J. Geophys. Res., 99, 23,723-23,734, 1994
© Copyright 1994 by the American Geophysical Union
Paper Number 94JA02182
J. T. Gosling
Los Alamos National Laboratory, Los Alamos, New Mexico
Abstract. We use ISEE magnetic field and plasma data to examine dayside magnetopause crossings under conditions of low Mach number and strongly northward interplanetary magnetic field (IMF). When the solar wind Mach number is low, the IMF strength and magnetosheath field strength are large, and we expect the effects of magnetic reconnection to be the strongest. When the IMF is strongly northward, we find that the location of the magnetopause boundary layer is very stationary in the space, and we observe many features that are common for both typical and low Mach numbers. However, under low Mach number conditions, we have observed some features that would be expected for cusp reconnection. The boundary layer near the subsolar region contains heated magnetosheath plasma with little hot magnetospheric component that has clearly entered the magnetosphere elsewhere. At least some of the structures present in the boundary layer are impulsive. Inside the boundary layer there is also clear evidence of accelerated flow from the cusp region for strongly northward IMF at low Mach number. Reconnection beyond the cusp can explain the observed field, plasma, and flow signatures. Therefore at low Mach number, reconnection is important in the formation of the boundary layer for northward IMF.
The magnetopause is the interface between the shocked solar wind in the magnetosheath and the Earth's magnetosphere. It is a complex plasma boundary consisting of both field and plasma transitions. The change of the magnetic field across the magnetopause is associated with a current layer, which is generally taken to be carried by deflected solar wind particles [Chapman and Ferraro, 1931]. Much of the incoming plasma flows tangential to this boundary at the magnetopause. On the sunward side of this boundary, the plasmas are dense and cold. On the earthward side of this boundary the plasmas are hot and tenuous. The region immediately earthward of the magnetopause was first studied more than 20 years ago and at low latitude was called low-latitude boundary layer (LLBL) [Hones et al., 1972; Akasofu et al., 1973]. Early observations revealed that the plasma inside this boundary layer has properties intermediate between those of the magnetosheath and the magnetosphere [Eastman et al., 1976; Haerendel et al., 1978; Eastman and Hones, 1979]. However, the observations from IMP 6 and Heos 2 differed in one important plasma property that led to two distinct mechanisms for the formation of the LLBL. In the IMP 6 data the density fell off gradually to the magnetospheric background level across the LLBL, and a local diffusion process was proposed [Eastman et al., 1976; Eastman and Hones, 1979]. However, in the Heos 2 data the density was roughly constant throughout the LLBL, and a mechanism of nonlocal entry from the cusp and/or heating of cold magnetospheric plasma were suggested [Haerendel et al., 1978]. The discrepancy in the two data sets may be due to the low time resolution of Heos 2 data [Eastman and Hones, 1979] or it may be due to the different spatial coverage of the dayside magnetopause by Heos 2 and IMP 6 [Paschmann et al., 1978].
ISEE and AMPTE spacecraft have provided both significantly more observations of the Earth's magnetosphere and data at higher time resolution than available previously. The structure of the magnetopause boundary layer and its plasma properties have been under active investigation using these data and data from other spacecraft [e.g., Paschmann et al., 1978, 1990; Sckopke et al., 1981; Ogilvie et al., 1984; Mitchell et al., 1987; Ogilvie and Fitzenreiter, 1989; Gosling et al., 1990; Song et al., 1990b; Takahashi et al., 1991]. It has long been recognized that the physical processes that govern structures of the magnetopause boundary layer change with the direction of the interplanetary magnetic field (IMF). The magnetopause has been more intensively studied when the IMF is southward, and it is found that its structures are affected mainly by the reconnection and related phenomena [Paschmann, 1979; Paschmann et al., 1979; Sonnerup et al., 1981; Mitchell et al., 1987; Gosling et al., 1990, 1991]. When the IMF is northward, the magnetic field transition has been found to be still a rotational discontinuity [Paschmann et al., 1990]. The plasma transition consists of multiple layers with relatively uniform structure inside each layer [Song et al., 1990b].
The structure of the magnetopause boundary layer depends upon more than just the IMF direction. It also depends upon the magnetic field strength and the ratio of plasma thermal pressure to the magnetic pressure, or beta. Observations of accelerated flow events have shown that the magnetic pressure has to be strong relative to the plasma pressure for reconnection [Paschmann et al., 1986; Scurry et al., 1994]. Statistically, the low solar wind Mach number is associated with large magnetic field strength and low plasma beta in the solar wind. Furthermore, low plasma beta in the shocked solar wind, or the magnetosheath, requires low Mach number upstream of the bow shock. Under the low Mach number condition the magnetic field plays a dominant role in any reconnection related physical processes occurring at the magnetopause boundary layer. We expect the effects of magnetic reconnection to be strongest, whether the reconnection occurs at high latitude for northward IMF or on the dayside for southward IMF. Reconnection should be least important when the solar wind plasma is weakly magnetized under conditions of very high Mach number and high plasma beta.
