Recent Voyager 1 data indicate that on 25 August 2012 at a distance of 121.7 AU from the Sun, sudden and unprecedented intensity changes were observed in anomalous and galactic cosmic rays


  • W. R. Webber,

    Corresponding author
    1. Department of Astronomy, New Mexico State University, Las Cruces, New Mexico, USA
    • Corresponding author: W. R. Webber, Department of Astronomy, New Mexico State University, P.O. Box 30001, 1320 Frenger St., Las Cruces, NM 88003, USA. (

    Search for more papers by this author
  • F. B. McDonald

    1. Institute of Physical Science and Technology, University of Maryland, College Park, Maryland, USA
    2. Deceased August 31, 2012
    Search for more papers by this author


[1] At the Voyager 1 (V1) spacecraft in the outer heliosphere, the intensities of both anomalous cosmic rays and galactic cosmic rays (GCRs) changed suddenly and decisively on 25 August (121.7 AU from the Sun). Within a matter of a few days, the intensity of 1.9–2.7 MeV protons and helium nuclei had decreased to less than 0.1 of their previous value, and eventually the intensities decreased by factors of at least 300–500. Also, on 25 August, the GCR protons, helium, and electrons increased suddenly in just 2 or 3 days by a factor of up to 2. The intensities of the GCR nuclei of all energies from 2 to 400 MeV then remained essentially constant with intensity levels and spectra that may represent the local GCR. The suddenness of these intensity changes indicates that V1 has crossed a well-defined boundary for energetic particles at this time possibly related to the heliopause.

1 Introduction

[2] The passage of the Voyagers 1 (V1) and 2 spacecraft through the outer heliosphere (heliosheath) has revealed a region quite unlike the inner heliosphere inside the heliospheric termination shock (HTS). The radial solar wind speed slows down from ~400 to ~130 km/s [Richardson et al., 2008] and later, at about 20 AU beyond the HTS, may decrease to very low values [Krimigis et al., 2011]. The anomalous and galactic cosmic ray (ACR and GCR) intensities hardly changed at the HTS contrary to theoretical expectations [Stone et al., 2005]. The magnetic field shows many distinct structures or features associated with the HTS and the heliosheath region beyond [Burlaga and Ness, 2010]. One of the largest of these structures was encountered by V1 at 2009.7 when the spacecraft was ~17 AU beyond the HTS crossing distance of 94 AU. At this time, the field direction suddenly changed from 90° to 270°, possibly indicating a sector crossing. Also, at this time, the GCR electron intensity increased by an unprecedented 30% and the radial intensity gradients of these electrons and higher-energy nuclei decreased by over a factor of 2 [Webber et al., 2012]. Sudden intensity increases of electrons and nuclei again occurred about 1.5 years later at a distance of ~116.5 AU or 22.5 AU beyond the HTS crossing distance. More recently, a new series of changes have been observed starting at about 2012.0 in both GCR and ACR. In particular, at about 2012.35 at ~120.5 AU from the Sun, large increases of both GCR nuclei and electrons were observed with little corresponding changes of ACR. In fact, throughout many of these unusual GCR intensity changes in the outer heliosphere, the ACR H, He, and O nuclei from ~1 to 50 MeV hardly changed at all, and any changes in ACR and GCR were not always correlated.

[3] The ACR represents the dominant energetic population in the heliosheath above ~1 MeV with intensities ~102–103 times those observed in the heliosphere inside the HTS. These ACR particles are accelerated somewhere in the heliosheath (several mechanisms are possible) and remain quasi-trapped there, leaking into the inner heliosphere where they are only weakly observed at the Earth. At the outer boundary of the heliosheath, these particles may also leak out into the interstellar region.

[4] On 25 August, when V1 was at 121.7 AU from the Sun, the intensity of the ACR component began to decrease rapidly. Within a few days, the intensity of this dominant energetic heliosheath component above 1–2 MeV decreased by more than 90–95%, reaching intensity levels not seen at V1 since it was well inside the HTS. At the same time, a sudden increase of a factor of ~2 occurred in lower-energy (6–100 MeV) electrons and ~30–50% for the higher-energy nuclei above 100 MeV. This simultaneous reduction of ACR intensities at lower energies and the abrupt increase in GCR intensities at higher energies have suddenly revealed one of the holy grails of GCR studies—the actual local interstellar spectra (LIS) of the GCR nuclei from H to Fe above ~10–20 MeV and possibly even to lower energies. For the multidimensional CRS instrument used here [Stone et al., 1977], the intrinsic backgrounds are so low that the observed reduction of ACR is at least a factor ~300–500, making the low-energy GCR measurements possible.

[5] This large decrease of ACR was preceded by two precursor temporary decreases starting on 28 July and 14 August. Thus, V1 may have crossed a boundary, which itself was very sharp, at least five times during this time period.

[6] It is this transition into a new region and some of its implications that we wish to summarize in this paper. Further details of these remarkable events will be presented in subsequent articles.

