ωc t)),
of the acceleration region, starting from which all protons will obFigure
15.The
The
sketch,
illustrating
the process
of and
solar
ions a also
Fig. 15.
sketch
illustrates
the process
of solar
wind wind
ions
acd T 0 are
constant
component
has
quite
different
shape.
Sec-out of D limits,
tain
keV.
in the
km.same
The energy
cycloid200
width
is:Protons, starting
acceleration
and
AMI
generation
by
BS
current
sheet
filaments
disceleration and AMI generation by BS current sheet filaments diswill obtain the energy <200 keV. b) Comparison of D for alpha2 · Zand
· e · for
Eand
A Ewould
ruption.Here
Here,
force
(EMF),
the
ruption.
AA isis the
the circuit
circuit
electromotive
force
(EMF),
B-the
J1 (blue
ondly,
the electromotive
energy
ratio
ofB isalpha-particles
protons
particles
Ymax
=
=protons.
2.088 ×Dotted
107 · lines
·
,show
kmthe same trajectories
2
self-inductanceEMF.
EMF.Electrons,
Electrons, having
having much
will
self-inductance
muchlower
lowergyroradia,
gyroradia,
will
Z B 2region width W = 5000 km
· mp · ωvalue
for theA larger
of the acceleration
c
ent dishave
been
to 4, not 2 as in our
observations. Of course,
sweptout
out
bythe
thedrift
drift
withoutequal
the
bebe
sweep
by
without
theacceleration.
acceleration.
instead of W= 4000 km.
The values of the ωc and Ymax for B = 10 and 20 nT and
−1 are given
’empty’
our spectra were measured notEright
thein Table
acceleration,
= 0.1 V mafter
4.
This assumption
confirmed
by
the
results
of
Let’s
examine
now
in
more
details
thetheacceleration
With
Ey = E, Exwas
= Eentirely
=
0,
B
=
B
=
0,
B
=
B
it
So
for
the
acceleration
of
all
protons
up to
same en- proz
x
y
z
but ofwith
a delay
depending
of
oftheprotons
to
the
ourgives:
full analysis
the ∼730
AMI events,
in particular: on time
cess
andflight
limitations
dimensions
ergy
level,
the
width of on
accelerationand
areaother
W incharacteristics
the direction
of
electric
field
must
be
5000
km
at
B
=
20
nT
and
cannotdvx = ω observation
point.
dvy
Z · e · E Rough estimate of these delays for the 6
20
000
km
at
B
=
10
nT.
For
heavy
ions
this
limitation
is
·
v
,
=
−ω
·
v
+
(3)
c y
c x
www.jn.net
Journalname
dt
dt
M
not
so
strong.
Figure
16a
shows
proton
trajectories
in
the
ace propaevents showed that they ranged from 1.5 to 24 s which cannot
ef f
ef f
ef f
′
celeration region with the thickness D (the current filament
Ann. Geophys., 29, 1439–1454, 2011
www.ann-geophys.net/29/1439/2011/
www.jn.net
Date
UT interval
Tel. T0,
s
A,
keV
B,
keV
C,
1/s
3. by the average energy of the proton line (see Table 2),
which is not inconsistent with the average potential drop
6.06.1996 12:24:30 - 12:25:00 1p 44672 26.65 34.82 0.082
across the magnetosphere due to the motional electric
V. N. Lutsenko
and- E.
A. Gavrilova:
Properties
the origin
beams
1449
9.04.2000
03:08:00
03:10:00
1p 11290
55.84and
32.89
0.058 of AMIfield
of the solar wind: : E sw = −(V sw × B sw ) ,
12.04.1996 22:22:20 - 22:23:20 2p 80552 22.5 29.74 0.04
10.08.1997 23:13:35 - 23:14:54 1p 83626 46.2 45.3 0.035
4. by short durations of AMI events (on average 104.7 s),
Table 4. Cyclotron
for different
by E 0.06
= 0.1 V m−1 .
max 74319
23.04.1997
20:38:30frequency
- 20:39:45and Y1p
48.9 ions
22.8
20.04.1999 13:43:20 - 13:44:25 1p 49402 80.43 59.47 0.09
5. by the character of time variations of proton line enerB = 20 nT
B = 10 nT
gies and intensities
at the event start. It is this character
Z
A
ωc , rad s−1
Ymax , km
change substantially the1results
analysis. 5217.5
1 of our
1.916
2
6
4
16
7 Model of AMI acceleration
0.958
0.719
10435
13913
Ymax , RE
0.818
1.636
2.181
−1
ofωvoltage
by the disruption
, RE
c , rad s ∆V Ychange
max , kmcanYbe
maxexpected
of the electric circuit, shown in Fig. 15,
0.958
20 870
3.272
0.479
41
740
6.544
6. by the total absence of monoenergetic
electrons in the
0.359
55 652
8.726
AMI events.
Previously, on the basis of our preliminary analysis of AMI
thickness) and disruption region width W . If W > Ymax , the
events we proposed a hypothesis (Lutsenko and Kudela,
creation of monoenergetic lines is impossible. One can see
1999;
2001)
in which
generation
thatLutsenko,
all protons,
starting
within
the limitsofofAMI
Deffobserved
, will obnear
BS
occurred
in
spikes
of
potential
electric
arising
tain the same final energy of 200 keV. Protons,field
starting
out
as aofresult
of
BS
current
sheet
filaments
disruptions
(see
Fig.a
Deff , will obtain lower energies. Figure 16b allows
15).comparison
Like eachof
current
circuit
these filamentsacceleration.
have an elec-It
protons
and alpha-particles
tromotive
(A inthat
Fig.the15),
the resistance
and
follows force
from EMF
this figure
effective
value DeffR is
alinductance
L. The
is distributed
that part Indeed,
of the
ways greater
forEMF
alpha-particles
thanalong
for protons.
p action of a motional
α /D
0
circuit
which
is exposed
to an
electric
for W
= 4000
km Deff
eff = 1.31 and for W = 5000 km
field
of thelines),
SW. the
After
the circuit
a sum
of the
(dashed
mentioned
ratiodisruption
will be equal
to 2.65.