The dependence of the structure of the magnetopause upon the Mach number has not been investigated systematically. Several authors studied the structures of the magnetopause for northward IMF. Song et al. [1990b] reported a detailed case study of the magnetopause for northward IMF. In this case the magnetosonic Mach number was 6, a typical value at 1 AU. Paschmann et al.  reported a survey of 22 magnetopause crossings for low local magnetic shear at the magnetopause. To determine the magnetic field effects, we examine in detail a limited number of cases under extreme conditions when the IMF and the magnetosheath field are strong rather than a large number of more typical conditions. The purpose of this paper is to document special features of magnetopause structures for low Mach number and strongly northward IMF in a preliminary study of the Mach number dependence of the magnetopause boundary layer. We discuss common features observed at both typical and low Mach numbers. We emphasize special features observed under low Mach number condition.
In this paper we present a detailed case study of two magnetopause crossings observed by the dual ISEE spacecraft under conditions of low Mach number and strongly northward IMF. The magnetosonic Mach numbers in the two cases presented in the next section are 4.2 and 3.1, respectively. The statistical survey of the solar wind parameters of the entire solar cycle 21 at 1 AU indicated that the magnetosonic Mach number is greater than 4 more than 75% of time and 3.5 more than 90% of time [Luhmann et al., 1993]. Thus, the Mach numbers in this study occur infrequently but should not be considered rare.
The data used in this study include those from the ISEE 1 and 2 magnetometer [Russell, 1978] and ISEE 2 Fast Plasma Experiment (FPE) [Bame et al., 1979] for the magnetopause crossing and from ISEE 3 and IMP-8 for upstream solar wind. The first case, the outbound crossing of ISEE 1 and 2 orbit 299 on October 5, 1979, is presented in section 2.1 and the second case, the inbound crossing of ISEE 1 and 2 orbit 314 on November 11, 1979, is presented in section 2.2. In both cases the solar wind is very steady. The average properties of the solar wind parameters near the magnetopause crossings are listed in Table 1. The ISEE 1 and 2 trajectories during these intervals projected on to the GSM XY and YZ planes are shown in Figure 1. Also shown in Figure 1 is the average magnetopause position in the XY plane.
|Table 1. Solar Wind Parameters|
|October 5, 1979||November 11, 1979|
|Magnetic field in GSM (nT)||(4.3, -1.2, 6.7)||(9.34, -9.64, 11.70)|
|Magnetic field strength (nT)||8.0||18.0|
|Ion temperature (K)||1.7E4||9.5E4|
|Magnetosonic Mach number||4.2||3.1|
2.1. October 5, 1979
Figure 2 shows the ISEE 2 data during the outbound magnetopause crossing on October 5, 1979. The top panel shows the magnetic field in GSM coordinates and the field strength with time a resolution of 4 s, which are averaged from the full resolution data (0.25 s) with an overlapped window of 12 s. The lower panels show ion density, temperature, ion beta, and velocity with a time resolution of 12 s. They are deduced from the two-dimensional FPE measurements. The two electrostatic analyzers of FPE in the ecliptic plane produces complete two-dimensional ion and electron distributions measurement within 55o elevation angle every spacecraft spin of 3 s. Highlighted by shading is an interval from 0217 to 0247 UT for the magnetopause boundary layer crossing, which will be discussed in detail. The spacecraft position in GSM is (9.61, -3.34, 5.04) RE at 0230 UT. The simultaneous observations of the IMF in GSM, the solar wind density, velocity, temperature, and dynamic pressure are shown on Figure 3 from both the ISEE 3 and IMP 8 since neither of them has complete data coverage during this interval. In Figure 3, the ISEE 3 data are lagged -70 min, which is the estimated time delay from ISEE 3 to the Earth's magnetosphere. The IMF is very steady and strongly northward during the interval. The clock angle of the IMF is 10o from the due northward direction. The magnetosonic Mach number is 4.2. The solar wind plasma beta is 0.2.
From Figure 2 it is very difficult to identify the magnetopause crossing from the magnetic field data alone. The magnetic field on both sides of the magnetopause has roughly the same strength and there is little magnetic shear across the magnetopause since the IMF is strongly northward. From the FPE plasma data it is clear that there exists a boundary layer (shaded area) where the average plasma properties are intermediate between their magnetospheric and magnetosheath values. We note that the plasma beta in the magnetosheath is very low and comparable to the beta in the magnetosphere. The structure of the boundary layer is shown more clearly in Figure 4 from 0215 to 0253 UT. In the upper four panels the high- resolution (0.25 s) magnetic field data are displayed in the boundary normal coordinate system, in which L is in the direction of the magnetospheric field and N is normal to the local magnetopause boundary. The next three panels show ion density, temperature and velocity from two-dimensional FPE data. In the bottom panel of Figure 4 is shown the north-south component of the ion flow velocity (Vz) with a time resolution of 50 s. The Vz component is deduced from three-dimensional FPE measurements at a slower rate than the two-dimensional measurements.
