On the stability of high-latitude plasma convection during northward IMF: an event study

TRITA-EE 2007:072

On the Stability of High-latitude Plasma Convection During Northward IMF: An Event Study

J. A. Cumnock, ^1,2 J. B. H. Baker, ³ and L. G. Blomberg ²

1

Center for Space Sciences, University of Texas at Dallas

2

Space & Plasma Physics, School of Electrical Engineering, KTH, Stockholm

3

Applied Physics Laboratory, Johns Hopkins University, Maryland

November 2007, Stockholm

Royal Institute of Technology Report TRITA-EE 2007:072

Abstract

We investigate the stability of the ionospheric convection pattern during northward IMF by studying an event where two DMSP satellites repeatedly traversed the Southern polar region. The Cluster satellite’s ionospheric footprint is located near the DMSP satellite tracks, moving slowly in the noon-midnight direction. TIMED/GUVI data confirm the presence of auroral activity at high latitude. SuperDARN plasma velocity data partially complete the picture. From the event studied we conclude that whereas the DMSP satellites observe local variations in the convection pattern between consecutive passes, Cluster confirms the existence of persistent sunward convection in the high-latitude ionosphere on a time scale of several hours.

1. Introduction

The primary source of energy driving ionospheric convection at high latitudes is the interaction of the solar wind with the Earth's magnetosphere. The flow of the solar wind past the magnetosphere induces a large-scale electrostatic potential drop across the magnetosphere, most of which projects into the ionosphere. This drives large-scale plasma convection and currents inside the magnetosphere and ionosphere, with their magnitudes and configurations dependent on the orientation of the interplanetary magnetic field (IMF). As the interplanetary magnetic field rotates from southward to northward the number of convection reversal boundaries (CRB’s) in the high-latitude plasma flow increases from two to four (or more) and the CRB’s move to higher latitudes.

Regions of sunward plasma flow at highest latitudes are characteristic of most convection signatures observed when the IMF has a northward component [Reiff and Burch, 1985;

Crooker, 1992; Cumnock et al., 1995 and references therein]. In addition, some of the CRB’s, located at highest latitudes may be associated with precipitating particles on closed field lines and transpolar arcs (TPA) or theta aurora [Nielsen et al., 1990;

Cumnock et al., 2002 and references therein]. The associated ionospheric convection patterns observed during northward IMF consist of a variety of configurations including one-cell patterns and multiple-cell patterns. The number of cells appears to have a first- order dependence on the ratio of B

to B

, resulting in an increasing number of cells as the IMF becomes more strongly northward [Potemra et al., 1984; Reiff and Burch, 1985;

Heelis et al., 1986; Cumnock et al., 1995, and references therein]. There are also a

variety of fundamentally different large-scale auroral distributions during northward IMF

[Elphinstone et al., 1991, Kullen et al., 2002, Cumnock and Blomberg, 2004]; including

TPA’s, which move across the entire polar region [Cumnock et al., 2002]. The associated

magnetospheric convection is also more complex than that which occurs during

southward IMF [Tanaka, 1999; Tanaka et al., 2004; Kullen et al., 2004; Cumnock et al.,

2006 and references therein]. This was first illustrated by Reiff [1982] in a schematic

(their Fig 4a) which shows the magnetospheric field lines mapped to various points in a

4-cell convection pattern in the ionosphere.

2

We investigate the stability of the ionospheric convection pattern during northward IMF by studying an event where two DMSP satellites repeatedly traversed the Southern polar region. The Cluster satellite’s ionospheric footprint is located near the DMSP satellite tracks, moving slowly in the noon-midnight direction.

2. Observations

2.1 Solar Wind

Figure 1 shows IMF measurements from one-minute averaged OMNI data (http://omniweb.gsfc.nasa.gov). The dataset consists of solar wind measurements from the solar wind monitor which is closest to the Earth’s bow shock and are then propagated in time to correspond to the solar wind conditions at the bow shock. During the time period of interest on 8 February 2004 the correction is one hour. IMF B

is mostly negative, both IMF B

and B

are generally positive. |B

/B

| is generally increasing until about 17 UT. The Solar wind density varies between 3 and 6 cm

, and its velocity varies between 360 and 430 km/s (not shown).