So
EMF
self-inductance
EMF of(Bthe
in acceleration
Fig. 15) will
be
we and
mustthe
expect
that after leaving
region,
applied
to the short
disruption
interval
W as a voltage
differ-at
alpha-particles
will
always have
an advantage
over protons
ence
∆V = Aof+monoenergetic
B.
creation
lines. It explains why the value of
ratio (Sp /Sα )calc calculated from the ratio Np /Nα is ∼2 times
lower than the observed average value of Sp /Sα (see Sect. 4).
Graphics in Fig. 16 and data in Table 4 allow us to make
rough estimates of the acceleration region size. The width of
the accelerating gap W should not exceed Ymax = 0.818 RE
for B = 20 nT. The thickness of the acceleration region D
can not be greater than 1.25 RE for a conservation of the observed advantage of alpha particles over protons in a factor
of ∼2 by monoenergetic lines formation.
8
The reason of current circuit disruption – connection
between AMI and HFA
The natural question arises: what is the reason of current
filaments disruptions? In the magnetotail neutral sheet where
AMI15.
were
observed,
the disruptions
the
Figure
Thealso
sketch,
illustrating
the process of
of filaments
solar windof
ions
current sheet
separating
oppositely
directed
acceleration
and AMI
generation
by BS current
sheetmagnetic
filamentsfields
dismay be
spontaneous.
Theirelectromotive
durations amount
∼20 s,B-the
after
ruption.
Here
A is the circuit
force to
(EMF),
which
the
conductivity
is
restored
(see
Figs.
14
and
15
from
self-inductance EMF. Electrons, having much lower gyroradia, will
Lutsenko
et the
al., drift
2008).
In the
of BS, the current sheet
be sweep
out by
without
the case
acceleration.
separates magnetic fields of about the same directions so one
should look for an external reason of disruptions.
This assumption was entirely confirmed by the results of
In the solar wind near BS, a phenomenon known as Hot
our full analysis of the ∼730 AMI events, in particular:
Flow Anomaly (HFA) (see e.g. Schwartz, 1995; Schwartz et
al., 2000) was often observed. HFA was observed when the
solar
wind had brought to the BS a tangential discontinuity in
www.jn.net
the magnetic field caused by a large-scale flat current sheet
(CS). Because of opposite directions of the magnetic fields
on both sides of the CS, the convection electric fields near the
CS-BS intersection line may be directed to the CS. This electric field changes the plasma flow direction, so that the solar
www.ann-geophys.net/29/1439/2011/
Figure 16. a) Proton trajectories in the acceleration region. D Fig. 16. (a) Proton trajectories in the acceleration region. D – the
the thickness of the disrupted current filament. W - the width of
thickness of the disrupted current filament. W – the width of the
the region with the potential electric field. De f f - effective thickness
region with the potential electric field. Deff – effective thickness
ofofthe
acceleration region, starting from which all protons will obthe acceleration region, starting from which all protons will obtain
the
starting out
out of
of D
De f f limits,
limits,
tain thesame
sameenergy
energy 200
200 keV.
keV. Protons,
Protons, starting
eff
will
obtain
the
energy
<200
keV.
b)
Comparison
of
D
for
alphae
f
f
will obtain the energy <200 keV. (b) Comparison of Deff for alphaparticles
and
for
protons.
Dotted
lines
show
the
same
trajectories
particles and for protons. Dotted lines show the same ′trajectories
for
acceleration region
regionwidth
widthWW0 =
= 5000
5000km
km
forthe
thelarger
larger value
value of
of the
the acceleration
instead
insteadof
ofW=
W =4000
4000km.
km.
Let’s examine now in more details the acceleration process
on dimensions
and other
windand
fluxlimitations
deflects through
a large angle
to thecharacteristics
CS position
and upon reaching the CS it even turns sunward (Schwartz,
1995). In these studies, the influence of HFA on such
a largeJournalname
scale BS parameter as its current sheet was not considered.
In Fig. 17 we would like to explain the physical nature
of BS current sheet and to show a possible reason for its
filaments disruption. Figure 17a illustrates the creation of
current carriers (ions and electrons) on a BS magnetic field
Ann. Geophys., 29, 1439–1454, 2011
V.N. Lutsenko and E. A. Gavrilova: Properties and
origin of Almost
Monoenergetic
Ion (AMI)
11
1450
V. the
N. Lutsenko
and E. A.
Gavrilova: Properties
andbeams
the origin of AMI beams
a)
c)
b)
E PRE
MSH
Solar Wind
E
MSH
MSH
Solar Wind
Solar Wind
i
i
B
Absorbing Screen
i
B
VSW
VSW
E
B POST
HFA
i
i
VSW
B
Current
BS
B PRE
Current
Disruption
Current
Disruption
i
Current Sheet
in the Solar Wind
B
BS
i
BS
E POST
Figure
HFA
phenomenon
a possible
cause
currentsheet
sheetfilaments
filamentsdisruptions.
disruptions.(a)
a) Creation
Creation of
of current
current carriers
carriers for
for the
Fig.