Several layers can be seen from Figure 2 and Figure 4:
Depletion layer: Immediately outside the magnetosphere in the magnetosheath there is a strong density depletion layer ( 0247 - 0315 UT). The inner edge of the depletion layer is the magnetopause, marked by a dashed line at 0247 UT in Figure 2 and Figure 4. This density depletion has been observed as a common feature in the magnetosheath for northward IMF and typical Mach numbers at 1 AU [e.g., Paschmann et al., 1978, 1993; Song et al., 1990a, b; Anderson et al., 1991; Phan et al., 1994]. There are strong Pc 1 waves ( 1 Hz) on the magnetosheath field lines. These waves are mainly transverse and appear mainly in the BM and BN components. This is also similar to the typical Mach number cases. For a typical Mach number case in the work by Song et al. [1990b] these waves are found to be generated in the density depletion layer.
Boundary layer: The boundary layer occurs from 0217 to 0247 UT, with the outermost boundary at the end of the density depletion. The innermost boundary is not well defined since there is no sharp transition to the magnetosphere. The density and temperature gradually reach their magnetospheric values near 0217 UT.
From the density and temperature the plasmas are magnetosheathlike on average in the outer part of the boundary layer ( 0226 - 0247 UT). The inner boundary layer ( 0217 - 0126 UT) is dominated by the magnetospheric particles, as evidenced by the higher temperature. The Pc 1 waves, which are very strong in the depletion layer, are also present in the outer boundary layer with reduced intensity, but are absent in the inner boundary layer (prior to 0226:37 UT as marked by a solid line in Figure 4). These properties are same as those reported by Song et al. [1990b] in boundary layers under typical Mach number condition. One difference is that the outer and inner boundary layers are not separated by a sharp transition in ion temperature in the low Mach number case.
Pulses and accelerated flows in the boundary layer: The special feature in this low Mach number case is that there is much structure inside the boundary layer as shown in Figure 4. The most striking characteristic is a series of pulses in field strength, density, velocity, and temperature. The pulses in the magnetic field occur only in the BL component and are compressional. These pulses are thus associated with the changes in field strength, but not field direction.
The Pc 1 waves are occasionally absent in the boundary layer (shaded intervals 1, 2, 3, and 4 in Figure 4). More detail can be seen in Figure 5. The upper two panels of Figure 5 show the high-pass filtered BM and BN components, respectively. The low cut off frequency for the high-pass filter is 0.1 Hz, and the Nyquist frequency is 2 Hz. The bottom panel of Figure 5 shows the magnetic field magnitude. The drops of Pc 1 wave amplitude coincide with pulses of field strength depressions and higher temperature plasmas (shaded interval intervals 1, 2, 3, and 4 in Figure 4). The properties of the waves are controlled by the plasma conditions. Changes in the wave pattern indicate changes in the plasma distribution function. At the magnetopause, such drastic changes may indicate changes in the field topology. Song et al. [1990b] have shown the correlation between the Pc 1 wave and the field topology. The decrease of Pc 1 wave amplitude in this case is consistent with the boundary layer being on closed field lines during these intervals. Similarly these waves are completely absent in the inner boundary layer and the magnetosphere. However, the energy spectrograms of ISEE 2 FPE ions (not shown) do not suggest a mixture of magnetospheric and magnetosheath plasma in these pulses. Rather, this plasma looks like heated sheath plasma with no hot magnetospheric plasma component present on the energy spectrograms.
This argument is supported by the two-dimensional ion and electron distribution functions observed inside these pulses. Figure 6 shows ISEE 2 FPE 3-s snapshots of ion (top panels) and electron (bottom panels) distribution functions at times marked by solid lines in Figure 5, all within the high-density and high-temperature pulses. In the ion distribution functions there are no hot ions from the terrestrial ring current which have energies corresponding to speeds well above 800 km/s, nor transmitted magnetosheath ions which would appear as a cooler and strongly flowing ions in the XY plane. The observation of electron distribution functions leads to the same conclusion. Thus the particles in these pulses are unlikely to be the hot magnetospheric plasmas entering the boundary layer locally, nor local magnetosheath plasma heated while crossing the magnetopause since there is little magnetic shear, nor a mixture of both. The observations suggest these pulses are heated magnetosheath plasma that entered from elsewhere.
For the rest of the time (unshaded intervals 1', 2', 3', and 4' in Figure 4), the particles are dominated by the cool magnetosheath plasma. Figure 7 shows 3-s snapshots of ion (top panels) and electron (bottom panels) distribution functions at times marked by dashed lines in Figure 5, all outside the high-density and high-temperature pulses. Only the cool sheath plasma component is observed in these density contours. The presence of Pc 1 waves suggests that the boundary layer is on open field lines where large temperature anisotropics have been reported [Song et al., 1993; Anderson et al., 1994]. However, the flows are accelerated to velocities well above the magnetosheath values during the unshaded intervals 1', 2', and 3', or near the inner edge of the boundary layer as shown in Figure 4. From the bottom panel of Figure 4, it is evident that the high speeds have very strong Vz component, that is roughly a field-aligned component. The positive sign of the Vz peak indicates the accelerated plasmas flow northward.
