Pages 699-702

Note:  The paper appearing in this CD-ROM has been slightly revised from that appearing in the hard copy of Advances in Space Research,Volume 20, Issue 4-5, pages 699-702.


Y. Kasaba 1, H. Matsumoto 1, , and R. R. Anderson 2

1 Radio Atmospheric Science Center, Kyoto University, Uji, Kyoto, 611, Japan,
2 Department of Physics and Astronomy, The University of Iowa, Iowa City, Iowa 52242, USA.


2fp emission generated in the terrestrial electron foreshock is observed with GEOTAIL at 10-30 RE from the Earth. We determine the source region by three statistical methods: mapping of 2fp flux, analysis of bifurcation phenomena with density discontinuity in the solar wind, and determination of propagation direction by spin modulation. We show that the source is generally on the tangential field line, but not concentrated around contact point of the tangential field line on the bow shock surface. Typical distance of the source from the contact point is about 10-40 RE.


Electromagnetic 2fp emission has been frequently observed in the upstream of the terrestrial bow shock at twice the solar wind electron plasma frequency, fp. In the upstream of the bow shock, backstreaming electron beams generate strong Langmuir waves and form the electron foreshock region at the downstream of the interplanetary magnetic filed (IMF) lines tangent to the bow shock (Filbert and Kellogg, 1979; Fitzenreiter, 1995). 2fp emission is believed to be generated from a coupling of these Langmuir waves by the same plasma emission mechanisms as type III solar radio bursts. However, these generation mechanisms of 2fp emission are not established well still now, and the foreshock region provides a valuable laboratory to us.
GEOTAIL entered into the 'Near-Tail Orbit' from November 1994 (Figure 1), and frequently passed electron foreshock. Therefore, we could get a large sample of in situ observation around the source region. In this study, we examine these data sets and investigate a geometry of the 2fp source region with three kinds of statistical analyses: mapping of 2fp flux, analysis of bifurcation phenomena with density discontinuity in the solar wind, and determination of propagation direction by spin modulation.


We use 1 min sampling data sets of GEOTAIL Plasma Wave Instrument (PWI) from No. 26, 1994 to Nov. 9, 1995. We examine both Langmuir waves and 2fp emission using Sweep Frequency Analyzer (SFA), a part of PWI. Frequency of Langmuir waves are automatically counted by peak searching within +-10 kHz of local fp defined by Low Energy Particle Experiment (LEP) density data. On the other hand, as the 2fp emission is not generated in the vicinity of GEOTAIL, we need to estimate twice of fp in the source region. We first assume that the source of 2fp emission lies around XGSE = -15 RE, and then estimate fp in the source from fp at s/c with time gap between GEOTAIL and the source by solar wind convection. In this analysis, we have to remove contamination in the 2fp flux by saturation effects and other terrestrial waves. When SFA saturates by intense Langmuir waves, many harmonics of Langmuir waves are enhanced and contaminate the 2fp flux. We reject data when 3fp flux exceeds -155 dB Vm-1Hz-1/2. This method is also effective to reject other broadband emissions, such as auroral kilometric radiation or continuum radiation. Part of the 'saturated' 2fp emission might include locally excited electrostatic 2fp emission. We can not pick them up by the present method. The foreshock geometry is defined by the IMF direction and the bow shock location, as indicated in Figure 2 (cf. Filbert and Kellogg, 1979). In this study, both are expressed by modified GSE coordinates corrected with solar wind aberration, written as (Xab, Yab, Zab). We should correct variations of bow shock locations by the solar wind ram pressure, Psw. We assume the bow shock locations, as

(Binsack and Vasyliunas, 1968; Filbert and Kellogg, 1979). The constants as0 an bs0 are 14.6 RE and 25.6 RE, respectively. We compare the predicted and the observed bow shock location for 30 GEOTAIL bow shock crossings. The predicted bow shock locations agree well with the observed locations within 2 RE error, but have larger error for |Yab|>20 RE. To analyze the data sets with different IMF directions together, we use two parameters 'Diff' and 'Dist' for foreshock geometry description. Diff is s/c depth toward downstream from the tangential field line, and Dist is s/c distance from the contact point along the tangential field line.


Intensity map of Langmuir waves in the Earth's foreshock was first presented by ISEE-3 (Greenstadt et al., 1995). We expand their results with higher time and spatial resolution. In Figure 3, we plot the averaged flux of Langmuir waves and of 2fp emission on the 'Diff-Dist' coordinates. Spatial resolution is 2 RE. We have confirmed previous results that strong Langmuir waves are concentrated around Diff=0 RE, close to the tangential field lines, and weak Langmuir waves extend downstream of the tangential field lines at |Dist|>30 RE. 2fp emission is not concentrated on the tangential field line because 2fp emission can propagate and its flux is strongly affected by kinetic energy flux of the solar wind. Generally, 2fp emission is distributed around the tangential field line.
We also find two interesting facts. First, strong Langmuir waves and 2fp emission are relatively weak in |Dist|<10 RE, close to the contact point, even on the tangential field line. Second, the flux of 2fp emission decreases beyond |Dist|>40 RE.


`Bifurcation' of the 2fp emission is sometimes observed with passage of the density discontinuity in the solar wind (Lacombe et al., 1988; Reiner et al., 1996). An example is shown in Figure 4. These phenomena are caused by progressive variation of the plasma frequency in the extended source region. New 2fp emission appears when the discontinuity reaches the sunward edge of the 2fp source region, and old 2fp emission vanishes when the discontinuity reaches the tailward edge the 2fp source region. Therefore, we can define the sunward/tailward edges of the source region on the Sun-Earth line, Xsunward and Xtailward, from appearance and vanishing time of 2fp emissions.