17. 17.
TheThe
HFA
phenomenon
as aaspossible
cause
forofBSBScurrent
the BS
BS
current sheet.
sheet. (b)
b) How
current
Howthe
theBS
BScurrent
currentsheet
sheetcan
canbe
bedisrupted.
disrupted.c)(c)How
HowthetheHFA
HFAcan
canplay
playa role
a roleofinthe
thescreen
screenininFig.
Fig.17b.
17b.
current carriers
(ionsions
and and
electrons)
on acrossing
BS magnetic
field
jump.
Solar wind
electrons
this jump
jump.
Solar
wind
ions
and
electrons
crossing
this
jump
perform a short gradient drift, shifting them along a pertanform a component
short gradient
driftmotional
shifting electric
them along
gential
of the
field aoftangential
the solar
component
theserve
motional
electric
field for
of the
solarcurrent.
wind.
wind.
Thus,of
they
as charge
carriers
the BS
Thus,
they
serve
as
charge
carriers
for
the
BS
current.
Then
Then they move downstream to the MSH, being replaced
by
they
move
downstream
to
the
MSH
being
replaced
by
new
new SW particles. Let us suppose that we have set someSW particles.
that we have
set protecting
somewherea
where
upstreamLet
of us
thesuppose
BS an absorbing
screen
upstream
of
the
BS
an
absorbing
screen
protecting
a part of
of
part of BS from the solar wind flux (Fig. 17b). Creation
BS
from
the
solar
wind
flux
(Fig.
17b).
Creation
of
current
current carriers here stops and the BS current sheet or, better
carriers
hereofstops
and the BS
current disrupted.
sheet or, better
said,
said,
a part
its filaments,
becomes
Figure
17ca
part
of
its
filaments,
becomes
disrupted.
Figure
17c
shows
shows how the arrival of CS in the solar wind, creation of the
how the
of CSofinplasma
the solar
wind,
of the HFA
HFA,
andarrival
deviation
flow
leadcreation
to a situation
with
and
deviation
of
plasma
flow
lead
to
a
situation
with
disrupdisruption of the BS current similar to the screen action
in
tion of
the BS
similar
to the
in Fig.
17b.
Fig.
17b.
An current
observer
situated
at screen
a smallaction
distance
behind
An BS
observer
situated
at a small
behind
the BS
will
the
will see
a decrease
of thedistance
antisunward
plasma
flux
to
see
a
decrease
of
the
antisunward
plasma
flux
to
about
a
zero
about a zero level, while the fluxes perpendicular to the sun
level, while
theincrease.
fluxes perpendicular to the sun direction must
direction
must
increase.
After the solar wind shifts the line of CS and BS intersecAfter
the (in
solar
shifts of
thethe
line
of CS
and in
BSFig.
intersection
further
thewind
direction
green
arrow
17c),
tion
further
(in
the
direction
of
the
green
arrow
in
Fig.
the antisunward plasma flux will not only be restored17c)
but
the antisunward
not onlyofbe
but
can
become moreplasma
intenseflux
duewill
to addition
therestored
solar wind
can
become
more
intense
due
to
addition
of
the
solar
wind
plasma deflected and detained by HFA (see, e.g. Sect. 9.2,
plasma
and detained
by HFAand
(seecan
e.g.arise
Sect.in 9.2,
Fig.
23).deflected
The disruption
disappears
the
Fig.
23).
The
disruption
disappeared
and
can
arise
in the
neighboring filaments.
neighboring
filaments.
We have paid
attention to a similarity of conditions for
We
have
paid
to a of
similarity
generation of AMIattention
and creation
HFA: of conditions for
generation of AMI and creation of HFA:
1. Arrival of the CS in the solar wind,
1. Arrival of the CS in the solar wind,
2.
2. The
The CS
CS is
is almost
almost perpendicular
perpendicular to
to BS
BS at
at times
times of
of AMI
AMI
and
HFA
generation
(the
angle
between
normals
and HFA generation. (the angle between normals N
Ncs
cs
◦◦ ),
and
N
is
close
to
90
bs
and N bs is close to 90 ),
3. Quasi-perpendicular orientation of the magnetic field
◦◦
(the angle Tbn
bn is close to 90 ) at least on one side of
CS (to reflect ions), and
4. The electric field at least at one CS side is directed to
the CS.
www.jn.net
Ann.
Geophys., 29, 1439–1454, 2011
The
0.12 per
given by
The rate
rate of
of HFA
HFA occurrence
occurrence (∼
(∼0.12
per hour),
hour) given
by
Schwartz
at
al.
(2000)
is
consistent
with
our
estimates
Schwartz et al. (2000) is consistent with our estimates of
of
AMI
0.038 per
AMI events
events occurrence
occurrence rates
rates (>
(>0.038
per hour).
hour). Our
Ouranalyanalsis
ysisofofthe
theAMI
AMIevents
eventsand
andpresence
presence of
of CS
CS in
in the
the solar
solar wind
wind
has
shown
that:
has shown that:
∼ 40% of AMI events were accompanied by a strong, sin∼40 % of AMI events were accompanied by a strong,
gle CS,
singleevents
CS, - by weak or numerous CS’s which compli∼ 30%
cates∼30
the analysis,
% events – by weak or numerous CSs – what com∼ 30%
events
- by the absence of CS which points to the
plicates
the analysis,
possibility of other reasons of current disruptions.