Scale lengths: We have also examined the simultaneous data from the dual ISEE 1 and 2 magnetometers to separate the temporal variation from the spatial variation and to estimate the velocity and the spatial extent of the boundary layers. During the time interval of the magnetopause boundary crossing, ISEE 1 and 2 travel outbound in an identical trajectory (see Figure 1) but ISEE 2 leads ISEE 1 about 11 min. The spacecraft velocity is 2.47 km/s along the orbit and 2.11 km/s along the boundary normal direction. Since the pulses in the magnetic field occur mainly in the field-aligned direction, we show simultaneous observations of only the magnetic field strength in Figure 8. The top panel of Figure 8 shows time series of the magnetic field strength with the highest time resolution. Both spacecraft observe the impulsive structures. ISEE 2 observes them earlier than ISEE 1 does. The time delay of the field observations is approximately the same as their orbital time delay, or 11 min. This observation suggests that the boundary layer as a whole is nearly stationary in space and the process that causes the pulses persist over a long period of time. The pulses in the boundary layer are substructures and each individual pulse may be a temporal structure inside the boundary layer. This can been seen clearly from the bottom panel of Figure 8, in which we show the magnetic field strength versus the spacecraft position X. Both ISEE 1 and 2 observe the pulses in nearly the same spatial range, while the observations are nearly 11 min apart. The average velocity of the boundary motion as a whole can be estimated by using the first pulses as the timing of the inner edges of the boundary layer. This velocity is extremely small, 0.07 km/s as calculated from the time delay (664 s) and the spatial separation of the first pulse observed by ISEE 1 and 2 (48.57 km along the boundary normal direction).
We also note in Figure 8 that at one point from 0245 to 0246 UT, the outermost spacecraft, ISEE 2, is in a heated plasma pulse and the innermost spacecraft, ISEE 1, is in the sheath-like plasma. This argues strongly that the pulses are not caused by boundary motion and is consistent with the extremely small boundary velocity obtained above.
Since the spatial extent of the boundary layer as a whole appears to be nearly stationary, the observed layers are mainly spatial variations due to the motion of the spacecraft for this particular case. We can estimate the spatial scale of these layers from the velocity of the spacecraft relative to the boundary layer (2.11 km/s along the boundary normal direction) and ignore the velocity of the boundary layer motion as a whole. The density depletion layer in the magnetosheath is observed from 0247 to 0315 UT immediately outside the magnetopause boundary layer. Thus the thickness of the depletion layer is 3545 km. From the time rate of change of density we can deduce that the density varies at a rate of 0.0051 cm-3 per second in the depletion layer. This gives a density gradient of 0.0024 cm-3/km in the depletion layer. The duration of the boundary layer is 35 min. The thickness of the boundary layer is 4430 km, or 0.7 RE. The pulses in the boundary layer appear to be temporal structures since pulses at ISEE 1 do not have one-to-one correlation to those at ISEE 2 as you can see from the bottom panel of Figure 8. The two spacecraft observations are unable to resolve their spatial structures.
2.2. November 11, 1979
Figure 9 shows the ISEE 2 data during the inbound magnetopause crossing from 2130 to 2330 UT on November 11, 1979. Similar to Figure 2, the top panel is the time series of the magnetic field in GSM coordinates with a time resolution of 4 s. The bottom panels show ion density, temperature, plasma beta, and velocity with time resolution of 12 s. The spacecraft trajectory during this interval is shown in Figure 1. At 2230 UT, the ISEE 2 position in GSM coordinates is (8.58, -0.63, -1.96) RE. The spacecraft is very close to the subsolar point. The ISEE 3 solar wind data for this interval are shown in Figure 10. The estimated time delay from ISEE 3 to the Earth's magnetosphere is about 50 min. The average solar wind parameters corresponding to the magnetopause boundary crossing are listed in Table 1. During this interval, the IMF is also strongly northward with clock angle of 39o from the due northward direction. The magnetosonic Mach number is 3.1 and the solar wind plasma beta is 0.2.
The ISEE 2 data in Figure 9 have many features similar to the October 5, 1979, case. The magnetic shear is very small due to the strong IMF strength and northward Bz component. The plasma beta is very low throughout this interval. The magnetopause crossing is not obvious in the lower resolution magnetic field data but can be see clearly from the two-dimensional FPE plasma data. There exists a boundary layer, highlighted by shading, in which the plasma properties are intermediate between their magnetospheric and magnetosheath values. Figure 11 shows details of the boundary layer from 2220 to 2300 UT using the highest time resolution magnetic field data. The magnetic field data are displayed in the boundary normal coordinate system. The solid line at 2225 UT marks the entry into the boundary layer, that is, the crossing of the magnetopause. We can note the following points from Figure 9 and Figure 11:
Depletion layer: In the magnetosheath immediately outside the boundary layer, there is a strong density depletion layer, where the density decreases gradually as the magnetopause is approached. Pc 1 waves are also present in the depletion layer. These properties are similar to the previous case and common for northward IMF regardless of the Mach number. Since the spacecraft is very close to the subsolar stagnation point, the flow velocity in the XY plane decreases to nearly zero when the spacecraft gets close to the magnetopause. The density depletion layer ends at the magnetopause at 2225 UT (the solid lines in Figure 9 and Figure 11).
