Response of magnetotail twisting to variations in IMF B y : A THEMIS case study 1–2 January 2009
T. Pitkänen1, M. Hamrin1, A. Kullen2, R. Maggiolo3, T. Karlsson2, H. Nilsson4, and P. Norqvist1
1Department of Physics, Umeå University, Umeå, Sweden,2Space and Plasma Physics, School of Electrical Engineering, Royal Institute of Technology, Stockholm, Sweden,3Belgian Institute for Space Aeronomy, Brussels, Belgium,4Swedish Institute of Space Physics, Kiruna, Sweden
Abstract Theoretical considerations, observations, and simulations have shown that the B
ycomponent of the interplanetary magnetic field (IMF) may cause twisting of the magnetotail. However, the fundamental issues, the temporal and spatial responses of the magnetotail in the twisting process, are still unresolved.
We report unique multipoint observations of the response of the magnetotail to the variations in IMF B
yon 1–2 January 2009. For the first time, estimates of the tail twisting response time at different (Time History of Events and Macroscale Interactions during Substorms, THEMIS) distances in the same event are inferred.
Using cross-correlation and timing analyses, we find that the tail twisting propagates from farther out toward the Earth and the response time increases significantly to the inner magnetosphere.
1. Introduction
The dawn-dusk, i.e., the B
ycomponent of the interplanetary magnetic field (IMF), may affect the large-scale structure and topology of the Earth’s magnetotail causing a twisting of the tail [e.g., Fairfield, 1979; Cowley, 1981; Kaymaz et al., 1994; Owen et al., 1995; Kullen and Janhunen, 2004a, 2004b; Tsyganenko et al., 2015]. For nonzero IMF B
ythe plasma sheet and cross-tail current/neutral sheet can get rotated along the Earth-Sun axis, and an additional IMF B
ycomponent appears in tail lobes and the plasma sheet that has the same sign as IMF B
y. The additional B
ycomponent in the plasma sheet is connected to a twisting of the closed field lines out of the meridional symmetry.
It is, however, not established how long it takes for the magnetotail to respond to changing IMF conditions;
i.e., what is the timescale associated with the tail twisting process. McComas et al. [1986] have discussed one cross-tail current sheet crossing of the ISEE satellites in the postmidnight sector (ISEE 1: [XGSM, YGSM]
∼
[−18,
−3]R
E, where GSM yields for geocentric solar magnetic). The crossing occurred during strongly northward IMF
∼1 h after the IMF had completed a long excursion of strong negativeB
y(at [−2, 22] R
E) and plasma sheet B
ymeasured by ISEE spacecraft was still collinear with IMF B
yof the excursion.
Motoba et al. [2011] inferred
∼1 h tail twisting response time to IMF in the postmidnight plasma sheet bound-ary layer (XGSM
∼ −11 to−14R
E, YGSM
∼−4 to−1R
E), measured from the bow shock nose during southward IMF. Even though some of the Cluster spacecraft had a separation of a couple of R
E, the response times esti- mated from the measurements by different spacecraft were practically the same. Recently, Rong et al. [2015]
have presented two events when an IMF effect was observed by Cluster in the postmidnight tail lobes after
∼1 h ([−18,−4]
R
E) and
∼1.5 h ([−16,−5]R
E), respectively, measured from the bow shock nose. These were associated with northward IMF.
Based on correlation results between IMF B
yand the initial magnetic local time location of transpolar arcs, two independent statistical studies [Fear and Milan, 2012; Kullen et al., 2015] found similar time delays, indicating a response time of the initial arc location on IMF B
yof at least 1–2 h. It is generally assumed that the formation of a transpolar arc is tightly connected to changes in the magnetotail due to IMF B
yeffects [e.g., Kullen and Janhunen, 2004a, 2004b; Milan et al., 2005]. As transpolar arcs are a predominantly northward IMF phenomenon, the results of the two studies give a good estimate of the tail response time on new IMF B
yconditions during northward IMF.
The tail twisting response time can also be studied using magnetospheric models. To our knowledge, first, such a study has been carried out by Walker et al. [1999], who used an MHD simulation and found that the
RESEARCH LETTER
10.1002/2016GL070068
Key Points:
• Magnetotail twisting response time to IMFByvariations
• Tail twisting propagates from farther in the tail toward the Earth
• The response time increases significantly at the inner magnetosphere
Correspondence to:
T. Pitkänen,
Citation:
Pitkänen, T., M. Hamrin, A. Kullen, R. Maggiolo, T. Karlsson, H. Nilsson, and P. Norqvist (2016), Response of magnetotail twisting to variations in IMFBy: A THEMIS case study 1–2 January 2009,Geophys. Res.