∼30 %these
events
– by the
absence of should
CS – that
points to
However
general
considerations
be supported
the
possibility
of
other
reasons
for
current
disruptions.
by direct observations of CS, HFA and AMI in one particular
event. For
such
analysis
it is necessary
to select
observaHowever,
these
general
considerations
should
be supported
tions
occurred
in the solar
wind
or inand
theAMI
magnetosheath
close
by direct
observations
of CS,
HFA,
in one particular
enough
(1 to
2 Ranalysis
Otherwisetoitselect
will be
difficult to
E ) to theitBS.
event. For
such
is necessary
observations
detect
AMI and
HFA
onwind
the same
We must also close
bear
that occurred
in the
solar
or in S/C.
the magnetosheath
in
mind
that
AMI
propagate
from
the
acceleration
place
as a
enough (1 to 2 RE ) to the BS. Otherwise, it will be difficult
narrow
beam
and
not
always
can
be
detected
on
the
S/C
even
to detect AMI and HFA on the same S/C. We must also bear
if
position
is close
to the acceleration
place. Onplace
the other
in its
mind
that AMI
propagate
from the acceleration
as a
hand
depending
on
the
direction
of
magnetic
field
connectnarrow beam and can not always be detected on the S/C,
even
ing
acceleration
with
the S/C the
AMIOn
observation
if itsthe
position
is closeplace
to the
acceleration
place.
the other
time
can
be
as
before,
as
after
the
moment
of
the
CS
HFA
hand, depending on the direction of magnetic field and
connectobservation
(delay
of
1
to
10
min).
In
such
cases
AMI
can
ing the acceleration place with the S/C, the AMI observation
be
connected
with
the
other
HFA
generated
by
the
same
CS
time can be as before, as after the moment of the CS and HFA
several
minutes
before
or
after
in
a
neighboring
place.
observation (delay of 1 to 10 min). In such cases, AMI can
be connected with the other HFA generated by the same CS
several
minutesof
before
or after
a neighboring
place.
9
Examples
several
AMIinevents
analysis
To
analysis
of AMI AMI
eventsevents
and toanalysis
check the fulfillment
9 make
Examples
of several
of 4 conditions specified in the previous section it is necessary
to find athe
CSAMI
which
can be
responsible
for generation
To analyze
events
and
check fulfillment
of theof4
HFA
and AMI
usinginmagnetic
field section,
data from
or several
conditions
specified
the previous
it isone
necessary
to
S/C’s
thewhich
solar wind.
it is necessary
to determine
the
find a inCS
can beThen
responsible
for generation
of HFA
CS
its motion
intersecandparameters
AMI usingallowing
magnetica modeling
field data of
from
one or by
several
S/C
tion
thewind.
BS (aThen
direction
CS normal
N cs and velocity
in thewith
solar
it is of
necessary
to determine
the CS
V
along
this normal).
Themotion
minimum
variance
parameters
allowing
a modeling
of its
by interseccsn of its motion
analysis
theBS
magnetic
fieldof
near
CS allows
finding
the
tion withofthe
(a direction
CSthe
normal
Ncs and
velocity
Journalname
www.ann-geophys.net/29/1439/2011/
9.1 AMI events on 23 April 1997 in the solar
wind near
1451
the BS
V. N. Lutsenko and E. A. Gavrilova: Properties and the origin of AMI beams
Vcsn of its motion along this normal). The minimum variance
analysis of the magnetic field near the CS allows finding the
normal to the current sheet Ncs . Then the normal component of the CS velocity Vcsn can be found by using Ncs and
measured solar wind velocity Vsw . Among the two possible
Ncs directions, the direction with Ncsx > 0 has always been
selected. In that case for non-fulfillment of the 4-th condition, it is necessary that simultaneously Enb < 0 and Ena > 0.
Here, Enb and Ena are the projections of the solar wind electric field on Ncs before and after CS coming, respectively. In
all other cases the condition 4 is fulfilled. Below we give several examples of almost simultaneous observation of the CS,
HFA, and AMI on Interball-1. The AMI observations were
made by the DOK-2 instrument. The presence of HFA (the
solar wind flux direction change) was identified by the VDP
instrument (Safrankova et al., 1997), with 15 s time resolution. VDP had 4 ion detectors (Faraday cups). One (F1) was
mounted to look sunward along the S/C spin axis and three
other (F2, F3, and F5) were mounted to look perpendicular to the spin axis with 90◦ and 180◦ azimuthal intervals.
As Interball-1 had no direct electric field measurements, the
E SW was calculated by using Wind measurements of V SW
and B SW often made far from the HFA and AMI generation
position. It may explain why in some cases below, the
4-th
Figure
condition was not fulfilled.
Fig. 18. Positions of Interball-1 and Geotail during AMI events on
18.
Positions of Interball-1 and Geotail during AMI events
23 April 1997.
on 23 April 1997.
9.1
AMI events on 23 April 1997 in the solar wind near
the BS
time interval from 18:43 to 18:49 UT. The spectra of DOK-2
andpositions
the EPIC device
for this event and
were Geotail
also
Figure(2p-telescope)
18 presents
of Interball-1
durshown
above
(see
Figs.
3
and
11).
Figure
23
presents
the
Figure 18 presents positions of Interball-1 and Geotailing
during
the observations
on Interball-1 of two AMI events (AMImagnetic fields for 3 S/C and plasma fluxes (according to the
the observations on Interball-1 of two AMI events (AMI-1
1 and
the Times
spectra
of which
are on
shown
in Fig. 19. The
VDP data).
of AMI
observations
S/C Interball-1
and AMI-2), the spectra of which are shown in Fig. 19.
