Three‐dimensional propagation of coronal mass ejections (CMEs) in a structured solar wind flow: 2. CME launched adjacent to the streamer belt

A three‐dimensional (3‐D) numerical hydrodynamic model is used to investigate the evolution of coronal mass ejections (CMEs) launched at several heliographic positions into a tilted‐dipole ambient solar wind (SW) flow, which is appropriate around solar activity minimum and declining phase. The CME is injected as an overpressured plasma cloud. Results show that the motion and local appearance of a CME in interplanetary space is strongly affected by its interaction with the background SW velocity and density structures. The most complicated scenario occurs when a CME is injected directly into the slow stream‐belt flow. Temporal profiles corresponding to the interaction of fast CME and slow SW flows, to the merging of CME‐driven and corotating disturbances at the fast stream leading edge, and to expanding structures at the trailing edge of the preceding fast stream are presented. The simplest configuration is when a CME is injected just to the west of the streamer belt. In that case the CME‐driven disturbance evolves and propagates freely into the preceding fast stream, untouched by the streamer belt slow flow near which it originates, except for a small segment at low latitudes. When a CME is injected just to the east of the streamer belt, the CME expansion and shock evolution are constrained by the slow, dense flow ahead of the fast stream in which the CME propagates initially. Low‐latitude portions of the CME are trapped between the slow streamer belt flow and the leading edge of the oncoming fast stream. In this case, CME‐driven and corotating shock structures can merge. These results are related to a previous study, where the CME is launched into the midst of the slow streamer belt flow. Numerical results confirm recent Ulysses findings about the overexpanding CME structure, the extreme latitudinal distortions in the CME shape, and the inclination of shock fronts. These results also provide a basis for interpreting variations in the radial width of the CME and the shock stand‐off distance. Simulations of spatial structures resulting from 3‐D dynamic interactions help illuminate differing temporal onsets and profiles of observed CMEs.