Lett.,43, 7822–7830, doi:10.1002/2016GL070068.
Received 19 JUN 2016 Accepted 24 JUL 2016
Accepted article online 27 JUL 2016 Published online 8 AUG 2016
©2016. American Geophysical Union.
All Rights Reserved.
Geophysical Research Letters 10.1002/2016GL070068
plasma sheet twist at XGSM
∼−20R
Eoccurred 45–60 min after an IMF B
ysign reversal during southward IMF at the dayside magnetopause. In addition, Kullen and Janhunen [2004a, 2004b] have investigated magnetotail twisting using the GUMICS-4 MHD code [Janhunen et al., 1996, 2012]. From their Figure 5, which corresponds to northward IMF, one can infer the equatorial tail B
yto change its direction in the central tail (XGSM
∼−12to
−20R
E)
∼22 min after the IMF has reversed its sign at the dayside magnetopause. Recently,Tenfjord et al.
[2015] have suggested few tens of minutes response time for the nightside near-Earth closed field lines, based on a model run using the Lyon-Fedder-Mobarry MHD model [Lyon et al., 2004; Merkin and Lyon, 2010]. Tenfjord et al. [2015] changed IMF B
yfrom zero to positive after 30 min during southward IMF.
Kullen and Janhunen [2004a, 2004b] and Walker et al. [1999] suggest that the twisting propagates from the tail flanks inward and from near-Earth tailward. Tenfjord et al. [2015] propose a propagation from near-Earth tailward, but the tail closed field line twist would be induced asymmetrically determined by the IMF B
ydirection (section 3).
In this letter, we report unique multipoint observations of the response of the magnetotail to the variations in IMF B
yon 1–2 January 2009. For the first time, estimates of the tail twisting response time at different (Time History of Events and Macroscale Interactions during Substorms, THEMIS) distances in the same event are inferred.
2. Observations
The time interval studied extends from 15:00 UT 1 January to 09:00 UT 2 January 2009. For the solar wind data, we use OMNI data at a 1 min time resolution and propagated to the bow shock nose (http://omniweb.
gsfc.nasa.gov/). The magnetotail data are provided by the five satellites of the Time History of Events and Macroscale Interactions during Substorms (THEMIS) mission [Angelopoulos, 2008]. Magnetic field data from the fluxgate magnetometer (FGM) [Auster et al., 2008] and particle data from the electrostatic analyzer (ESA) [McFadden et al., 2008] on board the satellites are studied. In section 2.2, the THEMIS data are presented in the
∼3 s resolution, but in the latter sections the data have been averaged to 1 min resolution to make itcomparable with the IMF data. Throughout the study, we will present data in the geocentric solar magnetic (GSM) coordinates.
2.1. Solar Wind and Geomagnetic Activity
During the time interval of interest, the solar wind speed mostly stays between 400 and 450 km/s, with minimum and maximum excursions of 369 and 453 km/s, respectively (data not shown). The solar wind pro- ton density varies between 1.1 and 3.7 cm
−3. The proton density maximum is measured during a period of enhanced density between
∼02:30 and 05:00 UT. This includes a transient major change in IMF B
xbetween
∼
02:30 and 04:10 UT with an excursion from negative IMF B
xvalues to positive at∼ 03:17–03:59 UT (Figure 1a).
The generally low proton density with typical solar wind speed lead to a prevailing low
≲1 nPa solar wind dynamic pressure with a maximum of
∼1.2 nPa at the proton density maximum. IMFB
zis relatively weak and fluctuates around zero with a maximum amplitude of
∼3.4 nT (Figure 1a). IMFB
yshows two major negative incursions, one shorter and more rapid
∼16:48–18:16 UT and one longer and more gradual∼21:04–21:13 UT(Figure 1b). In the following, we will focus on the gradual negative incursion of IMF B
y. Before 17:00 UT, the AE indices indicate an occurrence of a small substorm (AE maximum
∼300 nT). After that geomagnetic activityremains low.