Boundary layer: The boundary layer is observed from 2225 to 2252 UT. The transition from the magnetosheath to the boundary layer is sharp, marked by a sudden increases in temperature and a flat density profile. The intensity of the high-frequency waves decreases upon entry into the boundary layer, and eventually the waves disappear. The outer part of the boundary layer is quite uniform except for an isolated pulse of enhanced temperature and decreased density near 2241 UT. Otherwise, the ion density and temperature stay at a plateau level. There is no sharp transition from the boundary layer to the magnetosphere proper. Similar to the previous case, the density and temperature gradually approach their magnetospheric values, although the transition at 0252 UT is quite sharp. These properties are generally consistent with those observed for typical Mach numbers.
Accelerated flow in the boundary layer: The special feature in this low Mach number case is the high speed flow earthward of the magnetopause in the boundary layer. The ions are significantly accelerated at the outer edge of the boundary layer, associated with the sharp increase of the ion temperature. The accelerated flow velocity has strong +Vx and +Vz components. Combining with the magnetic field direction, we find that the accelerated plasma flows northward and mainly tangent to the magnetopause and field-aligned. The maximum field-aligned flow velocity reaches about 155 km/s northward. Sunward of the magnetopause, the same velocity component is about 60 km/s southward. At this time the spacecraft is located about 2 RE south of the magnetospheric equator, thus the accelerated flow is from the high latitude region to the equator. These observations strongly suggest that the ions in the high speed flow have not entered and accelerated locally across the magnetopause due to the lack of magnetic shear at the magnetopause. This can be verified by examining the condition of tangential momentum balance between plasma and magnetic field [e.g., Paschmann et al., 1979; Sonnerup et al., 1981]. If reconnection were occurring locally the plasma would have been accelerated by the j B force as it crossed the magnetopause current layer. For the November 11, 1979, event the tangential velocity jump across the magnetopause is about 215 km/s at its maximum. The shear of tangential magnetic field across the magnetopause current is about 4o and tangential field jump is -16.3 nT. Using the observed proton density of 8 cm-3 and about 10% alpha ion content (from solar wind data), the tangential velocity jump was estimated to be about 110 km/s according to tangential momentum balance. Such a jump can only provide a 50 km/s field aligned velocity earthward of magnetopause, which was much smaller than the observed field-aligned velocity. Thus we believe the reconnection occurs remotely in this case.
Scale lengths: The combination of ISEE 1 and 2 magnetic field data are used to calculate the velocity of the magnetopause motion. For this interval, ISEE 1 and 2 travel inbound in an identical trajectory and ISEE 2 leads ISEE 1 by about 23 min. The spacecraft velocity is 3.05 km/s along the orbit and 2.05 km/s along the boundary normal direction. Figure 12 shows the simultaneous observation of the magnetic field strength from ISEE 1 and 2. The top panel of Figure 12 shows the time series of the field strength and the bottom panel shows the field strength versus the X position of the spacecraft. There is a decrease of the field strength, a signature of the boundary layer entry, in both the ISEE 1 and 2 data. However, ISEE 2 enters the boundary layer 1742 s earlier than ISEE 1 does. The separation of the spacecraft position at the boundary layer entry (2254:30 UT for ISEE 1 and 2225:28 for ISEE 2) is 785 km along the boundary normal direction. Thus the velocity of the magnetopause motion is 0.45 km/s, and the magnetopause motion is inward.
From the speed of the boundary we find that the velocity of spacecraft relative to the magnetopause is 2.50 km/s along the boundary normal direction. From the plasma time series we find that the thickness of the depletion layer is 2700 km and the density gradient is 0.0056 cm^-3/km. The boundary layer thickness is 4500 km, or 0.7 Re.
We have presented above observations of the structure of the magnetopause boundary layer under low Mach number and northward IMF conditions. There are many features that are common for the northward IMF regardless of Mach number, including: (1) a nearly stationary boundary layer with a thickness 0.7 RE and an average speed < 1 km/s; (2) a strong density depletion layer immediately outside the magnetopause with a thickness 3000 km and a density gradient of the order of a few 10^-3 cm^-3/km; (3) an overall flat density profile in the boundary layer; and (4) no sharp transition from the boundary layer to the magnetosphere.
However, we have seen some properties in the low Mach number cases that have not been reported in previous study under typical Mach number. They are (1) at least in one case (October 5, 1979) impulsive structures are present in the boundary layer and the process that causes them persist over a long time period; (2) the pulses contain mainly heated magnetosheath plasma entered elsewhere without locally transmitted magnetosheath plasma and hot magnetospheric plasma components; and (3) there is also clear evidence of accelerated flows inside the boundary layer in both cases. These accelerated flows have a strong Vz (field-aligned) component, indicating that they come from high-latitude region.
