Comment on "Interplanetary coronal mass ejections that are undetected by solar coronagraphs" by T. A. Howard and G. M. Simnett

Howard and Simnett (HS) employed a new technique for associating LASCO CMEs to SMEI ICMEs. In order to extrapolate the SMEI data back to the LASCO field of view they used nonlinear trajectories, dependent on a speed and direction, what is more realistic than the linear extrapolation with only one parameter (a speed). However, there are two errors and one mistake in their procedure: (1) HS used two free parameters of the direction, whereas only one can be freely selected, because the second is provided by SMEI data. As a result, the directions determined by HS are incorrect. (2) HS overlooked that, since the trajectory depends on more than one parameter, there is a broad set of trajectories, for various speeds and directions, matching the event, and thus a broad range of the onset times. HS select only one trajectory for each SMEI event. Therefore the associations made by them are incomplete, and they should be reexamined. As long as it is not done any conclusion about CMEs undetected by solar coronagraphs are premature. (3) HS made some mistake in determination of the SMEI speeds. The speeds given in their Table 1 are about twice as high as those demanded to obtain the onset times given in the table. It explains why the SMEI speed distribution is excessively shifted toward high speeds ; hence, there is no reason to search for a physical explanation.

💡 Research Summary

Howard and Simnett (HS) introduced a novel method for linking interplanetary coronal mass ejections (ICMEs) observed by the Solar Mass Ejection Imager (SMEI) to coronal mass ejections (CMEs) recorded by the LASCO coronagraph. Their approach replaces the traditional linear back‑extrapolation—characterized by a single speed parameter—with a nonlinear trajectory model that depends on both a propagation speed and a direction vector. The authors argue that this two‑parameter model is more realistic because it accounts for the three‑dimensional geometry of CME propagation.

The present comment identifies three fundamental flaws in HS’s methodology that undermine their principal claim: that a substantial number of ICMEs are associated with CMEs that were completely missed by LASCO.

1. Over‑parameterization of direction – SMEI provides a measured position angle (PA) for each ICME, which uniquely determines one component of the propagation direction (e.g., the azimuth in heliographic coordinates). Consequently, only one directional degree of freedom (typically the latitude or elevation) can be varied freely. HS, however, treated both azimuth and elevation as independent free parameters. By allowing both to float, they generated direction solutions that are not constrained by the actual SMEI observation, leading to systematically erroneous propagation vectors.

2. Neglect of the solution manifold – The nonlinear trajectory equation, with speed (v) and the single free directional angle (θ) as variables, admits a continuum of (v, θ) pairs that reproduce the same SMEI time‑height point. Each pair corresponds to a distinct CME launch time at the Sun. HS selected a single (v, θ) combination for each event without explaining the selection criterion and without exploring the full family of admissible solutions. This practice collapses a potentially wide range of launch times into a single value, artificially narrowing the temporal association window and inflating the apparent success rate of CME‑ICME matches.

3. Systematic speed overestimation – The speeds listed in HS’s Table 1 are roughly twice the magnitude required to reproduce the onset times they report when inserted into the trajectory formula. This discrepancy indicates a computational or transcription error in the speed determination step. As a result, the derived speed distribution is heavily biased toward high speeds, prompting HS to invoke a physical explanation for an apparently anomalous excess of fast ICMEs. In reality, the bias is an artifact of the erroneous speed values; correcting the speeds would shift the distribution toward lower, more typical CME speeds and eliminate the need for a speculative physical interpretation.

Collectively, these errors cast serious doubt on the robustness of HS’s CME‑ICME associations. The incorrect direction vectors, the omission of the full (v, θ) solution space, and the inflated speed estimates all contribute to an incomplete and potentially misleading picture of which LASCO CMEs correspond to SMEI ICMEs. Consequently, the central assertion—that a significant population of ICMEs originates from CMEs that were entirely undetected by coronagraphs—remains unproven.

The comment recommends a rigorous re‑analysis that (i) fixes the direction component supplied by SMEI, (ii) performs a systematic scan of the remaining directional degree of freedom and speed to map the entire admissible solution manifold, and (iii) validates speed calculations against independent CME catalogs. Only after such a comprehensive treatment can one reliably assess the fraction of “missing” CMEs and draw meaningful conclusions about CME detection completeness.

In summary, while HS’s effort to improve CME‑ICME linkage is commendable, the methodological oversights identified here invalidate their quantitative results. A corrected, exhaustive approach is essential before any claim about coronagraph‑invisible CMEs can be accepted.