Two teams analysing cosmic-ray data from Voyager I cannot agree if it has become the first spacecraft to reach the termination shock of the solar wind
Where is the edge of the solar system? The accepted answer is that the solar system ends where the “solar wind” – the material that flows from the top of the Sun’s atmosphere – becomes so spread out that it is no longer able to push its way through the interstellar gas. However, before it gets this far the solar wind must pass through another important boundary known as the “termination shock”. This is the region beyond which the solar wind, which is initially supersonic, suddenly becomes much slower. Its flow then becomes subsonic and eventually merges with the interstellar gas (see figure).
The actual termination shock is almost certainly neither stationary nor smooth and it is expected to move back and forth, both as a whole and in localized regions, as the solar wind fluctuates. Most recent estimates place the shock at a distance of between 90 and 120 astronomical units (AU) from the Sun, where 1 AU is the mean distance from the Sun to the Earth (about 150 million km).
The precise nature of this boundary, and its distance from the Sun, have remained the subject of theory and speculation, even as spacecraft searching for it – such as the two Voyager missions launched in 1977 – have moved further and further away from the Sun. Voyager I is currently about 90 AU from the Sun, whereas Voyager II is 73 AU away on a different trajectory.
Although Voyager I is clearly out in front, its plasma detector malfunctioned after it passed Saturn, so it cannot make a direct detection of the shock by simply measuring the gas velocity or density. However, data from other instruments that measure charged particles, magnetic fields and radio waves should show characteristic changes as the shock is crossed. Two groups of researchers have now reported the results of charged-particle measurements – but they have reached opposite conclusions. However, magnetic-field and radio-wave data suggest that the shock has not yet been crossed.
Particle puzzles
We already know a good deal about the termination shock and the region beyond it from Earth-based measurements of high-energy cosmic rays. Some of these cosmic rays are described as “anomalous” because when they were first discovered in the 1970s they could not be explained by the then current models. We now know that they are accelerated by the electric and magnetic fields at the termination shock. Earth-based instruments measure the number of cosmic rays as a function of energy (to obtain their spectrum) and direction of origin (to detect any anisotropies). Clearly, measurements of anomalous cosmic-ray particles – including helium, oxygen and neon nuclei and protons – in the vicinity of the termination shock could provide much more information.
Recently, two groups with instruments on Voyager I reported an event in which the intensities of anomalous cosmic rays increased by more than a factor of 10 for a period of seven months in 2002-2003. The researchers argue that these increases did not originate at the Sun, and although both groups suggest that the event may have been related to the termination shock, they differ considerably in their conclusions.
Tom Krimigis of the Johns Hopkins University in the US and co-workers claim that the event is the result of Voyager I crossing the termination shock – twice. More precisely, they argue that the shock moved inward, crossing Voyager in the process, and seven months later moved out again, crossing Voyager for a second time. In this picture the increased flux of cosmic rays was the result of Voyager being in the post-shock flow (S M Krimigis et al. 2003 Nature 426 45-48).
However, Frank McDonald of the University of Maryland and colleagues argue that although the event may well be related to proximity to the shock, Voyager I did not actually cross the shock boundary (F B McDonald et al. 2003 Nature 426 48-50). It is important to note that the results obtained from the two instruments, although somewhat different, basically agree and that the difference between the two teams is in the interpretation.
Krimigis and co-workers use three main facts to make their case. First, the average anisotropy of cosmic-ray protons with energies of about 1 MeV is small and directed radially outward. If this is interpreted as being entirely the result of convection with the ambient solar wind, then the speed is low and consistent with what is expected in the post-shock region. It should also be noted that they observe large fluctuating anisotropies along the magnetic field, which come from the solar direction and which are difficult to interpret.
Second, the energy spectra at energies well below 1 MeV per nucleon show behaviour that is expected at or behind the shock. Finally, the composition of the enhanced intensity of energetic particles is not consistent with solar particles, but it is consistent with anomalous cosmic rays.
On the other hand, McDonald and co-workers insist that the event was a precursor to crossing the shock, and that the shock was not actually crossed. They make two main points. First, the spectrum of anomalous cosmic rays at energies near 25 MeV per nucleon shows a peak, which is a signature of the spacecraft being a significant distance upstream of the shock (i.e. on the solar side). They also point out that the large – and unexpected – anisotropies that are observed by both teams are very highly variable and generally in the azimuthal direction. Finally, and correctly, they point out that the average anisotropy is actually the result of the combined effect of diffusion (which is neglected by Krimigis and co-workers) and convection, and so cannot be used unambiguously to determine the underlying plasma velocity.
Bottom lines
These measurements of energetic particles are both noteworthy because they strongly suggest that Voyager I is, at the very least, quite near the shock. To this author, the most convincing data are the energy spectra. However, the simultaneous observation of a smooth power law from MeV energies to less than 0.1 MeV, and also a peak at about 25 MeV per nucleon, are difficult to fit into one picture. The former suggests that Voyager was at or downstream of the shock, while the latter suggests equally strongly that it is still a significant distance upstream. The different interpretations show that the shock is probably considerably different from what we had expected.
Voyager I also sent back magnetic-field and radio-wave data during this period. Both sets of data should show characteristic signatures of crossing the shock , but nothing was found. A significant increase in magnetic-field strength would have been detected for the entire duration of the event if Voyager were indeed in the post-shock flow. This was not seen (L F Burlaga et al. 2003 Geophys. Res. Lett. 30 2072). This is a very important finding that supports the argument that the shock was not crossed. It is not at all clear how to slow down the solar wind, with or without a shock, without an associated increase in the magnetic field. The radio-wave data show no evidence of crossing the shock either (D A Gurnett et al. 2003 Geophys. Res. Lett. at press).
At the time of writing, I learned of another event similar to the 2002 event, but with only about 40% of its amplitude. It began in August 2003 and is still in progress.
Irrespective of how the current controversy is resolved, Voyager I is moving away from the Sun at a speed of about 4 AU per year and the termination shock or its equivalent will eventually be crossed. Voyager I will then enter a totally new region of space called the heliosheath and, if it lasts long enough, go on to become the first spacecraft to leave the solar system and observe the plasma between the stars.