2.2 TIMED

The Global Ultraviolet Imager (GUVI) on the TIMED (Thermosphere Ionosphere Mesosphere Energetics and Dynamics) satellite is a cross-track far-ultraviolet (115 to 180 nm) scanning imaging spectrograph that provides horizon-to-horizon images in five selectable wavelength intervals (HI 121.6 nm, OI 130.4 nm, OI 135.6 nm, and N2 Lyman-Birge-Hopfield bands 140 to 150 nm and 165 to 180 nm) [Christensen et al., 2003]. Figure 2 shows three GUVI images of the Southern Hemisphere aurora in corrected geomagnetic coordinates; the sun is toward the top, with magnetic local time denoted by the dashed lines, the lowest latitude is at 60 degrees. The 17 minute scans are centered at 13:18, 14:56 and 16:35 UT. At 13:18 UT the dawnside oval has expanded poleward as expected in the Southern Hemisphere for positive IMF B

and Northward IMF [Hones et al., 1989 and Cumnock et al., 1997]. The dawnside oval expands further poleward as IMF B

decreases with respect to B

(second and third images). Individual arcs are seen embedded in the expanded dawnside oval in all three images.

2.3 Cluster

Figure 3 shows the Cluster 4 (Tango) orbital track (from 12 to 17 UT); plotted by the Orbit Visualization Tool (http://ovt.irfu.se), in both the x-z (a) and y-z (b) GSE planes, with a grid denoting the magnetopause. High-altitude satellites, like Cluster, are more or less stationary on the time scale of an orbit of a low-altitude satellite like DMSP. During the time shown the satellite passes through the magnetopause into the dayside magnetosphere, moving from a distance of about 12 R

to 7 R

from Earth. All 4 Cluster satellites are very close to one another at this time, thus measuring similar large-scale features.

The Electric Field and Wave instrument (EFW) on Cluster [Gustafsson et al., 2001]

measures DC electric fields in the range 0.1 to 700 mV/m in two axes. The spin-fit

electric field data are available almost continuously during the 57-hour orbit. Figure 4

shows Cluster 4 EFW data from 8 February 2004 during the same time period as Figure

3. The four panels show the (1) negative of the satellite potential, (2) sunward electric field (~GSE x-component), (3) duskward electric field (~GSE y-component), and (4) Sigma, the standard deviation in the fitting of a sine wave to the electric field data in the spinning frame of the satellite, an indicator of the level of fluctuations of the electric field.

The satellite potential is an indicator of the ambient plasma density; its value goes more positive in thin plasmas, such as on open field lines [Gustafsson et al., 2001]. Note that Vps in Figure 4 is the negative of the satellite potential. The satellite potential (and ion data) indicates that the satellite passes the bow shock between 5 and 6 UT (not shown).

The satellite then enters the dayside magnetosheath, first passing into the dayside magnetosphere at about 14 UT, however, the spacecraft potential and Sigma indicate that the magnetopause moves several times with respect to the satellite between 14 and 15 UT.

2.4 DMSP and SuperDARN

The Defense Meteorological Satellite Program's (DMSP) satellites provide measurements of ionospheric plasma flows and precipitating particles for several Southern Hemisphere passes during the time period of interest. The line-of-sight velocity data provided by the SuperDARN radar network, together with DMSP measurements, are fitted to an IMF- driven model to allow the pattern to be estimated where no data are present.

DMSP F13 passes in the dawn-to-dusk direction and F14 from about 8 to 19 MLT.

