Field Aligned Currents in the High Latitude, High Altitude Magnetosphere: POLAR Initial Results

C. T. Russell1, X-W Zhou1, Guan Le1, P. H. Reiff 2, J. G. Luhmann3, C. A. Cattell4 and H. Kawano5


1. Institute of Geophysics and Planetary Physics, University of California, Los Angeles, CA 90095-1567
2. Center for Space Physics, Rice University, Houston, TX 77251-1892
3. Space Sciences Laboratory, University of California, Berkeley, CA 94720
4. School of Physics & Astronomy, University of Minnesota, Minneapolis, MN 55455-0112
5. Nagoya University, STE Lab., Honahara 3-13, Toyokawa, Aichi 442, Japan

Originally published in:
Geophysical Research Letters, 24, 1455-1458, 1997.



Magnetic field measurements obtained by the POLAR spacecraft are used to map out the region 1 and region 2 field-aligned current systems in the dayside hemisphere at radial distances of close to 8 Earth radii above the northern polar cap. The extrapolated invariant latitude of these currents is quite variable. The equatorward edge of the region 1 system ranges from 73 to 83 degrees invariant latitude. The sense of both systems reverses across noon as expected. The strength of the currents also varies, averaging about 0.03 A/m for the region 1 currents and 0.01 A/m for the region 2 currents. Over the dayside hemisphere this amounts to about 1.5 million amps flowing down and up the field lines in the region 1 system and about 0.5 million amps in the region 2 system. These numbers and the behavior of the current are quite consistent with low-altitude observations.



The coupling of the solar wind to the Earth's magnetosphere is most clearly manifested in the high-latitude magnetosphere by the auroral phenomena observed there and the field-aligned currents that couple the outer magnetosphere to the ionosphere in these regions. These current systems have been investigated at low altitudes in the magnetosphere and the ionosphere by several missions including most notably TRIAD [e.g. Iijima and Potemra, 1976]; MAGSAT [e.g. Iijima and Shibaji, 1987] and VIKING [e.g. Erlandson et al., 1988]. These missions have delineated the morphology of these currents and their dependence on the interplanetary north-south and east-west components [see Iijima and Shibaji 1987 and Erlandson et al. 1988, respectively]. The POLAR mission, part of the ISTP program, was designed to address this high-latitude coupling through a combination of remote sensing of the aurora and in situ, direct measurement of the fields and plasmas that produce the aurora and transmit the stress from the solar wind to the magnetosphere. The magnetic field investigation on POLAR [Russell et al. 1995] enables us to probe these current systems at high altitudes close to their source.


It is the purpose of this note to report on the nature of these high-altitude currents seen over the first three months of the magnetometer operation over the sunlit hemisphere. We illustrate these currents by showing a few examples of the perturbations seen and providing a statistical examination of their occurrence.



The fluxgate magnetometer consists of two triads of orthogonal sensors mounted 2 meters apart on a boom extending 7.2 meters from the center of the spacecraft. Each sensor triad (inboard and outboard) is equipped with flippers that can interchange the sensor along the spin axis with one in the spin plane. The inboard magnetometer has a maximum range of 46,700 nT with a lower gain setting of 5860 nT. The outboard magnetometer has a maximum range of 5525 nT with a lower gain setting of 694 nT. All measurements are digitized to 16 bits. The finest digitization is 11 pT and the coarsest is 700 pT. The basic magnetometer is extremely linear. As evidence for this linearity, when a sine wave is fit to the spin plane sensors in strong quiet fields, the stepping of the individual bits of the analogue to digital converter (one part is 65,536) is the only non-linearity noticeable. There is no evidence for non-linear coupling between sensors due to high transverse fields or other non-linear effects in the measurements.

Internal to the magnetometer the field is sampled at a rate of 500 vectors per second. They are then filtered and resampled to 108 samples per second and supplied to other instruments on the spacecraft. A 100 Hz sample is prepared for transmission to Earth on command as burst mode data and a 120 millisecond sample is filtered and sent to the ground continually when burst mode data are not being transmitted. All averaging is performed with recursive filters in the instrument. In addition to these data in the spinning satellite frame, a crude despun vector is obtained once per spin, immediately after the sun sensor detects the sun direction. These measurements were initially obtained from the 120 ms data stream and hence could be sampled up to 7.2 late. Since 1732 UT, October 2, 1996, the data have been obtained from the 10 ms data and can be only up to 0.6 of rotation late. It is possible to synchronize the internal sampling of the magnetometer and the spacecraft clock to the nearest 1 ms. From this synchronization, then it is possible to correct the once-per-spin data to an angular accuracy of 0.06. We have done so herein and used these data in this paper.


Orbits and Coordinates

Fig. 1a. Projection of the POLAR orbit into the noon-midnight plane in solar magnetic coordinates for the period 2200 UT March 24, 1996 to 1500 UT on March 25, 1996.

Fig. 1b. Projection of the same orbit into the dawn-dusk plane in solar magnetic coordinates. The black bar indicates the time of the Region 1/Region 2 current sheet traversal seen in Figure 2.

Two projections of the orbit beginning late on March 24, 1996 are shown in Figures 1a and b. These plots show the X-Z projection (view from dawn) and the Y-Z projections (viewed from the sun) in solar magnetic coordinates. We use the solar magnetic coordinate system in which the Z-axis is along the magnetic dipole direction and the X-Z plane contains the solar direction because the POLAR spacecraft remains generally within the magnetosphere. However, we do expect some motion of magnetospheric boundaries in this coordinate system as the dipole tilt varies just as there is motion of boundaries in solar magnetospheric coordinates. With time the plane of the orbit "precesses" with respect to the Earth-sun line due to the motion of the Earth around the sun with approximately a one-year period. The orbital plane crossed the noon-midnight meridian in mid April. Since the orbital period of 17.5 hours is not an integral multiple of a day, the dipole orientation varies from orbit to orbit so that the magnetic local time of successive passes is not a simple progression. Thus afternoon-side data can be found after mid April and morning-side data prior to mid April despite the overall progression from the afternoon to morning side during this period.

The overall configuration of the magnetospheric magnetic field is well known at least qualitatively and useful empirical models of the field exist. We are most interested here in the sharp deviations of the magnetic field from the general swept-back field configuration. These sharp changes are associated with field-aligned current sheets that feed into the auroral oval. We will use as our comparison external contribution to the field the recent 1995 Tsyganenko model [N. A. Tsyganenko, personal communication, 1995] using the conditions AE=200, Dst=0 and solar wind dynamic pressure equals 2nPA in the model. We use this model herein instead of a more recent model because the 1995 model does not include explicit field-aligned currents and we wish to use the differences from the model to identify and measure these currents. For the internal field we use the 1995 International Geomagnetic Reference Field at epoch of date. This model provides an accurate reference field at low altitudes outside the regions of field-aligned currents [Zhou et al., 1997] and deviates significantly only at highest altitudes or at the most disturbed conditions. We will use it as our baseline and display herein only the residual field after subtracting the model field. We will display these residuals in a field-aligned coordinate system, with the Z-direction along the model magnetic field, the Y-direction eastward perpendicular to the model field and X completing a right-handed orthogonal set.


Field-Aligned Currents Across the Noon Sector

Fig. 2. Magnetic field measured by the POLAR magnetometer over the period 0000 to 0300 UT on March 25, 1996. The top panel shows the eastward component of the field perpendicular to the expected dipole field. The bottom two panels show the inclination and declination of the magnetic field, as observed and from the Tsyganenko 1995 model.

The top panel of Figure 2 shows the eastward component of the magnetic field with the Tsyganenko 1995 model field removed for the period 0000 to 0300 UT on March 25, 1996. The lower two panels show the inclination and declination of the magnetic field as observed and as predicted by the Tsyganenko 1995 model. The inclination is the angle between the observed magnetic field and the local horizontal plane consistent with its definition in ground-based studies. The declination is the angle of the field projection in the local horizontal plane as measured from the local dipole magnetic meridian. Qualitatively the Tsyganenko model predicts the observed direction of the magnetic field. At lowest altitudes (on the left of Figure 2) the model and observations agree quite well. At highest altitudes the direction of the field generally oscillates around the predicted one, with the large swing in declination occurring at lower altitudes than predicted. These comparisons are discussed in more detail by Zhou et al [1997].

The feature of interest here is the increase in the eastward component of the field from about 0004 to 0010 UT and the following decrease from 0010 to 0020 UT. The sense of these changes is that of the region 2 and region 1 currents [Iijima and Potemra, 1976]. The first current sheet has the deflection expected for a downward flowing current and the second for an upward flowing current as we expect in crossing the region 1 and 2 currents on the afternoon side. We note that the very small perturbation of the field from the model field at the beginning of the plot indicates that the magnetic field is close to its dipolar orientation. As it crosses through the region 2 current, the magnetic field rotates so that its high altitude end bends toward the sun and as it crosses the region 1 current it bends away from the sun. These directions are those expected due to, first, the sunward convection of plasma "pulling" on the auroral plasma and, second, to the polar-cap field lines pulled back into the tail. The gradual change in By after 0020 is consistent with the spreading of the field lines in the polar cap, albeit some of that spreading is present in the model field that serves as the baseline for this plot.

Fig. 3. Magnetic local time and invariant latitude corresponding to crossing upward and downward current sheets in the high altitude magnetosphere.

In Figure 3 we show the locations where we encountered these downward and upward currents over the first three months of magnetometer operation. The interplanetary magnetic field during these observations was quite varied. The YZ65M clock angle was within 45o of northward 19% of the crossings and southward 11%. It was dawnward 61% and duskward 9% of the time. (In calculating these conditions we have determined the time for the solar wind to reach the nose of the magnetopause along the X direction at 400 km/s.) We have mapped the observations to the ionosphere using dipole magnetic local time and invariant latitude calculated from the Tsyganenko 95 model. Clearly the latitudes of the region 1 and 2 currents are variable as are their extents. On occasion the region 2 current in this sector is too weak to be detected above the natural magnetospheric fluctuation level. We have not recorded any currents whose accompanying field deflection is less than 10 nT. As evident from Figure 3 the region 1/region 2 current pattern originally reported at low altitudes is preserved in the higher altitude POLAR measurements at distances up to 9 Re.

Fig. 4. Changes in the eastward component of the magnetic field and the equivalent current intensity seen while crossing the high-latitude region 1 and region 2 current systems.

Fig. 5. Current intensity seen at POLAR extrapolated down to ionospheric altitudes and fitted with logistic curves. The dotted line gives the one standard deviation error bars.

Figure 4 emphasizes the reversal of current flow across noon with downward region 1 before noon and upward afternoon with the reverse behavior for region 2. Here we show the magnitude of the change in the eastward component associated with crossing the current sheets. We note that a region 1 current sheet is always identifiable but a region 2 current is too weak to resolve about one-third of the time, as can be seen by the relative number of data points in Figure 4. The median region 1 field change was about 40 nT corresponding to 0.03 Amps/m and the median region 2 field change was about 15 nT corresponding to 0.01 Amps/m. When extrapolated to ionospheric levels along dipolar field lines, these currents should increase in strength as R1.5 so that changes in the field and equivalent current will be about 15 times larger in the ionosphere than seen at POLAR apogee. Figure 5 shows the data of Figure 4 extrapolated into the ionosphere and fitted with logistic curves of the form I = a (1 + exp {-b (MLT-c)})-1 +d where I is the current intensity, MLT is the magnetic local time and a, b, c, d are the parameters to be determined. The upper and lower curves are one standard deviation "error bars" calculated with the bootstrap method of Kawano and Higuchi [1995]. The fits suggest that there may be some dawn-dusk asymmetry in the current strength and the current reversal may be centered off the noon meridian but these asymmetries are all within the two sigma error bars and, to the accuracy we can determine from the present data, the currents are dawn-dusk symmetric and reverse around local noon. Over the dayside hemisphere, thus, there are about 1.5 million amperes flowing down in the dawn region 1 currents and up in the afternoon region 1 currents. There are about one half million amperes flowing on each side of the region 2 system.


Summary and Conclusion

POLAR magnetic field measurements obtained during the first three months of operation show that the low-altitude field-aligned current map as expected to high altitudes along the boundaries of the polar cap. Over the dayside hemisphere about 1.5 M amps are flowing on each side of the region 1 system and about 0.5 M amps are flowing in the region 2 currents. The reversal of the currents occurs very close to noon. The region 1 currents decrease in strength between 1000 and 1300 MLT as they are reversing. This behavior is not clearly seen in the weaker region 2 currents. The strength of these currents were otherwise moderately steady over the course of these observations. However, the latitude of the current system varied markedly from about 73 to 83 for the equatorward edge of the region 1 current. The observations reported here were obtained over a wide range of IMF magnetic field directions and for both steady and changing conditions. We have not yet examined the solar wind control of the latitude of these currents and their amplitude but we expect that this variability is associated with the north-south component of the interplanetary magnetic field. The studies with MAGSAT [Iijima and Shibaji, 1987] and with VIKING [Erlandson et al. 1988], in particular, indicate that both the north-south and the east-west components of the interplanetary magnetic field will alter the location and amplitude of the currents. Finally, now that plasma data are being made routinely available we will begin to place these current measurements more properly in the context of the plasma population in which they are observed.



This research was supported by the National Aeronautics and Space Administration under research grant NAG 5-3171. We wish to express our appreciation to J. D. Means, D. Dearborn, D. Pierce, R. C. Snare, and M. Larson, who assisted in the construction of the magnetometer and especially to Ron Harten of Lockheed Martin who oversaw the integration of this investigation on the spacecraft. We also wish to thank N. Tsyganenko who made his 1995 field model available to us in advance of publication.



Erlandson, R. E., L. J. Zanetti, T. A. Potemra, P. F. Bythrow and R. Lundin, IMF By dependence of region 1 Birkeland currents near noon, J. Geophys. Res., 93, 9804-9814, 1988.

Iijima, T. and T. A. Potemra, The amplitude distribution of field-aligned currents at northern high latitudes observed by TRIAD, J. Geophys. Res., 81, 2165-2174, 1976.

Iijima, T. and T. Shibaji, Global characteristics of northward IMF - associated (NBZ) field-aligned currents, J. Geophys. Res., 92, 2408-2424, 1987.

Kawano, H. and T. Higuchi, The bootstrap method in space physics: error estimation for minimum variance analysis, Geophys Res. Lett., 22, 307-310, 1995.

Russell, C. T., R. C. Snare, J. D. Means, D. Pierce, D. Dearborn, M. Larson, G. Barr, and G. Le, The GGS/POLAR magnetic fields investigation, Space Sci. Rev., 71, 563-582, 1995.

Zhou. X-W, C. T. Russell, Guan Le and N. Tsyganenko, Comparison of observed and model magnetic fields at high altitudes above the polar cap: POLAR initial results, Geophys. Res. Lett., submitted, 1997.

Back to CT Russell's page More On-line Resources
Back to the SSC Home Page