Xs/c and T1 are location and time when the discontinuity passed s/c. T2 and T3 are the appearance time of new 2fp emission and the vanishing time of old 2fp emission. Vsw is the solar wind velocity. For this analysis, we assumed that the normal of discontinuity in each case is parallel to the Sun-Earth line for simplification. We find 32 cases of such bifurcation phenomena from February to October 1995, under a condition that Diff is within 2 RE for 5 min.
Figure 5 shows results in Xab-sqrt(Yab2+Zab2) coordinates. Circle indicates contact point at each event defined from IMF direction. Bars indicate extension of the 2fp source region along Xab axis. In Figure 5, we can find that sunward extension of the source region is limited up to 5 RE from the contact point. On the other hand, tailward extension reaches up to 10-20 RE. These features are inconsistent with symmetric existence of the source obtained in MAPPING results. One of possible causes is that electron beams on the tangential field line are disturbed and destroyed when crossing the discontinuity. On this point of view, 2fp emission is hard to be generated beyond upstream-side of the discontinuity, and sunward edge of 2fp source region is generally under-estimated.


Propagation direction of electromagnetic waves are determined by spin modulation. Analysis of spin modulation for 2fp emission has been done using data from far upstream by ISEE-3 and WIND (Hoang et al., 1981; Reiner et al., 1996). They concluded that 2fp source lies within the electron foreshock in the vicinity of the bow shock. On this study, we statistically analyze propagation direction for 21 cases observed from Feb. to Nov. 1995. We should mention that the propagation direction of 2fp emission is generally scattered in GEOTAIL data. This might be caused by fast variation of 2fp flux, scattering in the ion foreshock or at the bow shock, and/or too close s/c locations to the source. The propagation direction of the selected samples stay within 10 deg. for 30 min. For all cases studied, IMF direction stay within 10 deg. for 30 min, and |phi_B| <30 deg., coincidentally.
Figure 6 shows two examples of the propagation directions for different IMF-B angle ranges. The propagation angles seem to be parallel with IMF direction, or to point to the tangential field line with |Dist|>10 RE. Those indicate that source region of 2fp emission is NOT concentrated around contact point, nor distribute far away. Those are consistent with our MAPPING results.


We have defined 2fp source region by three methods: analysis of mapping, bifurcation phenomena, and propagation direction. We confirm that the 2fp source region is on the tangential field line, coexisting with strong Langmuir waves. This is one of direct evidence that 2fp emission is generated from intense Langmuir waves. We also find that both Langmuir wave and 2fp emission is not strong around contact point, and that 2fp source region does not extend too far from the contact point. Typical distance of major 2fp source from contact point is about 10-40 RE. In summary, our current conclusion is as follows: (1) The near contact point region (|Dist|<10 RE) is not the source region of 2fp emission. This may be due to a fact that a sharp electron beam is not formed through the velocity filter in the region close to the contact point, because of a shortage of flight time. Another possible mechanism is some blocking effect of non-linear generation mechanisms of 2fp emission due to presence of ion beam and/or ion acoustic waves. (2) The source region is limited up to 40 RE. This may be due to the consumption of free energy of the electron beam disabling the capability of 2fp excitation.
In-situ features of waves and particles on the tangential field line and comparison of 2fp flux and direction by GEOTAIL and WIND are important clues to understanding the 2fp emission features presented in this paper. We also hope to compare our results with numerical experiments and detailed wave data around other planets. These results will be presented in near future.


We would like to thank M. J. Reiner and G. K. Crawford for their valuable discussion. We thank GEOTAIL MGF team (S. Kokubun and T. Yamamoto, PI) and LEP team (T. Mukai, PI) for supplying magnetic field and particle data in the solar wind. We also thank all members of GEOTAIL PWI team for their help and interest.


Binsack, J. H., and V. M. Vasyliunas, Simultaneous IMP 2 and OGO 1 observations of bow shock compression, J. Geophys. Res., 73, 429 (1968).

Filbert, P. C., and P. K. Kellogg, Electrostatic noise at the plasma frequency beyond the Earth's bow shock, J. Geophys. Res., 84, 1369 (1979).

Fitzenreiter, R. J., The electron foreshock, Adv. Space Res., 15, 9 (1995). Greenstadt, E. W., G. K. Crawford, R. J. Strangeway, S. L. Moses, and F. V. Coroniti, Spatial distribution of electron plasma oscillations in the Earth's foreshock at ISEE 3, J. Geophys. Res., 100, 19933 (1995).

Hoang, S., J. Fainberg, J. L. Steinberg, R. G. Stone, and R. H. Zwickl, J. Geophys. Res., 86, 4531 (1981).

Lacombe, C., C. C. Harvey, S. Hoang, A. Mangeney, J. L. Steinberg and D. Burgess, ISEE observations of emission at twice the solar wind plasma frequency, Ann. Geophys., 6, 1, 113 (1988).

Reiner, M. J., M. L. Kaiser, J. Fainberg, M. D. Desch, and R. G. Stone, 2fp radio emission from the vicinity of the Earth's foreshock: WIND observations, Geophys. Res. Lett., 23, 1247 (1996).