Figures 5 and 6 show (at the top) the combined SuperDARN/DMSP plasma flows in the Southern Hemisphere. The red X marks the Cluster footprint. A 28 minute integration period is used because of the 15 minute transit time of the satellite and the fact that F14 precedes F13 by about 9 minutes. For each of the time periods similar flow patterns are seen in both F13 and F14 data consistent with a fairly stable IMF during the time it takes both satellites to cross the auroral region. The lower panels show the corresponding F13 precipitating particle and plasma flow data (plotted dawn (on left side of plot) to dusk (right side)) which include, from top to bottom, (1) electron and ion integral energy flux, (2) electron and ion average energy, (3) precipitating electron spectrogram, (4) precipitating ion spectrogram, and (5) cross-track horizontal plasma drift.

3. Discussion and Summary

The topology of the convection cells in the ionosphere as identified from

DMSP/SuperDARN data (and correlated with auroral signatures) can be associated with

features in the Cluster electric field data. For example, a dawnward pointing electric field

measured in the dayside magnetosphere (negative GSE y-component) in general maps to

sunward flow in the ionosphere. By making some simplifying assumptions (for example,

square flux tube cross-sections, and equipotential magnetic field lines), the electric field

measured in the magnetosphere can easily be mapped to the ionosphere and converted to

E x B drift.

4

Figure 5 shows the ionospheric convection from 1346 to 1414 UT. DMSP F14 reaches its highest latitude at 1355 UT, F13 at 1404 UT. Figure 1 shows that during this time period IMF |B

| is less than IMF |B

|. IMF B

and B

are positive resulting in an expanded dawnside sunward flow region in the Southern hemisphere (part of a pattern with two regions of sunward flow located equatorward of an antisunward flow region). The resulting pattern is that of a spatially dominant cell with anti-clockwise rotation and a smaller clockwise rotating cell on the duskside. DMSP particle data show that the dawnside sunward flow is co-located with an expanded auroral region (left side of the lower panel). An arc signature is located at the poleward edge of the dawnside oval expansion. The energy flux and average energy of the precipitating electrons are almost as large as those measured in the auroral oval. Dark red arrows denote the arc poleward of the oval separated from the oval by a small gap, as seen in the electron and ion spectra and the corresponding location in the ionospheric plasma flow. The DMSP particle measurements are consistent with the signature seen in the first two GUVI images (Figure 2) taken just before and after the DMSP pass. The particle data also indicate that the high-latitude antisunward flow region is on open field lines (identified by polar rain in the electron signature and a lack of ion precipitation). The Cluster footprint is located in the very weak antisunward flow just poleward of the dawnside CRB. This flow maps to a weak or nonexistent E-field in the dusk-dawn direction as the Cluster satellite enters the dayside magnetosphere.

Figure 6 shows the ionospheric convection from 1526 to 1554 UT. DMSP F14 reaches its highest latitude at 1534 UT, F13 at 1543 UT. Figure 1 shows that IMF |B

| has decreased with respect to IMF |B

| resulting in the expansion and bifurcation of the dawnside sunward flow region forming a total of three sunward flow regions and two antisunward flow regions. The highest latitude sunward flow region extends across the noon-midnight meridian resulting in “reverse convection” at highest latitudes. DMSP particle data show arc signatures (associated with closed field lines) located in the equatorward (dawnside) half of the high-latitude sunward flow region. There are arc signatures closer to the noon-midnight meridian (see red arrows) than in the previous pass; however the satellite track is at lower latitudes than in the previous pass. The DMSP satellites pass over the high-latitude region between the times the second and third GUVI images were taken (Figure 2). The Cluster footprint is located just to the nightside of the high-latitude sunward flow region. A dawnward E-field in the dayside magnetosphere maps to the high-latitude sunward flow region in the ionosphere measured by DMSP at approximately 1543 UT. Cluster observes a 1 mV/m dawnward electric field. Mapped to the ionosphere, assuming no parallel potential drop and that the field scales as the square root of the magnitude of the magnetic field, and converted to an E x B drift velocity, this corresponds to a sunward ionospheric plasma flow velocity of 500 m/s, in good qualitative agreement with DMSP observations in the vicinity of the Cluster magnetic footprint.

The Cluster footprint moves from the dayside to the nightside of the dawnside polar

region from about 1530 to 1730 UT. DMSP particle data (not shown in all cases)

indicate that Cluster maps to an expanded dawnside auroral oval in the ionosphere on

closed field lines (for example, see Figure 6). Cluster measures dawnward electric field

continuously during this period, implying sunward flow in the dawnside ionosphere also between DMSP passes. The Cluster E-field measurements are consistent with weak dawnward flows then stronger more structured flows as the Cluster footprint moves toward the nightside auroral region. Whereas the DMSP flow observations indicate that the details of the convection pattern change during the event, the Cluster data confirm a stable presence of generally sunward flow in the ionosphere for an extended period of time while the IMF remains northward.

Acknowledgements.

The authors thank Fred Rich for providing the DMSP F13 and F14 SSIES and SSJ4 data;

the ACE MAG and SWEPAM instrument teams and the ACE Science Centre for providing the ACE data. We thank K. Stasiewicz, M. Khotyaintsev and Y. Khotyaintsev for providing the Orbit Visualization Tool (http://ovt.irfu.se). Work at the University of Texas at Dallas was supported by NSF grants ATM0536868 and ATM0228209 and NASA grant NNG04GG84G. Work at the Royal Institute of Technology was supported by the Swedish National Space Board and the Alfvén Laboratory Centre for Space and Fusion Plasma Physics. The work at the Johns Hopkins University Applied Physics Laboratory was supported by NASA grants NNG04GA12G and NNG05GE25G.

References

Christensen, A. B., et al., Initial observations with the Global Ultraviolet Imager (GUVI) in the NASA TIMED satellite mission, J. Geophys. Res., 108(A12), 1451, doi:10.1029/2003JA009918, 2003.

Crooker, N. U., Reverse convection, J. Geophys. Res., 97, 19363, 1992.

Cumnock, J. A., L. G. Blomberg, I. I. Alexeev, E. S. Belenkaya, S. Yu. Bobrovnikov and V. V. Kalegaev, Simultaneous Polar Aurorae and Modelled Convection Patterns in Both Hemispheres, Adv. Space Research, 38, 1685, doi:10.1016/j.asr.2005.04.105, 2006.

Cumnock, J. A. and L. G. Blomberg, Transpolar Arc Evolution and Associated Potential Patterns, Ann. Geophys., 22, 1213, 2004.

Cumnock, J. A., J. R. Sharber, R. A. Heelis, L. G. Blomberg, G. A. Germany, J. F. Spann and W. R. Coley, Interplanetary magnetic field control of theta aurora development, J.

Geophys. Res., 107(A7), 1108, doi:10.1029/2001JA009126, 2002.

Cumnock J. A., J. R. Sharber, M. R. Hairston, R. A. Heelis, and J. D. Craven, Evolution of the global auroral pattern during positive IMF B

and varying IMF B

conditions, J.

Geophys. Res., 102, 17489, 1997.

Cumnock, J. A., R. A. Heelis, M. R. Hairston and P. T. Newell, The high latitude ionospheric convection pattern during steady northward interplanetary magnetic field, J.

Geophys. Res., 100, 14537, 1995.

6

Elphinstone, R. D., D. Hearn, J. S. Murphree, and L. L. Cogger, Mapping using the Tsyganenko long magnetospheric model and its relationship to Viking auroral images, J.

Geophys. Res., 96, 1467, 1991.

Gustafsson, G., et al., First results of electric field and density observations by Cluster EFW based on initial months of operation, Ann. Geophys., 19, 1219, 2001.

Heelis, R. A., P. H. Reiff, J. D. Winningham and W. B. Hanson, Ionospheric convection signatures observed by DE 2 during northward interplanetary magnetic field, J. Geophys.

Res., 91, 5817, 1986.

Hones, E. W. J., J. D. Craven, L. A. Frank, D. S. Evans, and P. T. Newell, The horse- collar aurora: A frequent pattern of the aurora in quiet times, Geophys. Res. Lett., 16, 37, 1989.