2 Data

[7] In Figure 1, we show the time history from 2012.2 to the present of GCR nuclei >200 MeV and electrons of 6–14 MeV as well as ~1 MeV H nuclei as representative of the lower-energy energetic particle population. This figure can be thought of as a follow-up to Figure 2 in the recent paper on Voyager observations beyond 111 AU by Webber et al. [2012]. The complexity of the intensity changes in the time period beyond 2012 is evident. The suddenness and decisiveness of the last change in the sequence at a time of 2012.65 (25 August) are extraordinary. The increase at 2012.35, which occurred for GCR particles but not for the lowest-energy ACR, and also the two precursor changes at 2012.57 and 2012.615, which applied to both GCR and ACR, are also significant and will be discussed later. In the two precursor intensity changes, the ACR decreased considerably and the GCR increased, mimicking the last change at 2012.65 seen in Figure 1.

Figure 1.

Five day running average intensities of 0.5 MeV protons times 0.25, curve A (mainly ACR), 6–14 MeV GCR electrons times 80, curve B, and >200 MeV protons times 2.0, curve C, from 2012.2 to the end of data. The various intensity jumps for both ACR and GCR as indicated by the shaded regions are discussed in the text.

Figure 2.

Weekly average of proton spectra from 2 to 400 MeV measured at V1 during the rapid decrease period starting from about 2012.61 to the end of data (each week = 0.07 AU of V1 outward movements). The reference time period in blue is from 2011.8 to 2012.2. The ratio of intensities in percent between the final time period to those in the reference period is shown along the bottom of the graph below the average energy of each energy interval. The increasing ratio (percentage) above 10 MeV indicates the increasing contribution of GCR to the total intensity in that interval. The numbers to the left of each weekly average spectrum are the weeks of the year 2012, with 41–44 in red representing a 4 week average as the intensities stabilize. Other features of the figure are discussed in the text.

[8] In the time period of just a few weeks after 2012.65 (1 week = 0.07 AU of outward movement by V1), the ACR intensity between 2 and 10 MeV dropped to levels of less than 1% of the values seen earlier at these energies, hereafter referred to as the “heliocliff,” and the GCR nuclei and electron intensities increased by ~1.5 and 2.0 times, respectively, to the highest levels observed since the launch of V1. These GCR intensities have remained almost constant at these high levels for 6 months.

[9] To see how these intensity changes affect the energy spectra of protons and helium nuclei being measured at V1, we show Figures 2 and 3. For protons in Figure 2, note that the CRS particle telescopes measure the proton spectra from ~2 to 400 MeV, and for helium nuclei in Figure 3, the energy range is ~2–600 MeV. Over the entire energy range, the CRIS telescopes use a two- or three-parameter analysis [Stone et al., 1977], which practically eliminates the background that is inherent in a single parameter threshold type of analysis. This enables us to accurately measure these low intensities

Figure 3.

Same as in Figure 2 except this are He data from 2 to 600 MeV/nuc.

[10] It is seen that for both protons and helium nuclei, the large reduction in ACR intensity at lower energies produces intensity levels only seen previously in the inner heliosphere at quiet times. The intensities of both H and He nuclei that are remaining after about 2012.75 look much like some earlier predictions of possible GCR spectra, perhaps down to ~10 MeV and below. So the “boundary” that V1 has just crossed several times between 2012.51 and 2012.65 is an extremely effective barrier for the ACR, reducing the intensities by >99% in less than 0.2 AU. It is also an effective barrier for GCR flowing inward equivalent to a large decrease in modulation potential.

[11] As for the spectra that are observed for GCR protons and He after about 2012.75, they will be discussed briefly next in this paper and more fully in subsequent papers. It should be noted that this is, as yet, only a brief glimpse of these GCR and ACR particles at a location still in close proximity to the boundary just crossed. Further surprises may be in store as V1 proceeds outward.

3 Discussion and Summary and Conclusions

[12] The sudden and large decreases seen in all energetic ACR type particles above ~0.5 MeV at 2012.65 is the most dominant feature for these particles at V1 since its launch 35 years ago. This intensity decrease, its suddenness, and associated anisotropies for nuclei above ~1 MeV define the features of the “boundary” just crossed. This is illustrated in Figure 4 which shows an expanded intensity versus time curve for the >0.5 MeV rate. This shows the suddenness of this event and the precursors, with the intensities decreasing by factors of 2–3 in 1 day or less.

Figure 4.

Expanded intensity versus time profile of >0.5 MeV daily average rate for the time period 2012.5 to 2012.8. Times of barrier crossings are labeled 1–5.