These observations support the idea that direct plasma entry from the magnetosheath via diffusive processes is not a major contributor to the plasma population inside the boundary layer. The overall density is quite constant within the boundary layer, that is, the density gradient expected from diffusive entry is absent. The special features observed in the low mach number cases provide additional evidence for the nonlocal entry of the plasma inside the boundary. First, during the high-density and high-temperature pulses in the October 5, 1979, case, the plasma distribution functions indicated that these pulses contain neither hot magnetospheric plasma nor the transmitted magnetosheath plasma. They are heated magnetosheath plasmas that entered the boundary layer from elsewhere. Second, the accelerated flows are observed in spite of lack of local field shear for both cases. The flows have a strong field-aligned component. Diffusive processes can not explain the plasma accelerations observed within the boundary.
An alternative explanation for the boundary layer structure is reconnection above the cusps for northward IMF [e.g., Dungey, 1961; Crooker, 1979; Cowley, 1981; Paschmann et al., 1990; Gosling et al., 1991; Song and Russell, 1992], and observations in this study support this idea. In the observation of October 5, 1979, crossing the boundary layer consists of several pulses of heated sheath plasma. High-frequency waves near 1 Hz are present for most of the boundary layer but are absent during these pulses. These waves, common for northward IMF, are found to be generated and maintained in the density depletion layer and do not propagate into the magnetosphere [Song et al., 1990b]. It might be caused by the fact that the magnetic field lines are alternately connected to the magnetosphere during these pulses and to the solar wind for the rest of the time since the change of plasma distribution function might be expected at the boundary of open and closed field lines. We note that the presence and absence of 1 Hz waves due to the change of plasma distribution function maybe an indication of change of field topology. Although it may not be an unambiguous measurement of the boundary between open and close field lines, it is a very important problem and needs further investigation.
The observations of the plasmas during these pulses are consistent with the reconnection picture. On the field lines that are apparently connected to the solar wind (open field lines), the plasmas are basically sheathlike with reduced density. However, the pulses of plasmas on closed field lines look like heated sheath plasma with no hot magnetospheric plasma component and have densities much higher than those adjacent to them. Thus they are clearly not a mixture of magnetospheric and magnetosheath plasmas from adjacent regions. Rather, they could be magnetosheath plasmas which originate and are heated in cusp regions due to high-latitude reconnection. The accelerated flow on open field lines may also be the results of the reconnection at high latitude.
As the solar wind plasma enters the magnetosphere from the reconnection site, the j BN force will accelerate the plasma flow in directions tangent to the magnetopause boundary and away from the reconnection site due to the tangent momentum balance. For the northward IMF geometry when the reconnection occurs poleward of cusp region (either northern cusp or southern cusp), the accelerated plasma streams both tailward and equatorward from the cusp reconnection site. Thus, accelerated flow can move from the cusp reconnection site to the observation site near the equator for both northern and southern cusp reconnection. In both cases studied the accelerated flow velocity has a strong positive Vz peak (northward), indicating that the reconnection occurs above southern cusp region.
We have used ISEE magnetometer and FPE data to examine the structure of the magnetopause boundary layer under conditions of low Mach number and strongly northward IMF. When the solar wind Mach number is low, the IMF and magnetosheath field strength are large, and we expect effects of reconnection to be the strongest. In this study we have observed some structures that are common for both typical and low Mach numbers when the IMF is northward. However, we find some structures in the boundary layer that occur under low Mach number when the IMF is northward. These include impulsive structures in the boundary layer, nonlocal entry of heated magnetosheath plasma into the boundary layer, and clear evidence of accelerated flow for strongly northward IMF. Reconnection over the cusp regions can explain the observed field, plasma and flow signatures. Thus, we conclude that reconnection is important in the formation of the boundary layer for northward IMF at low Mach number, as well as for southward IMF.
Acknowledgments. The work at UCLA was supported by the National Science Foundation under research grant ATM91-11913. The work at Los Alamos National Laboratory was performed under the auspices of the U.S. Department of Energy and supported in part by NASA.
The Editor thanks R. L. Kaufmann and another referee for their assistance in evaluating this paper.
Akasofu, S. I., E. W. Hones Jr., S. J. Bame, J. R. Asbridge, and A. T. Y. Lui, Magnetotail and boundary layer plasma at geocentric distance of 18 Re: Vela 5 and 6 observations, J. Geophys. Res., 78, 7257, 1973.
Anderson, B. J., S. A. Fuselier, and D. Murr, Electromagnetic ion cyclotron waves observed in the plasma depletion layer, Geophys. Res. Lett., 18, 1955, 1991.
Anderson, B. J., S. A. Fuselier, S. P. Gary, and R. E. Denton, Magnetic spectral signatures in the Earth's magnetosheath and plasma depletion layer, J. Geophys. Res., 99, 5877, 1994.
