7. A current sheet
The heliospheric current sheet is considered by many as the most likely cause for most if not all DEs. In fact one can find in the literature statements as strong as: "[Analysis of ] 19 DEs in Halley's comet leaves little doubt that DEs are associated primarily with crossings of the HCS and apparently not with any other properties of the solar wind such as high speed streams, dense regions, or dynamic pressure increases" (Brandt et al. 1999, p. 76). Such a statement relies explicitly on the assumption that all (or most) DEs have a common cause. It relies implicitly on the assumption that DEs form a homogeneous class of phenomena.
We start our calculation from model 'slow1' (in fact we start from the final state of the fast to slow transition reported in Sect. 6). At time t=0 the IMF in the solar wind is replaced by its negative value.
There is numerical diffusion in the numerical code due to limited resolution of the computational grid. Thus, magnetic field of opposite polarity diffuses and cancels. Energy conservation transforms magnetic into thermal energy and so replaces magnetic by thermal pressure. Magnetic pressure acts with an adiabatic index , while thermal pressure acts only with . Thus by converting magnetic into thermal energy the plasma becomes more compressible. This has the consequence that the plasma near the current sheet is adiabatically compressed. The density is enhanced.
Fig. 15 shows that there is no deflection of the tail.
The point of maximum brightness is in the stationary model 'slow1' at a distance of 20 000 km behind the nucleus. After 1 h it moves closer to the nucleus to a distance of 10 000 km and then recedes in an accelerated motion into the tail with an acceleration of 3.3 m s-2. We recall from Sect. 3 that this is only slightly below the acceleration in the stationary model 'slow1'. Therefore, the cloud is simply convected by the flow.
This blob in the tail is brighter than the coma for three hours. Due to the diffusion (or reconnection) the field around the HCS is depleted. Therefore, less momentum is transported to the central part of the tail where the flow remains slower and denser. This leaves its signature in the ions per tail length profile in the upper panel of Fig. 15. After the passage of the current sheet the tail returns to its original appearance, which is insensitive to the sign of the magnetic field.
The blob in the tail can also easily be followed in the images shown in Fig. 14. It is most conspicuous after 4 and 5 hours when the tail looks like a snake after swallowing a rabbit.
When the field is rotated by an angle less than 180o then diffusion and reconnection can also deplete magnetic field but only the antiparallel component which is a fraction of . In the case of a 90o rotation this amounts to 50%. The cloud in the tail found by Schmidt-Voigt (1989), p. 444, in a simulation of a 90o field rotation can be explained by the same mechanism as described above in connection with our results for the current sheet.
The results of this section rely on the diffusion of magnetic field. The simulations simply use the diffusion caused by the numerical scheme. But as already noted by Schmidt-Voigt (1989) the numerical diffusion exaggerates the field diffusion in a cometary plasma by several orders of magnitude. Thus, the results of this section are not very realistic.
© European Southern Observatory (ESO) 2000
Online publication: June 8, 2000