2.2. Overview of THEMIS Observations
Figures 1g and 1h show the orbits of the THEMIS satellites during the time interval of interest. After 01:03 UT, all the five spacecraft have moved to the nightside and are traversing toward dawn and the apogee of the orbits in the postmidnight sector. For clarity, the trajectory of THEMIS D (thd) is not shown since THEMIS A (tha) and thd have a small separation. The red and green solid lines indicate the time intervals 19:04–06:01 UT and 21:25–07:21 UT for THEMIS B (thb) and THEMIS C (thc) spacecraft, respectively. These time intervals are used in the cross-correlation analysis between the IMF and THEMIS data in section 2.3. The two markers for each spacecraft indicate the positions of the spacecraft at 02:55 UT and 03:52 UT when thb and tha measure a beginning of a significant increase of B
yassociated with the preceding IMF B
ychange, respectively. These findings will be discussed in detail in section 2.4.
In Figures 1c and 1d magnetic field data measured by thb are displayed. The thb-B
zdata show that before
∼18:30 UT, the component is relatively strong and decreases with geocentric distance as the satellite moves
Figure 1.(a, b) IMF components, (c, d) THEMIS B (thb), and (e, f ) THEMIS C (thc)Bfield components during the time interval of interest. (g) Dashed lines: orbits of the THEMIS spacecraft in the GSMXYplane between 15:00 UT 1 January to 09:00 UT 2 January 2009 (trajectories only for four of five satellites drawn). Solid lines: the THEMIS B (thb) and C (thc) spacecraft trajectories during the time intervals used in the cross-correlation analysis (see details in section 2.3). The markers indicate the time 02:55 UT (03:52 UT) when thb (tha) observed the beginning of the significant increase ofBy associated with the latter IMFBychange (see section 2.4). The estimates for the propagation delay of the tail twisting change from thb toward the inner magnetosphere obtained in this study are given with arrows indicating associated distances. (h) Same as Figure 1g but for the GSMXZplane.
Geophysical Research Letters 10.1002/2016GL070068
away from the Earth (Figure 1d). After that a more tail-like field is measured with a smaller B
zcontribution.
The thb-B
xdata indicate that the spacecraft stays during most of the time above the neutral sheet. A striking feature in the thb magnetic field data is that after
∼19:00 UT, theB
ycomponent (thb-B
y) closely follows the behavior of the IMF B
ycomponent with some delay. As IMF B
ygradually changed from positive to negative, thb-B
ybehaved similarly. The occasional high-frequency variations with large amplitudes seen in the thb-B
ydata, e.g., the one occurring right after 21:30 UT, are associated with flow burst activity seen in the bulk velocity data (not shown). The magnitudes of thb-B
x(Figure 1d) and the ion beta (not shown) indicate that thb was in the plasma sheet. The magnetic pressure observed by thb stays quasi-stable between
∼20:00 UT and∼04:50 UT suggesting the field compression is not playing a role with the decrease and increase trends of
thb-B
y.
Figures 1e and 1f display the magnetic field data measured by thc. The satellite is interpreted to exit the signif- icantly dipolar inner-magnetospheric field lines at
∼01:00 UT. However, thc-B
ybegins to follow IMF B
yalready after
∼23:00 UT, similarly as thb-By. As well as for thb, the abrupt high-frequency disturbances in thc-B
y, like the one just before 01:30 UT, are associated with flow burst activity and the measurements are made in the plasma sheet (not shown). Similarly, no indication of the field compression driving the thc-B
ychange trends is seen.
Tsyganenko and Fairfield [2004] found that the geomagnetic dipole tilt angle may affect tail plasma sheet B
y. However, the dipole tilt effects on B
yappear only in certain regions in the tail. According to Petrukovich [2009, Figure 7], the dipole tilt effects should be negligible (in the region passed by thb) or cause a constant, small additional B
ycomponent (in the region passed by thc), which cannot explain the transition from negative to positive thc-B
ymeasurements.
2.3. Tail Twisting Response From Cross Correlation
To study the magnetotail response to the variations of IMF B
y, we have carried out a cross-correlation analysis between IMF B
yand the THEMIS B
ycomponents. The cross-correlation analysis is made for the two outermost spacecraft, thb and thc, since no clear and long analogous response in B
y(positive-negative-positive) is observed by the other spacecraft.
Figures 2a and 2b show IMF B
yand the IMF clock angle. The IMF clock angle is the angle between the pro- jection vector of the IMF on the YZ plane and the Z axis, increasing from the Z axis toward the positive Y direction. The clock angle values 90
∘and 270
∘thus correspond pure positive and negative IMF B
yin the YZ plane, respectively.