Kullen, A. and P. Janhunen, Relation of polar arcs to magnetotail twisting and IMF Rotations: A systematic MHD simulation study, Ann. Geophys., 22, 951, 2004.

Kullen, A., M. Brittnacher, J. A. Cumnock and L. G. Blomberg, Solar wind dependence of the occurrence and motion of polar auroral arcs: A statistical study, J. Geophys. Res., 107(A11), 1362, doi:10.1029/2002JA009245, 2002.

Nielsen, E., J. D. Craven, L. A. Frank and R. A. Heelis, Ionospheric flows associated with a transpolar arc, J. Geophys. Res., 95, 21169, 1990.

Potemra, T. A., L. J. Zanetti, P. F. Bythrow, A. T. Y. Lui and T. Iijima, B

-dependent convection patterns during northward interplanetary magnetic field, J. Geophys. Res., 89, 9753, 1984.

Reiff, P. H., Sunward convection in both polar caps, J. Geophys. Res., 87, 5976, 1982.

Reiff, P. H. and J. L. Burch, IMF B

-dependent plasma flow and Birkeland currents in the dayside magnetosphere, 2, A global model for northward and southward IMF, J.

Geophys. Res., 90, 1595, 1985.

Tanaka, T., Configuration of the magnetosphere-ionosphere convection system under northward IMF conditions with nonzero IMF B

, J. Geophys. Res., 104, 14683, 1999.

Tanaka, T., T. Obara and M. Kunitake, Formation of the theta aurora by a transient

convection during northward interplanetary magnetic field, J. Geophys. Res., 109,

A09201, doi:10.1029/2003JA010271, 2004.

Figure 1. IMF measurements from one-minute averaged OMNI data during the time

period of interest on 8 February 2004. From top to bottom are IMF magnitude, B

, B

and

B

. The time scale is shifted one hour to reflect the propagation time from the satellite to

the Earth.

8

Figure 2. Three GUVI images of the Southern Hemisphere aurora in corrected

geomagnetic coordinates; the sun is toward the top, with magnetic local time denoted by

the dashed lines, the lowest latitude is at 60 degrees. The 17 minute scans are centered at

13:18, 14:56 and 16:35 UT.

Figure 3. The Cluster 4 (Tango) orbital track plotted by the Orbit Visualization Tool

(http://ovt.irfu.se), in both the x-z (a) and y-z (b) GSE planes, with a grid denoting the

magnetopause.

10

Figure 4. Cluster 4 EFW data from 8 February 2004 during the same time period as

Figure 3. The four panels show the (1) negative of the satellite potential, (2) sunward

electric field (~GSE x-component), (3) duskward electric field (~GSE y-component), and

(4) Sigma, the standard deviation in the fitting of a sine wave to the electric field data in

the spinning frame of the satellite, an indicator of the level of fluctuations of the electric

field.

Figure 5. Shown are the combined SuperDARN/DMSP plasma flows from 1346 to 1414 UT on 8 Feb 2004 in the Southern Hemisphere. The red X marks the Cluster footprint.

Dark red arrows denote the arc signature as seen in the electron and ion spectra and the

corresponding location in the ionospheric plasma flow. A 28 minute integration period is

used because of the 20 minute transit time of the satellite and the fact that F14 precedes

12

F13 by about 9 minutes. Flow patterns seen in both F13 and F14 data are consistent with

a stable IMF during the time it takes both satellites to cross the auroral region. The lower

panels show the corresponding F13 precipitating particle and plasma flow data (plotted

dawn (on left side of plot) to dusk (right side) which include, from top to bottom, (1)

electron and ion integral energy flux, (2) electron and ion average energy, (3)

precipitating electron spectrogram, (4) precipitating ion spectrogram, and (5) cross-track

horizontal plasma drift.

Figure 6. Same format as Figure 5, but from 1526 to 1554 UT.