The AMI-2)
Geotail
are
shown
by
short
magenta
lines
in
correspondtimes of AMI-1 and AMI-2 observations are markedtimes
on the of and
AMI-1 and AMI-2 observations are marked on the
ing panels. The HFA is clearly seen here in the bottom panel
third panel of Fig. 20, where magnetic fields (Wind, Geothird panel
of Fig.
where
magnetic
in time
interval20,
from
18:47:29
to 18:50:52fields
UT as a(Wind,
drop of Geotai
tail and Interball-1) and solar wind fluxes (Interball-1) are
F1 flux to about
andwind
increasefluxes
of F2, F3,
and F5 fluxes. are preand Interball-1)
and zero
solar
(Interball-1)
presented.
As
both
of
these
S/C
were
in
the
MSH
rather
far from the
Here we have two strong current sheets in the solar
wind
sented.
model
BS
(−2.6
and
−1.42
R
),
the
CS6
distinctly
visible in
E
CS1 and CS2 which can be responsible for AMI-1 and AMIthe
Wind
magnetic
field
is
not
seen
in
Interball-1
and
Geotail
Here
we
have
two
strong
current
sheets
in
the
solar wind
2. Interball-1 approached to the model BS by 0.65 RE bemagnetic
field
data.
Nevertheless,
we
have
shown
that
just
tween CS1 and CS2, what can explain the absenceCS1
of ob-and CS2 which can be responsible for AMI-1 and AMIthe current sheet CS6 can be responsible for AMI genera-
vious signs of HFA near the CS1 and their presence near
tion. The results of this tangential discontinuity analysis and
the CS2. A double jump in magnetic field and plasma flux
calculated position of the CS6 at the moments of HFA and
near the CS2 on Interball-1 may be caused by short osJournalname
AMI observations are given in Fig. 24. It is seen from this
cillations of the BS position. Figures 21 and 22 present
figure that the first 3 requirements are fulfilled. The electric
the positions of CS1 and CS2 at observations of AMIfield on one side of the CS6 (after the current sheet) is di1 and AMI-2, as well as HFA (only for AMI-2). For
rected to the CS6: Enb = −1.50, Ena = −0.44 mV m−1 , i.e.
the AMI-2 event, all 4 requirements are fulfilled (Enb =
all 4 requirements are fulfilled.
2.21 mV m−1 , Ena = −1.36 mV m−1 ). For AMI-1, all the
requirements are fulfilled apart from the 4-th requirement
(Enb = −1.25 mV m−1 , Ena = 1.89 mV m−1 ).
10 Conclusions
9.2 AMI event on 16 April 1996 in the MSH near the BS
1. The study of the large data array on AMI beams observed by Interball-1 near the BS showed that AMI
On 16 April 1996, Interball-1 and Geotail were both in the
beams are a fairly widespread phenomenon. The
MSH near the BS (see Fig. 2) observing an AMI event in
www.ann-geophys.net/29/1439/2011/
Ann. Geophys., 29, 1439–1454, 2011
V.N. Lutsenko and E. A. Gavrilova: Properties and the origin of Almos
1452
V. N.beams
Lutsenko and
A. Gavrilova:
and the
origin ofand
AMI
va: Properties
and the origin of Almost Monoenergetic Ion (AMI)
V.N.E.Lutsenko
and E.Properties
A. Gavrilova:
Properties
thebeams
origin of Almos
compoN cs and
possible
ways seondition
y. Here
ric field
ll other
eral exS, HFA
e made
he solar
VDP inresoluF1)was
nd three
r to the
terballwas calW often
It may
was not
−
−
−2
−3
Figur
Figure
and H
and
HF
18:09
18:09:
Geot
Geota
just t
just
th
eratio
eratio
and c
and
andca
A
and
thisAF
this
F
electr
electri
is dir
isthe
dire
4
the 4 r
nd near
tail durs (AMI19. The
d on the
Geotail
are pre-
ar wind
d AMI-
Figure 19. Spectra of two AMI events on 23 April 1997.
Fig. 19. Spectra of two AMI events on 23 April 1997.
2. Interball-1
approachedfrequency
to the model
BSand
at MSH
0.65 Rduring
E beaverage observation
in SW
tween
CS1
and
CS2
what
can
explain
the
absence
of
obvious
the 5-year work of Interball-1 was 0.038 per hour.
signs of HFA near the CS1 and their presence near the CS2.
A2.double
jump inofmagnetic
field and
plasma
nearnumthe
The statistics
AMI properties
based
on aflux
greater
CS2 ber
on Interball-1
may be
oscillations
of
of observations
as caused
well as by
theshort
comparison
of the
the BS
position.
and
22 some
present
the experiments
positions of
DOK-2
data Figures
with the21
data
from
other
CS1 confirm
and CS2our
at observations
and of
AMI-2,
and also
assumptions of
onAMI-1
the nature
AMI lines
and
HFAmechanism
(only for AMI-2).
For the AMI-2 event all 4 requireof their formation.
ments are fulfilled (Enb = 2.21 mV/m, Ena = -1.36 mV/m).
3. AMI-1
The formation
of AMI beams
is a result
of the
For
all the requirements
are fulfilled
apart
fromsolar
the
wind ions acceleration
in small
themV/m).
BS with a
4-th requirement
(Enb = -1.25
mV/m,regions
Ena = on
1.89
potential electric field created by disruptions of BS current sheet filaments. Such acceleration mechanism was
9.2 confirmed,
AMI eventin
onparticular:
16 April 1996 in the MSH near the BS
On 16 –April
1996
Interball-1
and Geotail
were bothofinprothe
by the
dynamics
of energies
and intensities
MSH near
the
BS
(see
Fig.
2)
observing
an
AMI
event
in
ton lines after the event start,
time interval from 18:43 to 18:49 UT. The spectra of DOK-2
– by complete
of monoenergetic
electrons
(2p-telescope)
and the absence
EPIC device
for this event were
also
in
the
25–400
keV
energy
range
during
AMI
shown above (see Fig. 3 and Fig. 11). Figure 23 the
presents
events,
the magnetic
fields for 3 S/C’s and plasma fluxes (according to –thebyVDP
data).
Times of of
AMI
S/C’s
ratios
of energies
theobservations
three AMI on
lines
atInterball-1
and Geotail
are and
shown
by short magenta
tributed
to H, He,
(CNO)-group,
and lines in
corresponding panels. The HFA is well seen here in the bot– by the marked difference between the ratios of
tom panel in time interval from 18:47:29 to 18:50:52 UT as
numbers of accelerated protons and alpha-particles
a drop of F1 flux to about zero and increase of F2, F3 and
in AMI spectra and the same ratios calculated from
Ann. Geophys., 29, 1439–1454, 2011
www.jn.net
Figure 20. Magnetic fields and solar wind fluxes (data from Wind,
Figure
Magneticfields
fieldsand
and solarwind
windfluxes
fluxes(data
(data fromWind,
Wind,
Fig. 20.20.Magnetic
Interball-1
and Geotail)
during solar
two AMI events
on 23 from
April 1997.
Interball-1
and
Geotail)
during
two
AMI
events
on
23
April
1997.
Interball-1 and Geotail) during two AMI events on 23 April 1997.
Intersection of the Current Sheet CS1 with the BS at AMI−1 Observation (Interball−1)
23.04.1997
at 17:31:42−17:32:06
0.7134,
0.5244] (Interball−1)
(Wind)
Intersection
of the
Current Sheet CS1 UT,
withNcs1=[0.4648,
the BS at AMI−1
Observation
23.04.1997 at 17:31:42−17:32:06 UT, Ncs1=[0.4648, 0.7134, 0.5244] (Wind)
CS1 at AMI
CS1 at AMI
30
30
Interball−1
Interball−1
Ncs1
Ncs1
20
20
Xgse
Xgse
MI events
2
Xgse
Xgse
4
0
Geotail
Geotail
0
B, Tbn=80.7°
B, Tbn=80.7°
30
30
−10
−10
−20
−20
−30
−30
20
20
−20
−20
−10
−10
10
10
0
0
−10 Zgse
−10 Zgse
00
Ygse
Ygse
10
10
−20
−20
20
20
30
30
−30
−30
Figure
Intersections
of the
during
AMI-1
obFig. 21.21.
Intersections
of the
CS1CS1
andand
the the
BS BS
during
AMI-1
obserFigure
21.
Intersections
during
AMI-1
observation
Interball-1
April
1997
17:31:42-17:32:06
UT.
vation onon
Interball-1
onon
23 23
April
1997
at 17:31:42–17:32:06
UT.UT.
servation
on
Interball-1
on
at 17:31:42-17:32:06
the densities of these ions in the solar wind plasma
F5
MSH rather
rather far
far
F5fluxes.
fluxes. As
As both
both of
of these
these S/C’s were in the MSH
(enrichment in alpha-particles relative to protons).
from
CS6 distinctly
distinctlyvisvisfromthe
themodel
model BS
BS (-2.6
(-2.6 and -1.42 RE ) the CS6
iblein
in the
the Wind
Wind magnetic
magnetic field is not seen in
ible
in Interball-1
Interball-1 and
and
www.ann-geophys.net/29/1439/2011/
www.jn.net
www.jn.net
1.
1. T
s
b
a
5
2.2. T
b
D
c
m
Ncs1
Ncs1
10
10
10
10 C
3.3. T
w
p
r
c
Ackno
Figure 23.
Magnetic fields on Wind, Geotail, Interball-1 and
plasma fluxes on Interball-1 during AMI event in the MSH on 16
origin of Almost
(AMI) beams
13ofApril
V. N. Monoenergetic
Lutsenko and E. Ion
A. Gavrilova:
Properties and the origin
AMI1996.
beams
1453
Intersections of the Current Sheet CS2 with the BS at AMI−2 and HFA Observations
23.04.1997 at 18:09:10−18:09:24 UT, Ncs2=[0.2469, 0.7836, 0.5701] (Wind)
Intersection of the Current Sheet CS6 with the BS at AMI and HFA Observations on 16.04.1996
at 18:43:26−18:45:35 UT (Interball−1) and 18:43:10−18:46:26 UT (Geotail) Ncs6=[0.2671, −0.2965, 0.9169] (Wind)
CS6 at AMI
(Interball−1)
start and end
CS2 position at AMI−2
(start and end)
CS2 position at HFA
(start and end)
CS6 at AMI−2
(Geotail)
start and end
CS6 at HFA
(Interball−1)
start and end
Interball−1 Ncs6
(in MSH)
lund f
DOKperime
servat
tion.
were
http://
Refer
Wind
20
40
Geotail
(in MSH)
10
Ncs2
20
Xgse
Interball−1
Xgse
Ncs2
Geotail
0
B, Tbn=78.7°
(Geotail )
20
0
Zgse
−20
−10
0
Ygse
10
B, Tbn=79.6°
(Interball−1)
−20
30
B, Tbn=87.5°
−20
−30
0
−10
30
20
20
10
10
0
Zgse
−20
20
50
40
30
0
Ygse
−10
−10
−20
−20
−30
−30
Figure
Intersections
of the
andBS
theduring
BS during
Fig. 22.22.Intersections
of the
CS2CS2
and the
AMI-2AMI-2
and
Fig. 24. Intersections of the current sheet CS6 and BS during obserand
HFA
observations
Interball-1
April
1997
18:09:10- Figure
HFA
observations
on on
Interball-1
on on
23 23
April
1997
at at
18:09:10–
24.AMI
Intersections
current
sheet
CS6 and
BS durvations
and HFA
onof
16the
April
1996 by
Interball-1
(18:43:26–
14
V.N.
Lutsenko
and
E.
A.
Gavrilova:
Properties
andofthe
origin
of Almost
Monoenergetic
Ion (AMI)
beams
18:09:24UT.
UT.
18:09:24
ing
observations
ofGeotail
AMI and
HFA on 16 AprilUT).
1996 by Interball-1
18:45:35
UT) and
(18:43:10–18:46:26
(18:43:26-18:45:35 UT) and Geotail (18:43:10- 18:46:26 UT).
Wind,
997.
– by the marked difference between the ratios of
Geotail magnetic field data. Nevertheless we have shown that
numbers of accelerated protons and alpha-particles
4. The study of angular distributions and temporal characjust the current sheet CS6 can be responsible for AMI genin AMI spectra and the same ratios calculated from
teristics of AMI beams as well as the modeling of ions
eration. The results of this tangential discontinuity analysis
the densities of these ions in the solar wind plasma
motion in the acceleration region showed that their acand calculated position of the CS6 at the moments of HFA Journalname
(enrichment in alpha-particles relative to protons).
celeration occurs during a short time (∼2 min) in parts
and AMI observations are given in Fig. 24. It is seen from
the study
BS, characteristic
dimensions ofand
which
don’t ex4.ofThe
of angular distributions
temporal
charthis Figure that the first 3 requirements are fulfilled. The
ceed
0.8 to 1.25
RE . beams as well as the modeling of
acteristics
of
AMI
electric field on one side of the CS6 (after the current sheet)
ions motion in the acceleration region showed that their
is directed to the CS6: Enb = -1.50, Ena = -0.44 mV/m, i.e. all
5.
The
main reason
for BS
current
sheet time
disruptions
was in
acceleration
occurs
during
a short
(∼ 2 min)
the 4 requirements are fulfilled.
determined:
an interplanetary
sheetof(tanthe parts of Ittheis BS,
characteristic current
dimensions
which
gential
discontinuity)
brought
don’t exceed
0.8 to 1.25
RE . by the solar wind to the
10 Conclusions
1. The study of the large data array on AMI beams observed by Interball-1 near the BS showed that AMI
beams are a fairly widespread phenomenon. The average observation frequency in SW and MSH during the
5-year work of Interball-1 was 0.038 per hour.
2. The statistics of AMI properties based on a greater number of observations as well as the comparison of the
DOK-2 data with the data from some other experiments
confirm our assumptions on the nature of AMI lines and
mechanism of their formation.
30
3. The formation of AMI beams is a result of the solar
wind ions acceleration in small regions on the BS with a
potential electric field created by disruptions of BS current sheet filaments. Such acceleration mechanism was
confirmed, in particular:
1 ob6 UT.
Fig. 23. Magnetic fields on Wind, Geotail, Interball-1 and plasma
Figure– 23.
Magnetic
fields onenergies
Wind, Geotail,
Interball-1 proand
the
dynamics
fluxes onby
Interball-1
during of
AMI event inand
the intensities
MSH on 16 of
April
plasma fluxes
on Interball-1
during
AMI
event in the MSH on 16
ton
lines
after
the
event
start,
1996.
er far
y vis1 and
– by complete absence of monoenergetic electrons
in the 25-400 keV energy range during the AMI
Intersection of the Current Sheet CS6 with the BS at AMI and HFA Observations on 16.04.1996
at 18:43:26−18:45:35 UT (Interball−1) and 18:43:10−18:46:26 UT (Geotail) Ncs6=[0.2671, −0.2965, 0.9169] (Wind)
events,
CS6 at AMI
at AMI−2
– by ratios of energies
of the three CS6
AMI
lines at(Interball−1)
(Geotail)
www.ann-geophys.net/29/1439/2011/
start and end
start and end
tributed to H, He and (CNO)-group,
CS6
at
HFA
Ncs6
April 1996.
Interball−1
(in MSH)
(Interball−1)
start and end
BS. Under certain conditions it leads to formation of
5.a The
of BSthe
current
disruptions
HFA.main
The reason
HFA blocks
accesssheet
of the
solar windwas
determined.
It
is
an
interplanetary
current
(tanflux to some small part of the BS which stops sheet
the cregential
discontinuity)
bytothe
wind
to the
ation
of current
carriers. brought
This leads
thesolar
current
circuit
BS. Under certain conditions it leads to formation of a
disruption.
HFA. The HFA blocks the access of the solar wind flux
some
small
the of
BSAMI
which
stops the
creation
6. Ittowas
shown
thatpart
the of
study
generation
process
of
current
carriers
here.
This
leads
to
the
current
circuit
may give new information on a large-scale structure,
disruption.
properties, and dynamics of BS current sheet which canbe obtained
nowthe
by other
means.
6.not
It was
shown that
studyknown
of AMI
generation process
may give new information on a large-scale structure,
7. New
arguments
were found
in support
of H.which
Alfvén
properties
and dynamics
of BS
current sheet
canideas
on
a
great
role
which
electric
currents
and
not be obtained now by other known means. electric current sheets play in space plasmas.
7. New arguments were found in support of H. Alfvén
ideas on a great
role
whichto electric
currents
and for
elecAcknowledgements.
We are
grateful
A. Lui and
S. Nylund
tric
current
sheets
play
in
space
plasmas.
submission of selected EPIC data for comparison with DOK-2
data, to G. N. Zastenker for provision of VDP experiment data and
Acknowledgements. We are grateful to Dr. A. Lui and S. Nyto A. Klassen for information on AMI observations in STEREO
lund for submission of selected EPIC data for comparison with
project before of corresponding publication. Magnetic field data
DOK-2
Dr. G.
Zastenker
for taken
provision
from
Winddata,
and to
Geotail
(3 s N.
resolution)
were
from of
theVDP
web- experiment
data Space
and toFlight
Dr. A.
Klassen
for information on AMI obsite
of Goddard
Center:
http://cdaweb.gsfc.nasa.gov.
servations
in STEREO
project
before
corresponding
Guest Editor
M. Gedalin
thanks
two of
anonymous
refereespublicafor
tion.
Magnetic
field
data
from
Wind
and
Geotail
(3
s resolution)
their help in evaluating this paper.
were taken from the web-site of Goddard Space Flight Center:
http://cdaweb.gsfc.nasa.gov
Ann. Geophys., 29, 1439–1454, 2011
References
Wind
20
Journalname
Alfvén, H.: Electric Currents in Cosmic Plasmas, Revs. Geophys.
Alfvén
Spa
Ander
stre
Klasse
S.,
STE
mo
Lutsen
Stu
in t
Lutsen
the
26,
Lutsen
Alf
Phy
1454
V. N. Lutsenko and E. A. Gavrilova: Properties and the origin of AMI beams
References
Alfvén, H.: Electric Currents in Cosmic Plasmas, Revs. Geophys.
Space Phys., 15, 271–284, 1977.
Anderson, K. A.: Measurements of bow shock particles far upstream from the Earth, J. Geophys. Res., 86, 4445–4454, 1981.
Klassen, A., Gómez-Herrero, R., Müller-Mellin, R., Böttcher, S.,
Heber, B., Wimmer-Schweingruber, R., and Mason, G. M.:
STEREO/SEPT observations of upstream particle events: almost monoenergetic ion beams, Ann. Geophys., 27, 2077–2085,
doi:10.5194/angeo-27-2077-2009, 2009.
Lutsenko, V. N.: Almost Monoenergetic Ions: New Support for
Alfvén Ideas on the Role of Electric Currents in Space Plasmas,
Phys. Chem. Earth (C), 26, 615–619, 2001.
Lutsenko, V. N. and Kudela, K.: Almost monoenergetic ions near
the Earth’s magnetospheric boundaries, Geophys. Res. Lett., 26,
413–415, 1999.
Lutsenko, V. N., Kudela, K., and Sarris, E. T.: The DOK-2 Experiment to Study Energetic Particles by the Tail and Auroral Probe
Satellites in the Interball project, Cosmic Research, 36, 98–107,
1998.
Lutsenko, V. N., Gretchko, T. V., Kobelev, A. V., Styazhkin, V.
A., and Kudela, K.: “Wavy” Energetic Ion Dispersion Events
and PC-5 Type Magnetic Field Pulsations, Adv. Space Res., 30,
1783–1786, 2002.
Ann. Geophys., 29, 1439–1454, 2011
Lutsenko, V. N., Kirpichev, I. P., Grechko, T. V., and Delcourt, D.:
Source Positions for Energetic Particles Responsible for the Fine
Dispersion Structures: Numerical Simulation Results, Planet.
Space Sci., 53, 275–281, 2005.
Lutsenko, V. N., Gavrilova, E. A., and Grechko, T. V.: Statistics
of fine dispersion structures events in energetic particle spectra:
their origin and role in the outer magnetosphere, Ann. Geophys.,
26, 2097–2110, doi:10.5194/angeo-26-2097-2008, 2008.
Safrankova, J., Zastenker, G., Nemecek, Z., Fedorov, A., Simersky, M., and Prech, L.: Small scale observation of magnetopause
motion: preliminary results of the INTERBALL project, Ann.
Geophys., 15, 562–569, doi:10.1007/s00585-997-0562-8, 1997.
Schwartz, S. J.: Hot flow anomalies near the Earth’s bow shock,
Adv. Space Res., 15, 107–116, 1995.
Schwartz, S. J., Paschmann, G., Sckopke, N., Bauer, T. M., Dunlop,
M., Fazakerley, N., and Thomsen, M. F.: Conditions for formation of hot flow anomalies at Earth’s bow shock, J. Geophys.
Res., 105, 12639–12650, 2000.
Williams, D. J., McEntire, R. W., C. Schlemn II, C., Lui, A. T.
Y., Gloeckler, G., Christon, S. P., and Gliem, F.: GEOTAIL energetic particles and ion composition instrument, J. Geomagn.
Geoelectr., 46, 39–57, 1994.
www.ann-geophys.net/29/1439/2011/