Figure 2c shows thb-B
y(solid line) as well as the model B
yvalues of the T96 magnetosphere model [Tsyganenko, 1996] at the thb position. The input parameters for the model have been averaged over the preceding 60 min period (black dash-dotted line). The red dash-dotted line indicates the T96 model with zero input IMF B
y. It is obvious that in this case the T96 model cannot mimic the observed large changes in the tail magnetic field configuration apparently caused by the variations in IMF B
y. Figure 2d shows the residual
ΔByobtained by subtracting B
yvalues of the T96 model with IMF B
y=0as input (red dash-dotted line in Figure 2c) from the measured B
y. The residual
ΔBywill be used in the cross-correlation analysis. The subtraction of the background, nontwisted magnetic field is expected to remove the static spatial effects from the tail B
ymea- surement caused by the moving satellite (e.g., changing azimuthal flaring of the field lines) and it has been also utilized by Motoba et al. [2011]. Figures 2e and 2f show the same as Figures 2c and 2d for thc. Also in this case the T96 model differs significantly from the observations.
The IMF B
ytime interval for the cross correlation is 19:20–02:48 UT between the two local B
ymaxima (yellow shading in Figure 3a). For thb and thc, we similarly selected the time intervals 19:04–06:01 UT and 21:25–07:21 UT, respectively, from the residual
ΔBydata (yellow shading in Figures 2d and 2f ). Note that the IMF and THEMIS time intervals have different lengths. For cross correlation, we made the IMF time series equal to the length of the THEMIS time series. For thb, this was done by adding zeros on the both sides of the IMF B
yand the clock angle series to make them span the same time interval as the thb time series. In the case of thc, zeros were added in the end of the IMF and in the beginning of the thc time series.
The sample cross correlation [e.g., Chatfield, 2004] computed using 658 data point pairs peaks at a 98 min lag
for thb independently of the IMF time series (correlograms not shown). The corresponding cross-correlation
coefficients are 0.63 and 0.65 for IMF B
yand the clock angle, respectively. The peak lags for thc computed
Figure 2.(a) IMFByand (b) clock angle. (c) The solid line is theBycomponent of the tail magnetic field measured by thb. The black (red) dash-dotted line indicates the T96 modelBy(with zero input IMFBy) when the model input parameters have been averaged over the preceding 60 min period. (d) The residual thbΔByafter subtracting the T96 model with zero input IMFBy. (e, f ) Same as Figures 2c and 2d but for thc. The yellow shading marks the time intervals used in the cross-correlation analysis.
using 722 data point pairs are 118 min (IMF B
y) and 116 min (the clock angle). The two-min difference arises due to a longer period of high clock angles compared to the corresponding period of large negative IMF B
y00–02 UT. The corresponding cross-correlation coefficients are 0.67 and 0.68, respectively.
The results imply that the magnetotail farther out responds first to variations in IMF B
y. Note that during the time intervals used in the cross-correlation analysis, thb and thc moved from
∼−11R
Eto
−20REand from
∼−5
R
Eto
−13REin XGSM, respectively (Figures 1g and 1h). The associated time intervals are long and the
Geophysical Research Letters 10.1002/2016GL070068
Figure 3.(a) IMFByand (b) clock angle. (c, e) Same as, e.g., Figure 2c for thb, thc, and tha. The vertical dashed lines and arrows indicate the onsets of the significant decrease and increases in the clock angle and the THEMISBycomponents, respectively.
spacecraft positions vary over a large range, which is expected to have an effect to the timing accuracy. How- ever, as thc is always located at shorter geocentric distances than thb, it is clear that the magnetotail farther out responds first.
2.4. Tail Twisting Response FromByIncrease
We also investigated the magnetotail response by comparing the onset times of the significant increases in
the THEMIS B
ycomponents after the latter IMF B
ychange from negative back to positive. Figure 3 shows from
top to bottom IMF B
y, the IMF clock angle, and the B
ycomponents measured by thb, thc, and tha. Also, the
T96 model fields are shown for a comparison with the actual tail measurements. The onset times have been
identified from the
ΔBycurves (e.g., Figures 2d and 2f ) and are marked into Figures 3c–3e by green vertical
dashed lines and arrows. No clear onset time of the increase in IMF B
ycan be seen (Figure 3a), but instead, one can identify an onset in the clock angle data at 01:17 UT (Figure 3b).
