VLF Waves in the Foreshock

R. J. Strangeway* and G. W. Crawford**

*Institute of Geophysics and Planetary Physics,
University of California at Los Angeles

**Radio Atmosphere Science Center
Kyoto University at Kyoko 611, Japan
Now at SRI International, Menlo Park, California 94025, U. S. A.

Adv. Space Res., vol 15, (8/9)29-(8/9)42, 1995
Copyright 1995 by COSPAR

Next: Down-shifted Plasma Oscillations
Previous: VLF Emissions at Venus: Examples
Top: Title and Abstract

VLF Emissions at Venus: Mappings

      Although the PVO plasma wave instrument suffers from a lack of frequency resolution, there are advantages to studying VLF data at Venus. First of all, the PVO orbit had apoapsis around 12 Rv, much larger than the size of the Venus obstacle, and so the spacecraft spent a significant amount of time on each orbit in the unshocked solar wind. Furthermore, the spacecraft was in the solar wind both in the sub-solar region, and also well past the planetary terminator, thus allowing a much larger area of the foreshock to be mapped. Last the spacecraft was in orbit around Venus for 14 years, and we can carry out large statistical studies. A preliminary study /29/ using data from nearly one Venus year (200 orbits) showed the usefulness of imaging the Venus foreshock using VLF data. Subsequently /20/, this study was extended using a separate set of 650 orbits. The results of this study are presented in Figures 5 and 6.

Fig. 5.VLF map of the Venus foreshock for nominal Parker spiral (after /20/).

Fig. 6.VLF map of the Venus foreshock for perpendicular IMF (after /20/).

      In Figure 5 the data have been restricted to those intervals for which the IMF was within 10° of the nominal Parker spiral in the B-v plane. As discussed in /20/ and /29/, the technique employed in generating the maps is to calculate for each observation the distance along the tangent field line and depth behind the tangent line using the instantaneous IMF, and a model bow shock that has been scaled to the observed bow shock location for each orbit. In scaling the bow shock, we have only changed the semilatus rectum (L) of the conic of revolution that specifies the shock. This conic of revolution is given by

where R is radial distance from the focus, is the eccentricity, and sz is the solar zenith angle with respect to the focus. In generating the maps we used a bow shock model with = 1.03, and focus at 0.45 Rv sunward of the planet /20, 21, 30/. When matching bow shock models to actual observations, it is preferable to keep the shock shape (specified by ) and focus fixed, and allow the size (specified by L) to change /31/.

      Having calculated depth and distance, these are then converted to a Cartesian coordinate location with respect to the point of tangency in the B-v plane by assuming the IMF is at the nominal Parker spiral angle (35° at Venus), even though the instantaneous IMF may be at some other angle. This is done to prevent "smearing" of the tangent field line, which would occur if the instantaneous IMF orientation was used to specify the location. Finally, all the parallel B-v planes are mapped to the equatorial plane, using the point of tangency in each plane as the common reference point. Although we have mixed flank bow shock intersections with sub-solar intersections, this does not appear to drastically alter the statistical results /20/. Once the data have been mapped to a common foreshock geometry, the data are accumulated in bins with l l Rv resolution.

      At the top left of Figure 5 we show the 9th decile of the 30 kHz wave intensity. Although the data have been accumulated in l l Rv bins we have interpolated the data in generating the maps. The color scale shows the log 10 of the wave intensity. The small black dot is the point of tangency. At the bottom left of the figure we show the log 10 of the number of samples per l l Rv bin. Throughout most of the distribution we have over 100 samples per bin, and thus we have high stastitical confidence in the results. In the bottom panel we have superimposed the reference bow shock model used in generating the maps, given by L = 1.69 Rv /21, 26/.

      It is clear that the 30 kHz emissions occur mainly along the tangent field line. In addition, the wave intensity only reaches a maximum a few Rv away from the point of tangency. Moreover, the wave intensity shows a marked decrease some 15 Rv upstream of the point of tangency. Similar mapping studies using ISEE-3 data show wave emissions extending up to ~ 100 Rv along the tangent field-line /32/. Thus the electron foreshock emissions seem in part to be controlled by the scale size of the shock, consistent with limits being imposed on the Fast Fermi process through shock curvature, as noted earlier. If the drop-off was due to some inherent properties of the plasma instability and subsequent saturation processes, we would expect the drop-off scale to be independent of shock-size, instead being controlled by some scale dependent on the ambient plasma parameters. Since the solar wind is very similar at the Earth and Venus, such a scale factor would be the same at both planets.

      We also note that in Figure 5 the electron foreshock emissions that are upstream of the tangent point appear to be stronger than the emissions in the downstream region. This could be attributed to two possible causes /20/. The first is that the solar wind electron distribution is asymmetric, there is a heat flux from the sun, and electrons energized at the bow shock will be flowing in the same direction as this heat flux in the downstream foreshock, thus reducing the slope in the distribution function, and hence the growth rate for instability. The second possibility can be seen on inspection of (1). While the electrons are coming from regions with roughly the same bn, the shock normal is almost parallel to the solar wind flow in the sub-solar region, while the shock normal is more nearly perpendicular along the flanks. The vndependence of the de Hoffman-Teller velocity implies that the electron acceleration will be weaker for the downstream foreshock. However, the interplay between changes in vn and changes in bn requires further analysis, similar to /18/, but taking into account both upstream and downstream foreshocks.

      In this paper we have concentrated on the electron foreshock emissions, but one or two remarks are in order concerning the ion foreshock emissions at Venus, as revealed in the Figure 5 At the right of Figure 5 we include maps of the 9th decile of wave intensity measured at 5.4 kHz, and the standard deviation of the trace of the magnetic field. The 5.4 kHz channel is used to monitor the ion acoustic emissions in the foreshock, while the magnetic field deviation is a proxy for the presence of ULF waves. The line intersecting the model bow shock behind the tangent field line is the ULF boundary obtained from terrestrial studies /33/, but modified for the nominal Parker spiral orientation at Venus /21/. The ULF waves are most intense near the bow shock, and tend to be confined to locations behind the ULF boundary. The 5.4 kHz emissions, on the other hand, appear to be confined to locations even further downstream of the model ULF boundary. It is normally assumed, and Figure 4 appears to bear this out, that the ULF waves and VLF ion acoustic waves are both seen together. However, Figure 5 implies that this is only the case for observations deep in the ion foreshock. This suggests that different plasma populations are responsible for the different waves. Unfortunately, no similar mapping studies of the terrestrial ion foreshock have been carried out to determine if this result is unique to Venus. The smaller shock scale size at Venus will probably result in weaker Fermi and shock-drift acceleration, This, together with the lack of magnetosphere to act as a reservoir of energetic leakage particles, leads us to expect that the ion foreshock at Venus will be populated by lower energies and fluxes than at the Earth.

      In Figure 6 we repeat the foreshock analysis, but now for IMF orientations nearly perpendicular to the solar wind flow. We find that the electron foreshock emissions are still present, but with intensities that lie between those observed in the upstream and downstream foreshock as shown in Figure 5 That the 30 kHz emissions are weaker than in the upstream foreshock of Figure 5 is consistent with our arguments concerning the vn dependence of the electron reflection process. The ion foreshock VLF emissions are almost completely absent. There is perhaps a hint of ULF emissions close to the shock behind the model ULF boundary.

Next: Down-shifted Plasma Oscillations
Previous: VLF Emissions at Venus: Examples
Top: Title and Abstract

Go to R. J. Strangeway's Homepage

Converted to HTML by Shaharoh Bolling
Last Updated: Feb. 16, 2000