[13] The possible multiple boundary crossings are labeled 1–5 in the figure, with the crossings 1, 3, and 5 representing crossings in which V1 moves from inside to outside. Crossings 1 and 5 are particularly interesting. V1 appears to have first crossed the boundary during the tracking period on 28 July. The intensity >0.5 MeV decreases rapidly at a rate ~5% per hour. On the following day, the intensity bottoms out at a value ~40% of that in the middle of the previous day. During decrease 5, the intensity also drops to ~30% of its initial value in just 1 day. For a stationary boundary, these decreases would correspond to an intensity decrease e-folding distance of less than ~0.01 AU.

[14] If V1 is passing through a stationary medium, the decreases/increases would be caused by the passage of V1 through ribbons of field connected to the region beyond the barrier. These ribbons would be ~0.01 and ~0.02 AU thick, respectively, corresponding to the ~3 day event starting on 28 July and the 6–7 day event starting on 14 August.

[15] During the 28 days between the first precursor decrease and the “final” decrease, V1 moves outward 0.3 AU. In a nonstationary scenario, since V1 moves from inside to outside the barrier on crossings 1, 3, and 5, the barrier itself must be moving. The intensity decreases on 28 July and 13 August could then be the result of outward motion of the barrier location, and the intensity-time profiles of these decreases could be the result of speed variations of the barrier movements. The times between decreases 1, 3, and 5, which have remarkably similar initial intensity time profiles, are 16 and 12 days, respectively, about 0.5 of a solar rotation period. So the precursor decreases could, in this case, be the result of the boundary movement rather than V1 moving through stationary structures just inside the barrier itself.

[16] Future study of the variations of the remaining nuclei below ~10 MeV after 2012.78 and magnetic field data will help to determine whether this component is really a part of a low-energy galactic component or whether it is a weak “halo” of ACR around the heliosheath region. In any case, there are good possibilities for determining the local diffusion coefficient in the region beyond the barrier just by simply studying the temporal behavior and anisotropies of the H, He, and O intensities as V1 moves further beyond the barrier.

[17] We should note that both the electron spectrum and the spectra of heavier nuclei from Be to Fe and including C and O nuclei are also being measured at V1, although these data are not reported here. These new in situ measurements can be used to compare with galactic propagated spectra and understand better the propagation and source characteristics of these never before studied low-energy GCR. This study will include the singly ionized components such as N, O, Ne, and Ar that are part of the ACR component in the heliosheath, and the secondary components B and F that are created only during propagation in the galaxy.

[18] The electron spectra from ~2 to 100 MeV that V1 measures will be particularly valuable for understanding the propagation characteristics of the lowest-rigidity particles in the galaxy. The electrons have the advantage that their spectrum in the galaxy below ~1 GeV can be deduced from the galactic radio synchrotron spectrum observed between 1 and 100 MHz. In fact, it will now be possible to obtain a consistency check between the Voyager-measured electron spectrum and that predicted from the galactic radio synchrotron spectrum. This comparison utilizes the average galactic plasma parameters (ne, T) and the magnetic field (B), which is also measured by Voyager.

[19] The observed intensities of galactic H and He nuclei after about 2012.75 appear to peak at energies between ~30 and 60 MeV and then to decrease slightly at lower energies similar to that described in some galactic propagation models [see, e.g., Putze et al., 2010]. There is no evidence of even a small residual solar modulation which would rapidly decrease the intensity of the lowest-energy H and He nuclei. So, from this, it appears that V1 has exited the main solar modulation region, revealing H and He spectra characteristic of those to be expected in the local interstellar medium (LISM).

[20] Finally, what about the “boundary” or the “heliocliff” that Voyager crossed several times between 28 July and 25 August? Well, it constitutes an almost impenetrable barrier for energetic heliospheric ACR nuclei accelerated and confined within the boundary. It is also, by far, the most significant barrier to the GCR particles that has been encountered by V1 in the outer heliosheath for both energetic GCR nuclei and low-energy GCR electrons, both of which are moving inward.

[21] The increase which occurred earlier on 8 May 2012 contributes ~0.5 of the total increase of GCR but does not produce a significant decrease of ACR.

[22] If the GCR and ACR intensities continue to remain at their present levels, then indeed this “heliocliff” region displays many of the properties of a “classical” heliopause, perhaps a much more impressive barrier to inward and outward transport of energetic particles than would have been anticipated.

[23] There are many interesting effects that could be observed beyond this barrier as the heliospheric neutral H is mediated in the LISM. These include possible low-level modulation of GCR and effects on the plasma of the LISM [see, e.g., Zank et al., 2013]. Future observations at V1 will hopefully settle many of these issues as the spacecraft proceeds further into this uncharted region.


[24] This article was conceived by our Voyager colleague, Frank B. McDonald, who is no longer with us. Frank, we have been working together for over 55 years to reach the goal of actually observing the interstellar spectra of cosmic rays, possibly now achieved almost on the day of your passing. You wanted so badly to be able to finish this article that you had already started. Together we did it. Bon Voyage!

[25] This work, under the able direction of project manager Ed Stone, is supported by JPL. The data used here come from the Web sites and, maintained by Nand Lal and Bryant Heikkila.