Bame, S. J., J. R. Asbridge, H. E. Felthauser, J. P. Glore, G. Paschmann, P. Hemmerich, K. Lehmann, and H. Rosenbauer, ISEE 1 and ISEE 2 fast plasma experiment and the ISEE 1 solar wind experiment, IEEE Trans. Geosci. Electron., GE-16, 216, 1979.
Chapman, S., and V. C. A. Ferraro, A new theory of magnetic storms, J. Geophys. Res., 36, 77, 1931.
Crooker, N. U., Dayside merging and cusp geometry, J. Geophys. Res., 84, 951, 1979.
Cowley, S. W. H., Magnetospheric and ionospheric flow and the interplanetary field, in The Physical Basis of the Ionosphere in the Solar-Terrestrial System, p. 4, AGARD Conf. Publ., NATO, Neuilly-sur-Seine, 1981.
Dungey, J. W., Interplanetary magnetic field and the auroral zones, Phys. Rev. Lett., 6, 47, 1961.
Eastman, T. E., and E. W. Hones Jr., Characteristics of the magnetospheric boundary layer and magnetopause layer as observed by IMP 6, J. Geophys. Res., 84, 2019, 1979.
Eastman, T. E., E. W. Hones Jr., S. J. Bame, and J. R. Asbridge, The magnetospheric boundary layer: Site of plasma, momentum and energy transfer from the magnetosheath into the magnetosphere, Geophys. Res. Lett., 3, 685, 1976.
Gosling, J. T., M. F. Thomsen, S. J. Bame, T. G. Onsager, and C. T. Russell, The electron edge of the low latitude boundary layer during accelerated flow events, Geophys. Res. Lett., 17, 1833, 1990.
Gosling, J. T., M. F. Thomsen, S. J. Bame, R. C. Elphic, and C. T. Russell, Observations of reconnection of interplanetary and lobe magnetic field lines at the high-latitude magnetopause, J. Geophys. Res., 96, 10,097, 1991.
Haerendel, G., G. Paschmann, N. Sckopke, H. Rosenbauer, and P. C. Hedgecock, The frontside boundary layer of the magnetosphere and the problem of reconnection, J. Geophys. Res., 83, 3195, 1978.
Hones, E. W., Jr., J. R. Asbridge, S. J. Bame, M. D. Montgomery, S. Singer, and S. I. Akasofu, Measurements of magnetotail plasma flow made with Vela 4B, J. Geophys. Res., 77, 5503, 1972.
Luhmann, J. G., T.-Z. Zhang, S. M. Petrinec, and C. T. Russell, Solar cycle 21 effects on the interplanetary magnetic field and related parameters at 0.7 and 1.0 AU, J. Geophys. Res., 98, 5559, 1993.
Mitchell, D. G., F. Kutchko, D. J. Williams, T. E. Eastman, L. A. Frank, and C. T. Russell, An extended study of the low-latitude boundary layer on the dawn and dusk flanks of the magnetosphere, J. Geophys. Res., 92, 7394, 1987.
Ogilvie, E. W., and R. J. Fitzenreiter, The Kelvin-Helmholtz instability at the magnetopause and inner boundary layer, J. Geophys. Res., 94, 15,113, 1989.
Ogilvie, K. W., R. J. Fitzenreiter, and J. D. Scudder, Observations of electron beams in the low-latitude boundary layer, J. Geophys. Res., 89, 10,723, 1984.
Paschmann, G., Plasma structure of the magnetopause and boundary layer, in Magnetospheric Boundary Layers, Eur. Space Agency Spec. Publ., SP-148, 25, 1979.
Paschmann, G., N. Sckopke, G. Haerendel, I. Papamastorakis, S. J. Bame, J. R. Asbridge, J. T. Gosling, E. W. Hones Jr., and E. R. Tech, ISEE plasma observations near the subsolar magnetopause, Space Sci. Rev., 22, 717, 1978.
Paschmann, G., B. U. O. Sonnerup, I. Papamastorakis, N. Sckopke, G. Haerendel, S. J. Bame, J. R. Asbridge, J. T. Gosling, C. T. Russell, and R. C. Elphic, Plasma acceleration at the Earth's magnetopause: Evidence for reconnection, Nature, 282, 243, 1979.
Paschmann, G.,I. Papamastorakis, W. Baumjohann, N. Sckopke, C. W. Carlson, B. U. O. Sonnerup, and H. Luehr, The magnetopause for large magnetic shear: AMPTE/IRM observations, J. Geophys. Res., 91, 11,099, 1986.
Paschmann, G., B. Sonnerup, I. Papamastorakis, W. Baumjohann, N. Sckopke, and H. Luehr, The magnetopause and boundary layer for small magnetic shear: Convection electric fields and reconnection, Geophys. Res. Lett., 17, 1829, 1990.
Paschmann, G., W. Baumjohann, N. Sckopke, T.-D. Pan and H. Luehr, Structure of the dayside magnetopause for low magnetic shear, J. Geophys. Res., 98, 13,409, 1993.