The first onset is observed first by thb at 02:55 UT (Figures 3c and 1g), 98 min after the IMF clock angle onset, which matches with the delay obtained by the cross-correlation analysis. Spacecraft thc observes the onset 13 min later at 03:08 UT (Figure 3d). The 13 min delay between the onsets at thb and thc (111 min between the IMF clock angle onset and the thc onset) is of the order of the 20 min (116 min) delay obtained using the cross-correlation analysis. These correspondences suggest that the timing uncertainties arising from the long time interval and associated large satellite location change may not significantly influence the cross-correlation analysis.
The tha satellite in the inner magnetosphere observes a signature of an onset of the B
yincrease at 03:52 UT.
If the increase in tha-B
yis associated with the same process causing the B
ychanges measured by thb and thc, the disturbance reaches the tha spacecraft in the inner magnetosphere 57 min after thb (44 min after thc).
This is a significant longer delay compared to the delay between thb and thc. We have also checked the delay to the satellite, which followed tha (Figures 1g and 1h). We found a possible, even though weaker, signature of an onset of the-B
yincrease at 04:06 UT, i.e., 71 min after thb (data not shown). Also, thd shows similar magnetic field signatures as tha, as the two spacecraft have a small separation (not shown).
The negative bay observed by tha between
∼05:20 and 06:00 UT is also seen by the other THEMIS inner-magnetosphere satellites. It is associated with small amplitude (10–20 km/s) velocity oscillations in the earthward-dawnward to tailward-duskward direction and could be an inner-magnetospheric signature of oscillatory flow braking of fast earthward flows [e.g., Panov et al., 2010, 2013].
3. Summary and Discussion
We have presented unique multipoint observations of the response of the magnetotail to variations of IMF B
yon 1–2 January 2009, during which the THEMIS satellites are aligned well inside the magnetospheric tail at different geocentric distances in the midnight-postmidnight sector. For the first time, both the temporal and spatial magnetotail response to IMF B
yduring one single event can be inferred.
During the event, IMF B
yshowed a gradual negative incursion from the positive direction lasting
∼5 h. ThisIMF behavior was conveyed to the tail plasma sheet B
ywith a delay depending on the tailward distance from the Earth, as measured by the THEMIS satellites. The IMF B
zwas weak and fluctuated around zero.
By using both the cross-correlation analysis and the comparison of the onset times of magnetic field changes at the bow shock nose and in the tail, we have inferred the response times of the B
ycomponent to the IMF B
ychanges at different tail distances. We interpret the tail B
ychanges as changes in the field line twisting in the magnetotail. The results indicate that the magnetotail responded in
∼100 min, 110–120 min, and160–170 min at thb, thc, and tha/the distances, respectively, to the IMF lagged to the bow shock nose (OMNI data) (Figures 1g and 1h). These results suggest that the magnetotail farther out responds first and the response time increases significantly as geocentric distance decreases to inner-magnetospheric distances.
The 100 min and 110–120 min delays in tail B
yresponse for thb and thc, respectively, are longer than the 60 min delay estimate deduced from McComas et al. [1986] and the 60–90 min delays inferred by Rong et al.
[2015], the latter obtained in the tail lobes. The 60 min delay by Motoba et al. [2011] lies in the same range as the previous estimates. Notably, the measurements analyzed in all these studies have been made during more active geomagnetic conditions than the measurements analyzed in the present study. Thus, the quiet geomagnetic conditions may explain the longer time responses in the present study. Specifically, the obser- vational evidence presented by Rong et al. [2015] that stronger convection is associated with shorter response time in the lobes supports this explanation.
Also, as discussed in detail by Kullen et al. [2015], there exists a very high correlation between IMF B
yand the initial transpolar arc location with a delay of about 1–2 h up to many hours back in time (which appears due to the statistically rather constant IMF B
yvalues the hours before transpolar arc events). As the transpolar arcs appear during predominantly northward IMF [Fear and Milan, 2012; Kullen et al., 2015, and references therein]
and their formation is strongly controlled by IMF B
y-induced changes in the magnetotail [Kullen and Janhunen,
2004a, 2004b], the time delay of at least 1–2 h should correspond to the tail response time on new IMF B
yconditions during quiet times. However, it is not known whether the way how IMF B
yaffects the tail is the
same for northward and southward IMF as tail reconnection is predominantly a southward IMF phenomenon.
Geophysical Research Letters 10.1002/2016GL070068
Between 23 and 03 UT, we find that the fraction of induced tail B
y(ΔB
y/IMF B
y) varies between 200 and 500%
for both thb and thc (data not shown). It is known that the induced B
yis nonuniformly distributed within the plasma sheet with highest values at the neutral sheet [e.g., Kaymaz et al., 1994; Kullen and Janhunen, 2004a, 2004b]. However, the induced-B
yvalues found in the present study are still a magnitude higher than in previous observational and simulation studies even at the neutral sheet [e.g., Kaymaz et al., 1994; Kullen and Janhunen, 2004a, 2004b; Cao et al., 2014, and references therein]. In the inner magnetosphere at the geosyn- chronous orbit, the proportionality between magnetospheric B
yand IMF B
yhas statistically shown to reach
∼80% in the midnight sector [Wing et al., 1995]. At∼22:00 UT, thc is located at about that region and we
find thc-ΔB
y∼−2.1 nT and IMFB
y∼ +0.2 nT, when using the 118 min lag for IMFB
y. Even using a longer, 160–170 min lag for the IMF, which was deduced using the inner-magnetospheric THEMIS data, the signs remain opposite. However, the results by Wing et al.[1995] show large scattering for both positive and negative IMF B
ysuggesting that observations of a lack of/an opposite extra B
yin the inner-magnetospheric distances are not totally unusual. In conclusion, the present event may represent a somewhat extreme event during which the mechanism generating the twisted closed tail field lines is working effectively. A more plausible contributing factor to the high efficiency could be the limited ability of T96 to model the reference, non-IMF B
y-affected field. The values of the solar wind and IMF parameters are not found to be extreme.
The IMF B
y“penetration” into the tail lobes can be understood to be caused by dawn-dusk asymmetric dayside reconnection under nonzero, favorably strong or dominating IMF B
y. On the contrary, the question how does additional B
yconvey into the closed tail field lines is yet unanswered. While the propagation of the IMF-induced B
yfrom outside (flanks)-in (central tail) appears in the MHD simulations by Walker et al. [1999]
and Kullen and Janhunen [2004a, 2004b], the results of the present study regarding the propagation in the Sun-Earth direction suggest contrarywise. The strong northward IMF conditions in the simulation study by Kullen and Janhunen [2004a, 2004b] (high-latitude lobe reconnection and no tail reconnection seen in the simulations) could be part of the explanation why the propagation of the IMF-induced new B
ydirection is directed in a different direction in the present event as compared to their simulations.
Hau and Erickson [1995] have proposed that IMF-induced plasma sheet B
ycould be generated by the differ- ential azimuthal shear motion of magnetic field lines and that the earthward convection could also enhance B
y(in the direction of IMF B
y). Khurana et al. [1996] have presented an analogous model according to which the shear flows between the northern and the southern halves of the plasma sheet are driven by opposite cross-tail flows in the northern and southern lobe/mantle regions generating B
yon the closed field lines. The cross-tail flows in the lobes/mantles would be driven by magnetic pressure gradients generated by asymmet- ric addition of the magnetic flux into the tail lobes during dawn-dusk asymmetric dayside reconnection. The results studied by Tenfjord et al. [2015] are in accordance with this scenario and suggest that the timescale in which the nightside near-Earth closed field lines respond to IMF B
yis faster than the convection time from the dayside magnetosphere boundaries to the tail reconnection site. On the other hand, an indication of a correlation between the IMF B
yand the YGSM component of the earthward convective fast flows has been found, which is consistent with untwisting rather than twisting of the tail magnetic field lines [e.g., Grocott et al., 2007; Pitkänen et al., 2013, 2015]. A guide field reconnection determined by the IMF B
ydirection could play a role in conveying the IMF B
yeffects into the plasma sheet.
The present THEMIS observations invoke further questions like why should the response time of the tail twisting delay from farther out from the tail to inner-magnetospheric distances? What is the mechanism generating the tail twist? To answer these questions, the associated plasma flows must be investigated. The analysis of the magnetospheric flows associated with this event is beyond the scope of the present study, but it is planned for a future study.
References
Angelopoulos, V. (2008), The THEMIS mission,Space Sci. Rev.,141, 5–34, doi:10.1007/s11214-008-9336-1.
Auster, H. U., et al. (2008), The THEMIS fluxgate magnetometer,Space Sci. Rev.,141, 235–264, doi:10.1007/s11214-008-9365-9.
Cao, J. B., et al. (2014), Dependence of IMFBypenetration into the neutral sheet on IMFBzand geomagnetic activity,J. Geophys. Res. Space Physics,119, 5279–5285, doi:10.1002/2014JA019827.
