[1] Recently, Manchester and Zurbuchen [2006] used results from a global MHD simulation of a coronal mass ejection (CME) propagating in the solar wind to propose a new interpretation for the properties of a set of nine heliospheric disturbances observed by the Ulysses spacecraft at high heliographic latitudes. These events were observed predominantly, but not exclusively, during the declining phase of the solar cycle. Gosling et al. [1994a, 1994b] originally suggested that such events could be produced by the overexpansion of high pressure (either magnetic, thermal, or some combination thereof) CMEs, leading to their name: “Overexpanding” CMEs. In this scenario (modeled numerically in two dimensions by Riley et al. [1997]) a fast CME erupting at low heliographic latitudes, into a bimodal ambient solar wind, would expand in all dimensions, penetrating into the high-speed wind at high latitudes. At low latitudes, the interplanetary CME (ICME) propagates significantly faster than the ambient solar wind and drives a forward (F) shock ahead of it. At high latitudes, however, the ICME becomes embedded within the fast coronal hole stream. The ejecta's high internal pressure, relative to the surrounding wind, drives a fast-mode wave away from it in almost all directions, leading to the observed F wave (or shock) ahead of it, and the R wave (or shock) behind it. [2] In contrast, Manchester and Zurbuchen [2006] offered an interpretation based on their simulation of a fast ICME, which lacked an enhanced internal pressure, interacting with an idealized axisymmetric bimodal ambient wind that, nevertheless, was traveling considerably slower than the ICME at all heliographic latitudes. Their simulation produced a F shock at all latitudes driven by the relative motion between the ICME and the ambient wind ahead. They found that the slow wind ahead of the ICME was deflected to higher latitudes, whereas behind the ICME the fast wind from the polar coronal hole was deflected to lower latitudes. A compression region, which subsequently steepened into a R shock, formed where those deflected flows collided poleward of the ICME. The authors thus suggested that the F/R shock pairs observed at high heliographic latitudes resulted from this sort of interaction rather than from overexpansion of ICMEs and that Ulysses never actually encountered the ICMEs producing the observed disturbances. Moreover, their simulation indicated that in such disturbances the plasma immediately downstream of the F shock is deflected poleward while immediately downstream of the R shock it is deflected equatorward. The purpose of this comment is to argue that the observations continue to favor the original interpretation. [3] A total of nine events [Gosling et al., 1998; Reisenfeld et al., 2003] have been observed at high latitudes that can be broadly described by the following characteristics: (1) They are found at high heliographic latitudes (preferentially during the declining phase of the solar cycle and at solar minimum), embedded within high-speed streams from coronal holes; (2) The disturbance profiles are relatively symmetric; (3) Following the passage of the F shock, the disturbances have declining speed profiles indicating expansion; (4) At their centers the disturbances have relative minima in pressure, density, temperature, and magnetic field strength, with enhancements in these quantities on either side; (5) In most of the events, rotations in the magnetic field suggest the presence of a magnetic flux rope; (6) They are usually bounded by F and R shocks or waves, which stand off from the edges of the identified ejecta by approximately the same distances; (7) The identified ejecta have speeds that are comparable to the background high-speed wind within which they are immersed; (8) The events usually contain intervals where counterstreaming suprathermal electrons (CSEs) are present, suggesting the presence of closed magnetic field lines; and (9) In one case, a low-latitude counterpart was observed by an in-ecliptic spacecraft. Figure 1 displays a particularly good event, which displays essentially all of the these signatures [Gosling et al., 1994a, 1994b]. Table 1 summarizes the basic properties of all nine events. The characterization of the density, temperature, He++/H+, and β within the event as “low,” “high,” and/or “variable” was based on comparisons with the ambient solar wind ahead, and is, to some extent, subjective. Nevertheless, given the complexity that often exists within these profiles, a more quantitative measure would not have been more insightful. Clearly, all events were associated with low plasma-β, anomalous He++/H+ ratio, and low density and temperature. Furthermore, all but one contained CSEs, and the majority of events appeared to contain a flux rope. [4] Manchester and Zurbuchen [2006] argued that in addition to reproducing the basic features of the observations, their interpretation addresses some deficiencies of the overexpansion explanation. Specifically: (1) that the compositional signatures of the events more closely match the ambient high-speed solar wind than that of ICMEs; (2) that a R shock is present at all; (3) that these events are observed only at high latitudes; and (4) that these events are not observed at solar maximum. As outlined below, however, we find that the overexpansion model is consistent with all of these points. Moreover, as we will demonstrate, and has been mentioned by Manchester and Zurbuchen [2006], their model fails to explain at least two crucial aspects of the observations. [5] Consider first the issue of ion charge states. As summarized in Table 1, these events are typically associated with anomalous (high or low) He++/H+ ratios, although there does not appear to be a systematic trend to these variations. In contrast, the He++/H+ ratio in the ambient solar wind is relatively constant, at ∼4.4% [Feldman et al., 1996]. This suggests that the material in the interiors of these events is commonly associated with unique source regions. Studies of ICMEs have demonstrated that ionic composition measurements, while extremely useful, are not a necessary and sufficient criteria for detecting ICMEs [e.g., Lepri et al., 2001]. Moreover, interpretation of composition variations is difficult. We do not know, for example, the basic spatial distribution of composition/elemental abundance tracers within ICMEs. [6] Next consider the presence of the R shock. Manchester and Zurbuchen [2006] suggest that since their R shock results from interactions in the solar wind well away from