Phan, T.-D., G. Paschmann, W. Baumjohann, N. Sckopke, and H. Luehr, The magnetosheath region adjacent to the dayside magnetopause: AMPTE/IRM observations, J. Geophys. Res., 99, 121, 1994.
Russell, C. T., The ISEE 1 and 2 fluxgate magnetometers, IEEE Trans. Geosci. Electron., GE-16, 239, 1978.
Russell, C. T., and R. C. Elphic, Initial ISEE magnetometer results: Magnetopause observations, Space Sci. Rev., 22, 681, 1978.
Sckopke, N., G. Paschmann, G. Haerendel, B. U. O. Sonnerup, S. J. Bame, T. G. Forbes, E. W. Hones Jr., and C. T. Russell, Structure of the low-latitude boundary layer, J. Geophys. Res., 86, 2099, 1981.
Scurry, L., C. T. Russell, and J. T. Gosling, A statistical study of accelerated flow events at the dayside magnetopause, J. Geophys. Res., 99, 14,815, 1994.
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, 1992.
Song, P., C. T. Russell, J. T. Gosling, M. F. Thomsen, and R. C. Elphic, Observations of the density profile in the magnetosheath near the stagnation streamline, Geophys. Res. Lett., 17, 2035, 1990a.
Song, P., R. C. Elphic, C. T. Russell, J. T. Gosling, and C. A. Cattell, Structure and properties of the subsolar magnetopause for northward IMF: ISEE observations, J. Geophys. Res., 95, 6375, 1990b.
Song, P., C. T. Russell, and C. Y. Huang, Wave properties near the subsolar magnetopause: Pc 1 waves in the sheath transition layer, J. Geophys. Res., 98, 5907, 1993.
Sonnerup, B. U. O., G. Paschmann, I. Papamastorakis, N. Sckopke, G. Haerendel, S. J. Bame, J. R. Asbridge, J. T. Gosling, and C. T. Russell, Evidence for magnetic field reconnection at the Earth's magnetopause, J. Geophys. Res., 86, 10,049, 1981.
Takahashi, K., D. G. Sibeck, P. T. Newell, and H. E. Spence, ULF waves in the low-latitude boundary layer and their relationship to magnetospheric pulsations: A multisatellite observation, J. Geophys. Res., 96, 9503, 1991.
G. Le and C. T. Russell, Institute of Geophysics and Planetary Physics, University of California, Los Angeles, CA 90024-1567 (e-mail: firstname.lastname@example.org; email@example.com).
(Received June 25, 1993; revised August 12, 1994; accepted August 18, 1994.)
Figure 1. ISEE 1 and 2 trajectories near the magnetopause crossings in GSM XY and YX planes. (a) 0130-0330 UT, October 5, 1979. (b) Orbit 2130-2330 UT, November 11, 1979.
Figure 2. ISEE 2 magnetic field and two-dimensional FPE ion density, temperature, and bulk velocity for magnetopause crossing from 0130 to 0330 UT, October 5, 1979.
Figure 3. Simultaneous solar wind data for October 5, 1979, magnetopause crossing.
Figure 4. High-resolution magnetic field data and FPE ion data showing the structure of boundary layer for the October 5, 1979, magnetopause crossing. The bottom panel is the Vz component (north-south) from three-dimensional FPE measurements at a lower rate.
Figure 5. High-pass-filtered magnetic BM and BN components and the magnetic field strength for the October 5, 1979, magnetopause crossing.
Figure 6. Three second snapshots of two-dimensional (top) ion and (bottom) electron velocity distribution function observed inside the high-density and high-temperature pulses for the October 5, 1979, case. The corresponding times are marked by solid lines in Figure 5. The distributions are shown as contours of constant phase-space density spaced logarithmically. The numbers on dotted circles is the velocity scale in kilometers per second.
Figure 7. Three second snapshots of two-dimensional (top) ion and (bottom) electron velocity distribution function observed outside the high density and high temperature pulses for the October 5, 1979, case. The corresponding times are marked by dashed lines in Figure 5.
Figure 8. (Upper) Time series of magnetic field strength from both ISEE 1 and ISEE 2. (Lower) The magnetic field strength from both ISEE 1 and ISEE 2 as a function of the spacecraft X coordinate.
Figure 9. ISEE 2 magnetic field and two-dimensional FPE ion density, temperature, and bulk velocity for magnetopause crossing from 2130 to 2330 UT, November 11, 1979.
Figure 10. Simultaneous solar wind data for November 11, 1979, magnetopause crossing.
Figure 11. High-resolution magnetic field data and FPE ion data showing the structure of boundary layer for November 11, 1979, magnetopause crossing. The bottom panel is the Vz component (north-south) from three-dimensional FPE measurements at a lower rate.
Figure 12. (Upper) Time series of magnetic field strength from both ISEE 1 and ISEE 2. (Lower) The magnetic field strength from both ISEE 1 and ISEE 2 as a function of the spacecraft X coordinate.