Chatfield, C. (2004),The Analysis of Time Series: An Introduction, Chapman & Hall/CRC, Boca Raton, Fla.
Cowley, S. W. H. (1981), Magnetospheric asymmetries associated with the Y -component of the IMF,Planet. Space Sci.,29, 79–96, doi:10.1016/0032-0633(81)90141-0.
Fairfield, D. H. (1979), On the average configuration of the geomagnetic field,J. Geophys. Res.,84(A5), 1950–1958, doi:10.1029/JA084iA05p01950.
Acknowledgments
The authors thank the THEMIS FGM and ESA PIs and teams as well as the PI of the THEMIS mission for the THEMIS data. The THEMIS mission is hosted by UCLA/IGPP and UC Berkeley/SSL. In addition, we acknowledge GSFC SPDF/OMNIWeb for the solar wind data and WDC for Geomagnetism, Kyoto, for theAE index data. The authors would also like to thank A. Vaivads, C. Norgren, D.B. Graham, and A. De Spiegeleer for useful discussions. The work by T. Pitkänen was supported by the SNSB grant 77/14.
Fear, R. C., and S. E. Milan (2012), The IMF dependence of the local time of transpolar arcs: Implications for formation mechanism,J. Geophys.
Res.,117, A03213, doi:10.1029/2011JA017209.
Grocott, A., et al. (2007), Multi-scale observations of magnetotail flux transport during IMF-northward non-substorm intervals, Ann. Geophys.,25, 1709–1720, doi:10.5194/angeo-25-1709-2007.
Hau, L.-N., and G. M. Erickson (1995), Penetration of the interplanetary magnetic fieldByinto Earth’s plasma sheet,21,100, 21,745–21,751.
Janhunen, P., H. E. J. Koskinen, and T. I. Pulkkinen (1996), A new global ionosphere-magnetosphere coupling simulation utilizing locally varying time step, paper ESA SP-389 presented at T3rd International Conference on Substorms, Versailles, France, 12–17 May.
Janhunen, P., et al. (2012), The GUMICS-4 global MHD magnetosphere-ionosphere coupling simulation,J. Atmos. Sol. Ter. Phys.,80, 48–59, doi:10.1016/j.jastp.2012.03.006.
Kaymaz, Z., G. L. Siscoe, J. G. Luhmann, R. P. Lepping, and C. T. Russell (1994), Interplanetary magnetic field control of magnetotail field geometry: IMP 8 observations,J. Geophys. Res.,99(A6), 11,113–11,126, doi:10.1029/94JA00300.
Khurana, K. K., R. J. Walker, and T. Ogino (1996), Magnetospheric convection in the presence of interplanetary magnetic fieldBy: A conceptual model and simulations,J. Geophys. Res.,101, 4907–4916.
Kullen, A., and P. Janhunen (2004a), Relation of polar auroral arcs to magnetotail twisting and IMF rotation: A systematic MHD simulation study,Ann. Geophys.,22, 951–970, doi:10.5194/angeo-22-951-2004.
Kullen, A., and P. Janhunen (2004b), Erratum to Relation of polar auroral arcs to magnetotail twisting and IMF rotation: A systematic MHD simulation study,Ann. Geophys.,22, 2655–2655, doi:10.5194/angeo-22-2655-2004.
Kullen, A., R. C. Fear, S. E. Milan, J. A. Carter, and T. Karlsson (2015), The statistical difference between bending arcs and regular polar arcs, J. Geophys. Res. Space Physics,120, 10,443–10,465, doi:10.1002/2015JA021298.
Lyon, J. G., J. A. Fedder, and C. M. Mobarry (2004), The Lyon-Fedder-Mobarry (LFM) global MHD magnetospheric simulation code,J. Atmos.
Sol. Terr. Phys.,66(15–16), 1333–1350, doi:10.1016/j.jastp.2004.03.020.
McComas, D. J., C. T. Russell, R. C. Elphic, and S. J. Bame (1986), The near-Earth cross-tail current sheet: Detailed ISEE 1 and 2 case studies, J. Geophys. Res.,91(A4), 4287–4301, doi:10.1029/JA091iA04p04287.
McFadden, J. P., C. W. Carlson, D. Larson, J. Bonnell, F. Mozer, V. Angelopoulos, K.-H. Glassmeier, and U. Auster (2008), THEMIS ESA first science results and performance issues,Space Sci. Rev.,141, 477–508, doi:10.1007/s11214-008-9433-1.
Merkin, V. G., and J. G. Lyon (2010), Effects of the low-latitude ionospheric boundary condition on the global magnetosphere,J. Geophys.
Res.,115, A10202, doi:10.1029/2010JA015461.
Milan, S. E., B. Hubert, and A. Grocott (2005), Formation and motion of a transpolar arc in response to dayside and nightside reconnection, J. Geophys. Res.,110, A01212, doi:10.1029/2004JA010835.
Motoba, T., K. Hosokawa, Y. Ogawa, N. Sato, A. Kadokura, S. C. Buchert, and H. Rème (2011), In situ evidence for interplanetary magnetic field induced tail twisting associated with relative displacement of conjugate auroral features,J. Geophys. Res.,116, A04209, doi:10.1029/2010JA016206.
Owen, C. J., J. A. Slavin, I. G. Richardson, N. Murphy, and R. J. Hynds (1995), Average motion, structure and orientation of the distant magnetotail determined from remote sensing of the edge of the plasma sheet boundary layer withE>35 keV ions,J. Geophys. Res.,100, 185–204, doi:10.1029/94JA02417.
Panov, E. V., et al. (2010), Multiple overshoot and rebound of a bursty bulk flow,Geophys. Res. Lett.,37, L08103, doi:10.1029/2009GL041971.
Panov, E. V., M. V. Kubyshkina, R. Nakamura, W. Baumjohann, V. Angelopoulos, V. A. Sergeev, and A. A. Petrukovich (2013), Oscillatory flow braking in the magnetotail: THEMIS statistics,Geophys. Res. Lett.,40, 2505–2510, doi:10.1002/grl.50407.
Petrukovich, A. A. (2009), Dipole tilt effects in plasma sheetBy: Statistical model and extreme values,Ann. Geophys.,27, 1343–1352, doi:10.5194/angeo-27-1343-2009.
Pitkänen, T., M. Hamrin, P. Norqvist, T. Karlsson, and H. Nilsson (2013), IMF dependence of the azimuthal direction of earthward magnetotail fast flows,Geophys. Res. Lett.,40, 5598–5604, doi:10.1002/2013GL058136.
Pitkänen, T., M. Hamrin, P. Norqvist, T. Karlsson, H. Nilsson, A. Kullen, S. M. Imber, and S. E. Milan (2015), Azimuthal velocity shear within an Earthward fast flow—Further evidence for magnetotail untwisting?,Ann. Geophys.,33, 245–255, doi:10.5194/angeo-33-245-2015.
Rong, Z. J., A. T. Y. Lui, W. X. Wan, Y. Y. Yang, C. Shen, A. A. Petrukovich, Y. C. Zhang, T. L. Zhang, and Y. Wei (2015), Time delay of interplanetary magnetic field penetration into Earth’s magnetotail,J. Geophys. Res. Space Physics,119, 3406–3414, doi:10.1002/2014JA020452.
Tenfjord, P., N. Østgaard, K. Snekvik, K. M. Laundal, J. P. Reistad, S. Haaland, and S. E. Milan (2015), How the IMFByinduces aBycomponent in the closed magnetosphere and how it leads to asymmetric currents and convection patterns in the two hemispheres,J. Geophys. Res.
Space Physics,120, 9368–9384, doi:10.1002/2015JA021579.
Tsyganenko, N. A. (1996), Effects of the solar wind conditions on the global magnetospheric configuration as deduced from data-based field models, paper ESA SP 389 presented at 3rd International Conference of Substorms (ICS-3), pp. 181–185, Versailles, France, 12–17 May.
Tsyganenko, N. A., and D. H. Fairfield (2004), Global shape of the magnetotail current sheet as derived from Geotail and Polar data, J. Geophys. Res.,109, A03218, doi:10.1029/2003JA010062.
Tsyganenko, N. A., V. A. Andreeva, and E. I. Gordeev (2015), Internally and externally induced deformations of the magnetospheric equatorial current as inferred from spacecraft data,Ann. Geophys.,33, 1–11, doi:10.5194/angeo-33-1-2015.
Walker, R. J., R. L. Richard, T. Ogino, and M. Ashour-Abdalla (1999), The response of the magnetotail to changes in the IMF orientation:
The magnetotail’s long memory,Phys. Chem. Earth,24, 221–227.
Wing, S., P. T. Newell, D. G. Sibeck, and K. B. Baker (1995), A large statistical study of the entry of interplanetary magnetic field Y -component into the magnetosphere,Geophys. Res. Lett.,22, 2083–2086.