the Sun, it does not suffer from the problem of how that wave escapes from the Sun if it forms below the magnetosonic critical point. The overexpansion model makes no claims as to the heliocentric radius where the R wave forms. As envisaged by Riley et al. [1997], a compact, high-pressure ejecta is launched at low latitudes, and predominantly into slow solar wind. As it propagates away from the Sun it expands and penetrates into the higher-speed streams above (and below) it that originate in coronal holes. Where it penetrates the high-speed wind is the point that must be compared with the magnetosonic critical point. There is no reason why this could not occur at heliocentric distances beyond 10–15 RS. [7] Consider now the point that these events are observed only at high latitudes and not at low latitudes. As we have discussed in relation to the overexpansion model [Gosling et al., 1995; Riley et al., 1997], in general, at low latitudes, there is actually a more complex evolution of waves, including the formation of two pairs of F and R waves. As a fast ICME overtakes and compresses the ambient solar wind ahead, it develops a region of high pressure at the leading edge of the ICME that is bounded by a F and R wave. Since the ICME initially has a higher pressure than the surrounding plasma, as it expands a second pair of F and R waves forms centered on the ICME and propagating outward. The R wave associated with the expansion of the compressed region at the leading edge of the ICME interacts with the F wave associated with the expansion of the ICME, weakening both in the process. In addition, the R wave associated with the expansion of the CME does not develop into a shock at low latitudes because the ICME propagates away from slow ambient solar wind behind it faster than the wave can propagate. The net result is that at low latitudes, where the ICME is traveling significantly faster than the ambient solar wind ahead and behind, only one wave remains and steepens into a F shock ahead of the ejecta. One could ask, however, why the type of events observed at high latitudes has not been observed at low latitudes. To produce such an event, a CME would have to have been launched with an internal pressure significantly higher than the ambient slow solar wind, yet at speeds comparable to the slow wind. One possible explanation is that only fast, explosive events are likely to have this high internal pressure. Thus they only manifest the over-expansion mechanism when they become embedded within the fast wind. That is, the events capable of overexpanding must necessarily be significantly faster than the ambient slow solar wind and hence are observed as the events we described above. [8] One distinguishing feature of the Manchester and Zurbuchen [2006] interpretation is that the spacecraft does not sample the ICME directly in these events, but rather a complex disturbance poleward of the ICME. We do not doubt that disturbances driven by ICMEs propagate to higher latitudes than the ejecta itself reaches. In fact, Riley et al. [1997] found that the F and R shocks generated through overexpansion extended considerably further poleward than the meridional expansion of the ejecta, suggesting that it might be possible to observe symmetric profiles of F/R shock pairs without intercepting the actual ICMEs. In this case, some plasma and magnetic signatures of the overexpansion might still exist. For example, relative minima in the plasma temperature and density would still occur near the center of the disturbances due to the poleward propagation of the waves (see third panel of Plate 3 of Riley et al. [1997]). Additionally, from more recent MHD simulations [Riley et al., 2003], we suggest that rotations in the magnetic field might be associated with draping of ambient interplanetary magnetic field lines within the sheath region behind the F shock that is driven by relative motion between a fast ICME and the ambient wind at low latitudes. On the other hand, we would not anticipate either the presence of CSEs for these cases, and, to a lesser extent, anomalous He++/H+ ratios. As Table 1 clearly shows, CSEs were present to some extent within the interiors of all but one event. [9] The Manchester and Zurbuchen [2006] interpretation predicts the presence of systematic meridional flow deflections at the F and R shocks (or waves). Their Figures 3d–3f, for example, suggest meridional deflections of up to ±25 km s−1 at 1 AU. In contrast, for the events in Table 1 we find (1) no obvious meridional flow deflections downstream of the 6/93 R shock, the 4/94 F shock, the 10/96 F shock, the 9/01 R shock, the 10/01 F shock and the 10/01 R shock, (2) the meridional deflections were equatorward, rather than poleward, for the remaining three F waves and (3) the meridional deflections were poleward, rather than equatorward, for the rest of the R waves except for the 2/93 event. That is, any systematic trends in the data were opposite to those predicted by the Manchester and Zurbuchen model. (It is worth remarking that the meridional flows across F and R waves associated with tilted corotating interaction regions (CIRs) are considerably more systematic and distinguishable from the events discussed here [Gosling et al., 1993; Riley et al., 1996]). [10] In this comment, we have argued that the high-latitude forward-reverse wave disturbance observations are adequately explained by the over-expansion model and that the deficiencies of the model in this respect noted by Manchester and Zurbuchen [2006] are more apparent than real. Moreover, we find that their own model fails to explain the high-latitude disturbance observations in two important respects. First, it fails to explain why ICME-like signatures are observed within most of these events, and second, it predicts systematic meridional flow deflections associated with these events that are not observed. [11] PR gratefully acknowledges the support of the National Aeronautics and Space Administration (LWS, GI, and Theory programs) and the National Science Foundation (SHINE and CISM Programs). JTG was supported by the Ulysses program and by NASA GI grant NNG055GJ55G. The authors would like to thank Ruth Skoug for her help in preparing Figure 1. [13] Amitava Bhattacharjee thanks the reviewers for their assistance in evaluating this paper. Please note: The publisher is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.
No takes yet. Share an insight, caveat, or question.
Riley et al. (2007) studied this question.
Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context: