Cusp observations of high- and low-latitude reconnection for northward IMF: An alternate view

Abstract. During high solar wind dynamic pressure, northward interplanetary magnetic field (IMF) intervals, the POLAR/TIMAS instrument observes angular distribution functions, that have been interpreted in terms of reconnection, below the latitude of the cusp, of nearly parallel magnetic field lines from the magnetosheath with those in the magnetosphere. If true, this process would lead to erosion of the dayside magnetosphere by northward IMF, a process not consistent with the observed control by IMF B_z of the location of the dayside magnetopause, of the polar cusp, or of geomagnetic activity. However, we note that the TIMAS observations do not require such an interpretation and that a simpler interpretation exists, consistent with the current paradigm of IMF effects on the magnetosphere.

1. Introduction

As envisioned by Dungey [1961, 1963], the interplanetary magnetic field (IMF) can reconnect with the magnetospheric magnetic field when the IMF is southward and when it is northward. When the IMF is due southward, reconnection is believed to be initiated in the subsolar region and transfers magnetic flux from the closed dayside magnetosphere into the tail. Reconnection across the tail current sheet returns the opened flux to the closed magnetosphere. This process couples the momentum of the solar wind to the magnetosphere. It was originally proposed to explain how the solar wind powers geomagnetic activity. When the IMF is due northward, the site of reconnection is thought to be behind the cusp, in both the north and the south, adding solar wind containing flux tubes to the dayside magnetosphere and causing a day-to-night flow within the closed magnetosphere.

Much evidence supports the existence of both modes of reconnection, but the rate of reconnection appears to be much less when the IMF is due northward than due southward. This can be discerned from the relative insensitivity of the dayside magnetopause location to northward IMF compared to that of southward IMF [e.g., Petrinec and Russell, 1993; Shue et al., 1998]; the relative insensitivity of the location of the polar cusp to northward IMF compared to that for southward IMF [e.g., Zhou et al., 2000]; and the relative insensitivity of geomagnetic activity to northward IMF compared to that for southward IMF [Scurry and Russell, 1991].

There is some controversy on the location of the reconnection site as a function of the clock angle of the IMF, the angle of the IMF in the plane perpendicular to the solar wind flow. Crooker [1979] proposed that the site moves in response to changing IMF orientation, so reconnection is initiated where the IMF and magnetospheric fields are antiparallel. This was extended by Luhmann et al. [1984] to take into account the realistic orientation of the magnetic field in the magnetosheath and the magnetosphere. Support for this hypothesis has been provided by observations of the control of the local time of the polar cusp by the IMF clock angle [Newell et al., 1989; Zhou et al., 2000]. In observation and in theory the reconnection site moves poleward of the polar cusp when the IMF is northward and is almost directly behind the cusp when it is nearly due northward.

Recently Fuselier et al. [1997] have interpreted the electron and ion signatures near the magnetopause to indicate that on occasion the site of reconnection is equatorward of the cusp even though the IMF is very nearly due northward. Even more recently, Fuselier et al. [2000] have examined Polar/Toroidal Imaging Mass Angle spectrograph (TIMAS) observations in the vicinity of the cusp as further evidence for this behavior. They base their interpretation on the implication that flow on a magnetospherically oriented field line must be away from the reconnection site and not toward it. However, as they note, this interpretation leads to several difficulties. They state: "It is difficult to see how the effective location of the spacecraft can jump across the cusp rapidly." They further note that "Surprisingly, most of the events consistent with reconnection equatorward of the cusp that are presented here have the smallest IMF clock angles." In their words "the correlation between equatorward of the cusp reconnection and small clock angles would appear to be contrary to expectations of reconnection." In an attempt to resolve this dilemma they postulate that dynamic pressure can control the site of reconnection. Herein we note that a simpler explanation of the observed particle distributions exists, one consistent with previous studies of magnetospheric behavior and by Occam's razor, one more likely to be correct.

2. Alternative Explanation

Fuselier et al. [2000] examine in detail two periods in which the IMF was northward and the solar wind dynamic pressure was high, 0300-0800 UT on June 20, 1996, and 1100-1600 on April 11, 1997, with pressures reaching about 5.5 nPa and 8 nPa, respectively, during the moments of interest. Their Figures 1 to 4 are repeated here but with different interpretive sketches in the top panels of Figures 1 and 3. The interpretive sketches differ as follows: Firstly, since the entry into the "cusp" plasma is occasioned by an increase in the solar wind dynamic pressure the "trajectory" of the spacecraft through the near-cusp region has been taken to be radially away from the Earth. This is a departure from the arbitrary motion of the spacecraft that Fuselier et al. [2000] could not explain. This motion corresponds to a self-similar compression of the magnetosphere. Secondly, we draw the current sheet only in the region in which the magnetic field "reverses" across it because the current becomes thick and weak away from this region. Thirdly, we avoid the term magnetopause which some might apply to the current sheet near the cusp, but others might apply to the boundary between those field lines connected to the Earth by one or both feet and those totally unconnected to the Earth. This might be a region of both velocity and magnetic field shear and is the outermost line in these two interpretive panels. While this latter definition might be most similar to the operational definition used in most studies that define the size of the magnetosphere, it has a sudden inward step at the reconnection point where field lines become disconnected from the Earth if reconnection is taking place with open magnetic field lines. The magnetopause here is best considered to be the entire region of the cusp and the various boundary layers surrounding it.

2.1. June 20, 1996 Observations

Figure 1 shows an overview of the June 20, 1996 cusp event for strongly northward IMF. The top panel shows a noon-midnight meridian projection of the orbit from 0300 to 0800 UT and the expected magnetic configuration for northward IMF. The schematic magnetosphere is in its uncompressed state, but the solar wind dynamic pressure as shown in the bottom panel was about 4 times its nominal value at this time. We can illustrate the effect of this increased pressure by increasing the radial position of the spacecraft by 25% to account for the inverse sixth root dependence of the magnetopause size on the solar wind dynamic pressure. This increase will put the spacecraft close to the expected location of the reconnection-associated current layer across which the magnetic field reverses near the cusp. Two locations, L and M, at which ion distribution functions are given in Figure 2 are shown here. The M location occurred when the solar wind pressure was the highest and is placed at the greatest distance. The magnetic field at this location is northward and antisunward in accord with the sketch. The magnetic field at location L is southward and sunward and, as noted in the sketch, should lie below the current sheet.

Figure 1. (top) Noon-midnight meridian cut through the magnetosphere showing the Polar spacecraft location for June 20, 1996, from 0300 to 0800 UT. (middle) The three geocentric solar magnetospheric (GSM) components and the magnitude of the magnetic field measured in the cusp from 0300 to 0800 UT. (bottom) The solar wind dynamic pressure measured by the Wind spacecraft upstream from the Earth (the time is corrected for the convection time to the shock, through the magnetosheath, and into the cusp). The observed (solid lines) and model (dashed lines) magnetic field agree reasonably well before 0450 UT and after 0700 UT. Between 0450 and 0700 UT the measured magnetic field often has a dayside magnetospheric orientation (Bx < Bz > 0) when the model magnetic field predicts a lobe-like orientation (Bx > Bz < 0). The intervals of dayside magnetospheric orientation are correlated with high solar wind dynamic pressure [after Fuselier et al., 2000].

The angular distributions of the ions shown in Figure 2 are quite different at the two locations. The spectrum L, with two antiparallel beams, has a field orientation appropriate to a location poleward of the cusp and below the current sheet. The beam in the top half plane cuts off at zero parallel velocity and is quite D shaped. It appears to be solar wind plasma that has recently passed through the reconnection region. The beam in the bottom half of the plane, flowing antiparallel to B, is most probably the solar wind plasma mirrored above the polar cap. We agree with the analysis of Fuselier et al. [2000] that this location is probably close to the reconnection site and about 8 R_E from the mirror point in the ionosphere.

Figure 2. Two-dimensional cuts in the velocity space distribution in a plane containing the magnetic field (y axis) and the Earth-Sun line (top panels) and one-dimensional cuts along the magnetic field (bottom panels). In the bottom panels the solid curve is the phase space density and the dashed line the 1 level. These distributions are from the cusp interval in Figure 1 when the magnetic field had a lobe-like orientation (left-hand side) and when it had a dayside magnetospheric-like orientation (right-hand side). When the magnetic field has a lobe-like orientation, the proton distribution consists of two distinct populations. The distribution on the right-hand side resembles a nearly isotropic distribution with a small bulk flow parallel to the magnetic field [Fuselier et al., 2000].

The second distribution function, M is nearly isotropic and shows no two beam structure nor cutoffs. We believe that it is important to note that in Figure 1 the magnetic field strength at M is much less than at L, indicating that M, while on the far side of the current sheet from the Earth and the far side of location L, is also closer to the current sheet. Thus the field line passing through M does not come so close to the reconnection site as that through L. Alternatively, one can say that the flux tube on which M resides passed through the reconnection point earlier than did the tube through L. Hence bouncing ions have had the chance to isotropize on this "older" flux tube that we expect to be closed by nearly contemporaneous southern cusp reconnection for this nearly due northward IMF condition. We note that there is little bulk velocity at M, either along or across the magnetic field.

In summary our interpretation of this event does not differ greatly from that of Fuselier et al. [2000]. The main difference here is that we interpret the effect of increased solar wind pressure in terms of effective radial motion, while Fuselier et al. [2000] include much, apparently arbitrary, translation in response to the solar wind pressure change.

2.2. April 11, 1997, Observations

Figure 3 shows an overview of the April 11, 1997 event again for strongly northward IMF. The format of this figure is identical to that of Figure 1. The bottom panel shows that the solar wind dynamic pressure was about 4-5 times its nominal value at the time of observations L and M. The nose of the magnetopause drawn by Fuselier indicates that it was scaled to a pressure about twice the nominal pressure. Thus more correctly the portion of the trajectory shown in Figure 3 should be about 15% closer to the magnetopause than is shown. This would place it very close to the current sheet as sketched. As before we clearly have observations at L and M that are on opposite sides of the current sheet but here because of the increased pressure, the M location appears to be farther from the current sheet than the L location. Thus the field line through M passes closer to the reconnection site than the field line through L. Alternatively stated, the flux tube that passes through M crossed the reconnection site much more recently than did the flux tube crossing L.

Figure 3. (top) Noon-midnight meridian cut through the magnetosphere showing the Polar spacecraft location for April 11, 1997, from 1100 to 1600 UT. (middle) The three geocentric solar magnetospheric (GSM) components and the magnitude of the magnetic field measured in the cusp from 1100 to 1600 UT. (bottom) The solar wind dynamic pressure measured by the Wind spacecraft upstream from the Earth. The format is the same as in Figure 1. The observed (solid lines) and model (dashed lines) magnetic field agree reasonably well except for the interval from 1430 to 1450 UT. In this interval, the observed magnetic field has a dayside magnetospheric-like orientation and this interval is correlated with high solar wind dynamic pressure [after Fuselier et al., 2000].

The angular ion distribution functions are shown in Figure 4 for these two locations. We again interpret the isotropic distribution at L as a sign of age in a very low bulk-velocity region on recently closed field lines. The distribution at M shows the incoming solar wind beam and reflected ions but has an important difference from the similar distribution on June 20, 1996. Here the ion cutoff velocity is negative. Fuselier et al. [2000] cannot explain this negative cutoff and therefore dismiss it as an artifact. However, it has a simple explanation. At M on the far side of the current sheet, solar wind plasma has access to the spacecraft from both the sunward and the antisunward directions. High negative velocities of course cannot reach the spacecraft because the field line has undergone reconnection at some distance farther from the Sun, but low negative velocities can still reach the spacecraft because the plasma is slowly drifting toward the Earth perpendicular to the magnetic field. This access is not possible inside the current sheet distance at the location at which Fuselier et al. [2000] place their M. In the work in progress (G. Le et al., unpublished manuscript, 1999) we show the consistency of global magnetohydradynamic (MHD) simulations, AMIE-derived polar-cap convection patterns, and the in situ Polar spacecraft observations for this event.

Figure 4. Two-dimensional cuts in the velocity space distribution in a plane containing the magnetic field (y axis) and the Earth-Sun line (top panels) and one-dimensional cuts along the magnetic field (bottom panels). The format is the same as in Figure 2. When the magnetic field has a dayside magnetospheric-like orientation, the proton distribution (right-hand side) consists of two distinct populations. The distribution on the left-hand side resembles a nearly isotropic distribution with a small bulk flow parallel to the magnetic field [Fuselier et al., 2000].

In summary our interpretation of this second event differs greatly from that of Fuselier et al. [2000]. Our interpretation of the pressure increase providing a radial probing of the reconnection current sheet allows us to understand the angular distributions in terms of age of the flux tube since it reconnected. Our interpretation places the observation points at locations that make sense physically and avoids arbitrary translation of observation points to match the field geometry. Most importantly, it allows us to interpret non-D-shaped distribution functions with negative flows. Perhaps, surprisingly, the bulk velocity of the incoming, magnetosheath-derived plasma is low, but it needs to be low for reconnection to proceed.

3. Discussion and Conclusions

The interpretation of the June 20, 1991, and April 11, 1997, events by Fuselier et al. [2000] has several unsettling aspects. First, they invoke effects by the dynamic pressure on the magnetosphere which involve more than just a change of scale. They propose that the effective location of the cusp moves significantly in latitude in addition to the motion due to the magnetosphere changing its scale size when the pressure changes. They also propose that the unexpected reconnection for very low shear is a product of these high-pressure conditions. However, in our discussion above we see that it is possible to avoid these postulates. We can interpret the data on these two days with just radial scale changes. No motion of the cusp is needed and the reconnection site for strongly northward IMF can be kept poleward of the cusp. OccamÆs razor, which states that the simplest explanation is to be preferred, certainly favors this interpretation.

We emphasize the low velocities of the plasma in this region. In the absence of reconnection postcusp we might expect that the plasma reaches bulk velocities near that of the solar wind as it crosses the terminator region. It is possible that this reduction in velocity from that of the free-stream magnetosheath flow is accomplished by a slow-mode wave, much as postulated for the edge of the plasma sheet during tail reconnection. Similar low velocities near the cusp for northward IMF have been observed on the widely studied May 24, 1996, event [Russell et al., 1998]. These low velocities are also an important consideration in the interpretation of Fuselier et al. [2000] that the spacecraft is poleward of the reconnection site. Since we would expect these low velocities when the magnetic field is mostly due northward as reconnection poleward of the north and south cusps adds a new flux tube to the magnetosphere [Song and Russell, 1992], we can understand why this "reconnection" appeared to occur for small clock angles. The correlation with dynamic pressure occurs because the region in which these distributions can be sampled requires polar to be near the current sheet and the boundary with the flowing solar wind.

Our interpretation is also to be preferred because it preserves other, earlier understanding of the behavior of the magnetosphere in response to northward IMF. Under these conditions there is little erosion of the dayside magnetosphere, the cusp remains roughly at constant latitude, and geomagnetic activity remains quiet. Such behavior would not prevail if there were subcusp reconnection for northward IMF.

Acknowledgments

The authors wish to thank K. W. Ogilvie and R. P. Lepping for providing solar wind data through the CDAWeb. Discussions with R. J. Strangeway were most helpful. This research was supported by the National Aeronautics and Space Administration under grant NAG5-7721.

Hiroshi Matsumoto thanks M. F. Thomsen and R. L. Kessell for their assistance in evaluating this paper.

Crooker, N. U., Dayside merging and cusp geometry, J. Geophys. Res., 84, 951-959, 1979.

Dungey, J. W., Interplanetary magnetic field and the auroral zones, Phys. Rev. Lett., 6, 47-48, 1961.

Dungey, J. W., The structure of the exosphere or adventures in velocity space, in Geophysics: The Earth's Environment, edited by C. Dewitt, J. Hieblolt, and A. Lebeau, pp. 505-550, Gordon and Breach, New York, 1963.

Fuselier, S. M., K. J. Trattner and S. M. Petrinec, Cusp observations of high- and low-latitude reconnection for northward IMF, J. Geophys. Res., submitted 1999.

Luhmann, J. G., R. J. Walker, C. T. Russell, N. U. Crooker, J. R. Spreiter, and S. S. Stahara, Patterns of potential magnetic field merging sites on the dayside magnetopause, J. Geophys. Res., 89, 1739-1742, 1984.

Newell, P. T., C. I. Meng, D. G. Sibeck, and R. P. Lepping, Some low-altitude cusp dependencies interplanetary magnetic field, J. Geophys. Res., 94, 8921-8927, 1989.

Petrinec, S. M., and C. T. Russell, External and internal influences on the size of the dayside terrestrial magnetosphere, Geophys. Res. Lett., 20 (5), 339-342, 1993.

Russell, C. T., J. A. Fedder, S. P. Slinker, X-W. Zhou, G. Le, J. G. Luhmann, F. R. Fenrich, M. O. Chandler, T. E. Moore and S. A. Fuselier, Entry of the POLAR spacecraft into the polar cusp under northward IMF conditions, Geophys. Res. Lett., 25, 3015-3018, 1998.

Scurry, L., and C. T. Russell, Proxy studies of energy transfer in the magnetosphere, J. Geophys. Res., 96, A6, 9541-9548, 1991.

Shue, J.-H., P. Song, C. T. Russell, J. T. Steinberg, J. K. Chao, G. Zastenker, O. L. Vaisberg, S. Kokubun, H. J. Singer, T. R. Detman, and H. Kawano, Magnetopause location under extreme solar wind conditions, J. Geophys. Res., 103, 17691-17700, 1998.

Song, P., and C. T. Russell, Model of the formation of the low-latitude boundary layer for strongly northward interplanetary magnetic field, J. Geophys. Res., 97, 1411-1420, 1992.

Zhou, X. W., C. T. Russell, G. Le, S. A. Fuselier and J. D. Scudder, Solar wind control of the polar cusp at high altitude, J. Geophys. Res., in press 1999.

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